Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Pull in the 'net' tree to get Daniel Borkmann's flow dissector
infrastructure change.
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/Documentation/ABI/testing/sysfs-class-net-mesh b/Documentation/ABI/testing/sysfs-class-net-mesh
index bc41da6..bdcd8b4 100644
--- a/Documentation/ABI/testing/sysfs-class-net-mesh
+++ b/Documentation/ABI/testing/sysfs-class-net-mesh
@@ -67,6 +67,14 @@
Defines the penalty which will be applied to an
originator message's tq-field on every hop.
+What: /sys/class/net/<mesh_iface>/mesh/network_coding
+Date: Nov 2012
+Contact: Martin Hundeboll <martin@hundeboll.net>
+Description:
+ Controls whether Network Coding (using some magic
+ to send fewer wifi packets but still the same
+ content) is enabled or not.
+
What: /sys/class/net/<mesh_iface>/mesh/orig_interval
Date: May 2010
Contact: Marek Lindner <lindner_marek@yahoo.de>
diff --git a/Documentation/devicetree/bindings/net/cpsw.txt b/Documentation/devicetree/bindings/net/cpsw.txt
index ecfdf75..4f2ca6b 100644
--- a/Documentation/devicetree/bindings/net/cpsw.txt
+++ b/Documentation/devicetree/bindings/net/cpsw.txt
@@ -15,16 +15,22 @@
- mac_control : Specifies Default MAC control register content
for the specific platform
- slaves : Specifies number for slaves
-- cpts_active_slave : Specifies the slave to use for time stamping
+- active_slave : Specifies the slave to use for time stamping,
+ ethtool and SIOCGMIIPHY
- cpts_clock_mult : Numerator to convert input clock ticks into nanoseconds
- cpts_clock_shift : Denominator to convert input clock ticks into nanoseconds
-- phy_id : Specifies slave phy id
-- mac-address : Specifies slave MAC address
Optional properties:
- ti,hwmods : Must be "cpgmac0"
- no_bd_ram : Must be 0 or 1
- dual_emac : Specifies Switch to act as Dual EMAC
+
+Slave Properties:
+Required properties:
+- phy_id : Specifies slave phy id
+- mac-address : Specifies slave MAC address
+
+Optional properties:
- dual_emac_res_vlan : Specifies VID to be used to segregate the ports
Note: "ti,hwmods" field is used to fetch the base address and irq
@@ -47,7 +53,7 @@
rx_descs = <64>;
mac_control = <0x20>;
slaves = <2>;
- cpts_active_slave = <0>;
+ active_slave = <0>;
cpts_clock_mult = <0x80000000>;
cpts_clock_shift = <29>;
cpsw_emac0: slave@0 {
@@ -73,7 +79,7 @@
rx_descs = <64>;
mac_control = <0x20>;
slaves = <2>;
- cpts_active_slave = <0>;
+ active_slave = <0>;
cpts_clock_mult = <0x80000000>;
cpts_clock_shift = <29>;
cpsw_emac0: slave@0 {
diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index dc2dc87..17953e2 100644
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -29,7 +29,7 @@
neigh/default/gc_thresh1 - INTEGER
Minimum number of entries to keep. Garbage collector will not
purge entries if there are fewer than this number.
- Default: 256
+ Default: 128
neigh/default/gc_thresh3 - INTEGER
Maximum number of neighbor entries allowed. Increase this
@@ -175,14 +175,6 @@
is inherited.
[see setsockopt(listenfd, SOL_TCP, TCP_CONGESTION, "name" ...) ]
-tcp_cookie_size - INTEGER
- Default size of TCP Cookie Transactions (TCPCT) option, that may be
- overridden on a per socket basis by the TCPCT socket option.
- Values greater than the maximum (16) are interpreted as the maximum.
- Values greater than zero and less than the minimum (8) are interpreted
- as the minimum. Odd values are interpreted as the next even value.
- Default: 0 (off).
-
tcp_dsack - BOOLEAN
Allows TCP to send "duplicate" SACKs.
@@ -190,7 +182,9 @@
Enable Early Retransmit (ER), per RFC 5827. ER lowers the threshold
for triggering fast retransmit when the amount of outstanding data is
small and when no previously unsent data can be transmitted (such
- that limited transmit could be used).
+ that limited transmit could be used). Also controls the use of
+ Tail loss probe (TLP) that converts RTOs occuring due to tail
+ losses into fast recovery (draft-dukkipati-tcpm-tcp-loss-probe-01).
Possible values:
0 disables ER
1 enables ER
@@ -198,7 +192,9 @@
by a fourth of RTT. This mitigates connection falsely
recovers when network has a small degree of reordering
(less than 3 packets).
- Default: 2
+ 3 enables delayed ER and TLP.
+ 4 enables TLP only.
+ Default: 3
tcp_ecn - INTEGER
Control use of Explicit Congestion Notification (ECN) by TCP.
diff --git a/arch/arm/boot/dts/am33xx.dtsi b/arch/arm/boot/dts/am33xx.dtsi
index 0957645..91fe4f1 100644
--- a/arch/arm/boot/dts/am33xx.dtsi
+++ b/arch/arm/boot/dts/am33xx.dtsi
@@ -349,7 +349,7 @@
rx_descs = <64>;
mac_control = <0x20>;
slaves = <2>;
- cpts_active_slave = <0>;
+ active_slave = <0>;
cpts_clock_mult = <0x80000000>;
cpts_clock_shift = <29>;
reg = <0x4a100000 0x800
diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index a3fde52..65c30ea 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -511,12 +511,16 @@
static int send_connect(struct c4iw_ep *ep)
{
struct cpl_act_open_req *req;
+ struct cpl_t5_act_open_req *t5_req;
struct sk_buff *skb;
u64 opt0;
u32 opt2;
unsigned int mtu_idx;
int wscale;
- int wrlen = roundup(sizeof *req, 16);
+ int size = is_t4(ep->com.dev->rdev.lldi.adapter_type) ?
+ sizeof(struct cpl_act_open_req) :
+ sizeof(struct cpl_t5_act_open_req);
+ int wrlen = roundup(size, 16);
PDBG("%s ep %p atid %u\n", __func__, ep, ep->atid);
@@ -552,17 +556,36 @@
opt2 |= WND_SCALE_EN(1);
t4_set_arp_err_handler(skb, NULL, act_open_req_arp_failure);
- req = (struct cpl_act_open_req *) skb_put(skb, wrlen);
- INIT_TP_WR(req, 0);
- OPCODE_TID(req) = cpu_to_be32(
- MK_OPCODE_TID(CPL_ACT_OPEN_REQ, ((ep->rss_qid<<14)|ep->atid)));
- req->local_port = ep->com.local_addr.sin_port;
- req->peer_port = ep->com.remote_addr.sin_port;
- req->local_ip = ep->com.local_addr.sin_addr.s_addr;
- req->peer_ip = ep->com.remote_addr.sin_addr.s_addr;
- req->opt0 = cpu_to_be64(opt0);
- req->params = cpu_to_be32(select_ntuple(ep->com.dev, ep->dst, ep->l2t));
- req->opt2 = cpu_to_be32(opt2);
+ if (is_t4(ep->com.dev->rdev.lldi.adapter_type)) {
+ req = (struct cpl_act_open_req *) skb_put(skb, wrlen);
+ INIT_TP_WR(req, 0);
+ OPCODE_TID(req) = cpu_to_be32(
+ MK_OPCODE_TID(CPL_ACT_OPEN_REQ,
+ ((ep->rss_qid << 14) | ep->atid)));
+ req->local_port = ep->com.local_addr.sin_port;
+ req->peer_port = ep->com.remote_addr.sin_port;
+ req->local_ip = ep->com.local_addr.sin_addr.s_addr;
+ req->peer_ip = ep->com.remote_addr.sin_addr.s_addr;
+ req->opt0 = cpu_to_be64(opt0);
+ req->params = cpu_to_be32(select_ntuple(ep->com.dev,
+ ep->dst, ep->l2t));
+ req->opt2 = cpu_to_be32(opt2);
+ } else {
+ t5_req = (struct cpl_t5_act_open_req *) skb_put(skb, wrlen);
+ INIT_TP_WR(t5_req, 0);
+ OPCODE_TID(t5_req) = cpu_to_be32(
+ MK_OPCODE_TID(CPL_ACT_OPEN_REQ,
+ ((ep->rss_qid << 14) | ep->atid)));
+ t5_req->local_port = ep->com.local_addr.sin_port;
+ t5_req->peer_port = ep->com.remote_addr.sin_port;
+ t5_req->local_ip = ep->com.local_addr.sin_addr.s_addr;
+ t5_req->peer_ip = ep->com.remote_addr.sin_addr.s_addr;
+ t5_req->opt0 = cpu_to_be64(opt0);
+ t5_req->params = cpu_to_be64(V_FILTER_TUPLE(
+ select_ntuple(ep->com.dev, ep->dst, ep->l2t)));
+ t5_req->opt2 = cpu_to_be32(opt2);
+ }
+
set_bit(ACT_OPEN_REQ, &ep->com.history);
return c4iw_l2t_send(&ep->com.dev->rdev, skb, ep->l2t);
}
@@ -1676,9 +1699,9 @@
case CPL_ERR_CONN_TIMEDOUT:
break;
case CPL_ERR_TCAM_FULL:
+ dev->rdev.stats.tcam_full++;
if (dev->rdev.lldi.enable_fw_ofld_conn) {
mutex_lock(&dev->rdev.stats.lock);
- dev->rdev.stats.tcam_full++;
mutex_unlock(&dev->rdev.stats.lock);
send_fw_act_open_req(ep,
GET_TID_TID(GET_AOPEN_ATID(
@@ -2875,12 +2898,14 @@
static void build_cpl_pass_accept_req(struct sk_buff *skb, int stid , u8 tos)
{
u32 l2info;
- u16 vlantag, len, hdr_len;
+ u16 vlantag, len, hdr_len, eth_hdr_len;
u8 intf;
struct cpl_rx_pkt *cpl = cplhdr(skb);
struct cpl_pass_accept_req *req;
struct tcp_options_received tmp_opt;
+ struct c4iw_dev *dev;
+ dev = *((struct c4iw_dev **) (skb->cb + sizeof(void *)));
/* Store values from cpl_rx_pkt in temporary location. */
vlantag = (__force u16) cpl->vlan;
len = (__force u16) cpl->len;
@@ -2896,7 +2921,7 @@
*/
memset(&tmp_opt, 0, sizeof(tmp_opt));
tcp_clear_options(&tmp_opt);
- tcp_parse_options(skb, &tmp_opt, NULL, 0, NULL);
+ tcp_parse_options(skb, &tmp_opt, 0, NULL);
req = (struct cpl_pass_accept_req *)__skb_push(skb, sizeof(*req));
memset(req, 0, sizeof(*req));
@@ -2904,14 +2929,16 @@
V_SYN_MAC_IDX(G_RX_MACIDX(
(__force int) htonl(l2info))) |
F_SYN_XACT_MATCH);
+ eth_hdr_len = is_t4(dev->rdev.lldi.adapter_type) ?
+ G_RX_ETHHDR_LEN((__force int) htonl(l2info)) :
+ G_RX_T5_ETHHDR_LEN((__force int) htonl(l2info));
req->hdr_len = cpu_to_be32(V_SYN_RX_CHAN(G_RX_CHAN(
(__force int) htonl(l2info))) |
V_TCP_HDR_LEN(G_RX_TCPHDR_LEN(
(__force int) htons(hdr_len))) |
V_IP_HDR_LEN(G_RX_IPHDR_LEN(
(__force int) htons(hdr_len))) |
- V_ETH_HDR_LEN(G_RX_ETHHDR_LEN(
- (__force int) htonl(l2info))));
+ V_ETH_HDR_LEN(G_RX_ETHHDR_LEN(eth_hdr_len)));
req->vlan = (__force __be16) vlantag;
req->len = (__force __be16) len;
req->tos_stid = cpu_to_be32(PASS_OPEN_TID(stid) |
@@ -2999,7 +3026,7 @@
u16 window;
struct port_info *pi;
struct net_device *pdev;
- u16 rss_qid;
+ u16 rss_qid, eth_hdr_len;
int step;
u32 tx_chan;
struct neighbour *neigh;
@@ -3028,7 +3055,10 @@
goto reject;
}
- if (G_RX_ETHHDR_LEN(ntohl(cpl->l2info)) == ETH_HLEN) {
+ eth_hdr_len = is_t4(dev->rdev.lldi.adapter_type) ?
+ G_RX_ETHHDR_LEN(htonl(cpl->l2info)) :
+ G_RX_T5_ETHHDR_LEN(htonl(cpl->l2info));
+ if (eth_hdr_len == ETH_HLEN) {
eh = (struct ethhdr *)(req + 1);
iph = (struct iphdr *)(eh + 1);
} else {
diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c
index 80069ad..ae65601 100644
--- a/drivers/infiniband/hw/cxgb4/device.c
+++ b/drivers/infiniband/hw/cxgb4/device.c
@@ -41,10 +41,20 @@
#define DRV_VERSION "0.1"
MODULE_AUTHOR("Steve Wise");
-MODULE_DESCRIPTION("Chelsio T4 RDMA Driver");
+MODULE_DESCRIPTION("Chelsio T4/T5 RDMA Driver");
MODULE_LICENSE("Dual BSD/GPL");
MODULE_VERSION(DRV_VERSION);
+static int allow_db_fc_on_t5;
+module_param(allow_db_fc_on_t5, int, 0644);
+MODULE_PARM_DESC(allow_db_fc_on_t5,
+ "Allow DB Flow Control on T5 (default = 0)");
+
+static int allow_db_coalescing_on_t5;
+module_param(allow_db_coalescing_on_t5, int, 0644);
+MODULE_PARM_DESC(allow_db_coalescing_on_t5,
+ "Allow DB Coalescing on T5 (default = 0)");
+
struct uld_ctx {
struct list_head entry;
struct cxgb4_lld_info lldi;
@@ -614,7 +624,7 @@
{
return infop->vr->stag.size > 0 && infop->vr->pbl.size > 0 &&
infop->vr->rq.size > 0 && infop->vr->qp.size > 0 &&
- infop->vr->cq.size > 0 && infop->vr->ocq.size > 0;
+ infop->vr->cq.size > 0;
}
static struct c4iw_dev *c4iw_alloc(const struct cxgb4_lld_info *infop)
@@ -627,6 +637,22 @@
pci_name(infop->pdev));
return ERR_PTR(-ENOSYS);
}
+ if (!ocqp_supported(infop))
+ pr_info("%s: On-Chip Queues not supported on this device.\n",
+ pci_name(infop->pdev));
+
+ if (!is_t4(infop->adapter_type)) {
+ if (!allow_db_fc_on_t5) {
+ db_fc_threshold = 100000;
+ pr_info("DB Flow Control Disabled.\n");
+ }
+
+ if (!allow_db_coalescing_on_t5) {
+ db_coalescing_threshold = -1;
+ pr_info("DB Coalescing Disabled.\n");
+ }
+ }
+
devp = (struct c4iw_dev *)ib_alloc_device(sizeof(*devp));
if (!devp) {
printk(KERN_ERR MOD "Cannot allocate ib device\n");
@@ -678,8 +704,8 @@
int i;
if (!vers_printed++)
- printk(KERN_INFO MOD "Chelsio T4 RDMA Driver - version %s\n",
- DRV_VERSION);
+ pr_info("Chelsio T4/T5 RDMA Driver - version %s\n",
+ DRV_VERSION);
ctx = kzalloc(sizeof *ctx, GFP_KERNEL);
if (!ctx) {
diff --git a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
index 7eec5e1..485183a 100644
--- a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
+++ b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
@@ -162,7 +162,7 @@
return min((int)T4_MAX_NUM_STAG, (int)(rdev->lldi.vr->stag.size >> 5));
}
-#define C4IW_WR_TO (10*HZ)
+#define C4IW_WR_TO (30*HZ)
struct c4iw_wr_wait {
struct completion completion;
@@ -369,7 +369,6 @@
DEFINE_DMA_UNMAP_ADDR(mapping);
dma_addr_t dma_addr;
struct c4iw_dev *dev;
- int size;
};
static inline struct c4iw_fr_page_list *to_c4iw_fr_page_list(
@@ -817,6 +816,15 @@
return wscale;
}
+static inline int ocqp_supported(const struct cxgb4_lld_info *infop)
+{
+#if defined(__i386__) || defined(__x86_64__) || defined(CONFIG_PPC64)
+ return infop->vr->ocq.size > 0;
+#else
+ return 0;
+#endif
+}
+
u32 c4iw_id_alloc(struct c4iw_id_table *alloc);
void c4iw_id_free(struct c4iw_id_table *alloc, u32 obj);
int c4iw_id_table_alloc(struct c4iw_id_table *alloc, u32 start, u32 num,
@@ -930,6 +938,8 @@
extern c4iw_handler_func c4iw_handlers[NUM_CPL_CMDS];
extern int c4iw_max_read_depth;
extern int db_fc_threshold;
+extern int db_coalescing_threshold;
+extern int use_dsgl;
#endif
diff --git a/drivers/infiniband/hw/cxgb4/mem.c b/drivers/infiniband/hw/cxgb4/mem.c
index 903a92d..4cb8eb2 100644
--- a/drivers/infiniband/hw/cxgb4/mem.c
+++ b/drivers/infiniband/hw/cxgb4/mem.c
@@ -30,16 +30,76 @@
* SOFTWARE.
*/
+#include <linux/module.h>
+#include <linux/moduleparam.h>
#include <rdma/ib_umem.h>
#include <linux/atomic.h>
#include "iw_cxgb4.h"
+int use_dsgl = 1;
+module_param(use_dsgl, int, 0644);
+MODULE_PARM_DESC(use_dsgl, "Use DSGL for PBL/FastReg (default=1)");
+
#define T4_ULPTX_MIN_IO 32
#define C4IW_MAX_INLINE_SIZE 96
+#define T4_ULPTX_MAX_DMA 1024
+#define C4IW_INLINE_THRESHOLD 128
-static int write_adapter_mem(struct c4iw_rdev *rdev, u32 addr, u32 len,
- void *data)
+static int inline_threshold = C4IW_INLINE_THRESHOLD;
+module_param(inline_threshold, int, 0644);
+MODULE_PARM_DESC(inline_threshold, "inline vs dsgl threshold (default=128)");
+
+static int _c4iw_write_mem_dma_aligned(struct c4iw_rdev *rdev, u32 addr,
+ u32 len, dma_addr_t data, int wait)
+{
+ struct sk_buff *skb;
+ struct ulp_mem_io *req;
+ struct ulptx_sgl *sgl;
+ u8 wr_len;
+ int ret = 0;
+ struct c4iw_wr_wait wr_wait;
+
+ addr &= 0x7FFFFFF;
+
+ if (wait)
+ c4iw_init_wr_wait(&wr_wait);
+ wr_len = roundup(sizeof(*req) + sizeof(*sgl), 16);
+
+ skb = alloc_skb(wr_len, GFP_KERNEL | __GFP_NOFAIL);
+ if (!skb)
+ return -ENOMEM;
+ set_wr_txq(skb, CPL_PRIORITY_CONTROL, 0);
+
+ req = (struct ulp_mem_io *)__skb_put(skb, wr_len);
+ memset(req, 0, wr_len);
+ INIT_ULPTX_WR(req, wr_len, 0, 0);
+ req->wr.wr_hi = cpu_to_be32(FW_WR_OP(FW_ULPTX_WR) |
+ (wait ? FW_WR_COMPL(1) : 0));
+ req->wr.wr_lo = wait ? (__force __be64)&wr_wait : 0;
+ req->wr.wr_mid = cpu_to_be32(FW_WR_LEN16(DIV_ROUND_UP(wr_len, 16)));
+ req->cmd = cpu_to_be32(ULPTX_CMD(ULP_TX_MEM_WRITE));
+ req->cmd |= cpu_to_be32(V_T5_ULP_MEMIO_ORDER(1));
+ req->dlen = cpu_to_be32(ULP_MEMIO_DATA_LEN(len>>5));
+ req->len16 = cpu_to_be32(DIV_ROUND_UP(wr_len-sizeof(req->wr), 16));
+ req->lock_addr = cpu_to_be32(ULP_MEMIO_ADDR(addr));
+
+ sgl = (struct ulptx_sgl *)(req + 1);
+ sgl->cmd_nsge = cpu_to_be32(ULPTX_CMD(ULP_TX_SC_DSGL) |
+ ULPTX_NSGE(1));
+ sgl->len0 = cpu_to_be32(len);
+ sgl->addr0 = cpu_to_be64(data);
+
+ ret = c4iw_ofld_send(rdev, skb);
+ if (ret)
+ return ret;
+ if (wait)
+ ret = c4iw_wait_for_reply(rdev, &wr_wait, 0, 0, __func__);
+ return ret;
+}
+
+static int _c4iw_write_mem_inline(struct c4iw_rdev *rdev, u32 addr, u32 len,
+ void *data)
{
struct sk_buff *skb;
struct ulp_mem_io *req;
@@ -47,6 +107,12 @@
u8 wr_len, *to_dp, *from_dp;
int copy_len, num_wqe, i, ret = 0;
struct c4iw_wr_wait wr_wait;
+ __be32 cmd = cpu_to_be32(ULPTX_CMD(ULP_TX_MEM_WRITE));
+
+ if (is_t4(rdev->lldi.adapter_type))
+ cmd |= cpu_to_be32(ULP_MEMIO_ORDER(1));
+ else
+ cmd |= cpu_to_be32(V_T5_ULP_MEMIO_IMM(1));
addr &= 0x7FFFFFF;
PDBG("%s addr 0x%x len %u\n", __func__, addr, len);
@@ -77,7 +143,7 @@
req->wr.wr_mid = cpu_to_be32(
FW_WR_LEN16(DIV_ROUND_UP(wr_len, 16)));
- req->cmd = cpu_to_be32(ULPTX_CMD(ULP_TX_MEM_WRITE) | (1<<23));
+ req->cmd = cmd;
req->dlen = cpu_to_be32(ULP_MEMIO_DATA_LEN(
DIV_ROUND_UP(copy_len, T4_ULPTX_MIN_IO)));
req->len16 = cpu_to_be32(DIV_ROUND_UP(wr_len-sizeof(req->wr),
@@ -107,6 +173,67 @@
return ret;
}
+int _c4iw_write_mem_dma(struct c4iw_rdev *rdev, u32 addr, u32 len, void *data)
+{
+ u32 remain = len;
+ u32 dmalen;
+ int ret = 0;
+ dma_addr_t daddr;
+ dma_addr_t save;
+
+ daddr = dma_map_single(&rdev->lldi.pdev->dev, data, len, DMA_TO_DEVICE);
+ if (dma_mapping_error(&rdev->lldi.pdev->dev, daddr))
+ return -1;
+ save = daddr;
+
+ while (remain > inline_threshold) {
+ if (remain < T4_ULPTX_MAX_DMA) {
+ if (remain & ~T4_ULPTX_MIN_IO)
+ dmalen = remain & ~(T4_ULPTX_MIN_IO-1);
+ else
+ dmalen = remain;
+ } else
+ dmalen = T4_ULPTX_MAX_DMA;
+ remain -= dmalen;
+ ret = _c4iw_write_mem_dma_aligned(rdev, addr, dmalen, daddr,
+ !remain);
+ if (ret)
+ goto out;
+ addr += dmalen >> 5;
+ data += dmalen;
+ daddr += dmalen;
+ }
+ if (remain)
+ ret = _c4iw_write_mem_inline(rdev, addr, remain, data);
+out:
+ dma_unmap_single(&rdev->lldi.pdev->dev, save, len, DMA_TO_DEVICE);
+ return ret;
+}
+
+/*
+ * write len bytes of data into addr (32B aligned address)
+ * If data is NULL, clear len byte of memory to zero.
+ */
+static int write_adapter_mem(struct c4iw_rdev *rdev, u32 addr, u32 len,
+ void *data)
+{
+ if (is_t5(rdev->lldi.adapter_type) && use_dsgl) {
+ if (len > inline_threshold) {
+ if (_c4iw_write_mem_dma(rdev, addr, len, data)) {
+ printk_ratelimited(KERN_WARNING
+ "%s: dma map"
+ " failure (non fatal)\n",
+ pci_name(rdev->lldi.pdev));
+ return _c4iw_write_mem_inline(rdev, addr, len,
+ data);
+ } else
+ return 0;
+ } else
+ return _c4iw_write_mem_inline(rdev, addr, len, data);
+ } else
+ return _c4iw_write_mem_inline(rdev, addr, len, data);
+}
+
/*
* Build and write a TPT entry.
* IN: stag key, pdid, perm, bind_enabled, zbva, to, len, page_size,
@@ -760,19 +887,23 @@
struct c4iw_fr_page_list *c4pl;
struct c4iw_dev *dev = to_c4iw_dev(device);
dma_addr_t dma_addr;
- int size = sizeof *c4pl + page_list_len * sizeof(u64);
+ int pll_len = roundup(page_list_len * sizeof(u64), 32);
- c4pl = dma_alloc_coherent(&dev->rdev.lldi.pdev->dev, size,
- &dma_addr, GFP_KERNEL);
+ c4pl = kmalloc(sizeof(*c4pl), GFP_KERNEL);
if (!c4pl)
return ERR_PTR(-ENOMEM);
+ c4pl->ibpl.page_list = dma_alloc_coherent(&dev->rdev.lldi.pdev->dev,
+ pll_len, &dma_addr,
+ GFP_KERNEL);
+ if (!c4pl->ibpl.page_list) {
+ kfree(c4pl);
+ return ERR_PTR(-ENOMEM);
+ }
dma_unmap_addr_set(c4pl, mapping, dma_addr);
c4pl->dma_addr = dma_addr;
c4pl->dev = dev;
- c4pl->size = size;
- c4pl->ibpl.page_list = (u64 *)(c4pl + 1);
- c4pl->ibpl.max_page_list_len = page_list_len;
+ c4pl->ibpl.max_page_list_len = pll_len;
return &c4pl->ibpl;
}
@@ -781,8 +912,10 @@
{
struct c4iw_fr_page_list *c4pl = to_c4iw_fr_page_list(ibpl);
- dma_free_coherent(&c4pl->dev->rdev.lldi.pdev->dev, c4pl->size,
- c4pl, dma_unmap_addr(c4pl, mapping));
+ dma_free_coherent(&c4pl->dev->rdev.lldi.pdev->dev,
+ c4pl->ibpl.max_page_list_len,
+ c4pl->ibpl.page_list, dma_unmap_addr(c4pl, mapping));
+ kfree(c4pl);
}
int c4iw_dereg_mr(struct ib_mr *ib_mr)
diff --git a/drivers/infiniband/hw/cxgb4/provider.c b/drivers/infiniband/hw/cxgb4/provider.c
index e084fdc..7e94c9a 100644
--- a/drivers/infiniband/hw/cxgb4/provider.c
+++ b/drivers/infiniband/hw/cxgb4/provider.c
@@ -162,8 +162,14 @@
*/
if (addr >= rdev->oc_mw_pa)
vma->vm_page_prot = t4_pgprot_wc(vma->vm_page_prot);
- else
- vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
+ else {
+ if (is_t5(rdev->lldi.adapter_type))
+ vma->vm_page_prot =
+ t4_pgprot_wc(vma->vm_page_prot);
+ else
+ vma->vm_page_prot =
+ pgprot_noncached(vma->vm_page_prot);
+ }
ret = io_remap_pfn_range(vma, vma->vm_start,
addr >> PAGE_SHIFT,
len, vma->vm_page_prot);
@@ -263,7 +269,7 @@
dev = to_c4iw_dev(ibdev);
memset(props, 0, sizeof *props);
memcpy(&props->sys_image_guid, dev->rdev.lldi.ports[0]->dev_addr, 6);
- props->hw_ver = dev->rdev.lldi.adapter_type;
+ props->hw_ver = CHELSIO_CHIP_RELEASE(dev->rdev.lldi.adapter_type);
props->fw_ver = dev->rdev.lldi.fw_vers;
props->device_cap_flags = dev->device_cap_flags;
props->page_size_cap = T4_PAGESIZE_MASK;
@@ -346,7 +352,8 @@
struct c4iw_dev *c4iw_dev = container_of(dev, struct c4iw_dev,
ibdev.dev);
PDBG("%s dev 0x%p\n", __func__, dev);
- return sprintf(buf, "%d\n", c4iw_dev->rdev.lldi.adapter_type);
+ return sprintf(buf, "%d\n",
+ CHELSIO_CHIP_RELEASE(c4iw_dev->rdev.lldi.adapter_type));
}
static ssize_t show_fw_ver(struct device *dev, struct device_attribute *attr,
diff --git a/drivers/infiniband/hw/cxgb4/qp.c b/drivers/infiniband/hw/cxgb4/qp.c
index 17ba4f8..9fe6f1e 100644
--- a/drivers/infiniband/hw/cxgb4/qp.c
+++ b/drivers/infiniband/hw/cxgb4/qp.c
@@ -42,10 +42,21 @@
module_param(ocqp_support, int, 0644);
MODULE_PARM_DESC(ocqp_support, "Support on-chip SQs (default=1)");
-int db_fc_threshold = 2000;
+int db_fc_threshold = 1000;
module_param(db_fc_threshold, int, 0644);
-MODULE_PARM_DESC(db_fc_threshold, "QP count/threshold that triggers automatic "
- "db flow control mode (default = 2000)");
+MODULE_PARM_DESC(db_fc_threshold,
+ "QP count/threshold that triggers"
+ " automatic db flow control mode (default = 1000)");
+
+int db_coalescing_threshold;
+module_param(db_coalescing_threshold, int, 0644);
+MODULE_PARM_DESC(db_coalescing_threshold,
+ "QP count/threshold that triggers"
+ " disabling db coalescing (default = 0)");
+
+static int max_fr_immd = T4_MAX_FR_IMMD;
+module_param(max_fr_immd, int, 0644);
+MODULE_PARM_DESC(max_fr_immd, "fastreg threshold for using DSGL instead of immedate");
static void set_state(struct c4iw_qp *qhp, enum c4iw_qp_state state)
{
@@ -76,7 +87,7 @@
static int alloc_oc_sq(struct c4iw_rdev *rdev, struct t4_sq *sq)
{
- if (!ocqp_support || !t4_ocqp_supported())
+ if (!ocqp_support || !ocqp_supported(&rdev->lldi))
return -ENOSYS;
sq->dma_addr = c4iw_ocqp_pool_alloc(rdev, sq->memsize);
if (!sq->dma_addr)
@@ -129,7 +140,7 @@
int wr_len;
struct c4iw_wr_wait wr_wait;
struct sk_buff *skb;
- int ret;
+ int ret = 0;
int eqsize;
wq->sq.qid = c4iw_get_qpid(rdev, uctx);
@@ -169,17 +180,14 @@
}
if (user) {
- ret = alloc_oc_sq(rdev, &wq->sq);
+ if (alloc_oc_sq(rdev, &wq->sq) && alloc_host_sq(rdev, &wq->sq))
+ goto free_hwaddr;
+ } else {
+ ret = alloc_host_sq(rdev, &wq->sq);
if (ret)
goto free_hwaddr;
+ }
- ret = alloc_host_sq(rdev, &wq->sq);
- if (ret)
- goto free_sq;
- } else
- ret = alloc_host_sq(rdev, &wq->sq);
- if (ret)
- goto free_hwaddr;
memset(wq->sq.queue, 0, wq->sq.memsize);
dma_unmap_addr_set(&wq->sq, mapping, wq->sq.dma_addr);
@@ -532,7 +540,7 @@
}
static int build_fastreg(struct t4_sq *sq, union t4_wr *wqe,
- struct ib_send_wr *wr, u8 *len16)
+ struct ib_send_wr *wr, u8 *len16, u8 t5dev)
{
struct fw_ri_immd *imdp;
@@ -554,28 +562,51 @@
wqe->fr.va_hi = cpu_to_be32(wr->wr.fast_reg.iova_start >> 32);
wqe->fr.va_lo_fbo = cpu_to_be32(wr->wr.fast_reg.iova_start &
0xffffffff);
- WARN_ON(pbllen > T4_MAX_FR_IMMD);
- imdp = (struct fw_ri_immd *)(&wqe->fr + 1);
- imdp->op = FW_RI_DATA_IMMD;
- imdp->r1 = 0;
- imdp->r2 = 0;
- imdp->immdlen = cpu_to_be32(pbllen);
- p = (__be64 *)(imdp + 1);
- rem = pbllen;
- for (i = 0; i < wr->wr.fast_reg.page_list_len; i++) {
- *p = cpu_to_be64((u64)wr->wr.fast_reg.page_list->page_list[i]);
- rem -= sizeof *p;
- if (++p == (__be64 *)&sq->queue[sq->size])
- p = (__be64 *)sq->queue;
+
+ if (t5dev && use_dsgl && (pbllen > max_fr_immd)) {
+ struct c4iw_fr_page_list *c4pl =
+ to_c4iw_fr_page_list(wr->wr.fast_reg.page_list);
+ struct fw_ri_dsgl *sglp;
+
+ for (i = 0; i < wr->wr.fast_reg.page_list_len; i++) {
+ wr->wr.fast_reg.page_list->page_list[i] = (__force u64)
+ cpu_to_be64((u64)
+ wr->wr.fast_reg.page_list->page_list[i]);
+ }
+
+ sglp = (struct fw_ri_dsgl *)(&wqe->fr + 1);
+ sglp->op = FW_RI_DATA_DSGL;
+ sglp->r1 = 0;
+ sglp->nsge = cpu_to_be16(1);
+ sglp->addr0 = cpu_to_be64(c4pl->dma_addr);
+ sglp->len0 = cpu_to_be32(pbllen);
+
+ *len16 = DIV_ROUND_UP(sizeof(wqe->fr) + sizeof(*sglp), 16);
+ } else {
+ imdp = (struct fw_ri_immd *)(&wqe->fr + 1);
+ imdp->op = FW_RI_DATA_IMMD;
+ imdp->r1 = 0;
+ imdp->r2 = 0;
+ imdp->immdlen = cpu_to_be32(pbllen);
+ p = (__be64 *)(imdp + 1);
+ rem = pbllen;
+ for (i = 0; i < wr->wr.fast_reg.page_list_len; i++) {
+ *p = cpu_to_be64(
+ (u64)wr->wr.fast_reg.page_list->page_list[i]);
+ rem -= sizeof(*p);
+ if (++p == (__be64 *)&sq->queue[sq->size])
+ p = (__be64 *)sq->queue;
+ }
+ BUG_ON(rem < 0);
+ while (rem) {
+ *p = 0;
+ rem -= sizeof(*p);
+ if (++p == (__be64 *)&sq->queue[sq->size])
+ p = (__be64 *)sq->queue;
+ }
+ *len16 = DIV_ROUND_UP(sizeof(wqe->fr) + sizeof(*imdp)
+ + pbllen, 16);
}
- BUG_ON(rem < 0);
- while (rem) {
- *p = 0;
- rem -= sizeof *p;
- if (++p == (__be64 *)&sq->queue[sq->size])
- p = (__be64 *)sq->queue;
- }
- *len16 = DIV_ROUND_UP(sizeof wqe->fr + sizeof *imdp + pbllen, 16);
return 0;
}
@@ -676,7 +707,10 @@
case IB_WR_FAST_REG_MR:
fw_opcode = FW_RI_FR_NSMR_WR;
swsqe->opcode = FW_RI_FAST_REGISTER;
- err = build_fastreg(&qhp->wq.sq, wqe, wr, &len16);
+ err = build_fastreg(&qhp->wq.sq, wqe, wr, &len16,
+ is_t5(
+ qhp->rhp->rdev.lldi.adapter_type) ?
+ 1 : 0);
break;
case IB_WR_LOCAL_INV:
if (wr->send_flags & IB_SEND_FENCE)
@@ -1448,6 +1482,9 @@
rhp->db_state = NORMAL;
idr_for_each(&rhp->qpidr, enable_qp_db, NULL);
}
+ if (db_coalescing_threshold >= 0)
+ if (rhp->qpcnt <= db_coalescing_threshold)
+ cxgb4_enable_db_coalescing(rhp->rdev.lldi.ports[0]);
spin_unlock_irq(&rhp->lock);
atomic_dec(&qhp->refcnt);
wait_event(qhp->wait, !atomic_read(&qhp->refcnt));
@@ -1559,11 +1596,15 @@
spin_lock_irq(&rhp->lock);
if (rhp->db_state != NORMAL)
t4_disable_wq_db(&qhp->wq);
- if (++rhp->qpcnt > db_fc_threshold && rhp->db_state == NORMAL) {
+ rhp->qpcnt++;
+ if (rhp->qpcnt > db_fc_threshold && rhp->db_state == NORMAL) {
rhp->rdev.stats.db_state_transitions++;
rhp->db_state = FLOW_CONTROL;
idr_for_each(&rhp->qpidr, disable_qp_db, NULL);
}
+ if (db_coalescing_threshold >= 0)
+ if (rhp->qpcnt > db_coalescing_threshold)
+ cxgb4_disable_db_coalescing(rhp->rdev.lldi.ports[0]);
ret = insert_handle_nolock(rhp, &rhp->qpidr, qhp, qhp->wq.sq.qid);
spin_unlock_irq(&rhp->lock);
if (ret)
diff --git a/drivers/infiniband/hw/cxgb4/t4.h b/drivers/infiniband/hw/cxgb4/t4.h
index 16f26ab..ebcb03b 100644
--- a/drivers/infiniband/hw/cxgb4/t4.h
+++ b/drivers/infiniband/hw/cxgb4/t4.h
@@ -84,7 +84,7 @@
sizeof(struct fw_ri_isgl)) / sizeof(struct fw_ri_sge))
#define T4_MAX_FR_IMMD ((T4_SQ_NUM_BYTES - sizeof(struct fw_ri_fr_nsmr_wr) - \
sizeof(struct fw_ri_immd)) & ~31UL)
-#define T4_MAX_FR_DEPTH (T4_MAX_FR_IMMD / sizeof(u64))
+#define T4_MAX_FR_DEPTH (1024 / sizeof(u64))
#define T4_RQ_NUM_SLOTS 2
#define T4_RQ_NUM_BYTES (T4_EQ_ENTRY_SIZE * T4_RQ_NUM_SLOTS)
@@ -280,15 +280,6 @@
#endif
}
-static inline int t4_ocqp_supported(void)
-{
-#if defined(__i386__) || defined(__x86_64__) || defined(CONFIG_PPC64)
- return 1;
-#else
- return 0;
-#endif
-}
-
enum {
T4_SQ_ONCHIP = (1<<0),
};
diff --git a/drivers/isdn/capi/capidrv.c b/drivers/isdn/capi/capidrv.c
index 832bc80..cc9f192 100644
--- a/drivers/isdn/capi/capidrv.c
+++ b/drivers/isdn/capi/capidrv.c
@@ -469,8 +469,7 @@
{
struct ncci_datahandle_queue *n, **pp;
- n = (struct ncci_datahandle_queue *)
- kmalloc(sizeof(struct ncci_datahandle_queue), GFP_ATOMIC);
+ n = kmalloc(sizeof(struct ncci_datahandle_queue), GFP_ATOMIC);
if (!n) {
printk(KERN_ERR "capidrv: kmalloc ncci_datahandle failed\n");
return -1;
diff --git a/drivers/isdn/hisax/fsm.c b/drivers/isdn/hisax/fsm.c
index 1bb2910..c7a9471 100644
--- a/drivers/isdn/hisax/fsm.c
+++ b/drivers/isdn/hisax/fsm.c
@@ -26,7 +26,7 @@
{
int i;
- fsm->jumpmatrix = (FSMFNPTR *)
+ fsm->jumpmatrix =
kzalloc(sizeof(FSMFNPTR) * fsm->state_count * fsm->event_count, GFP_KERNEL);
if (!fsm->jumpmatrix)
return -ENOMEM;
diff --git a/drivers/isdn/hisax/hfc_sx.c b/drivers/isdn/hisax/hfc_sx.c
index 90f34ae..dc4574f 100644
--- a/drivers/isdn/hisax/hfc_sx.c
+++ b/drivers/isdn/hisax/hfc_sx.c
@@ -1479,7 +1479,7 @@
release_region(cs->hw.hfcsx.base, 2);
return (0);
}
- if (!(cs->hw.hfcsx.extra = (void *)
+ if (!(cs->hw.hfcsx.extra =
kmalloc(sizeof(struct hfcsx_extra), GFP_ATOMIC))) {
release_region(cs->hw.hfcsx.base, 2);
printk(KERN_WARNING "HFC-SX: unable to allocate memory\n");
diff --git a/drivers/net/caif/Kconfig b/drivers/net/caif/Kconfig
index 60c2142..a966128 100644
--- a/drivers/net/caif/Kconfig
+++ b/drivers/net/caif/Kconfig
@@ -32,13 +32,6 @@
help to synchronize to the next transfer in case of over or under-runs.
This option also needs to be enabled on the modem.
-config CAIF_SHM
- tristate "CAIF shared memory protocol driver"
- depends on CAIF && U5500_MBOX
- default n
- ---help---
- The CAIF shared memory protocol driver for the STE UX5500 platform.
-
config CAIF_HSI
tristate "CAIF HSI transport driver"
depends on CAIF
diff --git a/drivers/net/caif/Makefile b/drivers/net/caif/Makefile
index 91dff86..15a9d2fc7 100644
--- a/drivers/net/caif/Makefile
+++ b/drivers/net/caif/Makefile
@@ -7,9 +7,5 @@
cfspi_slave-objs := caif_spi.o caif_spi_slave.o
obj-$(CONFIG_CAIF_SPI_SLAVE) += cfspi_slave.o
-# Shared memory
-caif_shm-objs := caif_shmcore.o caif_shm_u5500.o
-obj-$(CONFIG_CAIF_SHM) += caif_shm.o
-
# HSI interface
obj-$(CONFIG_CAIF_HSI) += caif_hsi.o
diff --git a/drivers/net/caif/caif_shm_u5500.c b/drivers/net/caif/caif_shm_u5500.c
deleted file mode 100644
index 89d76b7..0000000
--- a/drivers/net/caif/caif_shm_u5500.c
+++ /dev/null
@@ -1,128 +0,0 @@
-/*
- * Copyright (C) ST-Ericsson AB 2010
- * Contact: Sjur Brendeland / sjur.brandeland@stericsson.com
- * Author: Amarnath Revanna / amarnath.bangalore.revanna@stericsson.com
- * License terms: GNU General Public License (GPL) version 2
- */
-
-#define pr_fmt(fmt) KBUILD_MODNAME ":" fmt
-
-#include <linux/init.h>
-#include <linux/module.h>
-#include <linux/netdevice.h>
-#include <mach/mbox-db5500.h>
-#include <net/caif/caif_shm.h>
-
-MODULE_LICENSE("GPL");
-MODULE_DESCRIPTION("CAIF Shared Memory protocol driver");
-
-#define MAX_SHM_INSTANCES 1
-
-enum {
- MBX_ACC0,
- MBX_ACC1,
- MBX_DSP
-};
-
-static struct shmdev_layer shmdev_lyr[MAX_SHM_INSTANCES];
-
-static unsigned int shm_start;
-static unsigned int shm_size;
-
-module_param(shm_size, uint , 0440);
-MODULE_PARM_DESC(shm_total_size, "Start of SHM shared memory");
-
-module_param(shm_start, uint , 0440);
-MODULE_PARM_DESC(shm_total_start, "Total Size of SHM shared memory");
-
-static int shmdev_send_msg(u32 dev_id, u32 mbx_msg)
-{
- /* Always block until msg is written successfully */
- mbox_send(shmdev_lyr[dev_id].hmbx, mbx_msg, true);
- return 0;
-}
-
-static int shmdev_mbx_setup(void *pshmdrv_cb, struct shmdev_layer *pshm_dev,
- void *pshm_drv)
-{
- /*
- * For UX5500, we have only 1 SHM instance which uses MBX0
- * for communication with the peer modem
- */
- pshm_dev->hmbx = mbox_setup(MBX_ACC0, pshmdrv_cb, pshm_drv);
-
- if (!pshm_dev->hmbx)
- return -ENODEV;
- else
- return 0;
-}
-
-static int __init caif_shmdev_init(void)
-{
- int i, result;
-
- /* Loop is currently overkill, there is only one instance */
- for (i = 0; i < MAX_SHM_INSTANCES; i++) {
-
- shmdev_lyr[i].shm_base_addr = shm_start;
- shmdev_lyr[i].shm_total_sz = shm_size;
-
- if (((char *)shmdev_lyr[i].shm_base_addr == NULL)
- || (shmdev_lyr[i].shm_total_sz <= 0)) {
- pr_warn("ERROR,"
- "Shared memory Address and/or Size incorrect"
- ", Bailing out ...\n");
- result = -EINVAL;
- goto clean;
- }
-
- pr_info("SHM AREA (instance %d) STARTS"
- " AT %p\n", i, (char *)shmdev_lyr[i].shm_base_addr);
-
- shmdev_lyr[i].shm_id = i;
- shmdev_lyr[i].pshmdev_mbxsend = shmdev_send_msg;
- shmdev_lyr[i].pshmdev_mbxsetup = shmdev_mbx_setup;
-
- /*
- * Finally, CAIF core module is called with details in place:
- * 1. SHM base address
- * 2. SHM size
- * 3. MBX handle
- */
- result = caif_shmcore_probe(&shmdev_lyr[i]);
- if (result) {
- pr_warn("ERROR[%d],"
- "Could not probe SHM core (instance %d)"
- " Bailing out ...\n", result, i);
- goto clean;
- }
- }
-
- return 0;
-
-clean:
- /*
- * For now, we assume that even if one instance of SHM fails, we bail
- * out of the driver support completely. For this, we need to release
- * any memory allocated and unregister any instance of SHM net device.
- */
- for (i = 0; i < MAX_SHM_INSTANCES; i++) {
- if (shmdev_lyr[i].pshm_netdev)
- unregister_netdev(shmdev_lyr[i].pshm_netdev);
- }
- return result;
-}
-
-static void __exit caif_shmdev_exit(void)
-{
- int i;
-
- for (i = 0; i < MAX_SHM_INSTANCES; i++) {
- caif_shmcore_remove(shmdev_lyr[i].pshm_netdev);
- kfree((void *)shmdev_lyr[i].shm_base_addr);
- }
-
-}
-
-module_init(caif_shmdev_init);
-module_exit(caif_shmdev_exit);
diff --git a/drivers/net/caif/caif_shmcore.c b/drivers/net/caif/caif_shmcore.c
deleted file mode 100644
index bce8bac..0000000
--- a/drivers/net/caif/caif_shmcore.c
+++ /dev/null
@@ -1,747 +0,0 @@
-/*
- * Copyright (C) ST-Ericsson AB 2010
- * Contact: Sjur Brendeland / sjur.brandeland@stericsson.com
- * Authors: Amarnath Revanna / amarnath.bangalore.revanna@stericsson.com,
- * Daniel Martensson / daniel.martensson@stericsson.com
- * License terms: GNU General Public License (GPL) version 2
- */
-
-#define pr_fmt(fmt) KBUILD_MODNAME ":" fmt
-
-#include <linux/spinlock.h>
-#include <linux/sched.h>
-#include <linux/list.h>
-#include <linux/netdevice.h>
-#include <linux/if_arp.h>
-#include <linux/io.h>
-
-#include <net/caif/caif_device.h>
-#include <net/caif/caif_shm.h>
-
-#define NR_TX_BUF 6
-#define NR_RX_BUF 6
-#define TX_BUF_SZ 0x2000
-#define RX_BUF_SZ 0x2000
-
-#define CAIF_NEEDED_HEADROOM 32
-
-#define CAIF_FLOW_ON 1
-#define CAIF_FLOW_OFF 0
-
-#define LOW_WATERMARK 3
-#define HIGH_WATERMARK 4
-
-/* Maximum number of CAIF buffers per shared memory buffer. */
-#define SHM_MAX_FRMS_PER_BUF 10
-
-/*
- * Size in bytes of the descriptor area
- * (With end of descriptor signalling)
- */
-#define SHM_CAIF_DESC_SIZE ((SHM_MAX_FRMS_PER_BUF + 1) * \
- sizeof(struct shm_pck_desc))
-
-/*
- * Offset to the first CAIF frame within a shared memory buffer.
- * Aligned on 32 bytes.
- */
-#define SHM_CAIF_FRM_OFS (SHM_CAIF_DESC_SIZE + (SHM_CAIF_DESC_SIZE % 32))
-
-/* Number of bytes for CAIF shared memory header. */
-#define SHM_HDR_LEN 1
-
-/* Number of padding bytes for the complete CAIF frame. */
-#define SHM_FRM_PAD_LEN 4
-
-#define CAIF_MAX_MTU 4096
-
-#define SHM_SET_FULL(x) (((x+1) & 0x0F) << 0)
-#define SHM_GET_FULL(x) (((x >> 0) & 0x0F) - 1)
-
-#define SHM_SET_EMPTY(x) (((x+1) & 0x0F) << 4)
-#define SHM_GET_EMPTY(x) (((x >> 4) & 0x0F) - 1)
-
-#define SHM_FULL_MASK (0x0F << 0)
-#define SHM_EMPTY_MASK (0x0F << 4)
-
-struct shm_pck_desc {
- /*
- * Offset from start of shared memory area to start of
- * shared memory CAIF frame.
- */
- u32 frm_ofs;
- u32 frm_len;
-};
-
-struct buf_list {
- unsigned char *desc_vptr;
- u32 phy_addr;
- u32 index;
- u32 len;
- u32 frames;
- u32 frm_ofs;
- struct list_head list;
-};
-
-struct shm_caif_frm {
- /* Number of bytes of padding before the CAIF frame. */
- u8 hdr_ofs;
-};
-
-struct shmdrv_layer {
- /* caif_dev_common must always be first in the structure*/
- struct caif_dev_common cfdev;
-
- u32 shm_tx_addr;
- u32 shm_rx_addr;
- u32 shm_base_addr;
- u32 tx_empty_available;
- spinlock_t lock;
-
- struct list_head tx_empty_list;
- struct list_head tx_pend_list;
- struct list_head tx_full_list;
- struct list_head rx_empty_list;
- struct list_head rx_pend_list;
- struct list_head rx_full_list;
-
- struct workqueue_struct *pshm_tx_workqueue;
- struct workqueue_struct *pshm_rx_workqueue;
-
- struct work_struct shm_tx_work;
- struct work_struct shm_rx_work;
-
- struct sk_buff_head sk_qhead;
- struct shmdev_layer *pshm_dev;
-};
-
-static int shm_netdev_open(struct net_device *shm_netdev)
-{
- netif_wake_queue(shm_netdev);
- return 0;
-}
-
-static int shm_netdev_close(struct net_device *shm_netdev)
-{
- netif_stop_queue(shm_netdev);
- return 0;
-}
-
-int caif_shmdrv_rx_cb(u32 mbx_msg, void *priv)
-{
- struct buf_list *pbuf;
- struct shmdrv_layer *pshm_drv;
- struct list_head *pos;
- u32 avail_emptybuff = 0;
- unsigned long flags = 0;
-
- pshm_drv = priv;
-
- /* Check for received buffers. */
- if (mbx_msg & SHM_FULL_MASK) {
- int idx;
-
- spin_lock_irqsave(&pshm_drv->lock, flags);
-
- /* Check whether we have any outstanding buffers. */
- if (list_empty(&pshm_drv->rx_empty_list)) {
-
- /* Release spin lock. */
- spin_unlock_irqrestore(&pshm_drv->lock, flags);
-
- /* We print even in IRQ context... */
- pr_warn("No empty Rx buffers to fill: "
- "mbx_msg:%x\n", mbx_msg);
-
- /* Bail out. */
- goto err_sync;
- }
-
- pbuf =
- list_entry(pshm_drv->rx_empty_list.next,
- struct buf_list, list);
- idx = pbuf->index;
-
- /* Check buffer synchronization. */
- if (idx != SHM_GET_FULL(mbx_msg)) {
-
- /* We print even in IRQ context... */
- pr_warn(
- "phyif_shm_mbx_msg_cb: RX full out of sync:"
- " idx:%d, msg:%x SHM_GET_FULL(mbx_msg):%x\n",
- idx, mbx_msg, SHM_GET_FULL(mbx_msg));
-
- spin_unlock_irqrestore(&pshm_drv->lock, flags);
-
- /* Bail out. */
- goto err_sync;
- }
-
- list_del_init(&pbuf->list);
- list_add_tail(&pbuf->list, &pshm_drv->rx_full_list);
-
- spin_unlock_irqrestore(&pshm_drv->lock, flags);
-
- /* Schedule RX work queue. */
- if (!work_pending(&pshm_drv->shm_rx_work))
- queue_work(pshm_drv->pshm_rx_workqueue,
- &pshm_drv->shm_rx_work);
- }
-
- /* Check for emptied buffers. */
- if (mbx_msg & SHM_EMPTY_MASK) {
- int idx;
-
- spin_lock_irqsave(&pshm_drv->lock, flags);
-
- /* Check whether we have any outstanding buffers. */
- if (list_empty(&pshm_drv->tx_full_list)) {
-
- /* We print even in IRQ context... */
- pr_warn("No TX to empty: msg:%x\n", mbx_msg);
-
- spin_unlock_irqrestore(&pshm_drv->lock, flags);
-
- /* Bail out. */
- goto err_sync;
- }
-
- pbuf =
- list_entry(pshm_drv->tx_full_list.next,
- struct buf_list, list);
- idx = pbuf->index;
-
- /* Check buffer synchronization. */
- if (idx != SHM_GET_EMPTY(mbx_msg)) {
-
- spin_unlock_irqrestore(&pshm_drv->lock, flags);
-
- /* We print even in IRQ context... */
- pr_warn("TX empty "
- "out of sync:idx:%d, msg:%x\n", idx, mbx_msg);
-
- /* Bail out. */
- goto err_sync;
- }
- list_del_init(&pbuf->list);
-
- /* Reset buffer parameters. */
- pbuf->frames = 0;
- pbuf->frm_ofs = SHM_CAIF_FRM_OFS;
-
- list_add_tail(&pbuf->list, &pshm_drv->tx_empty_list);
-
- /* Check the available no. of buffers in the empty list */
- list_for_each(pos, &pshm_drv->tx_empty_list)
- avail_emptybuff++;
-
- /* Check whether we have to wake up the transmitter. */
- if ((avail_emptybuff > HIGH_WATERMARK) &&
- (!pshm_drv->tx_empty_available)) {
- pshm_drv->tx_empty_available = 1;
- spin_unlock_irqrestore(&pshm_drv->lock, flags);
- pshm_drv->cfdev.flowctrl
- (pshm_drv->pshm_dev->pshm_netdev,
- CAIF_FLOW_ON);
-
-
- /* Schedule the work queue. if required */
- if (!work_pending(&pshm_drv->shm_tx_work))
- queue_work(pshm_drv->pshm_tx_workqueue,
- &pshm_drv->shm_tx_work);
- } else
- spin_unlock_irqrestore(&pshm_drv->lock, flags);
- }
-
- return 0;
-
-err_sync:
- return -EIO;
-}
-
-static void shm_rx_work_func(struct work_struct *rx_work)
-{
- struct shmdrv_layer *pshm_drv;
- struct buf_list *pbuf;
- unsigned long flags = 0;
- struct sk_buff *skb;
- char *p;
- int ret;
-
- pshm_drv = container_of(rx_work, struct shmdrv_layer, shm_rx_work);
-
- while (1) {
-
- struct shm_pck_desc *pck_desc;
-
- spin_lock_irqsave(&pshm_drv->lock, flags);
-
- /* Check for received buffers. */
- if (list_empty(&pshm_drv->rx_full_list)) {
- spin_unlock_irqrestore(&pshm_drv->lock, flags);
- break;
- }
-
- pbuf =
- list_entry(pshm_drv->rx_full_list.next, struct buf_list,
- list);
- list_del_init(&pbuf->list);
- spin_unlock_irqrestore(&pshm_drv->lock, flags);
-
- /* Retrieve pointer to start of the packet descriptor area. */
- pck_desc = (struct shm_pck_desc *) pbuf->desc_vptr;
-
- /*
- * Check whether descriptor contains a CAIF shared memory
- * frame.
- */
- while (pck_desc->frm_ofs) {
- unsigned int frm_buf_ofs;
- unsigned int frm_pck_ofs;
- unsigned int frm_pck_len;
- /*
- * Check whether offset is within buffer limits
- * (lower).
- */
- if (pck_desc->frm_ofs <
- (pbuf->phy_addr - pshm_drv->shm_base_addr))
- break;
- /*
- * Check whether offset is within buffer limits
- * (higher).
- */
- if (pck_desc->frm_ofs >
- ((pbuf->phy_addr - pshm_drv->shm_base_addr) +
- pbuf->len))
- break;
-
- /* Calculate offset from start of buffer. */
- frm_buf_ofs =
- pck_desc->frm_ofs - (pbuf->phy_addr -
- pshm_drv->shm_base_addr);
-
- /*
- * Calculate offset and length of CAIF packet while
- * taking care of the shared memory header.
- */
- frm_pck_ofs =
- frm_buf_ofs + SHM_HDR_LEN +
- (*(pbuf->desc_vptr + frm_buf_ofs));
- frm_pck_len =
- (pck_desc->frm_len - SHM_HDR_LEN -
- (*(pbuf->desc_vptr + frm_buf_ofs)));
-
- /* Check whether CAIF packet is within buffer limits */
- if ((frm_pck_ofs + pck_desc->frm_len) > pbuf->len)
- break;
-
- /* Get a suitable CAIF packet and copy in data. */
- skb = netdev_alloc_skb(pshm_drv->pshm_dev->pshm_netdev,
- frm_pck_len + 1);
-
- if (skb == NULL) {
- pr_info("OOM: Try next frame in descriptor\n");
- break;
- }
-
- p = skb_put(skb, frm_pck_len);
- memcpy(p, pbuf->desc_vptr + frm_pck_ofs, frm_pck_len);
-
- skb->protocol = htons(ETH_P_CAIF);
- skb_reset_mac_header(skb);
- skb->dev = pshm_drv->pshm_dev->pshm_netdev;
-
- /* Push received packet up the stack. */
- ret = netif_rx_ni(skb);
-
- if (!ret) {
- pshm_drv->pshm_dev->pshm_netdev->stats.
- rx_packets++;
- pshm_drv->pshm_dev->pshm_netdev->stats.
- rx_bytes += pck_desc->frm_len;
- } else
- ++pshm_drv->pshm_dev->pshm_netdev->stats.
- rx_dropped;
- /* Move to next packet descriptor. */
- pck_desc++;
- }
-
- spin_lock_irqsave(&pshm_drv->lock, flags);
- list_add_tail(&pbuf->list, &pshm_drv->rx_pend_list);
-
- spin_unlock_irqrestore(&pshm_drv->lock, flags);
-
- }
-
- /* Schedule the work queue. if required */
- if (!work_pending(&pshm_drv->shm_tx_work))
- queue_work(pshm_drv->pshm_tx_workqueue, &pshm_drv->shm_tx_work);
-
-}
-
-static void shm_tx_work_func(struct work_struct *tx_work)
-{
- u32 mbox_msg;
- unsigned int frmlen, avail_emptybuff, append = 0;
- unsigned long flags = 0;
- struct buf_list *pbuf = NULL;
- struct shmdrv_layer *pshm_drv;
- struct shm_caif_frm *frm;
- struct sk_buff *skb;
- struct shm_pck_desc *pck_desc;
- struct list_head *pos;
-
- pshm_drv = container_of(tx_work, struct shmdrv_layer, shm_tx_work);
-
- do {
- /* Initialize mailbox message. */
- mbox_msg = 0x00;
- avail_emptybuff = 0;
-
- spin_lock_irqsave(&pshm_drv->lock, flags);
-
- /* Check for pending receive buffers. */
- if (!list_empty(&pshm_drv->rx_pend_list)) {
-
- pbuf = list_entry(pshm_drv->rx_pend_list.next,
- struct buf_list, list);
-
- list_del_init(&pbuf->list);
- list_add_tail(&pbuf->list, &pshm_drv->rx_empty_list);
- /*
- * Value index is never changed,
- * so read access should be safe.
- */
- mbox_msg |= SHM_SET_EMPTY(pbuf->index);
- }
-
- skb = skb_peek(&pshm_drv->sk_qhead);
-
- if (skb == NULL)
- goto send_msg;
- /* Check the available no. of buffers in the empty list */
- list_for_each(pos, &pshm_drv->tx_empty_list)
- avail_emptybuff++;
-
- if ((avail_emptybuff < LOW_WATERMARK) &&
- pshm_drv->tx_empty_available) {
- /* Update blocking condition. */
- pshm_drv->tx_empty_available = 0;
- spin_unlock_irqrestore(&pshm_drv->lock, flags);
- pshm_drv->cfdev.flowctrl
- (pshm_drv->pshm_dev->pshm_netdev,
- CAIF_FLOW_OFF);
- spin_lock_irqsave(&pshm_drv->lock, flags);
- }
- /*
- * We simply return back to the caller if we do not have space
- * either in Tx pending list or Tx empty list. In this case,
- * we hold the received skb in the skb list, waiting to
- * be transmitted once Tx buffers become available
- */
- if (list_empty(&pshm_drv->tx_empty_list))
- goto send_msg;
-
- /* Get the first free Tx buffer. */
- pbuf = list_entry(pshm_drv->tx_empty_list.next,
- struct buf_list, list);
- do {
- if (append) {
- skb = skb_peek(&pshm_drv->sk_qhead);
- if (skb == NULL)
- break;
- }
-
- frm = (struct shm_caif_frm *)
- (pbuf->desc_vptr + pbuf->frm_ofs);
-
- frm->hdr_ofs = 0;
- frmlen = 0;
- frmlen += SHM_HDR_LEN + frm->hdr_ofs + skb->len;
-
- /* Add tail padding if needed. */
- if (frmlen % SHM_FRM_PAD_LEN)
- frmlen += SHM_FRM_PAD_LEN -
- (frmlen % SHM_FRM_PAD_LEN);
-
- /*
- * Verify that packet, header and additional padding
- * can fit within the buffer frame area.
- */
- if (frmlen >= (pbuf->len - pbuf->frm_ofs))
- break;
-
- if (!append) {
- list_del_init(&pbuf->list);
- append = 1;
- }
-
- skb = skb_dequeue(&pshm_drv->sk_qhead);
- if (skb == NULL)
- break;
- /* Copy in CAIF frame. */
- skb_copy_bits(skb, 0, pbuf->desc_vptr +
- pbuf->frm_ofs + SHM_HDR_LEN +
- frm->hdr_ofs, skb->len);
-
- pshm_drv->pshm_dev->pshm_netdev->stats.tx_packets++;
- pshm_drv->pshm_dev->pshm_netdev->stats.tx_bytes +=
- frmlen;
- dev_kfree_skb_irq(skb);
-
- /* Fill in the shared memory packet descriptor area. */
- pck_desc = (struct shm_pck_desc *) (pbuf->desc_vptr);
- /* Forward to current frame. */
- pck_desc += pbuf->frames;
- pck_desc->frm_ofs = (pbuf->phy_addr -
- pshm_drv->shm_base_addr) +
- pbuf->frm_ofs;
- pck_desc->frm_len = frmlen;
- /* Terminate packet descriptor area. */
- pck_desc++;
- pck_desc->frm_ofs = 0;
- /* Update buffer parameters. */
- pbuf->frames++;
- pbuf->frm_ofs += frmlen + (frmlen % 32);
-
- } while (pbuf->frames < SHM_MAX_FRMS_PER_BUF);
-
- /* Assign buffer as full. */
- list_add_tail(&pbuf->list, &pshm_drv->tx_full_list);
- append = 0;
- mbox_msg |= SHM_SET_FULL(pbuf->index);
-send_msg:
- spin_unlock_irqrestore(&pshm_drv->lock, flags);
-
- if (mbox_msg)
- pshm_drv->pshm_dev->pshmdev_mbxsend
- (pshm_drv->pshm_dev->shm_id, mbox_msg);
- } while (mbox_msg);
-}
-
-static int shm_netdev_tx(struct sk_buff *skb, struct net_device *shm_netdev)
-{
- struct shmdrv_layer *pshm_drv;
-
- pshm_drv = netdev_priv(shm_netdev);
-
- skb_queue_tail(&pshm_drv->sk_qhead, skb);
-
- /* Schedule Tx work queue. for deferred processing of skbs*/
- if (!work_pending(&pshm_drv->shm_tx_work))
- queue_work(pshm_drv->pshm_tx_workqueue, &pshm_drv->shm_tx_work);
-
- return 0;
-}
-
-static const struct net_device_ops netdev_ops = {
- .ndo_open = shm_netdev_open,
- .ndo_stop = shm_netdev_close,
- .ndo_start_xmit = shm_netdev_tx,
-};
-
-static void shm_netdev_setup(struct net_device *pshm_netdev)
-{
- struct shmdrv_layer *pshm_drv;
- pshm_netdev->netdev_ops = &netdev_ops;
-
- pshm_netdev->mtu = CAIF_MAX_MTU;
- pshm_netdev->type = ARPHRD_CAIF;
- pshm_netdev->hard_header_len = CAIF_NEEDED_HEADROOM;
- pshm_netdev->tx_queue_len = 0;
- pshm_netdev->destructor = free_netdev;
-
- pshm_drv = netdev_priv(pshm_netdev);
-
- /* Initialize structures in a clean state. */
- memset(pshm_drv, 0, sizeof(struct shmdrv_layer));
-
- pshm_drv->cfdev.link_select = CAIF_LINK_LOW_LATENCY;
-}
-
-int caif_shmcore_probe(struct shmdev_layer *pshm_dev)
-{
- int result, j;
- struct shmdrv_layer *pshm_drv = NULL;
-
- pshm_dev->pshm_netdev = alloc_netdev(sizeof(struct shmdrv_layer),
- "cfshm%d", shm_netdev_setup);
- if (!pshm_dev->pshm_netdev)
- return -ENOMEM;
-
- pshm_drv = netdev_priv(pshm_dev->pshm_netdev);
- pshm_drv->pshm_dev = pshm_dev;
-
- /*
- * Initialization starts with the verification of the
- * availability of MBX driver by calling its setup function.
- * MBX driver must be available by this time for proper
- * functioning of SHM driver.
- */
- if ((pshm_dev->pshmdev_mbxsetup
- (caif_shmdrv_rx_cb, pshm_dev, pshm_drv)) != 0) {
- pr_warn("Could not config. SHM Mailbox,"
- " Bailing out.....\n");
- free_netdev(pshm_dev->pshm_netdev);
- return -ENODEV;
- }
-
- skb_queue_head_init(&pshm_drv->sk_qhead);
-
- pr_info("SHM DEVICE[%d] PROBED BY DRIVER, NEW SHM DRIVER"
- " INSTANCE AT pshm_drv =0x%p\n",
- pshm_drv->pshm_dev->shm_id, pshm_drv);
-
- if (pshm_dev->shm_total_sz <
- (NR_TX_BUF * TX_BUF_SZ + NR_RX_BUF * RX_BUF_SZ)) {
-
- pr_warn("ERROR, Amount of available"
- " Phys. SHM cannot accommodate current SHM "
- "driver configuration, Bailing out ...\n");
- free_netdev(pshm_dev->pshm_netdev);
- return -ENOMEM;
- }
-
- pshm_drv->shm_base_addr = pshm_dev->shm_base_addr;
- pshm_drv->shm_tx_addr = pshm_drv->shm_base_addr;
-
- if (pshm_dev->shm_loopback)
- pshm_drv->shm_rx_addr = pshm_drv->shm_tx_addr;
- else
- pshm_drv->shm_rx_addr = pshm_dev->shm_base_addr +
- (NR_TX_BUF * TX_BUF_SZ);
-
- spin_lock_init(&pshm_drv->lock);
- INIT_LIST_HEAD(&pshm_drv->tx_empty_list);
- INIT_LIST_HEAD(&pshm_drv->tx_pend_list);
- INIT_LIST_HEAD(&pshm_drv->tx_full_list);
-
- INIT_LIST_HEAD(&pshm_drv->rx_empty_list);
- INIT_LIST_HEAD(&pshm_drv->rx_pend_list);
- INIT_LIST_HEAD(&pshm_drv->rx_full_list);
-
- INIT_WORK(&pshm_drv->shm_tx_work, shm_tx_work_func);
- INIT_WORK(&pshm_drv->shm_rx_work, shm_rx_work_func);
-
- pshm_drv->pshm_tx_workqueue =
- create_singlethread_workqueue("shm_tx_work");
- pshm_drv->pshm_rx_workqueue =
- create_singlethread_workqueue("shm_rx_work");
-
- for (j = 0; j < NR_TX_BUF; j++) {
- struct buf_list *tx_buf =
- kmalloc(sizeof(struct buf_list), GFP_KERNEL);
-
- if (tx_buf == NULL) {
- free_netdev(pshm_dev->pshm_netdev);
- return -ENOMEM;
- }
- tx_buf->index = j;
- tx_buf->phy_addr = pshm_drv->shm_tx_addr + (TX_BUF_SZ * j);
- tx_buf->len = TX_BUF_SZ;
- tx_buf->frames = 0;
- tx_buf->frm_ofs = SHM_CAIF_FRM_OFS;
-
- if (pshm_dev->shm_loopback)
- tx_buf->desc_vptr = (unsigned char *)tx_buf->phy_addr;
- else
- /*
- * FIXME: the result of ioremap is not a pointer - arnd
- */
- tx_buf->desc_vptr =
- ioremap(tx_buf->phy_addr, TX_BUF_SZ);
-
- list_add_tail(&tx_buf->list, &pshm_drv->tx_empty_list);
- }
-
- for (j = 0; j < NR_RX_BUF; j++) {
- struct buf_list *rx_buf =
- kmalloc(sizeof(struct buf_list), GFP_KERNEL);
-
- if (rx_buf == NULL) {
- free_netdev(pshm_dev->pshm_netdev);
- return -ENOMEM;
- }
- rx_buf->index = j;
- rx_buf->phy_addr = pshm_drv->shm_rx_addr + (RX_BUF_SZ * j);
- rx_buf->len = RX_BUF_SZ;
-
- if (pshm_dev->shm_loopback)
- rx_buf->desc_vptr = (unsigned char *)rx_buf->phy_addr;
- else
- rx_buf->desc_vptr =
- ioremap(rx_buf->phy_addr, RX_BUF_SZ);
- list_add_tail(&rx_buf->list, &pshm_drv->rx_empty_list);
- }
-
- pshm_drv->tx_empty_available = 1;
- result = register_netdev(pshm_dev->pshm_netdev);
- if (result)
- pr_warn("ERROR[%d], SHM could not, "
- "register with NW FRMWK Bailing out ...\n", result);
-
- return result;
-}
-
-void caif_shmcore_remove(struct net_device *pshm_netdev)
-{
- struct buf_list *pbuf;
- struct shmdrv_layer *pshm_drv = NULL;
-
- pshm_drv = netdev_priv(pshm_netdev);
-
- while (!(list_empty(&pshm_drv->tx_pend_list))) {
- pbuf =
- list_entry(pshm_drv->tx_pend_list.next,
- struct buf_list, list);
-
- list_del(&pbuf->list);
- kfree(pbuf);
- }
-
- while (!(list_empty(&pshm_drv->tx_full_list))) {
- pbuf =
- list_entry(pshm_drv->tx_full_list.next,
- struct buf_list, list);
- list_del(&pbuf->list);
- kfree(pbuf);
- }
-
- while (!(list_empty(&pshm_drv->tx_empty_list))) {
- pbuf =
- list_entry(pshm_drv->tx_empty_list.next,
- struct buf_list, list);
- list_del(&pbuf->list);
- kfree(pbuf);
- }
-
- while (!(list_empty(&pshm_drv->rx_full_list))) {
- pbuf =
- list_entry(pshm_drv->tx_full_list.next,
- struct buf_list, list);
- list_del(&pbuf->list);
- kfree(pbuf);
- }
-
- while (!(list_empty(&pshm_drv->rx_pend_list))) {
- pbuf =
- list_entry(pshm_drv->tx_pend_list.next,
- struct buf_list, list);
- list_del(&pbuf->list);
- kfree(pbuf);
- }
-
- while (!(list_empty(&pshm_drv->rx_empty_list))) {
- pbuf =
- list_entry(pshm_drv->rx_empty_list.next,
- struct buf_list, list);
- list_del(&pbuf->list);
- kfree(pbuf);
- }
-
- /* Destroy work queues. */
- destroy_workqueue(pshm_drv->pshm_tx_workqueue);
- destroy_workqueue(pshm_drv->pshm_rx_workqueue);
-
- unregister_netdev(pshm_netdev);
-}
diff --git a/drivers/net/ethernet/adi/bfin_mac.c b/drivers/net/ethernet/adi/bfin_mac.c
index a175d0b..ee70577 100644
--- a/drivers/net/ethernet/adi/bfin_mac.c
+++ b/drivers/net/ethernet/adi/bfin_mac.c
@@ -188,10 +188,9 @@
/* allocate a new skb for next time receive */
new_skb = netdev_alloc_skb(dev, PKT_BUF_SZ + NET_IP_ALIGN);
- if (!new_skb) {
- pr_notice("init: low on mem - packet dropped\n");
+ if (!new_skb)
goto init_error;
- }
+
skb_reserve(new_skb, NET_IP_ALIGN);
/* Invidate the data cache of skb->data range when it is write back
* cache. It will prevent overwritting the new data from DMA
@@ -1236,7 +1235,6 @@
new_skb = netdev_alloc_skb(dev, PKT_BUF_SZ + NET_IP_ALIGN);
if (!new_skb) {
- netdev_notice(dev, "rx: low on mem - packet dropped\n");
dev->stats.rx_dropped++;
goto out;
}
diff --git a/drivers/net/ethernet/aeroflex/greth.c b/drivers/net/ethernet/aeroflex/greth.c
index 0be2195..2692954 100644
--- a/drivers/net/ethernet/aeroflex/greth.c
+++ b/drivers/net/ethernet/aeroflex/greth.c
@@ -1464,35 +1464,23 @@
}
/* Allocate TX descriptor ring in coherent memory */
- greth->tx_bd_base = (struct greth_bd *) dma_alloc_coherent(greth->dev,
- 1024,
- &greth->tx_bd_base_phys,
- GFP_KERNEL);
-
+ greth->tx_bd_base = dma_alloc_coherent(greth->dev, 1024,
+ &greth->tx_bd_base_phys,
+ GFP_KERNEL | __GFP_ZERO);
if (!greth->tx_bd_base) {
- if (netif_msg_probe(greth))
- dev_err(&dev->dev, "could not allocate descriptor memory.\n");
err = -ENOMEM;
goto error3;
}
- memset(greth->tx_bd_base, 0, 1024);
-
/* Allocate RX descriptor ring in coherent memory */
- greth->rx_bd_base = (struct greth_bd *) dma_alloc_coherent(greth->dev,
- 1024,
- &greth->rx_bd_base_phys,
- GFP_KERNEL);
-
+ greth->rx_bd_base = dma_alloc_coherent(greth->dev, 1024,
+ &greth->rx_bd_base_phys,
+ GFP_KERNEL | __GFP_ZERO);
if (!greth->rx_bd_base) {
- if (netif_msg_probe(greth))
- dev_err(greth->dev, "could not allocate descriptor memory.\n");
err = -ENOMEM;
goto error4;
}
- memset(greth->rx_bd_base, 0, 1024);
-
/* Get MAC address from: module param, OF property or ID prom */
for (i = 0; i < 6; i++) {
if (macaddr[i] != 0)
diff --git a/drivers/net/ethernet/amd/7990.c b/drivers/net/ethernet/amd/7990.c
index 6e722dc..65926a9 100644
--- a/drivers/net/ethernet/amd/7990.c
+++ b/drivers/net/ethernet/amd/7990.c
@@ -318,8 +318,6 @@
struct sk_buff *skb = netdev_alloc_skb(dev, len + 2);
if (!skb) {
- printk ("%s: Memory squeeze, deferring packet.\n",
- dev->name);
dev->stats.rx_dropped++;
rd->mblength = 0;
rd->rmd1_bits = LE_R1_OWN;
diff --git a/drivers/net/ethernet/amd/a2065.c b/drivers/net/ethernet/amd/a2065.c
index 3789aff..0866e76 100644
--- a/drivers/net/ethernet/amd/a2065.c
+++ b/drivers/net/ethernet/amd/a2065.c
@@ -293,7 +293,6 @@
struct sk_buff *skb = netdev_alloc_skb(dev, len + 2);
if (!skb) {
- netdev_warn(dev, "Memory squeeze, deferring packet\n");
dev->stats.rx_dropped++;
rd->mblength = 0;
rd->rmd1_bits = LE_R1_OWN;
diff --git a/drivers/net/ethernet/amd/am79c961a.c b/drivers/net/ethernet/amd/am79c961a.c
index 60e2b70..9793767 100644
--- a/drivers/net/ethernet/amd/am79c961a.c
+++ b/drivers/net/ethernet/amd/am79c961a.c
@@ -528,7 +528,6 @@
dev->stats.rx_packets++;
} else {
am_writeword (dev, hdraddr + 2, RMD_OWN);
- printk (KERN_WARNING "%s: memory squeeze, dropping packet.\n", dev->name);
dev->stats.rx_dropped++;
break;
}
diff --git a/drivers/net/ethernet/amd/ariadne.c b/drivers/net/ethernet/amd/ariadne.c
index 98f4522..c178eb4 100644
--- a/drivers/net/ethernet/amd/ariadne.c
+++ b/drivers/net/ethernet/amd/ariadne.c
@@ -193,7 +193,6 @@
skb = netdev_alloc_skb(dev, pkt_len + 2);
if (skb == NULL) {
- netdev_warn(dev, "Memory squeeze, deferring packet\n");
for (i = 0; i < RX_RING_SIZE; i++)
if (lowb(priv->rx_ring[(entry + i) % RX_RING_SIZE]->RMD1) & RF_OWN)
break;
diff --git a/drivers/net/ethernet/amd/atarilance.c b/drivers/net/ethernet/amd/atarilance.c
index 84219df..e8d0ef5 100644
--- a/drivers/net/ethernet/amd/atarilance.c
+++ b/drivers/net/ethernet/amd/atarilance.c
@@ -996,8 +996,6 @@
else {
skb = netdev_alloc_skb(dev, pkt_len + 2);
if (skb == NULL) {
- DPRINTK( 1, ( "%s: Memory squeeze, deferring packet.\n",
- dev->name ));
for( i = 0; i < RX_RING_SIZE; i++ )
if (MEM->rx_head[(entry+i) & RX_RING_MOD_MASK].flag &
RMD1_OWN_CHIP)
@@ -1149,9 +1147,7 @@
static int __init atarilance_module_init(void)
{
atarilance_dev = atarilance_probe(-1);
- if (IS_ERR(atarilance_dev))
- return PTR_ERR(atarilance_dev);
- return 0;
+ return PTR_RET(atarilance_dev);
}
static void __exit atarilance_module_exit(void)
diff --git a/drivers/net/ethernet/amd/au1000_eth.c b/drivers/net/ethernet/amd/au1000_eth.c
index de774d4..688aede 100644
--- a/drivers/net/ethernet/amd/au1000_eth.c
+++ b/drivers/net/ethernet/amd/au1000_eth.c
@@ -727,7 +727,6 @@
frmlen -= 4; /* Remove FCS */
skb = netdev_alloc_skb(dev, frmlen + 2);
if (skb == NULL) {
- netdev_err(dev, "Memory squeeze, dropping packet.\n");
dev->stats.rx_dropped++;
continue;
}
diff --git a/drivers/net/ethernet/amd/declance.c b/drivers/net/ethernet/amd/declance.c
index baca0bd..3d86ffe 100644
--- a/drivers/net/ethernet/amd/declance.c
+++ b/drivers/net/ethernet/amd/declance.c
@@ -607,8 +607,6 @@
skb = netdev_alloc_skb(dev, len + 2);
if (skb == 0) {
- printk("%s: Memory squeeze, deferring packet.\n",
- dev->name);
dev->stats.rx_dropped++;
*rds_ptr(rd, mblength, lp->type) = 0;
*rds_ptr(rd, rmd1, lp->type) =
diff --git a/drivers/net/ethernet/amd/mvme147.c b/drivers/net/ethernet/amd/mvme147.c
index 9af3c30..a51497c 100644
--- a/drivers/net/ethernet/amd/mvme147.c
+++ b/drivers/net/ethernet/amd/mvme147.c
@@ -188,9 +188,7 @@
int __init init_module(void)
{
dev_mvme147_lance = mvme147lance_probe(-1);
- if (IS_ERR(dev_mvme147_lance))
- return PTR_ERR(dev_mvme147_lance);
- return 0;
+ return PTR_RET(dev_mvme147_lance);
}
void __exit cleanup_module(void)
diff --git a/drivers/net/ethernet/amd/ni65.c b/drivers/net/ethernet/amd/ni65.c
index 013b651..26fc0ce 100644
--- a/drivers/net/ethernet/amd/ni65.c
+++ b/drivers/net/ethernet/amd/ni65.c
@@ -1238,7 +1238,7 @@
int __init init_module(void)
{
dev_ni65 = ni65_probe(-1);
- return IS_ERR(dev_ni65) ? PTR_ERR(dev_ni65) : 0;
+ return PTR_RET(dev_ni65);
}
void __exit cleanup_module(void)
diff --git a/drivers/net/ethernet/amd/pcnet32.c b/drivers/net/ethernet/amd/pcnet32.c
index 797f847..ed21307 100644
--- a/drivers/net/ethernet/amd/pcnet32.c
+++ b/drivers/net/ethernet/amd/pcnet32.c
@@ -1166,7 +1166,6 @@
skb = netdev_alloc_skb(dev, pkt_len + NET_IP_ALIGN);
if (skb == NULL) {
- netif_err(lp, drv, dev, "Memory squeeze, dropping packet\n");
dev->stats.rx_dropped++;
return;
}
diff --git a/drivers/net/ethernet/amd/sun3lance.c b/drivers/net/ethernet/amd/sun3lance.c
index 74b3891b..4375abe 100644
--- a/drivers/net/ethernet/amd/sun3lance.c
+++ b/drivers/net/ethernet/amd/sun3lance.c
@@ -812,9 +812,6 @@
else {
skb = netdev_alloc_skb(dev, pkt_len + 2);
if (skb == NULL) {
- DPRINTK( 1, ( "%s: Memory squeeze, deferring packet.\n",
- dev->name ));
-
dev->stats.rx_dropped++;
head->msg_length = 0;
head->flag |= RMD1_OWN_CHIP;
@@ -943,9 +940,7 @@
int __init init_module(void)
{
sun3lance_dev = sun3lance_probe(-1);
- if (IS_ERR(sun3lance_dev))
- return PTR_ERR(sun3lance_dev);
- return 0;
+ return PTR_RET(sun3lance_dev);
}
void __exit cleanup_module(void)
diff --git a/drivers/net/ethernet/amd/sunlance.c b/drivers/net/ethernet/amd/sunlance.c
index 6a40290..f47b780 100644
--- a/drivers/net/ethernet/amd/sunlance.c
+++ b/drivers/net/ethernet/amd/sunlance.c
@@ -536,8 +536,6 @@
skb = netdev_alloc_skb(dev, len + 2);
if (skb == NULL) {
- printk(KERN_INFO "%s: Memory squeeze, deferring packet.\n",
- dev->name);
dev->stats.rx_dropped++;
rd->mblength = 0;
rd->rmd1_bits = LE_R1_OWN;
@@ -708,8 +706,6 @@
skb = netdev_alloc_skb(dev, len + 2);
if (skb == NULL) {
- printk(KERN_INFO "%s: Memory squeeze, deferring packet.\n",
- dev->name);
dev->stats.rx_dropped++;
sbus_writew(0, &rd->mblength);
sbus_writeb(LE_R1_OWN, &rd->rmd1_bits);
@@ -1377,10 +1373,9 @@
dma_alloc_coherent(&op->dev,
sizeof(struct lance_init_block),
&lp->init_block_dvma, GFP_ATOMIC);
- if (!lp->init_block_mem) {
- printk(KERN_ERR "SunLance: Cannot allocate consistent DMA memory.\n");
+ if (!lp->init_block_mem)
goto fail;
- }
+
lp->pio_buffer = 0;
lp->init_ring = lance_init_ring_dvma;
lp->rx = lance_rx_dvma;
diff --git a/drivers/net/ethernet/apple/macmace.c b/drivers/net/ethernet/apple/macmace.c
index a206779..4ce8ceb 100644
--- a/drivers/net/ethernet/apple/macmace.c
+++ b/drivers/net/ethernet/apple/macmace.c
@@ -386,20 +386,16 @@
/* Allocate the DMA ring buffers */
mp->tx_ring = dma_alloc_coherent(mp->device,
- N_TX_RING * MACE_BUFF_SIZE,
- &mp->tx_ring_phys, GFP_KERNEL);
- if (mp->tx_ring == NULL) {
- printk(KERN_ERR "%s: unable to allocate DMA tx buffers\n", dev->name);
+ N_TX_RING * MACE_BUFF_SIZE,
+ &mp->tx_ring_phys, GFP_KERNEL);
+ if (mp->tx_ring == NULL)
goto out1;
- }
mp->rx_ring = dma_alloc_coherent(mp->device,
- N_RX_RING * MACE_BUFF_SIZE,
- &mp->rx_ring_phys, GFP_KERNEL);
- if (mp->rx_ring == NULL) {
- printk(KERN_ERR "%s: unable to allocate DMA rx buffers\n", dev->name);
+ N_RX_RING * MACE_BUFF_SIZE,
+ &mp->rx_ring_phys, GFP_KERNEL);
+ if (mp->rx_ring == NULL)
goto out2;
- }
mace_dma_off(dev);
diff --git a/drivers/net/ethernet/atheros/atl1e/atl1e_main.c b/drivers/net/ethernet/atheros/atl1e/atl1e_main.c
index 92f4734..e1f1b2a 100644
--- a/drivers/net/ethernet/atheros/atl1e/atl1e_main.c
+++ b/drivers/net/ethernet/atheros/atl1e/atl1e_main.c
@@ -1420,11 +1420,9 @@
packet_size = ((prrs->word1 >> RRS_PKT_SIZE_SHIFT) &
RRS_PKT_SIZE_MASK) - 4; /* CRC */
skb = netdev_alloc_skb_ip_align(netdev, packet_size);
- if (skb == NULL) {
- netdev_warn(netdev,
- "Memory squeeze, deferring packet\n");
+ if (skb == NULL)
goto skip_pkt;
- }
+
memcpy(skb->data, (u8 *)(prrs + 1), packet_size);
skb_put(skb, packet_size);
skb->protocol = eth_type_trans(skb, netdev);
diff --git a/drivers/net/ethernet/atheros/atlx/atl2.c b/drivers/net/ethernet/atheros/atlx/atl2.c
index 1278b47..a046b6f 100644
--- a/drivers/net/ethernet/atheros/atlx/atl2.c
+++ b/drivers/net/ethernet/atheros/atlx/atl2.c
@@ -437,9 +437,6 @@
/* alloc new buffer */
skb = netdev_alloc_skb_ip_align(netdev, rx_size);
if (NULL == skb) {
- printk(KERN_WARNING
- "%s: Mem squeeze, deferring packet.\n",
- netdev->name);
/*
* Check that some rx space is free. If not,
* free one and mark stats->rx_dropped++.
diff --git a/drivers/net/ethernet/broadcom/bcm63xx_enet.c b/drivers/net/ethernet/broadcom/bcm63xx_enet.c
index 7d81e05..0b3e23e 100644
--- a/drivers/net/ethernet/broadcom/bcm63xx_enet.c
+++ b/drivers/net/ethernet/broadcom/bcm63xx_enet.c
@@ -862,27 +862,25 @@
/* allocate rx dma ring */
size = priv->rx_ring_size * sizeof(struct bcm_enet_desc);
- p = dma_alloc_coherent(kdev, size, &priv->rx_desc_dma, GFP_KERNEL);
+ p = dma_alloc_coherent(kdev, size, &priv->rx_desc_dma,
+ GFP_KERNEL | __GFP_ZERO);
if (!p) {
- dev_err(kdev, "cannot allocate rx ring %u\n", size);
ret = -ENOMEM;
goto out_freeirq_tx;
}
- memset(p, 0, size);
priv->rx_desc_alloc_size = size;
priv->rx_desc_cpu = p;
/* allocate tx dma ring */
size = priv->tx_ring_size * sizeof(struct bcm_enet_desc);
- p = dma_alloc_coherent(kdev, size, &priv->tx_desc_dma, GFP_KERNEL);
+ p = dma_alloc_coherent(kdev, size, &priv->tx_desc_dma,
+ GFP_KERNEL | __GFP_ZERO);
if (!p) {
- dev_err(kdev, "cannot allocate tx ring\n");
ret = -ENOMEM;
goto out_free_rx_ring;
}
- memset(p, 0, size);
priv->tx_desc_alloc_size = size;
priv->tx_desc_cpu = p;
@@ -1619,7 +1617,6 @@
struct resource *res_mem, *res_irq, *res_irq_rx, *res_irq_tx;
struct mii_bus *bus;
const char *clk_name;
- unsigned int iomem_size;
int i, ret;
/* stop if shared driver failed, assume driver->probe will be
@@ -1644,17 +1641,12 @@
if (ret)
goto out;
- iomem_size = resource_size(res_mem);
- if (!request_mem_region(res_mem->start, iomem_size, "bcm63xx_enet")) {
- ret = -EBUSY;
+ priv->base = devm_request_and_ioremap(&pdev->dev, res_mem);
+ if (priv->base == NULL) {
+ ret = -ENOMEM;
goto out;
}
- priv->base = ioremap(res_mem->start, iomem_size);
- if (priv->base == NULL) {
- ret = -ENOMEM;
- goto out_release_mem;
- }
dev->irq = priv->irq = res_irq->start;
priv->irq_rx = res_irq_rx->start;
priv->irq_tx = res_irq_tx->start;
@@ -1674,9 +1666,9 @@
priv->mac_clk = clk_get(&pdev->dev, clk_name);
if (IS_ERR(priv->mac_clk)) {
ret = PTR_ERR(priv->mac_clk);
- goto out_unmap;
+ goto out;
}
- clk_enable(priv->mac_clk);
+ clk_prepare_enable(priv->mac_clk);
/* initialize default and fetch platform data */
priv->rx_ring_size = BCMENET_DEF_RX_DESC;
@@ -1705,7 +1697,7 @@
priv->phy_clk = NULL;
goto out_put_clk_mac;
}
- clk_enable(priv->phy_clk);
+ clk_prepare_enable(priv->phy_clk);
}
/* do minimal hardware init to be able to probe mii bus */
@@ -1733,7 +1725,8 @@
* if a slave is not present on hw */
bus->phy_mask = ~(1 << priv->phy_id);
- bus->irq = kmalloc(sizeof(int) * PHY_MAX_ADDR, GFP_KERNEL);
+ bus->irq = devm_kzalloc(&pdev->dev, sizeof(int) * PHY_MAX_ADDR,
+ GFP_KERNEL);
if (!bus->irq) {
ret = -ENOMEM;
goto out_free_mdio;
@@ -1794,10 +1787,8 @@
return 0;
out_unregister_mdio:
- if (priv->mii_bus) {
+ if (priv->mii_bus)
mdiobus_unregister(priv->mii_bus);
- kfree(priv->mii_bus->irq);
- }
out_free_mdio:
if (priv->mii_bus)
@@ -1807,19 +1798,13 @@
/* turn off mdc clock */
enet_writel(priv, 0, ENET_MIISC_REG);
if (priv->phy_clk) {
- clk_disable(priv->phy_clk);
+ clk_disable_unprepare(priv->phy_clk);
clk_put(priv->phy_clk);
}
out_put_clk_mac:
- clk_disable(priv->mac_clk);
+ clk_disable_unprepare(priv->mac_clk);
clk_put(priv->mac_clk);
-
-out_unmap:
- iounmap(priv->base);
-
-out_release_mem:
- release_mem_region(res_mem->start, iomem_size);
out:
free_netdev(dev);
return ret;
@@ -1833,7 +1818,6 @@
{
struct bcm_enet_priv *priv;
struct net_device *dev;
- struct resource *res;
/* stop netdevice */
dev = platform_get_drvdata(pdev);
@@ -1845,7 +1829,6 @@
if (priv->has_phy) {
mdiobus_unregister(priv->mii_bus);
- kfree(priv->mii_bus->irq);
mdiobus_free(priv->mii_bus);
} else {
struct bcm63xx_enet_platform_data *pd;
@@ -1856,17 +1839,12 @@
bcm_enet_mdio_write_mii);
}
- /* release device resources */
- iounmap(priv->base);
- res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
- release_mem_region(res->start, resource_size(res));
-
/* disable hw block clocks */
if (priv->phy_clk) {
- clk_disable(priv->phy_clk);
+ clk_disable_unprepare(priv->phy_clk);
clk_put(priv->phy_clk);
}
- clk_disable(priv->mac_clk);
+ clk_disable_unprepare(priv->mac_clk);
clk_put(priv->mac_clk);
platform_set_drvdata(pdev, NULL);
@@ -1889,31 +1867,20 @@
static int bcm_enet_shared_probe(struct platform_device *pdev)
{
struct resource *res;
- unsigned int iomem_size;
res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
if (!res)
return -ENODEV;
- iomem_size = resource_size(res);
- if (!request_mem_region(res->start, iomem_size, "bcm63xx_enet_dma"))
- return -EBUSY;
-
- bcm_enet_shared_base = ioremap(res->start, iomem_size);
- if (!bcm_enet_shared_base) {
- release_mem_region(res->start, iomem_size);
+ bcm_enet_shared_base = devm_request_and_ioremap(&pdev->dev, res);
+ if (!bcm_enet_shared_base)
return -ENOMEM;
- }
+
return 0;
}
static int bcm_enet_shared_remove(struct platform_device *pdev)
{
- struct resource *res;
-
- iounmap(bcm_enet_shared_base);
- res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
- release_mem_region(res->start, resource_size(res));
return 0;
}
diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c
index da5f439..eec0af4 100644
--- a/drivers/net/ethernet/broadcom/bgmac.c
+++ b/drivers/net/ethernet/broadcom/bgmac.c
@@ -13,6 +13,7 @@
#include <linux/delay.h>
#include <linux/etherdevice.h>
#include <linux/mii.h>
+#include <linux/phy.h>
#include <linux/interrupt.h>
#include <linux/dma-mapping.h>
#include <bcm47xx_nvram.h>
@@ -244,10 +245,8 @@
/* Alloc skb */
slot->skb = netdev_alloc_skb(bgmac->net_dev, BGMAC_RX_BUF_SIZE);
- if (!slot->skb) {
- bgmac_err(bgmac, "Allocation of skb failed!\n");
+ if (!slot->skb)
return -ENOMEM;
- }
/* Poison - if everything goes fine, hardware will overwrite it */
rx = (struct bgmac_rx_header *)slot->skb->data;
@@ -1313,6 +1312,73 @@
};
/**************************************************
+ * MII
+ **************************************************/
+
+static int bgmac_mii_read(struct mii_bus *bus, int mii_id, int regnum)
+{
+ return bgmac_phy_read(bus->priv, mii_id, regnum);
+}
+
+static int bgmac_mii_write(struct mii_bus *bus, int mii_id, int regnum,
+ u16 value)
+{
+ return bgmac_phy_write(bus->priv, mii_id, regnum, value);
+}
+
+static int bgmac_mii_register(struct bgmac *bgmac)
+{
+ struct mii_bus *mii_bus;
+ int i, err = 0;
+
+ mii_bus = mdiobus_alloc();
+ if (!mii_bus)
+ return -ENOMEM;
+
+ mii_bus->name = "bgmac mii bus";
+ sprintf(mii_bus->id, "%s-%d-%d", "bgmac", bgmac->core->bus->num,
+ bgmac->core->core_unit);
+ mii_bus->priv = bgmac;
+ mii_bus->read = bgmac_mii_read;
+ mii_bus->write = bgmac_mii_write;
+ mii_bus->parent = &bgmac->core->dev;
+ mii_bus->phy_mask = ~(1 << bgmac->phyaddr);
+
+ mii_bus->irq = kmalloc_array(PHY_MAX_ADDR, sizeof(int), GFP_KERNEL);
+ if (!mii_bus->irq) {
+ err = -ENOMEM;
+ goto err_free_bus;
+ }
+ for (i = 0; i < PHY_MAX_ADDR; i++)
+ mii_bus->irq[i] = PHY_POLL;
+
+ err = mdiobus_register(mii_bus);
+ if (err) {
+ bgmac_err(bgmac, "Registration of mii bus failed\n");
+ goto err_free_irq;
+ }
+
+ bgmac->mii_bus = mii_bus;
+
+ return err;
+
+err_free_irq:
+ kfree(mii_bus->irq);
+err_free_bus:
+ mdiobus_free(mii_bus);
+ return err;
+}
+
+static void bgmac_mii_unregister(struct bgmac *bgmac)
+{
+ struct mii_bus *mii_bus = bgmac->mii_bus;
+
+ mdiobus_unregister(mii_bus);
+ kfree(mii_bus->irq);
+ mdiobus_free(mii_bus);
+}
+
+/**************************************************
* BCMA bus ops
**************************************************/
@@ -1404,11 +1470,18 @@
if (core->bus->sprom.boardflags_lo & BGMAC_BFL_ENETADM)
bgmac_warn(bgmac, "Support for ADMtek ethernet switch not implemented\n");
+ err = bgmac_mii_register(bgmac);
+ if (err) {
+ bgmac_err(bgmac, "Cannot register MDIO\n");
+ err = -ENOTSUPP;
+ goto err_dma_free;
+ }
+
err = register_netdev(bgmac->net_dev);
if (err) {
bgmac_err(bgmac, "Cannot register net device\n");
err = -ENOTSUPP;
- goto err_dma_free;
+ goto err_mii_unregister;
}
netif_carrier_off(net_dev);
@@ -1417,6 +1490,8 @@
return 0;
+err_mii_unregister:
+ bgmac_mii_unregister(bgmac);
err_dma_free:
bgmac_dma_free(bgmac);
@@ -1433,6 +1508,7 @@
netif_napi_del(&bgmac->napi);
unregister_netdev(bgmac->net_dev);
+ bgmac_mii_unregister(bgmac);
bgmac_dma_free(bgmac);
bcma_set_drvdata(core, NULL);
free_netdev(bgmac->net_dev);
diff --git a/drivers/net/ethernet/broadcom/bgmac.h b/drivers/net/ethernet/broadcom/bgmac.h
index 4ede614..98d4b5f 100644
--- a/drivers/net/ethernet/broadcom/bgmac.h
+++ b/drivers/net/ethernet/broadcom/bgmac.h
@@ -399,6 +399,7 @@
struct bcma_device *cmn; /* Reference to CMN core for BCM4706 */
struct net_device *net_dev;
struct napi_struct napi;
+ struct mii_bus *mii_bus;
/* DMA */
struct bgmac_dma_ring tx_ring[BGMAC_MAX_TX_RINGS];
diff --git a/drivers/net/ethernet/broadcom/bnx2.c b/drivers/net/ethernet/broadcom/bnx2.c
index 2f0ba8f..e709296 100644
--- a/drivers/net/ethernet/broadcom/bnx2.c
+++ b/drivers/net/ethernet/broadcom/bnx2.c
@@ -854,12 +854,11 @@
sizeof(struct statistics_block);
status_blk = dma_alloc_coherent(&bp->pdev->dev, bp->status_stats_size,
- &bp->status_blk_mapping, GFP_KERNEL);
+ &bp->status_blk_mapping,
+ GFP_KERNEL | __GFP_ZERO);
if (status_blk == NULL)
goto alloc_mem_err;
- memset(status_blk, 0, bp->status_stats_size);
-
bnapi = &bp->bnx2_napi[0];
bnapi->status_blk.msi = status_blk;
bnapi->hw_tx_cons_ptr =
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
index e4605a9..a4729c7 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
@@ -492,7 +492,6 @@
struct bnx2x_fastpath {
struct bnx2x *bp; /* parent */
-#define BNX2X_NAPI_WEIGHT 128
struct napi_struct napi;
union host_hc_status_block status_blk;
/* chip independed shortcuts into sb structure */
@@ -613,9 +612,10 @@
* START_BD - describes packed
* START_BD(splitted) - includes unpaged data segment for GSO
* PARSING_BD - for TSO and CSUM data
+ * PARSING_BD2 - for encapsulation data
* Frag BDs - decribes pages for frags
*/
-#define BDS_PER_TX_PKT 3
+#define BDS_PER_TX_PKT 4
#define MAX_BDS_PER_TX_PKT (MAX_SKB_FRAGS + BDS_PER_TX_PKT)
/* max BDs per tx packet including next pages */
#define MAX_DESC_PER_TX_PKT (MAX_BDS_PER_TX_PKT + \
@@ -730,18 +730,24 @@
#define SKB_CS(skb) (*(u16 *)(skb_transport_header(skb) + \
skb->csum_offset))
-#define pbd_tcp_flags(skb) (ntohl(tcp_flag_word(tcp_hdr(skb)))>>16 & 0xff)
+#define pbd_tcp_flags(tcp_hdr) (ntohl(tcp_flag_word(tcp_hdr))>>16 & 0xff)
-#define XMIT_PLAIN 0
-#define XMIT_CSUM_V4 0x1
-#define XMIT_CSUM_V6 0x2
-#define XMIT_CSUM_TCP 0x4
-#define XMIT_GSO_V4 0x8
-#define XMIT_GSO_V6 0x10
+#define XMIT_PLAIN 0
+#define XMIT_CSUM_V4 (1 << 0)
+#define XMIT_CSUM_V6 (1 << 1)
+#define XMIT_CSUM_TCP (1 << 2)
+#define XMIT_GSO_V4 (1 << 3)
+#define XMIT_GSO_V6 (1 << 4)
+#define XMIT_CSUM_ENC_V4 (1 << 5)
+#define XMIT_CSUM_ENC_V6 (1 << 6)
+#define XMIT_GSO_ENC_V4 (1 << 7)
+#define XMIT_GSO_ENC_V6 (1 << 8)
-#define XMIT_CSUM (XMIT_CSUM_V4 | XMIT_CSUM_V6)
-#define XMIT_GSO (XMIT_GSO_V4 | XMIT_GSO_V6)
+#define XMIT_CSUM_ENC (XMIT_CSUM_ENC_V4 | XMIT_CSUM_ENC_V6)
+#define XMIT_GSO_ENC (XMIT_GSO_ENC_V4 | XMIT_GSO_ENC_V6)
+#define XMIT_CSUM (XMIT_CSUM_V4 | XMIT_CSUM_V6 | XMIT_CSUM_ENC)
+#define XMIT_GSO (XMIT_GSO_V4 | XMIT_GSO_V6 | XMIT_GSO_ENC)
/* stuff added to make the code fit 80Col */
#define CQE_TYPE(cqe_fp_flags) ((cqe_fp_flags) & ETH_FAST_PATH_RX_CQE_TYPE)
@@ -1215,6 +1221,7 @@
BNX2X_SP_RTNL_ENABLE_SRIOV,
BNX2X_SP_RTNL_VFPF_MCAST,
BNX2X_SP_RTNL_VFPF_STORM_RX_MODE,
+ BNX2X_SP_RTNL_HYPERVISOR_VLAN,
};
@@ -1281,6 +1288,8 @@
dma_addr_t pf2vf_bulletin_mapping;
struct pf_vf_bulletin_content old_bulletin;
+
+ u16 requested_nr_virtfn;
#endif /* CONFIG_BNX2X_SRIOV */
struct net_device *dev;
@@ -1944,12 +1953,9 @@
void bnx2x_igu_clear_sb_gen(struct bnx2x *bp, u8 func, u8 idu_sb_id,
bool is_pf);
-#define BNX2X_ILT_ZALLOC(x, y, size) \
- do { \
- x = dma_alloc_coherent(&bp->pdev->dev, size, y, GFP_KERNEL); \
- if (x) \
- memset(x, 0, size); \
- } while (0)
+#define BNX2X_ILT_ZALLOC(x, y, size) \
+ x = dma_alloc_coherent(&bp->pdev->dev, size, y, \
+ GFP_KERNEL | __GFP_ZERO)
#define BNX2X_ILT_FREE(x, y, size) \
do { \
@@ -2286,7 +2292,7 @@
DMAE_REG_GO_C12, DMAE_REG_GO_C13, DMAE_REG_GO_C14, DMAE_REG_GO_C15
};
-void bnx2x_set_ethtool_ops(struct net_device *netdev);
+void bnx2x_set_ethtool_ops(struct bnx2x *bp, struct net_device *netdev);
void bnx2x_notify_link_changed(struct bnx2x *bp);
#define BNX2X_MF_SD_PROTOCOL(bp) \
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
index 4046f97..db6912b 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
@@ -451,7 +451,8 @@
* Compute number of aggregated segments, and gso_type.
*/
static void bnx2x_set_gro_params(struct sk_buff *skb, u16 parsing_flags,
- u16 len_on_bd, unsigned int pkt_len)
+ u16 len_on_bd, unsigned int pkt_len,
+ u16 num_of_coalesced_segs)
{
/* TPA aggregation won't have either IP options or TCP options
* other than timestamp or IPv6 extension headers.
@@ -480,8 +481,7 @@
/* tcp_gro_complete() will copy NAPI_GRO_CB(skb)->count
* to skb_shinfo(skb)->gso_segs
*/
- NAPI_GRO_CB(skb)->count = DIV_ROUND_UP(pkt_len - hdrs_len,
- skb_shinfo(skb)->gso_size);
+ NAPI_GRO_CB(skb)->count = num_of_coalesced_segs;
}
static int bnx2x_alloc_rx_sge(struct bnx2x *bp,
@@ -537,7 +537,8 @@
/* This is needed in order to enable forwarding support */
if (frag_size)
bnx2x_set_gro_params(skb, tpa_info->parsing_flags, len_on_bd,
- le16_to_cpu(cqe->pkt_len));
+ le16_to_cpu(cqe->pkt_len),
+ le16_to_cpu(cqe->num_of_coalesced_segs));
#ifdef BNX2X_STOP_ON_ERROR
if (pages > min_t(u32, 8, MAX_SKB_FRAGS) * SGE_PAGES) {
@@ -3086,11 +3087,11 @@
* to ease the pain of our fellow microcode engineers
* we use one mapping for both BDs
*/
-static noinline u16 bnx2x_tx_split(struct bnx2x *bp,
- struct bnx2x_fp_txdata *txdata,
- struct sw_tx_bd *tx_buf,
- struct eth_tx_start_bd **tx_bd, u16 hlen,
- u16 bd_prod, int nbd)
+static u16 bnx2x_tx_split(struct bnx2x *bp,
+ struct bnx2x_fp_txdata *txdata,
+ struct sw_tx_bd *tx_buf,
+ struct eth_tx_start_bd **tx_bd, u16 hlen,
+ u16 bd_prod)
{
struct eth_tx_start_bd *h_tx_bd = *tx_bd;
struct eth_tx_bd *d_tx_bd;
@@ -3098,11 +3099,10 @@
int old_len = le16_to_cpu(h_tx_bd->nbytes);
/* first fix first BD */
- h_tx_bd->nbd = cpu_to_le16(nbd);
h_tx_bd->nbytes = cpu_to_le16(hlen);
- DP(NETIF_MSG_TX_QUEUED, "TSO split header size is %d (%x:%x) nbd %d\n",
- h_tx_bd->nbytes, h_tx_bd->addr_hi, h_tx_bd->addr_lo, h_tx_bd->nbd);
+ DP(NETIF_MSG_TX_QUEUED, "TSO split header size is %d (%x:%x)\n",
+ h_tx_bd->nbytes, h_tx_bd->addr_hi, h_tx_bd->addr_lo);
/* now get a new data BD
* (after the pbd) and fill it */
@@ -3131,7 +3131,7 @@
#define bswab32(b32) ((__force __le32) swab32((__force __u32) (b32)))
#define bswab16(b16) ((__force __le16) swab16((__force __u16) (b16)))
-static inline __le16 bnx2x_csum_fix(unsigned char *t_header, u16 csum, s8 fix)
+static __le16 bnx2x_csum_fix(unsigned char *t_header, u16 csum, s8 fix)
{
__sum16 tsum = (__force __sum16) csum;
@@ -3146,30 +3146,47 @@
return bswab16(tsum);
}
-static inline u32 bnx2x_xmit_type(struct bnx2x *bp, struct sk_buff *skb)
+static u32 bnx2x_xmit_type(struct bnx2x *bp, struct sk_buff *skb)
{
u32 rc;
+ __u8 prot = 0;
+ __be16 protocol;
if (skb->ip_summed != CHECKSUM_PARTIAL)
- rc = XMIT_PLAIN;
+ return XMIT_PLAIN;
- else {
- if (vlan_get_protocol(skb) == htons(ETH_P_IPV6)) {
- rc = XMIT_CSUM_V6;
- if (ipv6_hdr(skb)->nexthdr == IPPROTO_TCP)
+ protocol = vlan_get_protocol(skb);
+ if (protocol == htons(ETH_P_IPV6)) {
+ rc = XMIT_CSUM_V6;
+ prot = ipv6_hdr(skb)->nexthdr;
+ } else {
+ rc = XMIT_CSUM_V4;
+ prot = ip_hdr(skb)->protocol;
+ }
+
+ if (!CHIP_IS_E1x(bp) && skb->encapsulation) {
+ if (inner_ip_hdr(skb)->version == 6) {
+ rc |= XMIT_CSUM_ENC_V6;
+ if (inner_ipv6_hdr(skb)->nexthdr == IPPROTO_TCP)
rc |= XMIT_CSUM_TCP;
-
} else {
- rc = XMIT_CSUM_V4;
- if (ip_hdr(skb)->protocol == IPPROTO_TCP)
+ rc |= XMIT_CSUM_ENC_V4;
+ if (inner_ip_hdr(skb)->protocol == IPPROTO_TCP)
rc |= XMIT_CSUM_TCP;
}
}
+ if (prot == IPPROTO_TCP)
+ rc |= XMIT_CSUM_TCP;
- if (skb_is_gso_v6(skb))
- rc |= XMIT_GSO_V6 | XMIT_CSUM_TCP | XMIT_CSUM_V6;
- else if (skb_is_gso(skb))
- rc |= XMIT_GSO_V4 | XMIT_CSUM_V4 | XMIT_CSUM_TCP;
+ if (skb_is_gso_v6(skb)) {
+ rc |= (XMIT_GSO_V6 | XMIT_CSUM_TCP | XMIT_CSUM_V6);
+ if (rc & XMIT_CSUM_ENC)
+ rc |= XMIT_GSO_ENC_V6;
+ } else if (skb_is_gso(skb)) {
+ rc |= (XMIT_GSO_V4 | XMIT_CSUM_V4 | XMIT_CSUM_TCP);
+ if (rc & XMIT_CSUM_ENC)
+ rc |= XMIT_GSO_ENC_V4;
+ }
return rc;
}
@@ -3254,14 +3271,23 @@
}
#endif
-static inline void bnx2x_set_pbd_gso_e2(struct sk_buff *skb, u32 *parsing_data,
- u32 xmit_type)
+static void bnx2x_set_pbd_gso_e2(struct sk_buff *skb, u32 *parsing_data,
+ u32 xmit_type)
{
+ struct ipv6hdr *ipv6;
+
*parsing_data |= (skb_shinfo(skb)->gso_size <<
ETH_TX_PARSE_BD_E2_LSO_MSS_SHIFT) &
ETH_TX_PARSE_BD_E2_LSO_MSS;
- if ((xmit_type & XMIT_GSO_V6) &&
- (ipv6_hdr(skb)->nexthdr == NEXTHDR_IPV6))
+
+ if (xmit_type & XMIT_GSO_ENC_V6)
+ ipv6 = inner_ipv6_hdr(skb);
+ else if (xmit_type & XMIT_GSO_V6)
+ ipv6 = ipv6_hdr(skb);
+ else
+ ipv6 = NULL;
+
+ if (ipv6 && ipv6->nexthdr == NEXTHDR_IPV6)
*parsing_data |= ETH_TX_PARSE_BD_E2_IPV6_WITH_EXT_HDR;
}
@@ -3272,13 +3298,13 @@
* @pbd: parse BD
* @xmit_type: xmit flags
*/
-static inline void bnx2x_set_pbd_gso(struct sk_buff *skb,
- struct eth_tx_parse_bd_e1x *pbd,
- u32 xmit_type)
+static void bnx2x_set_pbd_gso(struct sk_buff *skb,
+ struct eth_tx_parse_bd_e1x *pbd,
+ u32 xmit_type)
{
pbd->lso_mss = cpu_to_le16(skb_shinfo(skb)->gso_size);
pbd->tcp_send_seq = bswab32(tcp_hdr(skb)->seq);
- pbd->tcp_flags = pbd_tcp_flags(skb);
+ pbd->tcp_flags = pbd_tcp_flags(tcp_hdr(skb));
if (xmit_type & XMIT_GSO_V4) {
pbd->ip_id = bswab16(ip_hdr(skb)->id);
@@ -3298,6 +3324,40 @@
}
/**
+ * bnx2x_set_pbd_csum_enc - update PBD with checksum and return header length
+ *
+ * @bp: driver handle
+ * @skb: packet skb
+ * @parsing_data: data to be updated
+ * @xmit_type: xmit flags
+ *
+ * 57712/578xx related, when skb has encapsulation
+ */
+static u8 bnx2x_set_pbd_csum_enc(struct bnx2x *bp, struct sk_buff *skb,
+ u32 *parsing_data, u32 xmit_type)
+{
+ *parsing_data |=
+ ((((u8 *)skb_inner_transport_header(skb) - skb->data) >> 1) <<
+ ETH_TX_PARSE_BD_E2_L4_HDR_START_OFFSET_W_SHIFT) &
+ ETH_TX_PARSE_BD_E2_L4_HDR_START_OFFSET_W;
+
+ if (xmit_type & XMIT_CSUM_TCP) {
+ *parsing_data |= ((inner_tcp_hdrlen(skb) / 4) <<
+ ETH_TX_PARSE_BD_E2_TCP_HDR_LENGTH_DW_SHIFT) &
+ ETH_TX_PARSE_BD_E2_TCP_HDR_LENGTH_DW;
+
+ return skb_inner_transport_header(skb) +
+ inner_tcp_hdrlen(skb) - skb->data;
+ }
+
+ /* We support checksum offload for TCP and UDP only.
+ * No need to pass the UDP header length - it's a constant.
+ */
+ return skb_inner_transport_header(skb) +
+ sizeof(struct udphdr) - skb->data;
+}
+
+/**
* bnx2x_set_pbd_csum_e2 - update PBD with checksum and return header length
*
* @bp: driver handle
@@ -3305,15 +3365,15 @@
* @parsing_data: data to be updated
* @xmit_type: xmit flags
*
- * 57712 related
+ * 57712/578xx related
*/
-static inline u8 bnx2x_set_pbd_csum_e2(struct bnx2x *bp, struct sk_buff *skb,
- u32 *parsing_data, u32 xmit_type)
+static u8 bnx2x_set_pbd_csum_e2(struct bnx2x *bp, struct sk_buff *skb,
+ u32 *parsing_data, u32 xmit_type)
{
*parsing_data |=
((((u8 *)skb_transport_header(skb) - skb->data) >> 1) <<
- ETH_TX_PARSE_BD_E2_TCP_HDR_START_OFFSET_W_SHIFT) &
- ETH_TX_PARSE_BD_E2_TCP_HDR_START_OFFSET_W;
+ ETH_TX_PARSE_BD_E2_L4_HDR_START_OFFSET_W_SHIFT) &
+ ETH_TX_PARSE_BD_E2_L4_HDR_START_OFFSET_W;
if (xmit_type & XMIT_CSUM_TCP) {
*parsing_data |= ((tcp_hdrlen(skb) / 4) <<
@@ -3328,17 +3388,15 @@
return skb_transport_header(skb) + sizeof(struct udphdr) - skb->data;
}
-static inline void bnx2x_set_sbd_csum(struct bnx2x *bp, struct sk_buff *skb,
- struct eth_tx_start_bd *tx_start_bd, u32 xmit_type)
+/* set FW indication according to inner or outer protocols if tunneled */
+static void bnx2x_set_sbd_csum(struct bnx2x *bp, struct sk_buff *skb,
+ struct eth_tx_start_bd *tx_start_bd,
+ u32 xmit_type)
{
tx_start_bd->bd_flags.as_bitfield |= ETH_TX_BD_FLAGS_L4_CSUM;
- if (xmit_type & XMIT_CSUM_V4)
- tx_start_bd->bd_flags.as_bitfield |=
- ETH_TX_BD_FLAGS_IP_CSUM;
- else
- tx_start_bd->bd_flags.as_bitfield |=
- ETH_TX_BD_FLAGS_IPV6;
+ if (xmit_type & (XMIT_CSUM_ENC_V6 | XMIT_CSUM_V6))
+ tx_start_bd->bd_flags.as_bitfield |= ETH_TX_BD_FLAGS_IPV6;
if (!(xmit_type & XMIT_CSUM_TCP))
tx_start_bd->bd_flags.as_bitfield |= ETH_TX_BD_FLAGS_IS_UDP;
@@ -3352,9 +3410,9 @@
* @pbd: parse BD to be updated
* @xmit_type: xmit flags
*/
-static inline u8 bnx2x_set_pbd_csum(struct bnx2x *bp, struct sk_buff *skb,
- struct eth_tx_parse_bd_e1x *pbd,
- u32 xmit_type)
+static u8 bnx2x_set_pbd_csum(struct bnx2x *bp, struct sk_buff *skb,
+ struct eth_tx_parse_bd_e1x *pbd,
+ u32 xmit_type)
{
u8 hlen = (skb_network_header(skb) - skb->data) >> 1;
@@ -3400,6 +3458,72 @@
return hlen;
}
+static void bnx2x_update_pbds_gso_enc(struct sk_buff *skb,
+ struct eth_tx_parse_bd_e2 *pbd_e2,
+ struct eth_tx_parse_2nd_bd *pbd2,
+ u16 *global_data,
+ u32 xmit_type)
+{
+ u16 inner_hlen_w = 0;
+ u8 outerip_off, outerip_len = 0;
+
+ /* IP len */
+ inner_hlen_w = (skb_inner_transport_header(skb) -
+ skb_inner_network_header(skb)) >> 1;
+
+ /* transport len */
+ if (xmit_type & XMIT_CSUM_TCP)
+ inner_hlen_w += inner_tcp_hdrlen(skb) >> 1;
+ else
+ inner_hlen_w += sizeof(struct udphdr) >> 1;
+
+ pbd2->fw_ip_hdr_to_payload_w = inner_hlen_w;
+
+ if (xmit_type & XMIT_CSUM_ENC_V4) {
+ struct iphdr *iph = inner_ip_hdr(skb);
+
+ pbd2->fw_ip_csum_wo_len_flags_frag =
+ bswab16(csum_fold((~iph->check) -
+ iph->tot_len - iph->frag_off));
+ } else {
+ pbd2->fw_ip_hdr_to_payload_w =
+ inner_hlen_w - ((sizeof(struct ipv6hdr)) >> 1);
+ }
+
+ pbd2->tcp_send_seq = bswab32(inner_tcp_hdr(skb)->seq);
+
+ pbd2->tcp_flags = pbd_tcp_flags(inner_tcp_hdr(skb));
+
+ if (xmit_type & XMIT_GSO_V4) {
+ pbd2->hw_ip_id = bswab16(ip_hdr(skb)->id);
+
+ pbd_e2->data.tunnel_data.pseudo_csum =
+ bswab16(~csum_tcpudp_magic(
+ inner_ip_hdr(skb)->saddr,
+ inner_ip_hdr(skb)->daddr,
+ 0, IPPROTO_TCP, 0));
+
+ outerip_len = ip_hdr(skb)->ihl << 1;
+ } else {
+ pbd_e2->data.tunnel_data.pseudo_csum =
+ bswab16(~csum_ipv6_magic(
+ &inner_ipv6_hdr(skb)->saddr,
+ &inner_ipv6_hdr(skb)->daddr,
+ 0, IPPROTO_TCP, 0));
+ }
+
+ outerip_off = (skb_network_header(skb) - skb->data) >> 1;
+
+ *global_data |=
+ outerip_off |
+ (!!(xmit_type & XMIT_CSUM_V6) <<
+ ETH_TX_PARSE_2ND_BD_IP_HDR_TYPE_OUTER_SHIFT) |
+ (outerip_len <<
+ ETH_TX_PARSE_2ND_BD_IP_HDR_LEN_OUTER_W_SHIFT) |
+ ((skb->protocol == cpu_to_be16(ETH_P_8021Q)) <<
+ ETH_TX_PARSE_2ND_BD_LLC_SNAP_EN_SHIFT);
+}
+
/* called with netif_tx_lock
* bnx2x_tx_int() runs without netif_tx_lock unless it needs to call
* netif_wake_queue()
@@ -3415,6 +3539,7 @@
struct eth_tx_bd *tx_data_bd, *total_pkt_bd = NULL;
struct eth_tx_parse_bd_e1x *pbd_e1x = NULL;
struct eth_tx_parse_bd_e2 *pbd_e2 = NULL;
+ struct eth_tx_parse_2nd_bd *pbd2 = NULL;
u32 pbd_e2_parsing_data = 0;
u16 pkt_prod, bd_prod;
int nbd, txq_index;
@@ -3482,7 +3607,7 @@
mac_type = MULTICAST_ADDRESS;
}
-#if (MAX_SKB_FRAGS >= MAX_FETCH_BD - 3)
+#if (MAX_SKB_FRAGS >= MAX_FETCH_BD - BDS_PER_TX_PKT)
/* First, check if we need to linearize the skb (due to FW
restrictions). No need to check fragmentation if page size > 8K
(there will be no violation to FW restrictions) */
@@ -3530,12 +3655,9 @@
first_bd = tx_start_bd;
tx_start_bd->bd_flags.as_bitfield = ETH_TX_BD_FLAGS_START_BD;
- SET_FLAG(tx_start_bd->general_data,
- ETH_TX_START_BD_PARSE_NBDS,
- 0);
- /* header nbd */
- SET_FLAG(tx_start_bd->general_data, ETH_TX_START_BD_HDR_NBDS, 1);
+ /* header nbd: indirectly zero other flags! */
+ tx_start_bd->general_data = 1 << ETH_TX_START_BD_HDR_NBDS_SHIFT;
/* remember the first BD of the packet */
tx_buf->first_bd = txdata->tx_bd_prod;
@@ -3555,19 +3677,16 @@
/* when transmitting in a vf, start bd must hold the ethertype
* for fw to enforce it
*/
-#ifndef BNX2X_STOP_ON_ERROR
- if (IS_VF(bp)) {
-#endif
+ if (IS_VF(bp))
tx_start_bd->vlan_or_ethertype =
cpu_to_le16(ntohs(eth->h_proto));
-#ifndef BNX2X_STOP_ON_ERROR
- } else {
+ else
/* used by FW for packet accounting */
tx_start_bd->vlan_or_ethertype = cpu_to_le16(pkt_prod);
- }
-#endif
}
+ nbd = 2; /* start_bd + pbd + frags (updated when pages are mapped) */
+
/* turn on parsing and get a BD */
bd_prod = TX_BD(NEXT_TX_IDX(bd_prod));
@@ -3577,23 +3696,58 @@
if (!CHIP_IS_E1x(bp)) {
pbd_e2 = &txdata->tx_desc_ring[bd_prod].parse_bd_e2;
memset(pbd_e2, 0, sizeof(struct eth_tx_parse_bd_e2));
- /* Set PBD in checksum offload case */
- if (xmit_type & XMIT_CSUM)
+
+ if (xmit_type & XMIT_CSUM_ENC) {
+ u16 global_data = 0;
+
+ /* Set PBD in enc checksum offload case */
+ hlen = bnx2x_set_pbd_csum_enc(bp, skb,
+ &pbd_e2_parsing_data,
+ xmit_type);
+
+ /* turn on 2nd parsing and get a BD */
+ bd_prod = TX_BD(NEXT_TX_IDX(bd_prod));
+
+ pbd2 = &txdata->tx_desc_ring[bd_prod].parse_2nd_bd;
+
+ memset(pbd2, 0, sizeof(*pbd2));
+
+ pbd_e2->data.tunnel_data.ip_hdr_start_inner_w =
+ (skb_inner_network_header(skb) -
+ skb->data) >> 1;
+
+ if (xmit_type & XMIT_GSO_ENC)
+ bnx2x_update_pbds_gso_enc(skb, pbd_e2, pbd2,
+ &global_data,
+ xmit_type);
+
+ pbd2->global_data = cpu_to_le16(global_data);
+
+ /* add addition parse BD indication to start BD */
+ SET_FLAG(tx_start_bd->general_data,
+ ETH_TX_START_BD_PARSE_NBDS, 1);
+ /* set encapsulation flag in start BD */
+ SET_FLAG(tx_start_bd->general_data,
+ ETH_TX_START_BD_TUNNEL_EXIST, 1);
+ nbd++;
+ } else if (xmit_type & XMIT_CSUM) {
+ /* Set PBD in checksum offload case w/o encapsulation */
hlen = bnx2x_set_pbd_csum_e2(bp, skb,
&pbd_e2_parsing_data,
xmit_type);
+ }
- if (IS_MF_SI(bp) || IS_VF(bp)) {
- /* fill in the MAC addresses in the PBD - for local
- * switching
- */
- bnx2x_set_fw_mac_addr(&pbd_e2->src_mac_addr_hi,
- &pbd_e2->src_mac_addr_mid,
- &pbd_e2->src_mac_addr_lo,
+ /* Add the macs to the parsing BD this is a vf */
+ if (IS_VF(bp)) {
+ /* override GRE parameters in BD */
+ bnx2x_set_fw_mac_addr(&pbd_e2->data.mac_addr.src_hi,
+ &pbd_e2->data.mac_addr.src_mid,
+ &pbd_e2->data.mac_addr.src_lo,
eth->h_source);
- bnx2x_set_fw_mac_addr(&pbd_e2->dst_mac_addr_hi,
- &pbd_e2->dst_mac_addr_mid,
- &pbd_e2->dst_mac_addr_lo,
+
+ bnx2x_set_fw_mac_addr(&pbd_e2->data.mac_addr.dst_hi,
+ &pbd_e2->data.mac_addr.dst_mid,
+ &pbd_e2->data.mac_addr.dst_lo,
eth->h_dest);
}
@@ -3615,14 +3769,13 @@
/* Setup the data pointer of the first BD of the packet */
tx_start_bd->addr_hi = cpu_to_le32(U64_HI(mapping));
tx_start_bd->addr_lo = cpu_to_le32(U64_LO(mapping));
- nbd = 2; /* start_bd + pbd + frags (updated when pages are mapped) */
tx_start_bd->nbytes = cpu_to_le16(skb_headlen(skb));
pkt_size = tx_start_bd->nbytes;
DP(NETIF_MSG_TX_QUEUED,
- "first bd @%p addr (%x:%x) nbd %d nbytes %d flags %x vlan %x\n",
+ "first bd @%p addr (%x:%x) nbytes %d flags %x vlan %x\n",
tx_start_bd, tx_start_bd->addr_hi, tx_start_bd->addr_lo,
- le16_to_cpu(tx_start_bd->nbd), le16_to_cpu(tx_start_bd->nbytes),
+ le16_to_cpu(tx_start_bd->nbytes),
tx_start_bd->bd_flags.as_bitfield,
le16_to_cpu(tx_start_bd->vlan_or_ethertype));
@@ -3635,10 +3788,12 @@
tx_start_bd->bd_flags.as_bitfield |= ETH_TX_BD_FLAGS_SW_LSO;
- if (unlikely(skb_headlen(skb) > hlen))
+ if (unlikely(skb_headlen(skb) > hlen)) {
+ nbd++;
bd_prod = bnx2x_tx_split(bp, txdata, tx_buf,
&tx_start_bd, hlen,
- bd_prod, ++nbd);
+ bd_prod);
+ }
if (!CHIP_IS_E1x(bp))
bnx2x_set_pbd_gso_e2(skb, &pbd_e2_parsing_data,
xmit_type);
@@ -3728,9 +3883,13 @@
if (pbd_e2)
DP(NETIF_MSG_TX_QUEUED,
"PBD (E2) @%p dst %x %x %x src %x %x %x parsing_data %x\n",
- pbd_e2, pbd_e2->dst_mac_addr_hi, pbd_e2->dst_mac_addr_mid,
- pbd_e2->dst_mac_addr_lo, pbd_e2->src_mac_addr_hi,
- pbd_e2->src_mac_addr_mid, pbd_e2->src_mac_addr_lo,
+ pbd_e2,
+ pbd_e2->data.mac_addr.dst_hi,
+ pbd_e2->data.mac_addr.dst_mid,
+ pbd_e2->data.mac_addr.dst_lo,
+ pbd_e2->data.mac_addr.src_hi,
+ pbd_e2->data.mac_addr.src_mid,
+ pbd_e2->data.mac_addr.src_lo,
pbd_e2->parsing_data);
DP(NETIF_MSG_TX_QUEUED, "doorbell: nbd %d bd %u\n", nbd, bd_prod);
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
index aee7671..f9098d8 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
@@ -50,13 +50,13 @@
} \
} while (0)
-#define BNX2X_PCI_ALLOC(x, y, size) \
- do { \
- x = dma_alloc_coherent(&bp->pdev->dev, size, y, GFP_KERNEL); \
- if (x == NULL) \
- goto alloc_mem_err; \
- memset((void *)x, 0, size); \
- } while (0)
+#define BNX2X_PCI_ALLOC(x, y, size) \
+do { \
+ x = dma_alloc_coherent(&bp->pdev->dev, size, y, \
+ GFP_KERNEL | __GFP_ZERO); \
+ if (x == NULL) \
+ goto alloc_mem_err; \
+} while (0)
#define BNX2X_ALLOC(x, size) \
do { \
@@ -496,7 +496,10 @@
/* setup_tc callback */
int bnx2x_setup_tc(struct net_device *dev, u8 num_tc);
+int bnx2x_get_vf_config(struct net_device *dev, int vf,
+ struct ifla_vf_info *ivi);
int bnx2x_set_vf_mac(struct net_device *dev, int queue, u8 *mac);
+int bnx2x_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, u8 qos);
/* select_queue callback */
u16 bnx2x_select_queue(struct net_device *dev, struct sk_buff *skb);
@@ -834,7 +837,7 @@
/* Add NAPI objects */
for_each_rx_queue_cnic(bp, i)
netif_napi_add(bp->dev, &bnx2x_fp(bp, i, napi),
- bnx2x_poll, BNX2X_NAPI_WEIGHT);
+ bnx2x_poll, NAPI_POLL_WEIGHT);
}
static inline void bnx2x_add_all_napi(struct bnx2x *bp)
@@ -844,7 +847,7 @@
/* Add NAPI objects */
for_each_eth_queue(bp, i)
netif_napi_add(bp->dev, &bnx2x_fp(bp, i, napi),
- bnx2x_poll, BNX2X_NAPI_WEIGHT);
+ bnx2x_poll, NAPI_POLL_WEIGHT);
}
static inline void bnx2x_del_all_napi_cnic(struct bnx2x *bp)
@@ -970,6 +973,9 @@
else /* CHIP_IS_E1X */
start_params->network_cos_mode = FW_WRR;
+ start_params->gre_tunnel_mode = IPGRE_TUNNEL;
+ start_params->gre_tunnel_rss = GRE_INNER_HEADERS_RSS;
+
return bnx2x_func_state_change(bp, &func_params);
}
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c
index edfa67a..324d691 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c
@@ -3232,7 +3232,32 @@
.get_ts_info = ethtool_op_get_ts_info,
};
-void bnx2x_set_ethtool_ops(struct net_device *netdev)
+static const struct ethtool_ops bnx2x_vf_ethtool_ops = {
+ .get_settings = bnx2x_get_settings,
+ .set_settings = bnx2x_set_settings,
+ .get_drvinfo = bnx2x_get_drvinfo,
+ .get_msglevel = bnx2x_get_msglevel,
+ .set_msglevel = bnx2x_set_msglevel,
+ .get_link = bnx2x_get_link,
+ .get_coalesce = bnx2x_get_coalesce,
+ .get_ringparam = bnx2x_get_ringparam,
+ .set_ringparam = bnx2x_set_ringparam,
+ .get_sset_count = bnx2x_get_sset_count,
+ .get_strings = bnx2x_get_strings,
+ .get_ethtool_stats = bnx2x_get_ethtool_stats,
+ .get_rxnfc = bnx2x_get_rxnfc,
+ .set_rxnfc = bnx2x_set_rxnfc,
+ .get_rxfh_indir_size = bnx2x_get_rxfh_indir_size,
+ .get_rxfh_indir = bnx2x_get_rxfh_indir,
+ .set_rxfh_indir = bnx2x_set_rxfh_indir,
+ .get_channels = bnx2x_get_channels,
+ .set_channels = bnx2x_set_channels,
+};
+
+void bnx2x_set_ethtool_ops(struct bnx2x *bp, struct net_device *netdev)
{
- SET_ETHTOOL_OPS(netdev, &bnx2x_ethtool_ops);
+ if (IS_PF(bp))
+ SET_ETHTOOL_OPS(netdev, &bnx2x_ethtool_ops);
+ else /* vf */
+ SET_ETHTOOL_OPS(netdev, &bnx2x_vf_ethtool_ops);
}
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_fw_defs.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_fw_defs.h
index e5f8083..40f22c6 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_fw_defs.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_fw_defs.h
@@ -30,31 +30,31 @@
* IRO[138].m2) + ((sbId) * IRO[138].m3))
#define CSTORM_IGU_MODE_OFFSET (IRO[157].base)
#define CSTORM_ISCSI_CQ_SIZE_OFFSET(pfId) \
- (IRO[316].base + ((pfId) * IRO[316].m1))
-#define CSTORM_ISCSI_CQ_SQN_SIZE_OFFSET(pfId) \
(IRO[317].base + ((pfId) * IRO[317].m1))
+#define CSTORM_ISCSI_CQ_SQN_SIZE_OFFSET(pfId) \
+ (IRO[318].base + ((pfId) * IRO[318].m1))
#define CSTORM_ISCSI_EQ_CONS_OFFSET(pfId, iscsiEqId) \
- (IRO[309].base + ((pfId) * IRO[309].m1) + ((iscsiEqId) * IRO[309].m2))
-#define CSTORM_ISCSI_EQ_NEXT_EQE_ADDR_OFFSET(pfId, iscsiEqId) \
- (IRO[311].base + ((pfId) * IRO[311].m1) + ((iscsiEqId) * IRO[311].m2))
-#define CSTORM_ISCSI_EQ_NEXT_PAGE_ADDR_OFFSET(pfId, iscsiEqId) \
(IRO[310].base + ((pfId) * IRO[310].m1) + ((iscsiEqId) * IRO[310].m2))
-#define CSTORM_ISCSI_EQ_NEXT_PAGE_ADDR_VALID_OFFSET(pfId, iscsiEqId) \
+#define CSTORM_ISCSI_EQ_NEXT_EQE_ADDR_OFFSET(pfId, iscsiEqId) \
(IRO[312].base + ((pfId) * IRO[312].m1) + ((iscsiEqId) * IRO[312].m2))
-#define CSTORM_ISCSI_EQ_PROD_OFFSET(pfId, iscsiEqId) \
- (IRO[308].base + ((pfId) * IRO[308].m1) + ((iscsiEqId) * IRO[308].m2))
-#define CSTORM_ISCSI_EQ_SB_INDEX_OFFSET(pfId, iscsiEqId) \
- (IRO[314].base + ((pfId) * IRO[314].m1) + ((iscsiEqId) * IRO[314].m2))
-#define CSTORM_ISCSI_EQ_SB_NUM_OFFSET(pfId, iscsiEqId) \
+#define CSTORM_ISCSI_EQ_NEXT_PAGE_ADDR_OFFSET(pfId, iscsiEqId) \
+ (IRO[311].base + ((pfId) * IRO[311].m1) + ((iscsiEqId) * IRO[311].m2))
+#define CSTORM_ISCSI_EQ_NEXT_PAGE_ADDR_VALID_OFFSET(pfId, iscsiEqId) \
(IRO[313].base + ((pfId) * IRO[313].m1) + ((iscsiEqId) * IRO[313].m2))
+#define CSTORM_ISCSI_EQ_PROD_OFFSET(pfId, iscsiEqId) \
+ (IRO[309].base + ((pfId) * IRO[309].m1) + ((iscsiEqId) * IRO[309].m2))
+#define CSTORM_ISCSI_EQ_SB_INDEX_OFFSET(pfId, iscsiEqId) \
+ (IRO[315].base + ((pfId) * IRO[315].m1) + ((iscsiEqId) * IRO[315].m2))
+#define CSTORM_ISCSI_EQ_SB_NUM_OFFSET(pfId, iscsiEqId) \
+ (IRO[314].base + ((pfId) * IRO[314].m1) + ((iscsiEqId) * IRO[314].m2))
#define CSTORM_ISCSI_HQ_SIZE_OFFSET(pfId) \
- (IRO[315].base + ((pfId) * IRO[315].m1))
+ (IRO[316].base + ((pfId) * IRO[316].m1))
#define CSTORM_ISCSI_NUM_OF_TASKS_OFFSET(pfId) \
- (IRO[307].base + ((pfId) * IRO[307].m1))
+ (IRO[308].base + ((pfId) * IRO[308].m1))
#define CSTORM_ISCSI_PAGE_SIZE_LOG_OFFSET(pfId) \
- (IRO[306].base + ((pfId) * IRO[306].m1))
+ (IRO[307].base + ((pfId) * IRO[307].m1))
#define CSTORM_ISCSI_PAGE_SIZE_OFFSET(pfId) \
- (IRO[305].base + ((pfId) * IRO[305].m1))
+ (IRO[306].base + ((pfId) * IRO[306].m1))
#define CSTORM_RECORD_SLOW_PATH_OFFSET(funcId) \
(IRO[151].base + ((funcId) * IRO[151].m1))
#define CSTORM_SP_STATUS_BLOCK_DATA_OFFSET(pfId) \
@@ -114,7 +114,7 @@
#define TSTORM_ISCSI_RQ_SIZE_OFFSET(pfId) \
(IRO[268].base + ((pfId) * IRO[268].m1))
#define TSTORM_ISCSI_TCP_LOCAL_ADV_WND_OFFSET(pfId) \
- (IRO[277].base + ((pfId) * IRO[277].m1))
+ (IRO[278].base + ((pfId) * IRO[278].m1))
#define TSTORM_ISCSI_TCP_VARS_FLAGS_OFFSET(pfId) \
(IRO[264].base + ((pfId) * IRO[264].m1))
#define TSTORM_ISCSI_TCP_VARS_LSB_LOCAL_MAC_ADDR_OFFSET(pfId) \
@@ -136,35 +136,32 @@
#define USTORM_ASSERT_LIST_INDEX_OFFSET (IRO[177].base)
#define USTORM_ASSERT_LIST_OFFSET(assertListEntry) \
(IRO[176].base + ((assertListEntry) * IRO[176].m1))
-#define USTORM_CQE_PAGE_NEXT_OFFSET(portId, clientId) \
- (IRO[205].base + ((portId) * IRO[205].m1) + ((clientId) * \
- IRO[205].m2))
#define USTORM_ETH_PAUSE_ENABLED_OFFSET(portId) \
(IRO[183].base + ((portId) * IRO[183].m1))
#define USTORM_FCOE_EQ_PROD_OFFSET(pfId) \
- (IRO[318].base + ((pfId) * IRO[318].m1))
+ (IRO[319].base + ((pfId) * IRO[319].m1))
#define USTORM_FUNC_EN_OFFSET(funcId) \
(IRO[178].base + ((funcId) * IRO[178].m1))
#define USTORM_ISCSI_CQ_SIZE_OFFSET(pfId) \
- (IRO[282].base + ((pfId) * IRO[282].m1))
-#define USTORM_ISCSI_CQ_SQN_SIZE_OFFSET(pfId) \
(IRO[283].base + ((pfId) * IRO[283].m1))
-#define USTORM_ISCSI_ERROR_BITMAP_OFFSET(pfId) \
- (IRO[287].base + ((pfId) * IRO[287].m1))
-#define USTORM_ISCSI_GLOBAL_BUF_PHYS_ADDR_OFFSET(pfId) \
+#define USTORM_ISCSI_CQ_SQN_SIZE_OFFSET(pfId) \
(IRO[284].base + ((pfId) * IRO[284].m1))
-#define USTORM_ISCSI_NUM_OF_TASKS_OFFSET(pfId) \
- (IRO[280].base + ((pfId) * IRO[280].m1))
-#define USTORM_ISCSI_PAGE_SIZE_LOG_OFFSET(pfId) \
- (IRO[279].base + ((pfId) * IRO[279].m1))
-#define USTORM_ISCSI_PAGE_SIZE_OFFSET(pfId) \
- (IRO[278].base + ((pfId) * IRO[278].m1))
-#define USTORM_ISCSI_R2TQ_SIZE_OFFSET(pfId) \
- (IRO[281].base + ((pfId) * IRO[281].m1))
-#define USTORM_ISCSI_RQ_BUFFER_SIZE_OFFSET(pfId) \
+#define USTORM_ISCSI_ERROR_BITMAP_OFFSET(pfId) \
+ (IRO[288].base + ((pfId) * IRO[288].m1))
+#define USTORM_ISCSI_GLOBAL_BUF_PHYS_ADDR_OFFSET(pfId) \
(IRO[285].base + ((pfId) * IRO[285].m1))
-#define USTORM_ISCSI_RQ_SIZE_OFFSET(pfId) \
+#define USTORM_ISCSI_NUM_OF_TASKS_OFFSET(pfId) \
+ (IRO[281].base + ((pfId) * IRO[281].m1))
+#define USTORM_ISCSI_PAGE_SIZE_LOG_OFFSET(pfId) \
+ (IRO[280].base + ((pfId) * IRO[280].m1))
+#define USTORM_ISCSI_PAGE_SIZE_OFFSET(pfId) \
+ (IRO[279].base + ((pfId) * IRO[279].m1))
+#define USTORM_ISCSI_R2TQ_SIZE_OFFSET(pfId) \
+ (IRO[282].base + ((pfId) * IRO[282].m1))
+#define USTORM_ISCSI_RQ_BUFFER_SIZE_OFFSET(pfId) \
(IRO[286].base + ((pfId) * IRO[286].m1))
+#define USTORM_ISCSI_RQ_SIZE_OFFSET(pfId) \
+ (IRO[287].base + ((pfId) * IRO[287].m1))
#define USTORM_MEM_WORKAROUND_ADDRESS_OFFSET(pfId) \
(IRO[182].base + ((pfId) * IRO[182].m1))
#define USTORM_RECORD_SLOW_PATH_OFFSET(funcId) \
@@ -190,39 +187,39 @@
#define XSTORM_FUNC_EN_OFFSET(funcId) \
(IRO[47].base + ((funcId) * IRO[47].m1))
#define XSTORM_ISCSI_HQ_SIZE_OFFSET(pfId) \
- (IRO[295].base + ((pfId) * IRO[295].m1))
-#define XSTORM_ISCSI_LOCAL_MAC_ADDR0_OFFSET(pfId) \
- (IRO[298].base + ((pfId) * IRO[298].m1))
-#define XSTORM_ISCSI_LOCAL_MAC_ADDR1_OFFSET(pfId) \
- (IRO[299].base + ((pfId) * IRO[299].m1))
-#define XSTORM_ISCSI_LOCAL_MAC_ADDR2_OFFSET(pfId) \
- (IRO[300].base + ((pfId) * IRO[300].m1))
-#define XSTORM_ISCSI_LOCAL_MAC_ADDR3_OFFSET(pfId) \
- (IRO[301].base + ((pfId) * IRO[301].m1))
-#define XSTORM_ISCSI_LOCAL_MAC_ADDR4_OFFSET(pfId) \
- (IRO[302].base + ((pfId) * IRO[302].m1))
-#define XSTORM_ISCSI_LOCAL_MAC_ADDR5_OFFSET(pfId) \
- (IRO[303].base + ((pfId) * IRO[303].m1))
-#define XSTORM_ISCSI_LOCAL_VLAN_OFFSET(pfId) \
- (IRO[304].base + ((pfId) * IRO[304].m1))
-#define XSTORM_ISCSI_NUM_OF_TASKS_OFFSET(pfId) \
- (IRO[294].base + ((pfId) * IRO[294].m1))
-#define XSTORM_ISCSI_PAGE_SIZE_LOG_OFFSET(pfId) \
- (IRO[293].base + ((pfId) * IRO[293].m1))
-#define XSTORM_ISCSI_PAGE_SIZE_OFFSET(pfId) \
- (IRO[292].base + ((pfId) * IRO[292].m1))
-#define XSTORM_ISCSI_R2TQ_SIZE_OFFSET(pfId) \
- (IRO[297].base + ((pfId) * IRO[297].m1))
-#define XSTORM_ISCSI_SQ_SIZE_OFFSET(pfId) \
(IRO[296].base + ((pfId) * IRO[296].m1))
+#define XSTORM_ISCSI_LOCAL_MAC_ADDR0_OFFSET(pfId) \
+ (IRO[299].base + ((pfId) * IRO[299].m1))
+#define XSTORM_ISCSI_LOCAL_MAC_ADDR1_OFFSET(pfId) \
+ (IRO[300].base + ((pfId) * IRO[300].m1))
+#define XSTORM_ISCSI_LOCAL_MAC_ADDR2_OFFSET(pfId) \
+ (IRO[301].base + ((pfId) * IRO[301].m1))
+#define XSTORM_ISCSI_LOCAL_MAC_ADDR3_OFFSET(pfId) \
+ (IRO[302].base + ((pfId) * IRO[302].m1))
+#define XSTORM_ISCSI_LOCAL_MAC_ADDR4_OFFSET(pfId) \
+ (IRO[303].base + ((pfId) * IRO[303].m1))
+#define XSTORM_ISCSI_LOCAL_MAC_ADDR5_OFFSET(pfId) \
+ (IRO[304].base + ((pfId) * IRO[304].m1))
+#define XSTORM_ISCSI_LOCAL_VLAN_OFFSET(pfId) \
+ (IRO[305].base + ((pfId) * IRO[305].m1))
+#define XSTORM_ISCSI_NUM_OF_TASKS_OFFSET(pfId) \
+ (IRO[295].base + ((pfId) * IRO[295].m1))
+#define XSTORM_ISCSI_PAGE_SIZE_LOG_OFFSET(pfId) \
+ (IRO[294].base + ((pfId) * IRO[294].m1))
+#define XSTORM_ISCSI_PAGE_SIZE_OFFSET(pfId) \
+ (IRO[293].base + ((pfId) * IRO[293].m1))
+#define XSTORM_ISCSI_R2TQ_SIZE_OFFSET(pfId) \
+ (IRO[298].base + ((pfId) * IRO[298].m1))
+#define XSTORM_ISCSI_SQ_SIZE_OFFSET(pfId) \
+ (IRO[297].base + ((pfId) * IRO[297].m1))
#define XSTORM_ISCSI_TCP_VARS_ADV_WND_SCL_OFFSET(pfId) \
- (IRO[291].base + ((pfId) * IRO[291].m1))
+ (IRO[292].base + ((pfId) * IRO[292].m1))
#define XSTORM_ISCSI_TCP_VARS_FLAGS_OFFSET(pfId) \
- (IRO[290].base + ((pfId) * IRO[290].m1))
+ (IRO[291].base + ((pfId) * IRO[291].m1))
#define XSTORM_ISCSI_TCP_VARS_TOS_OFFSET(pfId) \
- (IRO[289].base + ((pfId) * IRO[289].m1))
+ (IRO[290].base + ((pfId) * IRO[290].m1))
#define XSTORM_ISCSI_TCP_VARS_TTL_OFFSET(pfId) \
- (IRO[288].base + ((pfId) * IRO[288].m1))
+ (IRO[289].base + ((pfId) * IRO[289].m1))
#define XSTORM_RATE_SHAPING_PER_VN_VARS_OFFSET(pfId) \
(IRO[44].base + ((pfId) * IRO[44].m1))
#define XSTORM_RECORD_SLOW_PATH_OFFSET(funcId) \
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h
index 037860e..12f00a4 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h
@@ -114,6 +114,10 @@
#define EPIO_CFG_EPIO30 0x0000001f
#define EPIO_CFG_EPIO31 0x00000020
+struct mac_addr {
+ u32 upper;
+ u32 lower;
+};
struct shared_hw_cfg { /* NVRAM Offset */
/* Up to 16 bytes of NULL-terminated string */
@@ -508,7 +512,22 @@
#define PORT_HW_CFG_PAUSE_ON_HOST_RING_DISABLED 0x00000000
#define PORT_HW_CFG_PAUSE_ON_HOST_RING_ENABLED 0x00000001
- u32 reserved0[6]; /* 0x178 */
+ /* SFP+ Tx Equalization: NIC recommended and tested value is 0xBEB2
+ * LOM recommended and tested value is 0xBEB2. Using a different
+ * value means using a value not tested by BRCM
+ */
+ u32 sfi_tap_values; /* 0x178 */
+ #define PORT_HW_CFG_TX_EQUALIZATION_MASK 0x0000FFFF
+ #define PORT_HW_CFG_TX_EQUALIZATION_SHIFT 0
+
+ /* SFP+ Tx driver broadcast IDRIVER: NIC recommended and tested
+ * value is 0x2. LOM recommended and tested value is 0x2. Using a
+ * different value means using a value not tested by BRCM
+ */
+ #define PORT_HW_CFG_TX_DRV_BROADCAST_MASK 0x000F0000
+ #define PORT_HW_CFG_TX_DRV_BROADCAST_SHIFT 16
+
+ u32 reserved0[5]; /* 0x17c */
u32 aeu_int_mask; /* 0x190 */
@@ -2821,8 +2840,8 @@
#define BCM_5710_FW_MAJOR_VERSION 7
#define BCM_5710_FW_MINOR_VERSION 8
-#define BCM_5710_FW_REVISION_VERSION 2
-#define BCM_5710_FW_ENGINEERING_VERSION 0
+#define BCM_5710_FW_REVISION_VERSION 17
+#define BCM_5710_FW_ENGINEERING_VERSION 0
#define BCM_5710_FW_COMPILE_FLAGS 1
@@ -3513,11 +3532,14 @@
#define CLIENT_INIT_TX_DATA_BCAST_ACCEPT_ALL_SHIFT 2
#define CLIENT_INIT_TX_DATA_ACCEPT_ANY_VLAN (0x1<<3)
#define CLIENT_INIT_TX_DATA_ACCEPT_ANY_VLAN_SHIFT 3
-#define CLIENT_INIT_TX_DATA_RESERVED1 (0xFFF<<4)
-#define CLIENT_INIT_TX_DATA_RESERVED1_SHIFT 4
+#define CLIENT_INIT_TX_DATA_RESERVED0 (0xFFF<<4)
+#define CLIENT_INIT_TX_DATA_RESERVED0_SHIFT 4
u8 default_vlan_flg;
u8 force_default_pri_flg;
- __le32 reserved3;
+ u8 tunnel_lso_inc_ip_id;
+ u8 refuse_outband_vlan_flg;
+ u8 tunnel_non_lso_pcsum_location;
+ u8 reserved1;
};
/*
@@ -3551,6 +3573,11 @@
__le16 silent_vlan_mask;
u8 silent_vlan_removal_flg;
u8 silent_vlan_change_flg;
+ u8 refuse_outband_vlan_flg;
+ u8 refuse_outband_vlan_change_flg;
+ u8 tx_switching_flg;
+ u8 tx_switching_change_flg;
+ __le32 reserved1;
__le32 echo;
};
@@ -3620,7 +3647,8 @@
*/
struct eth_classify_mac_cmd {
struct eth_classify_cmd_header header;
- __le32 reserved0;
+ __le16 reserved0;
+ __le16 inner_mac;
__le16 mac_lsb;
__le16 mac_mid;
__le16 mac_msb;
@@ -3633,7 +3661,8 @@
*/
struct eth_classify_pair_cmd {
struct eth_classify_cmd_header header;
- __le32 reserved0;
+ __le16 reserved0;
+ __le16 inner_mac;
__le16 mac_lsb;
__le16 mac_mid;
__le16 mac_msb;
@@ -3855,8 +3884,68 @@
/*
- * Command for setting multicast classification for a client
+ * destination and source mac address.
*/
+struct eth_mac_addresses {
+#if defined(__BIG_ENDIAN)
+ __le16 dst_mid;
+ __le16 dst_lo;
+#elif defined(__LITTLE_ENDIAN)
+ __le16 dst_lo;
+ __le16 dst_mid;
+#endif
+#if defined(__BIG_ENDIAN)
+ __le16 src_lo;
+ __le16 dst_hi;
+#elif defined(__LITTLE_ENDIAN)
+ __le16 dst_hi;
+ __le16 src_lo;
+#endif
+#if defined(__BIG_ENDIAN)
+ __le16 src_hi;
+ __le16 src_mid;
+#elif defined(__LITTLE_ENDIAN)
+ __le16 src_mid;
+ __le16 src_hi;
+#endif
+};
+
+/* tunneling related data */
+struct eth_tunnel_data {
+#if defined(__BIG_ENDIAN)
+ __le16 dst_mid;
+ __le16 dst_lo;
+#elif defined(__LITTLE_ENDIAN)
+ __le16 dst_lo;
+ __le16 dst_mid;
+#endif
+#if defined(__BIG_ENDIAN)
+ __le16 reserved0;
+ __le16 dst_hi;
+#elif defined(__LITTLE_ENDIAN)
+ __le16 dst_hi;
+ __le16 reserved0;
+#endif
+#if defined(__BIG_ENDIAN)
+ u8 reserved1;
+ u8 ip_hdr_start_inner_w;
+ __le16 pseudo_csum;
+#elif defined(__LITTLE_ENDIAN)
+ __le16 pseudo_csum;
+ u8 ip_hdr_start_inner_w;
+ u8 reserved1;
+#endif
+};
+
+/* union for mac addresses and for tunneling data.
+ * considered as tunneling data only if (tunnel_exist == 1).
+ */
+union eth_mac_addr_or_tunnel_data {
+ struct eth_mac_addresses mac_addr;
+ struct eth_tunnel_data tunnel_data;
+};
+
+/*Command for setting multicast classification for a client */
struct eth_multicast_rules_cmd {
u8 cmd_general_data;
#define ETH_MULTICAST_RULES_CMD_RX_CMD (0x1<<0)
@@ -3874,7 +3963,6 @@
struct regpair reserved3;
};
-
/*
* parameters for multicast classification ramrod
*/
@@ -3883,7 +3971,6 @@
struct eth_multicast_rules_cmd rules[MULTICAST_RULES_COUNT];
};
-
/*
* Place holder for ramrods protocol specific data
*/
@@ -3947,11 +4034,14 @@
#define ETH_RSS_UPDATE_RAMROD_DATA_IPV6_TCP_CAPABILITY_SHIFT 4
#define ETH_RSS_UPDATE_RAMROD_DATA_IPV6_UDP_CAPABILITY (0x1<<5)
#define ETH_RSS_UPDATE_RAMROD_DATA_IPV6_UDP_CAPABILITY_SHIFT 5
+#define ETH_RSS_UPDATE_RAMROD_DATA_EN_5_TUPLE_CAPABILITY (0x1<<6)
+#define ETH_RSS_UPDATE_RAMROD_DATA_EN_5_TUPLE_CAPABILITY_SHIFT 6
#define ETH_RSS_UPDATE_RAMROD_DATA_UPDATE_RSS_KEY (0x1<<7)
#define ETH_RSS_UPDATE_RAMROD_DATA_UPDATE_RSS_KEY_SHIFT 7
u8 rss_result_mask;
u8 rss_mode;
- __le32 __reserved2;
+ __le16 udp_4tuple_dst_port_mask;
+ __le16 udp_4tuple_dst_port_value;
u8 indirection_table[T_ETH_INDIRECTION_TABLE_SIZE];
__le32 rss_key[T_ETH_RSS_KEY];
__le32 echo;
@@ -4115,6 +4205,23 @@
MAX_ETH_TPA_UPDATE_COMMAND
};
+/* In case of LSO over IPv4 tunnel, whether to increment
+ * IP ID on external IP header or internal IP header
+ */
+enum eth_tunnel_lso_inc_ip_id {
+ EXT_HEADER,
+ INT_HEADER,
+ MAX_ETH_TUNNEL_LSO_INC_IP_ID
+};
+
+/* In case tunnel exist and L4 checksum offload,
+ * the pseudo checksum location, on packet or on BD.
+ */
+enum eth_tunnel_non_lso_pcsum_location {
+ PCSUM_ON_PKT,
+ PCSUM_ON_BD,
+ MAX_ETH_TUNNEL_NON_LSO_PCSUM_LOCATION
+};
/*
* Tx regular BD structure
@@ -4166,8 +4273,8 @@
#define ETH_TX_START_BD_FORCE_VLAN_MODE_SHIFT 4
#define ETH_TX_START_BD_PARSE_NBDS (0x3<<5)
#define ETH_TX_START_BD_PARSE_NBDS_SHIFT 5
-#define ETH_TX_START_BD_RESREVED (0x1<<7)
-#define ETH_TX_START_BD_RESREVED_SHIFT 7
+#define ETH_TX_START_BD_TUNNEL_EXIST (0x1<<7)
+#define ETH_TX_START_BD_TUNNEL_EXIST_SHIFT 7
};
/*
@@ -4216,15 +4323,10 @@
* Tx parsing BD structure for ETH E2
*/
struct eth_tx_parse_bd_e2 {
- __le16 dst_mac_addr_lo;
- __le16 dst_mac_addr_mid;
- __le16 dst_mac_addr_hi;
- __le16 src_mac_addr_lo;
- __le16 src_mac_addr_mid;
- __le16 src_mac_addr_hi;
+ union eth_mac_addr_or_tunnel_data data;
__le32 parsing_data;
-#define ETH_TX_PARSE_BD_E2_TCP_HDR_START_OFFSET_W (0x7FF<<0)
-#define ETH_TX_PARSE_BD_E2_TCP_HDR_START_OFFSET_W_SHIFT 0
+#define ETH_TX_PARSE_BD_E2_L4_HDR_START_OFFSET_W (0x7FF<<0)
+#define ETH_TX_PARSE_BD_E2_L4_HDR_START_OFFSET_W_SHIFT 0
#define ETH_TX_PARSE_BD_E2_TCP_HDR_LENGTH_DW (0xF<<11)
#define ETH_TX_PARSE_BD_E2_TCP_HDR_LENGTH_DW_SHIFT 11
#define ETH_TX_PARSE_BD_E2_IPV6_WITH_EXT_HDR (0x1<<15)
@@ -4236,8 +4338,51 @@
};
/*
- * The last BD in the BD memory will hold a pointer to the next BD memory
+ * Tx 2nd parsing BD structure for ETH packet
*/
+struct eth_tx_parse_2nd_bd {
+ __le16 global_data;
+#define ETH_TX_PARSE_2ND_BD_IP_HDR_START_OUTER_W (0xF<<0)
+#define ETH_TX_PARSE_2ND_BD_IP_HDR_START_OUTER_W_SHIFT 0
+#define ETH_TX_PARSE_2ND_BD_IP_HDR_TYPE_OUTER (0x1<<4)
+#define ETH_TX_PARSE_2ND_BD_IP_HDR_TYPE_OUTER_SHIFT 4
+#define ETH_TX_PARSE_2ND_BD_LLC_SNAP_EN (0x1<<5)
+#define ETH_TX_PARSE_2ND_BD_LLC_SNAP_EN_SHIFT 5
+#define ETH_TX_PARSE_2ND_BD_NS_FLG (0x1<<6)
+#define ETH_TX_PARSE_2ND_BD_NS_FLG_SHIFT 6
+#define ETH_TX_PARSE_2ND_BD_TUNNEL_UDP_EXIST (0x1<<7)
+#define ETH_TX_PARSE_2ND_BD_TUNNEL_UDP_EXIST_SHIFT 7
+#define ETH_TX_PARSE_2ND_BD_IP_HDR_LEN_OUTER_W (0x1F<<8)
+#define ETH_TX_PARSE_2ND_BD_IP_HDR_LEN_OUTER_W_SHIFT 8
+#define ETH_TX_PARSE_2ND_BD_RESERVED0 (0x7<<13)
+#define ETH_TX_PARSE_2ND_BD_RESERVED0_SHIFT 13
+ __le16 reserved1;
+ u8 tcp_flags;
+#define ETH_TX_PARSE_2ND_BD_FIN_FLG (0x1<<0)
+#define ETH_TX_PARSE_2ND_BD_FIN_FLG_SHIFT 0
+#define ETH_TX_PARSE_2ND_BD_SYN_FLG (0x1<<1)
+#define ETH_TX_PARSE_2ND_BD_SYN_FLG_SHIFT 1
+#define ETH_TX_PARSE_2ND_BD_RST_FLG (0x1<<2)
+#define ETH_TX_PARSE_2ND_BD_RST_FLG_SHIFT 2
+#define ETH_TX_PARSE_2ND_BD_PSH_FLG (0x1<<3)
+#define ETH_TX_PARSE_2ND_BD_PSH_FLG_SHIFT 3
+#define ETH_TX_PARSE_2ND_BD_ACK_FLG (0x1<<4)
+#define ETH_TX_PARSE_2ND_BD_ACK_FLG_SHIFT 4
+#define ETH_TX_PARSE_2ND_BD_URG_FLG (0x1<<5)
+#define ETH_TX_PARSE_2ND_BD_URG_FLG_SHIFT 5
+#define ETH_TX_PARSE_2ND_BD_ECE_FLG (0x1<<6)
+#define ETH_TX_PARSE_2ND_BD_ECE_FLG_SHIFT 6
+#define ETH_TX_PARSE_2ND_BD_CWR_FLG (0x1<<7)
+#define ETH_TX_PARSE_2ND_BD_CWR_FLG_SHIFT 7
+ u8 reserved2;
+ u8 tunnel_udp_hdr_start_w;
+ u8 fw_ip_hdr_to_payload_w;
+ __le16 fw_ip_csum_wo_len_flags_frag;
+ __le16 hw_ip_id;
+ __le32 tcp_send_seq;
+};
+
+/* The last BD in the BD memory will hold a pointer to the next BD memory */
struct eth_tx_next_bd {
__le32 addr_lo;
__le32 addr_hi;
@@ -4252,6 +4397,7 @@
struct eth_tx_bd reg_bd;
struct eth_tx_parse_bd_e1x parse_bd_e1x;
struct eth_tx_parse_bd_e2 parse_bd_e2;
+ struct eth_tx_parse_2nd_bd parse_2nd_bd;
struct eth_tx_next_bd next_bd;
};
@@ -4663,10 +4809,10 @@
RAMROD_CMD_ID_COMMON_STOP_TRAFFIC,
RAMROD_CMD_ID_COMMON_START_TRAFFIC,
RAMROD_CMD_ID_COMMON_AFEX_VIF_LISTS,
+ RAMROD_CMD_ID_COMMON_SET_TIMESYNC,
MAX_COMMON_SPQE_CMD_ID
};
-
/*
* Per-protocol connection types
*/
@@ -4863,7 +5009,7 @@
*/
struct malicious_vf_event_data {
u8 vf_id;
- u8 reserved0;
+ u8 err_id;
u16 reserved1;
u32 reserved2;
u32 reserved3;
@@ -4969,10 +5115,10 @@
EVENT_RING_OPCODE_CLASSIFICATION_RULES,
EVENT_RING_OPCODE_FILTERS_RULES,
EVENT_RING_OPCODE_MULTICAST_RULES,
+ EVENT_RING_OPCODE_SET_TIMESYNC,
MAX_EVENT_RING_OPCODE
};
-
/*
* Modes for fairness algorithm
*/
@@ -5010,14 +5156,18 @@
*/
struct function_start_data {
u8 function_mode;
- u8 reserved;
+ u8 allow_npar_tx_switching;
__le16 sd_vlan_tag;
__le16 vif_id;
u8 path_id;
u8 network_cos_mode;
+ u8 dmae_cmd_id;
+ u8 gre_tunnel_mode;
+ u8 gre_tunnel_rss;
+ u8 nvgre_clss_en;
+ __le16 reserved1[2];
};
-
struct function_update_data {
u8 vif_id_change_flg;
u8 afex_default_vlan_change_flg;
@@ -5027,14 +5177,19 @@
__le16 afex_default_vlan;
u8 allowed_priorities;
u8 network_cos_mode;
+ u8 lb_mode_en_change_flg;
u8 lb_mode_en;
u8 tx_switch_suspend_change_flg;
u8 tx_switch_suspend;
u8 echo;
- __le16 reserved1;
+ u8 reserved1;
+ u8 update_gre_cfg_flg;
+ u8 gre_tunnel_mode;
+ u8 gre_tunnel_rss;
+ u8 nvgre_clss_en;
+ u32 reserved3;
};
-
/*
* FW version stored in the Xstorm RAM
*/
@@ -5061,6 +5216,22 @@
#define __FW_VERSION_RESERVED_SHIFT 4
};
+/* GRE RSS Mode */
+enum gre_rss_mode {
+ GRE_OUTER_HEADERS_RSS,
+ GRE_INNER_HEADERS_RSS,
+ NVGRE_KEY_ENTROPY_RSS,
+ MAX_GRE_RSS_MODE
+};
+
+/* GRE Tunnel Mode */
+enum gre_tunnel_type {
+ NO_GRE_TUNNEL,
+ NVGRE_TUNNEL,
+ L2GRE_TUNNEL,
+ IPGRE_TUNNEL,
+ MAX_GRE_TUNNEL_TYPE
+};
/*
* Dynamic Host-Coalescing - Driver(host) counters
@@ -5224,6 +5395,26 @@
MAX_IP_VER
};
+/*
+ * Malicious VF error ID
+ */
+enum malicious_vf_error_id {
+ VF_PF_CHANNEL_NOT_READY,
+ ETH_ILLEGAL_BD_LENGTHS,
+ ETH_PACKET_TOO_SHORT,
+ ETH_PAYLOAD_TOO_BIG,
+ ETH_ILLEGAL_ETH_TYPE,
+ ETH_ILLEGAL_LSO_HDR_LEN,
+ ETH_TOO_MANY_BDS,
+ ETH_ZERO_HDR_NBDS,
+ ETH_START_BD_NOT_SET,
+ ETH_ILLEGAL_PARSE_NBDS,
+ ETH_IPV6_AND_CHECKSUM,
+ ETH_VLAN_FLG_INCORRECT,
+ ETH_ILLEGAL_LSO_MSS,
+ ETH_TUNNEL_NOT_SUPPORTED,
+ MAX_MALICIOUS_VF_ERROR_ID
+};
/*
* Multi-function modes
@@ -5368,7 +5559,6 @@
union protocol_common_specific_data data;
};
-
/*
* The send queue element
*/
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c
index 77ebae0..e3fa808 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c
@@ -152,6 +152,7 @@
#define SFP_EEPROM_CON_TYPE_ADDR 0x2
#define SFP_EEPROM_CON_TYPE_VAL_LC 0x7
#define SFP_EEPROM_CON_TYPE_VAL_COPPER 0x21
+ #define SFP_EEPROM_CON_TYPE_VAL_RJ45 0x22
#define SFP_EEPROM_COMP_CODE_ADDR 0x3
@@ -3629,6 +3630,16 @@
* init configuration, and set/clear SGMII flag. Internal
* phy init is done purely in phy_init stage.
*/
+#define WC_TX_DRIVER(post2, idriver, ipre) \
+ ((post2 << MDIO_WC_REG_TX0_TX_DRIVER_POST2_COEFF_OFFSET) | \
+ (idriver << MDIO_WC_REG_TX0_TX_DRIVER_IDRIVER_OFFSET) | \
+ (ipre << MDIO_WC_REG_TX0_TX_DRIVER_IPRE_DRIVER_OFFSET))
+
+#define WC_TX_FIR(post, main, pre) \
+ ((post << MDIO_WC_REG_TX_FIR_TAP_POST_TAP_OFFSET) | \
+ (main << MDIO_WC_REG_TX_FIR_TAP_MAIN_TAP_OFFSET) | \
+ (pre << MDIO_WC_REG_TX_FIR_TAP_PRE_TAP_OFFSET))
+
static void bnx2x_warpcore_enable_AN_KR2(struct bnx2x_phy *phy,
struct link_params *params,
struct link_vars *vars)
@@ -3753,20 +3764,13 @@
/* Set Transmit PMD settings */
lane = bnx2x_get_warpcore_lane(phy, params);
bnx2x_cl45_write(bp, phy, MDIO_WC_DEVAD,
- MDIO_WC_REG_TX0_TX_DRIVER + 0x10*lane,
- ((0x02 << MDIO_WC_REG_TX0_TX_DRIVER_POST2_COEFF_OFFSET) |
- (0x06 << MDIO_WC_REG_TX0_TX_DRIVER_IDRIVER_OFFSET) |
- (0x09 << MDIO_WC_REG_TX0_TX_DRIVER_IPRE_DRIVER_OFFSET)));
+ MDIO_WC_REG_TX0_TX_DRIVER + 0x10*lane,
+ WC_TX_DRIVER(0x02, 0x06, 0x09));
/* Configure the next lane if dual mode */
if (phy->flags & FLAGS_WC_DUAL_MODE)
bnx2x_cl45_write(bp, phy, MDIO_WC_DEVAD,
MDIO_WC_REG_TX0_TX_DRIVER + 0x10*(lane+1),
- ((0x02 <<
- MDIO_WC_REG_TX0_TX_DRIVER_POST2_COEFF_OFFSET) |
- (0x06 <<
- MDIO_WC_REG_TX0_TX_DRIVER_IDRIVER_OFFSET) |
- (0x09 <<
- MDIO_WC_REG_TX0_TX_DRIVER_IPRE_DRIVER_OFFSET)));
+ WC_TX_DRIVER(0x02, 0x06, 0x09));
bnx2x_cl45_write(bp, phy, MDIO_WC_DEVAD,
MDIO_WC_REG_CL72_USERB0_CL72_OS_DEF_CTRL,
0x03f0);
@@ -3909,6 +3913,8 @@
{
struct bnx2x *bp = params->bp;
u16 misc1_val, tap_val, tx_driver_val, lane, val;
+ u32 cfg_tap_val, tx_drv_brdct, tx_equal;
+
/* Hold rxSeqStart */
bnx2x_cl45_read_or_write(bp, phy, MDIO_WC_DEVAD,
MDIO_WC_REG_DSC2B0_DSC_MISC_CTRL0, 0x8000);
@@ -3952,23 +3958,33 @@
if (is_xfi) {
misc1_val |= 0x5;
- tap_val = ((0x08 << MDIO_WC_REG_TX_FIR_TAP_POST_TAP_OFFSET) |
- (0x37 << MDIO_WC_REG_TX_FIR_TAP_MAIN_TAP_OFFSET) |
- (0x00 << MDIO_WC_REG_TX_FIR_TAP_PRE_TAP_OFFSET));
- tx_driver_val =
- ((0x00 << MDIO_WC_REG_TX0_TX_DRIVER_POST2_COEFF_OFFSET) |
- (0x02 << MDIO_WC_REG_TX0_TX_DRIVER_IDRIVER_OFFSET) |
- (0x03 << MDIO_WC_REG_TX0_TX_DRIVER_IPRE_DRIVER_OFFSET));
-
+ tap_val = WC_TX_FIR(0x08, 0x37, 0x00);
+ tx_driver_val = WC_TX_DRIVER(0x00, 0x02, 0x03);
} else {
+ cfg_tap_val = REG_RD(bp, params->shmem_base +
+ offsetof(struct shmem_region, dev_info.
+ port_hw_config[params->port].
+ sfi_tap_values));
+
+ tx_equal = cfg_tap_val & PORT_HW_CFG_TX_EQUALIZATION_MASK;
+
+ tx_drv_brdct = (cfg_tap_val &
+ PORT_HW_CFG_TX_DRV_BROADCAST_MASK) >>
+ PORT_HW_CFG_TX_DRV_BROADCAST_SHIFT;
+
misc1_val |= 0x9;
- tap_val = ((0x0f << MDIO_WC_REG_TX_FIR_TAP_POST_TAP_OFFSET) |
- (0x2b << MDIO_WC_REG_TX_FIR_TAP_MAIN_TAP_OFFSET) |
- (0x02 << MDIO_WC_REG_TX_FIR_TAP_PRE_TAP_OFFSET));
- tx_driver_val =
- ((0x03 << MDIO_WC_REG_TX0_TX_DRIVER_POST2_COEFF_OFFSET) |
- (0x02 << MDIO_WC_REG_TX0_TX_DRIVER_IDRIVER_OFFSET) |
- (0x06 << MDIO_WC_REG_TX0_TX_DRIVER_IPRE_DRIVER_OFFSET));
+
+ /* TAP values are controlled by nvram, if value there isn't 0 */
+ if (tx_equal)
+ tap_val = (u16)tx_equal;
+ else
+ tap_val = WC_TX_FIR(0x0f, 0x2b, 0x02);
+
+ if (tx_drv_brdct)
+ tx_driver_val = WC_TX_DRIVER(0x03, (u16)tx_drv_brdct,
+ 0x06);
+ else
+ tx_driver_val = WC_TX_DRIVER(0x03, 0x02, 0x06);
}
bnx2x_cl45_write(bp, phy, MDIO_WC_DEVAD,
MDIO_WC_REG_SERDESDIGITAL_MISC1, misc1_val);
@@ -4105,15 +4121,11 @@
/* Set Transmit PMD settings */
bnx2x_cl45_write(bp, phy, MDIO_WC_DEVAD,
MDIO_WC_REG_TX_FIR_TAP,
- ((0x12 << MDIO_WC_REG_TX_FIR_TAP_POST_TAP_OFFSET) |
- (0x2d << MDIO_WC_REG_TX_FIR_TAP_MAIN_TAP_OFFSET) |
- (0x00 << MDIO_WC_REG_TX_FIR_TAP_PRE_TAP_OFFSET) |
- MDIO_WC_REG_TX_FIR_TAP_ENABLE));
+ (WC_TX_FIR(0x12, 0x2d, 0x00) |
+ MDIO_WC_REG_TX_FIR_TAP_ENABLE));
bnx2x_cl45_write(bp, phy, MDIO_WC_DEVAD,
- MDIO_WC_REG_TX0_TX_DRIVER + 0x10*lane,
- ((0x02 << MDIO_WC_REG_TX0_TX_DRIVER_POST2_COEFF_OFFSET) |
- (0x02 << MDIO_WC_REG_TX0_TX_DRIVER_IDRIVER_OFFSET) |
- (0x02 << MDIO_WC_REG_TX0_TX_DRIVER_IPRE_DRIVER_OFFSET)));
+ MDIO_WC_REG_TX0_TX_DRIVER + 0x10*lane,
+ WC_TX_DRIVER(0x02, 0x02, 0x02));
}
static void bnx2x_warpcore_set_sgmii_speed(struct bnx2x_phy *phy,
@@ -8049,20 +8061,24 @@
break;
}
case SFP_EEPROM_CON_TYPE_VAL_LC:
+ case SFP_EEPROM_CON_TYPE_VAL_RJ45:
check_limiting_mode = 1;
if ((val[1] & (SFP_EEPROM_COMP_CODE_SR_MASK |
SFP_EEPROM_COMP_CODE_LR_MASK |
SFP_EEPROM_COMP_CODE_LRM_MASK)) == 0) {
- DP(NETIF_MSG_LINK, "1G Optic module detected\n");
+ DP(NETIF_MSG_LINK, "1G SFP module detected\n");
gport = params->port;
phy->media_type = ETH_PHY_SFP_1G_FIBER;
- phy->req_line_speed = SPEED_1000;
- if (!CHIP_IS_E1x(bp))
- gport = BP_PATH(bp) + (params->port << 1);
- netdev_err(bp->dev, "Warning: Link speed was forced to 1000Mbps."
- " Current SFP module in port %d is not"
- " compliant with 10G Ethernet\n",
- gport);
+ if (phy->req_line_speed != SPEED_1000) {
+ phy->req_line_speed = SPEED_1000;
+ if (!CHIP_IS_E1x(bp)) {
+ gport = BP_PATH(bp) +
+ (params->port << 1);
+ }
+ netdev_err(bp->dev,
+ "Warning: Link speed was forced to 1000Mbps. Current SFP module in port %d is not compliant with 10G Ethernet\n",
+ gport);
+ }
} else {
int idx, cfg_idx = 0;
DP(NETIF_MSG_LINK, "10G Optic module detected\n");
@@ -10281,7 +10297,8 @@
LINK_STATUS_LINK_PARTNER_10GXFD_CAPABLE;
/* Determine if EEE was negotiated */
- if (phy->type == PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM84833)
+ if ((phy->type == PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM84833) ||
+ (phy->type == PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM84834))
bnx2x_eee_an_resolve(phy, params, vars);
}
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index e81a747..4902d1e 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -75,8 +75,6 @@
#define FW_FILE_NAME_E1H "bnx2x/bnx2x-e1h-" FW_FILE_VERSION ".fw"
#define FW_FILE_NAME_E2 "bnx2x/bnx2x-e2-" FW_FILE_VERSION ".fw"
-#define MAC_LEADING_ZERO_CNT (ALIGN(ETH_ALEN, sizeof(u32)) - ETH_ALEN)
-
/* Time in jiffies before concluding the transmitter is hung */
#define TX_TIMEOUT (5*HZ)
@@ -2955,14 +2953,15 @@
__set_bit(BNX2X_Q_FLG_ACTIVE, &flags);
/* tx only connections collect statistics (on the same index as the
- * parent connection). The statistics are zeroed when the parent
- * connection is initialized.
+ * parent connection). The statistics are zeroed when the parent
+ * connection is initialized.
*/
__set_bit(BNX2X_Q_FLG_STATS, &flags);
if (zero_stats)
__set_bit(BNX2X_Q_FLG_ZERO_STATS, &flags);
+ __set_bit(BNX2X_Q_FLG_PCSUM_ON_PKT, &flags);
#ifdef BNX2X_STOP_ON_ERROR
__set_bit(BNX2X_Q_FLG_TX_SEC, &flags);
@@ -3227,16 +3226,29 @@
{
struct eth_stats_info *ether_stat =
&bp->slowpath->drv_info_to_mcp.ether_stat;
+ struct bnx2x_vlan_mac_obj *mac_obj =
+ &bp->sp_objs->mac_obj;
+ int i;
strlcpy(ether_stat->version, DRV_MODULE_VERSION,
ETH_STAT_INFO_VERSION_LEN);
- bp->sp_objs[0].mac_obj.get_n_elements(bp, &bp->sp_objs[0].mac_obj,
- DRV_INFO_ETH_STAT_NUM_MACS_REQUIRED,
- ether_stat->mac_local);
-
+ /* get DRV_INFO_ETH_STAT_NUM_MACS_REQUIRED macs, placing them in the
+ * mac_local field in ether_stat struct. The base address is offset by 2
+ * bytes to account for the field being 8 bytes but a mac address is
+ * only 6 bytes. Likewise, the stride for the get_n_elements function is
+ * 2 bytes to compensate from the 6 bytes of a mac to the 8 bytes
+ * allocated by the ether_stat struct, so the macs will land in their
+ * proper positions.
+ */
+ for (i = 0; i < DRV_INFO_ETH_STAT_NUM_MACS_REQUIRED; i++)
+ memset(ether_stat->mac_local + i, 0,
+ sizeof(ether_stat->mac_local[0]));
+ mac_obj->get_n_elements(bp, &bp->sp_objs[0].mac_obj,
+ DRV_INFO_ETH_STAT_NUM_MACS_REQUIRED,
+ ether_stat->mac_local + MAC_PAD, MAC_PAD,
+ ETH_ALEN);
ether_stat->mtu_size = bp->dev->mtu;
-
if (bp->dev->features & NETIF_F_RXCSUM)
ether_stat->feature_flags |= FEATURE_ETH_CHKSUM_OFFLOAD_MASK;
if (bp->dev->features & NETIF_F_TSO)
@@ -3258,8 +3270,7 @@
if (!CNIC_LOADED(bp))
return;
- memcpy(fcoe_stat->mac_local + MAC_LEADING_ZERO_CNT,
- bp->fip_mac, ETH_ALEN);
+ memcpy(fcoe_stat->mac_local + MAC_PAD, bp->fip_mac, ETH_ALEN);
fcoe_stat->qos_priority =
app->traffic_type_priority[LLFC_TRAFFIC_TYPE_FCOE];
@@ -3361,8 +3372,8 @@
if (!CNIC_LOADED(bp))
return;
- memcpy(iscsi_stat->mac_local + MAC_LEADING_ZERO_CNT,
- bp->cnic_eth_dev.iscsi_mac, ETH_ALEN);
+ memcpy(iscsi_stat->mac_local + MAC_PAD, bp->cnic_eth_dev.iscsi_mac,
+ ETH_ALEN);
iscsi_stat->qos_priority =
app->traffic_type_priority[LLFC_TRAFFIC_TYPE_ISCSI];
@@ -9525,6 +9536,10 @@
bnx2x_vfpf_storm_rx_mode(bp);
}
+ if (test_and_clear_bit(BNX2X_SP_RTNL_HYPERVISOR_VLAN,
+ &bp->sp_rtnl_state))
+ bnx2x_pf_set_vfs_vlan(bp);
+
/* work which needs rtnl lock not-taken (as it takes the lock itself and
* can be called from other contexts as well)
*/
@@ -9532,8 +9547,10 @@
/* enable SR-IOV if applicable */
if (IS_SRIOV(bp) && test_and_clear_bit(BNX2X_SP_RTNL_ENABLE_SRIOV,
- &bp->sp_rtnl_state))
+ &bp->sp_rtnl_state)) {
+ bnx2x_disable_sriov(bp);
bnx2x_enable_sriov(bp);
+ }
}
static void bnx2x_period_task(struct work_struct *work)
@@ -10034,8 +10051,12 @@
id = ((val & 0xffff) << 16);
val = REG_RD(bp, MISC_REG_CHIP_REV);
id |= ((val & 0xf) << 12);
- val = REG_RD(bp, MISC_REG_CHIP_METAL);
- id |= ((val & 0xff) << 4);
+
+ /* Metal is read from PCI regs, but we can't access >=0x400 from
+ * the configuration space (so we need to reg_rd)
+ */
+ val = REG_RD(bp, PCICFG_OFFSET + PCI_ID_VAL3);
+ id |= (((val >> 24) & 0xf) << 4);
val = REG_RD(bp, MISC_REG_BOND_ID);
id |= (val & 0xf);
bp->common.chip_id = id;
@@ -10812,14 +10833,12 @@
}
}
- if (IS_MF_STORAGE_SD(bp))
- /* Zero primary MAC configuration */
- memset(bp->dev->dev_addr, 0, ETH_ALEN);
-
- if (IS_MF_FCOE_AFEX(bp) || IS_MF_FCOE_SD(bp))
- /* use FIP MAC as primary MAC */
+ /* If this is a storage-only interface, use SAN mac as
+ * primary MAC. Notice that for SD this is already the case,
+ * as the SAN mac was copied from the primary MAC.
+ */
+ if (IS_MF_FCOE_AFEX(bp))
memcpy(bp->dev->dev_addr, fip_mac, ETH_ALEN);
-
} else {
val2 = SHMEM_RD(bp, dev_info.port_hw_config[port].
iscsi_mac_upper);
@@ -11056,6 +11075,9 @@
} else
BNX2X_DEV_INFO("illegal OV for SD\n");
break;
+ case SHARED_FEAT_CFG_FORCE_SF_MODE_FORCED_SF:
+ bp->mf_config[vn] = 0;
+ break;
default:
/* Unknown configuration: reset mf_config */
bp->mf_config[vn] = 0;
@@ -11402,26 +11424,6 @@
* net_device service functions
*/
-static int bnx2x_open_epilog(struct bnx2x *bp)
-{
- /* Enable sriov via delayed work. This must be done via delayed work
- * because it causes the probe of the vf devices to be run, which invoke
- * register_netdevice which must have rtnl lock taken. As we are holding
- * the lock right now, that could only work if the probe would not take
- * the lock. However, as the probe of the vf may be called from other
- * contexts as well (such as passthrough to vm failes) it can't assume
- * the lock is being held for it. Using delayed work here allows the
- * probe code to simply take the lock (i.e. wait for it to be released
- * if it is being held).
- */
- smp_mb__before_clear_bit();
- set_bit(BNX2X_SP_RTNL_ENABLE_SRIOV, &bp->sp_rtnl_state);
- smp_mb__after_clear_bit();
- schedule_delayed_work(&bp->sp_rtnl_task, 0);
-
- return 0;
-}
-
/* called with rtnl_lock */
static int bnx2x_open(struct net_device *dev)
{
@@ -11791,6 +11793,8 @@
.ndo_setup_tc = bnx2x_setup_tc,
#ifdef CONFIG_BNX2X_SRIOV
.ndo_set_vf_mac = bnx2x_set_vf_mac,
+ .ndo_set_vf_vlan = bnx2x_set_vf_vlan,
+ .ndo_get_vf_config = bnx2x_get_vf_config,
#endif
#ifdef NETDEV_FCOE_WWNN
.ndo_fcoe_get_wwn = bnx2x_fcoe_get_wwn,
@@ -11953,7 +11957,7 @@
dev->watchdog_timeo = TX_TIMEOUT;
dev->netdev_ops = &bnx2x_netdev_ops;
- bnx2x_set_ethtool_ops(dev);
+ bnx2x_set_ethtool_ops(bp, dev);
dev->priv_flags |= IFF_UNICAST_FLT;
@@ -11961,6 +11965,13 @@
NETIF_F_TSO | NETIF_F_TSO_ECN | NETIF_F_TSO6 |
NETIF_F_RXCSUM | NETIF_F_LRO | NETIF_F_GRO |
NETIF_F_RXHASH | NETIF_F_HW_VLAN_TX;
+ if (!CHIP_IS_E1x(bp)) {
+ dev->hw_features |= NETIF_F_GSO_GRE;
+ dev->hw_enc_features =
+ NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_SG |
+ NETIF_F_TSO | NETIF_F_TSO_ECN | NETIF_F_TSO6 |
+ NETIF_F_GSO_GRE;
+ }
dev->vlan_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
NETIF_F_TSO | NETIF_F_TSO_ECN | NETIF_F_TSO6 | NETIF_F_HIGHDMA;
@@ -12475,13 +12486,8 @@
goto init_one_exit;
}
- /* Enable SRIOV if capability found in configuration space.
- * Once the generic SR-IOV framework makes it in from the
- * pci tree this will be revised, to allow dynamic control
- * over the number of VFs. Right now, change the num of vfs
- * param below to enable SR-IOV.
- */
- rc = bnx2x_iov_init_one(bp, int_mode, 0/*num vfs*/);
+ /* Enable SRIOV if capability found in configuration space */
+ rc = bnx2x_iov_init_one(bp, int_mode, BNX2X_MAX_NUM_OF_VFS);
if (rc)
goto init_one_exit;
@@ -12493,16 +12499,6 @@
if (CHIP_IS_E1x(bp))
bp->flags |= NO_FCOE_FLAG;
- /* disable FCOE for 57840 device, until FW supports it */
- switch (ent->driver_data) {
- case BCM57840_O:
- case BCM57840_4_10:
- case BCM57840_2_20:
- case BCM57840_MFO:
- case BCM57840_MF:
- bp->flags |= NO_FCOE_FLAG;
- }
-
/* Set bp->num_queues for MSI-X mode*/
bnx2x_set_num_queues(bp);
@@ -12797,6 +12793,9 @@
.suspend = bnx2x_suspend,
.resume = bnx2x_resume,
.err_handler = &bnx2x_err_handler,
+#ifdef CONFIG_BNX2X_SRIOV
+ .sriov_configure = bnx2x_sriov_configure,
+#endif
};
static int __init bnx2x_init(void)
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
index 791eb2d..d22bc40 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
@@ -1491,10 +1491,6 @@
/* [R 4] This field indicates the type of the device. '0' - 2 Ports; '1' - 1
Port. */
#define MISC_REG_BOND_ID 0xa400
-/* [R 8] These bits indicate the metal revision of the chip. This value
- starts at 0x00 for each all-layer tape-out and increments by one for each
- tape-out. */
-#define MISC_REG_CHIP_METAL 0xa404
/* [R 16] These bits indicate the part number for the chip. */
#define MISC_REG_CHIP_NUM 0xa408
/* [R 4] These bits indicate the base revision of the chip. This value
@@ -6331,6 +6327,8 @@
#define PCI_PM_DATA_B 0x414
#define PCI_ID_VAL1 0x434
#define PCI_ID_VAL2 0x438
+#define PCI_ID_VAL3 0x43c
+
#define GRC_CONFIG_REG_PF_INIT_VF 0x624
#define GRC_CR_PF_INIT_VF_PF_FIRST_VF_NUM_MASK 0xf
/* First VF_NUM for PF is encoded in this register.
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c
index 7306416..5bdc1d6 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c
@@ -30,8 +30,6 @@
#define BNX2X_MAX_EMUL_MULTI 16
-#define MAC_LEADING_ZERO_CNT (ALIGN(ETH_ALEN, sizeof(u32)) - ETH_ALEN)
-
/**** Exe Queue interfaces ****/
/**
@@ -444,30 +442,21 @@
}
static int bnx2x_get_n_elements(struct bnx2x *bp, struct bnx2x_vlan_mac_obj *o,
- int n, u8 *buf)
+ int n, u8 *base, u8 stride, u8 size)
{
struct bnx2x_vlan_mac_registry_elem *pos;
- u8 *next = buf;
+ u8 *next = base;
int counter = 0;
/* traverse list */
list_for_each_entry(pos, &o->head, link) {
if (counter < n) {
- /* place leading zeroes in buffer */
- memset(next, 0, MAC_LEADING_ZERO_CNT);
-
- /* place mac after leading zeroes*/
- memcpy(next + MAC_LEADING_ZERO_CNT, pos->u.mac.mac,
- ETH_ALEN);
-
- /* calculate address of next element and
- * advance counter
- */
+ memcpy(next, &pos->u, size);
counter++;
- next = buf + counter * ALIGN(ETH_ALEN, sizeof(u32));
+ DP(BNX2X_MSG_SP, "copied element number %d to address %p element was:\n",
+ counter, next);
+ next += stride + size;
- DP(BNX2X_MSG_SP, "copied element number %d to address %p element was %pM\n",
- counter, next, pos->u.mac.mac);
}
}
return counter * ETH_ALEN;
@@ -487,7 +476,8 @@
/* Check if a requested MAC already exists */
list_for_each_entry(pos, &o->head, link)
- if (!memcmp(data->mac.mac, pos->u.mac.mac, ETH_ALEN))
+ if (!memcmp(data->mac.mac, pos->u.mac.mac, ETH_ALEN) &&
+ (data->mac.is_inner_mac == pos->u.mac.is_inner_mac))
return -EEXIST;
return 0;
@@ -520,7 +510,9 @@
list_for_each_entry(pos, &o->head, link)
if ((data->vlan_mac.vlan == pos->u.vlan_mac.vlan) &&
(!memcmp(data->vlan_mac.mac, pos->u.vlan_mac.mac,
- ETH_ALEN)))
+ ETH_ALEN)) &&
+ (data->vlan_mac.is_inner_mac ==
+ pos->u.vlan_mac.is_inner_mac))
return -EEXIST;
return 0;
@@ -538,7 +530,8 @@
DP(BNX2X_MSG_SP, "Checking MAC %pM for DEL command\n", data->mac.mac);
list_for_each_entry(pos, &o->head, link)
- if (!memcmp(data->mac.mac, pos->u.mac.mac, ETH_ALEN))
+ if ((!memcmp(data->mac.mac, pos->u.mac.mac, ETH_ALEN)) &&
+ (data->mac.is_inner_mac == pos->u.mac.is_inner_mac))
return pos;
return NULL;
@@ -573,7 +566,9 @@
list_for_each_entry(pos, &o->head, link)
if ((data->vlan_mac.vlan == pos->u.vlan_mac.vlan) &&
(!memcmp(data->vlan_mac.mac, pos->u.vlan_mac.mac,
- ETH_ALEN)))
+ ETH_ALEN)) &&
+ (data->vlan_mac.is_inner_mac ==
+ pos->u.vlan_mac.is_inner_mac))
return pos;
return NULL;
@@ -770,6 +765,8 @@
bnx2x_set_fw_mac_addr(&rule_entry->mac.mac_msb,
&rule_entry->mac.mac_mid,
&rule_entry->mac.mac_lsb, mac);
+ rule_entry->mac.inner_mac =
+ cpu_to_le16(elem->cmd_data.vlan_mac.u.mac.is_inner_mac);
/* MOVE: Add a rule that will add this MAC to the target Queue */
if (cmd == BNX2X_VLAN_MAC_MOVE) {
@@ -786,6 +783,9 @@
bnx2x_set_fw_mac_addr(&rule_entry->mac.mac_msb,
&rule_entry->mac.mac_mid,
&rule_entry->mac.mac_lsb, mac);
+ rule_entry->mac.inner_mac =
+ cpu_to_le16(elem->cmd_data.vlan_mac.
+ u.mac.is_inner_mac);
}
/* Set the ramrod data header */
@@ -974,7 +974,8 @@
bnx2x_set_fw_mac_addr(&rule_entry->pair.mac_msb,
&rule_entry->pair.mac_mid,
&rule_entry->pair.mac_lsb, mac);
-
+ rule_entry->pair.inner_mac =
+ cpu_to_le16(elem->cmd_data.vlan_mac.u.vlan_mac.is_inner_mac);
/* MOVE: Add a rule that will add this MAC to the target Queue */
if (cmd == BNX2X_VLAN_MAC_MOVE) {
rule_entry++;
@@ -991,6 +992,9 @@
bnx2x_set_fw_mac_addr(&rule_entry->pair.mac_msb,
&rule_entry->pair.mac_mid,
&rule_entry->pair.mac_lsb, mac);
+ rule_entry->pair.inner_mac =
+ cpu_to_le16(elem->cmd_data.vlan_mac.u.
+ vlan_mac.is_inner_mac);
}
/* Set the ramrod data header */
@@ -1854,6 +1858,7 @@
return rc;
}
list_del(&exeq_pos->link);
+ bnx2x_exe_queue_free_elem(bp, exeq_pos);
}
}
@@ -2012,6 +2017,7 @@
vlan_obj->check_move = bnx2x_check_move;
vlan_obj->ramrod_cmd =
RAMROD_CMD_ID_ETH_CLASSIFICATION_RULES;
+ vlan_obj->get_n_elements = bnx2x_get_n_elements;
/* Exe Queue */
bnx2x_exe_queue_init(bp,
@@ -4426,6 +4432,10 @@
tx_data->force_default_pri_flg =
test_bit(BNX2X_Q_FLG_FORCE_DEFAULT_PRI, flags);
+ tx_data->tunnel_non_lso_pcsum_location =
+ test_bit(BNX2X_Q_FLG_PCSUM_ON_PKT, flags) ? PCSUM_ON_PKT :
+ PCSUM_ON_BD;
+
tx_data->tx_status_block_id = params->fw_sb_id;
tx_data->tx_sb_index_number = params->sb_cq_index;
tx_data->tss_leading_client_id = params->tss_leading_cl_id;
@@ -5669,17 +5679,18 @@
memset(rdata, 0, sizeof(*rdata));
/* Fill the ramrod data with provided parameters */
- rdata->function_mode = (u8)start_params->mf_mode;
- rdata->sd_vlan_tag = cpu_to_le16(start_params->sd_vlan_tag);
- rdata->path_id = BP_PATH(bp);
- rdata->network_cos_mode = start_params->network_cos_mode;
+ rdata->function_mode = (u8)start_params->mf_mode;
+ rdata->sd_vlan_tag = cpu_to_le16(start_params->sd_vlan_tag);
+ rdata->path_id = BP_PATH(bp);
+ rdata->network_cos_mode = start_params->network_cos_mode;
+ rdata->gre_tunnel_mode = start_params->gre_tunnel_mode;
+ rdata->gre_tunnel_rss = start_params->gre_tunnel_rss;
- /*
- * No need for an explicit memory barrier here as long we would
- * need to ensure the ordering of writing to the SPQ element
- * and updating of the SPQ producer which involves a memory
- * read and we will have to put a full memory barrier there
- * (inside bnx2x_sp_post()).
+ /* No need for an explicit memory barrier here as long we would
+ * need to ensure the ordering of writing to the SPQ element
+ * and updating of the SPQ producer which involves a memory
+ * read and we will have to put a full memory barrier there
+ * (inside bnx2x_sp_post()).
*/
return bnx2x_sp_post(bp, RAMROD_CMD_ID_COMMON_FUNCTION_START, 0,
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.h
index ff90760..35479da 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.h
@@ -100,6 +100,7 @@
/************************* VLAN-MAC commands related parameters ***************/
struct bnx2x_mac_ramrod_data {
u8 mac[ETH_ALEN];
+ u8 is_inner_mac;
};
struct bnx2x_vlan_ramrod_data {
@@ -108,6 +109,7 @@
struct bnx2x_vlan_mac_ramrod_data {
u8 mac[ETH_ALEN];
+ u8 is_inner_mac;
u16 vlan;
};
@@ -313,8 +315,9 @@
*
* @return number of copied bytes
*/
- int (*get_n_elements)(struct bnx2x *bp, struct bnx2x_vlan_mac_obj *o,
- int n, u8 *buf);
+ int (*get_n_elements)(struct bnx2x *bp,
+ struct bnx2x_vlan_mac_obj *o, int n, u8 *base,
+ u8 stride, u8 size);
/**
* Checks if ADD-ramrod with the given params may be performed.
@@ -824,7 +827,8 @@
BNX2X_Q_FLG_TX_SEC,
BNX2X_Q_FLG_ANTI_SPOOF,
BNX2X_Q_FLG_SILENT_VLAN_REM,
- BNX2X_Q_FLG_FORCE_DEFAULT_PRI
+ BNX2X_Q_FLG_FORCE_DEFAULT_PRI,
+ BNX2X_Q_FLG_PCSUM_ON_PKT
};
/* Queue type options: queue type may be a compination of below. */
@@ -842,6 +846,7 @@
#define BNX2X_MULTI_TX_COS_E3B0 3
#define BNX2X_MULTI_TX_COS 3 /* Maximum possible */
+#define MAC_PAD (ALIGN(ETH_ALEN, sizeof(u32)) - ETH_ALEN)
struct bnx2x_queue_init_params {
struct {
@@ -1118,6 +1123,15 @@
/* Function cos mode */
u8 network_cos_mode;
+
+ /* NVGRE classification enablement */
+ u8 nvgre_clss_en;
+
+ /* NO_GRE_TUNNEL/NVGRE_TUNNEL/L2GRE_TUNNEL/IPGRE_TUNNEL */
+ u8 gre_tunnel_mode;
+
+ /* GRE_OUTER_HEADERS_RSS/GRE_INNER_HEADERS_RSS/NVGRE_KEY_ENTROPY_RSS */
+ u8 gre_tunnel_rss;
};
struct bnx2x_func_switch_update_params {
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c
index 6adfa20..faadd15 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c
@@ -20,7 +20,9 @@
#include "bnx2x.h"
#include "bnx2x_init.h"
#include "bnx2x_cmn.h"
+#include "bnx2x_sp.h"
#include <linux/crc32.h>
+#include <linux/if_vlan.h>
/* General service functions */
static void storm_memset_vf_to_pf(struct bnx2x *bp, u16 abs_fid,
@@ -555,8 +557,7 @@
rc = bnx2x_config_vlan_mac(bp, vlan_mac);
if (rc >= 0) {
cnt += pos->add ? 1 : -1;
- list_del(&pos->link);
- list_add(&pos->link, &rollback_list);
+ list_move(&pos->link, &rollback_list);
rc = 0;
} else if (rc == -EEXIST) {
rc = 0;
@@ -958,6 +959,12 @@
BNX2X_ERR("QSETUP[%d:%d] error: rc %d\n", vf->abs_vfid, qid, vfop->rc);
op_done:
case BNX2X_VFOP_QSETUP_DONE:
+ vf->cfg_flags |= VF_CFG_VLAN;
+ smp_mb__before_clear_bit();
+ set_bit(BNX2X_SP_RTNL_HYPERVISOR_VLAN,
+ &bp->sp_rtnl_state);
+ smp_mb__after_clear_bit();
+ schedule_delayed_work(&bp->sp_rtnl_task, 0);
bnx2x_vfop_end(bp, vf, vfop);
return;
default:
@@ -1459,7 +1466,6 @@
return bnx2x_is_pcie_pending(dev);
unknown_dev:
- BNX2X_ERR("Unknown device\n");
return false;
}
@@ -1964,8 +1970,10 @@
if (iov->total == 0)
goto failed;
- /* calculate the actual number of VFs */
- iov->nr_virtfn = min_t(u16, iov->total, (u16)num_vfs_param);
+ iov->nr_virtfn = min_t(u16, iov->total, num_vfs_param);
+
+ DP(BNX2X_MSG_IOV, "num_vfs_param was %d, nr_virtfn was %d\n",
+ num_vfs_param, iov->nr_virtfn);
/* allocate the vf array */
bp->vfdb->vfs = kzalloc(sizeof(struct bnx2x_virtf) *
@@ -3012,21 +3020,138 @@
vf->op_current = CHANNEL_TLV_NONE;
}
-void bnx2x_enable_sriov(struct bnx2x *bp)
+int bnx2x_sriov_configure(struct pci_dev *dev, int num_vfs_param)
{
- int rc = 0;
- /* disbale sriov in case it is still enabled */
- pci_disable_sriov(bp->pdev);
- DP(BNX2X_MSG_IOV, "sriov disabled\n");
+ struct bnx2x *bp = netdev_priv(pci_get_drvdata(dev));
- /* enable sriov */
- DP(BNX2X_MSG_IOV, "vf num (%d)\n", (bp->vfdb->sriov.nr_virtfn));
- rc = pci_enable_sriov(bp->pdev, (bp->vfdb->sriov.nr_virtfn));
- if (rc)
+ DP(BNX2X_MSG_IOV, "bnx2x_sriov_configure called with %d, BNX2X_NR_VIRTFN(bp) was %d\n",
+ num_vfs_param, BNX2X_NR_VIRTFN(bp));
+
+ /* HW channel is only operational when PF is up */
+ if (bp->state != BNX2X_STATE_OPEN) {
+ BNX2X_ERR("VF num configurtion via sysfs not supported while PF is down");
+ return -EINVAL;
+ }
+
+ /* we are always bound by the total_vfs in the configuration space */
+ if (num_vfs_param > BNX2X_NR_VIRTFN(bp)) {
+ BNX2X_ERR("truncating requested number of VFs (%d) down to maximum allowed (%d)\n",
+ num_vfs_param, BNX2X_NR_VIRTFN(bp));
+ num_vfs_param = BNX2X_NR_VIRTFN(bp);
+ }
+
+ bp->requested_nr_virtfn = num_vfs_param;
+ if (num_vfs_param == 0) {
+ pci_disable_sriov(dev);
+ return 0;
+ } else {
+ return bnx2x_enable_sriov(bp);
+ }
+}
+
+int bnx2x_enable_sriov(struct bnx2x *bp)
+{
+ int rc = 0, req_vfs = bp->requested_nr_virtfn;
+
+ rc = pci_enable_sriov(bp->pdev, req_vfs);
+ if (rc) {
BNX2X_ERR("pci_enable_sriov failed with %d\n", rc);
- else
- DP(BNX2X_MSG_IOV, "sriov enabled\n");
+ return rc;
+ }
+ DP(BNX2X_MSG_IOV, "sriov enabled (%d vfs)\n", req_vfs);
+ return req_vfs;
+}
+
+void bnx2x_pf_set_vfs_vlan(struct bnx2x *bp)
+{
+ int vfidx;
+ struct pf_vf_bulletin_content *bulletin;
+
+ DP(BNX2X_MSG_IOV, "configuring vlan for VFs from sp-task\n");
+ for_each_vf(bp, vfidx) {
+ bulletin = BP_VF_BULLETIN(bp, vfidx);
+ if (BP_VF(bp, vfidx)->cfg_flags & VF_CFG_VLAN)
+ bnx2x_set_vf_vlan(bp->dev, vfidx, bulletin->vlan, 0);
+ }
+}
+
+void bnx2x_disable_sriov(struct bnx2x *bp)
+{
+ pci_disable_sriov(bp->pdev);
+}
+
+static int bnx2x_vf_ndo_sanity(struct bnx2x *bp, int vfidx,
+ struct bnx2x_virtf *vf)
+{
+ if (!IS_SRIOV(bp)) {
+ BNX2X_ERR("vf ndo called though sriov is disabled\n");
+ return -EINVAL;
+ }
+
+ if (vfidx >= BNX2X_NR_VIRTFN(bp)) {
+ BNX2X_ERR("vf ndo called for uninitialized VF. vfidx was %d BNX2X_NR_VIRTFN was %d\n",
+ vfidx, BNX2X_NR_VIRTFN(bp));
+ return -EINVAL;
+ }
+
+ if (!vf) {
+ BNX2X_ERR("vf ndo called but vf was null. vfidx was %d\n",
+ vfidx);
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+int bnx2x_get_vf_config(struct net_device *dev, int vfidx,
+ struct ifla_vf_info *ivi)
+{
+ struct bnx2x *bp = netdev_priv(dev);
+ struct bnx2x_virtf *vf = BP_VF(bp, vfidx);
+ struct bnx2x_vlan_mac_obj *mac_obj = &bnx2x_vfq(vf, 0, mac_obj);
+ struct bnx2x_vlan_mac_obj *vlan_obj = &bnx2x_vfq(vf, 0, vlan_obj);
+ struct pf_vf_bulletin_content *bulletin = BP_VF_BULLETIN(bp, vfidx);
+ int rc;
+
+ /* sanity */
+ rc = bnx2x_vf_ndo_sanity(bp, vfidx, vf);
+ if (rc)
+ return rc;
+ if (!mac_obj || !vlan_obj || !bulletin) {
+ BNX2X_ERR("VF partially initialized\n");
+ return -EINVAL;
+ }
+
+ ivi->vf = vfidx;
+ ivi->qos = 0;
+ ivi->tx_rate = 10000; /* always 10G. TBA take from link struct */
+ ivi->spoofchk = 1; /*always enabled */
+ if (vf->state == VF_ENABLED) {
+ /* mac and vlan are in vlan_mac objects */
+ mac_obj->get_n_elements(bp, mac_obj, 1, (u8 *)&ivi->mac,
+ 0, ETH_ALEN);
+ vlan_obj->get_n_elements(bp, vlan_obj, 1, (u8 *)&ivi->vlan,
+ 0, VLAN_HLEN);
+ } else {
+ /* mac */
+ if (bulletin->valid_bitmap & (1 << MAC_ADDR_VALID))
+ /* mac configured by ndo so its in bulletin board */
+ memcpy(&ivi->mac, bulletin->mac, ETH_ALEN);
+ else
+ /* funtion has not been loaded yet. Show mac as 0s */
+ memset(&ivi->mac, 0, ETH_ALEN);
+
+ /* vlan */
+ if (bulletin->valid_bitmap & (1 << VLAN_VALID))
+ /* vlan configured by ndo so its in bulletin board */
+ memcpy(&ivi->vlan, &bulletin->vlan, VLAN_HLEN);
+ else
+ /* funtion has not been loaded yet. Show vlans as 0s */
+ memset(&ivi->vlan, 0, VLAN_HLEN);
+ }
+
+ return 0;
}
/* New mac for VF. Consider these cases:
@@ -3044,23 +3169,19 @@
* VF to configure any mac for itself except for this mac. In case of a race
* where the VF fails to see the new post on its bulletin board before sending a
* mac configuration request, the PF will simply fail the request and VF can try
- * again after consulting its bulletin board
+ * again after consulting its bulletin board.
*/
-int bnx2x_set_vf_mac(struct net_device *dev, int queue, u8 *mac)
+int bnx2x_set_vf_mac(struct net_device *dev, int vfidx, u8 *mac)
{
struct bnx2x *bp = netdev_priv(dev);
- int rc, q_logical_state, vfidx = queue;
+ int rc, q_logical_state;
struct bnx2x_virtf *vf = BP_VF(bp, vfidx);
struct pf_vf_bulletin_content *bulletin = BP_VF_BULLETIN(bp, vfidx);
- /* if SRIOV is disabled there is nothing to do (and somewhere, someone
- * has erred).
- */
- if (!IS_SRIOV(bp)) {
- BNX2X_ERR("bnx2x_set_vf_mac called though sriov is disabled\n");
- return -EINVAL;
- }
-
+ /* sanity */
+ rc = bnx2x_vf_ndo_sanity(bp, vfidx, vf);
+ if (rc)
+ return rc;
if (!is_valid_ether_addr(mac)) {
BNX2X_ERR("mac address invalid\n");
return -EINVAL;
@@ -3085,7 +3206,7 @@
if (vf->state == VF_ENABLED &&
q_logical_state == BNX2X_Q_LOGICAL_STATE_ACTIVE) {
/* configure the mac in device on this vf's queue */
- unsigned long flags = 0;
+ unsigned long ramrod_flags = 0;
struct bnx2x_vlan_mac_obj *mac_obj = &bnx2x_vfq(vf, 0, mac_obj);
/* must lock vfpf channel to protect against vf flows */
@@ -3106,14 +3227,133 @@
}
/* configure the new mac to device */
- __set_bit(RAMROD_COMP_WAIT, &flags);
+ __set_bit(RAMROD_COMP_WAIT, &ramrod_flags);
bnx2x_set_mac_one(bp, (u8 *)&bulletin->mac, mac_obj, true,
- BNX2X_ETH_MAC, &flags);
+ BNX2X_ETH_MAC, &ramrod_flags);
bnx2x_unlock_vf_pf_channel(bp, vf, CHANNEL_TLV_PF_SET_MAC);
}
- return rc;
+ return 0;
+}
+
+int bnx2x_set_vf_vlan(struct net_device *dev, int vfidx, u16 vlan, u8 qos)
+{
+ struct bnx2x *bp = netdev_priv(dev);
+ int rc, q_logical_state;
+ struct bnx2x_virtf *vf = BP_VF(bp, vfidx);
+ struct pf_vf_bulletin_content *bulletin = BP_VF_BULLETIN(bp, vfidx);
+
+ /* sanity */
+ rc = bnx2x_vf_ndo_sanity(bp, vfidx, vf);
+ if (rc)
+ return rc;
+
+ if (vlan > 4095) {
+ BNX2X_ERR("illegal vlan value %d\n", vlan);
+ return -EINVAL;
+ }
+
+ DP(BNX2X_MSG_IOV, "configuring VF %d with VLAN %d qos %d\n",
+ vfidx, vlan, 0);
+
+ /* update PF's copy of the VF's bulletin. No point in posting the vlan
+ * to the VF since it doesn't have anything to do with it. But it useful
+ * to store it here in case the VF is not up yet and we can only
+ * configure the vlan later when it does.
+ */
+ bulletin->valid_bitmap |= 1 << VLAN_VALID;
+ bulletin->vlan = vlan;
+
+ /* is vf initialized and queue set up? */
+ q_logical_state =
+ bnx2x_get_q_logical_state(bp, &bnx2x_vfq(vf, 0, sp_obj));
+ if (vf->state == VF_ENABLED &&
+ q_logical_state == BNX2X_Q_LOGICAL_STATE_ACTIVE) {
+ /* configure the vlan in device on this vf's queue */
+ unsigned long ramrod_flags = 0;
+ unsigned long vlan_mac_flags = 0;
+ struct bnx2x_vlan_mac_obj *vlan_obj =
+ &bnx2x_vfq(vf, 0, vlan_obj);
+ struct bnx2x_vlan_mac_ramrod_params ramrod_param;
+ struct bnx2x_queue_state_params q_params = {NULL};
+ struct bnx2x_queue_update_params *update_params;
+
+ memset(&ramrod_param, 0, sizeof(ramrod_param));
+
+ /* must lock vfpf channel to protect against vf flows */
+ bnx2x_lock_vf_pf_channel(bp, vf, CHANNEL_TLV_PF_SET_VLAN);
+
+ /* remove existing vlans */
+ __set_bit(RAMROD_COMP_WAIT, &ramrod_flags);
+ rc = vlan_obj->delete_all(bp, vlan_obj, &vlan_mac_flags,
+ &ramrod_flags);
+ if (rc) {
+ BNX2X_ERR("failed to delete vlans\n");
+ return -EINVAL;
+ }
+
+ /* send queue update ramrod to configure default vlan and silent
+ * vlan removal
+ */
+ __set_bit(RAMROD_COMP_WAIT, &q_params.ramrod_flags);
+ q_params.cmd = BNX2X_Q_CMD_UPDATE;
+ q_params.q_obj = &bnx2x_vfq(vf, 0, sp_obj);
+ update_params = &q_params.params.update;
+ __set_bit(BNX2X_Q_UPDATE_DEF_VLAN_EN_CHNG,
+ &update_params->update_flags);
+ __set_bit(BNX2X_Q_UPDATE_SILENT_VLAN_REM_CHNG,
+ &update_params->update_flags);
+
+ if (vlan == 0) {
+ /* if vlan is 0 then we want to leave the VF traffic
+ * untagged, and leave the incoming traffic untouched
+ * (i.e. do not remove any vlan tags).
+ */
+ __clear_bit(BNX2X_Q_UPDATE_DEF_VLAN_EN,
+ &update_params->update_flags);
+ __clear_bit(BNX2X_Q_UPDATE_SILENT_VLAN_REM,
+ &update_params->update_flags);
+ } else {
+ /* configure the new vlan to device */
+ __set_bit(RAMROD_COMP_WAIT, &ramrod_flags);
+ ramrod_param.vlan_mac_obj = vlan_obj;
+ ramrod_param.ramrod_flags = ramrod_flags;
+ ramrod_param.user_req.u.vlan.vlan = vlan;
+ ramrod_param.user_req.cmd = BNX2X_VLAN_MAC_ADD;
+ rc = bnx2x_config_vlan_mac(bp, &ramrod_param);
+ if (rc) {
+ BNX2X_ERR("failed to configure vlan\n");
+ return -EINVAL;
+ }
+
+ /* configure default vlan to vf queue and set silent
+ * vlan removal (the vf remains unaware of this vlan).
+ */
+ update_params = &q_params.params.update;
+ __set_bit(BNX2X_Q_UPDATE_DEF_VLAN_EN,
+ &update_params->update_flags);
+ __set_bit(BNX2X_Q_UPDATE_SILENT_VLAN_REM,
+ &update_params->update_flags);
+ update_params->def_vlan = vlan;
+ }
+
+ /* Update the Queue state */
+ rc = bnx2x_queue_state_change(bp, &q_params);
+ if (rc) {
+ BNX2X_ERR("Failed to configure default VLAN\n");
+ return rc;
+ }
+
+ /* clear the flag indicating that this VF needs its vlan
+ * (will only be set if the HV configured th Vlan before vf was
+ * and we were called because the VF came up later
+ */
+ vf->cfg_flags &= ~VF_CFG_VLAN;
+
+ bnx2x_unlock_vf_pf_channel(bp, vf, CHANNEL_TLV_PF_SET_VLAN);
+ }
+ return 0;
}
/* crc is the first field in the bulletin board. compute the crc over the
@@ -3165,6 +3405,10 @@
memcpy(bp->dev->dev_addr, bulletin.mac, ETH_ALEN);
}
+ /* the vlan in bulletin board is valid and is new */
+ if (bulletin.valid_bitmap & 1 << VLAN_VALID)
+ memcpy(&bulletin.vlan, &bp->old_bulletin.vlan, VLAN_HLEN);
+
/* copy new bulletin board to bp */
bp->old_bulletin = bulletin;
@@ -3196,3 +3440,26 @@
sizeof(union pf_vf_bulletin));
return -ENOMEM;
}
+
+int bnx2x_open_epilog(struct bnx2x *bp)
+{
+ /* Enable sriov via delayed work. This must be done via delayed work
+ * because it causes the probe of the vf devices to be run, which invoke
+ * register_netdevice which must have rtnl lock taken. As we are holding
+ * the lock right now, that could only work if the probe would not take
+ * the lock. However, as the probe of the vf may be called from other
+ * contexts as well (such as passthrough to vm failes) it can't assume
+ * the lock is being held for it. Using delayed work here allows the
+ * probe code to simply take the lock (i.e. wait for it to be released
+ * if it is being held). We only want to do this if the number of VFs
+ * was set before PF driver was loaded.
+ */
+ if (IS_SRIOV(bp) && BNX2X_NR_VIRTFN(bp)) {
+ smp_mb__before_clear_bit();
+ set_bit(BNX2X_SP_RTNL_ENABLE_SRIOV, &bp->sp_rtnl_state);
+ smp_mb__after_clear_bit();
+ schedule_delayed_work(&bp->sp_rtnl_task, 0);
+ }
+
+ return 0;
+}
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h
index b405017..a10bdb2 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h
@@ -193,6 +193,7 @@
#define VF_CFG_TPA 0x0004
#define VF_CFG_INT_SIMD 0x0008
#define VF_CACHE_LINE 0x0010
+#define VF_CFG_VLAN 0x0020
u8 state;
#define VF_FREE 0 /* VF ready to be acquired holds no resc */
@@ -752,11 +753,15 @@
enum sample_bulletin_result bnx2x_sample_bulletin(struct bnx2x *bp);
void bnx2x_vf_map_doorbells(struct bnx2x *bp);
int bnx2x_vf_pci_alloc(struct bnx2x *bp);
-void bnx2x_enable_sriov(struct bnx2x *bp);
+int bnx2x_enable_sriov(struct bnx2x *bp);
+void bnx2x_disable_sriov(struct bnx2x *bp);
static inline int bnx2x_vf_headroom(struct bnx2x *bp)
{
return bp->vfdb->sriov.nr_virtfn * BNX2X_CLIENTS_PER_VF;
}
+void bnx2x_pf_set_vfs_vlan(struct bnx2x *bp);
+int bnx2x_sriov_configure(struct pci_dev *dev, int num_vfs);
+int bnx2x_open_epilog(struct bnx2x *bp);
#else /* CONFIG_BNX2X_SRIOV */
@@ -779,7 +784,8 @@
static inline int bnx2x_iov_init_one(struct bnx2x *bp, int int_mode_param,
int num_vfs_param) {return 0; }
static inline void bnx2x_iov_remove_one(struct bnx2x *bp) {}
-static inline void bnx2x_enable_sriov(struct bnx2x *bp) {}
+static inline int bnx2x_enable_sriov(struct bnx2x *bp) {return 0; }
+static inline void bnx2x_disable_sriov(struct bnx2x *bp) {}
static inline int bnx2x_vfpf_acquire(struct bnx2x *bp,
u8 tx_count, u8 rx_count) {return 0; }
static inline int bnx2x_vfpf_release(struct bnx2x *bp) {return 0; }
@@ -804,6 +810,9 @@
static inline int bnx2x_vf_map_doorbells(struct bnx2x *bp) {return 0; }
static inline int bnx2x_vf_pci_alloc(struct bnx2x *bp) {return 0; }
+static inline void bnx2x_pf_set_vfs_vlan(struct bnx2x *bp) {}
+static inline int bnx2x_sriov_configure(struct pci_dev *dev, int num_vfs) {return 0; }
+static inline int bnx2x_open_epilog(struct bnx2x *bp) {return 0; }
#endif /* CONFIG_BNX2X_SRIOV */
#endif /* bnx2x_sriov.h */
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h
index bfc80ba..41708fa 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h
@@ -328,9 +328,15 @@
#define MAC_ADDR_VALID 0 /* alert the vf that a new mac address
* is available for it
*/
+#define VLAN_VALID 1 /* when set, the vf should not access
+ * the vfpf channel
+ */
u8 mac[ETH_ALEN];
- u8 padding[2];
+ u8 mac_padding[2];
+
+ u16 vlan;
+ u8 vlan_padding[6];
};
union pf_vf_bulletin {
@@ -353,6 +359,7 @@
CHANNEL_TLV_LIST_END,
CHANNEL_TLV_FLR,
CHANNEL_TLV_PF_SET_MAC,
+ CHANNEL_TLV_PF_SET_VLAN,
CHANNEL_TLV_MAX
};
diff --git a/drivers/net/ethernet/broadcom/sb1250-mac.c b/drivers/net/ethernet/broadcom/sb1250-mac.c
index e9b35da..e80bfb6 100644
--- a/drivers/net/ethernet/broadcom/sb1250-mac.c
+++ b/drivers/net/ethernet/broadcom/sb1250-mac.c
@@ -831,11 +831,8 @@
sb_new = netdev_alloc_skb(dev, ENET_PACKET_SIZE +
SMP_CACHE_BYTES * 2 +
NET_IP_ALIGN);
- if (sb_new == NULL) {
- pr_info("%s: sk_buff allocation failed\n",
- d->sbdma_eth->sbm_dev->name);
+ if (sb_new == NULL)
return -ENOBUFS;
- }
sbdma_align_skb(sb_new, SMP_CACHE_BYTES, NET_IP_ALIGN);
}
diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c
index 67d2663..dea7d7d 100644
--- a/drivers/net/ethernet/broadcom/tg3.c
+++ b/drivers/net/ethernet/broadcom/tg3.c
@@ -212,6 +212,7 @@
#define TG3_FW_UPDATE_FREQ_SEC (TG3_FW_UPDATE_TIMEOUT_SEC / 2)
#define FIRMWARE_TG3 "tigon/tg3.bin"
+#define FIRMWARE_TG357766 "tigon/tg357766.bin"
#define FIRMWARE_TG3TSO "tigon/tg3_tso.bin"
#define FIRMWARE_TG3TSO5 "tigon/tg3_tso5.bin"
@@ -3448,11 +3449,58 @@
#define TX_CPU_SCRATCH_SIZE 0x04000
/* tp->lock is held. */
-static int tg3_halt_cpu(struct tg3 *tp, u32 offset)
+static int tg3_pause_cpu(struct tg3 *tp, u32 cpu_base)
{
int i;
+ const int iters = 10000;
- BUG_ON(offset == TX_CPU_BASE && tg3_flag(tp, 5705_PLUS));
+ for (i = 0; i < iters; i++) {
+ tw32(cpu_base + CPU_STATE, 0xffffffff);
+ tw32(cpu_base + CPU_MODE, CPU_MODE_HALT);
+ if (tr32(cpu_base + CPU_MODE) & CPU_MODE_HALT)
+ break;
+ }
+
+ return (i == iters) ? -EBUSY : 0;
+}
+
+/* tp->lock is held. */
+static int tg3_rxcpu_pause(struct tg3 *tp)
+{
+ int rc = tg3_pause_cpu(tp, RX_CPU_BASE);
+
+ tw32(RX_CPU_BASE + CPU_STATE, 0xffffffff);
+ tw32_f(RX_CPU_BASE + CPU_MODE, CPU_MODE_HALT);
+ udelay(10);
+
+ return rc;
+}
+
+/* tp->lock is held. */
+static int tg3_txcpu_pause(struct tg3 *tp)
+{
+ return tg3_pause_cpu(tp, TX_CPU_BASE);
+}
+
+/* tp->lock is held. */
+static void tg3_resume_cpu(struct tg3 *tp, u32 cpu_base)
+{
+ tw32(cpu_base + CPU_STATE, 0xffffffff);
+ tw32_f(cpu_base + CPU_MODE, 0x00000000);
+}
+
+/* tp->lock is held. */
+static void tg3_rxcpu_resume(struct tg3 *tp)
+{
+ tg3_resume_cpu(tp, RX_CPU_BASE);
+}
+
+/* tp->lock is held. */
+static int tg3_halt_cpu(struct tg3 *tp, u32 cpu_base)
+{
+ int rc;
+
+ BUG_ON(cpu_base == TX_CPU_BASE && tg3_flag(tp, 5705_PLUS));
if (tg3_asic_rev(tp) == ASIC_REV_5906) {
u32 val = tr32(GRC_VCPU_EXT_CTRL);
@@ -3460,17 +3508,8 @@
tw32(GRC_VCPU_EXT_CTRL, val | GRC_VCPU_EXT_CTRL_HALT_CPU);
return 0;
}
- if (offset == RX_CPU_BASE) {
- for (i = 0; i < 10000; i++) {
- tw32(offset + CPU_STATE, 0xffffffff);
- tw32(offset + CPU_MODE, CPU_MODE_HALT);
- if (tr32(offset + CPU_MODE) & CPU_MODE_HALT)
- break;
- }
-
- tw32(offset + CPU_STATE, 0xffffffff);
- tw32_f(offset + CPU_MODE, CPU_MODE_HALT);
- udelay(10);
+ if (cpu_base == RX_CPU_BASE) {
+ rc = tg3_rxcpu_pause(tp);
} else {
/*
* There is only an Rx CPU for the 5750 derivative in the
@@ -3479,17 +3518,12 @@
if (tg3_flag(tp, IS_SSB_CORE))
return 0;
- for (i = 0; i < 10000; i++) {
- tw32(offset + CPU_STATE, 0xffffffff);
- tw32(offset + CPU_MODE, CPU_MODE_HALT);
- if (tr32(offset + CPU_MODE) & CPU_MODE_HALT)
- break;
- }
+ rc = tg3_txcpu_pause(tp);
}
- if (i >= 10000) {
+ if (rc) {
netdev_err(tp->dev, "%s timed out, %s CPU\n",
- __func__, offset == RX_CPU_BASE ? "RX" : "TX");
+ __func__, cpu_base == RX_CPU_BASE ? "RX" : "TX");
return -ENODEV;
}
@@ -3499,19 +3533,41 @@
return 0;
}
-struct fw_info {
- unsigned int fw_base;
- unsigned int fw_len;
- const __be32 *fw_data;
-};
+static int tg3_fw_data_len(struct tg3 *tp,
+ const struct tg3_firmware_hdr *fw_hdr)
+{
+ int fw_len;
+
+ /* Non fragmented firmware have one firmware header followed by a
+ * contiguous chunk of data to be written. The length field in that
+ * header is not the length of data to be written but the complete
+ * length of the bss. The data length is determined based on
+ * tp->fw->size minus headers.
+ *
+ * Fragmented firmware have a main header followed by multiple
+ * fragments. Each fragment is identical to non fragmented firmware
+ * with a firmware header followed by a contiguous chunk of data. In
+ * the main header, the length field is unused and set to 0xffffffff.
+ * In each fragment header the length is the entire size of that
+ * fragment i.e. fragment data + header length. Data length is
+ * therefore length field in the header minus TG3_FW_HDR_LEN.
+ */
+ if (tp->fw_len == 0xffffffff)
+ fw_len = be32_to_cpu(fw_hdr->len);
+ else
+ fw_len = tp->fw->size;
+
+ return (fw_len - TG3_FW_HDR_LEN) / sizeof(u32);
+}
/* tp->lock is held. */
static int tg3_load_firmware_cpu(struct tg3 *tp, u32 cpu_base,
u32 cpu_scratch_base, int cpu_scratch_size,
- struct fw_info *info)
+ const struct tg3_firmware_hdr *fw_hdr)
{
- int err, lock_err, i;
+ int err, i;
void (*write_op)(struct tg3 *, u32, u32);
+ int total_len = tp->fw->size;
if (cpu_base == TX_CPU_BASE && tg3_flag(tp, 5705_PLUS)) {
netdev_err(tp->dev,
@@ -3520,30 +3576,49 @@
return -EINVAL;
}
- if (tg3_flag(tp, 5705_PLUS))
+ if (tg3_flag(tp, 5705_PLUS) && tg3_asic_rev(tp) != ASIC_REV_57766)
write_op = tg3_write_mem;
else
write_op = tg3_write_indirect_reg32;
- /* It is possible that bootcode is still loading at this point.
- * Get the nvram lock first before halting the cpu.
- */
- lock_err = tg3_nvram_lock(tp);
- err = tg3_halt_cpu(tp, cpu_base);
- if (!lock_err)
- tg3_nvram_unlock(tp);
- if (err)
- goto out;
+ if (tg3_asic_rev(tp) != ASIC_REV_57766) {
+ /* It is possible that bootcode is still loading at this point.
+ * Get the nvram lock first before halting the cpu.
+ */
+ int lock_err = tg3_nvram_lock(tp);
+ err = tg3_halt_cpu(tp, cpu_base);
+ if (!lock_err)
+ tg3_nvram_unlock(tp);
+ if (err)
+ goto out;
- for (i = 0; i < cpu_scratch_size; i += sizeof(u32))
- write_op(tp, cpu_scratch_base + i, 0);
- tw32(cpu_base + CPU_STATE, 0xffffffff);
- tw32(cpu_base + CPU_MODE, tr32(cpu_base+CPU_MODE)|CPU_MODE_HALT);
- for (i = 0; i < (info->fw_len / sizeof(u32)); i++)
- write_op(tp, (cpu_scratch_base +
- (info->fw_base & 0xffff) +
- (i * sizeof(u32))),
- be32_to_cpu(info->fw_data[i]));
+ for (i = 0; i < cpu_scratch_size; i += sizeof(u32))
+ write_op(tp, cpu_scratch_base + i, 0);
+ tw32(cpu_base + CPU_STATE, 0xffffffff);
+ tw32(cpu_base + CPU_MODE,
+ tr32(cpu_base + CPU_MODE) | CPU_MODE_HALT);
+ } else {
+ /* Subtract additional main header for fragmented firmware and
+ * advance to the first fragment
+ */
+ total_len -= TG3_FW_HDR_LEN;
+ fw_hdr++;
+ }
+
+ do {
+ u32 *fw_data = (u32 *)(fw_hdr + 1);
+ for (i = 0; i < tg3_fw_data_len(tp, fw_hdr); i++)
+ write_op(tp, cpu_scratch_base +
+ (be32_to_cpu(fw_hdr->base_addr) & 0xffff) +
+ (i * sizeof(u32)),
+ be32_to_cpu(fw_data[i]));
+
+ total_len -= be32_to_cpu(fw_hdr->len);
+
+ /* Advance to next fragment */
+ fw_hdr = (struct tg3_firmware_hdr *)
+ ((void *)fw_hdr + be32_to_cpu(fw_hdr->len));
+ } while (total_len > 0);
err = 0;
@@ -3552,13 +3627,33 @@
}
/* tp->lock is held. */
+static int tg3_pause_cpu_and_set_pc(struct tg3 *tp, u32 cpu_base, u32 pc)
+{
+ int i;
+ const int iters = 5;
+
+ tw32(cpu_base + CPU_STATE, 0xffffffff);
+ tw32_f(cpu_base + CPU_PC, pc);
+
+ for (i = 0; i < iters; i++) {
+ if (tr32(cpu_base + CPU_PC) == pc)
+ break;
+ tw32(cpu_base + CPU_STATE, 0xffffffff);
+ tw32(cpu_base + CPU_MODE, CPU_MODE_HALT);
+ tw32_f(cpu_base + CPU_PC, pc);
+ udelay(1000);
+ }
+
+ return (i == iters) ? -EBUSY : 0;
+}
+
+/* tp->lock is held. */
static int tg3_load_5701_a0_firmware_fix(struct tg3 *tp)
{
- struct fw_info info;
- const __be32 *fw_data;
- int err, i;
+ const struct tg3_firmware_hdr *fw_hdr;
+ int err;
- fw_data = (void *)tp->fw->data;
+ fw_hdr = (struct tg3_firmware_hdr *)tp->fw->data;
/* Firmware blob starts with version numbers, followed by
start address and length. We are setting complete length.
@@ -3566,60 +3661,117 @@
Remainder is the blob to be loaded contiguously
from start address. */
- info.fw_base = be32_to_cpu(fw_data[1]);
- info.fw_len = tp->fw->size - 12;
- info.fw_data = &fw_data[3];
-
err = tg3_load_firmware_cpu(tp, RX_CPU_BASE,
RX_CPU_SCRATCH_BASE, RX_CPU_SCRATCH_SIZE,
- &info);
+ fw_hdr);
if (err)
return err;
err = tg3_load_firmware_cpu(tp, TX_CPU_BASE,
TX_CPU_SCRATCH_BASE, TX_CPU_SCRATCH_SIZE,
- &info);
+ fw_hdr);
if (err)
return err;
/* Now startup only the RX cpu. */
- tw32(RX_CPU_BASE + CPU_STATE, 0xffffffff);
- tw32_f(RX_CPU_BASE + CPU_PC, info.fw_base);
-
- for (i = 0; i < 5; i++) {
- if (tr32(RX_CPU_BASE + CPU_PC) == info.fw_base)
- break;
- tw32(RX_CPU_BASE + CPU_STATE, 0xffffffff);
- tw32(RX_CPU_BASE + CPU_MODE, CPU_MODE_HALT);
- tw32_f(RX_CPU_BASE + CPU_PC, info.fw_base);
- udelay(1000);
- }
- if (i >= 5) {
+ err = tg3_pause_cpu_and_set_pc(tp, RX_CPU_BASE,
+ be32_to_cpu(fw_hdr->base_addr));
+ if (err) {
netdev_err(tp->dev, "%s fails to set RX CPU PC, is %08x "
"should be %08x\n", __func__,
- tr32(RX_CPU_BASE + CPU_PC), info.fw_base);
+ tr32(RX_CPU_BASE + CPU_PC),
+ be32_to_cpu(fw_hdr->base_addr));
return -ENODEV;
}
- tw32(RX_CPU_BASE + CPU_STATE, 0xffffffff);
- tw32_f(RX_CPU_BASE + CPU_MODE, 0x00000000);
+
+ tg3_rxcpu_resume(tp);
+
+ return 0;
+}
+
+static int tg3_validate_rxcpu_state(struct tg3 *tp)
+{
+ const int iters = 1000;
+ int i;
+ u32 val;
+
+ /* Wait for boot code to complete initialization and enter service
+ * loop. It is then safe to download service patches
+ */
+ for (i = 0; i < iters; i++) {
+ if (tr32(RX_CPU_HWBKPT) == TG3_SBROM_IN_SERVICE_LOOP)
+ break;
+
+ udelay(10);
+ }
+
+ if (i == iters) {
+ netdev_err(tp->dev, "Boot code not ready for service patches\n");
+ return -EBUSY;
+ }
+
+ val = tg3_read_indirect_reg32(tp, TG3_57766_FW_HANDSHAKE);
+ if (val & 0xff) {
+ netdev_warn(tp->dev,
+ "Other patches exist. Not downloading EEE patch\n");
+ return -EEXIST;
+ }
return 0;
}
/* tp->lock is held. */
+static void tg3_load_57766_firmware(struct tg3 *tp)
+{
+ struct tg3_firmware_hdr *fw_hdr;
+
+ if (!tg3_flag(tp, NO_NVRAM))
+ return;
+
+ if (tg3_validate_rxcpu_state(tp))
+ return;
+
+ if (!tp->fw)
+ return;
+
+ /* This firmware blob has a different format than older firmware
+ * releases as given below. The main difference is we have fragmented
+ * data to be written to non-contiguous locations.
+ *
+ * In the beginning we have a firmware header identical to other
+ * firmware which consists of version, base addr and length. The length
+ * here is unused and set to 0xffffffff.
+ *
+ * This is followed by a series of firmware fragments which are
+ * individually identical to previous firmware. i.e. they have the
+ * firmware header and followed by data for that fragment. The version
+ * field of the individual fragment header is unused.
+ */
+
+ fw_hdr = (struct tg3_firmware_hdr *)tp->fw->data;
+ if (be32_to_cpu(fw_hdr->base_addr) != TG3_57766_FW_BASE_ADDR)
+ return;
+
+ if (tg3_rxcpu_pause(tp))
+ return;
+
+ /* tg3_load_firmware_cpu() will always succeed for the 57766 */
+ tg3_load_firmware_cpu(tp, 0, TG3_57766_FW_BASE_ADDR, 0, fw_hdr);
+
+ tg3_rxcpu_resume(tp);
+}
+
+/* tp->lock is held. */
static int tg3_load_tso_firmware(struct tg3 *tp)
{
- struct fw_info info;
- const __be32 *fw_data;
+ const struct tg3_firmware_hdr *fw_hdr;
unsigned long cpu_base, cpu_scratch_base, cpu_scratch_size;
- int err, i;
+ int err;
- if (tg3_flag(tp, HW_TSO_1) ||
- tg3_flag(tp, HW_TSO_2) ||
- tg3_flag(tp, HW_TSO_3))
+ if (!tg3_flag(tp, FW_TSO))
return 0;
- fw_data = (void *)tp->fw->data;
+ fw_hdr = (struct tg3_firmware_hdr *)tp->fw->data;
/* Firmware blob starts with version numbers, followed by
start address and length. We are setting complete length.
@@ -3627,10 +3779,7 @@
Remainder is the blob to be loaded contiguously
from start address. */
- info.fw_base = be32_to_cpu(fw_data[1]);
cpu_scratch_size = tp->fw_len;
- info.fw_len = tp->fw->size - 12;
- info.fw_data = &fw_data[3];
if (tg3_asic_rev(tp) == ASIC_REV_5705) {
cpu_base = RX_CPU_BASE;
@@ -3643,30 +3792,22 @@
err = tg3_load_firmware_cpu(tp, cpu_base,
cpu_scratch_base, cpu_scratch_size,
- &info);
+ fw_hdr);
if (err)
return err;
/* Now startup the cpu. */
- tw32(cpu_base + CPU_STATE, 0xffffffff);
- tw32_f(cpu_base + CPU_PC, info.fw_base);
-
- for (i = 0; i < 5; i++) {
- if (tr32(cpu_base + CPU_PC) == info.fw_base)
- break;
- tw32(cpu_base + CPU_STATE, 0xffffffff);
- tw32(cpu_base + CPU_MODE, CPU_MODE_HALT);
- tw32_f(cpu_base + CPU_PC, info.fw_base);
- udelay(1000);
- }
- if (i >= 5) {
+ err = tg3_pause_cpu_and_set_pc(tp, cpu_base,
+ be32_to_cpu(fw_hdr->base_addr));
+ if (err) {
netdev_err(tp->dev,
"%s fails to set CPU PC, is %08x should be %08x\n",
- __func__, tr32(cpu_base + CPU_PC), info.fw_base);
+ __func__, tr32(cpu_base + CPU_PC),
+ be32_to_cpu(fw_hdr->base_addr));
return -ENODEV;
}
- tw32(cpu_base + CPU_STATE, 0xffffffff);
- tw32_f(cpu_base + CPU_MODE, 0x00000000);
+
+ tg3_resume_cpu(tp, cpu_base);
return 0;
}
@@ -8039,11 +8180,9 @@
tnapi->rx_rcb = dma_alloc_coherent(&tp->pdev->dev,
TG3_RX_RCB_RING_BYTES(tp),
&tnapi->rx_rcb_mapping,
- GFP_KERNEL);
+ GFP_KERNEL | __GFP_ZERO);
if (!tnapi->rx_rcb)
goto err_out;
-
- memset(tnapi->rx_rcb, 0, TG3_RX_RCB_RING_BYTES(tp));
}
return 0;
@@ -8093,12 +8232,10 @@
tp->hw_stats = dma_alloc_coherent(&tp->pdev->dev,
sizeof(struct tg3_hw_stats),
&tp->stats_mapping,
- GFP_KERNEL);
+ GFP_KERNEL | __GFP_ZERO);
if (!tp->hw_stats)
goto err_out;
- memset(tp->hw_stats, 0, sizeof(struct tg3_hw_stats));
-
for (i = 0; i < tp->irq_cnt; i++) {
struct tg3_napi *tnapi = &tp->napi[i];
struct tg3_hw_status *sblk;
@@ -8106,11 +8243,10 @@
tnapi->hw_status = dma_alloc_coherent(&tp->pdev->dev,
TG3_HW_STATUS_SIZE,
&tnapi->status_mapping,
- GFP_KERNEL);
+ GFP_KERNEL | __GFP_ZERO);
if (!tnapi->hw_status)
goto err_out;
- memset(tnapi->hw_status, 0, TG3_HW_STATUS_SIZE);
sblk = tnapi->hw_status;
if (tg3_flag(tp, ENABLE_RSS)) {
@@ -9781,6 +9917,13 @@
return err;
}
+ if (tg3_asic_rev(tp) == ASIC_REV_57766) {
+ /* Ignore any errors for the firmware download. If download
+ * fails, the device will operate with EEE disabled
+ */
+ tg3_load_57766_firmware(tp);
+ }
+
if (tg3_flag(tp, TSO_CAPABLE)) {
err = tg3_load_tso_firmware(tp);
if (err)
@@ -10570,7 +10713,7 @@
static int tg3_request_firmware(struct tg3 *tp)
{
- const __be32 *fw_data;
+ const struct tg3_firmware_hdr *fw_hdr;
if (request_firmware(&tp->fw, tp->fw_needed, &tp->pdev->dev)) {
netdev_err(tp->dev, "Failed to load firmware \"%s\"\n",
@@ -10578,15 +10721,15 @@
return -ENOENT;
}
- fw_data = (void *)tp->fw->data;
+ fw_hdr = (struct tg3_firmware_hdr *)tp->fw->data;
/* Firmware blob starts with version numbers, followed by
* start address and _full_ length including BSS sections
* (which must be longer than the actual data, of course
*/
- tp->fw_len = be32_to_cpu(fw_data[2]); /* includes bss */
- if (tp->fw_len < (tp->fw->size - 12)) {
+ tp->fw_len = be32_to_cpu(fw_hdr->len); /* includes bss */
+ if (tp->fw_len < (tp->fw->size - TG3_FW_HDR_LEN)) {
netdev_err(tp->dev, "bogus length %d in \"%s\"\n",
tp->fw_len, tp->fw_needed);
release_firmware(tp->fw);
@@ -10885,7 +11028,15 @@
if (tp->fw_needed) {
err = tg3_request_firmware(tp);
- if (tg3_chip_rev_id(tp) == CHIPREV_ID_5701_A0) {
+ if (tg3_asic_rev(tp) == ASIC_REV_57766) {
+ if (err) {
+ netdev_warn(tp->dev, "EEE capability disabled\n");
+ tp->phy_flags &= ~TG3_PHYFLG_EEE_CAP;
+ } else if (!(tp->phy_flags & TG3_PHYFLG_EEE_CAP)) {
+ netdev_warn(tp->dev, "EEE capability restored\n");
+ tp->phy_flags |= TG3_PHYFLG_EEE_CAP;
+ }
+ } else if (tg3_chip_rev_id(tp) == CHIPREV_ID_5701_A0) {
if (err)
return err;
} else if (err) {
@@ -14515,6 +14666,7 @@
if (!(tp->phy_flags & TG3_PHYFLG_ANY_SERDES) &&
(tg3_asic_rev(tp) == ASIC_REV_5719 ||
tg3_asic_rev(tp) == ASIC_REV_5720 ||
+ tg3_asic_rev(tp) == ASIC_REV_57766 ||
tg3_asic_rev(tp) == ASIC_REV_5762 ||
(tg3_asic_rev(tp) == ASIC_REV_5717 &&
tg3_chip_rev_id(tp) != CHIPREV_ID_5717_A0) ||
@@ -15297,7 +15449,8 @@
} else if (tg3_asic_rev(tp) != ASIC_REV_5700 &&
tg3_asic_rev(tp) != ASIC_REV_5701 &&
tg3_chip_rev_id(tp) != CHIPREV_ID_5705_A0) {
- tg3_flag_set(tp, TSO_BUG);
+ tg3_flag_set(tp, FW_TSO);
+ tg3_flag_set(tp, TSO_BUG);
if (tg3_asic_rev(tp) == ASIC_REV_5705)
tp->fw_needed = FIRMWARE_TG3TSO5;
else
@@ -15308,7 +15461,7 @@
if (tg3_flag(tp, HW_TSO_1) ||
tg3_flag(tp, HW_TSO_2) ||
tg3_flag(tp, HW_TSO_3) ||
- tp->fw_needed) {
+ tg3_flag(tp, FW_TSO)) {
/* For firmware TSO, assume ASF is disabled.
* We'll disable TSO later if we discover ASF
* is enabled in tg3_get_eeprom_hw_cfg().
@@ -15323,6 +15476,9 @@
if (tg3_chip_rev_id(tp) == CHIPREV_ID_5701_A0)
tp->fw_needed = FIRMWARE_TG3;
+ if (tg3_asic_rev(tp) == ASIC_REV_57766)
+ tp->fw_needed = FIRMWARE_TG357766;
+
tp->irq_max = 1;
if (tg3_flag(tp, 5750_PLUS)) {
@@ -15595,7 +15751,7 @@
*/
tg3_get_eeprom_hw_cfg(tp);
- if (tp->fw_needed && tg3_flag(tp, ENABLE_ASF)) {
+ if (tg3_flag(tp, FW_TSO) && tg3_flag(tp, ENABLE_ASF)) {
tg3_flag_clear(tp, TSO_CAPABLE);
tg3_flag_clear(tp, TSO_BUG);
tp->fw_needed = NULL;
@@ -15783,6 +15939,11 @@
udelay(50);
tg3_nvram_init(tp);
+ /* If the device has an NVRAM, no need to load patch firmware */
+ if (tg3_asic_rev(tp) == ASIC_REV_57766 &&
+ !tg3_flag(tp, NO_NVRAM))
+ tp->fw_needed = NULL;
+
grc_misc_cfg = tr32(GRC_MISC_CFG);
grc_misc_cfg &= GRC_MISC_CFG_BOARD_ID_MASK;
diff --git a/drivers/net/ethernet/broadcom/tg3.h b/drivers/net/ethernet/broadcom/tg3.h
index 8d7d4c2..1cdc1b6 100644
--- a/drivers/net/ethernet/broadcom/tg3.h
+++ b/drivers/net/ethernet/broadcom/tg3.h
@@ -2222,6 +2222,12 @@
#define NIC_SRAM_MBUF_POOL_BASE5705 0x00010000
#define NIC_SRAM_MBUF_POOL_SIZE5705 0x0000e000
+#define TG3_SRAM_RXCPU_SCRATCH_BASE_57766 0x00030000
+#define TG3_SRAM_RXCPU_SCRATCH_SIZE_57766 0x00010000
+#define TG3_57766_FW_BASE_ADDR 0x00030000
+#define TG3_57766_FW_HANDSHAKE 0x0003fccc
+#define TG3_SBROM_IN_SERVICE_LOOP 0x51
+
#define TG3_SRAM_RX_STD_BDCACHE_SIZE_5700 128
#define TG3_SRAM_RX_STD_BDCACHE_SIZE_5755 64
#define TG3_SRAM_RX_STD_BDCACHE_SIZE_5906 32
@@ -3009,17 +3015,18 @@
TG3_FLAG_JUMBO_CAPABLE,
TG3_FLAG_CHIP_RESETTING,
TG3_FLAG_INIT_COMPLETE,
- TG3_FLAG_TSO_BUG,
TG3_FLAG_MAX_RXPEND_64,
- TG3_FLAG_TSO_CAPABLE,
TG3_FLAG_PCI_EXPRESS, /* BCM5785 + pci_is_pcie() */
TG3_FLAG_ASF_NEW_HANDSHAKE,
TG3_FLAG_HW_AUTONEG,
TG3_FLAG_IS_NIC,
TG3_FLAG_FLASH,
+ TG3_FLAG_FW_TSO,
TG3_FLAG_HW_TSO_1,
TG3_FLAG_HW_TSO_2,
TG3_FLAG_HW_TSO_3,
+ TG3_FLAG_TSO_CAPABLE,
+ TG3_FLAG_TSO_BUG,
TG3_FLAG_ICH_WORKAROUND,
TG3_FLAG_1SHOT_MSI,
TG3_FLAG_NO_FWARE_REPORTED,
@@ -3064,6 +3071,13 @@
TG3_FLAG_NUMBER_OF_FLAGS, /* Last entry in enum TG3_FLAGS */
};
+struct tg3_firmware_hdr {
+ __be32 version; /* unused for fragments */
+ __be32 base_addr;
+ __be32 len;
+};
+#define TG3_FW_HDR_LEN (sizeof(struct tg3_firmware_hdr))
+
struct tg3 {
/* begin "general, frequently-used members" cacheline section */
diff --git a/drivers/net/ethernet/brocade/bna/bfa_ioc.c b/drivers/net/ethernet/brocade/bna/bfa_ioc.c
index 3227fdd..f2b73ff 100644
--- a/drivers/net/ethernet/brocade/bna/bfa_ioc.c
+++ b/drivers/net/ethernet/brocade/bna/bfa_ioc.c
@@ -76,7 +76,7 @@
static void bfa_ioc_pf_failed(struct bfa_ioc *ioc);
static void bfa_ioc_pf_hwfailed(struct bfa_ioc *ioc);
static void bfa_ioc_pf_fwmismatch(struct bfa_ioc *ioc);
-static void bfa_ioc_boot(struct bfa_ioc *ioc, u32 boot_type,
+static void bfa_ioc_boot(struct bfa_ioc *ioc, enum bfi_fwboot_type boot_type,
u32 boot_param);
static u32 bfa_ioc_smem_pgnum(struct bfa_ioc *ioc, u32 fmaddr);
static void bfa_ioc_get_adapter_serial_num(struct bfa_ioc *ioc,
diff --git a/drivers/net/ethernet/brocade/bna/bnad.c b/drivers/net/ethernet/brocade/bna/bnad.c
index 7cce42d..d588f84 100644
--- a/drivers/net/ethernet/brocade/bna/bnad.c
+++ b/drivers/net/ethernet/brocade/bna/bnad.c
@@ -1264,9 +1264,8 @@
mem_info->mdl[i].len = mem_info->len;
mem_info->mdl[i].kva =
dma_alloc_coherent(&bnad->pcidev->dev,
- mem_info->len, &dma_pa,
- GFP_KERNEL);
-
+ mem_info->len, &dma_pa,
+ GFP_KERNEL);
if (mem_info->mdl[i].kva == NULL)
goto err_return;
diff --git a/drivers/net/ethernet/cadence/at91_ether.c b/drivers/net/ethernet/cadence/at91_ether.c
index 3becdb2..c6e40d6 100644
--- a/drivers/net/ethernet/cadence/at91_ether.c
+++ b/drivers/net/ethernet/cadence/at91_ether.c
@@ -47,22 +47,19 @@
int i;
lp->rx_ring = dma_alloc_coherent(&lp->pdev->dev,
- MAX_RX_DESCR * sizeof(struct macb_dma_desc),
- &lp->rx_ring_dma, GFP_KERNEL);
- if (!lp->rx_ring) {
- netdev_err(dev, "unable to alloc rx ring DMA buffer\n");
+ (MAX_RX_DESCR *
+ sizeof(struct macb_dma_desc)),
+ &lp->rx_ring_dma, GFP_KERNEL);
+ if (!lp->rx_ring)
return -ENOMEM;
- }
lp->rx_buffers = dma_alloc_coherent(&lp->pdev->dev,
- MAX_RX_DESCR * MAX_RBUFF_SZ,
- &lp->rx_buffers_dma, GFP_KERNEL);
+ MAX_RX_DESCR * MAX_RBUFF_SZ,
+ &lp->rx_buffers_dma, GFP_KERNEL);
if (!lp->rx_buffers) {
- netdev_err(dev, "unable to alloc rx data DMA buffer\n");
-
dma_free_coherent(&lp->pdev->dev,
- MAX_RX_DESCR * sizeof(struct macb_dma_desc),
- lp->rx_ring, lp->rx_ring_dma);
+ MAX_RX_DESCR * sizeof(struct macb_dma_desc),
+ lp->rx_ring, lp->rx_ring_dma);
lp->rx_ring = NULL;
return -ENOMEM;
}
@@ -209,7 +206,6 @@
netif_rx(skb);
} else {
lp->stats.rx_dropped++;
- netdev_notice(dev, "Memory squeeze, dropping packet.\n");
}
if (lp->rx_ring[lp->rx_tail].ctrl & MACB_BIT(RX_MHASH_MATCH))
@@ -519,18 +515,7 @@
},
};
-static int __init at91ether_init(void)
-{
- return platform_driver_probe(&at91ether_driver, at91ether_probe);
-}
-
-static void __exit at91ether_exit(void)
-{
- platform_driver_unregister(&at91ether_driver);
-}
-
-module_init(at91ether_init)
-module_exit(at91ether_exit)
+module_platform_driver_probe(at91ether_driver, at91ether_probe);
MODULE_LICENSE("GPL");
MODULE_DESCRIPTION("AT91RM9200 EMAC Ethernet driver");
diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c
index 7903943..3a5d680 100644
--- a/drivers/net/ethernet/cadence/macb.c
+++ b/drivers/net/ethernet/cadence/macb.c
@@ -1737,18 +1737,7 @@
},
};
-static int __init macb_init(void)
-{
- return platform_driver_probe(&macb_driver, macb_probe);
-}
-
-static void __exit macb_exit(void)
-{
- platform_driver_unregister(&macb_driver);
-}
-
-module_init(macb_init);
-module_exit(macb_exit);
+module_platform_driver_probe(macb_driver, macb_probe);
MODULE_LICENSE("GPL");
MODULE_DESCRIPTION("Cadence MACB/GEM Ethernet driver");
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
index 6db997c..681804b 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
@@ -54,6 +54,10 @@
#define FW_VERSION_MINOR 1
#define FW_VERSION_MICRO 0
+#define FW_VERSION_MAJOR_T5 0
+#define FW_VERSION_MINOR_T5 0
+#define FW_VERSION_MICRO_T5 0
+
#define CH_WARN(adap, fmt, ...) dev_warn(adap->pdev_dev, fmt, ## __VA_ARGS__)
enum {
@@ -66,7 +70,9 @@
enum {
MEM_EDC0,
MEM_EDC1,
- MEM_MC
+ MEM_MC,
+ MEM_MC0 = MEM_MC,
+ MEM_MC1
};
enum {
@@ -74,8 +80,10 @@
MEMWIN0_BASE = 0x1b800,
MEMWIN1_APERTURE = 32768,
MEMWIN1_BASE = 0x28000,
+ MEMWIN1_BASE_T5 = 0x52000,
MEMWIN2_APERTURE = 65536,
MEMWIN2_BASE = 0x30000,
+ MEMWIN2_BASE_T5 = 0x54000,
};
enum dev_master {
@@ -431,6 +439,7 @@
spinlock_t db_lock;
int db_disabled;
unsigned short db_pidx;
+ u64 udb;
};
struct sge_eth_txq { /* state for an SGE Ethernet Tx queue */
@@ -504,13 +513,44 @@
struct l2t_data;
+#define CHELSIO_CHIP_CODE(version, revision) (((version) << 4) | (revision))
+#define CHELSIO_CHIP_VERSION(code) ((code) >> 4)
+#define CHELSIO_CHIP_RELEASE(code) ((code) & 0xf)
+
+#define CHELSIO_T4 0x4
+#define CHELSIO_T5 0x5
+
+enum chip_type {
+ T4_A1 = CHELSIO_CHIP_CODE(CHELSIO_T4, 0),
+ T4_A2 = CHELSIO_CHIP_CODE(CHELSIO_T4, 1),
+ T4_A3 = CHELSIO_CHIP_CODE(CHELSIO_T4, 2),
+ T4_FIRST_REV = T4_A1,
+ T4_LAST_REV = T4_A3,
+
+ T5_A1 = CHELSIO_CHIP_CODE(CHELSIO_T5, 0),
+ T5_FIRST_REV = T5_A1,
+ T5_LAST_REV = T5_A1,
+};
+
+#ifdef CONFIG_PCI_IOV
+
+/* T4 supports SRIOV on PF0-3 and T5 on PF0-7. However, the Serial
+ * Configuration initialization for T5 only has SR-IOV functionality enabled
+ * on PF0-3 in order to simplify everything.
+ */
+#define NUM_OF_PF_WITH_SRIOV 4
+
+#endif
+
struct adapter {
void __iomem *regs;
+ void __iomem *bar2;
struct pci_dev *pdev;
struct device *pdev_dev;
unsigned int mbox;
unsigned int fn;
unsigned int flags;
+ enum chip_type chip;
int msg_enable;
@@ -673,6 +713,16 @@
VLAN_REWRITE
};
+static inline int is_t5(enum chip_type chip)
+{
+ return (chip >= T5_FIRST_REV && chip <= T5_LAST_REV);
+}
+
+static inline int is_t4(enum chip_type chip)
+{
+ return (chip >= T4_FIRST_REV && chip <= T4_LAST_REV);
+}
+
static inline u32 t4_read_reg(struct adapter *adap, u32 reg_addr)
{
return readl(adap->regs + reg_addr);
@@ -858,7 +908,8 @@
int start, int n, const u16 *rspq, unsigned int nrspq);
int t4_config_glbl_rss(struct adapter *adapter, int mbox, unsigned int mode,
unsigned int flags);
-int t4_mc_read(struct adapter *adap, u32 addr, __be32 *data, u64 *parity);
+int t4_mc_read(struct adapter *adap, int idx, u32 addr, __be32 *data,
+ u64 *parity);
int t4_edc_read(struct adapter *adap, int idx, u32 addr, __be32 *data,
u64 *parity);
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
index e707e31..e76cf03 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
@@ -68,8 +68,8 @@
#include "t4fw_api.h"
#include "l2t.h"
-#define DRV_VERSION "1.3.0-ko"
-#define DRV_DESC "Chelsio T4 Network Driver"
+#define DRV_VERSION "2.0.0-ko"
+#define DRV_DESC "Chelsio T4/T5 Network Driver"
/*
* Max interrupt hold-off timer value in us. Queues fall back to this value
@@ -229,11 +229,51 @@
CH_DEVICE(0x440a, 4),
CH_DEVICE(0x440d, 4),
CH_DEVICE(0x440e, 4),
+ CH_DEVICE(0x5001, 5),
+ CH_DEVICE(0x5002, 5),
+ CH_DEVICE(0x5003, 5),
+ CH_DEVICE(0x5004, 5),
+ CH_DEVICE(0x5005, 5),
+ CH_DEVICE(0x5006, 5),
+ CH_DEVICE(0x5007, 5),
+ CH_DEVICE(0x5008, 5),
+ CH_DEVICE(0x5009, 5),
+ CH_DEVICE(0x500A, 5),
+ CH_DEVICE(0x500B, 5),
+ CH_DEVICE(0x500C, 5),
+ CH_DEVICE(0x500D, 5),
+ CH_DEVICE(0x500E, 5),
+ CH_DEVICE(0x500F, 5),
+ CH_DEVICE(0x5010, 5),
+ CH_DEVICE(0x5011, 5),
+ CH_DEVICE(0x5012, 5),
+ CH_DEVICE(0x5013, 5),
+ CH_DEVICE(0x5401, 5),
+ CH_DEVICE(0x5402, 5),
+ CH_DEVICE(0x5403, 5),
+ CH_DEVICE(0x5404, 5),
+ CH_DEVICE(0x5405, 5),
+ CH_DEVICE(0x5406, 5),
+ CH_DEVICE(0x5407, 5),
+ CH_DEVICE(0x5408, 5),
+ CH_DEVICE(0x5409, 5),
+ CH_DEVICE(0x540A, 5),
+ CH_DEVICE(0x540B, 5),
+ CH_DEVICE(0x540C, 5),
+ CH_DEVICE(0x540D, 5),
+ CH_DEVICE(0x540E, 5),
+ CH_DEVICE(0x540F, 5),
+ CH_DEVICE(0x5410, 5),
+ CH_DEVICE(0x5411, 5),
+ CH_DEVICE(0x5412, 5),
+ CH_DEVICE(0x5413, 5),
{ 0, }
};
#define FW_FNAME "cxgb4/t4fw.bin"
+#define FW5_FNAME "cxgb4/t5fw.bin"
#define FW_CFNAME "cxgb4/t4-config.txt"
+#define FW5_CFNAME "cxgb4/t5-config.txt"
MODULE_DESCRIPTION(DRV_DESC);
MODULE_AUTHOR("Chelsio Communications");
@@ -241,6 +281,7 @@
MODULE_VERSION(DRV_VERSION);
MODULE_DEVICE_TABLE(pci, cxgb4_pci_tbl);
MODULE_FIRMWARE(FW_FNAME);
+MODULE_FIRMWARE(FW5_FNAME);
/*
* Normally we're willing to become the firmware's Master PF but will be happy
@@ -319,7 +360,10 @@
module_param(vf_acls, bool, 0644);
MODULE_PARM_DESC(vf_acls, "if set enable virtualization L2 ACL enforcement");
-static unsigned int num_vf[4];
+/* Configure the number of PCI-E Virtual Function which are to be instantiated
+ * on SR-IOV Capable Physical Functions.
+ */
+static unsigned int num_vf[NUM_OF_PF_WITH_SRIOV];
module_param_array(num_vf, uint, NULL, 0644);
MODULE_PARM_DESC(num_vf, "number of VFs for each of PFs 0-3");
@@ -1002,21 +1046,36 @@
static int upgrade_fw(struct adapter *adap)
{
int ret;
- u32 vers;
+ u32 vers, exp_major;
const struct fw_hdr *hdr;
const struct firmware *fw;
struct device *dev = adap->pdev_dev;
+ char *fw_file_name;
- ret = request_firmware(&fw, FW_FNAME, dev);
+ switch (CHELSIO_CHIP_VERSION(adap->chip)) {
+ case CHELSIO_T4:
+ fw_file_name = FW_FNAME;
+ exp_major = FW_VERSION_MAJOR;
+ break;
+ case CHELSIO_T5:
+ fw_file_name = FW5_FNAME;
+ exp_major = FW_VERSION_MAJOR_T5;
+ break;
+ default:
+ dev_err(dev, "Unsupported chip type, %x\n", adap->chip);
+ return -EINVAL;
+ }
+
+ ret = request_firmware(&fw, fw_file_name, dev);
if (ret < 0) {
- dev_err(dev, "unable to load firmware image " FW_FNAME
- ", error %d\n", ret);
+ dev_err(dev, "unable to load firmware image %s, error %d\n",
+ fw_file_name, ret);
return ret;
}
hdr = (const struct fw_hdr *)fw->data;
vers = ntohl(hdr->fw_ver);
- if (FW_HDR_FW_VER_MAJOR_GET(vers) != FW_VERSION_MAJOR) {
+ if (FW_HDR_FW_VER_MAJOR_GET(vers) != exp_major) {
ret = -EINVAL; /* wrong major version, won't do */
goto out;
}
@@ -1024,18 +1083,15 @@
/*
* If the flash FW is unusable or we found something newer, load it.
*/
- if (FW_HDR_FW_VER_MAJOR_GET(adap->params.fw_vers) != FW_VERSION_MAJOR ||
+ if (FW_HDR_FW_VER_MAJOR_GET(adap->params.fw_vers) != exp_major ||
vers > adap->params.fw_vers) {
dev_info(dev, "upgrading firmware ...\n");
ret = t4_fw_upgrade(adap, adap->mbox, fw->data, fw->size,
/*force=*/false);
if (!ret)
- dev_info(dev, "firmware successfully upgraded to "
- FW_FNAME " (%d.%d.%d.%d)\n",
- FW_HDR_FW_VER_MAJOR_GET(vers),
- FW_HDR_FW_VER_MINOR_GET(vers),
- FW_HDR_FW_VER_MICRO_GET(vers),
- FW_HDR_FW_VER_BUILD_GET(vers));
+ dev_info(dev,
+ "firmware upgraded to version %pI4 from %s\n",
+ &hdr->fw_ver, fw_file_name);
else
dev_err(dev, "firmware upgrade failed! err=%d\n", -ret);
} else {
@@ -1308,6 +1364,8 @@
"VLANinsertions ",
"GROpackets ",
"GROmerged ",
+ "WriteCoalSuccess ",
+ "WriteCoalFail ",
};
static int get_sset_count(struct net_device *dev, int sset)
@@ -1321,10 +1379,15 @@
}
#define T4_REGMAP_SIZE (160 * 1024)
+#define T5_REGMAP_SIZE (332 * 1024)
static int get_regs_len(struct net_device *dev)
{
- return T4_REGMAP_SIZE;
+ struct adapter *adap = netdev2adap(dev);
+ if (is_t4(adap->chip))
+ return T4_REGMAP_SIZE;
+ else
+ return T5_REGMAP_SIZE;
}
static int get_eeprom_len(struct net_device *dev)
@@ -1398,11 +1461,25 @@
{
struct port_info *pi = netdev_priv(dev);
struct adapter *adapter = pi->adapter;
+ u32 val1, val2;
t4_get_port_stats(adapter, pi->tx_chan, (struct port_stats *)data);
data += sizeof(struct port_stats) / sizeof(u64);
collect_sge_port_stats(adapter, pi, (struct queue_port_stats *)data);
+ data += sizeof(struct queue_port_stats) / sizeof(u64);
+ if (!is_t4(adapter->chip)) {
+ t4_write_reg(adapter, SGE_STAT_CFG, STATSOURCE_T5(7));
+ val1 = t4_read_reg(adapter, SGE_STAT_TOTAL);
+ val2 = t4_read_reg(adapter, SGE_STAT_MATCH);
+ *data = val1 - val2;
+ data++;
+ *data = val2;
+ data++;
+ } else {
+ memset(data, 0, 2 * sizeof(u64));
+ *data += 2;
+ }
}
/*
@@ -1413,7 +1490,8 @@
*/
static inline unsigned int mk_adap_vers(const struct adapter *ap)
{
- return 4 | (ap->params.rev << 10) | (1 << 16);
+ return CHELSIO_CHIP_VERSION(ap->chip) |
+ (CHELSIO_CHIP_RELEASE(ap->chip) << 10) | (1 << 16);
}
static void reg_block_dump(struct adapter *ap, void *buf, unsigned int start,
@@ -1428,7 +1506,7 @@
static void get_regs(struct net_device *dev, struct ethtool_regs *regs,
void *buf)
{
- static const unsigned int reg_ranges[] = {
+ static const unsigned int t4_reg_ranges[] = {
0x1008, 0x1108,
0x1180, 0x11b4,
0x11fc, 0x123c,
@@ -1648,13 +1726,452 @@
0x27e00, 0x27e04
};
+ static const unsigned int t5_reg_ranges[] = {
+ 0x1008, 0x1148,
+ 0x1180, 0x11b4,
+ 0x11fc, 0x123c,
+ 0x1280, 0x173c,
+ 0x1800, 0x18fc,
+ 0x3000, 0x3028,
+ 0x3060, 0x30d8,
+ 0x30e0, 0x30fc,
+ 0x3140, 0x357c,
+ 0x35a8, 0x35cc,
+ 0x35ec, 0x35ec,
+ 0x3600, 0x5624,
+ 0x56cc, 0x575c,
+ 0x580c, 0x5814,
+ 0x5890, 0x58bc,
+ 0x5940, 0x59dc,
+ 0x59fc, 0x5a18,
+ 0x5a60, 0x5a9c,
+ 0x5b9c, 0x5bfc,
+ 0x6000, 0x6040,
+ 0x6058, 0x614c,
+ 0x7700, 0x7798,
+ 0x77c0, 0x78fc,
+ 0x7b00, 0x7c54,
+ 0x7d00, 0x7efc,
+ 0x8dc0, 0x8de0,
+ 0x8df8, 0x8e84,
+ 0x8ea0, 0x8f84,
+ 0x8fc0, 0x90f8,
+ 0x9400, 0x9470,
+ 0x9600, 0x96f4,
+ 0x9800, 0x9808,
+ 0x9820, 0x983c,
+ 0x9850, 0x9864,
+ 0x9c00, 0x9c6c,
+ 0x9c80, 0x9cec,
+ 0x9d00, 0x9d6c,
+ 0x9d80, 0x9dec,
+ 0x9e00, 0x9e6c,
+ 0x9e80, 0x9eec,
+ 0x9f00, 0x9f6c,
+ 0x9f80, 0xa020,
+ 0xd004, 0xd03c,
+ 0xdfc0, 0xdfe0,
+ 0xe000, 0x11088,
+ 0x1109c, 0x1117c,
+ 0x11190, 0x11204,
+ 0x19040, 0x1906c,
+ 0x19078, 0x19080,
+ 0x1908c, 0x19124,
+ 0x19150, 0x191b0,
+ 0x191d0, 0x191e8,
+ 0x19238, 0x19290,
+ 0x193f8, 0x19474,
+ 0x19490, 0x194cc,
+ 0x194f0, 0x194f8,
+ 0x19c00, 0x19c60,
+ 0x19c94, 0x19e10,
+ 0x19e50, 0x19f34,
+ 0x19f40, 0x19f50,
+ 0x19f90, 0x19fe4,
+ 0x1a000, 0x1a06c,
+ 0x1a0b0, 0x1a120,
+ 0x1a128, 0x1a138,
+ 0x1a190, 0x1a1c4,
+ 0x1a1fc, 0x1a1fc,
+ 0x1e008, 0x1e00c,
+ 0x1e040, 0x1e04c,
+ 0x1e284, 0x1e290,
+ 0x1e2c0, 0x1e2c0,
+ 0x1e2e0, 0x1e2e0,
+ 0x1e300, 0x1e384,
+ 0x1e3c0, 0x1e3c8,
+ 0x1e408, 0x1e40c,
+ 0x1e440, 0x1e44c,
+ 0x1e684, 0x1e690,
+ 0x1e6c0, 0x1e6c0,
+ 0x1e6e0, 0x1e6e0,
+ 0x1e700, 0x1e784,
+ 0x1e7c0, 0x1e7c8,
+ 0x1e808, 0x1e80c,
+ 0x1e840, 0x1e84c,
+ 0x1ea84, 0x1ea90,
+ 0x1eac0, 0x1eac0,
+ 0x1eae0, 0x1eae0,
+ 0x1eb00, 0x1eb84,
+ 0x1ebc0, 0x1ebc8,
+ 0x1ec08, 0x1ec0c,
+ 0x1ec40, 0x1ec4c,
+ 0x1ee84, 0x1ee90,
+ 0x1eec0, 0x1eec0,
+ 0x1eee0, 0x1eee0,
+ 0x1ef00, 0x1ef84,
+ 0x1efc0, 0x1efc8,
+ 0x1f008, 0x1f00c,
+ 0x1f040, 0x1f04c,
+ 0x1f284, 0x1f290,
+ 0x1f2c0, 0x1f2c0,
+ 0x1f2e0, 0x1f2e0,
+ 0x1f300, 0x1f384,
+ 0x1f3c0, 0x1f3c8,
+ 0x1f408, 0x1f40c,
+ 0x1f440, 0x1f44c,
+ 0x1f684, 0x1f690,
+ 0x1f6c0, 0x1f6c0,
+ 0x1f6e0, 0x1f6e0,
+ 0x1f700, 0x1f784,
+ 0x1f7c0, 0x1f7c8,
+ 0x1f808, 0x1f80c,
+ 0x1f840, 0x1f84c,
+ 0x1fa84, 0x1fa90,
+ 0x1fac0, 0x1fac0,
+ 0x1fae0, 0x1fae0,
+ 0x1fb00, 0x1fb84,
+ 0x1fbc0, 0x1fbc8,
+ 0x1fc08, 0x1fc0c,
+ 0x1fc40, 0x1fc4c,
+ 0x1fe84, 0x1fe90,
+ 0x1fec0, 0x1fec0,
+ 0x1fee0, 0x1fee0,
+ 0x1ff00, 0x1ff84,
+ 0x1ffc0, 0x1ffc8,
+ 0x30000, 0x30030,
+ 0x30100, 0x30144,
+ 0x30190, 0x301d0,
+ 0x30200, 0x30318,
+ 0x30400, 0x3052c,
+ 0x30540, 0x3061c,
+ 0x30800, 0x30834,
+ 0x308c0, 0x30908,
+ 0x30910, 0x309ac,
+ 0x30a00, 0x30a04,
+ 0x30a0c, 0x30a2c,
+ 0x30a44, 0x30a50,
+ 0x30a74, 0x30c24,
+ 0x30d08, 0x30d14,
+ 0x30d1c, 0x30d20,
+ 0x30d3c, 0x30d50,
+ 0x31200, 0x3120c,
+ 0x31220, 0x31220,
+ 0x31240, 0x31240,
+ 0x31600, 0x31600,
+ 0x31608, 0x3160c,
+ 0x31a00, 0x31a1c,
+ 0x31e04, 0x31e20,
+ 0x31e38, 0x31e3c,
+ 0x31e80, 0x31e80,
+ 0x31e88, 0x31ea8,
+ 0x31eb0, 0x31eb4,
+ 0x31ec8, 0x31ed4,
+ 0x31fb8, 0x32004,
+ 0x32208, 0x3223c,
+ 0x32600, 0x32630,
+ 0x32a00, 0x32abc,
+ 0x32b00, 0x32b70,
+ 0x33000, 0x33048,
+ 0x33060, 0x3309c,
+ 0x330f0, 0x33148,
+ 0x33160, 0x3319c,
+ 0x331f0, 0x332e4,
+ 0x332f8, 0x333e4,
+ 0x333f8, 0x33448,
+ 0x33460, 0x3349c,
+ 0x334f0, 0x33548,
+ 0x33560, 0x3359c,
+ 0x335f0, 0x336e4,
+ 0x336f8, 0x337e4,
+ 0x337f8, 0x337fc,
+ 0x33814, 0x33814,
+ 0x3382c, 0x3382c,
+ 0x33880, 0x3388c,
+ 0x338e8, 0x338ec,
+ 0x33900, 0x33948,
+ 0x33960, 0x3399c,
+ 0x339f0, 0x33ae4,
+ 0x33af8, 0x33b10,
+ 0x33b28, 0x33b28,
+ 0x33b3c, 0x33b50,
+ 0x33bf0, 0x33c10,
+ 0x33c28, 0x33c28,
+ 0x33c3c, 0x33c50,
+ 0x33cf0, 0x33cfc,
+ 0x34000, 0x34030,
+ 0x34100, 0x34144,
+ 0x34190, 0x341d0,
+ 0x34200, 0x34318,
+ 0x34400, 0x3452c,
+ 0x34540, 0x3461c,
+ 0x34800, 0x34834,
+ 0x348c0, 0x34908,
+ 0x34910, 0x349ac,
+ 0x34a00, 0x34a04,
+ 0x34a0c, 0x34a2c,
+ 0x34a44, 0x34a50,
+ 0x34a74, 0x34c24,
+ 0x34d08, 0x34d14,
+ 0x34d1c, 0x34d20,
+ 0x34d3c, 0x34d50,
+ 0x35200, 0x3520c,
+ 0x35220, 0x35220,
+ 0x35240, 0x35240,
+ 0x35600, 0x35600,
+ 0x35608, 0x3560c,
+ 0x35a00, 0x35a1c,
+ 0x35e04, 0x35e20,
+ 0x35e38, 0x35e3c,
+ 0x35e80, 0x35e80,
+ 0x35e88, 0x35ea8,
+ 0x35eb0, 0x35eb4,
+ 0x35ec8, 0x35ed4,
+ 0x35fb8, 0x36004,
+ 0x36208, 0x3623c,
+ 0x36600, 0x36630,
+ 0x36a00, 0x36abc,
+ 0x36b00, 0x36b70,
+ 0x37000, 0x37048,
+ 0x37060, 0x3709c,
+ 0x370f0, 0x37148,
+ 0x37160, 0x3719c,
+ 0x371f0, 0x372e4,
+ 0x372f8, 0x373e4,
+ 0x373f8, 0x37448,
+ 0x37460, 0x3749c,
+ 0x374f0, 0x37548,
+ 0x37560, 0x3759c,
+ 0x375f0, 0x376e4,
+ 0x376f8, 0x377e4,
+ 0x377f8, 0x377fc,
+ 0x37814, 0x37814,
+ 0x3782c, 0x3782c,
+ 0x37880, 0x3788c,
+ 0x378e8, 0x378ec,
+ 0x37900, 0x37948,
+ 0x37960, 0x3799c,
+ 0x379f0, 0x37ae4,
+ 0x37af8, 0x37b10,
+ 0x37b28, 0x37b28,
+ 0x37b3c, 0x37b50,
+ 0x37bf0, 0x37c10,
+ 0x37c28, 0x37c28,
+ 0x37c3c, 0x37c50,
+ 0x37cf0, 0x37cfc,
+ 0x38000, 0x38030,
+ 0x38100, 0x38144,
+ 0x38190, 0x381d0,
+ 0x38200, 0x38318,
+ 0x38400, 0x3852c,
+ 0x38540, 0x3861c,
+ 0x38800, 0x38834,
+ 0x388c0, 0x38908,
+ 0x38910, 0x389ac,
+ 0x38a00, 0x38a04,
+ 0x38a0c, 0x38a2c,
+ 0x38a44, 0x38a50,
+ 0x38a74, 0x38c24,
+ 0x38d08, 0x38d14,
+ 0x38d1c, 0x38d20,
+ 0x38d3c, 0x38d50,
+ 0x39200, 0x3920c,
+ 0x39220, 0x39220,
+ 0x39240, 0x39240,
+ 0x39600, 0x39600,
+ 0x39608, 0x3960c,
+ 0x39a00, 0x39a1c,
+ 0x39e04, 0x39e20,
+ 0x39e38, 0x39e3c,
+ 0x39e80, 0x39e80,
+ 0x39e88, 0x39ea8,
+ 0x39eb0, 0x39eb4,
+ 0x39ec8, 0x39ed4,
+ 0x39fb8, 0x3a004,
+ 0x3a208, 0x3a23c,
+ 0x3a600, 0x3a630,
+ 0x3aa00, 0x3aabc,
+ 0x3ab00, 0x3ab70,
+ 0x3b000, 0x3b048,
+ 0x3b060, 0x3b09c,
+ 0x3b0f0, 0x3b148,
+ 0x3b160, 0x3b19c,
+ 0x3b1f0, 0x3b2e4,
+ 0x3b2f8, 0x3b3e4,
+ 0x3b3f8, 0x3b448,
+ 0x3b460, 0x3b49c,
+ 0x3b4f0, 0x3b548,
+ 0x3b560, 0x3b59c,
+ 0x3b5f0, 0x3b6e4,
+ 0x3b6f8, 0x3b7e4,
+ 0x3b7f8, 0x3b7fc,
+ 0x3b814, 0x3b814,
+ 0x3b82c, 0x3b82c,
+ 0x3b880, 0x3b88c,
+ 0x3b8e8, 0x3b8ec,
+ 0x3b900, 0x3b948,
+ 0x3b960, 0x3b99c,
+ 0x3b9f0, 0x3bae4,
+ 0x3baf8, 0x3bb10,
+ 0x3bb28, 0x3bb28,
+ 0x3bb3c, 0x3bb50,
+ 0x3bbf0, 0x3bc10,
+ 0x3bc28, 0x3bc28,
+ 0x3bc3c, 0x3bc50,
+ 0x3bcf0, 0x3bcfc,
+ 0x3c000, 0x3c030,
+ 0x3c100, 0x3c144,
+ 0x3c190, 0x3c1d0,
+ 0x3c200, 0x3c318,
+ 0x3c400, 0x3c52c,
+ 0x3c540, 0x3c61c,
+ 0x3c800, 0x3c834,
+ 0x3c8c0, 0x3c908,
+ 0x3c910, 0x3c9ac,
+ 0x3ca00, 0x3ca04,
+ 0x3ca0c, 0x3ca2c,
+ 0x3ca44, 0x3ca50,
+ 0x3ca74, 0x3cc24,
+ 0x3cd08, 0x3cd14,
+ 0x3cd1c, 0x3cd20,
+ 0x3cd3c, 0x3cd50,
+ 0x3d200, 0x3d20c,
+ 0x3d220, 0x3d220,
+ 0x3d240, 0x3d240,
+ 0x3d600, 0x3d600,
+ 0x3d608, 0x3d60c,
+ 0x3da00, 0x3da1c,
+ 0x3de04, 0x3de20,
+ 0x3de38, 0x3de3c,
+ 0x3de80, 0x3de80,
+ 0x3de88, 0x3dea8,
+ 0x3deb0, 0x3deb4,
+ 0x3dec8, 0x3ded4,
+ 0x3dfb8, 0x3e004,
+ 0x3e208, 0x3e23c,
+ 0x3e600, 0x3e630,
+ 0x3ea00, 0x3eabc,
+ 0x3eb00, 0x3eb70,
+ 0x3f000, 0x3f048,
+ 0x3f060, 0x3f09c,
+ 0x3f0f0, 0x3f148,
+ 0x3f160, 0x3f19c,
+ 0x3f1f0, 0x3f2e4,
+ 0x3f2f8, 0x3f3e4,
+ 0x3f3f8, 0x3f448,
+ 0x3f460, 0x3f49c,
+ 0x3f4f0, 0x3f548,
+ 0x3f560, 0x3f59c,
+ 0x3f5f0, 0x3f6e4,
+ 0x3f6f8, 0x3f7e4,
+ 0x3f7f8, 0x3f7fc,
+ 0x3f814, 0x3f814,
+ 0x3f82c, 0x3f82c,
+ 0x3f880, 0x3f88c,
+ 0x3f8e8, 0x3f8ec,
+ 0x3f900, 0x3f948,
+ 0x3f960, 0x3f99c,
+ 0x3f9f0, 0x3fae4,
+ 0x3faf8, 0x3fb10,
+ 0x3fb28, 0x3fb28,
+ 0x3fb3c, 0x3fb50,
+ 0x3fbf0, 0x3fc10,
+ 0x3fc28, 0x3fc28,
+ 0x3fc3c, 0x3fc50,
+ 0x3fcf0, 0x3fcfc,
+ 0x40000, 0x4000c,
+ 0x40040, 0x40068,
+ 0x40080, 0x40144,
+ 0x40180, 0x4018c,
+ 0x40200, 0x40298,
+ 0x402ac, 0x4033c,
+ 0x403f8, 0x403fc,
+ 0x41300, 0x413c4,
+ 0x41400, 0x4141c,
+ 0x41480, 0x414d0,
+ 0x44000, 0x44078,
+ 0x440c0, 0x44278,
+ 0x442c0, 0x44478,
+ 0x444c0, 0x44678,
+ 0x446c0, 0x44878,
+ 0x448c0, 0x449fc,
+ 0x45000, 0x45068,
+ 0x45080, 0x45084,
+ 0x450a0, 0x450b0,
+ 0x45200, 0x45268,
+ 0x45280, 0x45284,
+ 0x452a0, 0x452b0,
+ 0x460c0, 0x460e4,
+ 0x47000, 0x4708c,
+ 0x47200, 0x47250,
+ 0x47400, 0x47420,
+ 0x47600, 0x47618,
+ 0x47800, 0x47814,
+ 0x48000, 0x4800c,
+ 0x48040, 0x48068,
+ 0x48080, 0x48144,
+ 0x48180, 0x4818c,
+ 0x48200, 0x48298,
+ 0x482ac, 0x4833c,
+ 0x483f8, 0x483fc,
+ 0x49300, 0x493c4,
+ 0x49400, 0x4941c,
+ 0x49480, 0x494d0,
+ 0x4c000, 0x4c078,
+ 0x4c0c0, 0x4c278,
+ 0x4c2c0, 0x4c478,
+ 0x4c4c0, 0x4c678,
+ 0x4c6c0, 0x4c878,
+ 0x4c8c0, 0x4c9fc,
+ 0x4d000, 0x4d068,
+ 0x4d080, 0x4d084,
+ 0x4d0a0, 0x4d0b0,
+ 0x4d200, 0x4d268,
+ 0x4d280, 0x4d284,
+ 0x4d2a0, 0x4d2b0,
+ 0x4e0c0, 0x4e0e4,
+ 0x4f000, 0x4f08c,
+ 0x4f200, 0x4f250,
+ 0x4f400, 0x4f420,
+ 0x4f600, 0x4f618,
+ 0x4f800, 0x4f814,
+ 0x50000, 0x500cc,
+ 0x50400, 0x50400,
+ 0x50800, 0x508cc,
+ 0x50c00, 0x50c00,
+ 0x51000, 0x5101c,
+ 0x51300, 0x51308,
+ };
+
int i;
struct adapter *ap = netdev2adap(dev);
+ static const unsigned int *reg_ranges;
+ int arr_size = 0, buf_size = 0;
+
+ if (is_t4(ap->chip)) {
+ reg_ranges = &t4_reg_ranges[0];
+ arr_size = ARRAY_SIZE(t4_reg_ranges);
+ buf_size = T4_REGMAP_SIZE;
+ } else {
+ reg_ranges = &t5_reg_ranges[0];
+ arr_size = ARRAY_SIZE(t5_reg_ranges);
+ buf_size = T5_REGMAP_SIZE;
+ }
regs->version = mk_adap_vers(ap);
- memset(buf, 0, T4_REGMAP_SIZE);
- for (i = 0; i < ARRAY_SIZE(reg_ranges); i += 2)
+ memset(buf, 0, buf_size);
+ for (i = 0; i < arr_size; i += 2)
reg_block_dump(ap, buf, reg_ranges[i], reg_ranges[i + 1]);
}
@@ -2363,8 +2880,8 @@
int ret, ofst;
__be32 data[16];
- if (mem == MEM_MC)
- ret = t4_mc_read(adap, pos, data, NULL);
+ if ((mem == MEM_MC) || (mem == MEM_MC1))
+ ret = t4_mc_read(adap, mem % MEM_MC, pos, data, NULL);
else
ret = t4_edc_read(adap, mem, pos, data, NULL);
if (ret)
@@ -2405,18 +2922,37 @@
static int setup_debugfs(struct adapter *adap)
{
int i;
+ u32 size;
if (IS_ERR_OR_NULL(adap->debugfs_root))
return -1;
i = t4_read_reg(adap, MA_TARGET_MEM_ENABLE);
- if (i & EDRAM0_ENABLE)
- add_debugfs_mem(adap, "edc0", MEM_EDC0, 5);
- if (i & EDRAM1_ENABLE)
- add_debugfs_mem(adap, "edc1", MEM_EDC1, 5);
- if (i & EXT_MEM_ENABLE)
- add_debugfs_mem(adap, "mc", MEM_MC,
- EXT_MEM_SIZE_GET(t4_read_reg(adap, MA_EXT_MEMORY_BAR)));
+ if (i & EDRAM0_ENABLE) {
+ size = t4_read_reg(adap, MA_EDRAM0_BAR);
+ add_debugfs_mem(adap, "edc0", MEM_EDC0, EDRAM_SIZE_GET(size));
+ }
+ if (i & EDRAM1_ENABLE) {
+ size = t4_read_reg(adap, MA_EDRAM1_BAR);
+ add_debugfs_mem(adap, "edc1", MEM_EDC1, EDRAM_SIZE_GET(size));
+ }
+ if (is_t4(adap->chip)) {
+ size = t4_read_reg(adap, MA_EXT_MEMORY_BAR);
+ if (i & EXT_MEM_ENABLE)
+ add_debugfs_mem(adap, "mc", MEM_MC,
+ EXT_MEM_SIZE_GET(size));
+ } else {
+ if (i & EXT_MEM_ENABLE) {
+ size = t4_read_reg(adap, MA_EXT_MEMORY_BAR);
+ add_debugfs_mem(adap, "mc0", MEM_MC0,
+ EXT_MEM_SIZE_GET(size));
+ }
+ if (i & EXT_MEM1_ENABLE) {
+ size = t4_read_reg(adap, MA_EXT_MEMORY1_BAR);
+ add_debugfs_mem(adap, "mc1", MEM_MC1,
+ EXT_MEM_SIZE_GET(size));
+ }
+ }
if (adap->l2t)
debugfs_create_file("l2t", S_IRUSR, adap->debugfs_root, adap,
&t4_l2t_fops);
@@ -2747,10 +3283,18 @@
unsigned int cxgb4_dbfifo_count(const struct net_device *dev, int lpfifo)
{
struct adapter *adap = netdev2adap(dev);
- u32 v;
+ u32 v1, v2, lp_count, hp_count;
- v = t4_read_reg(adap, A_SGE_DBFIFO_STATUS);
- return lpfifo ? G_LP_COUNT(v) : G_HP_COUNT(v);
+ v1 = t4_read_reg(adap, A_SGE_DBFIFO_STATUS);
+ v2 = t4_read_reg(adap, SGE_DBFIFO_STATUS2);
+ if (is_t4(adap->chip)) {
+ lp_count = G_LP_COUNT(v1);
+ hp_count = G_HP_COUNT(v1);
+ } else {
+ lp_count = G_LP_COUNT_T5(v1);
+ hp_count = G_HP_COUNT_T5(v2);
+ }
+ return lpfifo ? lp_count : hp_count;
}
EXPORT_SYMBOL(cxgb4_dbfifo_count);
@@ -2853,6 +3397,25 @@
}
EXPORT_SYMBOL(cxgb4_sync_txq_pidx);
+void cxgb4_disable_db_coalescing(struct net_device *dev)
+{
+ struct adapter *adap;
+
+ adap = netdev2adap(dev);
+ t4_set_reg_field(adap, A_SGE_DOORBELL_CONTROL, F_NOCOALESCE,
+ F_NOCOALESCE);
+}
+EXPORT_SYMBOL(cxgb4_disable_db_coalescing);
+
+void cxgb4_enable_db_coalescing(struct net_device *dev)
+{
+ struct adapter *adap;
+
+ adap = netdev2adap(dev);
+ t4_set_reg_field(adap, A_SGE_DOORBELL_CONTROL, F_NOCOALESCE, 0);
+}
+EXPORT_SYMBOL(cxgb4_enable_db_coalescing);
+
static struct pci_driver cxgb4_driver;
static void check_neigh_update(struct neighbour *neigh)
@@ -2888,14 +3451,23 @@
static void drain_db_fifo(struct adapter *adap, int usecs)
{
- u32 v;
+ u32 v1, v2, lp_count, hp_count;
do {
+ v1 = t4_read_reg(adap, A_SGE_DBFIFO_STATUS);
+ v2 = t4_read_reg(adap, SGE_DBFIFO_STATUS2);
+ if (is_t4(adap->chip)) {
+ lp_count = G_LP_COUNT(v1);
+ hp_count = G_HP_COUNT(v1);
+ } else {
+ lp_count = G_LP_COUNT_T5(v1);
+ hp_count = G_HP_COUNT_T5(v2);
+ }
+
+ if (lp_count == 0 && hp_count == 0)
+ break;
set_current_state(TASK_UNINTERRUPTIBLE);
schedule_timeout(usecs_to_jiffies(usecs));
- v = t4_read_reg(adap, A_SGE_DBFIFO_STATUS);
- if (G_LP_COUNT(v) == 0 && G_HP_COUNT(v) == 0)
- break;
} while (1);
}
@@ -3004,24 +3576,62 @@
adap = container_of(work, struct adapter, db_drop_task);
+ if (is_t4(adap->chip)) {
+ disable_dbs(adap);
+ notify_rdma_uld(adap, CXGB4_CONTROL_DB_DROP);
+ drain_db_fifo(adap, 1);
+ recover_all_queues(adap);
+ enable_dbs(adap);
+ } else {
+ u32 dropped_db = t4_read_reg(adap, 0x010ac);
+ u16 qid = (dropped_db >> 15) & 0x1ffff;
+ u16 pidx_inc = dropped_db & 0x1fff;
+ unsigned int s_qpp;
+ unsigned short udb_density;
+ unsigned long qpshift;
+ int page;
+ u32 udb;
+
+ dev_warn(adap->pdev_dev,
+ "Dropped DB 0x%x qid %d bar2 %d coalesce %d pidx %d\n",
+ dropped_db, qid,
+ (dropped_db >> 14) & 1,
+ (dropped_db >> 13) & 1,
+ pidx_inc);
+
+ drain_db_fifo(adap, 1);
+
+ s_qpp = QUEUESPERPAGEPF1 * adap->fn;
+ udb_density = 1 << QUEUESPERPAGEPF0_GET(t4_read_reg(adap,
+ SGE_EGRESS_QUEUES_PER_PAGE_PF) >> s_qpp);
+ qpshift = PAGE_SHIFT - ilog2(udb_density);
+ udb = qid << qpshift;
+ udb &= PAGE_MASK;
+ page = udb / PAGE_SIZE;
+ udb += (qid - (page * udb_density)) * 128;
+
+ writel(PIDX(pidx_inc), adap->bar2 + udb + 8);
+
+ /* Re-enable BAR2 WC */
+ t4_set_reg_field(adap, 0x10b0, 1<<15, 1<<15);
+ }
+
t4_set_reg_field(adap, A_SGE_DOORBELL_CONTROL, F_DROPPED_DB, 0);
- disable_dbs(adap);
- notify_rdma_uld(adap, CXGB4_CONTROL_DB_DROP);
- drain_db_fifo(adap, 1);
- recover_all_queues(adap);
- enable_dbs(adap);
}
void t4_db_full(struct adapter *adap)
{
- t4_set_reg_field(adap, SGE_INT_ENABLE3,
- DBFIFO_HP_INT | DBFIFO_LP_INT, 0);
- queue_work(workq, &adap->db_full_task);
+ if (is_t4(adap->chip)) {
+ t4_set_reg_field(adap, SGE_INT_ENABLE3,
+ DBFIFO_HP_INT | DBFIFO_LP_INT, 0);
+ queue_work(workq, &adap->db_full_task);
+ }
}
void t4_db_dropped(struct adapter *adap)
{
- queue_work(workq, &adap->db_drop_task);
+ if (is_t4(adap->chip))
+ queue_work(workq, &adap->db_drop_task);
}
static void uld_attach(struct adapter *adap, unsigned int uld)
@@ -3566,17 +4176,27 @@
static void setup_memwin(struct adapter *adap)
{
- u32 bar0;
+ u32 bar0, mem_win0_base, mem_win1_base, mem_win2_base;
bar0 = pci_resource_start(adap->pdev, 0); /* truncation intentional */
+ if (is_t4(adap->chip)) {
+ mem_win0_base = bar0 + MEMWIN0_BASE;
+ mem_win1_base = bar0 + MEMWIN1_BASE;
+ mem_win2_base = bar0 + MEMWIN2_BASE;
+ } else {
+ /* For T5, only relative offset inside the PCIe BAR is passed */
+ mem_win0_base = MEMWIN0_BASE;
+ mem_win1_base = MEMWIN1_BASE_T5;
+ mem_win2_base = MEMWIN2_BASE_T5;
+ }
t4_write_reg(adap, PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_BASE_WIN, 0),
- (bar0 + MEMWIN0_BASE) | BIR(0) |
+ mem_win0_base | BIR(0) |
WINDOW(ilog2(MEMWIN0_APERTURE) - 10));
t4_write_reg(adap, PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_BASE_WIN, 1),
- (bar0 + MEMWIN1_BASE) | BIR(0) |
+ mem_win1_base | BIR(0) |
WINDOW(ilog2(MEMWIN1_APERTURE) - 10));
t4_write_reg(adap, PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_BASE_WIN, 2),
- (bar0 + MEMWIN2_BASE) | BIR(0) |
+ mem_win2_base | BIR(0) |
WINDOW(ilog2(MEMWIN2_APERTURE) - 10));
}
@@ -3745,6 +4365,7 @@
unsigned long mtype = 0, maddr = 0;
u32 finiver, finicsum, cfcsum;
int ret, using_flash;
+ char *fw_config_file, fw_config_file_path[256];
/*
* Reset device if necessary.
@@ -3761,7 +4382,21 @@
* then use that. Otherwise, use the configuration file stored
* in the adapter flash ...
*/
- ret = request_firmware(&cf, FW_CFNAME, adapter->pdev_dev);
+ switch (CHELSIO_CHIP_VERSION(adapter->chip)) {
+ case CHELSIO_T4:
+ fw_config_file = FW_CFNAME;
+ break;
+ case CHELSIO_T5:
+ fw_config_file = FW5_CFNAME;
+ break;
+ default:
+ dev_err(adapter->pdev_dev, "Device %d is not supported\n",
+ adapter->pdev->device);
+ ret = -EINVAL;
+ goto bye;
+ }
+
+ ret = request_firmware(&cf, fw_config_file, adapter->pdev_dev);
if (ret < 0) {
using_flash = 1;
mtype = FW_MEMTYPE_CF_FLASH;
@@ -3877,6 +4512,7 @@
if (ret < 0)
goto bye;
+ sprintf(fw_config_file_path, "/lib/firmware/%s", fw_config_file);
/*
* Return successfully and note that we're operating with parameters
* not supplied by the driver, rather than from hard-wired
@@ -3887,7 +4523,7 @@
"Configuration File %s, version %#x, computed checksum %#x\n",
(using_flash
? "in device FLASH"
- : "/lib/firmware/" FW_CFNAME),
+ : fw_config_file_path),
finiver, cfcsum);
return 0;
@@ -4814,7 +5450,8 @@
sprintf(bufp, "BASE-%s", base[pi->port_type]);
netdev_info(dev, "Chelsio %s rev %d %s %sNIC PCIe x%d%s%s\n",
- adap->params.vpd.id, adap->params.rev, buf,
+ adap->params.vpd.id,
+ CHELSIO_CHIP_RELEASE(adap->params.rev), buf,
is_offload(adap) ? "R" : "", adap->params.pci.width, spd,
(adap->flags & USING_MSIX) ? " MSI-X" :
(adap->flags & USING_MSI) ? " MSI" : "");
@@ -4854,10 +5491,11 @@
#define TSO_FLAGS (NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_TSO_ECN)
#define VLAN_FEAT (NETIF_F_SG | NETIF_F_IP_CSUM | TSO_FLAGS | \
NETIF_F_IPV6_CSUM | NETIF_F_HIGHDMA)
+#define SEGMENT_SIZE 128
static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
{
- int func, i, err;
+ int func, i, err, s_qpp, qpp, num_seg;
struct port_info *pi;
bool highdma = false;
struct adapter *adapter = NULL;
@@ -4934,7 +5572,34 @@
err = t4_prep_adapter(adapter);
if (err)
- goto out_unmap_bar;
+ goto out_unmap_bar0;
+
+ if (!is_t4(adapter->chip)) {
+ s_qpp = QUEUESPERPAGEPF1 * adapter->fn;
+ qpp = 1 << QUEUESPERPAGEPF0_GET(t4_read_reg(adapter,
+ SGE_EGRESS_QUEUES_PER_PAGE_PF) >> s_qpp);
+ num_seg = PAGE_SIZE / SEGMENT_SIZE;
+
+ /* Each segment size is 128B. Write coalescing is enabled only
+ * when SGE_EGRESS_QUEUES_PER_PAGE_PF reg value for the
+ * queue is less no of segments that can be accommodated in
+ * a page size.
+ */
+ if (qpp > num_seg) {
+ dev_err(&pdev->dev,
+ "Incorrect number of egress queues per page\n");
+ err = -EINVAL;
+ goto out_unmap_bar0;
+ }
+ adapter->bar2 = ioremap_wc(pci_resource_start(pdev, 2),
+ pci_resource_len(pdev, 2));
+ if (!adapter->bar2) {
+ dev_err(&pdev->dev, "cannot map device bar2 region\n");
+ err = -ENOMEM;
+ goto out_unmap_bar0;
+ }
+ }
+
setup_memwin(adapter);
err = adap_init0(adapter);
setup_memwin_rdma(adapter);
@@ -5063,6 +5728,9 @@
out_free_dev:
free_some_resources(adapter);
out_unmap_bar:
+ if (!is_t4(adapter->chip))
+ iounmap(adapter->bar2);
+ out_unmap_bar0:
iounmap(adapter->regs);
out_free_adapter:
kfree(adapter);
@@ -5113,6 +5781,8 @@
free_some_resources(adapter);
iounmap(adapter->regs);
+ if (!is_t4(adapter->chip))
+ iounmap(adapter->bar2);
kfree(adapter);
pci_disable_pcie_error_reporting(pdev);
pci_disable_device(pdev);
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
index e2bbc7f3e..4faf4d0 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
@@ -269,4 +269,7 @@
unsigned int skb_len, unsigned int pull_len);
int cxgb4_sync_txq_pidx(struct net_device *dev, u16 qid, u16 pidx, u16 size);
int cxgb4_flush_eq_cache(struct net_device *dev);
+void cxgb4_disable_db_coalescing(struct net_device *dev);
+void cxgb4_enable_db_coalescing(struct net_device *dev);
+
#endif /* !__CXGB4_OFLD_H */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/sge.c b/drivers/net/ethernet/chelsio/cxgb4/sge.c
index fe9a2ea..8b47b25 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/sge.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/sge.c
@@ -506,10 +506,14 @@
static inline void ring_fl_db(struct adapter *adap, struct sge_fl *q)
{
+ u32 val;
if (q->pend_cred >= 8) {
+ val = PIDX(q->pend_cred / 8);
+ if (!is_t4(adap->chip))
+ val |= DBTYPE(1);
wmb();
t4_write_reg(adap, MYPF_REG(SGE_PF_KDOORBELL), DBPRIO(1) |
- QID(q->cntxt_id) | PIDX(q->pend_cred / 8));
+ QID(q->cntxt_id) | val);
q->pend_cred &= 7;
}
}
@@ -812,6 +816,22 @@
*end = 0;
}
+/* This function copies 64 byte coalesced work request to
+ * memory mapped BAR2 space(user space writes).
+ * For coalesced WR SGE, fetches data from the FIFO instead of from Host.
+ */
+static void cxgb_pio_copy(u64 __iomem *dst, u64 *src)
+{
+ int count = 8;
+
+ while (count) {
+ writeq(*src, dst);
+ src++;
+ dst++;
+ count--;
+ }
+}
+
/**
* ring_tx_db - check and potentially ring a Tx queue's doorbell
* @adap: the adapter
@@ -822,11 +842,25 @@
*/
static inline void ring_tx_db(struct adapter *adap, struct sge_txq *q, int n)
{
+ unsigned int *wr, index;
+
wmb(); /* write descriptors before telling HW */
spin_lock(&q->db_lock);
if (!q->db_disabled) {
- t4_write_reg(adap, MYPF_REG(SGE_PF_KDOORBELL),
- QID(q->cntxt_id) | PIDX(n));
+ if (is_t4(adap->chip)) {
+ t4_write_reg(adap, MYPF_REG(SGE_PF_KDOORBELL),
+ QID(q->cntxt_id) | PIDX(n));
+ } else {
+ if (n == 1) {
+ index = q->pidx ? (q->pidx - 1) : (q->size - 1);
+ wr = (unsigned int *)&q->desc[index];
+ cxgb_pio_copy((u64 __iomem *)
+ (adap->bar2 + q->udb + 64),
+ (u64 *)wr);
+ } else
+ writel(n, adap->bar2 + q->udb + 8);
+ wmb();
+ }
}
q->db_pidx = q->pidx;
spin_unlock(&q->db_lock);
@@ -1555,7 +1589,6 @@
const struct pkt_gl *gl)
{
struct sk_buff *skb;
- struct cpl_trace_pkt *p;
skb = cxgb4_pktgl_to_skb(gl, RX_PULL_LEN, RX_PULL_LEN);
if (unlikely(!skb)) {
@@ -1563,8 +1596,11 @@
return 0;
}
- p = (struct cpl_trace_pkt *)skb->data;
- __skb_pull(skb, sizeof(*p));
+ if (is_t4(adap->chip))
+ __skb_pull(skb, sizeof(struct cpl_trace_pkt));
+ else
+ __skb_pull(skb, sizeof(struct cpl_t5_trace_pkt));
+
skb_reset_mac_header(skb);
skb->protocol = htons(0xffff);
skb->dev = adap->port[0];
@@ -1625,8 +1661,10 @@
const struct cpl_rx_pkt *pkt;
struct sge_eth_rxq *rxq = container_of(q, struct sge_eth_rxq, rspq);
struct sge *s = &q->adap->sge;
+ int cpl_trace_pkt = is_t4(q->adap->chip) ?
+ CPL_TRACE_PKT : CPL_TRACE_PKT_T5;
- if (unlikely(*(u8 *)rsp == CPL_TRACE_PKT))
+ if (unlikely(*(u8 *)rsp == cpl_trace_pkt))
return handle_trace_pkt(q->adap, si);
pkt = (const struct cpl_rx_pkt *)rsp;
@@ -2143,11 +2181,27 @@
static void init_txq(struct adapter *adap, struct sge_txq *q, unsigned int id)
{
+ q->cntxt_id = id;
+ if (!is_t4(adap->chip)) {
+ unsigned int s_qpp;
+ unsigned short udb_density;
+ unsigned long qpshift;
+ int page;
+
+ s_qpp = QUEUESPERPAGEPF1 * adap->fn;
+ udb_density = 1 << QUEUESPERPAGEPF0_GET((t4_read_reg(adap,
+ SGE_EGRESS_QUEUES_PER_PAGE_PF) >> s_qpp));
+ qpshift = PAGE_SHIFT - ilog2(udb_density);
+ q->udb = q->cntxt_id << qpshift;
+ q->udb &= PAGE_MASK;
+ page = q->udb / PAGE_SIZE;
+ q->udb += (q->cntxt_id - (page * udb_density)) * 128;
+ }
+
q->in_use = 0;
q->cidx = q->pidx = 0;
q->stops = q->restarts = 0;
q->stat = (void *)&q->desc[q->size];
- q->cntxt_id = id;
spin_lock_init(&q->db_lock);
adap->sge.egr_map[id - adap->sge.egr_start] = q;
}
@@ -2587,11 +2641,20 @@
* Set up to drop DOORBELL writes when the DOORBELL FIFO overflows
* and generate an interrupt when this occurs so we can recover.
*/
- t4_set_reg_field(adap, A_SGE_DBFIFO_STATUS,
- V_HP_INT_THRESH(M_HP_INT_THRESH) |
- V_LP_INT_THRESH(M_LP_INT_THRESH),
- V_HP_INT_THRESH(dbfifo_int_thresh) |
- V_LP_INT_THRESH(dbfifo_int_thresh));
+ if (is_t4(adap->chip)) {
+ t4_set_reg_field(adap, A_SGE_DBFIFO_STATUS,
+ V_HP_INT_THRESH(M_HP_INT_THRESH) |
+ V_LP_INT_THRESH(M_LP_INT_THRESH),
+ V_HP_INT_THRESH(dbfifo_int_thresh) |
+ V_LP_INT_THRESH(dbfifo_int_thresh));
+ } else {
+ t4_set_reg_field(adap, A_SGE_DBFIFO_STATUS,
+ V_LP_INT_THRESH_T5(M_LP_INT_THRESH_T5),
+ V_LP_INT_THRESH_T5(dbfifo_int_thresh));
+ t4_set_reg_field(adap, SGE_DBFIFO_STATUS2,
+ V_HP_INT_THRESH_T5(M_HP_INT_THRESH_T5),
+ V_HP_INT_THRESH_T5(dbfifo_int_thresh));
+ }
t4_set_reg_field(adap, A_SGE_DOORBELL_CONTROL, F_ENABLE_DROP,
F_ENABLE_DROP);
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
index 8049268..d02d4e8 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
@@ -282,6 +282,7 @@
* t4_mc_read - read from MC through backdoor accesses
* @adap: the adapter
* @addr: address of first byte requested
+ * @idx: which MC to access
* @data: 64 bytes of data containing the requested address
* @ecc: where to store the corresponding 64-bit ECC word
*
@@ -289,22 +290,38 @@
* that covers the requested address @addr. If @parity is not %NULL it
* is assigned the 64-bit ECC word for the read data.
*/
-int t4_mc_read(struct adapter *adap, u32 addr, __be32 *data, u64 *ecc)
+int t4_mc_read(struct adapter *adap, int idx, u32 addr, __be32 *data, u64 *ecc)
{
int i;
+ u32 mc_bist_cmd, mc_bist_cmd_addr, mc_bist_cmd_len;
+ u32 mc_bist_status_rdata, mc_bist_data_pattern;
- if (t4_read_reg(adap, MC_BIST_CMD) & START_BIST)
+ if (is_t4(adap->chip)) {
+ mc_bist_cmd = MC_BIST_CMD;
+ mc_bist_cmd_addr = MC_BIST_CMD_ADDR;
+ mc_bist_cmd_len = MC_BIST_CMD_LEN;
+ mc_bist_status_rdata = MC_BIST_STATUS_RDATA;
+ mc_bist_data_pattern = MC_BIST_DATA_PATTERN;
+ } else {
+ mc_bist_cmd = MC_REG(MC_P_BIST_CMD, idx);
+ mc_bist_cmd_addr = MC_REG(MC_P_BIST_CMD_ADDR, idx);
+ mc_bist_cmd_len = MC_REG(MC_P_BIST_CMD_LEN, idx);
+ mc_bist_status_rdata = MC_REG(MC_P_BIST_STATUS_RDATA, idx);
+ mc_bist_data_pattern = MC_REG(MC_P_BIST_DATA_PATTERN, idx);
+ }
+
+ if (t4_read_reg(adap, mc_bist_cmd) & START_BIST)
return -EBUSY;
- t4_write_reg(adap, MC_BIST_CMD_ADDR, addr & ~0x3fU);
- t4_write_reg(adap, MC_BIST_CMD_LEN, 64);
- t4_write_reg(adap, MC_BIST_DATA_PATTERN, 0xc);
- t4_write_reg(adap, MC_BIST_CMD, BIST_OPCODE(1) | START_BIST |
+ t4_write_reg(adap, mc_bist_cmd_addr, addr & ~0x3fU);
+ t4_write_reg(adap, mc_bist_cmd_len, 64);
+ t4_write_reg(adap, mc_bist_data_pattern, 0xc);
+ t4_write_reg(adap, mc_bist_cmd, BIST_OPCODE(1) | START_BIST |
BIST_CMD_GAP(1));
- i = t4_wait_op_done(adap, MC_BIST_CMD, START_BIST, 0, 10, 1);
+ i = t4_wait_op_done(adap, mc_bist_cmd, START_BIST, 0, 10, 1);
if (i)
return i;
-#define MC_DATA(i) MC_BIST_STATUS_REG(MC_BIST_STATUS_RDATA, i)
+#define MC_DATA(i) MC_BIST_STATUS_REG(mc_bist_status_rdata, i)
for (i = 15; i >= 0; i--)
*data++ = htonl(t4_read_reg(adap, MC_DATA(i)));
@@ -329,20 +346,39 @@
int t4_edc_read(struct adapter *adap, int idx, u32 addr, __be32 *data, u64 *ecc)
{
int i;
+ u32 edc_bist_cmd, edc_bist_cmd_addr, edc_bist_cmd_len;
+ u32 edc_bist_cmd_data_pattern, edc_bist_status_rdata;
- idx *= EDC_STRIDE;
- if (t4_read_reg(adap, EDC_BIST_CMD + idx) & START_BIST)
+ if (is_t4(adap->chip)) {
+ edc_bist_cmd = EDC_REG(EDC_BIST_CMD, idx);
+ edc_bist_cmd_addr = EDC_REG(EDC_BIST_CMD_ADDR, idx);
+ edc_bist_cmd_len = EDC_REG(EDC_BIST_CMD_LEN, idx);
+ edc_bist_cmd_data_pattern = EDC_REG(EDC_BIST_DATA_PATTERN,
+ idx);
+ edc_bist_status_rdata = EDC_REG(EDC_BIST_STATUS_RDATA,
+ idx);
+ } else {
+ edc_bist_cmd = EDC_REG_T5(EDC_H_BIST_CMD, idx);
+ edc_bist_cmd_addr = EDC_REG_T5(EDC_H_BIST_CMD_ADDR, idx);
+ edc_bist_cmd_len = EDC_REG_T5(EDC_H_BIST_CMD_LEN, idx);
+ edc_bist_cmd_data_pattern =
+ EDC_REG_T5(EDC_H_BIST_DATA_PATTERN, idx);
+ edc_bist_status_rdata =
+ EDC_REG_T5(EDC_H_BIST_STATUS_RDATA, idx);
+ }
+
+ if (t4_read_reg(adap, edc_bist_cmd) & START_BIST)
return -EBUSY;
- t4_write_reg(adap, EDC_BIST_CMD_ADDR + idx, addr & ~0x3fU);
- t4_write_reg(adap, EDC_BIST_CMD_LEN + idx, 64);
- t4_write_reg(adap, EDC_BIST_DATA_PATTERN + idx, 0xc);
- t4_write_reg(adap, EDC_BIST_CMD + idx,
+ t4_write_reg(adap, edc_bist_cmd_addr, addr & ~0x3fU);
+ t4_write_reg(adap, edc_bist_cmd_len, 64);
+ t4_write_reg(adap, edc_bist_cmd_data_pattern, 0xc);
+ t4_write_reg(adap, edc_bist_cmd,
BIST_OPCODE(1) | BIST_CMD_GAP(1) | START_BIST);
- i = t4_wait_op_done(adap, EDC_BIST_CMD + idx, START_BIST, 0, 10, 1);
+ i = t4_wait_op_done(adap, edc_bist_cmd, START_BIST, 0, 10, 1);
if (i)
return i;
-#define EDC_DATA(i) (EDC_BIST_STATUS_REG(EDC_BIST_STATUS_RDATA, i) + idx)
+#define EDC_DATA(i) (EDC_BIST_STATUS_REG(edc_bist_status_rdata, i))
for (i = 15; i >= 0; i--)
*data++ = htonl(t4_read_reg(adap, EDC_DATA(i)));
@@ -366,6 +402,7 @@
static int t4_mem_win_rw(struct adapter *adap, u32 addr, __be32 *data, int dir)
{
int i;
+ u32 win_pf = is_t4(adap->chip) ? 0 : V_PFNUM(adap->fn);
/*
* Setup offset into PCIE memory window. Address must be a
@@ -374,7 +411,7 @@
* values.)
*/
t4_write_reg(adap, PCIE_MEM_ACCESS_OFFSET,
- addr & ~(MEMWIN0_APERTURE - 1));
+ (addr & ~(MEMWIN0_APERTURE - 1)) | win_pf);
t4_read_reg(adap, PCIE_MEM_ACCESS_OFFSET);
/* Collecting data 4 bytes at a time upto MEMWIN0_APERTURE */
@@ -410,6 +447,7 @@
__be32 *buf, int dir)
{
u32 pos, start, end, offset, memoffset;
+ u32 edc_size, mc_size;
int ret = 0;
__be32 *data;
@@ -423,13 +461,21 @@
if (!data)
return -ENOMEM;
- /*
- * Offset into the region of memory which is being accessed
+ /* Offset into the region of memory which is being accessed
* MEM_EDC0 = 0
* MEM_EDC1 = 1
- * MEM_MC = 2
+ * MEM_MC = 2 -- T4
+ * MEM_MC0 = 2 -- For T5
+ * MEM_MC1 = 3 -- For T5
*/
- memoffset = (mtype * (5 * 1024 * 1024));
+ edc_size = EDRAM_SIZE_GET(t4_read_reg(adap, MA_EDRAM0_BAR));
+ if (mtype != MEM_MC1)
+ memoffset = (mtype * (edc_size * 1024 * 1024));
+ else {
+ mc_size = EXT_MEM_SIZE_GET(t4_read_reg(adap,
+ MA_EXT_MEMORY_BAR));
+ memoffset = (MEM_MC0 * edc_size + mc_size) * 1024 * 1024;
+ }
/* Determine the PCIE_MEM_ACCESS_OFFSET */
addr = addr + memoffset;
@@ -497,9 +543,9 @@
}
#define EEPROM_STAT_ADDR 0x7bfc
-#define VPD_LEN 512
#define VPD_BASE 0x400
#define VPD_BASE_OLD 0
+#define VPD_LEN 1024
/**
* t4_seeprom_wp - enable/disable EEPROM write protection
@@ -856,6 +902,7 @@
{
u32 api_vers[2];
int ret, major, minor, micro;
+ int exp_major, exp_minor, exp_micro;
ret = get_fw_version(adapter, &adapter->params.fw_vers);
if (!ret)
@@ -870,17 +917,35 @@
major = FW_HDR_FW_VER_MAJOR_GET(adapter->params.fw_vers);
minor = FW_HDR_FW_VER_MINOR_GET(adapter->params.fw_vers);
micro = FW_HDR_FW_VER_MICRO_GET(adapter->params.fw_vers);
- memcpy(adapter->params.api_vers, api_vers,
- sizeof(adapter->params.api_vers));
- if (major != FW_VERSION_MAJOR) { /* major mismatch - fail */
- dev_err(adapter->pdev_dev,
- "card FW has major version %u, driver wants %u\n",
- major, FW_VERSION_MAJOR);
+ switch (CHELSIO_CHIP_VERSION(adapter->chip)) {
+ case CHELSIO_T4:
+ exp_major = FW_VERSION_MAJOR;
+ exp_minor = FW_VERSION_MINOR;
+ exp_micro = FW_VERSION_MICRO;
+ break;
+ case CHELSIO_T5:
+ exp_major = FW_VERSION_MAJOR_T5;
+ exp_minor = FW_VERSION_MINOR_T5;
+ exp_micro = FW_VERSION_MICRO_T5;
+ break;
+ default:
+ dev_err(adapter->pdev_dev, "Unsupported chip type, %x\n",
+ adapter->chip);
return -EINVAL;
}
- if (minor == FW_VERSION_MINOR && micro == FW_VERSION_MICRO)
+ memcpy(adapter->params.api_vers, api_vers,
+ sizeof(adapter->params.api_vers));
+
+ if (major != exp_major) { /* major mismatch - fail */
+ dev_err(adapter->pdev_dev,
+ "card FW has major version %u, driver wants %u\n",
+ major, exp_major);
+ return -EINVAL;
+ }
+
+ if (minor == exp_minor && micro == exp_micro)
return 0; /* perfect match */
/* Minor/micro version mismatch. Report it but often it's OK. */
@@ -1246,6 +1311,45 @@
{ 0 }
};
+ static struct intr_info t5_pcie_intr_info[] = {
+ { MSTGRPPERR, "Master Response Read Queue parity error",
+ -1, 1 },
+ { MSTTIMEOUTPERR, "Master Timeout FIFO parity error", -1, 1 },
+ { MSIXSTIPERR, "MSI-X STI SRAM parity error", -1, 1 },
+ { MSIXADDRLPERR, "MSI-X AddrL parity error", -1, 1 },
+ { MSIXADDRHPERR, "MSI-X AddrH parity error", -1, 1 },
+ { MSIXDATAPERR, "MSI-X data parity error", -1, 1 },
+ { MSIXDIPERR, "MSI-X DI parity error", -1, 1 },
+ { PIOCPLGRPPERR, "PCI PIO completion Group FIFO parity error",
+ -1, 1 },
+ { PIOREQGRPPERR, "PCI PIO request Group FIFO parity error",
+ -1, 1 },
+ { TARTAGPERR, "PCI PCI target tag FIFO parity error", -1, 1 },
+ { MSTTAGQPERR, "PCI master tag queue parity error", -1, 1 },
+ { CREQPERR, "PCI CMD channel request parity error", -1, 1 },
+ { CRSPPERR, "PCI CMD channel response parity error", -1, 1 },
+ { DREQWRPERR, "PCI DMA channel write request parity error",
+ -1, 1 },
+ { DREQPERR, "PCI DMA channel request parity error", -1, 1 },
+ { DRSPPERR, "PCI DMA channel response parity error", -1, 1 },
+ { HREQWRPERR, "PCI HMA channel count parity error", -1, 1 },
+ { HREQPERR, "PCI HMA channel request parity error", -1, 1 },
+ { HRSPPERR, "PCI HMA channel response parity error", -1, 1 },
+ { CFGSNPPERR, "PCI config snoop FIFO parity error", -1, 1 },
+ { FIDPERR, "PCI FID parity error", -1, 1 },
+ { VFIDPERR, "PCI INTx clear parity error", -1, 1 },
+ { MAGRPPERR, "PCI MA group FIFO parity error", -1, 1 },
+ { PIOTAGPERR, "PCI PIO tag parity error", -1, 1 },
+ { IPRXHDRGRPPERR, "PCI IP Rx header group parity error",
+ -1, 1 },
+ { IPRXDATAGRPPERR, "PCI IP Rx data group parity error", -1, 1 },
+ { RPLPERR, "PCI IP replay buffer parity error", -1, 1 },
+ { IPSOTPERR, "PCI IP SOT buffer parity error", -1, 1 },
+ { TRGT1GRPPERR, "PCI TRGT1 group FIFOs parity error", -1, 1 },
+ { READRSPERR, "Outbound read error", -1, 0 },
+ { 0 }
+ };
+
int fat;
fat = t4_handle_intr_status(adapter,
@@ -1254,7 +1358,10 @@
t4_handle_intr_status(adapter,
PCIE_CORE_UTL_PCI_EXPRESS_PORT_STATUS,
pcie_port_intr_info) +
- t4_handle_intr_status(adapter, PCIE_INT_CAUSE, pcie_intr_info);
+ t4_handle_intr_status(adapter, PCIE_INT_CAUSE,
+ is_t4(adapter->chip) ?
+ pcie_intr_info : t5_pcie_intr_info);
+
if (fat)
t4_fatal_err(adapter);
}
@@ -1664,7 +1771,14 @@
*/
static void xgmac_intr_handler(struct adapter *adap, int port)
{
- u32 v = t4_read_reg(adap, PORT_REG(port, XGMAC_PORT_INT_CAUSE));
+ u32 v, int_cause_reg;
+
+ if (is_t4(adap->chip))
+ int_cause_reg = PORT_REG(port, XGMAC_PORT_INT_CAUSE);
+ else
+ int_cause_reg = T5_PORT_REG(port, MAC_PORT_INT_CAUSE);
+
+ v = t4_read_reg(adap, int_cause_reg);
v &= TXFIFO_PRTY_ERR | RXFIFO_PRTY_ERR;
if (!v)
@@ -2126,7 +2240,9 @@
u32 bgmap = get_mps_bg_map(adap, idx);
#define GET_STAT(name) \
- t4_read_reg64(adap, PORT_REG(idx, MPS_PORT_STAT_##name##_L))
+ t4_read_reg64(adap, \
+ (is_t4(adap->chip) ? PORT_REG(idx, MPS_PORT_STAT_##name##_L) : \
+ T5_PORT_REG(idx, MPS_PORT_STAT_##name##_L)))
#define GET_STAT_COM(name) t4_read_reg64(adap, MPS_STAT_##name##_L)
p->tx_octets = GET_STAT(TX_PORT_BYTES);
@@ -2205,14 +2321,26 @@
void t4_wol_magic_enable(struct adapter *adap, unsigned int port,
const u8 *addr)
{
+ u32 mag_id_reg_l, mag_id_reg_h, port_cfg_reg;
+
+ if (is_t4(adap->chip)) {
+ mag_id_reg_l = PORT_REG(port, XGMAC_PORT_MAGIC_MACID_LO);
+ mag_id_reg_h = PORT_REG(port, XGMAC_PORT_MAGIC_MACID_HI);
+ port_cfg_reg = PORT_REG(port, XGMAC_PORT_CFG2);
+ } else {
+ mag_id_reg_l = T5_PORT_REG(port, MAC_PORT_MAGIC_MACID_LO);
+ mag_id_reg_h = T5_PORT_REG(port, MAC_PORT_MAGIC_MACID_HI);
+ port_cfg_reg = T5_PORT_REG(port, MAC_PORT_CFG2);
+ }
+
if (addr) {
- t4_write_reg(adap, PORT_REG(port, XGMAC_PORT_MAGIC_MACID_LO),
+ t4_write_reg(adap, mag_id_reg_l,
(addr[2] << 24) | (addr[3] << 16) |
(addr[4] << 8) | addr[5]);
- t4_write_reg(adap, PORT_REG(port, XGMAC_PORT_MAGIC_MACID_HI),
+ t4_write_reg(adap, mag_id_reg_h,
(addr[0] << 8) | addr[1]);
}
- t4_set_reg_field(adap, PORT_REG(port, XGMAC_PORT_CFG2), MAGICEN,
+ t4_set_reg_field(adap, port_cfg_reg, MAGICEN,
addr ? MAGICEN : 0);
}
@@ -2235,16 +2363,23 @@
u64 mask0, u64 mask1, unsigned int crc, bool enable)
{
int i;
+ u32 port_cfg_reg;
+
+ if (is_t4(adap->chip))
+ port_cfg_reg = PORT_REG(port, XGMAC_PORT_CFG2);
+ else
+ port_cfg_reg = T5_PORT_REG(port, MAC_PORT_CFG2);
if (!enable) {
- t4_set_reg_field(adap, PORT_REG(port, XGMAC_PORT_CFG2),
- PATEN, 0);
+ t4_set_reg_field(adap, port_cfg_reg, PATEN, 0);
return 0;
}
if (map > 0xff)
return -EINVAL;
-#define EPIO_REG(name) PORT_REG(port, XGMAC_PORT_EPIO_##name)
+#define EPIO_REG(name) \
+ (is_t4(adap->chip) ? PORT_REG(port, XGMAC_PORT_EPIO_##name) : \
+ T5_PORT_REG(port, MAC_PORT_EPIO_##name))
t4_write_reg(adap, EPIO_REG(DATA1), mask0 >> 32);
t4_write_reg(adap, EPIO_REG(DATA2), mask1);
@@ -2322,24 +2457,24 @@
* @addr: address of first byte requested aligned on 32b.
* @data: len bytes to hold the data read
* @len: amount of data to read from window. Must be <=
- * MEMWIN0_APERATURE after adjusting for 16B alignment
- * requirements of the the memory window.
+ * MEMWIN0_APERATURE after adjusting for 16B for T4 and
+ * 128B for T5 alignment requirements of the the memory window.
*
* Read len bytes of data from MC starting at @addr.
*/
int t4_mem_win_read_len(struct adapter *adap, u32 addr, __be32 *data, int len)
{
- int i;
- int off;
+ int i, off;
+ u32 win_pf = is_t4(adap->chip) ? 0 : V_PFNUM(adap->fn);
- /*
- * Align on a 16B boundary.
+ /* Align on a 2KB boundary.
*/
- off = addr & 15;
+ off = addr & MEMWIN0_APERTURE;
if ((addr & 3) || (len + off) > MEMWIN0_APERTURE)
return -EINVAL;
- t4_write_reg(adap, PCIE_MEM_ACCESS_OFFSET, addr & ~15);
+ t4_write_reg(adap, PCIE_MEM_ACCESS_OFFSET,
+ (addr & ~MEMWIN0_APERTURE) | win_pf);
t4_read_reg(adap, PCIE_MEM_ACCESS_OFFSET);
for (i = 0; i < len; i += 4)
@@ -3162,6 +3297,9 @@
int i, ret;
struct fw_vi_mac_cmd c;
struct fw_vi_mac_exact *p;
+ unsigned int max_naddr = is_t4(adap->chip) ?
+ NUM_MPS_CLS_SRAM_L_INSTANCES :
+ NUM_MPS_T5_CLS_SRAM_L_INSTANCES;
if (naddr > 7)
return -EINVAL;
@@ -3187,8 +3325,8 @@
u16 index = FW_VI_MAC_CMD_IDX_GET(ntohs(p->valid_to_idx));
if (idx)
- idx[i] = index >= NEXACT_MAC ? 0xffff : index;
- if (index < NEXACT_MAC)
+ idx[i] = index >= max_naddr ? 0xffff : index;
+ if (index < max_naddr)
ret++;
else if (hash)
*hash |= (1ULL << hash_mac_addr(addr[i]));
@@ -3221,6 +3359,9 @@
int ret, mode;
struct fw_vi_mac_cmd c;
struct fw_vi_mac_exact *p = c.u.exact;
+ unsigned int max_mac_addr = is_t4(adap->chip) ?
+ NUM_MPS_CLS_SRAM_L_INSTANCES :
+ NUM_MPS_T5_CLS_SRAM_L_INSTANCES;
if (idx < 0) /* new allocation */
idx = persist ? FW_VI_MAC_ADD_PERSIST_MAC : FW_VI_MAC_ADD_MAC;
@@ -3238,7 +3379,7 @@
ret = t4_wr_mbox(adap, mbox, &c, sizeof(c), &c);
if (ret == 0) {
ret = FW_VI_MAC_CMD_IDX_GET(ntohs(p->valid_to_idx));
- if (ret >= NEXACT_MAC)
+ if (ret >= max_mac_addr)
ret = -ENOMEM;
}
return ret;
@@ -3547,7 +3688,8 @@
*/
int t4_prep_adapter(struct adapter *adapter)
{
- int ret;
+ int ret, ver;
+ uint16_t device_id;
ret = t4_wait_dev_ready(adapter);
if (ret < 0)
@@ -3562,6 +3704,28 @@
return ret;
}
+ /* Retrieve adapter's device ID
+ */
+ pci_read_config_word(adapter->pdev, PCI_DEVICE_ID, &device_id);
+ ver = device_id >> 12;
+ switch (ver) {
+ case CHELSIO_T4:
+ adapter->chip = CHELSIO_CHIP_CODE(CHELSIO_T4,
+ adapter->params.rev);
+ break;
+ case CHELSIO_T5:
+ adapter->chip = CHELSIO_CHIP_CODE(CHELSIO_T5,
+ adapter->params.rev);
+ break;
+ default:
+ dev_err(adapter->pdev_dev, "Device %d is not supported\n",
+ device_id);
+ return -EINVAL;
+ }
+
+ /* Reassign the updated revision field */
+ adapter->params.rev = adapter->chip;
+
init_cong_ctrl(adapter->params.a_wnd, adapter->params.b_wnd);
/*
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.h b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.h
index f534ed7..1d1623b 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.h
@@ -47,7 +47,6 @@
TCB_SIZE = 128, /* TCB size */
NMTUS = 16, /* size of MTU table */
NCCTRL_WIN = 32, /* # of congestion control windows */
- NEXACT_MAC = 336, /* # of exact MAC address filters */
L2T_SIZE = 4096, /* # of L2T entries */
MBOX_LEN = 64, /* mailbox size in bytes */
TRACE_LEN = 112, /* length of trace data and mask */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h b/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h
index 261d177..47656ac 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h
@@ -74,6 +74,7 @@
CPL_PASS_ESTABLISH = 0x41,
CPL_RX_DATA_DDP = 0x42,
CPL_PASS_ACCEPT_REQ = 0x44,
+ CPL_TRACE_PKT_T5 = 0x48,
CPL_RDMA_READ_REQ = 0x60,
@@ -287,6 +288,23 @@
__be32 opt2;
};
+#define S_FILTER_TUPLE 24
+#define M_FILTER_TUPLE 0xFFFFFFFFFF
+#define V_FILTER_TUPLE(x) ((x) << S_FILTER_TUPLE)
+#define G_FILTER_TUPLE(x) (((x) >> S_FILTER_TUPLE) & M_FILTER_TUPLE)
+struct cpl_t5_act_open_req {
+ WR_HDR;
+ union opcode_tid ot;
+ __be16 local_port;
+ __be16 peer_port;
+ __be32 local_ip;
+ __be32 peer_ip;
+ __be64 opt0;
+ __be32 rsvd;
+ __be32 opt2;
+ __be64 params;
+};
+
struct cpl_act_open_req6 {
WR_HDR;
union opcode_tid ot;
@@ -566,6 +584,11 @@
#define V_RX_ETHHDR_LEN(x) ((x) << S_RX_ETHHDR_LEN)
#define G_RX_ETHHDR_LEN(x) (((x) >> S_RX_ETHHDR_LEN) & M_RX_ETHHDR_LEN)
+#define S_RX_T5_ETHHDR_LEN 0
+#define M_RX_T5_ETHHDR_LEN 0x3F
+#define V_RX_T5_ETHHDR_LEN(x) ((x) << S_RX_T5_ETHHDR_LEN)
+#define G_RX_T5_ETHHDR_LEN(x) (((x) >> S_RX_T5_ETHHDR_LEN) & M_RX_T5_ETHHDR_LEN)
+
#define S_RX_MACIDX 8
#define M_RX_MACIDX 0x1FF
#define V_RX_MACIDX(x) ((x) << S_RX_MACIDX)
@@ -612,6 +635,28 @@
__be64 tstamp;
};
+struct cpl_t5_trace_pkt {
+ __u8 opcode;
+ __u8 intf;
+#if defined(__LITTLE_ENDIAN_BITFIELD)
+ __u8 runt:4;
+ __u8 filter_hit:4;
+ __u8:6;
+ __u8 err:1;
+ __u8 trunc:1;
+#else
+ __u8 filter_hit:4;
+ __u8 runt:4;
+ __u8 trunc:1;
+ __u8 err:1;
+ __u8:6;
+#endif
+ __be16 rsvd;
+ __be16 len;
+ __be64 tstamp;
+ __be64 rsvd1;
+};
+
struct cpl_l2t_write_req {
WR_HDR;
union opcode_tid ot;
@@ -742,4 +787,12 @@
#define ULP_MEMIO_LOCK(x) ((x) << 31)
};
+#define S_T5_ULP_MEMIO_IMM 23
+#define V_T5_ULP_MEMIO_IMM(x) ((x) << S_T5_ULP_MEMIO_IMM)
+#define F_T5_ULP_MEMIO_IMM V_T5_ULP_MEMIO_IMM(1U)
+
+#define S_T5_ULP_MEMIO_ORDER 22
+#define V_T5_ULP_MEMIO_ORDER(x) ((x) << S_T5_ULP_MEMIO_ORDER)
+#define F_T5_ULP_MEMIO_ORDER V_T5_ULP_MEMIO_ORDER(1U)
+
#endif /* __T4_MSG_H */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h b/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h
index 83ec5f7..ef146c0 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h
@@ -68,9 +68,14 @@
#define QID_SHIFT 15
#define QID(x) ((x) << QID_SHIFT)
#define DBPRIO(x) ((x) << 14)
+#define DBTYPE(x) ((x) << 13)
#define PIDX_MASK 0x00003fffU
#define PIDX_SHIFT 0
#define PIDX(x) ((x) << PIDX_SHIFT)
+#define S_PIDX_T5 0
+#define M_PIDX_T5 0x1fffU
+#define PIDX_T5(x) (((x) >> S_PIDX_T5) & M_PIDX_T5)
+
#define SGE_PF_GTS 0x4
#define INGRESSQID_MASK 0xffff0000U
@@ -152,6 +157,8 @@
#define QUEUESPERPAGEPF0_MASK 0x0000000fU
#define QUEUESPERPAGEPF0_GET(x) ((x) & QUEUESPERPAGEPF0_MASK)
+#define QUEUESPERPAGEPF1 4
+
#define SGE_INT_CAUSE1 0x1024
#define SGE_INT_CAUSE2 0x1030
#define SGE_INT_CAUSE3 0x103c
@@ -234,6 +241,10 @@
#define SGE_DOORBELL_CONTROL 0x10a8
#define ENABLE_DROP (1 << 13)
+#define S_NOCOALESCE 26
+#define V_NOCOALESCE(x) ((x) << S_NOCOALESCE)
+#define F_NOCOALESCE V_NOCOALESCE(1U)
+
#define SGE_TIMER_VALUE_0_AND_1 0x10b8
#define TIMERVALUE0_MASK 0xffff0000U
#define TIMERVALUE0_SHIFT 16
@@ -272,17 +283,36 @@
#define S_HP_INT_THRESH 28
#define M_HP_INT_THRESH 0xfU
#define V_HP_INT_THRESH(x) ((x) << S_HP_INT_THRESH)
+#define S_LP_INT_THRESH_T5 18
+#define V_LP_INT_THRESH_T5(x) ((x) << S_LP_INT_THRESH_T5)
+#define M_LP_COUNT_T5 0x3ffffU
+#define G_LP_COUNT_T5(x) (((x) >> S_LP_COUNT) & M_LP_COUNT_T5)
#define M_HP_COUNT 0x7ffU
#define S_HP_COUNT 16
#define G_HP_COUNT(x) (((x) >> S_HP_COUNT) & M_HP_COUNT)
#define S_LP_INT_THRESH 12
#define M_LP_INT_THRESH 0xfU
+#define M_LP_INT_THRESH_T5 0xfffU
#define V_LP_INT_THRESH(x) ((x) << S_LP_INT_THRESH)
#define M_LP_COUNT 0x7ffU
#define S_LP_COUNT 0
#define G_LP_COUNT(x) (((x) >> S_LP_COUNT) & M_LP_COUNT)
#define A_SGE_DBFIFO_STATUS 0x10a4
+#define SGE_STAT_TOTAL 0x10e4
+#define SGE_STAT_MATCH 0x10e8
+
+#define SGE_STAT_CFG 0x10ec
+#define S_STATSOURCE_T5 9
+#define STATSOURCE_T5(x) ((x) << S_STATSOURCE_T5)
+
+#define SGE_DBFIFO_STATUS2 0x1118
+#define M_HP_COUNT_T5 0x3ffU
+#define G_HP_COUNT_T5(x) ((x) & M_HP_COUNT_T5)
+#define S_HP_INT_THRESH_T5 10
+#define M_HP_INT_THRESH_T5 0xfU
+#define V_HP_INT_THRESH_T5(x) ((x) << S_HP_INT_THRESH_T5)
+
#define S_ENABLE_DROP 13
#define V_ENABLE_DROP(x) ((x) << S_ENABLE_DROP)
#define F_ENABLE_DROP V_ENABLE_DROP(1U)
@@ -331,8 +361,27 @@
#define MSIADDRHPERR 0x00000002U
#define MSIADDRLPERR 0x00000001U
+#define READRSPERR 0x20000000U
+#define TRGT1GRPPERR 0x10000000U
+#define IPSOTPERR 0x08000000U
+#define IPRXDATAGRPPERR 0x02000000U
+#define IPRXHDRGRPPERR 0x01000000U
+#define MAGRPPERR 0x00400000U
+#define VFIDPERR 0x00200000U
+#define HREQWRPERR 0x00010000U
+#define DREQWRPERR 0x00002000U
+#define MSTTAGQPERR 0x00000400U
+#define PIOREQGRPPERR 0x00000100U
+#define PIOCPLGRPPERR 0x00000080U
+#define MSIXSTIPERR 0x00000004U
+#define MSTTIMEOUTPERR 0x00000002U
+#define MSTGRPPERR 0x00000001U
+
#define PCIE_NONFAT_ERR 0x3010
#define PCIE_MEM_ACCESS_BASE_WIN 0x3068
+#define S_PCIEOFST 10
+#define M_PCIEOFST 0x3fffffU
+#define GET_PCIEOFST(x) (((x) >> S_PCIEOFST) & M_PCIEOFST)
#define PCIEOFST_MASK 0xfffffc00U
#define BIR_MASK 0x00000300U
#define BIR_SHIFT 8
@@ -342,6 +391,9 @@
#define WINDOW(x) ((x) << WINDOW_SHIFT)
#define PCIE_MEM_ACCESS_OFFSET 0x306c
+#define S_PFNUM 0
+#define V_PFNUM(x) ((x) << S_PFNUM)
+
#define PCIE_FW 0x30b8
#define PCIE_FW_ERR 0x80000000U
#define PCIE_FW_INIT 0x40000000U
@@ -407,12 +459,18 @@
#define MC_BIST_STATUS_RDATA 0x7688
+#define MA_EDRAM0_BAR 0x77c0
+#define MA_EDRAM1_BAR 0x77c4
+#define EDRAM_SIZE_MASK 0xfffU
+#define EDRAM_SIZE_GET(x) ((x) & EDRAM_SIZE_MASK)
+
#define MA_EXT_MEMORY_BAR 0x77c8
#define EXT_MEM_SIZE_MASK 0x00000fffU
#define EXT_MEM_SIZE_SHIFT 0
#define EXT_MEM_SIZE_GET(x) (((x) & EXT_MEM_SIZE_MASK) >> EXT_MEM_SIZE_SHIFT)
#define MA_TARGET_MEM_ENABLE 0x77d8
+#define EXT_MEM1_ENABLE 0x00000010U
#define EXT_MEM_ENABLE 0x00000004U
#define EDRAM1_ENABLE 0x00000002U
#define EDRAM0_ENABLE 0x00000001U
@@ -431,6 +489,7 @@
#define MA_PCIE_FW 0x30b8
#define MA_PARITY_ERROR_STATUS 0x77f4
+#define MA_EXT_MEMORY1_BAR 0x7808
#define EDC_0_BASE_ADDR 0x7900
#define EDC_BIST_CMD 0x7904
@@ -801,6 +860,15 @@
#define MPS_PORT_STAT_RX_PORT_PPP7_H 0x60c
#define MPS_PORT_STAT_RX_PORT_LESS_64B_L 0x610
#define MPS_PORT_STAT_RX_PORT_LESS_64B_H 0x614
+#define MAC_PORT_CFG2 0x818
+#define MAC_PORT_MAGIC_MACID_LO 0x824
+#define MAC_PORT_MAGIC_MACID_HI 0x828
+#define MAC_PORT_EPIO_DATA0 0x8c0
+#define MAC_PORT_EPIO_DATA1 0x8c4
+#define MAC_PORT_EPIO_DATA2 0x8c8
+#define MAC_PORT_EPIO_DATA3 0x8cc
+#define MAC_PORT_EPIO_OP 0x8d0
+
#define MPS_CMN_CTL 0x9000
#define NUMPORTS_MASK 0x00000003U
#define NUMPORTS_SHIFT 0
@@ -1063,6 +1131,7 @@
#define ADDRESS_SHIFT 0
#define ADDRESS(x) ((x) << ADDRESS_SHIFT)
+#define MAC_PORT_INT_CAUSE 0x8dc
#define XGMAC_PORT_INT_CAUSE 0x10dc
#define A_TP_TX_MOD_QUEUE_REQ_MAP 0x7e28
@@ -1101,4 +1170,33 @@
#define V_PORT(x) ((x) << S_PORT)
#define F_PORT V_PORT(1U)
+#define NUM_MPS_CLS_SRAM_L_INSTANCES 336
+#define NUM_MPS_T5_CLS_SRAM_L_INSTANCES 512
+
+#define T5_PORT0_BASE 0x30000
+#define T5_PORT_STRIDE 0x4000
+#define T5_PORT_BASE(idx) (T5_PORT0_BASE + (idx) * T5_PORT_STRIDE)
+#define T5_PORT_REG(idx, reg) (T5_PORT_BASE(idx) + (reg))
+
+#define MC_0_BASE_ADDR 0x40000
+#define MC_1_BASE_ADDR 0x48000
+#define MC_STRIDE (MC_1_BASE_ADDR - MC_0_BASE_ADDR)
+#define MC_REG(reg, idx) (reg + MC_STRIDE * idx)
+
+#define MC_P_BIST_CMD 0x41400
+#define MC_P_BIST_CMD_ADDR 0x41404
+#define MC_P_BIST_CMD_LEN 0x41408
+#define MC_P_BIST_DATA_PATTERN 0x4140c
+#define MC_P_BIST_STATUS_RDATA 0x41488
+#define EDC_T50_BASE_ADDR 0x50000
+#define EDC_H_BIST_CMD 0x50004
+#define EDC_H_BIST_CMD_ADDR 0x50008
+#define EDC_H_BIST_CMD_LEN 0x5000c
+#define EDC_H_BIST_DATA_PATTERN 0x50010
+#define EDC_H_BIST_STATUS_RDATA 0x50028
+
+#define EDC_T51_BASE_ADDR 0x50800
+#define EDC_STRIDE_T5 (EDC_T51_BASE_ADDR - EDC_T50_BASE_ADDR)
+#define EDC_REG_T5(reg, idx) (reg + EDC_STRIDE_T5 * idx)
+
#endif /* __T4_REGS_H */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h
index a0dcccd..9344432 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h
@@ -574,7 +574,7 @@
__be16 vlantci;
};
-#define FW_CMD_MAX_TIMEOUT 3000
+#define FW_CMD_MAX_TIMEOUT 10000
/*
* If a host driver does a HELLO and discovers that there's already a MASTER
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/adapter.h b/drivers/net/ethernet/chelsio/cxgb4vf/adapter.h
index 68eaa9c..be5c7ef 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/adapter.h
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/adapter.h
@@ -344,6 +344,7 @@
unsigned long registered_device_map;
unsigned long open_device_map;
unsigned long flags;
+ enum chip_type chip;
struct adapter_params params;
/* queue and interrupt resources */
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
index 56b46ab..7fcac20 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
@@ -54,8 +54,8 @@
/*
* Generic information about the driver.
*/
-#define DRV_VERSION "1.0.0"
-#define DRV_DESC "Chelsio T4 Virtual Function (VF) Network Driver"
+#define DRV_VERSION "2.0.0-ko"
+#define DRV_DESC "Chelsio T4/T5 Virtual Function (VF) Network Driver"
/*
* Module Parameters.
@@ -1050,7 +1050,7 @@
/*
* Chip version 4, revision 0x3f (cxgb4vf).
*/
- return 4 | (0x3f << 10);
+ return CHELSIO_CHIP_VERSION(adapter->chip) | (0x3f << 10);
}
/*
@@ -2099,6 +2099,15 @@
return err;
}
+ switch (adapter->pdev->device >> 12) {
+ case CHELSIO_T4:
+ adapter->chip = CHELSIO_CHIP_CODE(CHELSIO_T4, 0);
+ break;
+ case CHELSIO_T5:
+ adapter->chip = CHELSIO_CHIP_CODE(CHELSIO_T5, 0);
+ break;
+ }
+
/*
* Grab basic operational parameters. These will predominantly have
* been set up by the Physical Function Driver or will be hard coded
@@ -2888,6 +2897,26 @@
CH_DEVICE(0x480a, 0), /* T404-bt */
CH_DEVICE(0x480d, 0), /* T480-cr */
CH_DEVICE(0x480e, 0), /* T440-lp-cr */
+ CH_DEVICE(0x5800, 0), /* T580-dbg */
+ CH_DEVICE(0x5801, 0), /* T520-cr */
+ CH_DEVICE(0x5802, 0), /* T522-cr */
+ CH_DEVICE(0x5803, 0), /* T540-cr */
+ CH_DEVICE(0x5804, 0), /* T520-bch */
+ CH_DEVICE(0x5805, 0), /* T540-bch */
+ CH_DEVICE(0x5806, 0), /* T540-ch */
+ CH_DEVICE(0x5807, 0), /* T520-so */
+ CH_DEVICE(0x5808, 0), /* T520-cx */
+ CH_DEVICE(0x5809, 0), /* T520-bt */
+ CH_DEVICE(0x580a, 0), /* T504-bt */
+ CH_DEVICE(0x580b, 0), /* T520-sr */
+ CH_DEVICE(0x580c, 0), /* T504-bt */
+ CH_DEVICE(0x580d, 0), /* T580-cr */
+ CH_DEVICE(0x580e, 0), /* T540-lp-cr */
+ CH_DEVICE(0x580f, 0), /* Amsterdam */
+ CH_DEVICE(0x5810, 0), /* T580-lp-cr */
+ CH_DEVICE(0x5811, 0), /* T520-lp-cr */
+ CH_DEVICE(0x5812, 0), /* T560-cr */
+ CH_DEVICE(0x5813, 0), /* T580-cr */
{ 0, }
};
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/sge.c b/drivers/net/ethernet/chelsio/cxgb4vf/sge.c
index 9488032..61dfb2a 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/sge.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/sge.c
@@ -528,17 +528,21 @@
*/
static inline void ring_fl_db(struct adapter *adapter, struct sge_fl *fl)
{
+ u32 val;
+
/*
* The SGE keeps track of its Producer and Consumer Indices in terms
* of Egress Queue Units so we can only tell it about integral numbers
* of multiples of Free List Entries per Egress Queue Units ...
*/
if (fl->pend_cred >= FL_PER_EQ_UNIT) {
+ val = PIDX(fl->pend_cred / FL_PER_EQ_UNIT);
+ if (!is_t4(adapter->chip))
+ val |= DBTYPE(1);
wmb();
t4_write_reg(adapter, T4VF_SGE_BASE_ADDR + SGE_VF_KDOORBELL,
DBPRIO(1) |
- QID(fl->cntxt_id) |
- PIDX(fl->pend_cred / FL_PER_EQ_UNIT));
+ QID(fl->cntxt_id) | val);
fl->pend_cred %= FL_PER_EQ_UNIT;
}
}
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h
index 283f9d0..53cbfed 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h
@@ -38,6 +38,25 @@
#include "../cxgb4/t4fw_api.h"
+#define CHELSIO_CHIP_CODE(version, revision) (((version) << 4) | (revision))
+#define CHELSIO_CHIP_VERSION(code) ((code) >> 4)
+#define CHELSIO_CHIP_RELEASE(code) ((code) & 0xf)
+
+#define CHELSIO_T4 0x4
+#define CHELSIO_T5 0x5
+
+enum chip_type {
+ T4_A1 = CHELSIO_CHIP_CODE(CHELSIO_T4, 0),
+ T4_A2 = CHELSIO_CHIP_CODE(CHELSIO_T4, 1),
+ T4_A3 = CHELSIO_CHIP_CODE(CHELSIO_T4, 2),
+ T4_FIRST_REV = T4_A1,
+ T4_LAST_REV = T4_A3,
+
+ T5_A1 = CHELSIO_CHIP_CODE(CHELSIO_T5, 0),
+ T5_FIRST_REV = T5_A1,
+ T5_LAST_REV = T5_A1,
+};
+
/*
* The "len16" field of a Firmware Command Structure ...
*/
@@ -232,6 +251,11 @@
return t4vf_wr_mbox_core(adapter, cmd, size, rpl, false);
}
+static inline int is_t4(enum chip_type chip)
+{
+ return (chip >= T4_FIRST_REV && chip <= T4_LAST_REV);
+}
+
int t4vf_wait_dev_ready(struct adapter *);
int t4vf_port_init(struct adapter *, int);
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c
index 7127c7b..9f96dc3 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c
@@ -1027,8 +1027,11 @@
unsigned nfilters = 0;
unsigned int rem = naddr;
struct fw_vi_mac_cmd cmd, rpl;
+ unsigned int max_naddr = is_t4(adapter->chip) ?
+ NUM_MPS_CLS_SRAM_L_INSTANCES :
+ NUM_MPS_T5_CLS_SRAM_L_INSTANCES;
- if (naddr > FW_CLS_TCAM_NUM_ENTRIES)
+ if (naddr > max_naddr)
return -EINVAL;
for (offset = 0; offset < naddr; /**/) {
@@ -1069,10 +1072,10 @@
if (idx)
idx[offset+i] =
- (index >= FW_CLS_TCAM_NUM_ENTRIES
+ (index >= max_naddr
? 0xffff
: index);
- if (index < FW_CLS_TCAM_NUM_ENTRIES)
+ if (index < max_naddr)
nfilters++;
else if (hash)
*hash |= (1ULL << hash_mac_addr(addr[offset+i]));
@@ -1118,6 +1121,9 @@
struct fw_vi_mac_exact *p = &cmd.u.exact[0];
size_t len16 = DIV_ROUND_UP(offsetof(struct fw_vi_mac_cmd,
u.exact[1]), 16);
+ unsigned int max_naddr = is_t4(adapter->chip) ?
+ NUM_MPS_CLS_SRAM_L_INSTANCES :
+ NUM_MPS_T5_CLS_SRAM_L_INSTANCES;
/*
* If this is a new allocation, determine whether it should be
@@ -1140,7 +1146,7 @@
if (ret == 0) {
p = &rpl.u.exact[0];
ret = FW_VI_MAC_CMD_IDX_GET(be16_to_cpu(p->valid_to_idx));
- if (ret >= FW_CLS_TCAM_NUM_ENTRIES)
+ if (ret >= max_naddr)
ret = -ENOMEM;
}
return ret;
diff --git a/drivers/net/ethernet/cirrus/cs89x0.c b/drivers/net/ethernet/cirrus/cs89x0.c
index 1384469..aaa0499 100644
--- a/drivers/net/ethernet/cirrus/cs89x0.c
+++ b/drivers/net/ethernet/cirrus/cs89x0.c
@@ -478,9 +478,6 @@
/* Malloc up new buffer. */
skb = netdev_alloc_skb(dev, length + 2);
if (skb == NULL) {
- /* I don't think we want to do this to a stressed system */
- cs89_dbg(0, err, "%s: Memory squeeze, dropping packet\n",
- dev->name);
dev->stats.rx_dropped++;
/* AKPM: advance bp to the next frame */
@@ -731,9 +728,6 @@
/* Malloc up new buffer. */
skb = netdev_alloc_skb(dev, length + 2);
if (skb == NULL) {
-#if 0 /* Again, this seems a cruel thing to do */
- pr_warn("%s: Memory squeeze, dropping packet\n", dev->name);
-#endif
dev->stats.rx_dropped++;
return;
}
@@ -1978,18 +1972,6 @@
.remove = cs89x0_platform_remove,
};
-static int __init cs89x0_init(void)
-{
- return platform_driver_probe(&cs89x0_driver, cs89x0_platform_probe);
-}
-
-module_init(cs89x0_init);
-
-static void __exit cs89x0_cleanup(void)
-{
- platform_driver_unregister(&cs89x0_driver);
-}
-
-module_exit(cs89x0_cleanup);
+module_platform_driver_probe(cs89x0_driver, cs89x0_platform_probe);
#endif /* CONFIG_CS89x0_PLATFORM */
diff --git a/drivers/net/ethernet/cirrus/ep93xx_eth.c b/drivers/net/ethernet/cirrus/ep93xx_eth.c
index 354cbb7..67b0388 100644
--- a/drivers/net/ethernet/cirrus/ep93xx_eth.c
+++ b/drivers/net/ethernet/cirrus/ep93xx_eth.c
@@ -887,18 +887,7 @@
},
};
-static int __init ep93xx_eth_init_module(void)
-{
- printk(KERN_INFO DRV_MODULE_NAME " version " DRV_MODULE_VERSION " loading\n");
- return platform_driver_register(&ep93xx_eth_driver);
-}
+module_platform_driver(ep93xx_eth_driver);
-static void __exit ep93xx_eth_cleanup_module(void)
-{
- platform_driver_unregister(&ep93xx_eth_driver);
-}
-
-module_init(ep93xx_eth_init_module);
-module_exit(ep93xx_eth_cleanup_module);
MODULE_LICENSE("GPL");
MODULE_ALIAS("platform:ep93xx-eth");
diff --git a/drivers/net/ethernet/davicom/dm9000.c b/drivers/net/ethernet/davicom/dm9000.c
index 8cdf025..f38f677 100644
--- a/drivers/net/ethernet/davicom/dm9000.c
+++ b/drivers/net/ethernet/davicom/dm9000.c
@@ -1687,22 +1687,7 @@
.remove = dm9000_drv_remove,
};
-static int __init
-dm9000_init(void)
-{
- printk(KERN_INFO "%s Ethernet Driver, V%s\n", CARDNAME, DRV_VERSION);
-
- return platform_driver_register(&dm9000_driver);
-}
-
-static void __exit
-dm9000_cleanup(void)
-{
- platform_driver_unregister(&dm9000_driver);
-}
-
-module_init(dm9000_init);
-module_exit(dm9000_cleanup);
+module_platform_driver(dm9000_driver);
MODULE_AUTHOR("Sascha Hauer, Ben Dooks");
MODULE_DESCRIPTION("Davicom DM9000 network driver");
diff --git a/drivers/net/ethernet/dec/tulip/xircom_cb.c b/drivers/net/ethernet/dec/tulip/xircom_cb.c
index 88feced..cdbcd16 100644
--- a/drivers/net/ethernet/dec/tulip/xircom_cb.c
+++ b/drivers/net/ethernet/dec/tulip/xircom_cb.c
@@ -236,17 +236,14 @@
private->rx_buffer = dma_alloc_coherent(d, 8192,
&private->rx_dma_handle,
GFP_KERNEL);
- if (private->rx_buffer == NULL) {
- pr_err("%s: no memory for rx buffer\n", __func__);
+ if (private->rx_buffer == NULL)
goto rx_buf_fail;
- }
+
private->tx_buffer = dma_alloc_coherent(d, 8192,
&private->tx_dma_handle,
GFP_KERNEL);
- if (private->tx_buffer == NULL) {
- pr_err("%s: no memory for tx buffer\n", __func__);
+ if (private->tx_buffer == NULL)
goto tx_buf_fail;
- }
SET_NETDEV_DEV(dev, &pdev->dev);
diff --git a/drivers/net/ethernet/dlink/dl2k.c b/drivers/net/ethernet/dlink/dl2k.c
index 110d26f..afa8e3a 100644
--- a/drivers/net/ethernet/dlink/dl2k.c
+++ b/drivers/net/ethernet/dlink/dl2k.c
@@ -580,12 +580,9 @@
skb = netdev_alloc_skb_ip_align(dev, np->rx_buf_sz);
np->rx_skbuff[i] = skb;
- if (skb == NULL) {
- printk (KERN_ERR
- "%s: alloc_list: allocate Rx buffer error! ",
- dev->name);
+ if (skb == NULL)
break;
- }
+
/* Rubicon now supports 40 bits of addressing space. */
np->rx_ring[i].fraginfo =
cpu_to_le64 ( pci_map_single (
diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h
index 29aff55..2e2700e 100644
--- a/drivers/net/ethernet/emulex/benet/be.h
+++ b/drivers/net/ethernet/emulex/benet/be.h
@@ -1,5 +1,5 @@
/*
- * Copyright (C) 2005 - 2011 Emulex
+ * Copyright (C) 2005 - 2013 Emulex
* All rights reserved.
*
* This program is free software; you can redistribute it and/or
diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c
index 3c9b4f1..f286ad2 100644
--- a/drivers/net/ethernet/emulex/benet/be_cmds.c
+++ b/drivers/net/ethernet/emulex/benet/be_cmds.c
@@ -1,5 +1,5 @@
/*
- * Copyright (C) 2005 - 2011 Emulex
+ * Copyright (C) 2005 - 2013 Emulex
* All rights reserved.
*
* This program is free software; you can redistribute it and/or
@@ -2667,10 +2667,8 @@
cmd.size = sizeof(struct be_cmd_req_set_mac_list);
cmd.va = dma_alloc_coherent(&adapter->pdev->dev, cmd.size,
&cmd.dma, GFP_KERNEL);
- if (!cmd.va) {
- dev_err(&adapter->pdev->dev, "Memory alloc failure\n");
+ if (!cmd.va)
return -ENOMEM;
- }
spin_lock_bh(&adapter->mcc_lock);
@@ -3202,6 +3200,31 @@
return status;
}
+int be_cmd_intr_set(struct be_adapter *adapter, bool intr_enable)
+{
+ struct be_mcc_wrb *wrb;
+ struct be_cmd_req_intr_set *req;
+ int status;
+
+ if (mutex_lock_interruptible(&adapter->mbox_lock))
+ return -1;
+
+ wrb = wrb_from_mbox(adapter);
+
+ req = embedded_payload(wrb);
+
+ be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
+ OPCODE_COMMON_SET_INTERRUPT_ENABLE, sizeof(*req),
+ wrb, NULL);
+
+ req->intr_enabled = intr_enable;
+
+ status = be_mbox_notify_wait(adapter);
+
+ mutex_unlock(&adapter->mbox_lock);
+ return status;
+}
+
int be_roce_mcc_cmd(void *netdev_handle, void *wrb_payload,
int wrb_payload_size, u16 *cmd_status, u16 *ext_status)
{
diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.h b/drivers/net/ethernet/emulex/benet/be_cmds.h
index 9697086..f2af855 100644
--- a/drivers/net/ethernet/emulex/benet/be_cmds.h
+++ b/drivers/net/ethernet/emulex/benet/be_cmds.h
@@ -1,5 +1,5 @@
/*
- * Copyright (C) 2005 - 2011 Emulex
+ * Copyright (C) 2005 - 2013 Emulex
* All rights reserved.
*
* This program is free software; you can redistribute it and/or
@@ -188,6 +188,7 @@
#define OPCODE_COMMON_GET_BEACON_STATE 70
#define OPCODE_COMMON_READ_TRANSRECV_DATA 73
#define OPCODE_COMMON_GET_PORT_NAME 77
+#define OPCODE_COMMON_SET_INTERRUPT_ENABLE 89
#define OPCODE_COMMON_GET_PHY_DETAILS 102
#define OPCODE_COMMON_SET_DRIVER_FUNCTION_CAP 103
#define OPCODE_COMMON_GET_CNTL_ADDITIONAL_ATTRIBUTES 121
@@ -1791,6 +1792,12 @@
u8 rsvd[3];
};
+struct be_cmd_req_intr_set {
+ struct be_cmd_req_hdr hdr;
+ u8 intr_enabled;
+ u8 rsvd[3];
+};
+
static inline bool check_privilege(struct be_adapter *adapter, u32 flags)
{
return flags & adapter->cmd_privileges ? true : false;
@@ -1938,3 +1945,4 @@
extern int be_cmd_get_if_id(struct be_adapter *adapter,
struct be_vf_cfg *vf_cfg, int vf_num);
extern int be_cmd_enable_vf(struct be_adapter *adapter, u8 domain);
+extern int be_cmd_intr_set(struct be_adapter *adapter, bool intr_enable);
diff --git a/drivers/net/ethernet/emulex/benet/be_ethtool.c b/drivers/net/ethernet/emulex/benet/be_ethtool.c
index 76b302f..07b7f27 100644
--- a/drivers/net/ethernet/emulex/benet/be_ethtool.c
+++ b/drivers/net/ethernet/emulex/benet/be_ethtool.c
@@ -1,5 +1,5 @@
/*
- * Copyright (C) 2005 - 2011 Emulex
+ * Copyright (C) 2005 - 2013 Emulex
* All rights reserved.
*
* This program is free software; you can redistribute it and/or
@@ -719,10 +719,8 @@
ddrdma_cmd.size = sizeof(struct be_cmd_req_ddrdma_test);
ddrdma_cmd.va = dma_alloc_coherent(&adapter->pdev->dev, ddrdma_cmd.size,
&ddrdma_cmd.dma, GFP_KERNEL);
- if (!ddrdma_cmd.va) {
- dev_err(&adapter->pdev->dev, "Memory allocation failure\n");
+ if (!ddrdma_cmd.va)
return -ENOMEM;
- }
for (i = 0; i < 2; i++) {
ret = be_cmd_ddr_dma_test(adapter, pattern[i],
@@ -845,11 +843,8 @@
eeprom_cmd.va = dma_alloc_coherent(&adapter->pdev->dev, eeprom_cmd.size,
&eeprom_cmd.dma, GFP_KERNEL);
- if (!eeprom_cmd.va) {
- dev_err(&adapter->pdev->dev,
- "Memory allocation failure. Could not read eeprom\n");
+ if (!eeprom_cmd.va)
return -ENOMEM;
- }
status = be_cmd_get_seeprom_data(adapter, &eeprom_cmd);
diff --git a/drivers/net/ethernet/emulex/benet/be_hw.h b/drivers/net/ethernet/emulex/benet/be_hw.h
index 62dc220..89e6d8c 100644
--- a/drivers/net/ethernet/emulex/benet/be_hw.h
+++ b/drivers/net/ethernet/emulex/benet/be_hw.h
@@ -1,5 +1,5 @@
/*
- * Copyright (C) 2005 - 2011 Emulex
+ * Copyright (C) 2005 - 2013 Emulex
* All rights reserved.
*
* This program is free software; you can redistribute it and/or
diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index 08e54f3..536afa2 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -1,5 +1,5 @@
/*
- * Copyright (C) 2005 - 2011 Emulex
+ * Copyright (C) 2005 - 2013 Emulex
* All rights reserved.
*
* This program is free software; you can redistribute it and/or
@@ -146,20 +146,16 @@
q->entry_size = entry_size;
mem->size = len * entry_size;
mem->va = dma_alloc_coherent(&adapter->pdev->dev, mem->size, &mem->dma,
- GFP_KERNEL);
+ GFP_KERNEL | __GFP_ZERO);
if (!mem->va)
return -ENOMEM;
- memset(mem->va, 0, mem->size);
return 0;
}
-static void be_intr_set(struct be_adapter *adapter, bool enable)
+static void be_reg_intr_set(struct be_adapter *adapter, bool enable)
{
u32 reg, enabled;
- if (adapter->eeh_error)
- return;
-
pci_read_config_dword(adapter->pdev, PCICFG_MEMBAR_CTRL_INT_CTRL_OFFSET,
®);
enabled = reg & MEMBAR_CTRL_INT_CTRL_HOSTINTR_MASK;
@@ -175,6 +171,22 @@
PCICFG_MEMBAR_CTRL_INT_CTRL_OFFSET, reg);
}
+static void be_intr_set(struct be_adapter *adapter, bool enable)
+{
+ int status = 0;
+
+ /* On lancer interrupts can't be controlled via this register */
+ if (lancer_chip(adapter))
+ return;
+
+ if (adapter->eeh_error)
+ return;
+
+ status = be_cmd_intr_set(adapter, enable);
+ if (status)
+ be_reg_intr_set(adapter, enable);
+}
+
static void be_rxq_notify(struct be_adapter *adapter, u16 qid, u16 posted)
{
u32 val = 0;
@@ -2435,9 +2447,6 @@
be_roce_dev_close(adapter);
- if (!lancer_chip(adapter))
- be_intr_set(adapter, false);
-
for_all_evt_queues(adapter, eqo, i)
napi_disable(&eqo->napi);
@@ -2525,9 +2534,6 @@
be_irq_register(adapter);
- if (!lancer_chip(adapter))
- be_intr_set(adapter, true);
-
for_all_rx_queues(adapter, rxo, i)
be_cq_notify(adapter, rxo->cq.id, true, 0);
@@ -2562,10 +2568,9 @@
cmd.size = sizeof(struct be_cmd_req_acpi_wol_magic_config);
cmd.va = dma_alloc_coherent(&adapter->pdev->dev, cmd.size, &cmd.dma,
- GFP_KERNEL);
+ GFP_KERNEL | __GFP_ZERO);
if (cmd.va == NULL)
return -1;
- memset(cmd.va, 0, cmd.size);
if (enable) {
status = pci_write_config_dword(adapter->pdev,
@@ -3457,11 +3462,9 @@
flash_cmd.size = sizeof(struct lancer_cmd_req_write_object)
+ LANCER_FW_DOWNLOAD_CHUNK;
flash_cmd.va = dma_alloc_coherent(&adapter->pdev->dev, flash_cmd.size,
- &flash_cmd.dma, GFP_KERNEL);
+ &flash_cmd.dma, GFP_KERNEL);
if (!flash_cmd.va) {
status = -ENOMEM;
- dev_err(&adapter->pdev->dev,
- "Memory allocation failure while flashing\n");
goto lancer_fw_exit;
}
@@ -3563,8 +3566,6 @@
&flash_cmd.dma, GFP_KERNEL);
if (!flash_cmd.va) {
status = -ENOMEM;
- dev_err(&adapter->pdev->dev,
- "Memory allocation failure while flashing\n");
goto be_fw_exit;
}
@@ -3791,12 +3792,13 @@
rx_filter->size = sizeof(struct be_cmd_req_rx_filter);
rx_filter->va = dma_alloc_coherent(&adapter->pdev->dev, rx_filter->size,
- &rx_filter->dma, GFP_KERNEL);
+ &rx_filter->dma,
+ GFP_KERNEL | __GFP_ZERO);
if (rx_filter->va == NULL) {
status = -ENOMEM;
goto free_mbox;
}
- memset(rx_filter->va, 0, rx_filter->size);
+
mutex_init(&adapter->mbox_lock);
spin_lock_init(&adapter->mcc_lock);
spin_lock_init(&adapter->mcc_cq_lock);
@@ -3838,10 +3840,9 @@
cmd->size = sizeof(struct be_cmd_req_get_stats_v1);
cmd->va = dma_alloc_coherent(&adapter->pdev->dev, cmd->size, &cmd->dma,
- GFP_KERNEL);
+ GFP_KERNEL | __GFP_ZERO);
if (cmd->va == NULL)
return -1;
- memset(cmd->va, 0, cmd->size);
return 0;
}
@@ -3853,6 +3854,7 @@
return;
be_roce_dev_remove(adapter);
+ be_intr_set(adapter, false);
cancel_delayed_work_sync(&adapter->func_recovery_work);
@@ -4142,11 +4144,11 @@
goto ctrl_clean;
}
- /* The INTR bit may be set in the card when probed by a kdump kernel
- * after a crash.
- */
- if (!lancer_chip(adapter))
- be_intr_set(adapter, false);
+ /* Wait for interrupts to quiesce after an FLR */
+ msleep(100);
+
+ /* Allow interrupts for other ULPs running on NIC function */
+ be_intr_set(adapter, true);
status = be_stats_init(adapter);
if (status)
diff --git a/drivers/net/ethernet/emulex/benet/be_roce.c b/drivers/net/ethernet/emulex/benet/be_roce.c
index 55d32aa..f3d126d 100644
--- a/drivers/net/ethernet/emulex/benet/be_roce.c
+++ b/drivers/net/ethernet/emulex/benet/be_roce.c
@@ -1,5 +1,5 @@
/*
- * Copyright (C) 2005 - 2011 Emulex
+ * Copyright (C) 2005 - 2013 Emulex
* All rights reserved.
*
* This program is free software; you can redistribute it and/or
diff --git a/drivers/net/ethernet/emulex/benet/be_roce.h b/drivers/net/ethernet/emulex/benet/be_roce.h
index db4ea80..2765729 100644
--- a/drivers/net/ethernet/emulex/benet/be_roce.h
+++ b/drivers/net/ethernet/emulex/benet/be_roce.h
@@ -1,5 +1,5 @@
/*
- * Copyright (C) 2005 - 2011 Emulex
+ * Copyright (C) 2005 - 2013 Emulex
* All rights reserved.
*
* This program is free software; you can redistribute it and/or
diff --git a/drivers/net/ethernet/faraday/ftgmac100.c b/drivers/net/ethernet/faraday/ftgmac100.c
index 7c361d1..21b85fb 100644
--- a/drivers/net/ethernet/faraday/ftgmac100.c
+++ b/drivers/net/ethernet/faraday/ftgmac100.c
@@ -780,12 +780,11 @@
priv->descs = dma_alloc_coherent(priv->dev,
sizeof(struct ftgmac100_descs),
- &priv->descs_dma_addr, GFP_KERNEL);
+ &priv->descs_dma_addr,
+ GFP_KERNEL | __GFP_ZERO);
if (!priv->descs)
return -ENOMEM;
- memset(priv->descs, 0, sizeof(struct ftgmac100_descs));
-
/* initialize RX ring */
ftgmac100_rxdes_set_end_of_ring(&priv->descs->rxdes[RX_QUEUE_ENTRIES - 1]);
@@ -1350,22 +1349,7 @@
},
};
-/******************************************************************************
- * initialization / finalization
- *****************************************************************************/
-static int __init ftgmac100_init(void)
-{
- pr_info("Loading version " DRV_VERSION " ...\n");
- return platform_driver_register(&ftgmac100_driver);
-}
-
-static void __exit ftgmac100_exit(void)
-{
- platform_driver_unregister(&ftgmac100_driver);
-}
-
-module_init(ftgmac100_init);
-module_exit(ftgmac100_exit);
+module_platform_driver(ftgmac100_driver);
MODULE_AUTHOR("Po-Yu Chuang <ratbert@faraday-tech.com>");
MODULE_DESCRIPTION("FTGMAC100 driver");
diff --git a/drivers/net/ethernet/faraday/ftmac100.c b/drivers/net/ethernet/faraday/ftmac100.c
index b5ea8fb..a6eda8d 100644
--- a/drivers/net/ethernet/faraday/ftmac100.c
+++ b/drivers/net/ethernet/faraday/ftmac100.c
@@ -732,13 +732,13 @@
{
int i;
- priv->descs = dma_alloc_coherent(priv->dev, sizeof(struct ftmac100_descs),
- &priv->descs_dma_addr, GFP_KERNEL);
+ priv->descs = dma_alloc_coherent(priv->dev,
+ sizeof(struct ftmac100_descs),
+ &priv->descs_dma_addr,
+ GFP_KERNEL | __GFP_ZERO);
if (!priv->descs)
return -ENOMEM;
- memset(priv->descs, 0, sizeof(struct ftmac100_descs));
-
/* initialize RX ring */
ftmac100_rxdes_set_end_of_ring(&priv->descs->rxdes[RX_QUEUE_ENTRIES - 1]);
diff --git a/drivers/net/ethernet/freescale/fec.c b/drivers/net/ethernet/freescale/fec.c
index 911d025..186222e 100644
--- a/drivers/net/ethernet/freescale/fec.c
+++ b/drivers/net/ethernet/freescale/fec.c
@@ -29,7 +29,6 @@
#include <linux/ioport.h>
#include <linux/slab.h>
#include <linux/interrupt.h>
-#include <linux/pci.h>
#include <linux/init.h>
#include <linux/delay.h>
#include <linux/netdevice.h>
@@ -743,8 +742,6 @@
skb = netdev_alloc_skb(ndev, pkt_len - 4 + NET_IP_ALIGN);
if (unlikely(!skb)) {
- printk("%s: Memory squeeze, dropping packet.\n",
- ndev->name);
ndev->stats.rx_dropped++;
} else {
skb_reserve(skb, NET_IP_ALIGN);
@@ -1602,11 +1599,9 @@
/* Allocate memory for buffer descriptors. */
cbd_base = dma_alloc_coherent(NULL, PAGE_SIZE, &fep->bd_dma,
- GFP_KERNEL);
- if (!cbd_base) {
- printk("FEC: allocate descriptor memory failed?\n");
+ GFP_KERNEL);
+ if (!cbd_base)
return -ENOMEM;
- }
spin_lock_init(&fep->hw_lock);
@@ -1739,16 +1734,10 @@
if (!r)
return -ENXIO;
- r = request_mem_region(r->start, resource_size(r), pdev->name);
- if (!r)
- return -EBUSY;
-
/* Init network device */
ndev = alloc_etherdev(sizeof(struct fec_enet_private));
- if (!ndev) {
- ret = -ENOMEM;
- goto failed_alloc_etherdev;
- }
+ if (!ndev)
+ return -ENOMEM;
SET_NETDEV_DEV(ndev, &pdev->dev);
@@ -1760,7 +1749,7 @@
(pdev->id_entry->driver_data & FEC_QUIRK_HAS_GBIT))
fep->pause_flag |= FEC_PAUSE_FLAG_AUTONEG;
- fep->hwp = ioremap(r->start, resource_size(r));
+ fep->hwp = devm_request_and_ioremap(&pdev->dev, r);
fep->pdev = pdev;
fep->dev_id = dev_id++;
@@ -1882,11 +1871,8 @@
clk_disable_unprepare(fep->clk_ptp);
failed_pin:
failed_clk:
- iounmap(fep->hwp);
failed_ioremap:
free_netdev(ndev);
-failed_alloc_etherdev:
- release_mem_region(r->start, resource_size(r));
return ret;
}
@@ -1896,7 +1882,6 @@
{
struct net_device *ndev = platform_get_drvdata(pdev);
struct fec_enet_private *fep = netdev_priv(ndev);
- struct resource *r;
int i;
unregister_netdev(ndev);
@@ -1912,13 +1897,8 @@
if (irq > 0)
free_irq(irq, ndev);
}
- iounmap(fep->hwp);
free_netdev(ndev);
- r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
- BUG_ON(!r);
- release_mem_region(r->start, resource_size(r));
-
platform_set_drvdata(pdev, NULL);
return 0;
diff --git a/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c b/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c
index 46df288..edc1200 100644
--- a/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c
+++ b/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c
@@ -177,8 +177,6 @@
received++;
netif_receive_skb(skb);
} else {
- dev_warn(fep->dev,
- "Memory squeeze, dropping packet.\n");
fep->stats.rx_dropped++;
skbn = skb;
}
@@ -309,8 +307,6 @@
received++;
netif_rx(skb);
} else {
- dev_warn(fep->dev,
- "Memory squeeze, dropping packet.\n");
fep->stats.rx_dropped++;
skbn = skb;
}
@@ -505,11 +501,9 @@
*/
for (i = 0, bdp = fep->rx_bd_base; i < fep->rx_ring; i++, bdp++) {
skb = netdev_alloc_skb(dev, ENET_RX_FRSIZE);
- if (skb == NULL) {
- dev_warn(fep->dev,
- "Memory squeeze, unable to allocate skb\n");
+ if (skb == NULL)
break;
- }
+
skb_align(skb, ENET_RX_ALIGN);
fep->rx_skbuff[i] = skb;
CBDW_BUFADDR(bdp,
@@ -593,13 +587,8 @@
/* Alloc new skb */
new_skb = netdev_alloc_skb(dev, skb->len + 4);
- if (!new_skb) {
- if (net_ratelimit()) {
- dev_warn(fep->dev,
- "Memory squeeze, dropping tx packet.\n");
- }
+ if (!new_skb)
return NULL;
- }
/* Make sure new skb is properly aligned */
skb_align(new_skb, 4);
diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c
index d2c5441..1b468a8 100644
--- a/drivers/net/ethernet/freescale/gianfar.c
+++ b/drivers/net/ethernet/freescale/gianfar.c
@@ -245,14 +245,13 @@
/* Allocate memory for the buffer descriptors */
vaddr = dma_alloc_coherent(dev,
- sizeof(struct txbd8) * priv->total_tx_ring_size +
- sizeof(struct rxbd8) * priv->total_rx_ring_size,
- &addr, GFP_KERNEL);
- if (!vaddr) {
- netif_err(priv, ifup, ndev,
- "Could not allocate buffer descriptors!\n");
+ (priv->total_tx_ring_size *
+ sizeof(struct txbd8)) +
+ (priv->total_rx_ring_size *
+ sizeof(struct rxbd8)),
+ &addr, GFP_KERNEL);
+ if (!vaddr)
return -ENOMEM;
- }
for (i = 0; i < priv->num_tx_queues; i++) {
tx_queue = priv->tx_queue[i];
diff --git a/drivers/net/ethernet/fujitsu/fmvj18x_cs.c b/drivers/net/ethernet/fujitsu/fmvj18x_cs.c
index 2418faf..84125707 100644
--- a/drivers/net/ethernet/fujitsu/fmvj18x_cs.c
+++ b/drivers/net/ethernet/fujitsu/fmvj18x_cs.c
@@ -1003,8 +1003,6 @@
}
skb = netdev_alloc_skb(dev, pkt_len + 2);
if (skb == NULL) {
- netdev_notice(dev, "Memory squeeze, dropping packet (len %d)\n",
- pkt_len);
outb(F_SKP_PKT, ioaddr + RX_SKIP);
dev->stats.rx_dropped++;
break;
diff --git a/drivers/net/ethernet/i825xx/82596.c b/drivers/net/ethernet/i825xx/82596.c
index 1c54e22..e388161 100644
--- a/drivers/net/ethernet/i825xx/82596.c
+++ b/drivers/net/ethernet/i825xx/82596.c
@@ -798,16 +798,14 @@
#ifdef __mc68000__
cache_clear(virt_to_phys(newskb->data), PKT_BUF_SZ);
#endif
- }
- else
+ } else {
skb = netdev_alloc_skb(dev, pkt_len + 2);
+ }
memory_squeeze:
if (skb == NULL) {
/* XXX tulip.c can defer packets here!! */
- printk(KERN_WARNING "%s: i596_rx Memory squeeze, dropping packet.\n", dev->name);
dev->stats.rx_dropped++;
- }
- else {
+ } else {
if (!rx_in_place) {
/* 16 byte align the data fields */
skb_reserve(skb, 2);
diff --git a/drivers/net/ethernet/i825xx/lib82596.c b/drivers/net/ethernet/i825xx/lib82596.c
index f045ea4..d653bac 100644
--- a/drivers/net/ethernet/i825xx/lib82596.c
+++ b/drivers/net/ethernet/i825xx/lib82596.c
@@ -715,14 +715,12 @@
rbd->v_data = newskb->data;
rbd->b_data = SWAP32(dma_addr);
DMA_WBACK_INV(dev, rbd, sizeof(struct i596_rbd));
- } else
+ } else {
skb = netdev_alloc_skb_ip_align(dev, pkt_len);
+ }
memory_squeeze:
if (skb == NULL) {
/* XXX tulip.c can defer packets here!! */
- printk(KERN_ERR
- "%s: i596_rx Memory squeeze, dropping packet.\n",
- dev->name);
dev->stats.rx_dropped++;
} else {
if (!rx_in_place) {
diff --git a/drivers/net/ethernet/ibm/ehea/ehea_main.c b/drivers/net/ethernet/ibm/ehea/ehea_main.c
index 328f47c..0296334 100644
--- a/drivers/net/ethernet/ibm/ehea/ehea_main.c
+++ b/drivers/net/ethernet/ibm/ehea/ehea_main.c
@@ -402,7 +402,6 @@
skb_arr_rq1[index] = netdev_alloc_skb(dev,
EHEA_L_PKT_SIZE);
if (!skb_arr_rq1[index]) {
- netdev_info(dev, "Unable to allocate enough skb in the array\n");
pr->rq1_skba.os_skbs = fill_wqes - i;
break;
}
@@ -432,10 +431,8 @@
for (i = 0; i < nr_rq1a; i++) {
skb_arr_rq1[i] = netdev_alloc_skb(dev, EHEA_L_PKT_SIZE);
- if (!skb_arr_rq1[i]) {
- netdev_info(dev, "Not enough memory to allocate skb array\n");
+ if (!skb_arr_rq1[i])
break;
- }
}
/* Ring doorbell */
ehea_update_rq1a(pr->qp, i - 1);
@@ -695,10 +692,8 @@
skb = netdev_alloc_skb(dev,
EHEA_L_PKT_SIZE);
- if (!skb) {
- netdev_err(dev, "Not enough memory to allocate skb\n");
+ if (!skb)
break;
- }
}
skb_copy_to_linear_data(skb, ((char *)cqe) + 64,
cqe->num_bytes_transfered - 4);
diff --git a/drivers/net/ethernet/ibm/emac/mal.c b/drivers/net/ethernet/ibm/emac/mal.c
index 1f7ecf5..610ed223 100644
--- a/drivers/net/ethernet/ibm/emac/mal.c
+++ b/drivers/net/ethernet/ibm/emac/mal.c
@@ -637,17 +637,12 @@
bd_size = sizeof(struct mal_descriptor) *
(NUM_TX_BUFF * mal->num_tx_chans +
NUM_RX_BUFF * mal->num_rx_chans);
- mal->bd_virt =
- dma_alloc_coherent(&ofdev->dev, bd_size, &mal->bd_dma,
- GFP_KERNEL);
+ mal->bd_virt = dma_alloc_coherent(&ofdev->dev, bd_size, &mal->bd_dma,
+ GFP_KERNEL | __GFP_ZERO);
if (mal->bd_virt == NULL) {
- printk(KERN_ERR
- "mal%d: out of memory allocating RX/TX descriptors!\n",
- index);
err = -ENOMEM;
goto fail_unmap;
}
- memset(mal->bd_virt, 0, bd_size);
for (i = 0; i < mal->num_tx_chans; ++i)
set_mal_dcrn(mal, MAL_TXCTPR(i), mal->bd_dma +
diff --git a/drivers/net/ethernet/ibm/ibmveth.c b/drivers/net/ethernet/ibm/ibmveth.c
index c859771..302d594 100644
--- a/drivers/net/ethernet/ibm/ibmveth.c
+++ b/drivers/net/ethernet/ibm/ibmveth.c
@@ -556,11 +556,9 @@
adapter->rx_queue.queue_len = sizeof(struct ibmveth_rx_q_entry) *
rxq_entries;
adapter->rx_queue.queue_addr =
- dma_alloc_coherent(dev, adapter->rx_queue.queue_len,
- &adapter->rx_queue.queue_dma, GFP_KERNEL);
-
+ dma_alloc_coherent(dev, adapter->rx_queue.queue_len,
+ &adapter->rx_queue.queue_dma, GFP_KERNEL);
if (!adapter->rx_queue.queue_addr) {
- netdev_err(netdev, "unable to allocate rx queue pages\n");
rc = -ENOMEM;
goto err_out;
}
diff --git a/drivers/net/ethernet/intel/e1000/e1000_ethtool.c b/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
index 43462d5..a9f9c79 100644
--- a/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
+++ b/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
@@ -1020,12 +1020,11 @@
txdr->size = txdr->count * sizeof(struct e1000_tx_desc);
txdr->size = ALIGN(txdr->size, 4096);
txdr->desc = dma_alloc_coherent(&pdev->dev, txdr->size, &txdr->dma,
- GFP_KERNEL);
+ GFP_KERNEL | __GFP_ZERO);
if (!txdr->desc) {
ret_val = 2;
goto err_nomem;
}
- memset(txdr->desc, 0, txdr->size);
txdr->next_to_use = txdr->next_to_clean = 0;
ew32(TDBAL, ((u64)txdr->dma & 0x00000000FFFFFFFF));
@@ -1075,12 +1074,11 @@
rxdr->size = rxdr->count * sizeof(struct e1000_rx_desc);
rxdr->desc = dma_alloc_coherent(&pdev->dev, rxdr->size, &rxdr->dma,
- GFP_KERNEL);
+ GFP_KERNEL | __GFP_ZERO);
if (!rxdr->desc) {
ret_val = 5;
goto err_nomem;
}
- memset(rxdr->desc, 0, rxdr->size);
rxdr->next_to_use = rxdr->next_to_clean = 0;
rctl = er32(RCTL);
diff --git a/drivers/net/ethernet/intel/e1000/e1000_main.c b/drivers/net/ethernet/intel/e1000/e1000_main.c
index 8502c62..d98e1d09 100644
--- a/drivers/net/ethernet/intel/e1000/e1000_main.c
+++ b/drivers/net/ethernet/intel/e1000/e1000_main.c
@@ -1516,8 +1516,6 @@
if (!txdr->desc) {
setup_tx_desc_die:
vfree(txdr->buffer_info);
- e_err(probe, "Unable to allocate memory for the Tx descriptor "
- "ring\n");
return -ENOMEM;
}
@@ -1707,10 +1705,7 @@
rxdr->desc = dma_alloc_coherent(&pdev->dev, rxdr->size, &rxdr->dma,
GFP_KERNEL);
-
if (!rxdr->desc) {
- e_err(probe, "Unable to allocate memory for the Rx descriptor "
- "ring\n");
setup_rx_desc_die:
vfree(rxdr->buffer_info);
return -ENOMEM;
@@ -1729,8 +1724,6 @@
if (!rxdr->desc) {
dma_free_coherent(&pdev->dev, rxdr->size, olddesc,
olddma);
- e_err(probe, "Unable to allocate memory for the Rx "
- "descriptor ring\n");
goto setup_rx_desc_die;
}
diff --git a/drivers/net/ethernet/intel/e1000e/80003es2lan.c b/drivers/net/ethernet/intel/e1000e/80003es2lan.c
index e099138..b71c850 100644
--- a/drivers/net/ethernet/intel/e1000e/80003es2lan.c
+++ b/drivers/net/ethernet/intel/e1000e/80003es2lan.c
@@ -37,7 +37,9 @@
* "index + 5".
*/
static const u16 e1000_gg82563_cable_length_table[] = {
- 0, 60, 115, 150, 150, 60, 115, 150, 180, 180, 0xFF };
+ 0, 60, 115, 150, 150, 60, 115, 150, 180, 180, 0xFF
+};
+
#define GG82563_CABLE_LENGTH_TABLE_SIZE \
ARRAY_SIZE(e1000_gg82563_cable_length_table)
@@ -116,7 +118,7 @@
nvm->type = e1000_nvm_eeprom_spi;
size = (u16)((eecd & E1000_EECD_SIZE_EX_MASK) >>
- E1000_EECD_SIZE_EX_SHIFT);
+ E1000_EECD_SIZE_EX_SHIFT);
/* Added to a constant, "size" becomes the left-shift value
* for setting word_size.
@@ -393,7 +395,7 @@
* before the device has completed the "Page Select" MDI
* transaction. So we wait 200us after each MDI command...
*/
- udelay(200);
+ usleep_range(200, 400);
/* ...and verify the command was successful. */
ret_val = e1000e_read_phy_reg_mdic(hw, page_select, &temp);
@@ -403,17 +405,17 @@
return -E1000_ERR_PHY;
}
- udelay(200);
+ usleep_range(200, 400);
ret_val = e1000e_read_phy_reg_mdic(hw,
- MAX_PHY_REG_ADDRESS & offset,
- data);
+ MAX_PHY_REG_ADDRESS & offset,
+ data);
- udelay(200);
+ usleep_range(200, 400);
} else {
ret_val = e1000e_read_phy_reg_mdic(hw,
- MAX_PHY_REG_ADDRESS & offset,
- data);
+ MAX_PHY_REG_ADDRESS & offset,
+ data);
}
e1000_release_phy_80003es2lan(hw);
@@ -462,7 +464,7 @@
* before the device has completed the "Page Select" MDI
* transaction. So we wait 200us after each MDI command...
*/
- udelay(200);
+ usleep_range(200, 400);
/* ...and verify the command was successful. */
ret_val = e1000e_read_phy_reg_mdic(hw, page_select, &temp);
@@ -472,17 +474,17 @@
return -E1000_ERR_PHY;
}
- udelay(200);
+ usleep_range(200, 400);
ret_val = e1000e_write_phy_reg_mdic(hw,
- MAX_PHY_REG_ADDRESS & offset,
- data);
+ MAX_PHY_REG_ADDRESS &
+ offset, data);
- udelay(200);
+ usleep_range(200, 400);
} else {
ret_val = e1000e_write_phy_reg_mdic(hw,
- MAX_PHY_REG_ADDRESS & offset,
- data);
+ MAX_PHY_REG_ADDRESS &
+ offset, data);
}
e1000_release_phy_80003es2lan(hw);
@@ -580,7 +582,7 @@
e_dbg("Waiting for forced speed/duplex link on GG82563 phy.\n");
ret_val = e1000e_phy_has_link_generic(hw, PHY_FORCE_LIMIT,
- 100000, &link);
+ 100000, &link);
if (ret_val)
return ret_val;
@@ -595,7 +597,7 @@
/* Try once more */
ret_val = e1000e_phy_has_link_generic(hw, PHY_FORCE_LIMIT,
- 100000, &link);
+ 100000, &link);
if (ret_val)
return ret_val;
}
@@ -666,14 +668,12 @@
s32 ret_val;
if (hw->phy.media_type == e1000_media_type_copper) {
- ret_val = e1000e_get_speed_and_duplex_copper(hw,
- speed,
- duplex);
+ ret_val = e1000e_get_speed_and_duplex_copper(hw, speed, duplex);
hw->phy.ops.cfg_on_link_up(hw);
} else {
ret_val = e1000e_get_speed_and_duplex_fiber_serdes(hw,
- speed,
- duplex);
+ speed,
+ duplex);
}
return ret_val;
@@ -754,9 +754,9 @@
/* Initialize identification LED */
ret_val = mac->ops.id_led_init(hw);
+ /* An error is not fatal and we should not stop init due to this */
if (ret_val)
e_dbg("Error initializing identification LED\n");
- /* This is not fatal and we should not stop init due to this */
/* Disabling VLAN filtering */
e_dbg("Initializing the IEEE VLAN\n");
@@ -784,14 +784,14 @@
/* Set the transmit descriptor write-back policy */
reg_data = er32(TXDCTL(0));
- reg_data = (reg_data & ~E1000_TXDCTL_WTHRESH) |
- E1000_TXDCTL_FULL_TX_DESC_WB | E1000_TXDCTL_COUNT_DESC;
+ reg_data = ((reg_data & ~E1000_TXDCTL_WTHRESH) |
+ E1000_TXDCTL_FULL_TX_DESC_WB | E1000_TXDCTL_COUNT_DESC);
ew32(TXDCTL(0), reg_data);
/* ...for both queues. */
reg_data = er32(TXDCTL(1));
- reg_data = (reg_data & ~E1000_TXDCTL_WTHRESH) |
- E1000_TXDCTL_FULL_TX_DESC_WB | E1000_TXDCTL_COUNT_DESC;
+ reg_data = ((reg_data & ~E1000_TXDCTL_WTHRESH) |
+ E1000_TXDCTL_FULL_TX_DESC_WB | E1000_TXDCTL_COUNT_DESC);
ew32(TXDCTL(1), reg_data);
/* Enable retransmit on late collisions */
@@ -818,13 +818,12 @@
/* default to true to enable the MDIC W/A */
hw->dev_spec.e80003es2lan.mdic_wa_enable = true;
- ret_val = e1000_read_kmrn_reg_80003es2lan(hw,
- E1000_KMRNCTRLSTA_OFFSET >>
- E1000_KMRNCTRLSTA_OFFSET_SHIFT,
- &i);
+ ret_val =
+ e1000_read_kmrn_reg_80003es2lan(hw, E1000_KMRNCTRLSTA_OFFSET >>
+ E1000_KMRNCTRLSTA_OFFSET_SHIFT, &i);
if (!ret_val) {
if ((i & E1000_KMRNCTRLSTA_OPMODE_MASK) ==
- E1000_KMRNCTRLSTA_OPMODE_INBAND_MDIO)
+ E1000_KMRNCTRLSTA_OPMODE_INBAND_MDIO)
hw->dev_spec.e80003es2lan.mdic_wa_enable = false;
}
@@ -891,7 +890,7 @@
{
struct e1000_phy_info *phy = &hw->phy;
s32 ret_val;
- u32 ctrl_ext;
+ u32 reg;
u16 data;
ret_val = e1e_rphy(hw, GG82563_PHY_MAC_SPEC_CTRL, &data);
@@ -954,22 +953,19 @@
}
/* Bypass Rx and Tx FIFO's */
- ret_val = e1000_write_kmrn_reg_80003es2lan(hw,
- E1000_KMRNCTRLSTA_OFFSET_FIFO_CTRL,
- E1000_KMRNCTRLSTA_FIFO_CTRL_RX_BYPASS |
- E1000_KMRNCTRLSTA_FIFO_CTRL_TX_BYPASS);
+ reg = E1000_KMRNCTRLSTA_OFFSET_FIFO_CTRL;
+ data = (E1000_KMRNCTRLSTA_FIFO_CTRL_RX_BYPASS |
+ E1000_KMRNCTRLSTA_FIFO_CTRL_TX_BYPASS);
+ ret_val = e1000_write_kmrn_reg_80003es2lan(hw, reg, data);
if (ret_val)
return ret_val;
- ret_val = e1000_read_kmrn_reg_80003es2lan(hw,
- E1000_KMRNCTRLSTA_OFFSET_MAC2PHY_OPMODE,
- &data);
+ reg = E1000_KMRNCTRLSTA_OFFSET_MAC2PHY_OPMODE;
+ ret_val = e1000_read_kmrn_reg_80003es2lan(hw, reg, &data);
if (ret_val)
return ret_val;
data |= E1000_KMRNCTRLSTA_OPMODE_E_IDLE;
- ret_val = e1000_write_kmrn_reg_80003es2lan(hw,
- E1000_KMRNCTRLSTA_OFFSET_MAC2PHY_OPMODE,
- data);
+ ret_val = e1000_write_kmrn_reg_80003es2lan(hw, reg, data);
if (ret_val)
return ret_val;
@@ -982,9 +978,9 @@
if (ret_val)
return ret_val;
- ctrl_ext = er32(CTRL_EXT);
- ctrl_ext &= ~(E1000_CTRL_EXT_LINK_MODE_MASK);
- ew32(CTRL_EXT, ctrl_ext);
+ reg = er32(CTRL_EXT);
+ reg &= ~E1000_CTRL_EXT_LINK_MODE_MASK;
+ ew32(CTRL_EXT, reg);
ret_val = e1e_rphy(hw, GG82563_PHY_PWR_MGMT_CTRL, &data);
if (ret_val)
@@ -1049,27 +1045,29 @@
* polling the phy; this fixes erroneous timeouts at 10Mbps.
*/
ret_val = e1000_write_kmrn_reg_80003es2lan(hw, GG82563_REG(0x34, 4),
- 0xFFFF);
+ 0xFFFF);
if (ret_val)
return ret_val;
ret_val = e1000_read_kmrn_reg_80003es2lan(hw, GG82563_REG(0x34, 9),
- ®_data);
+ ®_data);
if (ret_val)
return ret_val;
reg_data |= 0x3F;
ret_val = e1000_write_kmrn_reg_80003es2lan(hw, GG82563_REG(0x34, 9),
- reg_data);
+ reg_data);
if (ret_val)
return ret_val;
- ret_val = e1000_read_kmrn_reg_80003es2lan(hw,
- E1000_KMRNCTRLSTA_OFFSET_INB_CTRL,
- ®_data);
+ ret_val =
+ e1000_read_kmrn_reg_80003es2lan(hw,
+ E1000_KMRNCTRLSTA_OFFSET_INB_CTRL,
+ ®_data);
if (ret_val)
return ret_val;
reg_data |= E1000_KMRNCTRLSTA_INB_CTRL_DIS_PADDING;
- ret_val = e1000_write_kmrn_reg_80003es2lan(hw,
- E1000_KMRNCTRLSTA_OFFSET_INB_CTRL,
- reg_data);
+ ret_val =
+ e1000_write_kmrn_reg_80003es2lan(hw,
+ E1000_KMRNCTRLSTA_OFFSET_INB_CTRL,
+ reg_data);
if (ret_val)
return ret_val;
@@ -1096,7 +1094,7 @@
if (hw->phy.media_type == e1000_media_type_copper) {
ret_val = e1000e_get_speed_and_duplex_copper(hw, &speed,
- &duplex);
+ &duplex);
if (ret_val)
return ret_val;
@@ -1125,9 +1123,10 @@
u16 reg_data, reg_data2;
reg_data = E1000_KMRNCTRLSTA_HD_CTRL_10_100_DEFAULT;
- ret_val = e1000_write_kmrn_reg_80003es2lan(hw,
- E1000_KMRNCTRLSTA_OFFSET_HD_CTRL,
- reg_data);
+ ret_val =
+ e1000_write_kmrn_reg_80003es2lan(hw,
+ E1000_KMRNCTRLSTA_OFFSET_HD_CTRL,
+ reg_data);
if (ret_val)
return ret_val;
@@ -1171,9 +1170,10 @@
u32 i = 0;
reg_data = E1000_KMRNCTRLSTA_HD_CTRL_1000_DEFAULT;
- ret_val = e1000_write_kmrn_reg_80003es2lan(hw,
- E1000_KMRNCTRLSTA_OFFSET_HD_CTRL,
- reg_data);
+ ret_val =
+ e1000_write_kmrn_reg_80003es2lan(hw,
+ E1000_KMRNCTRLSTA_OFFSET_HD_CTRL,
+ reg_data);
if (ret_val)
return ret_val;
@@ -1220,7 +1220,7 @@
return ret_val;
kmrnctrlsta = ((offset << E1000_KMRNCTRLSTA_OFFSET_SHIFT) &
- E1000_KMRNCTRLSTA_OFFSET) | E1000_KMRNCTRLSTA_REN;
+ E1000_KMRNCTRLSTA_OFFSET) | E1000_KMRNCTRLSTA_REN;
ew32(KMRNCTRLSTA, kmrnctrlsta);
e1e_flush();
@@ -1255,7 +1255,7 @@
return ret_val;
kmrnctrlsta = ((offset << E1000_KMRNCTRLSTA_OFFSET_SHIFT) &
- E1000_KMRNCTRLSTA_OFFSET) | data;
+ E1000_KMRNCTRLSTA_OFFSET) | data;
ew32(KMRNCTRLSTA, kmrnctrlsta);
e1e_flush();
@@ -1419,4 +1419,3 @@
.phy_ops = &es2_phy_ops,
.nvm_ops = &es2_nvm_ops,
};
-
diff --git a/drivers/net/ethernet/intel/e1000e/82571.c b/drivers/net/ethernet/intel/e1000e/82571.c
index 2faffbde..7380442 100644
--- a/drivers/net/ethernet/intel/e1000e/82571.c
+++ b/drivers/net/ethernet/intel/e1000e/82571.c
@@ -184,7 +184,7 @@
default:
nvm->type = e1000_nvm_eeprom_spi;
size = (u16)((eecd & E1000_EECD_SIZE_EX_MASK) >>
- E1000_EECD_SIZE_EX_SHIFT);
+ E1000_EECD_SIZE_EX_SHIFT);
/* Added to a constant, "size" becomes the left-shift value
* for setting word_size.
*/
@@ -437,7 +437,7 @@
return ret_val;
phy->id = (u32)(phy_id << 16);
- udelay(20);
+ usleep_range(20, 40);
ret_val = e1e_rphy(hw, MII_PHYSID2, &phy_id);
if (ret_val)
return ret_val;
@@ -482,7 +482,7 @@
if (!(swsm & E1000_SWSM_SMBI))
break;
- udelay(50);
+ usleep_range(50, 100);
i++;
}
@@ -499,7 +499,7 @@
if (er32(SWSM) & E1000_SWSM_SWESMBI)
break;
- udelay(50);
+ usleep_range(50, 100);
}
if (i == fw_timeout) {
@@ -526,6 +526,7 @@
swsm &= ~(E1000_SWSM_SMBI | E1000_SWSM_SWESMBI);
ew32(SWSM, swsm);
}
+
/**
* e1000_get_hw_semaphore_82573 - Acquire hardware semaphore
* @hw: pointer to the HW structure
@@ -846,9 +847,9 @@
}
for (i = 0; i < words; i++) {
- eewr = (data[i] << E1000_NVM_RW_REG_DATA) |
- ((offset+i) << E1000_NVM_RW_ADDR_SHIFT) |
- E1000_NVM_RW_REG_START;
+ eewr = ((data[i] << E1000_NVM_RW_REG_DATA) |
+ ((offset + i) << E1000_NVM_RW_ADDR_SHIFT) |
+ E1000_NVM_RW_REG_START);
ret_val = e1000e_poll_eerd_eewr_done(hw, E1000_NVM_POLL_WRITE);
if (ret_val)
@@ -875,8 +876,7 @@
s32 timeout = PHY_CFG_TIMEOUT;
while (timeout) {
- if (er32(EEMNGCTL) &
- E1000_NVM_CFG_DONE_PORT_0)
+ if (er32(EEMNGCTL) & E1000_NVM_CFG_DONE_PORT_0)
break;
usleep_range(1000, 2000);
timeout--;
@@ -1022,7 +1022,7 @@
}
if (hw->nvm.type == e1000_nvm_flash_hw) {
- udelay(10);
+ usleep_range(10, 20);
ctrl_ext = er32(CTRL_EXT);
ctrl_ext |= E1000_CTRL_EXT_EE_RST;
ew32(CTRL_EXT, ctrl_ext);
@@ -1095,9 +1095,9 @@
/* Initialize identification LED */
ret_val = mac->ops.id_led_init(hw);
+ /* An error is not fatal and we should not stop init due to this */
if (ret_val)
e_dbg("Error initializing identification LED\n");
- /* This is not fatal and we should not stop init due to this */
/* Disabling VLAN filtering */
e_dbg("Initializing the IEEE VLAN\n");
@@ -1122,9 +1122,8 @@
/* Set the transmit descriptor write-back policy */
reg_data = er32(TXDCTL(0));
- reg_data = (reg_data & ~E1000_TXDCTL_WTHRESH) |
- E1000_TXDCTL_FULL_TX_DESC_WB |
- E1000_TXDCTL_COUNT_DESC;
+ reg_data = ((reg_data & ~E1000_TXDCTL_WTHRESH) |
+ E1000_TXDCTL_FULL_TX_DESC_WB | E1000_TXDCTL_COUNT_DESC);
ew32(TXDCTL(0), reg_data);
/* ...for both queues. */
@@ -1140,9 +1139,9 @@
break;
default:
reg_data = er32(TXDCTL(1));
- reg_data = (reg_data & ~E1000_TXDCTL_WTHRESH) |
- E1000_TXDCTL_FULL_TX_DESC_WB |
- E1000_TXDCTL_COUNT_DESC;
+ reg_data = ((reg_data & ~E1000_TXDCTL_WTHRESH) |
+ E1000_TXDCTL_FULL_TX_DESC_WB |
+ E1000_TXDCTL_COUNT_DESC);
ew32(TXDCTL(1), reg_data);
break;
}
@@ -1530,7 +1529,7 @@
status = er32(STATUS);
er32(RXCW);
/* SYNCH bit and IV bit are sticky */
- udelay(10);
+ usleep_range(10, 20);
rxcw = er32(RXCW);
if ((rxcw & E1000_RXCW_SYNCH) && !(rxcw & E1000_RXCW_IV)) {
@@ -1633,7 +1632,7 @@
* the IV bit and restart Autoneg
*/
for (i = 0; i < AN_RETRY_COUNT; i++) {
- udelay(10);
+ usleep_range(10, 20);
rxcw = er32(RXCW);
if ((rxcw & E1000_RXCW_SYNCH) &&
(rxcw & E1000_RXCW_C))
@@ -2066,4 +2065,3 @@
.phy_ops = &e82_phy_ops_bm,
.nvm_ops = &e82571_nvm_ops,
};
-
diff --git a/drivers/net/ethernet/intel/e1000e/82571.h b/drivers/net/ethernet/intel/e1000e/82571.h
index 85cb1a3..08e24dc 100644
--- a/drivers/net/ethernet/intel/e1000e/82571.h
+++ b/drivers/net/ethernet/intel/e1000e/82571.h
@@ -44,6 +44,8 @@
#define E1000_EIAC_82574 0x000DC /* Ext. Interrupt Auto Clear - RW */
#define E1000_EIAC_MASK_82574 0x01F00000
+#define E1000_IVAR_INT_ALLOC_VALID 0x8
+
/* Manageability Operation Mode mask */
#define E1000_NVM_INIT_CTRL2_MNGM 0x6000
diff --git a/drivers/net/ethernet/intel/e1000e/defines.h b/drivers/net/ethernet/intel/e1000e/defines.h
index fc3a4fe..b7c664f 100644
--- a/drivers/net/ethernet/intel/e1000e/defines.h
+++ b/drivers/net/ethernet/intel/e1000e/defines.h
@@ -66,7 +66,7 @@
#define E1000_CTRL_EXT_LINK_MODE_PCIE_SERDES 0x00C00000
#define E1000_CTRL_EXT_EIAME 0x01000000
#define E1000_CTRL_EXT_DRV_LOAD 0x10000000 /* Driver loaded bit for FW */
-#define E1000_CTRL_EXT_IAME 0x08000000 /* Interrupt acknowledge Auto-mask */
+#define E1000_CTRL_EXT_IAME 0x08000000 /* Int ACK Auto-mask */
#define E1000_CTRL_EXT_PBA_CLR 0x80000000 /* PBA Clear */
#define E1000_CTRL_EXT_LSECCK 0x00001000
#define E1000_CTRL_EXT_PHYPDEN 0x00100000
@@ -216,6 +216,8 @@
#define E1000_CTRL_MEHE 0x00080000 /* Memory Error Handling Enable */
#define E1000_CTRL_SWDPIN0 0x00040000 /* SWDPIN 0 value */
#define E1000_CTRL_SWDPIN1 0x00080000 /* SWDPIN 1 value */
+#define E1000_CTRL_ADVD3WUC 0x00100000 /* D3 WUC */
+#define E1000_CTRL_EN_PHY_PWR_MGMT 0x00200000 /* PHY PM enable */
#define E1000_CTRL_SWDPIO0 0x00400000 /* SWDPIN 0 Input or output */
#define E1000_CTRL_RST 0x04000000 /* Global reset */
#define E1000_CTRL_RFCE 0x08000000 /* Receive Flow Control enable */
@@ -239,12 +241,11 @@
#define E1000_STATUS_SPEED_1000 0x00000080 /* Speed 1000Mb/s */
#define E1000_STATUS_LAN_INIT_DONE 0x00000200 /* Lan Init Completion by NVM */
#define E1000_STATUS_PHYRA 0x00000400 /* PHY Reset Asserted */
-#define E1000_STATUS_GIO_MASTER_ENABLE 0x00080000 /* Status of Master requests. */
+#define E1000_STATUS_GIO_MASTER_ENABLE 0x00080000 /* Master Req status */
#define HALF_DUPLEX 1
#define FULL_DUPLEX 2
-
#define ADVERTISE_10_HALF 0x0001
#define ADVERTISE_10_FULL 0x0002
#define ADVERTISE_100_HALF 0x0004
@@ -311,6 +312,7 @@
/* SerDes Control */
#define E1000_SCTL_DISABLE_SERDES_LOOPBACK 0x0400
+#define E1000_SCTL_ENABLE_SERDES_LOOPBACK 0x0410
/* Receive Checksum Control */
#define E1000_RXCSUM_TUOFL 0x00000200 /* TCP / UDP checksum offload */
@@ -400,7 +402,8 @@
#define E1000_ICR_RXDMT0 0x00000010 /* Rx desc min. threshold (0) */
#define E1000_ICR_RXT0 0x00000080 /* Rx timer intr (ring 0) */
#define E1000_ICR_ECCER 0x00400000 /* Uncorrectable ECC Error */
-#define E1000_ICR_INT_ASSERTED 0x80000000 /* If this bit asserted, the driver should claim the interrupt */
+/* If this bit asserted, the driver should claim the interrupt */
+#define E1000_ICR_INT_ASSERTED 0x80000000
#define E1000_ICR_RXQ0 0x00100000 /* Rx Queue 0 Interrupt */
#define E1000_ICR_RXQ1 0x00200000 /* Rx Queue 1 Interrupt */
#define E1000_ICR_TXQ0 0x00400000 /* Tx Queue 0 Interrupt */
@@ -583,13 +586,13 @@
#define E1000_EECD_SEC1VAL 0x00400000 /* Sector One Valid */
#define E1000_EECD_SEC1VAL_VALID_MASK (E1000_EECD_AUTO_RD | E1000_EECD_PRES)
-#define E1000_NVM_RW_REG_DATA 16 /* Offset to data in NVM read/write registers */
-#define E1000_NVM_RW_REG_DONE 2 /* Offset to READ/WRITE done bit */
-#define E1000_NVM_RW_REG_START 1 /* Start operation */
-#define E1000_NVM_RW_ADDR_SHIFT 2 /* Shift to the address bits */
-#define E1000_NVM_POLL_WRITE 1 /* Flag for polling for write complete */
-#define E1000_NVM_POLL_READ 0 /* Flag for polling for read complete */
-#define E1000_FLASH_UPDATES 2000
+#define E1000_NVM_RW_REG_DATA 16 /* Offset to data in NVM r/w regs */
+#define E1000_NVM_RW_REG_DONE 2 /* Offset to READ/WRITE done bit */
+#define E1000_NVM_RW_REG_START 1 /* Start operation */
+#define E1000_NVM_RW_ADDR_SHIFT 2 /* Shift to the address bits */
+#define E1000_NVM_POLL_WRITE 1 /* Flag for polling write complete */
+#define E1000_NVM_POLL_READ 0 /* Flag for polling read complete */
+#define E1000_FLASH_UPDATES 2000
/* NVM Word Offsets */
#define NVM_COMPAT 0x0003
diff --git a/drivers/net/ethernet/intel/e1000e/e1000.h b/drivers/net/ethernet/intel/e1000e/e1000.h
index fcc7581..e371f77 100644
--- a/drivers/net/ethernet/intel/e1000e/e1000.h
+++ b/drivers/net/ethernet/intel/e1000e/e1000.h
@@ -61,7 +61,6 @@
#define e_notice(format, arg...) \
netdev_notice(adapter->netdev, format, ## arg)
-
/* Interrupt modes, as used by the IntMode parameter */
#define E1000E_INT_MODE_LEGACY 0
#define E1000E_INT_MODE_MSI 1
@@ -239,9 +238,8 @@
u16 tx_itr;
u16 rx_itr;
- /* Tx */
- struct e1000_ring *tx_ring /* One per active queue */
- ____cacheline_aligned_in_smp;
+ /* Tx - one ring per active queue */
+ struct e1000_ring *tx_ring ____cacheline_aligned_in_smp;
u32 tx_fifo_limit;
struct napi_struct napi;
@@ -487,8 +485,8 @@
extern void e1000e_free_rx_resources(struct e1000_ring *ring);
extern void e1000e_free_tx_resources(struct e1000_ring *ring);
extern struct rtnl_link_stats64 *e1000e_get_stats64(struct net_device *netdev,
- struct rtnl_link_stats64
- *stats);
+ struct rtnl_link_stats64
+ *stats);
extern void e1000e_set_interrupt_capability(struct e1000_adapter *adapter);
extern void e1000e_reset_interrupt_capability(struct e1000_adapter *adapter);
extern void e1000e_get_hw_control(struct e1000_adapter *adapter);
@@ -558,12 +556,14 @@
return hw->nvm.ops.update(hw);
}
-static inline s32 e1000_read_nvm(struct e1000_hw *hw, u16 offset, u16 words, u16 *data)
+static inline s32 e1000_read_nvm(struct e1000_hw *hw, u16 offset, u16 words,
+ u16 *data)
{
return hw->nvm.ops.read(hw, offset, words, data);
}
-static inline s32 e1000_write_nvm(struct e1000_hw *hw, u16 offset, u16 words, u16 *data)
+static inline s32 e1000_write_nvm(struct e1000_hw *hw, u16 offset, u16 words,
+ u16 *data)
{
return hw->nvm.ops.write(hw, offset, words, data);
}
@@ -597,7 +597,7 @@
s32 i = E1000_ICH_FWSM_PCIM2PCI_COUNT;
while ((er32(FWSM) & E1000_ICH_FWSM_PCIM2PCI) && --i)
- udelay(50);
+ usleep_range(50, 100);
return i;
}
diff --git a/drivers/net/ethernet/intel/e1000e/ethtool.c b/drivers/net/ethernet/intel/e1000e/ethtool.c
index f91a8f3..e835e7b 100644
--- a/drivers/net/ethernet/intel/e1000e/ethtool.c
+++ b/drivers/net/ethernet/intel/e1000e/ethtool.c
@@ -40,7 +40,7 @@
#include "e1000.h"
-enum {NETDEV_STATS, E1000_STATS};
+enum { NETDEV_STATS, E1000_STATS };
struct e1000_stats {
char stat_string[ETH_GSTRING_LEN];
@@ -121,6 +121,7 @@
"Interrupt test (offline)", "Loopback test (offline)",
"Link test (on/offline)"
};
+
#define E1000_TEST_LEN ARRAY_SIZE(e1000_gstrings_test)
static int e1000_get_settings(struct net_device *netdev,
@@ -197,8 +198,7 @@
/* MDI-X => 2; MDI =>1; Invalid =>0 */
if ((hw->phy.media_type == e1000_media_type_copper) &&
netif_carrier_ok(netdev))
- ecmd->eth_tp_mdix = hw->phy.is_mdix ? ETH_TP_MDI_X :
- ETH_TP_MDI;
+ ecmd->eth_tp_mdix = hw->phy.is_mdix ? ETH_TP_MDI_X : ETH_TP_MDI;
else
ecmd->eth_tp_mdix = ETH_TP_MDI_INVALID;
@@ -224,8 +224,7 @@
/* Fiber NICs only allow 1000 gbps Full duplex */
if ((adapter->hw.phy.media_type == e1000_media_type_fiber) &&
- spd != SPEED_1000 &&
- dplx != DUPLEX_FULL) {
+ (spd != SPEED_1000) && (dplx != DUPLEX_FULL)) {
goto err_inval;
}
@@ -298,12 +297,10 @@
hw->mac.autoneg = 1;
if (hw->phy.media_type == e1000_media_type_fiber)
hw->phy.autoneg_advertised = ADVERTISED_1000baseT_Full |
- ADVERTISED_FIBRE |
- ADVERTISED_Autoneg;
+ ADVERTISED_FIBRE | ADVERTISED_Autoneg;
else
hw->phy.autoneg_advertised = ecmd->advertising |
- ADVERTISED_TP |
- ADVERTISED_Autoneg;
+ ADVERTISED_TP | ADVERTISED_Autoneg;
ecmd->advertising = hw->phy.autoneg_advertised;
if (adapter->fc_autoneg)
hw->fc.requested_mode = e1000_fc_default;
@@ -346,7 +343,7 @@
struct e1000_hw *hw = &adapter->hw;
pause->autoneg =
- (adapter->fc_autoneg ? AUTONEG_ENABLE : AUTONEG_DISABLE);
+ (adapter->fc_autoneg ? AUTONEG_ENABLE : AUTONEG_DISABLE);
if (hw->fc.current_mode == e1000_fc_rx_pause) {
pause->rx_pause = 1;
@@ -435,7 +432,7 @@
memset(p, 0, E1000_REGS_LEN * sizeof(u32));
regs->version = (1 << 24) | (adapter->pdev->revision << 16) |
- adapter->pdev->device;
+ adapter->pdev->device;
regs_buff[0] = er32(CTRL);
regs_buff[1] = er32(STATUS);
@@ -503,8 +500,8 @@
first_word = eeprom->offset >> 1;
last_word = (eeprom->offset + eeprom->len - 1) >> 1;
- eeprom_buff = kmalloc(sizeof(u16) *
- (last_word - first_word + 1), GFP_KERNEL);
+ eeprom_buff = kmalloc(sizeof(u16) * (last_word - first_word + 1),
+ GFP_KERNEL);
if (!eeprom_buff)
return -ENOMEM;
@@ -515,7 +512,7 @@
} else {
for (i = 0; i < last_word - first_word + 1; i++) {
ret_val = e1000_read_nvm(hw, first_word + i, 1,
- &eeprom_buff[i]);
+ &eeprom_buff[i]);
if (ret_val)
break;
}
@@ -553,7 +550,8 @@
if (eeprom->len == 0)
return -EOPNOTSUPP;
- if (eeprom->magic != (adapter->pdev->vendor | (adapter->pdev->device << 16)))
+ if (eeprom->magic !=
+ (adapter->pdev->vendor | (adapter->pdev->device << 16)))
return -EFAULT;
if (adapter->flags & FLAG_READ_ONLY_NVM)
@@ -579,7 +577,7 @@
/* need read/modify/write of last changed EEPROM word */
/* only the first byte of the word is being modified */
ret_val = e1000_read_nvm(hw, last_word, 1,
- &eeprom_buff[last_word - first_word]);
+ &eeprom_buff[last_word - first_word]);
if (ret_val)
goto out;
@@ -618,8 +616,7 @@
{
struct e1000_adapter *adapter = netdev_priv(netdev);
- strlcpy(drvinfo->driver, e1000e_driver_name,
- sizeof(drvinfo->driver));
+ strlcpy(drvinfo->driver, e1000e_driver_name, sizeof(drvinfo->driver));
strlcpy(drvinfo->version, e1000e_driver_version,
sizeof(drvinfo->version));
@@ -627,10 +624,10 @@
* PCI-E controllers
*/
snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version),
- "%d.%d-%d",
- (adapter->eeprom_vers & 0xF000) >> 12,
- (adapter->eeprom_vers & 0x0FF0) >> 4,
- (adapter->eeprom_vers & 0x000F));
+ "%d.%d-%d",
+ (adapter->eeprom_vers & 0xF000) >> 12,
+ (adapter->eeprom_vers & 0x0FF0) >> 4,
+ (adapter->eeprom_vers & 0x000F));
strlcpy(drvinfo->bus_info, pci_name(adapter->pdev),
sizeof(drvinfo->bus_info));
@@ -756,7 +753,8 @@
{
u32 pat, val;
static const u32 test[] = {
- 0x5A5A5A5A, 0xA5A5A5A5, 0x00000000, 0xFFFFFFFF};
+ 0x5A5A5A5A, 0xA5A5A5A5, 0x00000000, 0xFFFFFFFF
+ };
for (pat = 0; pat < ARRAY_SIZE(test); pat++) {
E1000_WRITE_REG_ARRAY(&adapter->hw, reg, offset,
(test[pat] & write));
@@ -786,6 +784,7 @@
}
return 0;
}
+
#define REG_PATTERN_TEST_ARRAY(reg, offset, mask, write) \
do { \
if (reg_pattern_test(adapter, data, reg, offset, mask, write)) \
@@ -813,16 +812,16 @@
u32 wlock_mac = 0;
/* The status register is Read Only, so a write should fail.
- * Some bits that get toggled are ignored.
+ * Some bits that get toggled are ignored. There are several bits
+ * on newer hardware that are r/w.
*/
switch (mac->type) {
- /* there are several bits on newer hardware that are r/w */
case e1000_82571:
case e1000_82572:
case e1000_80003es2lan:
toggle = 0x7FFFF3FF;
break;
- default:
+ default:
toggle = 0x7FFFF033;
break;
}
@@ -928,7 +927,7 @@
}
/* If Checksum is not Correct return error else test passed */
- if ((checksum != (u16) NVM_SUM) && !(*data))
+ if ((checksum != (u16)NVM_SUM) && !(*data))
*data = 2;
return *data;
@@ -936,7 +935,7 @@
static irqreturn_t e1000_test_intr(int __always_unused irq, void *data)
{
- struct net_device *netdev = (struct net_device *) data;
+ struct net_device *netdev = (struct net_device *)data;
struct e1000_adapter *adapter = netdev_priv(netdev);
struct e1000_hw *hw = &adapter->hw;
@@ -969,8 +968,8 @@
if (!request_irq(irq, e1000_test_intr, IRQF_PROBE_SHARED, netdev->name,
netdev)) {
shared_int = 0;
- } else if (request_irq(irq, e1000_test_intr, IRQF_SHARED,
- netdev->name, netdev)) {
+ } else if (request_irq(irq, e1000_test_intr, IRQF_SHARED, netdev->name,
+ netdev)) {
*data = 1;
ret_val = -1;
goto out;
@@ -1080,28 +1079,33 @@
struct e1000_ring *tx_ring = &adapter->test_tx_ring;
struct e1000_ring *rx_ring = &adapter->test_rx_ring;
struct pci_dev *pdev = adapter->pdev;
+ struct e1000_buffer *buffer_info;
int i;
if (tx_ring->desc && tx_ring->buffer_info) {
for (i = 0; i < tx_ring->count; i++) {
- if (tx_ring->buffer_info[i].dma)
+ buffer_info = &tx_ring->buffer_info[i];
+
+ if (buffer_info->dma)
dma_unmap_single(&pdev->dev,
- tx_ring->buffer_info[i].dma,
- tx_ring->buffer_info[i].length,
- DMA_TO_DEVICE);
- if (tx_ring->buffer_info[i].skb)
- dev_kfree_skb(tx_ring->buffer_info[i].skb);
+ buffer_info->dma,
+ buffer_info->length,
+ DMA_TO_DEVICE);
+ if (buffer_info->skb)
+ dev_kfree_skb(buffer_info->skb);
}
}
if (rx_ring->desc && rx_ring->buffer_info) {
for (i = 0; i < rx_ring->count; i++) {
- if (rx_ring->buffer_info[i].dma)
+ buffer_info = &rx_ring->buffer_info[i];
+
+ if (buffer_info->dma)
dma_unmap_single(&pdev->dev,
- rx_ring->buffer_info[i].dma,
- 2048, DMA_FROM_DEVICE);
- if (rx_ring->buffer_info[i].skb)
- dev_kfree_skb(rx_ring->buffer_info[i].skb);
+ buffer_info->dma,
+ 2048, DMA_FROM_DEVICE);
+ if (buffer_info->skb)
+ dev_kfree_skb(buffer_info->skb);
}
}
@@ -1138,8 +1142,7 @@
tx_ring->count = E1000_DEFAULT_TXD;
tx_ring->buffer_info = kcalloc(tx_ring->count,
- sizeof(struct e1000_buffer),
- GFP_KERNEL);
+ sizeof(struct e1000_buffer), GFP_KERNEL);
if (!tx_ring->buffer_info) {
ret_val = 1;
goto err_nomem;
@@ -1156,8 +1159,8 @@
tx_ring->next_to_use = 0;
tx_ring->next_to_clean = 0;
- ew32(TDBAL(0), ((u64) tx_ring->dma & 0x00000000FFFFFFFF));
- ew32(TDBAH(0), ((u64) tx_ring->dma >> 32));
+ ew32(TDBAL(0), ((u64)tx_ring->dma & 0x00000000FFFFFFFF));
+ ew32(TDBAH(0), ((u64)tx_ring->dma >> 32));
ew32(TDLEN(0), tx_ring->count * sizeof(struct e1000_tx_desc));
ew32(TDH(0), 0);
ew32(TDT(0), 0);
@@ -1179,8 +1182,8 @@
tx_ring->buffer_info[i].skb = skb;
tx_ring->buffer_info[i].length = skb->len;
tx_ring->buffer_info[i].dma =
- dma_map_single(&pdev->dev, skb->data, skb->len,
- DMA_TO_DEVICE);
+ dma_map_single(&pdev->dev, skb->data, skb->len,
+ DMA_TO_DEVICE);
if (dma_mapping_error(&pdev->dev,
tx_ring->buffer_info[i].dma)) {
ret_val = 4;
@@ -1200,8 +1203,7 @@
rx_ring->count = E1000_DEFAULT_RXD;
rx_ring->buffer_info = kcalloc(rx_ring->count,
- sizeof(struct e1000_buffer),
- GFP_KERNEL);
+ sizeof(struct e1000_buffer), GFP_KERNEL);
if (!rx_ring->buffer_info) {
ret_val = 5;
goto err_nomem;
@@ -1220,16 +1222,16 @@
rctl = er32(RCTL);
if (!(adapter->flags2 & FLAG2_NO_DISABLE_RX))
ew32(RCTL, rctl & ~E1000_RCTL_EN);
- ew32(RDBAL(0), ((u64) rx_ring->dma & 0xFFFFFFFF));
- ew32(RDBAH(0), ((u64) rx_ring->dma >> 32));
+ ew32(RDBAL(0), ((u64)rx_ring->dma & 0xFFFFFFFF));
+ ew32(RDBAH(0), ((u64)rx_ring->dma >> 32));
ew32(RDLEN(0), rx_ring->size);
ew32(RDH(0), 0);
ew32(RDT(0), 0);
rctl = E1000_RCTL_EN | E1000_RCTL_BAM | E1000_RCTL_SZ_2048 |
- E1000_RCTL_UPE | E1000_RCTL_MPE | E1000_RCTL_LPE |
- E1000_RCTL_SBP | E1000_RCTL_SECRC |
- E1000_RCTL_LBM_NO | E1000_RCTL_RDMTS_HALF |
- (adapter->hw.mac.mc_filter_type << E1000_RCTL_MO_SHIFT);
+ E1000_RCTL_UPE | E1000_RCTL_MPE | E1000_RCTL_LPE |
+ E1000_RCTL_SBP | E1000_RCTL_SECRC |
+ E1000_RCTL_LBM_NO | E1000_RCTL_RDMTS_HALF |
+ (adapter->hw.mac.mc_filter_type << E1000_RCTL_MO_SHIFT);
ew32(RCTL, rctl);
for (i = 0; i < rx_ring->count; i++) {
@@ -1244,8 +1246,8 @@
skb_reserve(skb, NET_IP_ALIGN);
rx_ring->buffer_info[i].skb = skb;
rx_ring->buffer_info[i].dma =
- dma_map_single(&pdev->dev, skb->data, 2048,
- DMA_FROM_DEVICE);
+ dma_map_single(&pdev->dev, skb->data, 2048,
+ DMA_FROM_DEVICE);
if (dma_mapping_error(&pdev->dev,
rx_ring->buffer_info[i].dma)) {
ret_val = 8;
@@ -1296,7 +1298,7 @@
ew32(CTRL, ctrl_reg);
e1e_flush();
- udelay(500);
+ usleep_range(500, 1000);
return 0;
}
@@ -1322,7 +1324,7 @@
e1e_wphy(hw, PHY_REG(2, 21), phy_reg);
/* Assert SW reset for above settings to take effect */
hw->phy.ops.commit(hw);
- mdelay(1);
+ usleep_range(1000, 2000);
/* Force Full Duplex */
e1e_rphy(hw, PHY_REG(769, 16), &phy_reg);
e1e_wphy(hw, PHY_REG(769, 16), phy_reg | 0x000C);
@@ -1363,7 +1365,7 @@
/* force 1000, set loopback */
e1e_wphy(hw, MII_BMCR, 0x4140);
- mdelay(250);
+ msleep(250);
/* Now set up the MAC to the same speed/duplex as the PHY. */
ctrl_reg = er32(CTRL);
@@ -1395,7 +1397,7 @@
if (hw->phy.type == e1000_phy_m88)
e1000_phy_disable_receiver(adapter);
- udelay(500);
+ usleep_range(500, 1000);
return 0;
}
@@ -1431,8 +1433,7 @@
/* special write to serdes control register to enable SerDes analog
* loopback
*/
-#define E1000_SERDES_LB_ON 0x410
- ew32(SCTL, E1000_SERDES_LB_ON);
+ ew32(SCTL, E1000_SCTL_ENABLE_SERDES_LOOPBACK);
e1e_flush();
usleep_range(10000, 20000);
@@ -1526,8 +1527,7 @@
case e1000_82572:
if (hw->phy.media_type == e1000_media_type_fiber ||
hw->phy.media_type == e1000_media_type_internal_serdes) {
-#define E1000_SERDES_LB_OFF 0x400
- ew32(SCTL, E1000_SERDES_LB_OFF);
+ ew32(SCTL, E1000_SCTL_DISABLE_SERDES_LOOPBACK);
e1e_flush();
usleep_range(10000, 20000);
break;
@@ -1564,7 +1564,7 @@
frame_size &= ~1;
if (*(skb->data + 3) == 0xFF)
if ((*(skb->data + frame_size / 2 + 10) == 0xBE) &&
- (*(skb->data + frame_size / 2 + 12) == 0xAF))
+ (*(skb->data + frame_size / 2 + 12) == 0xAF))
return 0;
return 13;
}
@@ -1575,6 +1575,7 @@
struct e1000_ring *rx_ring = &adapter->test_rx_ring;
struct pci_dev *pdev = adapter->pdev;
struct e1000_hw *hw = &adapter->hw;
+ struct e1000_buffer *buffer_info;
int i, j, k, l;
int lc;
int good_cnt;
@@ -1595,14 +1596,17 @@
k = 0;
l = 0;
- for (j = 0; j <= lc; j++) { /* loop count loop */
- for (i = 0; i < 64; i++) { /* send the packets */
- e1000_create_lbtest_frame(tx_ring->buffer_info[k].skb,
- 1024);
+ /* loop count loop */
+ for (j = 0; j <= lc; j++) {
+ /* send the packets */
+ for (i = 0; i < 64; i++) {
+ buffer_info = &tx_ring->buffer_info[k];
+
+ e1000_create_lbtest_frame(buffer_info->skb, 1024);
dma_sync_single_for_device(&pdev->dev,
- tx_ring->buffer_info[k].dma,
- tx_ring->buffer_info[k].length,
- DMA_TO_DEVICE);
+ buffer_info->dma,
+ buffer_info->length,
+ DMA_TO_DEVICE);
k++;
if (k == tx_ring->count)
k = 0;
@@ -1612,13 +1616,16 @@
msleep(200);
time = jiffies; /* set the start time for the receive */
good_cnt = 0;
- do { /* receive the sent packets */
- dma_sync_single_for_cpu(&pdev->dev,
- rx_ring->buffer_info[l].dma, 2048,
- DMA_FROM_DEVICE);
+ /* receive the sent packets */
+ do {
+ buffer_info = &rx_ring->buffer_info[l];
- ret_val = e1000_check_lbtest_frame(
- rx_ring->buffer_info[l].skb, 1024);
+ dma_sync_single_for_cpu(&pdev->dev,
+ buffer_info->dma, 2048,
+ DMA_FROM_DEVICE);
+
+ ret_val = e1000_check_lbtest_frame(buffer_info->skb,
+ 1024);
if (!ret_val)
good_cnt++;
l++;
@@ -1637,7 +1644,7 @@
ret_val = 14; /* error code for time out error */
break;
}
- } /* end loop count loop */
+ }
return ret_val;
}
@@ -1696,7 +1703,7 @@
/* On some Phy/switch combinations, link establishment
* can take a few seconds more than expected.
*/
- msleep(5000);
+ msleep_interruptible(5000);
if (!(er32(STATUS) & E1000_STATUS_LU))
*data = 1;
@@ -1980,12 +1987,12 @@
for (i = 0; i < E1000_GLOBAL_STATS_LEN; i++) {
switch (e1000_gstrings_stats[i].type) {
case NETDEV_STATS:
- p = (char *) &net_stats +
- e1000_gstrings_stats[i].stat_offset;
+ p = (char *)&net_stats +
+ e1000_gstrings_stats[i].stat_offset;
break;
case E1000_STATS:
- p = (char *) adapter +
- e1000_gstrings_stats[i].stat_offset;
+ p = (char *)adapter +
+ e1000_gstrings_stats[i].stat_offset;
break;
default:
data[i] = 0;
@@ -1993,7 +2000,7 @@
}
data[i] = (e1000_gstrings_stats[i].sizeof_stat ==
- sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
+ sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
}
}
diff --git a/drivers/net/ethernet/intel/e1000e/hw.h b/drivers/net/ethernet/intel/e1000e/hw.h
index 1e6b889..84850f7 100644
--- a/drivers/net/ethernet/intel/e1000e/hw.h
+++ b/drivers/net/ethernet/intel/e1000e/hw.h
@@ -167,7 +167,7 @@
e1000_1000t_rx_status_undefined = 0xFF
};
-enum e1000_rev_polarity{
+enum e1000_rev_polarity {
e1000_rev_polarity_normal = 0,
e1000_rev_polarity_reversed,
e1000_rev_polarity_undefined = 0xFF
@@ -545,7 +545,7 @@
u16 mta_reg_count;
/* Maximum size of the MTA register table in all supported adapters */
- #define MAX_MTA_REG 128
+#define MAX_MTA_REG 128
u32 mta_shadow[MAX_MTA_REG];
u16 rar_entry_count;
diff --git a/drivers/net/ethernet/intel/e1000e/ich8lan.c b/drivers/net/ethernet/intel/e1000e/ich8lan.c
index 121a865..1cdec5f 100644
--- a/drivers/net/ethernet/intel/e1000e/ich8lan.c
+++ b/drivers/net/ethernet/intel/e1000e/ich8lan.c
@@ -61,15 +61,15 @@
/* Offset 04h HSFSTS */
union ich8_hws_flash_status {
struct ich8_hsfsts {
- u16 flcdone :1; /* bit 0 Flash Cycle Done */
- u16 flcerr :1; /* bit 1 Flash Cycle Error */
- u16 dael :1; /* bit 2 Direct Access error Log */
- u16 berasesz :2; /* bit 4:3 Sector Erase Size */
- u16 flcinprog :1; /* bit 5 flash cycle in Progress */
- u16 reserved1 :2; /* bit 13:6 Reserved */
- u16 reserved2 :6; /* bit 13:6 Reserved */
- u16 fldesvalid :1; /* bit 14 Flash Descriptor Valid */
- u16 flockdn :1; /* bit 15 Flash Config Lock-Down */
+ u16 flcdone:1; /* bit 0 Flash Cycle Done */
+ u16 flcerr:1; /* bit 1 Flash Cycle Error */
+ u16 dael:1; /* bit 2 Direct Access error Log */
+ u16 berasesz:2; /* bit 4:3 Sector Erase Size */
+ u16 flcinprog:1; /* bit 5 flash cycle in Progress */
+ u16 reserved1:2; /* bit 13:6 Reserved */
+ u16 reserved2:6; /* bit 13:6 Reserved */
+ u16 fldesvalid:1; /* bit 14 Flash Descriptor Valid */
+ u16 flockdn:1; /* bit 15 Flash Config Lock-Down */
} hsf_status;
u16 regval;
};
@@ -78,11 +78,11 @@
/* Offset 06h FLCTL */
union ich8_hws_flash_ctrl {
struct ich8_hsflctl {
- u16 flcgo :1; /* 0 Flash Cycle Go */
- u16 flcycle :2; /* 2:1 Flash Cycle */
- u16 reserved :5; /* 7:3 Reserved */
- u16 fldbcount :2; /* 9:8 Flash Data Byte Count */
- u16 flockdn :6; /* 15:10 Reserved */
+ u16 flcgo:1; /* 0 Flash Cycle Go */
+ u16 flcycle:2; /* 2:1 Flash Cycle */
+ u16 reserved:5; /* 7:3 Reserved */
+ u16 fldbcount:2; /* 9:8 Flash Data Byte Count */
+ u16 flockdn:6; /* 15:10 Reserved */
} hsf_ctrl;
u16 regval;
};
@@ -90,10 +90,10 @@
/* ICH Flash Region Access Permissions */
union ich8_hws_flash_regacc {
struct ich8_flracc {
- u32 grra :8; /* 0:7 GbE region Read Access */
- u32 grwa :8; /* 8:15 GbE region Write Access */
- u32 gmrag :8; /* 23:16 GbE Master Read Access Grant */
- u32 gmwag :8; /* 31:24 GbE Master Write Access Grant */
+ u32 grra:8; /* 0:7 GbE region Read Access */
+ u32 grwa:8; /* 8:15 GbE region Write Access */
+ u32 gmrag:8; /* 23:16 GbE Master Read Access Grant */
+ u32 gmwag:8; /* 31:24 GbE Master Write Access Grant */
} hsf_flregacc;
u16 regval;
};
@@ -312,7 +312,7 @@
mac_reg &= ~E1000_CTRL_LANPHYPC_VALUE;
ew32(CTRL, mac_reg);
e1e_flush();
- udelay(10);
+ usleep_range(10, 20);
mac_reg &= ~E1000_CTRL_LANPHYPC_OVERRIDE;
ew32(CTRL, mac_reg);
e1e_flush();
@@ -548,8 +548,8 @@
/* find total size of the NVM, then cut in half since the total
* size represents two separate NVM banks.
*/
- nvm->flash_bank_size = (sector_end_addr - sector_base_addr)
- << FLASH_SECTOR_ADDR_SHIFT;
+ nvm->flash_bank_size = ((sector_end_addr - sector_base_addr)
+ << FLASH_SECTOR_ADDR_SHIFT);
nvm->flash_bank_size /= 2;
/* Adjust to word count */
nvm->flash_bank_size /= sizeof(u16);
@@ -1134,9 +1134,9 @@
u32 fwsm;
fwsm = er32(FWSM);
- return (fwsm & E1000_ICH_FWSM_FW_VALID) &&
- ((fwsm & E1000_FWSM_MODE_MASK) ==
- (E1000_ICH_MNG_IAMT_MODE << E1000_FWSM_MODE_SHIFT));
+ return ((fwsm & E1000_ICH_FWSM_FW_VALID) &&
+ ((fwsm & E1000_FWSM_MODE_MASK) ==
+ (E1000_ICH_MNG_IAMT_MODE << E1000_FWSM_MODE_SHIFT)));
}
/**
@@ -1153,7 +1153,7 @@
fwsm = er32(FWSM);
return (fwsm & E1000_ICH_FWSM_FW_VALID) &&
- (fwsm & (E1000_ICH_MNG_IAMT_MODE << E1000_FWSM_MODE_SHIFT));
+ (fwsm & (E1000_ICH_MNG_IAMT_MODE << E1000_FWSM_MODE_SHIFT));
}
/**
@@ -1440,8 +1440,7 @@
word_addr = (u16)(cnf_base_addr << 1);
for (i = 0; i < cnf_size; i++) {
- ret_val = e1000_read_nvm(hw, (word_addr + i * 2), 1,
- ®_data);
+ ret_val = e1000_read_nvm(hw, (word_addr + i * 2), 1, ®_data);
if (ret_val)
goto release;
@@ -1501,13 +1500,13 @@
if (ret_val)
goto release;
- status_reg &= BM_CS_STATUS_LINK_UP |
- BM_CS_STATUS_RESOLVED |
- BM_CS_STATUS_SPEED_MASK;
+ status_reg &= (BM_CS_STATUS_LINK_UP |
+ BM_CS_STATUS_RESOLVED |
+ BM_CS_STATUS_SPEED_MASK);
if (status_reg == (BM_CS_STATUS_LINK_UP |
- BM_CS_STATUS_RESOLVED |
- BM_CS_STATUS_SPEED_1000))
+ BM_CS_STATUS_RESOLVED |
+ BM_CS_STATUS_SPEED_1000))
k1_enable = false;
}
@@ -1516,13 +1515,13 @@
if (ret_val)
goto release;
- status_reg &= HV_M_STATUS_LINK_UP |
- HV_M_STATUS_AUTONEG_COMPLETE |
- HV_M_STATUS_SPEED_MASK;
+ status_reg &= (HV_M_STATUS_LINK_UP |
+ HV_M_STATUS_AUTONEG_COMPLETE |
+ HV_M_STATUS_SPEED_MASK);
if (status_reg == (HV_M_STATUS_LINK_UP |
- HV_M_STATUS_AUTONEG_COMPLETE |
- HV_M_STATUS_SPEED_1000))
+ HV_M_STATUS_AUTONEG_COMPLETE |
+ HV_M_STATUS_SPEED_1000))
k1_enable = false;
}
@@ -1579,7 +1578,7 @@
if (ret_val)
return ret_val;
- udelay(20);
+ usleep_range(20, 40);
ctrl_ext = er32(CTRL_EXT);
ctrl_reg = er32(CTRL);
@@ -1589,11 +1588,11 @@
ew32(CTRL_EXT, ctrl_ext | E1000_CTRL_EXT_SPD_BYPS);
e1e_flush();
- udelay(20);
+ usleep_range(20, 40);
ew32(CTRL, ctrl_reg);
ew32(CTRL_EXT, ctrl_ext);
e1e_flush();
- udelay(20);
+ usleep_range(20, 40);
return 0;
}
@@ -1667,7 +1666,6 @@
return ret_val;
}
-
/**
* e1000_set_mdio_slow_mode_hv - Set slow MDIO access mode
* @hw: pointer to the HW structure
@@ -1834,7 +1832,7 @@
* SHRAL/H) and initial CRC values to the MAC
*/
for (i = 0; i < (hw->mac.rar_entry_count + 4); i++) {
- u8 mac_addr[ETH_ALEN] = {0};
+ u8 mac_addr[ETH_ALEN] = { 0 };
u32 addr_high, addr_low;
addr_high = er32(RAH(i));
@@ -1865,8 +1863,8 @@
ew32(RCTL, mac_reg);
ret_val = e1000e_read_kmrn_reg(hw,
- E1000_KMRNCTRLSTA_CTRL_OFFSET,
- &data);
+ E1000_KMRNCTRLSTA_CTRL_OFFSET,
+ &data);
if (ret_val)
return ret_val;
ret_val = e1000e_write_kmrn_reg(hw,
@@ -1875,8 +1873,8 @@
if (ret_val)
return ret_val;
ret_val = e1000e_read_kmrn_reg(hw,
- E1000_KMRNCTRLSTA_HD_CTRL,
- &data);
+ E1000_KMRNCTRLSTA_HD_CTRL,
+ &data);
if (ret_val)
return ret_val;
data &= ~(0xF << 8);
@@ -1923,8 +1921,8 @@
ew32(RCTL, mac_reg);
ret_val = e1000e_read_kmrn_reg(hw,
- E1000_KMRNCTRLSTA_CTRL_OFFSET,
- &data);
+ E1000_KMRNCTRLSTA_CTRL_OFFSET,
+ &data);
if (ret_val)
return ret_val;
ret_val = e1000e_write_kmrn_reg(hw,
@@ -1933,8 +1931,8 @@
if (ret_val)
return ret_val;
ret_val = e1000e_read_kmrn_reg(hw,
- E1000_KMRNCTRLSTA_HD_CTRL,
- &data);
+ E1000_KMRNCTRLSTA_HD_CTRL,
+ &data);
if (ret_val)
return ret_val;
data &= ~(0xF << 8);
@@ -2100,7 +2098,7 @@
do {
data = er32(STATUS);
data &= E1000_STATUS_LAN_INIT_DONE;
- udelay(100);
+ usleep_range(100, 200);
} while ((!data) && --loop);
/* If basic configuration is incomplete before the above loop
@@ -2445,7 +2443,7 @@
/* Check bank 0 */
ret_val = e1000_read_flash_byte_ich8lan(hw, act_offset,
- &sig_byte);
+ &sig_byte);
if (ret_val)
return ret_val;
if ((sig_byte & E1000_ICH_NVM_VALID_SIG_MASK) ==
@@ -2456,8 +2454,8 @@
/* Check bank 1 */
ret_val = e1000_read_flash_byte_ich8lan(hw, act_offset +
- bank1_offset,
- &sig_byte);
+ bank1_offset,
+ &sig_byte);
if (ret_val)
return ret_val;
if ((sig_byte & E1000_ICH_NVM_VALID_SIG_MASK) ==
@@ -2510,8 +2508,8 @@
ret_val = 0;
for (i = 0; i < words; i++) {
- if (dev_spec->shadow_ram[offset+i].modified) {
- data[i] = dev_spec->shadow_ram[offset+i].value;
+ if (dev_spec->shadow_ram[offset + i].modified) {
+ data[i] = dev_spec->shadow_ram[offset + i].value;
} else {
ret_val = e1000_read_flash_word_ich8lan(hw,
act_offset + i,
@@ -2696,8 +2694,8 @@
if (size < 1 || size > 2 || offset > ICH_FLASH_LINEAR_ADDR_MASK)
return -E1000_ERR_NVM;
- flash_linear_addr = (ICH_FLASH_LINEAR_ADDR_MASK & offset) +
- hw->nvm.flash_base_addr;
+ flash_linear_addr = ((ICH_FLASH_LINEAR_ADDR_MASK & offset) +
+ hw->nvm.flash_base_addr);
do {
udelay(1);
@@ -2714,8 +2712,9 @@
ew32flash(ICH_FLASH_FADDR, flash_linear_addr);
- ret_val = e1000_flash_cycle_ich8lan(hw,
- ICH_FLASH_READ_COMMAND_TIMEOUT);
+ ret_val =
+ e1000_flash_cycle_ich8lan(hw,
+ ICH_FLASH_READ_COMMAND_TIMEOUT);
/* Check if FCERR is set to 1, if set to 1, clear it
* and try the whole sequence a few more times, else
@@ -2774,8 +2773,8 @@
nvm->ops.acquire(hw);
for (i = 0; i < words; i++) {
- dev_spec->shadow_ram[offset+i].modified = true;
- dev_spec->shadow_ram[offset+i].value = data[i];
+ dev_spec->shadow_ram[offset + i].modified = true;
+ dev_spec->shadow_ram[offset + i].value = data[i];
}
nvm->ops.release(hw);
@@ -2844,8 +2843,8 @@
data = dev_spec->shadow_ram[i].value;
} else {
ret_val = e1000_read_flash_word_ich8lan(hw, i +
- old_bank_offset,
- &data);
+ old_bank_offset,
+ &data);
if (ret_val)
break;
}
@@ -2863,7 +2862,7 @@
/* Convert offset to bytes. */
act_offset = (i + new_bank_offset) << 1;
- udelay(100);
+ usleep_range(100, 200);
/* Write the bytes to the new bank. */
ret_val = e1000_retry_write_flash_byte_ich8lan(hw,
act_offset,
@@ -2871,10 +2870,10 @@
if (ret_val)
break;
- udelay(100);
+ usleep_range(100, 200);
ret_val = e1000_retry_write_flash_byte_ich8lan(hw,
- act_offset + 1,
- (u8)(data >> 8));
+ act_offset + 1,
+ (u8)(data >> 8));
if (ret_val)
break;
}
@@ -3050,8 +3049,8 @@
offset > ICH_FLASH_LINEAR_ADDR_MASK)
return -E1000_ERR_NVM;
- flash_linear_addr = (ICH_FLASH_LINEAR_ADDR_MASK & offset) +
- hw->nvm.flash_base_addr;
+ flash_linear_addr = ((ICH_FLASH_LINEAR_ADDR_MASK & offset) +
+ hw->nvm.flash_base_addr);
do {
udelay(1);
@@ -3062,7 +3061,7 @@
hsflctl.regval = er16flash(ICH_FLASH_HSFCTL);
/* 0b/1b corresponds to 1 or 2 byte size, respectively. */
- hsflctl.hsf_ctrl.fldbcount = size -1;
+ hsflctl.hsf_ctrl.fldbcount = size - 1;
hsflctl.hsf_ctrl.flcycle = ICH_CYCLE_WRITE;
ew16flash(ICH_FLASH_HSFCTL, hsflctl.regval);
@@ -3078,8 +3077,9 @@
/* check if FCERR is set to 1 , if set to 1, clear it
* and try the whole sequence a few more times else done
*/
- ret_val = e1000_flash_cycle_ich8lan(hw,
- ICH_FLASH_WRITE_COMMAND_TIMEOUT);
+ ret_val =
+ e1000_flash_cycle_ich8lan(hw,
+ ICH_FLASH_WRITE_COMMAND_TIMEOUT);
if (!ret_val)
break;
@@ -3138,7 +3138,7 @@
for (program_retries = 0; program_retries < 100; program_retries++) {
e_dbg("Retrying Byte %2.2X at offset %u\n", byte, offset);
- udelay(100);
+ usleep_range(100, 200);
ret_val = e1000_write_flash_byte_ich8lan(hw, offset, byte);
if (!ret_val)
break;
@@ -3209,8 +3209,10 @@
flash_linear_addr = hw->nvm.flash_base_addr;
flash_linear_addr += (bank) ? flash_bank_size : 0;
- for (j = 0; j < iteration ; j++) {
+ for (j = 0; j < iteration; j++) {
do {
+ u32 timeout = ICH_FLASH_ERASE_COMMAND_TIMEOUT;
+
/* Steps */
ret_val = e1000_flash_cycle_init_ich8lan(hw);
if (ret_val)
@@ -3230,8 +3232,7 @@
flash_linear_addr += (j * sector_size);
ew32flash(ICH_FLASH_FADDR, flash_linear_addr);
- ret_val = e1000_flash_cycle_ich8lan(hw,
- ICH_FLASH_ERASE_COMMAND_TIMEOUT);
+ ret_val = e1000_flash_cycle_ich8lan(hw, timeout);
if (!ret_val)
break;
@@ -3270,8 +3271,7 @@
return ret_val;
}
- if (*data == ID_LED_RESERVED_0000 ||
- *data == ID_LED_RESERVED_FFFF)
+ if (*data == ID_LED_RESERVED_0000 || *data == ID_LED_RESERVED_FFFF)
*data = ID_LED_DEFAULT_ICH8LAN;
return 0;
@@ -3511,9 +3511,9 @@
/* Initialize identification LED */
ret_val = mac->ops.id_led_init(hw);
+ /* An error is not fatal and we should not stop init due to this */
if (ret_val)
e_dbg("Error initializing identification LED\n");
- /* This is not fatal and we should not stop init due to this */
/* Setup the receive address. */
e1000e_init_rx_addrs(hw, mac->rar_entry_count);
@@ -3541,16 +3541,16 @@
/* Set the transmit descriptor write-back policy for both queues */
txdctl = er32(TXDCTL(0));
- txdctl = (txdctl & ~E1000_TXDCTL_WTHRESH) |
- E1000_TXDCTL_FULL_TX_DESC_WB;
- txdctl = (txdctl & ~E1000_TXDCTL_PTHRESH) |
- E1000_TXDCTL_MAX_TX_DESC_PREFETCH;
+ txdctl = ((txdctl & ~E1000_TXDCTL_WTHRESH) |
+ E1000_TXDCTL_FULL_TX_DESC_WB);
+ txdctl = ((txdctl & ~E1000_TXDCTL_PTHRESH) |
+ E1000_TXDCTL_MAX_TX_DESC_PREFETCH);
ew32(TXDCTL(0), txdctl);
txdctl = er32(TXDCTL(1));
- txdctl = (txdctl & ~E1000_TXDCTL_WTHRESH) |
- E1000_TXDCTL_FULL_TX_DESC_WB;
- txdctl = (txdctl & ~E1000_TXDCTL_PTHRESH) |
- E1000_TXDCTL_MAX_TX_DESC_PREFETCH;
+ txdctl = ((txdctl & ~E1000_TXDCTL_WTHRESH) |
+ E1000_TXDCTL_FULL_TX_DESC_WB);
+ txdctl = ((txdctl & ~E1000_TXDCTL_PTHRESH) |
+ E1000_TXDCTL_MAX_TX_DESC_PREFETCH);
ew32(TXDCTL(1), txdctl);
/* ICH8 has opposite polarity of no_snoop bits.
@@ -3559,7 +3559,7 @@
if (mac->type == e1000_ich8lan)
snoop = PCIE_ICH8_SNOOP_ALL;
else
- snoop = (u32) ~(PCIE_NO_SNOOP_ALL);
+ snoop = (u32)~(PCIE_NO_SNOOP_ALL);
e1000e_set_pcie_no_snoop(hw, snoop);
ctrl_ext = er32(CTRL_EXT);
@@ -3575,6 +3575,7 @@
return ret_val;
}
+
/**
* e1000_initialize_hw_bits_ich8lan - Initialize required hardware bits
* @hw: pointer to the HW structure
@@ -3686,8 +3687,7 @@
*/
hw->fc.current_mode = hw->fc.requested_mode;
- e_dbg("After fix-ups FlowControl is now = %x\n",
- hw->fc.current_mode);
+ e_dbg("After fix-ups FlowControl is now = %x\n", hw->fc.current_mode);
/* Continue to configure the copper link. */
ret_val = hw->mac.ops.setup_physical_interface(hw);
@@ -3737,12 +3737,12 @@
if (ret_val)
return ret_val;
ret_val = e1000e_read_kmrn_reg(hw, E1000_KMRNCTRLSTA_INBAND_PARAM,
- ®_data);
+ ®_data);
if (ret_val)
return ret_val;
reg_data |= 0x3F;
ret_val = e1000e_write_kmrn_reg(hw, E1000_KMRNCTRLSTA_INBAND_PARAM,
- reg_data);
+ reg_data);
if (ret_val)
return ret_val;
@@ -3815,8 +3815,7 @@
return ret_val;
if ((hw->mac.type == e1000_ich8lan) &&
- (hw->phy.type == e1000_phy_igp_3) &&
- (*speed == SPEED_1000)) {
+ (hw->phy.type == e1000_phy_igp_3) && (*speed == SPEED_1000)) {
ret_val = e1000_kmrn_lock_loss_workaround_ich8lan(hw);
}
@@ -3899,7 +3898,7 @@
* /disabled - false).
**/
void e1000e_set_kmrn_lock_loss_workaround_ich8lan(struct e1000_hw *hw,
- bool state)
+ bool state)
{
struct e1000_dev_spec_ich8lan *dev_spec = &hw->dev_spec.ich8lan;
@@ -3981,12 +3980,12 @@
return;
ret_val = e1000e_read_kmrn_reg(hw, E1000_KMRNCTRLSTA_DIAG_OFFSET,
- ®_data);
+ ®_data);
if (ret_val)
return;
reg_data |= E1000_KMRNCTRLSTA_DIAG_NELPBK;
ret_val = e1000e_write_kmrn_reg(hw, E1000_KMRNCTRLSTA_DIAG_OFFSET,
- reg_data);
+ reg_data);
if (ret_val)
return;
reg_data &= ~E1000_KMRNCTRLSTA_DIAG_NELPBK;
diff --git a/drivers/net/ethernet/intel/e1000e/mac.c b/drivers/net/ethernet/intel/e1000e/mac.c
index b78e021..b25cc43 100644
--- a/drivers/net/ethernet/intel/e1000e/mac.c
+++ b/drivers/net/ethernet/intel/e1000e/mac.c
@@ -596,7 +596,7 @@
* serdes media type.
*/
/* SYNCH bit and IV bit are sticky. */
- udelay(10);
+ usleep_range(10, 20);
rxcw = er32(RXCW);
if (rxcw & E1000_RXCW_SYNCH) {
if (!(rxcw & E1000_RXCW_IV)) {
@@ -613,7 +613,7 @@
status = er32(STATUS);
if (status & E1000_STATUS_LU) {
/* SYNCH bit and IV bit are sticky, so reread rxcw. */
- udelay(10);
+ usleep_range(10, 20);
rxcw = er32(RXCW);
if (rxcw & E1000_RXCW_SYNCH) {
if (!(rxcw & E1000_RXCW_IV)) {
@@ -1382,7 +1382,7 @@
if (!(swsm & E1000_SWSM_SMBI))
break;
- udelay(50);
+ usleep_range(50, 100);
i++;
}
@@ -1400,7 +1400,7 @@
if (er32(SWSM) & E1000_SWSM_SWESMBI)
break;
- udelay(50);
+ usleep_range(50, 100);
}
if (i == timeout) {
@@ -1712,7 +1712,7 @@
while (timeout) {
if (!(er32(STATUS) & E1000_STATUS_GIO_MASTER_ENABLE))
break;
- udelay(100);
+ usleep_range(100, 200);
timeout--;
}
diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c
index 948b86ff..142ca39 100644
--- a/drivers/net/ethernet/intel/e1000e/netdev.c
+++ b/drivers/net/ethernet/intel/e1000e/netdev.c
@@ -219,9 +219,8 @@
if (netdev) {
dev_info(&adapter->pdev->dev, "Net device Info\n");
pr_info("Device Name state trans_start last_rx\n");
- pr_info("%-15s %016lX %016lX %016lX\n",
- netdev->name, netdev->state, netdev->trans_start,
- netdev->last_rx);
+ pr_info("%-15s %016lX %016lX %016lX\n", netdev->name,
+ netdev->state, netdev->trans_start, netdev->last_rx);
}
/* Print Registers */
@@ -755,8 +754,7 @@
cpu_to_le64(ps_page->dma);
}
- skb = __netdev_alloc_skb_ip_align(netdev,
- adapter->rx_ps_bsize0,
+ skb = __netdev_alloc_skb_ip_align(netdev, adapter->rx_ps_bsize0,
gfp);
if (!skb) {
@@ -850,8 +848,8 @@
if (!buffer_info->dma)
buffer_info->dma = dma_map_page(&pdev->dev,
- buffer_info->page, 0,
- PAGE_SIZE,
+ buffer_info->page, 0,
+ PAGE_SIZE,
DMA_FROM_DEVICE);
rx_desc = E1000_RX_DESC_EXT(*rx_ring, i);
@@ -937,10 +935,8 @@
cleaned = true;
cleaned_count++;
- dma_unmap_single(&pdev->dev,
- buffer_info->dma,
- adapter->rx_buffer_len,
- DMA_FROM_DEVICE);
+ dma_unmap_single(&pdev->dev, buffer_info->dma,
+ adapter->rx_buffer_len, DMA_FROM_DEVICE);
buffer_info->dma = 0;
length = le16_to_cpu(rx_desc->wb.upper.length);
@@ -1068,8 +1064,8 @@
static void e1000_print_hw_hang(struct work_struct *work)
{
struct e1000_adapter *adapter = container_of(work,
- struct e1000_adapter,
- print_hang_task);
+ struct e1000_adapter,
+ print_hang_task);
struct net_device *netdev = adapter->netdev;
struct e1000_ring *tx_ring = adapter->tx_ring;
unsigned int i = tx_ring->next_to_clean;
@@ -1082,8 +1078,7 @@
if (test_bit(__E1000_DOWN, &adapter->state))
return;
- if (!adapter->tx_hang_recheck &&
- (adapter->flags2 & FLAG2_DMA_BURST)) {
+ if (!adapter->tx_hang_recheck && (adapter->flags2 & FLAG2_DMA_BURST)) {
/* May be block on write-back, flush and detect again
* flush pending descriptor writebacks to memory
*/
@@ -1125,19 +1120,10 @@
"PHY 1000BASE-T Status <%x>\n"
"PHY Extended Status <%x>\n"
"PCI Status <%x>\n",
- readl(tx_ring->head),
- readl(tx_ring->tail),
- tx_ring->next_to_use,
- tx_ring->next_to_clean,
- tx_ring->buffer_info[eop].time_stamp,
- eop,
- jiffies,
- eop_desc->upper.fields.status,
- er32(STATUS),
- phy_status,
- phy_1000t_status,
- phy_ext_status,
- pci_status);
+ readl(tx_ring->head), readl(tx_ring->tail), tx_ring->next_to_use,
+ tx_ring->next_to_clean, tx_ring->buffer_info[eop].time_stamp,
+ eop, jiffies, eop_desc->upper.fields.status, er32(STATUS),
+ phy_status, phy_1000t_status, phy_ext_status, pci_status);
/* Suggest workaround for known h/w issue */
if ((hw->mac.type == e1000_pchlan) && (er32(CTRL) & E1000_CTRL_TFCE))
@@ -1430,7 +1416,7 @@
e1000_rx_hash(netdev, rx_desc->wb.lower.hi_dword.rss, skb);
if (rx_desc->wb.upper.header_status &
- cpu_to_le16(E1000_RXDPS_HDRSTAT_HDRSP))
+ cpu_to_le16(E1000_RXDPS_HDRSTAT_HDRSP))
adapter->rx_hdr_split++;
e1000_receive_skb(adapter, netdev, skb, staterr,
@@ -1468,7 +1454,7 @@
* e1000_consume_page - helper function
**/
static void e1000_consume_page(struct e1000_buffer *bi, struct sk_buff *skb,
- u16 length)
+ u16 length)
{
bi->page = NULL;
skb->len += length;
@@ -1495,7 +1481,8 @@
unsigned int i;
int cleaned_count = 0;
bool cleaned = false;
- unsigned int total_rx_bytes=0, total_rx_packets=0;
+ unsigned int total_rx_bytes = 0, total_rx_packets = 0;
+ struct skb_shared_info *shinfo;
i = rx_ring->next_to_clean;
rx_desc = E1000_RX_DESC_EXT(*rx_ring, i);
@@ -1541,7 +1528,6 @@
rx_ring->rx_skb_top = NULL;
goto next_desc;
}
-
#define rxtop (rx_ring->rx_skb_top)
if (!(staterr & E1000_RXD_STAT_EOP)) {
/* this descriptor is only the beginning (or middle) */
@@ -1549,12 +1535,13 @@
/* this is the beginning of a chain */
rxtop = skb;
skb_fill_page_desc(rxtop, 0, buffer_info->page,
- 0, length);
+ 0, length);
} else {
/* this is the middle of a chain */
- skb_fill_page_desc(rxtop,
- skb_shinfo(rxtop)->nr_frags,
- buffer_info->page, 0, length);
+ shinfo = skb_shinfo(rxtop);
+ skb_fill_page_desc(rxtop, shinfo->nr_frags,
+ buffer_info->page, 0,
+ length);
/* re-use the skb, only consumed the page */
buffer_info->skb = skb;
}
@@ -1563,9 +1550,10 @@
} else {
if (rxtop) {
/* end of the chain */
- skb_fill_page_desc(rxtop,
- skb_shinfo(rxtop)->nr_frags,
- buffer_info->page, 0, length);
+ shinfo = skb_shinfo(rxtop);
+ skb_fill_page_desc(rxtop, shinfo->nr_frags,
+ buffer_info->page, 0,
+ length);
/* re-use the current skb, we only consumed the
* page
*/
@@ -1590,10 +1578,10 @@
skb_put(skb, length);
} else {
skb_fill_page_desc(skb, 0,
- buffer_info->page, 0,
- length);
+ buffer_info->page, 0,
+ length);
e1000_consume_page(buffer_info, skb,
- length);
+ length);
}
}
}
@@ -1666,8 +1654,7 @@
DMA_FROM_DEVICE);
else if (adapter->clean_rx == e1000_clean_jumbo_rx_irq)
dma_unmap_page(&pdev->dev, buffer_info->dma,
- PAGE_SIZE,
- DMA_FROM_DEVICE);
+ PAGE_SIZE, DMA_FROM_DEVICE);
else if (adapter->clean_rx == e1000_clean_rx_irq_ps)
dma_unmap_single(&pdev->dev, buffer_info->dma,
adapter->rx_ps_bsize0,
@@ -1720,7 +1707,8 @@
static void e1000e_downshift_workaround(struct work_struct *work)
{
struct e1000_adapter *adapter = container_of(work,
- struct e1000_adapter, downshift_task);
+ struct e1000_adapter,
+ downshift_task);
if (test_bit(__E1000_DOWN, &adapter->state))
return;
@@ -1913,7 +1901,6 @@
struct e1000_hw *hw = &adapter->hw;
struct e1000_ring *tx_ring = adapter->tx_ring;
-
adapter->total_tx_bytes = 0;
adapter->total_tx_packets = 0;
@@ -1970,7 +1957,6 @@
ew32(RFCTL, rfctl);
}
-#define E1000_IVAR_INT_ALLOC_VALID 0x8
/* Configure Rx vector */
rx_ring->ims_val = E1000_IMS_RXQ0;
adapter->eiac_mask |= rx_ring->ims_val;
@@ -2045,8 +2031,9 @@
if (adapter->flags & FLAG_HAS_MSIX) {
adapter->num_vectors = 3; /* RxQ0, TxQ0 and other */
adapter->msix_entries = kcalloc(adapter->num_vectors,
- sizeof(struct msix_entry),
- GFP_KERNEL);
+ sizeof(struct
+ msix_entry),
+ GFP_KERNEL);
if (adapter->msix_entries) {
for (i = 0; i < adapter->num_vectors; i++)
adapter->msix_entries[i].entry = i;
@@ -2490,7 +2477,7 @@
switch (itr_setting) {
case lowest_latency:
/* handle TSO and jumbo frames */
- if (bytes/packets > 8000)
+ if (bytes / packets > 8000)
retval = bulk_latency;
else if ((packets < 5) && (bytes > 512))
retval = low_latency;
@@ -2498,13 +2485,13 @@
case low_latency: /* 50 usec aka 20000 ints/s */
if (bytes > 10000) {
/* this if handles the TSO accounting */
- if (bytes/packets > 8000)
+ if (bytes / packets > 8000)
retval = bulk_latency;
- else if ((packets < 10) || ((bytes/packets) > 1200))
+ else if ((packets < 10) || ((bytes / packets) > 1200))
retval = bulk_latency;
else if ((packets > 35))
retval = lowest_latency;
- } else if (bytes/packets > 2000) {
+ } else if (bytes / packets > 2000) {
retval = bulk_latency;
} else if (packets <= 2 && bytes < 512) {
retval = lowest_latency;
@@ -2556,8 +2543,8 @@
current_itr = max(adapter->rx_itr, adapter->tx_itr);
- switch (current_itr) {
/* counts and packets in update_itr are dependent on these numbers */
+ switch (current_itr) {
case lowest_latency:
new_itr = 70000;
break;
@@ -2578,8 +2565,7 @@
* increasing
*/
new_itr = new_itr > adapter->itr ?
- min(adapter->itr + (new_itr >> 2), new_itr) :
- new_itr;
+ min(adapter->itr + (new_itr >> 2), new_itr) : new_itr;
adapter->itr = new_itr;
adapter->rx_ring->itr_val = new_itr;
if (adapter->msix_entries)
@@ -2810,8 +2796,7 @@
u16 vid = adapter->hw.mng_cookie.vlan_id;
u16 old_vid = adapter->mng_vlan_id;
- if (adapter->hw.mng_cookie.status &
- E1000_MNG_DHCP_COOKIE_STATUS_VLAN) {
+ if (adapter->hw.mng_cookie.status & E1000_MNG_DHCP_COOKIE_STATUS_VLAN) {
e1000_vlan_rx_add_vid(netdev, vid);
adapter->mng_vlan_id = vid;
}
@@ -2827,7 +2812,7 @@
e1000_vlan_rx_add_vid(adapter->netdev, 0);
for_each_set_bit(vid, adapter->active_vlans, VLAN_N_VID)
- e1000_vlan_rx_add_vid(adapter->netdev, vid);
+ e1000_vlan_rx_add_vid(adapter->netdev, vid);
}
static void e1000_init_manageability_pt(struct e1000_adapter *adapter)
@@ -3002,8 +2987,8 @@
rctl = er32(RCTL);
rctl &= ~(3 << E1000_RCTL_MO_SHIFT);
rctl |= E1000_RCTL_EN | E1000_RCTL_BAM |
- E1000_RCTL_LBM_NO | E1000_RCTL_RDMTS_HALF |
- (adapter->hw.mac.mc_filter_type << E1000_RCTL_MO_SHIFT);
+ E1000_RCTL_LBM_NO | E1000_RCTL_RDMTS_HALF |
+ (adapter->hw.mac.mc_filter_type << E1000_RCTL_MO_SHIFT);
/* Do not Store bad packets */
rctl &= ~E1000_RCTL_SBP;
@@ -3089,19 +3074,17 @@
/* Enable Packet split descriptors */
rctl |= E1000_RCTL_DTYP_PS;
- psrctl |= adapter->rx_ps_bsize0 >>
- E1000_PSRCTL_BSIZE0_SHIFT;
+ psrctl |= adapter->rx_ps_bsize0 >> E1000_PSRCTL_BSIZE0_SHIFT;
switch (adapter->rx_ps_pages) {
case 3:
- psrctl |= PAGE_SIZE <<
- E1000_PSRCTL_BSIZE3_SHIFT;
+ psrctl |= PAGE_SIZE << E1000_PSRCTL_BSIZE3_SHIFT;
+ /* fall-through */
case 2:
- psrctl |= PAGE_SIZE <<
- E1000_PSRCTL_BSIZE2_SHIFT;
+ psrctl |= PAGE_SIZE << E1000_PSRCTL_BSIZE2_SHIFT;
+ /* fall-through */
case 1:
- psrctl |= PAGE_SIZE >>
- E1000_PSRCTL_BSIZE1_SHIFT;
+ psrctl |= PAGE_SIZE >> E1000_PSRCTL_BSIZE1_SHIFT;
break;
}
@@ -3275,7 +3258,7 @@
/* update_mc_addr_list expects a packed array of only addresses. */
i = 0;
netdev_for_each_mc_addr(ha, netdev)
- memcpy(mta_list + (i++ * ETH_ALEN), ha->addr, ETH_ALEN);
+ memcpy(mta_list + (i++ * ETH_ALEN), ha->addr, ETH_ALEN);
hw->mac.ops.update_mc_addr_list(hw, mta_list, i);
kfree(mta_list);
@@ -3752,8 +3735,7 @@
* but don't include ethernet FCS because hardware appends it
*/
min_tx_space = (adapter->max_frame_size +
- sizeof(struct e1000_tx_desc) -
- ETH_FCS_LEN) * 2;
+ sizeof(struct e1000_tx_desc) - ETH_FCS_LEN) * 2;
min_tx_space = ALIGN(min_tx_space, 1024);
min_tx_space >>= 10;
/* software strips receive CRC, so leave room for it */
@@ -3856,13 +3838,13 @@
if ((adapter->max_frame_size * 2) > (pba << 10)) {
if (!(adapter->flags2 & FLAG2_DISABLE_AIM)) {
dev_info(&adapter->pdev->dev,
- "Interrupt Throttle Rate turned off\n");
+ "Interrupt Throttle Rate off\n");
adapter->flags2 |= FLAG2_DISABLE_AIM;
e1000e_write_itr(adapter, 0);
}
} else if (adapter->flags2 & FLAG2_DISABLE_AIM) {
dev_info(&adapter->pdev->dev,
- "Interrupt Throttle Rate turned on\n");
+ "Interrupt Throttle Rate on\n");
adapter->flags2 &= ~FLAG2_DISABLE_AIM;
adapter->itr = 20000;
e1000e_write_itr(adapter, adapter->itr);
@@ -4261,8 +4243,7 @@
e1000e_power_up_phy(adapter);
adapter->mng_vlan_id = E1000_MNG_VLAN_NONE;
- if ((adapter->hw.mng_cookie.status &
- E1000_MNG_DHCP_COOKIE_STATUS_VLAN))
+ if ((adapter->hw.mng_cookie.status & E1000_MNG_DHCP_COOKIE_STATUS_VLAN))
e1000_update_mng_vlan(adapter);
/* DMA latency requirement to workaround jumbo issue */
@@ -4365,8 +4346,7 @@
/* kill manageability vlan ID if supported, but not if a vlan with
* the same ID is registered on the host OS (let 8021q kill it)
*/
- if (adapter->hw.mng_cookie.status &
- E1000_MNG_DHCP_COOKIE_STATUS_VLAN)
+ if (adapter->hw.mng_cookie.status & E1000_MNG_DHCP_COOKIE_STATUS_VLAN)
e1000_vlan_rx_kill_vid(netdev, adapter->mng_vlan_id);
/* If AMT is enabled, let the firmware know that the network
@@ -4382,6 +4362,7 @@
return 0;
}
+
/**
* e1000_set_mac - Change the Ethernet Address of the NIC
* @netdev: network interface device structure
@@ -4432,7 +4413,8 @@
static void e1000e_update_phy_task(struct work_struct *work)
{
struct e1000_adapter *adapter = container_of(work,
- struct e1000_adapter, update_phy_task);
+ struct e1000_adapter,
+ update_phy_task);
if (test_bit(__E1000_DOWN, &adapter->state))
return;
@@ -4449,7 +4431,7 @@
**/
static void e1000_update_phy_info(unsigned long data)
{
- struct e1000_adapter *adapter = (struct e1000_adapter *) data;
+ struct e1000_adapter *adapter = (struct e1000_adapter *)data;
if (test_bit(__E1000_DOWN, &adapter->state))
return;
@@ -4616,18 +4598,16 @@
* our own version based on RUC and ROC
*/
netdev->stats.rx_errors = adapter->stats.rxerrc +
- adapter->stats.crcerrs + adapter->stats.algnerrc +
- adapter->stats.ruc + adapter->stats.roc +
- adapter->stats.cexterr;
+ adapter->stats.crcerrs + adapter->stats.algnerrc +
+ adapter->stats.ruc + adapter->stats.roc + adapter->stats.cexterr;
netdev->stats.rx_length_errors = adapter->stats.ruc +
- adapter->stats.roc;
+ adapter->stats.roc;
netdev->stats.rx_crc_errors = adapter->stats.crcerrs;
netdev->stats.rx_frame_errors = adapter->stats.algnerrc;
netdev->stats.rx_missed_errors = adapter->stats.mpc;
/* Tx Errors */
- netdev->stats.tx_errors = adapter->stats.ecol +
- adapter->stats.latecol;
+ netdev->stats.tx_errors = adapter->stats.ecol + adapter->stats.latecol;
netdev->stats.tx_aborted_errors = adapter->stats.ecol;
netdev->stats.tx_window_errors = adapter->stats.latecol;
netdev->stats.tx_carrier_errors = adapter->stats.tncrs;
@@ -4785,7 +4765,7 @@
**/
static void e1000_watchdog(unsigned long data)
{
- struct e1000_adapter *adapter = (struct e1000_adapter *) data;
+ struct e1000_adapter *adapter = (struct e1000_adapter *)data;
/* Do the rest outside of interrupt context */
schedule_work(&adapter->watchdog_task);
@@ -4796,7 +4776,8 @@
static void e1000_watchdog_task(struct work_struct *work)
{
struct e1000_adapter *adapter = container_of(work,
- struct e1000_adapter, watchdog_task);
+ struct e1000_adapter,
+ watchdog_task);
struct net_device *netdev = adapter->netdev;
struct e1000_mac_info *mac = &adapter->hw.mac;
struct e1000_phy_info *phy = &adapter->hw.phy;
@@ -4830,8 +4811,8 @@
/* update snapshot of PHY registers on LSC */
e1000_phy_read_status(adapter);
mac->ops.get_link_up_info(&adapter->hw,
- &adapter->link_speed,
- &adapter->link_duplex);
+ &adapter->link_speed,
+ &adapter->link_duplex);
e1000_print_link_info(adapter);
/* check if SmartSpeed worked */
@@ -4944,7 +4925,7 @@
adapter->flags |= FLAG_RESTART_NOW;
else
pm_schedule_suspend(netdev->dev.parent,
- LINK_TIMEOUT);
+ LINK_TIMEOUT);
}
}
@@ -4979,8 +4960,8 @@
*/
u32 goc = (adapter->gotc + adapter->gorc) / 10000;
u32 dif = (adapter->gotc > adapter->gorc ?
- adapter->gotc - adapter->gorc :
- adapter->gorc - adapter->gotc) / 10000;
+ adapter->gotc - adapter->gorc :
+ adapter->gorc - adapter->gotc) / 10000;
u32 itr = goc > 0 ? (dif * 6000 / goc + 2000) : 8000;
e1000e_write_itr(adapter, itr);
@@ -5059,14 +5040,14 @@
iph->tot_len = 0;
iph->check = 0;
tcp_hdr(skb)->check = ~csum_tcpudp_magic(iph->saddr, iph->daddr,
- 0, IPPROTO_TCP, 0);
+ 0, IPPROTO_TCP, 0);
cmd_length = E1000_TXD_CMD_IP;
ipcse = skb_transport_offset(skb) - 1;
} else if (skb_is_gso_v6(skb)) {
ipv6_hdr(skb)->payload_len = 0;
tcp_hdr(skb)->check = ~csum_ipv6_magic(&ipv6_hdr(skb)->saddr,
- &ipv6_hdr(skb)->daddr,
- 0, IPPROTO_TCP, 0);
+ &ipv6_hdr(skb)->daddr,
+ 0, IPPROTO_TCP, 0);
ipcse = 0;
}
ipcss = skb_network_offset(skb);
@@ -5075,7 +5056,7 @@
tucso = (void *)&(tcp_hdr(skb)->check) - (void *)skb->data;
cmd_length |= (E1000_TXD_CMD_DEXT | E1000_TXD_CMD_TSE |
- E1000_TXD_CMD_TCP | (skb->len - (hdr_len)));
+ E1000_TXD_CMD_TCP | (skb->len - (hdr_len)));
i = tx_ring->next_to_use;
context_desc = E1000_CONTEXT_DESC(*tx_ring, i);
@@ -5145,8 +5126,7 @@
context_desc->lower_setup.ip_config = 0;
context_desc->upper_setup.tcp_fields.tucss = css;
- context_desc->upper_setup.tcp_fields.tucso =
- css + skb->csum_offset;
+ context_desc->upper_setup.tcp_fields.tucso = css + skb->csum_offset;
context_desc->upper_setup.tcp_fields.tucse = 0;
context_desc->tcp_seg_setup.data = 0;
context_desc->cmd_and_length = cpu_to_le32(cmd_len);
@@ -5219,7 +5199,8 @@
buffer_info->time_stamp = jiffies;
buffer_info->next_to_watch = i;
buffer_info->dma = skb_frag_dma_map(&pdev->dev, frag,
- offset, size, DMA_TO_DEVICE);
+ offset, size,
+ DMA_TO_DEVICE);
buffer_info->mapped_as_page = true;
if (dma_mapping_error(&pdev->dev, buffer_info->dma))
goto dma_error;
@@ -5268,7 +5249,7 @@
if (tx_flags & E1000_TX_FLAGS_TSO) {
txd_lower |= E1000_TXD_CMD_DEXT | E1000_TXD_DTYP_D |
- E1000_TXD_CMD_TSE;
+ E1000_TXD_CMD_TSE;
txd_upper |= E1000_TXD_POPTS_TXSM << 8;
if (tx_flags & E1000_TX_FLAGS_IPV4)
@@ -5299,8 +5280,8 @@
buffer_info = &tx_ring->buffer_info[i];
tx_desc = E1000_TX_DESC(*tx_ring, i);
tx_desc->buffer_addr = cpu_to_le64(buffer_info->dma);
- tx_desc->lower.data =
- cpu_to_le32(txd_lower | buffer_info->length);
+ tx_desc->lower.data = cpu_to_le32(txd_lower |
+ buffer_info->length);
tx_desc->upper.data = cpu_to_le32(txd_upper);
i++;
@@ -5350,11 +5331,11 @@
if (skb->len <= MINIMUM_DHCP_PACKET_SIZE)
return 0;
- if (((struct ethhdr *) skb->data)->h_proto != htons(ETH_P_IP))
+ if (((struct ethhdr *)skb->data)->h_proto != htons(ETH_P_IP))
return 0;
{
- const struct iphdr *ip = (struct iphdr *)((u8 *)skb->data+14);
+ const struct iphdr *ip = (struct iphdr *)((u8 *)skb->data + 14);
struct udphdr *udp;
if (ip->protocol != IPPROTO_UDP)
@@ -5579,7 +5560,7 @@
* Returns the address of the device statistics structure.
**/
struct rtnl_link_stats64 *e1000e_get_stats64(struct net_device *netdev,
- struct rtnl_link_stats64 *stats)
+ struct rtnl_link_stats64 *stats)
{
struct e1000_adapter *adapter = netdev_priv(netdev);
@@ -5600,18 +5581,15 @@
* our own version based on RUC and ROC
*/
stats->rx_errors = adapter->stats.rxerrc +
- adapter->stats.crcerrs + adapter->stats.algnerrc +
- adapter->stats.ruc + adapter->stats.roc +
- adapter->stats.cexterr;
- stats->rx_length_errors = adapter->stats.ruc +
- adapter->stats.roc;
+ adapter->stats.crcerrs + adapter->stats.algnerrc +
+ adapter->stats.ruc + adapter->stats.roc + adapter->stats.cexterr;
+ stats->rx_length_errors = adapter->stats.ruc + adapter->stats.roc;
stats->rx_crc_errors = adapter->stats.crcerrs;
stats->rx_frame_errors = adapter->stats.algnerrc;
stats->rx_missed_errors = adapter->stats.mpc;
/* Tx Errors */
- stats->tx_errors = adapter->stats.ecol +
- adapter->stats.latecol;
+ stats->tx_errors = adapter->stats.ecol + adapter->stats.latecol;
stats->tx_aborted_errors = adapter->stats.ecol;
stats->tx_window_errors = adapter->stats.latecol;
stats->tx_carrier_errors = adapter->stats.tncrs;
@@ -5680,9 +5658,9 @@
/* adjust allocation if LPE protects us, and we aren't using SBP */
if ((max_frame == ETH_FRAME_LEN + ETH_FCS_LEN) ||
- (max_frame == ETH_FRAME_LEN + VLAN_HLEN + ETH_FCS_LEN))
+ (max_frame == ETH_FRAME_LEN + VLAN_HLEN + ETH_FCS_LEN))
adapter->rx_buffer_len = ETH_FRAME_LEN + VLAN_HLEN
- + ETH_FCS_LEN;
+ + ETH_FCS_LEN;
if (netif_running(netdev))
e1000e_up(adapter);
@@ -5861,7 +5839,7 @@
phy_reg &= ~(BM_RCTL_MO_MASK);
if (mac_reg & E1000_RCTL_MO_3)
phy_reg |= (((mac_reg & E1000_RCTL_MO_3) >> E1000_RCTL_MO_SHIFT)
- << BM_RCTL_MO_SHIFT);
+ << BM_RCTL_MO_SHIFT);
if (mac_reg & E1000_RCTL_BAM)
phy_reg |= BM_RCTL_BAM;
if (mac_reg & E1000_RCTL_PMCF)
@@ -5930,10 +5908,6 @@
}
ctrl = er32(CTRL);
- /* advertise wake from D3Cold */
- #define E1000_CTRL_ADVD3WUC 0x00100000
- /* phy power management enable */
- #define E1000_CTRL_EN_PHY_PWR_MGMT 0x00200000
ctrl |= E1000_CTRL_ADVD3WUC;
if (!(adapter->flags2 & FLAG2_HAS_PHY_WAKEUP))
ctrl |= E1000_CTRL_EN_PHY_PWR_MGMT;
@@ -5977,8 +5951,6 @@
*/
e1000e_release_hw_control(adapter);
- pci_clear_master(pdev);
-
/* The pci-e switch on some quad port adapters will report a
* correctable error when the MAC transitions from D0 to D3. To
* prevent this we need to mask off the correctable errors on the
@@ -6077,24 +6049,24 @@
e1e_rphy(&adapter->hw, BM_WUS, &phy_data);
if (phy_data) {
e_info("PHY Wakeup cause - %s\n",
- phy_data & E1000_WUS_EX ? "Unicast Packet" :
- phy_data & E1000_WUS_MC ? "Multicast Packet" :
- phy_data & E1000_WUS_BC ? "Broadcast Packet" :
- phy_data & E1000_WUS_MAG ? "Magic Packet" :
- phy_data & E1000_WUS_LNKC ?
- "Link Status Change" : "other");
+ phy_data & E1000_WUS_EX ? "Unicast Packet" :
+ phy_data & E1000_WUS_MC ? "Multicast Packet" :
+ phy_data & E1000_WUS_BC ? "Broadcast Packet" :
+ phy_data & E1000_WUS_MAG ? "Magic Packet" :
+ phy_data & E1000_WUS_LNKC ?
+ "Link Status Change" : "other");
}
e1e_wphy(&adapter->hw, BM_WUS, ~0);
} else {
u32 wus = er32(WUS);
if (wus) {
e_info("MAC Wakeup cause - %s\n",
- wus & E1000_WUS_EX ? "Unicast Packet" :
- wus & E1000_WUS_MC ? "Multicast Packet" :
- wus & E1000_WUS_BC ? "Broadcast Packet" :
- wus & E1000_WUS_MAG ? "Magic Packet" :
- wus & E1000_WUS_LNKC ? "Link Status Change" :
- "other");
+ wus & E1000_WUS_EX ? "Unicast Packet" :
+ wus & E1000_WUS_MC ? "Multicast Packet" :
+ wus & E1000_WUS_BC ? "Broadcast Packet" :
+ wus & E1000_WUS_MAG ? "Magic Packet" :
+ wus & E1000_WUS_LNKC ? "Link Status Change" :
+ "other");
}
ew32(WUS, ~0);
}
@@ -6369,7 +6341,7 @@
e_info("(PCI Express:2.5GT/s:%s) %pM\n",
/* bus width */
((hw->bus.width == e1000_bus_width_pcie_x4) ? "Width x4" :
- "Width x1"),
+ "Width x1"),
/* MAC address */
netdev->dev_addr);
e_info("Intel(R) PRO/%s Network Connection\n",
@@ -6479,7 +6451,7 @@
resource_size_t flash_start, flash_len;
static int cards_found;
u16 aspm_disable_flag = 0;
- int i, err, pci_using_dac;
+ int bars, i, err, pci_using_dac;
u16 eeprom_data = 0;
u16 eeprom_apme_mask = E1000_EEPROM_APME;
@@ -6506,15 +6478,16 @@
err = dma_set_coherent_mask(&pdev->dev,
DMA_BIT_MASK(32));
if (err) {
- dev_err(&pdev->dev, "No usable DMA configuration, aborting\n");
+ dev_err(&pdev->dev,
+ "No usable DMA configuration, aborting\n");
goto err_dma;
}
}
}
- err = pci_request_selected_regions_exclusive(pdev,
- pci_select_bars(pdev, IORESOURCE_MEM),
- e1000e_driver_name);
+ bars = pci_select_bars(pdev, IORESOURCE_MEM);
+ err = pci_request_selected_regions_exclusive(pdev, bars,
+ e1000e_driver_name);
if (err)
goto err_pci_reg;
@@ -6683,11 +6656,11 @@
init_timer(&adapter->watchdog_timer);
adapter->watchdog_timer.function = e1000_watchdog;
- adapter->watchdog_timer.data = (unsigned long) adapter;
+ adapter->watchdog_timer.data = (unsigned long)adapter;
init_timer(&adapter->phy_info_timer);
adapter->phy_info_timer.function = e1000_update_phy_info;
- adapter->phy_info_timer.data = (unsigned long) adapter;
+ adapter->phy_info_timer.data = (unsigned long)adapter;
INIT_WORK(&adapter->reset_task, e1000_reset_task);
INIT_WORK(&adapter->watchdog_task, e1000_watchdog_task);
@@ -6795,7 +6768,7 @@
free_netdev(netdev);
err_alloc_etherdev:
pci_release_selected_regions(pdev,
- pci_select_bars(pdev, IORESOURCE_MEM));
+ pci_select_bars(pdev, IORESOURCE_MEM));
err_pci_reg:
err_dma:
pci_disable_device(pdev);
@@ -6865,7 +6838,7 @@
if (adapter->hw.flash_address)
iounmap(adapter->hw.flash_address);
pci_release_selected_regions(pdev,
- pci_select_bars(pdev, IORESOURCE_MEM));
+ pci_select_bars(pdev, IORESOURCE_MEM));
free_netdev(netdev);
@@ -6886,7 +6859,8 @@
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82571EB_COPPER), board_82571 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82571EB_FIBER), board_82571 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82571EB_QUAD_COPPER), board_82571 },
- { PCI_VDEVICE(INTEL, E1000_DEV_ID_82571EB_QUAD_COPPER_LP), board_82571 },
+ { PCI_VDEVICE(INTEL, E1000_DEV_ID_82571EB_QUAD_COPPER_LP),
+ board_82571 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82571EB_QUAD_FIBER), board_82571 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82571EB_SERDES), board_82571 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82571EB_SERDES_DUAL), board_82571 },
@@ -6962,8 +6936,8 @@
#ifdef CONFIG_PM
static const struct dev_pm_ops e1000_pm_ops = {
SET_SYSTEM_SLEEP_PM_OPS(e1000_suspend, e1000_resume)
- SET_RUNTIME_PM_OPS(e1000_runtime_suspend,
- e1000_runtime_resume, e1000_idle)
+ SET_RUNTIME_PM_OPS(e1000_runtime_suspend, e1000_runtime_resume,
+ e1000_idle)
};
#endif
diff --git a/drivers/net/ethernet/intel/e1000e/nvm.c b/drivers/net/ethernet/intel/e1000e/nvm.c
index 84fecc2..44ddc0a 100644
--- a/drivers/net/ethernet/intel/e1000e/nvm.c
+++ b/drivers/net/ethernet/intel/e1000e/nvm.c
@@ -630,7 +630,7 @@
{
u32 ctrl_ext;
- udelay(10);
+ usleep_range(10, 20);
ctrl_ext = er32(CTRL_EXT);
ctrl_ext |= E1000_CTRL_EXT_EE_RST;
ew32(CTRL_EXT, ctrl_ext);
diff --git a/drivers/net/ethernet/intel/e1000e/param.c b/drivers/net/ethernet/intel/e1000e/param.c
index 98da75d..c16bd75 100644
--- a/drivers/net/ethernet/intel/e1000e/param.c
+++ b/drivers/net/ethernet/intel/e1000e/param.c
@@ -45,7 +45,7 @@
unsigned int copybreak = COPYBREAK_DEFAULT;
module_param(copybreak, uint, 0644);
MODULE_PARM_DESC(copybreak,
- "Maximum size of packet that is copied to a new buffer on receive");
+ "Maximum size of packet that is copied to a new buffer on receive");
/* All parameters are treated the same, as an integer array of values.
* This macro just reduces the need to repeat the same declaration code
@@ -143,7 +143,8 @@
*
* Default Value: 1 (enabled)
*/
-E1000_PARAM(WriteProtectNVM, "Write-protect NVM [WARNING: disabling this can lead to corrupted NVM]");
+E1000_PARAM(WriteProtectNVM,
+ "Write-protect NVM [WARNING: disabling this can lead to corrupted NVM]");
/* Enable CRC Stripping
*
@@ -160,13 +161,18 @@
const char *err;
int def;
union {
- struct { /* range_option info */
+ /* range_option info */
+ struct {
int min;
int max;
} r;
- struct { /* list_option info */
+ /* list_option info */
+ struct {
int nr;
- struct e1000_opt_list { int i; char *str; } *p;
+ struct e1000_opt_list {
+ int i;
+ char *str;
+ } *p;
} l;
} arg;
};
@@ -246,7 +252,8 @@
"Using defaults for all values\n");
}
- { /* Transmit Interrupt Delay */
+ /* Transmit Interrupt Delay */
+ {
static const struct e1000_option opt = {
.type = range_option,
.name = "Transmit Interrupt Delay",
@@ -265,7 +272,8 @@
adapter->tx_int_delay = opt.def;
}
}
- { /* Transmit Absolute Interrupt Delay */
+ /* Transmit Absolute Interrupt Delay */
+ {
static const struct e1000_option opt = {
.type = range_option,
.name = "Transmit Absolute Interrupt Delay",
@@ -284,7 +292,8 @@
adapter->tx_abs_int_delay = opt.def;
}
}
- { /* Receive Interrupt Delay */
+ /* Receive Interrupt Delay */
+ {
static struct e1000_option opt = {
.type = range_option,
.name = "Receive Interrupt Delay",
@@ -303,7 +312,8 @@
adapter->rx_int_delay = opt.def;
}
}
- { /* Receive Absolute Interrupt Delay */
+ /* Receive Absolute Interrupt Delay */
+ {
static const struct e1000_option opt = {
.type = range_option,
.name = "Receive Absolute Interrupt Delay",
@@ -322,7 +332,8 @@
adapter->rx_abs_int_delay = opt.def;
}
}
- { /* Interrupt Throttling Rate */
+ /* Interrupt Throttling Rate */
+ {
static const struct e1000_option opt = {
.type = range_option,
.name = "Interrupt Throttling Rate (ints/sec)",
@@ -392,7 +403,8 @@
break;
}
}
- { /* Interrupt Mode */
+ /* Interrupt Mode */
+ {
static struct e1000_option opt = {
.type = range_option,
.name = "Interrupt Mode",
@@ -435,7 +447,8 @@
kfree(opt.err);
#endif
}
- { /* Smart Power Down */
+ /* Smart Power Down */
+ {
static const struct e1000_option opt = {
.type = enable_option,
.name = "PHY Smart Power Down",
@@ -450,7 +463,8 @@
adapter->flags |= FLAG_SMART_POWER_DOWN;
}
}
- { /* CRC Stripping */
+ /* CRC Stripping */
+ {
static const struct e1000_option opt = {
.type = enable_option,
.name = "CRC Stripping",
@@ -470,27 +484,28 @@
adapter->flags2 |= FLAG2_DFLT_CRC_STRIPPING;
}
}
- { /* Kumeran Lock Loss Workaround */
+ /* Kumeran Lock Loss Workaround */
+ {
static const struct e1000_option opt = {
.type = enable_option,
.name = "Kumeran Lock Loss Workaround",
.err = "defaulting to Enabled",
.def = OPTION_ENABLED
};
+ bool enabled = opt.def;
if (num_KumeranLockLoss > bd) {
unsigned int kmrn_lock_loss = KumeranLockLoss[bd];
e1000_validate_option(&kmrn_lock_loss, &opt, adapter);
- if (hw->mac.type == e1000_ich8lan)
- e1000e_set_kmrn_lock_loss_workaround_ich8lan(hw,
- kmrn_lock_loss);
- } else {
- if (hw->mac.type == e1000_ich8lan)
- e1000e_set_kmrn_lock_loss_workaround_ich8lan(hw,
- opt.def);
+ enabled = kmrn_lock_loss;
}
+
+ if (hw->mac.type == e1000_ich8lan)
+ e1000e_set_kmrn_lock_loss_workaround_ich8lan(hw,
+ enabled);
}
- { /* Write-protect NVM */
+ /* Write-protect NVM */
+ {
static const struct e1000_option opt = {
.type = enable_option,
.name = "Write-protect NVM",
@@ -500,7 +515,8 @@
if (adapter->flags & FLAG_IS_ICH) {
if (num_WriteProtectNVM > bd) {
- unsigned int write_protect_nvm = WriteProtectNVM[bd];
+ unsigned int write_protect_nvm =
+ WriteProtectNVM[bd];
e1000_validate_option(&write_protect_nvm, &opt,
adapter);
if (write_protect_nvm)
diff --git a/drivers/net/ethernet/intel/e1000e/phy.c b/drivers/net/ethernet/intel/e1000e/phy.c
index 0930c13..60dbf02 100644
--- a/drivers/net/ethernet/intel/e1000e/phy.c
+++ b/drivers/net/ethernet/intel/e1000e/phy.c
@@ -37,7 +37,9 @@
/* Cable length tables */
static const u16 e1000_m88_cable_length_table[] = {
- 0, 50, 80, 110, 140, 140, E1000_CABLE_LENGTH_UNDEFINED };
+ 0, 50, 80, 110, 140, 140, E1000_CABLE_LENGTH_UNDEFINED
+};
+
#define M88E1000_CABLE_LENGTH_TABLE_SIZE \
ARRAY_SIZE(e1000_m88_cable_length_table)
@@ -49,7 +51,9 @@
66, 70, 75, 79, 83, 87, 91, 94, 98, 101, 104, 60, 66, 72, 77, 82,
87, 92, 96, 100, 104, 108, 111, 114, 117, 119, 121, 83, 89, 95,
100, 105, 109, 113, 116, 119, 122, 124, 104, 109, 114, 118, 121,
- 124};
+ 124
+};
+
#define IGP02E1000_CABLE_LENGTH_TABLE_SIZE \
ARRAY_SIZE(e1000_igp_2_cable_length_table)
@@ -67,8 +71,7 @@
manc = er32(MANC);
- return (manc & E1000_MANC_BLK_PHY_RST_ON_IDE) ?
- E1000_BLK_PHY_RESET : 0;
+ return (manc & E1000_MANC_BLK_PHY_RST_ON_IDE) ? E1000_BLK_PHY_RESET : 0;
}
/**
@@ -94,7 +97,7 @@
return ret_val;
phy->id = (u32)(phy_id << 16);
- udelay(20);
+ usleep_range(20, 40);
ret_val = e1e_rphy(hw, MII_PHYSID2, &phy_id);
if (ret_val)
return ret_val;
@@ -162,7 +165,7 @@
* the lower time out
*/
for (i = 0; i < (E1000_GEN_POLL_TIMEOUT * 3); i++) {
- udelay(50);
+ usleep_range(50, 100);
mdic = er32(MDIC);
if (mdic & E1000_MDIC_READY)
break;
@@ -175,13 +178,13 @@
e_dbg("MDI Error\n");
return -E1000_ERR_PHY;
}
- *data = (u16) mdic;
+ *data = (u16)mdic;
/* Allow some time after each MDIC transaction to avoid
* reading duplicate data in the next MDIC transaction.
*/
if (hw->mac.type == e1000_pch2lan)
- udelay(100);
+ usleep_range(100, 200);
return 0;
}
@@ -220,7 +223,7 @@
* the lower time out
*/
for (i = 0; i < (E1000_GEN_POLL_TIMEOUT * 3); i++) {
- udelay(50);
+ usleep_range(50, 100);
mdic = er32(MDIC);
if (mdic & E1000_MDIC_READY)
break;
@@ -238,7 +241,7 @@
* reading duplicate data in the next MDIC transaction.
*/
if (hw->mac.type == e1000_pch2lan)
- udelay(100);
+ usleep_range(100, 200);
return 0;
}
@@ -324,7 +327,7 @@
* semaphores before exiting.
**/
static s32 __e1000e_read_phy_reg_igp(struct e1000_hw *hw, u32 offset, u16 *data,
- bool locked)
+ bool locked)
{
s32 ret_val = 0;
@@ -391,7 +394,7 @@
* at the offset. Release any acquired semaphores before exiting.
**/
static s32 __e1000e_write_phy_reg_igp(struct e1000_hw *hw, u32 offset, u16 data,
- bool locked)
+ bool locked)
{
s32 ret_val = 0;
@@ -410,8 +413,7 @@
(u16)offset);
if (!ret_val)
ret_val = e1000e_write_phy_reg_mdic(hw, MAX_PHY_REG_ADDRESS &
- offset,
- data);
+ offset, data);
if (!locked)
hw->phy.ops.release(hw);
@@ -458,7 +460,7 @@
* Release any acquired semaphores before exiting.
**/
static s32 __e1000_read_kmrn_reg(struct e1000_hw *hw, u32 offset, u16 *data,
- bool locked)
+ bool locked)
{
u32 kmrnctrlsta;
@@ -531,7 +533,7 @@
* before exiting.
**/
static s32 __e1000_write_kmrn_reg(struct e1000_hw *hw, u32 offset, u16 data,
- bool locked)
+ bool locked)
{
u32 kmrnctrlsta;
@@ -772,8 +774,7 @@
phy_data |= M88E1000_EPSCR_TX_CLK_25;
- if ((phy->revision == 2) &&
- (phy->id == M88E1111_I_PHY_ID)) {
+ if ((phy->revision == 2) && (phy->id == M88E1111_I_PHY_ID)) {
/* 82573L PHY - set the downshift counter to 5x. */
phy_data &= ~M88EC018_EPSCR_DOWNSHIFT_COUNTER_MASK;
phy_data |= M88EC018_EPSCR_DOWNSHIFT_COUNTER_5X;
@@ -1296,7 +1297,7 @@
e_dbg("Waiting for forced speed/duplex link on M88 phy.\n");
ret_val = e1000e_phy_has_link_generic(hw, PHY_FORCE_LIMIT,
- 100000, &link);
+ 100000, &link);
if (ret_val)
return ret_val;
@@ -1319,7 +1320,7 @@
/* Try once more */
ret_val = e1000e_phy_has_link_generic(hw, PHY_FORCE_LIMIT,
- 100000, &link);
+ 100000, &link);
if (ret_val)
return ret_val;
}
@@ -1609,9 +1610,9 @@
ret_val = e1e_rphy(hw, M88E1000_PHY_SPEC_STATUS, &data);
if (!ret_val)
- phy->cable_polarity = (data & M88E1000_PSSR_REV_POLARITY)
- ? e1000_rev_polarity_reversed
- : e1000_rev_polarity_normal;
+ phy->cable_polarity = ((data & M88E1000_PSSR_REV_POLARITY)
+ ? e1000_rev_polarity_reversed
+ : e1000_rev_polarity_normal);
return ret_val;
}
@@ -1653,9 +1654,9 @@
ret_val = e1e_rphy(hw, offset, &data);
if (!ret_val)
- phy->cable_polarity = (data & mask)
- ? e1000_rev_polarity_reversed
- : e1000_rev_polarity_normal;
+ phy->cable_polarity = ((data & mask)
+ ? e1000_rev_polarity_reversed
+ : e1000_rev_polarity_normal);
return ret_val;
}
@@ -1685,9 +1686,9 @@
ret_val = e1e_rphy(hw, offset, &phy_data);
if (!ret_val)
- phy->cable_polarity = (phy_data & mask)
- ? e1000_rev_polarity_reversed
- : e1000_rev_polarity_normal;
+ phy->cable_polarity = ((phy_data & mask)
+ ? e1000_rev_polarity_reversed
+ : e1000_rev_polarity_normal);
return ret_val;
}
@@ -1733,7 +1734,7 @@
* Polls the PHY status register for link, 'iterations' number of times.
**/
s32 e1000e_phy_has_link_generic(struct e1000_hw *hw, u32 iterations,
- u32 usec_interval, bool *success)
+ u32 usec_interval, bool *success)
{
s32 ret_val = 0;
u16 i, phy_status;
@@ -1756,7 +1757,7 @@
if (phy_status & BMSR_LSTATUS)
break;
if (usec_interval >= 1000)
- mdelay(usec_interval/1000);
+ mdelay(usec_interval / 1000);
else
udelay(usec_interval);
}
@@ -1791,8 +1792,8 @@
if (ret_val)
return ret_val;
- index = (phy_data & M88E1000_PSSR_CABLE_LENGTH) >>
- M88E1000_PSSR_CABLE_LENGTH_SHIFT;
+ index = ((phy_data & M88E1000_PSSR_CABLE_LENGTH) >>
+ M88E1000_PSSR_CABLE_LENGTH_SHIFT);
if (index >= M88E1000_CABLE_LENGTH_TABLE_SIZE - 1)
return -E1000_ERR_PHY;
@@ -1824,10 +1825,10 @@
u16 cur_agc_index, max_agc_index = 0;
u16 min_agc_index = IGP02E1000_CABLE_LENGTH_TABLE_SIZE - 1;
static const u16 agc_reg_array[IGP02E1000_PHY_CHANNEL_NUM] = {
- IGP02E1000_PHY_AGC_A,
- IGP02E1000_PHY_AGC_B,
- IGP02E1000_PHY_AGC_C,
- IGP02E1000_PHY_AGC_D
+ IGP02E1000_PHY_AGC_A,
+ IGP02E1000_PHY_AGC_B,
+ IGP02E1000_PHY_AGC_C,
+ IGP02E1000_PHY_AGC_D
};
/* Read the AGC registers for all channels */
@@ -1841,8 +1842,8 @@
* that can be put into the lookup table to obtain the
* approximate cable length.
*/
- cur_agc_index = (phy_data >> IGP02E1000_AGC_LENGTH_SHIFT) &
- IGP02E1000_AGC_LENGTH_MASK;
+ cur_agc_index = ((phy_data >> IGP02E1000_AGC_LENGTH_SHIFT) &
+ IGP02E1000_AGC_LENGTH_MASK);
/* Array index bound check. */
if ((cur_agc_index >= IGP02E1000_CABLE_LENGTH_TABLE_SIZE) ||
@@ -1865,8 +1866,8 @@
agc_value /= (IGP02E1000_PHY_CHANNEL_NUM - 2);
/* Calculate cable length with the error range of +/- 10 meters. */
- phy->min_cable_length = ((agc_value - IGP02E1000_AGC_RANGE) > 0) ?
- (agc_value - IGP02E1000_AGC_RANGE) : 0;
+ phy->min_cable_length = (((agc_value - IGP02E1000_AGC_RANGE) > 0) ?
+ (agc_value - IGP02E1000_AGC_RANGE) : 0);
phy->max_cable_length = agc_value + IGP02E1000_AGC_RANGE;
phy->cable_length = (phy->min_cable_length + phy->max_cable_length) / 2;
@@ -2040,9 +2041,9 @@
return ret_val;
} else {
/* Polarity is forced */
- phy->cable_polarity = (data & IFE_PSC_FORCE_POLARITY)
- ? e1000_rev_polarity_reversed
- : e1000_rev_polarity_normal;
+ phy->cable_polarity = ((data & IFE_PSC_FORCE_POLARITY)
+ ? e1000_rev_polarity_reversed
+ : e1000_rev_polarity_normal);
}
ret_val = e1e_rphy(hw, IFE_PHY_MDIX_CONTROL, &data);
@@ -2119,7 +2120,7 @@
ew32(CTRL, ctrl);
e1e_flush();
- udelay(150);
+ usleep_range(150, 300);
phy->ops.release(hw);
@@ -2375,13 +2376,13 @@
/* Page is shifted left, PHY expects (page x 32) */
ret_val = e1000e_write_phy_reg_mdic(hw, page_select,
- (page << page_shift));
+ (page << page_shift));
if (ret_val)
goto release;
}
ret_val = e1000e_write_phy_reg_mdic(hw, MAX_PHY_REG_ADDRESS & offset,
- data);
+ data);
release:
hw->phy.ops.release(hw);
@@ -2433,13 +2434,13 @@
/* Page is shifted left, PHY expects (page x 32) */
ret_val = e1000e_write_phy_reg_mdic(hw, page_select,
- (page << page_shift));
+ (page << page_shift));
if (ret_val)
goto release;
}
ret_val = e1000e_read_phy_reg_mdic(hw, MAX_PHY_REG_ADDRESS & offset,
- data);
+ data);
release:
hw->phy.ops.release(hw);
return ret_val;
@@ -2674,7 +2675,7 @@
if (read) {
/* Read the Wakeup register page value using opcode 0x12 */
ret_val = e1000e_read_phy_reg_mdic(hw, BM_WUC_DATA_OPCODE,
- data);
+ data);
} else {
/* Write the Wakeup register page value using opcode 0x12 */
ret_val = e1000e_write_phy_reg_mdic(hw, BM_WUC_DATA_OPCODE,
@@ -2763,7 +2764,7 @@
if (page > 0 && page < HV_INTC_FC_PAGE_START) {
ret_val = e1000_access_phy_debug_regs_hv(hw, offset,
- data, true);
+ data, true);
goto out;
}
@@ -2786,8 +2787,7 @@
e_dbg("reading PHY page %d (or 0x%x shifted) reg 0x%x\n", page,
page << IGP_PAGE_SHIFT, reg);
- ret_val = e1000e_read_phy_reg_mdic(hw, MAX_PHY_REG_ADDRESS & reg,
- data);
+ ret_val = e1000e_read_phy_reg_mdic(hw, MAX_PHY_REG_ADDRESS & reg, data);
out:
if (!locked)
hw->phy.ops.release(hw);
@@ -2871,7 +2871,7 @@
if (page > 0 && page < HV_INTC_FC_PAGE_START) {
ret_val = e1000_access_phy_debug_regs_hv(hw, offset,
- &data, false);
+ &data, false);
goto out;
}
@@ -2910,7 +2910,7 @@
page << IGP_PAGE_SHIFT, reg);
ret_val = e1000e_write_phy_reg_mdic(hw, MAX_PHY_REG_ADDRESS & reg,
- data);
+ data);
out:
if (!locked)
@@ -2988,15 +2988,15 @@
* These accesses done with PHY address 2 and without using pages.
**/
static s32 e1000_access_phy_debug_regs_hv(struct e1000_hw *hw, u32 offset,
- u16 *data, bool read)
+ u16 *data, bool read)
{
s32 ret_val;
u32 addr_reg;
u32 data_reg;
/* This takes care of the difference with desktop vs mobile phy */
- addr_reg = (hw->phy.type == e1000_phy_82578) ?
- I82578_ADDR_REG : I82577_ADDR_REG;
+ addr_reg = ((hw->phy.type == e1000_phy_82578) ?
+ I82578_ADDR_REG : I82577_ADDR_REG);
data_reg = addr_reg + 1;
/* All operations in this function are phy address 2 */
@@ -3050,8 +3050,8 @@
if (ret_val)
return ret_val;
- data &= BM_CS_STATUS_LINK_UP | BM_CS_STATUS_RESOLVED |
- BM_CS_STATUS_SPEED_MASK;
+ data &= (BM_CS_STATUS_LINK_UP | BM_CS_STATUS_RESOLVED |
+ BM_CS_STATUS_SPEED_MASK);
if (data != (BM_CS_STATUS_LINK_UP | BM_CS_STATUS_RESOLVED |
BM_CS_STATUS_SPEED_1000))
@@ -3086,9 +3086,9 @@
ret_val = e1e_rphy(hw, I82577_PHY_STATUS_2, &data);
if (!ret_val)
- phy->cable_polarity = (data & I82577_PHY_STATUS2_REV_POLARITY)
- ? e1000_rev_polarity_reversed
- : e1000_rev_polarity_normal;
+ phy->cable_polarity = ((data & I82577_PHY_STATUS2_REV_POLARITY)
+ ? e1000_rev_polarity_reversed
+ : e1000_rev_polarity_normal);
return ret_val;
}
@@ -3215,8 +3215,8 @@
if (ret_val)
return ret_val;
- length = (phy_data & I82577_DSTATUS_CABLE_LENGTH) >>
- I82577_DSTATUS_CABLE_LENGTH_SHIFT;
+ length = ((phy_data & I82577_DSTATUS_CABLE_LENGTH) >>
+ I82577_DSTATUS_CABLE_LENGTH_SHIFT);
if (length == E1000_CABLE_LENGTH_UNDEFINED)
return -E1000_ERR_PHY;
diff --git a/drivers/net/ethernet/intel/igbvf/netdev.c b/drivers/net/ethernet/intel/igbvf/netdev.c
index d60cd43..bea46bb 100644
--- a/drivers/net/ethernet/intel/igbvf/netdev.c
+++ b/drivers/net/ethernet/intel/igbvf/netdev.c
@@ -447,7 +447,6 @@
tx_ring->desc = dma_alloc_coherent(&pdev->dev, tx_ring->size,
&tx_ring->dma, GFP_KERNEL);
-
if (!tx_ring->desc)
goto err;
@@ -488,7 +487,6 @@
rx_ring->desc = dma_alloc_coherent(&pdev->dev, rx_ring->size,
&rx_ring->dma, GFP_KERNEL);
-
if (!rx_ring->desc)
goto err;
diff --git a/drivers/net/ethernet/intel/ixgb/ixgb_main.c b/drivers/net/ethernet/intel/ixgb/ixgb_main.c
index ea48083..74464c3 100644
--- a/drivers/net/ethernet/intel/ixgb/ixgb_main.c
+++ b/drivers/net/ethernet/intel/ixgb/ixgb_main.c
@@ -717,14 +717,11 @@
txdr->size = ALIGN(txdr->size, 4096);
txdr->desc = dma_alloc_coherent(&pdev->dev, txdr->size, &txdr->dma,
- GFP_KERNEL);
+ GFP_KERNEL | __GFP_ZERO);
if (!txdr->desc) {
vfree(txdr->buffer_info);
- netif_err(adapter, probe, adapter->netdev,
- "Unable to allocate transmit descriptor memory\n");
return -ENOMEM;
}
- memset(txdr->desc, 0, txdr->size);
txdr->next_to_use = 0;
txdr->next_to_clean = 0;
@@ -807,8 +804,6 @@
if (!rxdr->desc) {
vfree(rxdr->buffer_info);
- netif_err(adapter, probe, adapter->netdev,
- "Unable to allocate receive descriptors\n");
return -ENOMEM;
}
memset(rxdr->desc, 0, rxdr->size);
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index db5611a..e56a3d1 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -7007,7 +7007,7 @@
int err;
if (!(adapter->flags & IXGBE_FLAG_SRIOV_ENABLED))
- return -EOPNOTSUPP;
+ return ndo_dflt_fdb_add(ndm, tb, dev, addr, flags);
/* Hardware does not support aging addresses so if a
* ndm_state is given only allow permanent addresses
@@ -7038,44 +7038,6 @@
return err;
}
-static int ixgbe_ndo_fdb_del(struct ndmsg *ndm, struct nlattr *tb[],
- struct net_device *dev,
- const unsigned char *addr)
-{
- struct ixgbe_adapter *adapter = netdev_priv(dev);
- int err = -EOPNOTSUPP;
-
- if (ndm->ndm_state & NUD_PERMANENT) {
- pr_info("%s: FDB only supports static addresses\n",
- ixgbe_driver_name);
- return -EINVAL;
- }
-
- if (adapter->flags & IXGBE_FLAG_SRIOV_ENABLED) {
- if (is_unicast_ether_addr(addr))
- err = dev_uc_del(dev, addr);
- else if (is_multicast_ether_addr(addr))
- err = dev_mc_del(dev, addr);
- else
- err = -EINVAL;
- }
-
- return err;
-}
-
-static int ixgbe_ndo_fdb_dump(struct sk_buff *skb,
- struct netlink_callback *cb,
- struct net_device *dev,
- int idx)
-{
- struct ixgbe_adapter *adapter = netdev_priv(dev);
-
- if (adapter->flags & IXGBE_FLAG_SRIOV_ENABLED)
- idx = ndo_dflt_fdb_dump(skb, cb, dev, idx);
-
- return idx;
-}
-
static int ixgbe_ndo_bridge_setlink(struct net_device *dev,
struct nlmsghdr *nlh)
{
@@ -7171,8 +7133,6 @@
.ndo_set_features = ixgbe_set_features,
.ndo_fix_features = ixgbe_fix_features,
.ndo_fdb_add = ixgbe_ndo_fdb_add,
- .ndo_fdb_del = ixgbe_ndo_fdb_del,
- .ndo_fdb_dump = ixgbe_ndo_fdb_dump,
.ndo_bridge_setlink = ixgbe_ndo_bridge_setlink,
.ndo_bridge_getlink = ixgbe_ndo_bridge_getlink,
};
diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h b/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h
index fc0af9a..fff0d98 100644
--- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h
+++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h
@@ -44,8 +44,8 @@
struct sk_buff *skb;
dma_addr_t dma;
unsigned long time_stamp;
+ union ixgbe_adv_tx_desc *next_to_watch;
u16 length;
- u16 next_to_watch;
u16 mapped_as_page;
};
diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
index c3db6cd..ac0c315 100644
--- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
+++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
@@ -76,12 +76,9 @@
* { Vendor ID, Device ID, SubVendor ID, SubDevice ID,
* Class, Class Mask, private data (not used) }
*/
-static struct pci_device_id ixgbevf_pci_tbl[] = {
- {PCI_VDEVICE(INTEL, IXGBE_DEV_ID_82599_VF),
- board_82599_vf},
- {PCI_VDEVICE(INTEL, IXGBE_DEV_ID_X540_VF),
- board_X540_vf},
-
+static DEFINE_PCI_DEVICE_TABLE(ixgbevf_pci_tbl) = {
+ {PCI_VDEVICE(INTEL, IXGBE_DEV_ID_82599_VF), board_82599_vf },
+ {PCI_VDEVICE(INTEL, IXGBE_DEV_ID_X540_VF), board_X540_vf },
/* required last entry */
{0, }
};
@@ -190,28 +187,37 @@
struct ixgbevf_adapter *adapter = q_vector->adapter;
union ixgbe_adv_tx_desc *tx_desc, *eop_desc;
struct ixgbevf_tx_buffer *tx_buffer_info;
- unsigned int i, eop, count = 0;
+ unsigned int i, count = 0;
unsigned int total_bytes = 0, total_packets = 0;
if (test_bit(__IXGBEVF_DOWN, &adapter->state))
return true;
i = tx_ring->next_to_clean;
- eop = tx_ring->tx_buffer_info[i].next_to_watch;
- eop_desc = IXGBEVF_TX_DESC(tx_ring, eop);
+ tx_buffer_info = &tx_ring->tx_buffer_info[i];
+ eop_desc = tx_buffer_info->next_to_watch;
- while ((eop_desc->wb.status & cpu_to_le32(IXGBE_TXD_STAT_DD)) &&
- (count < tx_ring->count)) {
+ do {
bool cleaned = false;
- rmb(); /* read buffer_info after eop_desc */
- /* eop could change between read and DD-check */
- if (unlikely(eop != tx_ring->tx_buffer_info[i].next_to_watch))
- goto cont_loop;
+
+ /* if next_to_watch is not set then there is no work pending */
+ if (!eop_desc)
+ break;
+
+ /* prevent any other reads prior to eop_desc */
+ read_barrier_depends();
+
+ /* if DD is not set pending work has not been completed */
+ if (!(eop_desc->wb.status & cpu_to_le32(IXGBE_TXD_STAT_DD)))
+ break;
+
+ /* clear next_to_watch to prevent false hangs */
+ tx_buffer_info->next_to_watch = NULL;
+
for ( ; !cleaned; count++) {
struct sk_buff *skb;
tx_desc = IXGBEVF_TX_DESC(tx_ring, i);
- tx_buffer_info = &tx_ring->tx_buffer_info[i];
- cleaned = (i == eop);
+ cleaned = (tx_desc == eop_desc);
skb = tx_buffer_info->skb;
if (cleaned && skb) {
@@ -234,12 +240,12 @@
i++;
if (i == tx_ring->count)
i = 0;
+
+ tx_buffer_info = &tx_ring->tx_buffer_info[i];
}
-cont_loop:
- eop = tx_ring->tx_buffer_info[i].next_to_watch;
- eop_desc = IXGBEVF_TX_DESC(tx_ring, eop);
- }
+ eop_desc = tx_buffer_info->next_to_watch;
+ } while (count < tx_ring->count);
tx_ring->next_to_clean = i;
@@ -2417,9 +2423,6 @@
&rx_ring->dma, GFP_KERNEL);
if (!rx_ring->desc) {
- hw_dbg(&adapter->hw,
- "Unable to allocate memory for "
- "the receive descriptor ring\n");
vfree(rx_ring->rx_buffer_info);
rx_ring->rx_buffer_info = NULL;
goto alloc_failed;
@@ -2806,8 +2809,7 @@
}
static int ixgbevf_tx_map(struct ixgbevf_ring *tx_ring,
- struct sk_buff *skb, u32 tx_flags,
- unsigned int first)
+ struct sk_buff *skb, u32 tx_flags)
{
struct ixgbevf_tx_buffer *tx_buffer_info;
unsigned int len;
@@ -2832,7 +2834,6 @@
size, DMA_TO_DEVICE);
if (dma_mapping_error(tx_ring->dev, tx_buffer_info->dma))
goto dma_error;
- tx_buffer_info->next_to_watch = i;
len -= size;
total -= size;
@@ -2862,7 +2863,6 @@
tx_buffer_info->dma))
goto dma_error;
tx_buffer_info->mapped_as_page = true;
- tx_buffer_info->next_to_watch = i;
len -= size;
total -= size;
@@ -2881,8 +2881,6 @@
else
i = i - 1;
tx_ring->tx_buffer_info[i].skb = skb;
- tx_ring->tx_buffer_info[first].next_to_watch = i;
- tx_ring->tx_buffer_info[first].time_stamp = jiffies;
return count;
@@ -2891,7 +2889,6 @@
/* clear timestamp and dma mappings for failed tx_buffer_info map */
tx_buffer_info->dma = 0;
- tx_buffer_info->next_to_watch = 0;
count--;
/* clear timestamp and dma mappings for remaining portion of packet */
@@ -2908,7 +2905,8 @@
}
static void ixgbevf_tx_queue(struct ixgbevf_ring *tx_ring, int tx_flags,
- int count, u32 paylen, u8 hdr_len)
+ int count, unsigned int first, u32 paylen,
+ u8 hdr_len)
{
union ixgbe_adv_tx_desc *tx_desc = NULL;
struct ixgbevf_tx_buffer *tx_buffer_info;
@@ -2959,6 +2957,16 @@
tx_desc->read.cmd_type_len |= cpu_to_le32(txd_cmd);
+ tx_ring->tx_buffer_info[first].time_stamp = jiffies;
+
+ /* Force memory writes to complete before letting h/w
+ * know there are new descriptors to fetch. (Only
+ * applicable for weak-ordered memory model archs,
+ * such as IA-64).
+ */
+ wmb();
+
+ tx_ring->tx_buffer_info[first].next_to_watch = tx_desc;
tx_ring->next_to_use = i;
}
@@ -3050,15 +3058,8 @@
tx_flags |= IXGBE_TX_FLAGS_CSUM;
ixgbevf_tx_queue(tx_ring, tx_flags,
- ixgbevf_tx_map(tx_ring, skb, tx_flags, first),
- skb->len, hdr_len);
- /*
- * Force memory writes to complete before letting h/w
- * know there are new descriptors to fetch. (Only
- * applicable for weak-ordered memory model archs,
- * such as IA-64).
- */
- wmb();
+ ixgbevf_tx_map(tx_ring, skb, tx_flags),
+ first, skb->len, hdr_len);
writel(tx_ring->next_to_use, adapter->hw.hw_addr + tx_ring->tail);
diff --git a/drivers/net/ethernet/marvell/mv643xx_eth.c b/drivers/net/ethernet/marvell/mv643xx_eth.c
index 6562c73..d1ecf4b 100644
--- a/drivers/net/ethernet/marvell/mv643xx_eth.c
+++ b/drivers/net/ethernet/marvell/mv643xx_eth.c
@@ -20,6 +20,8 @@
* Copyright (C) 2007-2008 Marvell Semiconductor
* Lennert Buytenhek <buytenh@marvell.com>
*
+ * Copyright (C) 2013 Michael Stapelberg <michael@stapelberg.de>
+ *
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version 2
@@ -1523,6 +1525,34 @@
return 0;
}
+static void
+mv643xx_eth_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
+{
+ struct mv643xx_eth_private *mp = netdev_priv(dev);
+ wol->supported = 0;
+ wol->wolopts = 0;
+ if (mp->phy)
+ phy_ethtool_get_wol(mp->phy, wol);
+}
+
+static int
+mv643xx_eth_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
+{
+ struct mv643xx_eth_private *mp = netdev_priv(dev);
+ int err;
+
+ if (mp->phy == NULL)
+ return -EOPNOTSUPP;
+
+ err = phy_ethtool_set_wol(mp->phy, wol);
+ /* Given that mv643xx_eth works without the marvell-specific PHY driver,
+ * this debugging hint is useful to have.
+ */
+ if (err == -EOPNOTSUPP)
+ netdev_info(dev, "The PHY does not support set_wol, was CONFIG_MARVELL_PHY enabled?\n");
+ return err;
+}
+
static int
mv643xx_eth_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
{
@@ -1708,6 +1738,8 @@
.get_ethtool_stats = mv643xx_eth_get_ethtool_stats,
.get_sset_count = mv643xx_eth_get_sset_count,
.get_ts_info = ethtool_op_get_ts_info,
+ .get_wol = mv643xx_eth_get_wol,
+ .set_wol = mv643xx_eth_set_wol,
};
diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index cd345b8..e48261e 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -1969,13 +1969,8 @@
rxq->descs = dma_alloc_coherent(pp->dev->dev.parent,
rxq->size * MVNETA_DESC_ALIGNED_SIZE,
&rxq->descs_phys, GFP_KERNEL);
- if (rxq->descs == NULL) {
- netdev_err(pp->dev,
- "rxq=%d: Can't allocate %d bytes for %d RX descr\n",
- rxq->id, rxq->size * MVNETA_DESC_ALIGNED_SIZE,
- rxq->size);
+ if (rxq->descs == NULL)
return -ENOMEM;
- }
BUG_ON(rxq->descs !=
PTR_ALIGN(rxq->descs, MVNETA_CPU_D_CACHE_LINE_SIZE));
@@ -2029,13 +2024,8 @@
txq->descs = dma_alloc_coherent(pp->dev->dev.parent,
txq->size * MVNETA_DESC_ALIGNED_SIZE,
&txq->descs_phys, GFP_KERNEL);
- if (txq->descs == NULL) {
- netdev_err(pp->dev,
- "txQ=%d: Can't allocate %d bytes for %d TX descr\n",
- txq->id, txq->size * MVNETA_DESC_ALIGNED_SIZE,
- txq->size);
+ if (txq->descs == NULL)
return -ENOMEM;
- }
/* Make sure descriptor address is cache line size aligned */
BUG_ON(txq->descs !=
diff --git a/drivers/net/ethernet/marvell/pxa168_eth.c b/drivers/net/ethernet/marvell/pxa168_eth.c
index 037ed86..339bb32 100644
--- a/drivers/net/ethernet/marvell/pxa168_eth.c
+++ b/drivers/net/ethernet/marvell/pxa168_eth.c
@@ -584,12 +584,14 @@
*/
if (pep->htpr == NULL) {
pep->htpr = dma_alloc_coherent(pep->dev->dev.parent,
- HASH_ADDR_TABLE_SIZE,
- &pep->htpr_dma, GFP_KERNEL);
+ HASH_ADDR_TABLE_SIZE,
+ &pep->htpr_dma,
+ GFP_KERNEL | __GFP_ZERO);
if (pep->htpr == NULL)
return -ENOMEM;
+ } else {
+ memset(pep->htpr, 0, HASH_ADDR_TABLE_SIZE);
}
- memset(pep->htpr, 0, HASH_ADDR_TABLE_SIZE);
wrl(pep, HTPR, pep->htpr_dma);
return 0;
}
@@ -1023,13 +1025,11 @@
size = pep->rx_ring_size * sizeof(struct rx_desc);
pep->rx_desc_area_size = size;
pep->p_rx_desc_area = dma_alloc_coherent(pep->dev->dev.parent, size,
- &pep->rx_desc_dma, GFP_KERNEL);
- if (!pep->p_rx_desc_area) {
- printk(KERN_ERR "%s: Cannot alloc RX ring (size %d bytes)\n",
- dev->name, size);
+ &pep->rx_desc_dma,
+ GFP_KERNEL | __GFP_ZERO);
+ if (!pep->p_rx_desc_area)
goto out;
- }
- memset((void *)pep->p_rx_desc_area, 0, size);
+
/* initialize the next_desc_ptr links in the Rx descriptors ring */
p_rx_desc = pep->p_rx_desc_area;
for (i = 0; i < rx_desc_num; i++) {
@@ -1086,13 +1086,10 @@
size = pep->tx_ring_size * sizeof(struct tx_desc);
pep->tx_desc_area_size = size;
pep->p_tx_desc_area = dma_alloc_coherent(pep->dev->dev.parent, size,
- &pep->tx_desc_dma, GFP_KERNEL);
- if (!pep->p_tx_desc_area) {
- printk(KERN_ERR "%s: Cannot allocate Tx Ring (size %d bytes)\n",
- dev->name, size);
+ &pep->tx_desc_dma,
+ GFP_KERNEL | __GFP_ZERO);
+ if (!pep->p_tx_desc_area)
goto out;
- }
- memset((void *)pep->p_tx_desc_area, 0, pep->tx_desc_area_size);
/* Initialize the next_desc_ptr links in the Tx descriptors ring */
p_tx_desc = pep->p_tx_desc_area;
for (i = 0; i < tx_desc_num; i++) {
diff --git a/drivers/net/ethernet/mellanox/mlx4/cmd.c b/drivers/net/ethernet/mellanox/mlx4/cmd.c
index fdc5f23..05267d7 100644
--- a/drivers/net/ethernet/mellanox/mlx4/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx4/cmd.c
@@ -1837,10 +1837,8 @@
priv->mfunc.vhcr = dma_alloc_coherent(&(dev->pdev->dev), PAGE_SIZE,
&priv->mfunc.vhcr_dma,
GFP_KERNEL);
- if (!priv->mfunc.vhcr) {
- mlx4_err(dev, "Couldn't allocate VHCR.\n");
+ if (!priv->mfunc.vhcr)
goto err_hcr;
- }
}
priv->cmd.pool = pci_pool_create("mlx4_cmd", dev->pdev,
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index 995d4b6..47b9960 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -1931,79 +1931,6 @@
}
-static int mlx4_en_fdb_add(struct ndmsg *ndm, struct nlattr *tb[],
- struct net_device *dev,
- const unsigned char *addr, u16 flags)
-{
- struct mlx4_en_priv *priv = netdev_priv(dev);
- struct mlx4_dev *mdev = priv->mdev->dev;
- int err;
-
- if (!mlx4_is_mfunc(mdev))
- return -EOPNOTSUPP;
-
- /* Hardware does not support aging addresses, allow only
- * permanent addresses if ndm_state is given
- */
- if (ndm->ndm_state && !(ndm->ndm_state & NUD_PERMANENT)) {
- en_info(priv, "Add FDB only supports static addresses\n");
- return -EINVAL;
- }
-
- if (is_unicast_ether_addr(addr) || is_link_local_ether_addr(addr))
- err = dev_uc_add_excl(dev, addr);
- else if (is_multicast_ether_addr(addr))
- err = dev_mc_add_excl(dev, addr);
- else
- err = -EINVAL;
-
- /* Only return duplicate errors if NLM_F_EXCL is set */
- if (err == -EEXIST && !(flags & NLM_F_EXCL))
- err = 0;
-
- return err;
-}
-
-static int mlx4_en_fdb_del(struct ndmsg *ndm,
- struct nlattr *tb[],
- struct net_device *dev,
- const unsigned char *addr)
-{
- struct mlx4_en_priv *priv = netdev_priv(dev);
- struct mlx4_dev *mdev = priv->mdev->dev;
- int err;
-
- if (!mlx4_is_mfunc(mdev))
- return -EOPNOTSUPP;
-
- if (ndm->ndm_state && !(ndm->ndm_state & NUD_PERMANENT)) {
- en_info(priv, "Del FDB only supports static addresses\n");
- return -EINVAL;
- }
-
- if (is_unicast_ether_addr(addr) || is_link_local_ether_addr(addr))
- err = dev_uc_del(dev, addr);
- else if (is_multicast_ether_addr(addr))
- err = dev_mc_del(dev, addr);
- else
- err = -EINVAL;
-
- return err;
-}
-
-static int mlx4_en_fdb_dump(struct sk_buff *skb,
- struct netlink_callback *cb,
- struct net_device *dev, int idx)
-{
- struct mlx4_en_priv *priv = netdev_priv(dev);
- struct mlx4_dev *mdev = priv->mdev->dev;
-
- if (mlx4_is_mfunc(mdev))
- idx = ndo_dflt_fdb_dump(skb, cb, dev, idx);
-
- return idx;
-}
-
static const struct net_device_ops mlx4_netdev_ops = {
.ndo_open = mlx4_en_open,
.ndo_stop = mlx4_en_close,
@@ -2025,9 +1952,6 @@
#ifdef CONFIG_RFS_ACCEL
.ndo_rx_flow_steer = mlx4_en_filter_rfs,
#endif
- .ndo_fdb_add = mlx4_en_fdb_add,
- .ndo_fdb_del = mlx4_en_fdb_del,
- .ndo_fdb_dump = mlx4_en_fdb_dump,
};
int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port,
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_selftest.c b/drivers/net/ethernet/mellanox/mlx4/en_selftest.c
index 3488c6d..2448f0d 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_selftest.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_selftest.c
@@ -58,10 +58,9 @@
/* build the pkt before xmit */
skb = netdev_alloc_skb(priv->dev, MLX4_LOOPBACK_TEST_PAYLOAD + ETH_HLEN + NET_IP_ALIGN);
- if (!skb) {
- en_err(priv, "-LOOPBACK_TEST_XMIT- failed to create skb for xmit\n");
+ if (!skb)
return -ENOMEM;
- }
+
skb_reserve(skb, NET_IP_ALIGN);
ethh = (struct ethhdr *)skb_put(skb, sizeof(struct ethhdr));
diff --git a/drivers/net/ethernet/micrel/ks8851_mll.c b/drivers/net/ethernet/micrel/ks8851_mll.c
index a343066..ddaf138 100644
--- a/drivers/net/ethernet/micrel/ks8851_mll.c
+++ b/drivers/net/ethernet/micrel/ks8851_mll.c
@@ -792,20 +792,35 @@
frame_hdr = ks->frame_head_info;
while (ks->frame_cnt--) {
+ if (unlikely(!(frame_hdr->sts & RXFSHR_RXFV) ||
+ frame_hdr->len >= RX_BUF_SIZE ||
+ frame_hdr->len <= 0)) {
+
+ /* discard an invalid packet */
+ ks_wrreg16(ks, KS_RXQCR, (ks->rc_rxqcr | RXQCR_RRXEF));
+ netdev->stats.rx_dropped++;
+ if (!(frame_hdr->sts & RXFSHR_RXFV))
+ netdev->stats.rx_frame_errors++;
+ else
+ netdev->stats.rx_length_errors++;
+ frame_hdr++;
+ continue;
+ }
+
skb = netdev_alloc_skb(netdev, frame_hdr->len + 16);
- if (likely(skb && (frame_hdr->sts & RXFSHR_RXFV) &&
- (frame_hdr->len < RX_BUF_SIZE) && frame_hdr->len)) {
+ if (likely(skb)) {
skb_reserve(skb, 2);
/* read data block including CRC 4 bytes */
ks_read_qmu(ks, (u16 *)skb->data, frame_hdr->len);
- skb_put(skb, frame_hdr->len);
+ skb_put(skb, frame_hdr->len - 4);
skb->protocol = eth_type_trans(skb, netdev);
netif_rx(skb);
+ /* exclude CRC size */
+ netdev->stats.rx_bytes += frame_hdr->len - 4;
+ netdev->stats.rx_packets++;
} else {
- pr_err("%s: err:skb alloc\n", __func__);
ks_wrreg16(ks, KS_RXQCR, (ks->rc_rxqcr | RXQCR_RRXEF));
- if (skb)
- dev_kfree_skb_irq(skb);
+ netdev->stats.rx_dropped++;
}
frame_hdr++;
}
@@ -877,6 +892,8 @@
ks_wrreg16(ks, KS_PMECR, pmecr | PMECR_WKEVT_LINK);
}
+ if (unlikely(status & IRQ_RXOI))
+ ks->netdev->stats.rx_over_errors++;
/* this should be the last in IRQ handler*/
ks_restore_cmd_reg(ks);
return IRQ_HANDLED;
@@ -1015,6 +1032,9 @@
if (likely(ks_tx_fifo_space(ks) >= skb->len + 12)) {
ks_write_qmu(ks, skb->data, skb->len);
+ /* add tx statistics */
+ netdev->stats.tx_bytes += skb->len;
+ netdev->stats.tx_packets++;
dev_kfree_skb(skb);
} else
retv = NETDEV_TX_BUSY;
diff --git a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
index 4f9937e..d5ffdc8 100644
--- a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
+++ b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
@@ -3592,10 +3592,9 @@
bytes = mgp->max_intr_slots * sizeof(*ss->rx_done.entry);
ss->rx_done.entry = dma_alloc_coherent(&pdev->dev, bytes,
&ss->rx_done.bus,
- GFP_KERNEL);
+ GFP_KERNEL | __GFP_ZERO);
if (ss->rx_done.entry == NULL)
goto abort;
- memset(ss->rx_done.entry, 0, bytes);
bytes = sizeof(*ss->fw_stats);
ss->fw_stats = dma_alloc_coherent(&pdev->dev, bytes,
&ss->fw_stats_bus,
diff --git a/drivers/net/ethernet/natsemi/jazzsonic.c b/drivers/net/ethernet/natsemi/jazzsonic.c
index b0b3615..c20766c 100644
--- a/drivers/net/ethernet/natsemi/jazzsonic.c
+++ b/drivers/net/ethernet/natsemi/jazzsonic.c
@@ -175,13 +175,13 @@
/* Allocate the entire chunk of memory for the descriptors.
Note that this cannot cross a 64K boundary. */
- if ((lp->descriptors = dma_alloc_coherent(lp->device,
- SIZEOF_SONIC_DESC * SONIC_BUS_SCALE(lp->dma_bitmode),
- &lp->descriptors_laddr, GFP_KERNEL)) == NULL) {
- printk(KERN_ERR "%s: couldn't alloc DMA memory for descriptors.\n",
- dev_name(lp->device));
+ lp->descriptors = dma_alloc_coherent(lp->device,
+ SIZEOF_SONIC_DESC *
+ SONIC_BUS_SCALE(lp->dma_bitmode),
+ &lp->descriptors_laddr,
+ GFP_KERNEL);
+ if (lp->descriptors == NULL)
goto out;
- }
/* Now set up the pointers to point to the appropriate places */
lp->cda = lp->descriptors;
diff --git a/drivers/net/ethernet/natsemi/macsonic.c b/drivers/net/ethernet/natsemi/macsonic.c
index 0ffde69..346a4e0 100644
--- a/drivers/net/ethernet/natsemi/macsonic.c
+++ b/drivers/net/ethernet/natsemi/macsonic.c
@@ -202,13 +202,13 @@
/* Allocate the entire chunk of memory for the descriptors.
Note that this cannot cross a 64K boundary. */
- if ((lp->descriptors = dma_alloc_coherent(lp->device,
- SIZEOF_SONIC_DESC * SONIC_BUS_SCALE(lp->dma_bitmode),
- &lp->descriptors_laddr, GFP_KERNEL)) == NULL) {
- printk(KERN_ERR "%s: couldn't alloc DMA memory for descriptors.\n",
- dev_name(lp->device));
+ lp->descriptors = dma_alloc_coherent(lp->device,
+ SIZEOF_SONIC_DESC *
+ SONIC_BUS_SCALE(lp->dma_bitmode),
+ &lp->descriptors_laddr,
+ GFP_KERNEL);
+ if (lp->descriptors == NULL)
return -ENOMEM;
- }
/* Now set up the pointers to point to the appropriate places */
lp->cda = lp->descriptors;
diff --git a/drivers/net/ethernet/natsemi/sonic.c b/drivers/net/ethernet/natsemi/sonic.c
index 46795e4..1bd419d 100644
--- a/drivers/net/ethernet/natsemi/sonic.c
+++ b/drivers/net/ethernet/natsemi/sonic.c
@@ -424,7 +424,6 @@
/* Malloc up new buffer. */
new_skb = netdev_alloc_skb(dev, SONIC_RBSIZE + 2);
if (new_skb == NULL) {
- printk(KERN_ERR "%s: Memory squeeze, dropping packet.\n", dev->name);
lp->stats.rx_dropped++;
break;
}
diff --git a/drivers/net/ethernet/natsemi/xtsonic.c b/drivers/net/ethernet/natsemi/xtsonic.c
index 5e4748e..c2e0256 100644
--- a/drivers/net/ethernet/natsemi/xtsonic.c
+++ b/drivers/net/ethernet/natsemi/xtsonic.c
@@ -197,14 +197,12 @@
* We also allocate extra space for a pointer to allow freeing
* this structure later on (in xtsonic_cleanup_module()).
*/
- lp->descriptors =
- dma_alloc_coherent(lp->device,
- SIZEOF_SONIC_DESC * SONIC_BUS_SCALE(lp->dma_bitmode),
- &lp->descriptors_laddr, GFP_KERNEL);
-
+ lp->descriptors = dma_alloc_coherent(lp->device,
+ SIZEOF_SONIC_DESC *
+ SONIC_BUS_SCALE(lp->dma_bitmode),
+ &lp->descriptors_laddr,
+ GFP_KERNEL);
if (lp->descriptors == NULL) {
- printk(KERN_ERR "%s: couldn't alloc DMA memory for "
- " descriptors.\n", dev_name(lp->device));
err = -ENOMEM;
goto out;
}
diff --git a/drivers/net/ethernet/neterion/s2io.c b/drivers/net/ethernet/neterion/s2io.c
index bfd8873..3371ff4 100644
--- a/drivers/net/ethernet/neterion/s2io.c
+++ b/drivers/net/ethernet/neterion/s2io.c
@@ -80,6 +80,7 @@
#include <linux/slab.h>
#include <linux/prefetch.h>
#include <net/tcp.h>
+#include <net/checksum.h>
#include <asm/div64.h>
#include <asm/irq.h>
@@ -8337,16 +8338,13 @@
{
struct iphdr *ip = lro->iph;
struct tcphdr *tcp = lro->tcph;
- __sum16 nchk;
struct swStat *swstats = &sp->mac_control.stats_info->sw_stat;
DBG_PRINT(INFO_DBG, "%s: Been here...\n", __func__);
/* Update L3 header */
+ csum_replace2(&ip->check, ip->tot_len, htons(lro->total_len));
ip->tot_len = htons(lro->total_len);
- ip->check = 0;
- nchk = ip_fast_csum((u8 *)lro->iph, ip->ihl);
- ip->check = nchk;
/* Update L4 header */
tcp->ack_seq = lro->tcp_ack;
diff --git a/drivers/net/ethernet/netx-eth.c b/drivers/net/ethernet/netx-eth.c
index 63e7af4..cb9e638 100644
--- a/drivers/net/ethernet/netx-eth.c
+++ b/drivers/net/ethernet/netx-eth.c
@@ -152,8 +152,6 @@
skb = netdev_alloc_skb(ndev, len);
if (unlikely(skb == NULL)) {
- printk(KERN_NOTICE "%s: Low memory, packet dropped.\n",
- ndev->name);
ndev->stats.rx_dropped++;
return;
}
diff --git a/drivers/net/ethernet/nuvoton/w90p910_ether.c b/drivers/net/ethernet/nuvoton/w90p910_ether.c
index 162da89..3df8287 100644
--- a/drivers/net/ethernet/nuvoton/w90p910_ether.c
+++ b/drivers/net/ethernet/nuvoton/w90p910_ether.c
@@ -287,23 +287,16 @@
ether = netdev_priv(dev);
pdev = ether->pdev;
- ether->tdesc = (struct tran_pdesc *)
- dma_alloc_coherent(&pdev->dev, sizeof(struct tran_pdesc),
- ðer->tdesc_phys, GFP_KERNEL);
-
- if (!ether->tdesc) {
- dev_err(&pdev->dev, "Failed to allocate memory for tx desc\n");
+ ether->tdesc = dma_alloc_coherent(&pdev->dev, sizeof(struct tran_pdesc),
+ ðer->tdesc_phys, GFP_KERNEL);
+ if (!ether->tdesc)
return -ENOMEM;
- }
- ether->rdesc = (struct recv_pdesc *)
- dma_alloc_coherent(&pdev->dev, sizeof(struct recv_pdesc),
- ðer->rdesc_phys, GFP_KERNEL);
-
+ ether->rdesc = dma_alloc_coherent(&pdev->dev, sizeof(struct recv_pdesc),
+ ðer->rdesc_phys, GFP_KERNEL);
if (!ether->rdesc) {
- dev_err(&pdev->dev, "Failed to allocate memory for rx desc\n");
dma_free_coherent(&pdev->dev, sizeof(struct tran_pdesc),
- ether->tdesc, ether->tdesc_phys);
+ ether->tdesc, ether->tdesc_phys);
return -ENOMEM;
}
@@ -737,7 +730,6 @@
data = ether->rdesc->recv_buf[ether->cur_rx];
skb = netdev_alloc_skb(dev, length + 2);
if (!skb) {
- dev_err(&pdev->dev, "get skb buffer error\n");
ether->stats.rx_dropped++;
return;
}
diff --git a/drivers/net/ethernet/nvidia/forcedeth.c b/drivers/net/ethernet/nvidia/forcedeth.c
index 0b8de12..b62262c 100644
--- a/drivers/net/ethernet/nvidia/forcedeth.c
+++ b/drivers/net/ethernet/nvidia/forcedeth.c
@@ -5025,7 +5025,6 @@
pkt_len = ETH_DATA_LEN;
tx_skb = netdev_alloc_skb(dev, pkt_len);
if (!tx_skb) {
- netdev_err(dev, "netdev_alloc_skb() failed during loopback test\n");
ret = 0;
goto out;
}
diff --git a/drivers/net/ethernet/nxp/lpc_eth.c b/drivers/net/ethernet/nxp/lpc_eth.c
index efa29b7..89d1b0e 100644
--- a/drivers/net/ethernet/nxp/lpc_eth.c
+++ b/drivers/net/ethernet/nxp/lpc_eth.c
@@ -1409,9 +1409,7 @@
dma_alloc_coherent(&pldat->pdev->dev,
pldat->dma_buff_size, &dma_handle,
GFP_KERNEL);
-
if (pldat->dma_buff_base_v == NULL) {
- dev_err(&pdev->dev, "error getting DMA region.\n");
ret = -ENOMEM;
goto err_out_free_irq;
}
diff --git a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c
index 39ab4d0..abd5fba 100644
--- a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c
+++ b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c
@@ -1469,13 +1469,11 @@
size = rx_ring->count * bufsz + PCH_GBE_RESERVE_MEMORY;
rx_ring->rx_buff_pool = dma_alloc_coherent(&pdev->dev, size,
- &rx_ring->rx_buff_pool_logic,
- GFP_KERNEL);
- if (!rx_ring->rx_buff_pool) {
- pr_err("Unable to allocate memory for the receive pool buffer\n");
+ &rx_ring->rx_buff_pool_logic,
+ GFP_KERNEL | __GFP_ZERO);
+ if (!rx_ring->rx_buff_pool)
return -ENOMEM;
- }
- memset(rx_ring->rx_buff_pool, 0, size);
+
rx_ring->rx_buff_pool_size = size;
for (i = 0; i < rx_ring->count; i++) {
buffer_info = &rx_ring->buffer_info[i];
@@ -1774,13 +1772,12 @@
tx_ring->size = tx_ring->count * (int)sizeof(struct pch_gbe_tx_desc);
tx_ring->desc = dma_alloc_coherent(&pdev->dev, tx_ring->size,
- &tx_ring->dma, GFP_KERNEL);
+ &tx_ring->dma,
+ GFP_KERNEL | __GFP_ZERO);
if (!tx_ring->desc) {
vfree(tx_ring->buffer_info);
- pr_err("Unable to allocate memory for the transmit descriptor ring\n");
return -ENOMEM;
}
- memset(tx_ring->desc, 0, tx_ring->size);
tx_ring->next_to_use = 0;
tx_ring->next_to_clean = 0;
@@ -1820,14 +1817,12 @@
rx_ring->size = rx_ring->count * (int)sizeof(struct pch_gbe_rx_desc);
rx_ring->desc = dma_alloc_coherent(&pdev->dev, rx_ring->size,
- &rx_ring->dma, GFP_KERNEL);
-
+ &rx_ring->dma,
+ GFP_KERNEL | __GFP_ZERO);
if (!rx_ring->desc) {
- pr_err("Unable to allocate memory for the receive descriptor ring\n");
vfree(rx_ring->buffer_info);
return -ENOMEM;
}
- memset(rx_ring->desc, 0, rx_ring->size);
rx_ring->next_to_clean = 0;
rx_ring->next_to_use = 0;
for (desNo = 0; desNo < rx_ring->count; desNo++) {
diff --git a/drivers/net/ethernet/pasemi/pasemi_mac.c b/drivers/net/ethernet/pasemi/pasemi_mac.c
index b1cfbb7..a5f0b5d 100644
--- a/drivers/net/ethernet/pasemi/pasemi_mac.c
+++ b/drivers/net/ethernet/pasemi/pasemi_mac.c
@@ -441,12 +441,11 @@
ring->buffers = dma_alloc_coherent(&mac->dma_pdev->dev,
RX_RING_SIZE * sizeof(u64),
- &ring->buf_dma, GFP_KERNEL);
+ &ring->buf_dma,
+ GFP_KERNEL | __GFP_ZERO);
if (!ring->buffers)
goto out_ring_desc;
- memset(ring->buffers, 0, RX_RING_SIZE * sizeof(u64));
-
write_dma_reg(PAS_DMA_RXCHAN_BASEL(chno),
PAS_DMA_RXCHAN_BASEL_BRBL(ring->chan.ring_dma));
diff --git a/drivers/net/ethernet/qlogic/netxen/netxen_nic.h b/drivers/net/ethernet/qlogic/netxen/netxen_nic.h
index eb3dfdb..322a36b 100644
--- a/drivers/net/ethernet/qlogic/netxen/netxen_nic.h
+++ b/drivers/net/ethernet/qlogic/netxen/netxen_nic.h
@@ -955,9 +955,10 @@
uint8_t mac_addr[ETH_ALEN+2];
} nx_mac_list_t;
-struct nx_vlan_ip_list {
+struct nx_ip_list {
struct list_head list;
__be32 ip_addr;
+ bool master;
};
/*
@@ -1605,7 +1606,7 @@
struct net_device *netdev;
struct pci_dev *pdev;
struct list_head mac_list;
- struct list_head vlan_ip_list;
+ struct list_head ip_list;
spinlock_t tx_clean_lock;
diff --git a/drivers/net/ethernet/qlogic/netxen/netxen_nic_init.c b/drivers/net/ethernet/qlogic/netxen/netxen_nic_init.c
index 4782dcf..7692dfd 100644
--- a/drivers/net/ethernet/qlogic/netxen/netxen_nic_init.c
+++ b/drivers/net/ethernet/qlogic/netxen/netxen_nic_init.c
@@ -27,6 +27,7 @@
#include <linux/delay.h>
#include <linux/slab.h>
#include <linux/if_vlan.h>
+#include <net/checksum.h>
#include "netxen_nic.h"
#include "netxen_nic_hw.h"
@@ -1641,9 +1642,8 @@
th = (struct tcphdr *)((skb->data + vhdr_len) + (iph->ihl << 2));
length = (iph->ihl << 2) + (th->doff << 2) + lro_length;
+ csum_replace2(&iph->check, iph->tot_len, htons(length));
iph->tot_len = htons(length);
- iph->check = 0;
- iph->check = ip_fast_csum((unsigned char *)iph, iph->ihl);
th->psh = push;
th->seq = htonl(seq_number);
diff --git a/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c b/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
index 501f492..7867aeb 100644
--- a/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
+++ b/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
@@ -90,7 +90,7 @@
static irqreturn_t netxen_msi_intr(int irq, void *data);
static irqreturn_t netxen_msix_intr(int irq, void *data);
-static void netxen_free_vlan_ip_list(struct netxen_adapter *);
+static void netxen_free_ip_list(struct netxen_adapter *, bool);
static void netxen_restore_indev_addr(struct net_device *dev, unsigned long);
static struct rtnl_link_stats64 *netxen_nic_get_stats(struct net_device *dev,
struct rtnl_link_stats64 *stats);
@@ -1450,7 +1450,7 @@
spin_lock_init(&adapter->tx_clean_lock);
INIT_LIST_HEAD(&adapter->mac_list);
- INIT_LIST_HEAD(&adapter->vlan_ip_list);
+ INIT_LIST_HEAD(&adapter->ip_list);
err = netxen_setup_pci_map(adapter);
if (err)
@@ -1585,7 +1585,7 @@
cancel_work_sync(&adapter->tx_timeout_task);
- netxen_free_vlan_ip_list(adapter);
+ netxen_free_ip_list(adapter, false);
netxen_nic_detach(adapter);
nx_decr_dev_ref_cnt(adapter);
@@ -3137,62 +3137,77 @@
}
static void
-netxen_free_vlan_ip_list(struct netxen_adapter *adapter)
+netxen_free_ip_list(struct netxen_adapter *adapter, bool master)
{
- struct nx_vlan_ip_list *cur;
- struct list_head *head = &adapter->vlan_ip_list;
+ struct nx_ip_list *cur, *tmp_cur;
- while (!list_empty(head)) {
- cur = list_entry(head->next, struct nx_vlan_ip_list, list);
- netxen_config_ipaddr(adapter, cur->ip_addr, NX_IP_DOWN);
- list_del(&cur->list);
- kfree(cur);
+ list_for_each_entry_safe(cur, tmp_cur, &adapter->ip_list, list) {
+ if (master) {
+ if (cur->master) {
+ netxen_config_ipaddr(adapter, cur->ip_addr,
+ NX_IP_DOWN);
+ list_del(&cur->list);
+ kfree(cur);
+ }
+ } else {
+ netxen_config_ipaddr(adapter, cur->ip_addr, NX_IP_DOWN);
+ list_del(&cur->list);
+ kfree(cur);
+ }
}
-
}
-static void
-netxen_list_config_vlan_ip(struct netxen_adapter *adapter,
+
+static bool
+netxen_list_config_ip(struct netxen_adapter *adapter,
struct in_ifaddr *ifa, unsigned long event)
{
struct net_device *dev;
- struct nx_vlan_ip_list *cur, *tmp_cur;
+ struct nx_ip_list *cur, *tmp_cur;
struct list_head *head;
+ bool ret = false;
dev = ifa->ifa_dev ? ifa->ifa_dev->dev : NULL;
if (dev == NULL)
- return;
-
- if (!is_vlan_dev(dev))
- return;
+ goto out;
switch (event) {
case NX_IP_UP:
- list_for_each(head, &adapter->vlan_ip_list) {
- cur = list_entry(head, struct nx_vlan_ip_list, list);
+ list_for_each(head, &adapter->ip_list) {
+ cur = list_entry(head, struct nx_ip_list, list);
if (cur->ip_addr == ifa->ifa_address)
- return;
+ goto out;
}
- cur = kzalloc(sizeof(struct nx_vlan_ip_list), GFP_ATOMIC);
+ cur = kzalloc(sizeof(struct nx_ip_list), GFP_ATOMIC);
if (cur == NULL)
- return;
-
+ goto out;
+ if (dev->priv_flags & IFF_802_1Q_VLAN)
+ dev = vlan_dev_real_dev(dev);
+ cur->master = !!netif_is_bond_master(dev);
cur->ip_addr = ifa->ifa_address;
- list_add_tail(&cur->list, &adapter->vlan_ip_list);
+ list_add_tail(&cur->list, &adapter->ip_list);
+ netxen_config_ipaddr(adapter, ifa->ifa_address, NX_IP_UP);
+ ret = true;
break;
case NX_IP_DOWN:
list_for_each_entry_safe(cur, tmp_cur,
- &adapter->vlan_ip_list, list) {
+ &adapter->ip_list, list) {
if (cur->ip_addr == ifa->ifa_address) {
list_del(&cur->list);
kfree(cur);
+ netxen_config_ipaddr(adapter, ifa->ifa_address,
+ NX_IP_DOWN);
+ ret = true;
break;
}
}
}
+out:
+ return ret;
}
+
static void
netxen_config_indev_addr(struct netxen_adapter *adapter,
struct net_device *dev, unsigned long event)
@@ -3209,14 +3224,10 @@
for_ifa(indev) {
switch (event) {
case NETDEV_UP:
- netxen_config_ipaddr(adapter,
- ifa->ifa_address, NX_IP_UP);
- netxen_list_config_vlan_ip(adapter, ifa, NX_IP_UP);
+ netxen_list_config_ip(adapter, ifa, NX_IP_UP);
break;
case NETDEV_DOWN:
- netxen_config_ipaddr(adapter,
- ifa->ifa_address, NX_IP_DOWN);
- netxen_list_config_vlan_ip(adapter, ifa, NX_IP_DOWN);
+ netxen_list_config_ip(adapter, ifa, NX_IP_DOWN);
break;
default:
break;
@@ -3231,23 +3242,78 @@
{
struct netxen_adapter *adapter = netdev_priv(netdev);
- struct nx_vlan_ip_list *pos, *tmp_pos;
+ struct nx_ip_list *pos, *tmp_pos;
unsigned long ip_event;
ip_event = (event == NETDEV_UP) ? NX_IP_UP : NX_IP_DOWN;
netxen_config_indev_addr(adapter, netdev, event);
- list_for_each_entry_safe(pos, tmp_pos, &adapter->vlan_ip_list, list) {
+ list_for_each_entry_safe(pos, tmp_pos, &adapter->ip_list, list) {
netxen_config_ipaddr(adapter, pos->ip_addr, ip_event);
}
}
+static inline bool
+netxen_config_checkdev(struct net_device *dev)
+{
+ struct netxen_adapter *adapter;
+
+ if (!is_netxen_netdev(dev))
+ return false;
+ adapter = netdev_priv(dev);
+ if (!adapter)
+ return false;
+ if (!netxen_destip_supported(adapter))
+ return false;
+ if (adapter->is_up != NETXEN_ADAPTER_UP_MAGIC)
+ return false;
+
+ return true;
+}
+
+/**
+ * netxen_config_master - configure addresses based on master
+ * @dev: netxen device
+ * @event: netdev event
+ */
+static void netxen_config_master(struct net_device *dev, unsigned long event)
+{
+ struct net_device *master, *slave;
+ struct netxen_adapter *adapter = netdev_priv(dev);
+
+ rcu_read_lock();
+ master = netdev_master_upper_dev_get_rcu(dev);
+ /*
+ * This is the case where the netxen nic is being
+ * enslaved and is dev_open()ed in bond_enslave()
+ * Now we should program the bond's (and its vlans')
+ * addresses in the netxen NIC.
+ */
+ if (master && netif_is_bond_master(master) &&
+ !netif_is_bond_slave(dev)) {
+ netxen_config_indev_addr(adapter, master, event);
+ for_each_netdev_rcu(&init_net, slave)
+ if (slave->priv_flags & IFF_802_1Q_VLAN &&
+ vlan_dev_real_dev(slave) == master)
+ netxen_config_indev_addr(adapter, slave, event);
+ }
+ rcu_read_unlock();
+ /*
+ * This is the case where the netxen nic is being
+ * released and is dev_close()ed in bond_release()
+ * just before IFF_BONDING is stripped.
+ */
+ if (!master && dev->priv_flags & IFF_BONDING)
+ netxen_free_ip_list(adapter, true);
+}
+
static int netxen_netdev_event(struct notifier_block *this,
unsigned long event, void *ptr)
{
struct netxen_adapter *adapter;
struct net_device *dev = (struct net_device *)ptr;
struct net_device *orig_dev = dev;
+ struct net_device *slave;
recheck:
if (dev == NULL)
@@ -3257,19 +3323,28 @@
dev = vlan_dev_real_dev(dev);
goto recheck;
}
-
- if (!is_netxen_netdev(dev))
- goto done;
-
- adapter = netdev_priv(dev);
-
- if (!adapter)
- goto done;
-
- if (adapter->is_up != NETXEN_ADAPTER_UP_MAGIC)
- goto done;
-
- netxen_config_indev_addr(adapter, orig_dev, event);
+ if (event == NETDEV_UP || event == NETDEV_DOWN) {
+ /* If this is a bonding device, look for netxen-based slaves*/
+ if (netif_is_bond_master(dev)) {
+ rcu_read_lock();
+ for_each_netdev_in_bond_rcu(dev, slave) {
+ if (!netxen_config_checkdev(slave))
+ continue;
+ adapter = netdev_priv(slave);
+ netxen_config_indev_addr(adapter,
+ orig_dev, event);
+ }
+ rcu_read_unlock();
+ } else {
+ if (!netxen_config_checkdev(dev))
+ goto done;
+ adapter = netdev_priv(dev);
+ /* Act only if the actual netxen is the target */
+ if (orig_dev == dev)
+ netxen_config_master(dev, event);
+ netxen_config_indev_addr(adapter, orig_dev, event);
+ }
+ }
done:
return NOTIFY_DONE;
}
@@ -3279,12 +3354,12 @@
unsigned long event, void *ptr)
{
struct netxen_adapter *adapter;
- struct net_device *dev;
-
+ struct net_device *dev, *slave;
struct in_ifaddr *ifa = (struct in_ifaddr *)ptr;
+ unsigned long ip_event;
dev = ifa->ifa_dev ? ifa->ifa_dev->dev : NULL;
-
+ ip_event = (event == NETDEV_UP) ? NX_IP_UP : NX_IP_DOWN;
recheck:
if (dev == NULL)
goto done;
@@ -3293,31 +3368,24 @@
dev = vlan_dev_real_dev(dev);
goto recheck;
}
-
- if (!is_netxen_netdev(dev))
- goto done;
-
- adapter = netdev_priv(dev);
-
- if (!adapter || !netxen_destip_supported(adapter))
- goto done;
-
- if (adapter->is_up != NETXEN_ADAPTER_UP_MAGIC)
- goto done;
-
- switch (event) {
- case NETDEV_UP:
- netxen_config_ipaddr(adapter, ifa->ifa_address, NX_IP_UP);
- netxen_list_config_vlan_ip(adapter, ifa, NX_IP_UP);
- break;
- case NETDEV_DOWN:
- netxen_config_ipaddr(adapter, ifa->ifa_address, NX_IP_DOWN);
- netxen_list_config_vlan_ip(adapter, ifa, NX_IP_DOWN);
- break;
- default:
- break;
+ if (event == NETDEV_UP || event == NETDEV_DOWN) {
+ /* If this is a bonding device, look for netxen-based slaves*/
+ if (netif_is_bond_master(dev)) {
+ rcu_read_lock();
+ for_each_netdev_in_bond_rcu(dev, slave) {
+ if (!netxen_config_checkdev(slave))
+ continue;
+ adapter = netdev_priv(slave);
+ netxen_list_config_ip(adapter, ifa, ip_event);
+ }
+ rcu_read_unlock();
+ } else {
+ if (!netxen_config_checkdev(dev))
+ goto done;
+ adapter = netdev_priv(dev);
+ netxen_list_config_ip(adapter, ifa, ip_event);
+ }
}
-
done:
return NOTIFY_DONE;
}
@@ -3334,7 +3402,7 @@
netxen_restore_indev_addr(struct net_device *dev, unsigned long event)
{ }
static void
-netxen_free_vlan_ip_list(struct netxen_adapter *adapter)
+netxen_free_ip_list(struct netxen_adapter *adapter, bool master)
{ }
#endif
diff --git a/drivers/net/ethernet/qlogic/qla3xxx.c b/drivers/net/ethernet/qlogic/qla3xxx.c
index 8fd38cb6..91a8fcd 100644
--- a/drivers/net/ethernet/qlogic/qla3xxx.c
+++ b/drivers/net/ethernet/qlogic/qla3xxx.c
@@ -312,7 +312,6 @@
lrg_buf_cb->skb = netdev_alloc_skb(qdev->ndev,
qdev->lrg_buffer_len);
if (unlikely(!lrg_buf_cb->skb)) {
- netdev_err(qdev->ndev, "failed netdev_alloc_skb()\n");
qdev->lrg_buf_skb_check++;
} else {
/*
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h
index ba3c72f..1577799 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h
@@ -38,8 +38,8 @@
#define _QLCNIC_LINUX_MAJOR 5
#define _QLCNIC_LINUX_MINOR 1
-#define _QLCNIC_LINUX_SUBVERSION 35
-#define QLCNIC_LINUX_VERSIONID "5.1.35"
+#define _QLCNIC_LINUX_SUBVERSION 37
+#define QLCNIC_LINUX_VERSIONID "5.1.37"
#define QLCNIC_DRV_IDC_VER 0x01
#define QLCNIC_DRIVER_VERSION ((_QLCNIC_LINUX_MAJOR << 16) |\
(_QLCNIC_LINUX_MINOR << 8) | (_QLCNIC_LINUX_SUBVERSION))
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c
index cd5ae88..56c3676 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c
@@ -15,36 +15,57 @@
#define RSS_HASHTYPE_IP_TCP 0x3
/* status descriptor mailbox data
- * @phy_addr: physical address of buffer
+ * @phy_addr_{low|high}: physical address of buffer
* @sds_ring_size: buffer size
* @intrpt_id: interrupt id
* @intrpt_val: source of interrupt
*/
struct qlcnic_sds_mbx {
- u64 phy_addr;
- u8 rsvd1[16];
+ u32 phy_addr_low;
+ u32 phy_addr_high;
+ u32 rsvd1[4];
+#if defined(__LITTLE_ENDIAN)
u16 sds_ring_size;
- u16 rsvd2[3];
+ u16 rsvd2;
+ u16 rsvd3[2];
u16 intrpt_id;
u8 intrpt_val;
- u8 rsvd3[5];
+ u8 rsvd4;
+#elif defined(__BIG_ENDIAN)
+ u16 rsvd2;
+ u16 sds_ring_size;
+ u16 rsvd3[2];
+ u8 rsvd4;
+ u8 intrpt_val;
+ u16 intrpt_id;
+#endif
+ u32 rsvd5;
} __packed;
/* receive descriptor buffer data
- * phy_addr_reg: physical address of regular buffer
- * phy_addr_jmb: physical address of jumbo buffer
+ * phy_addr_reg_{low|high}: physical address of regular buffer
+ * phy_addr_jmb_{low|high}: physical address of jumbo buffer
* reg_ring_sz: size of regular buffer
* reg_ring_len: no. of entries in regular buffer
* jmb_ring_len: no. of entries in jumbo buffer
* jmb_ring_sz: size of jumbo buffer
*/
struct qlcnic_rds_mbx {
- u64 phy_addr_reg;
- u64 phy_addr_jmb;
+ u32 phy_addr_reg_low;
+ u32 phy_addr_reg_high;
+ u32 phy_addr_jmb_low;
+ u32 phy_addr_jmb_high;
+#if defined(__LITTLE_ENDIAN)
u16 reg_ring_sz;
u16 reg_ring_len;
u16 jmb_ring_sz;
u16 jmb_ring_len;
+#elif defined(__BIG_ENDIAN)
+ u16 reg_ring_len;
+ u16 reg_ring_sz;
+ u16 jmb_ring_len;
+ u16 jmb_ring_sz;
+#endif
} __packed;
/* host producers for regular and jumbo rings */
@@ -61,6 +82,7 @@
* @phy_port: physical port id
*/
struct qlcnic_rcv_mbx_out {
+#if defined(__LITTLE_ENDIAN)
u8 rcv_num;
u8 sts_num;
u16 ctx_id;
@@ -68,32 +90,56 @@
u8 num_pci_func;
u8 phy_port;
u8 vport_id;
+#elif defined(__BIG_ENDIAN)
+ u16 ctx_id;
+ u8 sts_num;
+ u8 rcv_num;
+ u8 vport_id;
+ u8 phy_port;
+ u8 num_pci_func;
+ u8 state;
+#endif
u32 host_csmr[QLCNIC_MAX_RING_SETS];
struct __host_producer_mbx host_prod[QLCNIC_MAX_RING_SETS];
} __packed;
struct qlcnic_add_rings_mbx_out {
+#if defined(__LITTLE_ENDIAN)
u8 rcv_num;
u8 sts_num;
- u16 ctx_id;
+ u16 ctx_id;
+#elif defined(__BIG_ENDIAN)
+ u16 ctx_id;
+ u8 sts_num;
+ u8 rcv_num;
+#endif
u32 host_csmr[QLCNIC_MAX_RING_SETS];
struct __host_producer_mbx host_prod[QLCNIC_MAX_RING_SETS];
} __packed;
/* Transmit context mailbox inbox registers
- * @phys_addr: DMA address of the transmit buffer
- * @cnsmr_index: host consumer index
+ * @phys_addr_{low|high}: DMA address of the transmit buffer
+ * @cnsmr_index_{low|high}: host consumer index
* @size: legth of transmit buffer ring
* @intr_id: interrput id
* @src: src of interrupt
*/
struct qlcnic_tx_mbx {
- u64 phys_addr;
- u64 cnsmr_index;
+ u32 phys_addr_low;
+ u32 phys_addr_high;
+ u32 cnsmr_index_low;
+ u32 cnsmr_index_high;
+#if defined(__LITTLE_ENDIAN)
u16 size;
u16 intr_id;
u8 src;
u8 rsvd[3];
+#elif defined(__BIG_ENDIAN)
+ u16 intr_id;
+ u16 size;
+ u8 rsvd[3];
+ u8 src;
+#endif
} __packed;
/* Transmit context mailbox outbox registers
@@ -101,11 +147,18 @@
* @ctx_id: transmit context id
* @state: state of the transmit context
*/
+
struct qlcnic_tx_mbx_out {
u32 host_prod;
+#if defined(__LITTLE_ENDIAN)
u16 ctx_id;
u8 state;
u8 rsvd;
+#elif defined(__BIG_ENDIAN)
+ u8 rsvd;
+ u8 state;
+ u16 ctx_id;
+#endif
} __packed;
static const struct qlcnic_mailbox_metadata qlcnic_83xx_mbx_tbl[] = {
@@ -1004,7 +1057,8 @@
sds = &recv_ctx->sds_rings[i];
sds->consumer = 0;
memset(sds->desc_head, 0, STATUS_DESC_RINGSIZE(sds));
- sds_mbx.phy_addr = sds->phys_addr;
+ sds_mbx.phy_addr_low = LSD(sds->phys_addr);
+ sds_mbx.phy_addr_high = MSD(sds->phys_addr);
sds_mbx.sds_ring_size = sds->num_desc;
if (adapter->flags & QLCNIC_MSIX_ENABLED)
@@ -1090,7 +1144,8 @@
sds = &recv_ctx->sds_rings[i];
sds->consumer = 0;
memset(sds->desc_head, 0, STATUS_DESC_RINGSIZE(sds));
- sds_mbx.phy_addr = sds->phys_addr;
+ sds_mbx.phy_addr_low = LSD(sds->phys_addr);
+ sds_mbx.phy_addr_high = MSD(sds->phys_addr);
sds_mbx.sds_ring_size = sds->num_desc;
if (adapter->flags & QLCNIC_MSIX_ENABLED)
intrpt_id = ahw->intr_tbl[i].id;
@@ -1110,13 +1165,15 @@
rds = &recv_ctx->rds_rings[0];
rds->producer = 0;
memset(&rds_mbx, 0, rds_mbx_size);
- rds_mbx.phy_addr_reg = rds->phys_addr;
+ rds_mbx.phy_addr_reg_low = LSD(rds->phys_addr);
+ rds_mbx.phy_addr_reg_high = MSD(rds->phys_addr);
rds_mbx.reg_ring_sz = rds->dma_size;
rds_mbx.reg_ring_len = rds->num_desc;
/* Jumbo ring */
rds = &recv_ctx->rds_rings[1];
rds->producer = 0;
- rds_mbx.phy_addr_jmb = rds->phys_addr;
+ rds_mbx.phy_addr_jmb_low = LSD(rds->phys_addr);
+ rds_mbx.phy_addr_jmb_high = MSD(rds->phys_addr);
rds_mbx.jmb_ring_sz = rds->dma_size;
rds_mbx.jmb_ring_len = rds->num_desc;
buf = &cmd.req.arg[index];
@@ -1182,8 +1239,10 @@
memset(&mbx, 0, sizeof(struct qlcnic_tx_mbx));
/* setup mailbox inbox registerss */
- mbx.phys_addr = tx->phys_addr;
- mbx.cnsmr_index = tx->hw_cons_phys_addr;
+ mbx.phys_addr_low = LSD(tx->phys_addr);
+ mbx.phys_addr_high = MSD(tx->phys_addr);
+ mbx.cnsmr_index_low = LSD(tx->hw_cons_phys_addr);
+ mbx.cnsmr_index_high = MSD(tx->hw_cons_phys_addr);
mbx.size = tx->num_desc;
if (adapter->flags & QLCNIC_MSIX_ENABLED)
msix_id = ahw->intr_tbl[adapter->max_sds_rings + ring].id;
@@ -1373,6 +1432,51 @@
}
}
+int qlcnic_83xx_set_led(struct net_device *netdev,
+ enum ethtool_phys_id_state state)
+{
+ struct qlcnic_adapter *adapter = netdev_priv(netdev);
+ int err = -EIO, active = 1;
+
+ if (adapter->ahw->op_mode == QLCNIC_NON_PRIV_FUNC) {
+ netdev_warn(netdev,
+ "LED test is not supported in non-privileged mode\n");
+ return -EOPNOTSUPP;
+ }
+
+ switch (state) {
+ case ETHTOOL_ID_ACTIVE:
+ if (test_and_set_bit(__QLCNIC_LED_ENABLE, &adapter->state))
+ return -EBUSY;
+
+ if (test_bit(__QLCNIC_RESETTING, &adapter->state))
+ break;
+
+ err = qlcnic_83xx_config_led(adapter, active, 0);
+ if (err)
+ netdev_err(netdev, "Failed to set LED blink state\n");
+ break;
+ case ETHTOOL_ID_INACTIVE:
+ active = 0;
+
+ if (test_bit(__QLCNIC_RESETTING, &adapter->state))
+ break;
+
+ err = qlcnic_83xx_config_led(adapter, active, 0);
+ if (err)
+ netdev_err(netdev, "Failed to reset LED blink state\n");
+ break;
+
+ default:
+ return -EINVAL;
+ }
+
+ if (!active || err)
+ clear_bit(__QLCNIC_LED_ENABLE, &adapter->state);
+
+ return err;
+}
+
void qlcnic_83xx_register_nic_idc_func(struct qlcnic_adapter *adapter,
int enable)
{
@@ -1713,7 +1817,12 @@
(adapter->recv_ctx->context_id << 16);
mv.vlan = le16_to_cpu(vlan_id);
- memcpy(&mv.mac, addr, ETH_ALEN);
+ mv.mac_addr0 = addr[0];
+ mv.mac_addr1 = addr[1];
+ mv.mac_addr2 = addr[2];
+ mv.mac_addr3 = addr[3];
+ mv.mac_addr4 = addr[4];
+ mv.mac_addr5 = addr[5];
buf = &cmd.req.arg[2];
memcpy(buf, &mv, sizeof(struct qlcnic_macvlan_mbx));
err = qlcnic_issue_cmd(adapter, &cmd);
@@ -2163,7 +2272,7 @@
return 0;
}
-static int qlcnic_83xx_enable_flash_write_op(struct qlcnic_adapter *adapter)
+int qlcnic_83xx_enable_flash_write(struct qlcnic_adapter *adapter)
{
int ret;
u32 cmd;
@@ -2181,7 +2290,7 @@
return 0;
}
-static int qlcnic_83xx_disable_flash_write_op(struct qlcnic_adapter *adapter)
+int qlcnic_83xx_disable_flash_write(struct qlcnic_adapter *adapter)
{
int ret;
@@ -2255,7 +2364,7 @@
return -EIO;
if (adapter->ahw->fdt.mfg_id == adapter->flash_mfg_id) {
- ret = qlcnic_83xx_enable_flash_write_op(adapter);
+ ret = qlcnic_83xx_enable_flash_write(adapter);
if (ret) {
qlcnic_83xx_unlock_flash(adapter);
dev_err(&adapter->pdev->dev,
@@ -2297,7 +2406,7 @@
}
if (adapter->ahw->fdt.mfg_id == adapter->flash_mfg_id) {
- ret = qlcnic_83xx_disable_flash_write_op(adapter);
+ ret = qlcnic_83xx_disable_flash_write(adapter);
if (ret) {
qlcnic_83xx_unlock_flash(adapter);
dev_err(&adapter->pdev->dev,
@@ -2337,8 +2446,8 @@
u32 temp;
int ret = -EIO;
- if ((count < QLC_83XX_FLASH_BULK_WRITE_MIN) ||
- (count > QLC_83XX_FLASH_BULK_WRITE_MAX)) {
+ if ((count < QLC_83XX_FLASH_WRITE_MIN) ||
+ (count > QLC_83XX_FLASH_WRITE_MAX)) {
dev_err(&adapter->pdev->dev,
"%s: Invalid word count\n", __func__);
return -EIO;
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.h b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.h
index 61f81f6..fbb3d1d 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.h
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.h
@@ -12,6 +12,8 @@
#include <linux/etherdevice.h>
#include "qlcnic_hw.h"
+#define QLCNIC_83XX_BAR0_LENGTH 0x4000
+
/* Directly mapped registers */
#define QLC_83XX_CRB_WIN_BASE 0x3800
#define QLC_83XX_CRB_WIN_FUNC(f) (QLC_83XX_CRB_WIN_BASE+((f)*4))
@@ -94,8 +96,23 @@
};
struct qlcnic_macvlan_mbx {
- u8 mac[ETH_ALEN];
+#if defined(__LITTLE_ENDIAN)
+ u8 mac_addr0;
+ u8 mac_addr1;
+ u8 mac_addr2;
+ u8 mac_addr3;
+ u8 mac_addr4;
+ u8 mac_addr5;
u16 vlan;
+#elif defined(__BIG_ENDIAN)
+ u8 mac_addr3;
+ u8 mac_addr2;
+ u8 mac_addr1;
+ u8 mac_addr0;
+ u16 vlan;
+ u8 mac_addr5;
+ u8 mac_addr4;
+#endif
};
struct qlc_83xx_fw_info {
@@ -242,8 +259,8 @@
#define QLC_83XX_FLASH_BULK_WRITE_CMD 0xcadcadca
#define QLC_83XX_FLASH_READ_RETRY_COUNT 5000
#define QLC_83XX_FLASH_STATUS_READY 0x6
-#define QLC_83XX_FLASH_BULK_WRITE_MIN 2
-#define QLC_83XX_FLASH_BULK_WRITE_MAX 64
+#define QLC_83XX_FLASH_WRITE_MIN 2
+#define QLC_83XX_FLASH_WRITE_MAX 64
#define QLC_83XX_FLASH_STATUS_REG_POLL_DELAY 1
#define QLC_83XX_ERASE_MODE 1
#define QLC_83XX_WRITE_MODE 2
@@ -434,5 +451,8 @@
int qlcnic_83xx_get_registers(struct qlcnic_adapter *, u32 *);
int qlcnic_83xx_loopback_test(struct net_device *, u8);
int qlcnic_83xx_interrupt_test(struct net_device *);
+int qlcnic_83xx_set_led(struct net_device *, enum ethtool_phys_id_state);
int qlcnic_83xx_flash_test(struct qlcnic_adapter *);
+int qlcnic_83xx_enable_flash_write(struct qlcnic_adapter *);
+int qlcnic_83xx_disable_flash_write(struct qlcnic_adapter *);
#endif
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c
index 5c033f2..ba5ac69 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c
@@ -31,6 +31,7 @@
/* Template header */
struct qlc_83xx_reset_hdr {
+#if defined(__LITTLE_ENDIAN)
u16 version;
u16 signature;
u16 size;
@@ -39,14 +40,31 @@
u16 checksum;
u16 init_offset;
u16 start_offset;
+#elif defined(__BIG_ENDIAN)
+ u16 signature;
+ u16 version;
+ u16 entries;
+ u16 size;
+ u16 checksum;
+ u16 hdr_size;
+ u16 start_offset;
+ u16 init_offset;
+#endif
} __packed;
/* Command entry header. */
struct qlc_83xx_entry_hdr {
- u16 cmd;
- u16 size;
- u16 count;
- u16 delay;
+#if defined(__LITTLE_ENDIAN)
+ u16 cmd;
+ u16 size;
+ u16 count;
+ u16 delay;
+#elif defined(__BIG_ENDIAN)
+ u16 size;
+ u16 cmd;
+ u16 delay;
+ u16 count;
+#endif
} __packed;
/* Generic poll command */
@@ -60,10 +78,17 @@
u32 mask;
u32 xor_value;
u32 or_value;
+#if defined(__LITTLE_ENDIAN)
u8 shl;
u8 shr;
u8 index_a;
u8 rsvd;
+#elif defined(__BIG_ENDIAN)
+ u8 rsvd;
+ u8 index_a;
+ u8 shr;
+ u8 shl;
+#endif
} __packed;
/* Generic command with 2 DWORD */
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ctx.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ctx.c
index a69097c..2d9c23f 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ctx.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ctx.c
@@ -422,22 +422,20 @@
rq_size = SIZEOF_HOSTRQ_TX(struct qlcnic_hostrq_tx_ctx);
rq_addr = dma_alloc_coherent(&adapter->pdev->dev, rq_size,
- &rq_phys_addr, GFP_KERNEL);
+ &rq_phys_addr, GFP_KERNEL | __GFP_ZERO);
if (!rq_addr)
return -ENOMEM;
rsp_size = SIZEOF_CARDRSP_TX(struct qlcnic_cardrsp_tx_ctx);
rsp_addr = dma_alloc_coherent(&adapter->pdev->dev, rsp_size,
- &rsp_phys_addr, GFP_KERNEL);
+ &rsp_phys_addr, GFP_KERNEL | __GFP_ZERO);
if (!rsp_addr) {
err = -ENOMEM;
goto out_free_rq;
}
- memset(rq_addr, 0, rq_size);
prq = rq_addr;
- memset(rsp_addr, 0, rsp_size);
prsp = rsp_addr;
prq->host_rsp_dma_addr = cpu_to_le64(rsp_phys_addr);
@@ -532,20 +530,15 @@
ptr = (__le32 *)dma_alloc_coherent(&pdev->dev, sizeof(u32),
&tx_ring->hw_cons_phys_addr,
GFP_KERNEL);
-
- if (ptr == NULL) {
- dev_err(&pdev->dev, "failed to allocate tx consumer\n");
+ if (ptr == NULL)
return -ENOMEM;
- }
+
tx_ring->hw_consumer = ptr;
/* cmd desc ring */
addr = dma_alloc_coherent(&pdev->dev, TX_DESC_RINGSIZE(tx_ring),
&tx_ring->phys_addr,
GFP_KERNEL);
-
if (addr == NULL) {
- dev_err(&pdev->dev,
- "failed to allocate tx desc ring\n");
err = -ENOMEM;
goto err_out_free;
}
@@ -556,11 +549,9 @@
for (ring = 0; ring < adapter->max_rds_rings; ring++) {
rds_ring = &recv_ctx->rds_rings[ring];
addr = dma_alloc_coherent(&adapter->pdev->dev,
- RCV_DESC_RINGSIZE(rds_ring),
- &rds_ring->phys_addr, GFP_KERNEL);
+ RCV_DESC_RINGSIZE(rds_ring),
+ &rds_ring->phys_addr, GFP_KERNEL);
if (addr == NULL) {
- dev_err(&pdev->dev,
- "failed to allocate rds ring [%d]\n", ring);
err = -ENOMEM;
goto err_out_free;
}
@@ -572,11 +563,9 @@
sds_ring = &recv_ctx->sds_rings[ring];
addr = dma_alloc_coherent(&adapter->pdev->dev,
- STATUS_DESC_RINGSIZE(sds_ring),
- &sds_ring->phys_addr, GFP_KERNEL);
+ STATUS_DESC_RINGSIZE(sds_ring),
+ &sds_ring->phys_addr, GFP_KERNEL);
if (addr == NULL) {
- dev_err(&pdev->dev,
- "failed to allocate sds ring [%d]\n", ring);
err = -ENOMEM;
goto err_out_free;
}
@@ -753,10 +742,9 @@
size_t nic_size = sizeof(struct qlcnic_info_le);
nic_info_addr = dma_alloc_coherent(&adapter->pdev->dev, nic_size,
- &nic_dma_t, GFP_KERNEL);
+ &nic_dma_t, GFP_KERNEL | __GFP_ZERO);
if (!nic_info_addr)
return -ENOMEM;
- memset(nic_info_addr, 0, nic_size);
nic_info = nic_info_addr;
@@ -804,11 +792,10 @@
return err;
nic_info_addr = dma_alloc_coherent(&adapter->pdev->dev, nic_size,
- &nic_dma_t, GFP_KERNEL);
+ &nic_dma_t, GFP_KERNEL | __GFP_ZERO);
if (!nic_info_addr)
return -ENOMEM;
- memset(nic_info_addr, 0, nic_size);
nic_info = nic_info_addr;
nic_info->pci_func = cpu_to_le16(nic->pci_func);
@@ -854,10 +841,10 @@
size_t pci_size = npar_size * QLCNIC_MAX_PCI_FUNC;
pci_info_addr = dma_alloc_coherent(&adapter->pdev->dev, pci_size,
- &pci_info_dma_t, GFP_KERNEL);
+ &pci_info_dma_t,
+ GFP_KERNEL | __GFP_ZERO);
if (!pci_info_addr)
return -ENOMEM;
- memset(pci_info_addr, 0, pci_size);
npar = pci_info_addr;
qlcnic_alloc_mbx_args(&cmd, adapter, QLCNIC_CMD_GET_PCI_INFO);
@@ -949,12 +936,9 @@
}
stats_addr = dma_alloc_coherent(&adapter->pdev->dev, stats_size,
- &stats_dma_t, GFP_KERNEL);
- if (!stats_addr) {
- dev_err(&adapter->pdev->dev, "Unable to allocate memory\n");
+ &stats_dma_t, GFP_KERNEL | __GFP_ZERO);
+ if (!stats_addr)
return -ENOMEM;
- }
- memset(stats_addr, 0, stats_size);
arg1 = func | QLCNIC_STATS_VERSION << 8 | QLCNIC_STATS_PORT << 12;
arg1 |= rx_tx << 15 | stats_size << 16;
@@ -1003,13 +987,10 @@
return -ENOMEM;
stats_addr = dma_alloc_coherent(&adapter->pdev->dev, stats_size,
- &stats_dma_t, GFP_KERNEL);
- if (!stats_addr) {
- dev_err(&adapter->pdev->dev,
- "%s: Unable to allocate memory.\n", __func__);
+ &stats_dma_t, GFP_KERNEL | __GFP_ZERO);
+ if (!stats_addr)
return -ENOMEM;
- }
- memset(stats_addr, 0, stats_size);
+
qlcnic_alloc_mbx_args(&cmd, adapter, QLCNIC_CMD_GET_MAC_STATS);
cmd.req.arg[1] = stats_size << 16;
cmd.req.arg[2] = MSD(stats_dma_t);
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ethtool.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ethtool.c
index 5641f8e..f4f279d 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ethtool.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ethtool.c
@@ -149,7 +149,8 @@
static inline int qlcnic_82xx_statistics(void)
{
- return QLCNIC_STATS_LEN + ARRAY_SIZE(qlcnic_83xx_mac_stats_strings);
+ return ARRAY_SIZE(qlcnic_device_gstrings_stats) +
+ ARRAY_SIZE(qlcnic_83xx_mac_stats_strings);
}
static inline int qlcnic_83xx_statistics(void)
@@ -1070,8 +1071,7 @@
}
}
-static void
-qlcnic_fill_stats(u64 *data, void *stats, int type)
+static u64 *qlcnic_fill_stats(u64 *data, void *stats, int type)
{
if (type == QLCNIC_MAC_STATS) {
struct qlcnic_mac_statistics *mac_stats =
@@ -1120,6 +1120,7 @@
*data++ = QLCNIC_FILL_STATS(esw_stats->local_frames);
*data++ = QLCNIC_FILL_STATS(esw_stats->numbytes);
}
+ return data;
}
static void qlcnic_get_ethtool_stats(struct net_device *dev,
@@ -1147,7 +1148,7 @@
/* Retrieve MAC statistics from firmware */
memset(&mac_stats, 0, sizeof(struct qlcnic_mac_statistics));
qlcnic_get_mac_stats(adapter, &mac_stats);
- qlcnic_fill_stats(data, &mac_stats, QLCNIC_MAC_STATS);
+ data = qlcnic_fill_stats(data, &mac_stats, QLCNIC_MAC_STATS);
}
if (!(adapter->flags & QLCNIC_ESWITCH_ENABLED))
@@ -1159,7 +1160,7 @@
if (ret)
return;
- qlcnic_fill_stats(data, &port_stats.rx, QLCNIC_ESW_STATS);
+ data = qlcnic_fill_stats(data, &port_stats.rx, QLCNIC_ESW_STATS);
ret = qlcnic_get_port_stats(adapter, adapter->ahw->pci_func,
QLCNIC_QUERY_TX_COUNTER, &port_stats.tx);
if (ret)
@@ -1176,7 +1177,8 @@
int err = -EIO, active = 1;
if (qlcnic_83xx_check(adapter))
- return -EOPNOTSUPP;
+ return qlcnic_83xx_set_led(dev, state);
+
if (adapter->ahw->op_mode == QLCNIC_NON_PRIV_FUNC) {
netdev_warn(dev, "LED test not supported for non "
"privilege function\n");
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c
index 0e63006..891f12d 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c
@@ -9,6 +9,7 @@
#include <linux/if_vlan.h>
#include <net/ip.h>
#include <linux/ipv6.h>
+#include <net/checksum.h>
#include "qlcnic.h"
@@ -1132,9 +1133,8 @@
iph = (struct iphdr *)skb->data;
th = (struct tcphdr *)(skb->data + (iph->ihl << 2));
length = (iph->ihl << 2) + (th->doff << 2) + lro_length;
+ csum_replace2(&iph->check, iph->tot_len, htons(length));
iph->tot_len = htons(length);
- iph->check = 0;
- iph->check = ip_fast_csum((unsigned char *)iph, iph->ihl);
}
th->psh = push;
@@ -1595,9 +1595,8 @@
iph = (struct iphdr *)skb->data;
th = (struct tcphdr *)(skb->data + (iph->ihl << 2));
length = (iph->ihl << 2) + (th->doff << 2) + lro_length;
+ csum_replace2(&iph->check, iph->tot_len, htons(length));
iph->tot_len = htons(length);
- iph->check = 0;
- iph->check = ip_fast_csum((unsigned char *)iph, iph->ihl);
}
th->psh = push;
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
index 28a6d48..c6f9d5e 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
@@ -253,11 +253,8 @@
struct qlcnic_adapter *adapter = netdev_priv(netdev);
int err = -EOPNOTSUPP;
- if (!adapter->fdb_mac_learn) {
- pr_info("%s: Driver mac learn is enabled, FDB operation not allowed\n",
- __func__);
- return err;
- }
+ if (!adapter->fdb_mac_learn)
+ return ndo_dflt_fdb_del(ndm, tb, netdev, addr);
if (adapter->flags & QLCNIC_ESWITCH_ENABLED) {
if (is_unicast_ether_addr(addr))
@@ -277,11 +274,8 @@
struct qlcnic_adapter *adapter = netdev_priv(netdev);
int err = 0;
- if (!adapter->fdb_mac_learn) {
- pr_info("%s: Driver mac learn is enabled, FDB operation not allowed\n",
- __func__);
- return -EOPNOTSUPP;
- }
+ if (!adapter->fdb_mac_learn)
+ return ndo_dflt_fdb_add(ndm, tb, netdev, addr, flags);
if (!(adapter->flags & QLCNIC_ESWITCH_ENABLED)) {
pr_info("%s: FDB e-switch is not enabled\n", __func__);
@@ -306,11 +300,8 @@
{
struct qlcnic_adapter *adapter = netdev_priv(netdev);
- if (!adapter->fdb_mac_learn) {
- pr_info("%s: Driver mac learn is enabled, FDB operation not allowed\n",
- __func__);
- return -EOPNOTSUPP;
- }
+ if (!adapter->fdb_mac_learn)
+ return ndo_dflt_fdb_dump(skb, ncb, netdev, idx);
if (adapter->flags & QLCNIC_ESWITCH_ENABLED)
idx = ndo_dflt_fdb_dump(skb, ncb, netdev, idx);
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_minidump.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_minidump.c
index abbd22c..4b9bab1 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_minidump.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_minidump.c
@@ -810,11 +810,8 @@
tmp_addr = dma_alloc_coherent(&adapter->pdev->dev, temp_size,
&tmp_addr_t, GFP_KERNEL);
- if (!tmp_addr) {
- dev_err(&adapter->pdev->dev,
- "Can't get memory for FW dump template\n");
+ if (!tmp_addr)
return -ENOMEM;
- }
if (qlcnic_alloc_mbx_args(&cmd, adapter, QLCNIC_CMD_GET_TEMP_HDR)) {
err = -ENOMEM;
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sysfs.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sysfs.c
index 987fb6f..c77675d 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sysfs.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sysfs.c
@@ -21,8 +21,6 @@
#include <linux/aer.h>
#include <linux/log2.h>
-#include <linux/sysfs.h>
-
#define QLC_STATUS_UNSUPPORTED_CMD -2
int qlcnicvf_config_bridged_mode(struct qlcnic_adapter *adapter, u32 enable)
@@ -886,6 +884,244 @@
return size;
}
+static ssize_t qlcnic_83xx_sysfs_flash_read_handler(struct file *filp,
+ struct kobject *kobj,
+ struct bin_attribute *attr,
+ char *buf, loff_t offset,
+ size_t size)
+{
+ unsigned char *p_read_buf;
+ int ret, count;
+ struct device *dev = container_of(kobj, struct device, kobj);
+ struct qlcnic_adapter *adapter = dev_get_drvdata(dev);
+
+ if (!size)
+ return QL_STATUS_INVALID_PARAM;
+ if (!buf)
+ return QL_STATUS_INVALID_PARAM;
+
+ count = size / sizeof(u32);
+
+ if (size % sizeof(u32))
+ count++;
+
+ p_read_buf = kcalloc(size, sizeof(unsigned char), GFP_KERNEL);
+ if (!p_read_buf)
+ return -ENOMEM;
+ if (qlcnic_83xx_lock_flash(adapter) != 0) {
+ kfree(p_read_buf);
+ return -EIO;
+ }
+
+ ret = qlcnic_83xx_lockless_flash_read32(adapter, offset, p_read_buf,
+ count);
+
+ if (ret) {
+ qlcnic_83xx_unlock_flash(adapter);
+ kfree(p_read_buf);
+ return ret;
+ }
+
+ qlcnic_83xx_unlock_flash(adapter);
+ memcpy(buf, p_read_buf, size);
+ kfree(p_read_buf);
+
+ return size;
+}
+
+static int qlcnic_83xx_sysfs_flash_bulk_write(struct qlcnic_adapter *adapter,
+ char *buf, loff_t offset,
+ size_t size)
+{
+ int i, ret, count;
+ unsigned char *p_cache, *p_src;
+
+ p_cache = kcalloc(size, sizeof(unsigned char), GFP_KERNEL);
+ if (!p_cache)
+ return -ENOMEM;
+
+ memcpy(p_cache, buf, size);
+ p_src = p_cache;
+ count = size / sizeof(u32);
+
+ if (qlcnic_83xx_lock_flash(adapter) != 0) {
+ kfree(p_cache);
+ return -EIO;
+ }
+
+ if (adapter->ahw->fdt.mfg_id == adapter->flash_mfg_id) {
+ ret = qlcnic_83xx_enable_flash_write(adapter);
+ if (ret) {
+ kfree(p_cache);
+ qlcnic_83xx_unlock_flash(adapter);
+ return -EIO;
+ }
+ }
+
+ for (i = 0; i < count / QLC_83XX_FLASH_WRITE_MAX; i++) {
+ ret = qlcnic_83xx_flash_bulk_write(adapter, offset,
+ (u32 *)p_src,
+ QLC_83XX_FLASH_WRITE_MAX);
+
+ if (ret) {
+ if (adapter->ahw->fdt.mfg_id == adapter->flash_mfg_id) {
+ ret = qlcnic_83xx_disable_flash_write(adapter);
+ if (ret) {
+ kfree(p_cache);
+ qlcnic_83xx_unlock_flash(adapter);
+ return -EIO;
+ }
+ }
+
+ kfree(p_cache);
+ qlcnic_83xx_unlock_flash(adapter);
+ return -EIO;
+ }
+
+ p_src = p_src + sizeof(u32)*QLC_83XX_FLASH_WRITE_MAX;
+ offset = offset + sizeof(u32)*QLC_83XX_FLASH_WRITE_MAX;
+ }
+
+ if (adapter->ahw->fdt.mfg_id == adapter->flash_mfg_id) {
+ ret = qlcnic_83xx_disable_flash_write(adapter);
+ if (ret) {
+ kfree(p_cache);
+ qlcnic_83xx_unlock_flash(adapter);
+ return -EIO;
+ }
+ }
+
+ kfree(p_cache);
+ qlcnic_83xx_unlock_flash(adapter);
+
+ return 0;
+}
+
+static int qlcnic_83xx_sysfs_flash_write(struct qlcnic_adapter *adapter,
+ char *buf, loff_t offset, size_t size)
+{
+ int i, ret, count;
+ unsigned char *p_cache, *p_src;
+
+ p_cache = kcalloc(size, sizeof(unsigned char), GFP_KERNEL);
+ if (!p_cache)
+ return -ENOMEM;
+
+ memcpy(p_cache, buf, size);
+ p_src = p_cache;
+ count = size / sizeof(u32);
+
+ if (qlcnic_83xx_lock_flash(adapter) != 0) {
+ kfree(p_cache);
+ return -EIO;
+ }
+
+ if (adapter->ahw->fdt.mfg_id == adapter->flash_mfg_id) {
+ ret = qlcnic_83xx_enable_flash_write(adapter);
+ if (ret) {
+ kfree(p_cache);
+ qlcnic_83xx_unlock_flash(adapter);
+ return -EIO;
+ }
+ }
+
+ for (i = 0; i < count; i++) {
+ ret = qlcnic_83xx_flash_write32(adapter, offset, (u32 *)p_src);
+ if (ret) {
+ if (adapter->ahw->fdt.mfg_id == adapter->flash_mfg_id) {
+ ret = qlcnic_83xx_disable_flash_write(adapter);
+ if (ret) {
+ kfree(p_cache);
+ qlcnic_83xx_unlock_flash(adapter);
+ return -EIO;
+ }
+ }
+ kfree(p_cache);
+ qlcnic_83xx_unlock_flash(adapter);
+ return -EIO;
+ }
+
+ p_src = p_src + sizeof(u32);
+ offset = offset + sizeof(u32);
+ }
+
+ if (adapter->ahw->fdt.mfg_id == adapter->flash_mfg_id) {
+ ret = qlcnic_83xx_disable_flash_write(adapter);
+ if (ret) {
+ kfree(p_cache);
+ qlcnic_83xx_unlock_flash(adapter);
+ return -EIO;
+ }
+ }
+
+ kfree(p_cache);
+ qlcnic_83xx_unlock_flash(adapter);
+
+ return 0;
+}
+
+static ssize_t qlcnic_83xx_sysfs_flash_write_handler(struct file *filp,
+ struct kobject *kobj,
+ struct bin_attribute *attr,
+ char *buf, loff_t offset,
+ size_t size)
+{
+ int ret;
+ static int flash_mode;
+ unsigned long data;
+ struct device *dev = container_of(kobj, struct device, kobj);
+ struct qlcnic_adapter *adapter = dev_get_drvdata(dev);
+
+ if (!buf)
+ return QL_STATUS_INVALID_PARAM;
+
+ ret = kstrtoul(buf, 16, &data);
+
+ switch (data) {
+ case QLC_83XX_FLASH_SECTOR_ERASE_CMD:
+ flash_mode = QLC_83XX_ERASE_MODE;
+ ret = qlcnic_83xx_erase_flash_sector(adapter, offset);
+ if (ret) {
+ dev_err(&adapter->pdev->dev,
+ "%s failed at %d\n", __func__, __LINE__);
+ return -EIO;
+ }
+ break;
+
+ case QLC_83XX_FLASH_BULK_WRITE_CMD:
+ flash_mode = QLC_83XX_BULK_WRITE_MODE;
+ break;
+
+ case QLC_83XX_FLASH_WRITE_CMD:
+ flash_mode = QLC_83XX_WRITE_MODE;
+ break;
+ default:
+ if (flash_mode == QLC_83XX_BULK_WRITE_MODE) {
+ ret = qlcnic_83xx_sysfs_flash_bulk_write(adapter, buf,
+ offset, size);
+ if (ret) {
+ dev_err(&adapter->pdev->dev,
+ "%s failed at %d\n",
+ __func__, __LINE__);
+ return -EIO;
+ }
+ }
+
+ if (flash_mode == QLC_83XX_WRITE_MODE) {
+ ret = qlcnic_83xx_sysfs_flash_write(adapter, buf,
+ offset, size);
+ if (ret) {
+ dev_err(&adapter->pdev->dev,
+ "%s failed at %d\n", __func__,
+ __LINE__);
+ return -EIO;
+ }
+ }
+ }
+
+ return size;
+}
+
static struct device_attribute dev_attr_bridged_mode = {
.attr = {.name = "bridged_mode", .mode = (S_IRUGO | S_IWUSR)},
.show = qlcnic_show_bridged_mode,
@@ -960,6 +1196,13 @@
.write = qlcnic_sysfs_write_pm_config,
};
+static struct bin_attribute bin_attr_flash = {
+ .attr = {.name = "flash", .mode = (S_IRUGO | S_IWUSR)},
+ .size = 0,
+ .read = qlcnic_83xx_sysfs_flash_read_handler,
+ .write = qlcnic_83xx_sysfs_flash_write_handler,
+};
+
void qlcnic_create_sysfs_entries(struct qlcnic_adapter *adapter)
{
struct device *dev = &adapter->pdev->dev;
@@ -1048,10 +1291,18 @@
void qlcnic_83xx_add_sysfs(struct qlcnic_adapter *adapter)
{
+ struct device *dev = &adapter->pdev->dev;
+
qlcnic_create_diag_entries(adapter);
+
+ if (sysfs_create_bin_file(&dev->kobj, &bin_attr_flash))
+ dev_info(dev, "failed to create flash sysfs entry\n");
}
void qlcnic_83xx_remove_sysfs(struct qlcnic_adapter *adapter)
{
+ struct device *dev = &adapter->pdev->dev;
+
qlcnic_remove_diag_entries(adapter);
+ sysfs_remove_bin_file(&dev->kobj, &bin_attr_flash);
}
diff --git a/drivers/net/ethernet/qlogic/qlge/qlge_main.c b/drivers/net/ethernet/qlogic/qlge/qlge_main.c
index b13ab54..1dd778a 100644
--- a/drivers/net/ethernet/qlogic/qlge/qlge_main.c
+++ b/drivers/net/ethernet/qlogic/qlge/qlge_main.c
@@ -1211,8 +1211,6 @@
netdev_alloc_skb(qdev->ndev,
SMALL_BUFFER_SIZE);
if (sbq_desc->p.skb == NULL) {
- netif_err(qdev, probe, qdev->ndev,
- "Couldn't get an skb.\n");
rx_ring->sbq_clean_idx = clean_idx;
return;
}
@@ -1519,8 +1517,6 @@
skb = netdev_alloc_skb(ndev, length);
if (!skb) {
- netif_err(qdev, drv, qdev->ndev,
- "Couldn't get an skb, need to unwind!.\n");
rx_ring->rx_dropped++;
put_page(lbq_desc->p.pg_chunk.page);
return;
@@ -1605,8 +1601,6 @@
/* Allocate new_skb and copy */
new_skb = netdev_alloc_skb(qdev->ndev, length + NET_IP_ALIGN);
if (new_skb == NULL) {
- netif_err(qdev, probe, qdev->ndev,
- "No skb available, drop the packet.\n");
rx_ring->rx_dropped++;
return;
}
diff --git a/drivers/net/ethernet/rdc/r6040.c b/drivers/net/ethernet/rdc/r6040.c
index 5b4103d..e9dc849 100644
--- a/drivers/net/ethernet/rdc/r6040.c
+++ b/drivers/net/ethernet/rdc/r6040.c
@@ -224,11 +224,14 @@
break;
}
+ if (limit < 0)
+ return -ETIMEDOUT;
+
return ioread16(ioaddr + MMRD);
}
/* Write a word data from PHY Chip */
-static void r6040_phy_write(void __iomem *ioaddr,
+static int r6040_phy_write(void __iomem *ioaddr,
int phy_addr, int reg, u16 val)
{
int limit = MAC_DEF_TIMEOUT;
@@ -243,6 +246,8 @@
if (!(cmd & MDIO_WRITE))
break;
}
+
+ return (limit < 0) ? -ETIMEDOUT : 0;
}
static int r6040_mdiobus_read(struct mii_bus *bus, int phy_addr, int reg)
@@ -261,9 +266,7 @@
struct r6040_private *lp = netdev_priv(dev);
void __iomem *ioaddr = lp->base;
- r6040_phy_write(ioaddr, phy_addr, reg, value);
-
- return 0;
+ return r6040_phy_write(ioaddr, phy_addr, reg, value);
}
static int r6040_mdiobus_reset(struct mii_bus *bus)
@@ -347,7 +350,6 @@
do {
skb = netdev_alloc_skb(dev, MAX_BUF_SIZE);
if (!skb) {
- netdev_err(dev, "failed to alloc skb for rx\n");
rc = -ENOMEM;
goto err_exit;
}
diff --git a/drivers/net/ethernet/realtek/8139too.c b/drivers/net/ethernet/realtek/8139too.c
index 1276ac7..3ccedeb 100644
--- a/drivers/net/ethernet/realtek/8139too.c
+++ b/drivers/net/ethernet/realtek/8139too.c
@@ -2041,8 +2041,6 @@
netif_receive_skb (skb);
} else {
- if (net_ratelimit())
- netdev_warn(dev, "Memory squeeze, dropping packet\n");
dev->stats.rx_dropped++;
}
received++;
diff --git a/drivers/net/ethernet/realtek/atp.c b/drivers/net/ethernet/realtek/atp.c
index 9f2d416..d77d60e 100644
--- a/drivers/net/ethernet/realtek/atp.c
+++ b/drivers/net/ethernet/realtek/atp.c
@@ -782,8 +782,6 @@
skb = netdev_alloc_skb(dev, pkt_len + 2);
if (skb == NULL) {
- printk(KERN_ERR "%s: Memory squeeze, dropping packet.\n",
- dev->name);
dev->stats.rx_dropped++;
goto done;
}
diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c
index bf5e3cf..56884c4 100644
--- a/drivers/net/ethernet/renesas/sh_eth.c
+++ b/drivers/net/ethernet/renesas/sh_eth.c
@@ -908,11 +908,8 @@
/* Allocate all Rx descriptors. */
rx_ringsize = sizeof(struct sh_eth_rxdesc) * mdp->num_rx_ring;
mdp->rx_ring = dma_alloc_coherent(NULL, rx_ringsize, &mdp->rx_desc_dma,
- GFP_KERNEL);
-
+ GFP_KERNEL);
if (!mdp->rx_ring) {
- dev_err(&ndev->dev, "Cannot allocate Rx Ring (size %d bytes)\n",
- rx_ringsize);
ret = -ENOMEM;
goto desc_ring_free;
}
@@ -922,10 +919,8 @@
/* Allocate all Tx descriptors. */
tx_ringsize = sizeof(struct sh_eth_txdesc) * mdp->num_tx_ring;
mdp->tx_ring = dma_alloc_coherent(NULL, tx_ringsize, &mdp->tx_desc_dma,
- GFP_KERNEL);
+ GFP_KERNEL);
if (!mdp->tx_ring) {
- dev_err(&ndev->dev, "Cannot allocate Tx Ring (size %d bytes)\n",
- tx_ringsize);
ret = -ENOMEM;
goto desc_ring_free;
}
diff --git a/drivers/net/ethernet/seeq/ether3.c b/drivers/net/ethernet/seeq/ether3.c
index 3aca578..bdac936 100644
--- a/drivers/net/ethernet/seeq/ether3.c
+++ b/drivers/net/ethernet/seeq/ether3.c
@@ -651,8 +651,11 @@
skb->protocol = eth_type_trans(skb, dev);
netif_rx(skb);
received ++;
- } else
- goto dropping;
+ } else {
+ ether3_outw(next_ptr >> 8, REG_RECVEND);
+ dev->stats.rx_dropped++;
+ goto done;
+ }
} else {
struct net_device_stats *stats = &dev->stats;
ether3_outw(next_ptr >> 8, REG_RECVEND);
@@ -679,21 +682,6 @@
}
return maxcnt;
-
-dropping:{
- static unsigned long last_warned;
-
- ether3_outw(next_ptr >> 8, REG_RECVEND);
- /*
- * Don't print this message too many times...
- */
- if (time_after(jiffies, last_warned + 10 * HZ)) {
- last_warned = jiffies;
- printk("%s: memory squeeze, dropping packet.\n", dev->name);
- }
- dev->stats.rx_dropped++;
- goto done;
- }
}
/*
diff --git a/drivers/net/ethernet/seeq/sgiseeq.c b/drivers/net/ethernet/seeq/sgiseeq.c
index 0fde9ca..0ad5694 100644
--- a/drivers/net/ethernet/seeq/sgiseeq.c
+++ b/drivers/net/ethernet/seeq/sgiseeq.c
@@ -381,8 +381,6 @@
dev->stats.rx_packets++;
dev->stats.rx_bytes += len;
} else {
- printk(KERN_NOTICE "%s: Memory squeeze, deferring packet.\n",
- dev->name);
dev->stats.rx_dropped++;
}
} else {
diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 0bc0099..01b9920 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -22,6 +22,7 @@
#include <linux/topology.h>
#include <linux/gfp.h>
#include <linux/cpu_rmap.h>
+#include <linux/aer.h>
#include "net_driver.h"
#include "efx.h"
#include "nic.h"
@@ -71,21 +72,21 @@
const unsigned int efx_reset_type_max = RESET_TYPE_MAX;
const char *const efx_reset_type_names[] = {
- [RESET_TYPE_INVISIBLE] = "INVISIBLE",
- [RESET_TYPE_ALL] = "ALL",
- [RESET_TYPE_WORLD] = "WORLD",
- [RESET_TYPE_DISABLE] = "DISABLE",
- [RESET_TYPE_TX_WATCHDOG] = "TX_WATCHDOG",
- [RESET_TYPE_INT_ERROR] = "INT_ERROR",
- [RESET_TYPE_RX_RECOVERY] = "RX_RECOVERY",
- [RESET_TYPE_RX_DESC_FETCH] = "RX_DESC_FETCH",
- [RESET_TYPE_TX_DESC_FETCH] = "TX_DESC_FETCH",
- [RESET_TYPE_TX_SKIP] = "TX_SKIP",
- [RESET_TYPE_MC_FAILURE] = "MC_FAILURE",
+ [RESET_TYPE_INVISIBLE] = "INVISIBLE",
+ [RESET_TYPE_ALL] = "ALL",
+ [RESET_TYPE_RECOVER_OR_ALL] = "RECOVER_OR_ALL",
+ [RESET_TYPE_WORLD] = "WORLD",
+ [RESET_TYPE_RECOVER_OR_DISABLE] = "RECOVER_OR_DISABLE",
+ [RESET_TYPE_DISABLE] = "DISABLE",
+ [RESET_TYPE_TX_WATCHDOG] = "TX_WATCHDOG",
+ [RESET_TYPE_INT_ERROR] = "INT_ERROR",
+ [RESET_TYPE_RX_RECOVERY] = "RX_RECOVERY",
+ [RESET_TYPE_RX_DESC_FETCH] = "RX_DESC_FETCH",
+ [RESET_TYPE_TX_DESC_FETCH] = "TX_DESC_FETCH",
+ [RESET_TYPE_TX_SKIP] = "TX_SKIP",
+ [RESET_TYPE_MC_FAILURE] = "MC_FAILURE",
};
-#define EFX_MAX_MTU (9 * 1024)
-
/* Reset workqueue. If any NIC has a hardware failure then a reset will be
* queued onto this work queue. This is not a per-nic work queue, because
* efx_reset_work() acquires the rtnl lock, so resets are naturally serialised.
@@ -117,9 +118,12 @@
static int napi_weight = 64;
/* This is the time (in jiffies) between invocations of the hardware
- * monitor. On Falcon-based NICs, this will:
+ * monitor.
+ * On Falcon-based NICs, this will:
* - Check the on-board hardware monitor;
* - Poll the link state and reconfigure the hardware as necessary.
+ * On Siena-based NICs for power systems with EEH support, this will give EEH a
+ * chance to start.
*/
static unsigned int efx_monitor_interval = 1 * HZ;
@@ -203,13 +207,14 @@
#define EFX_ASSERT_RESET_SERIALISED(efx) \
do { \
if ((efx->state == STATE_READY) || \
+ (efx->state == STATE_RECOVERY) || \
(efx->state == STATE_DISABLED)) \
ASSERT_RTNL(); \
} while (0)
static int efx_check_disabled(struct efx_nic *efx)
{
- if (efx->state == STATE_DISABLED) {
+ if (efx->state == STATE_DISABLED || efx->state == STATE_RECOVERY) {
netif_err(efx, drv, efx->net_dev,
"device is disabled due to earlier errors\n");
return -EIO;
@@ -242,15 +247,9 @@
struct efx_rx_queue *rx_queue =
efx_channel_get_rx_queue(channel);
- /* Deliver last RX packet. */
- if (channel->rx_pkt) {
- __efx_rx_packet(channel, channel->rx_pkt);
- channel->rx_pkt = NULL;
- }
- if (rx_queue->enabled) {
- efx_rx_strategy(channel);
+ efx_rx_flush_packet(channel);
+ if (rx_queue->enabled)
efx_fast_push_rx_descriptors(rx_queue);
- }
}
return spent;
@@ -625,20 +624,51 @@
*/
static void efx_start_datapath(struct efx_nic *efx)
{
+ bool old_rx_scatter = efx->rx_scatter;
struct efx_tx_queue *tx_queue;
struct efx_rx_queue *rx_queue;
struct efx_channel *channel;
+ size_t rx_buf_len;
/* Calculate the rx buffer allocation parameters required to
* support the current MTU, including padding for header
* alignment and overruns.
*/
- efx->rx_buffer_len = (max(EFX_PAGE_IP_ALIGN, NET_IP_ALIGN) +
- EFX_MAX_FRAME_LEN(efx->net_dev->mtu) +
- efx->type->rx_buffer_hash_size +
- efx->type->rx_buffer_padding);
- efx->rx_buffer_order = get_order(efx->rx_buffer_len +
- sizeof(struct efx_rx_page_state));
+ efx->rx_dma_len = (efx->type->rx_buffer_hash_size +
+ EFX_MAX_FRAME_LEN(efx->net_dev->mtu) +
+ efx->type->rx_buffer_padding);
+ rx_buf_len = (sizeof(struct efx_rx_page_state) +
+ EFX_PAGE_IP_ALIGN + efx->rx_dma_len);
+ if (rx_buf_len <= PAGE_SIZE) {
+ efx->rx_scatter = false;
+ efx->rx_buffer_order = 0;
+ } else if (efx->type->can_rx_scatter) {
+ BUILD_BUG_ON(sizeof(struct efx_rx_page_state) +
+ EFX_PAGE_IP_ALIGN + EFX_RX_USR_BUF_SIZE >
+ PAGE_SIZE / 2);
+ efx->rx_scatter = true;
+ efx->rx_dma_len = EFX_RX_USR_BUF_SIZE;
+ efx->rx_buffer_order = 0;
+ } else {
+ efx->rx_scatter = false;
+ efx->rx_buffer_order = get_order(rx_buf_len);
+ }
+
+ efx_rx_config_page_split(efx);
+ if (efx->rx_buffer_order)
+ netif_dbg(efx, drv, efx->net_dev,
+ "RX buf len=%u; page order=%u batch=%u\n",
+ efx->rx_dma_len, efx->rx_buffer_order,
+ efx->rx_pages_per_batch);
+ else
+ netif_dbg(efx, drv, efx->net_dev,
+ "RX buf len=%u step=%u bpp=%u; page batch=%u\n",
+ efx->rx_dma_len, efx->rx_page_buf_step,
+ efx->rx_bufs_per_page, efx->rx_pages_per_batch);
+
+ /* RX filters also have scatter-enabled flags */
+ if (efx->rx_scatter != old_rx_scatter)
+ efx_filter_update_rx_scatter(efx);
/* We must keep at least one descriptor in a TX ring empty.
* We could avoid this when the queue size does not exactly
@@ -655,16 +685,12 @@
efx_for_each_channel_tx_queue(tx_queue, channel)
efx_init_tx_queue(tx_queue);
- /* The rx buffer allocation strategy is MTU dependent */
- efx_rx_strategy(channel);
-
efx_for_each_channel_rx_queue(rx_queue, channel) {
efx_init_rx_queue(rx_queue);
efx_nic_generate_fill_event(rx_queue);
}
- WARN_ON(channel->rx_pkt != NULL);
- efx_rx_strategy(channel);
+ WARN_ON(channel->rx_pkt_n_frags);
}
if (netif_device_present(efx->net_dev))
@@ -683,7 +709,7 @@
BUG_ON(efx->port_enabled);
/* Only perform flush if dma is enabled */
- if (dev->is_busmaster) {
+ if (dev->is_busmaster && efx->state != STATE_RECOVERY) {
rc = efx_nic_flush_queues(efx);
if (rc && EFX_WORKAROUND_7803(efx)) {
@@ -1596,13 +1622,15 @@
efx_start_port(efx);
efx_start_datapath(efx);
- /* Start the hardware monitor if there is one. Otherwise (we're link
- * event driven), we have to poll the PHY because after an event queue
- * flush, we could have a missed a link state change */
- if (efx->type->monitor != NULL) {
+ /* Start the hardware monitor if there is one */
+ if (efx->type->monitor != NULL)
queue_delayed_work(efx->workqueue, &efx->monitor_work,
efx_monitor_interval);
- } else {
+
+ /* If link state detection is normally event-driven, we have
+ * to poll now because we could have missed a change
+ */
+ if (efx_nic_rev(efx) >= EFX_REV_SIENA_A0) {
mutex_lock(&efx->mac_lock);
if (efx->phy_op->poll(efx))
efx_link_status_changed(efx);
@@ -2309,7 +2337,9 @@
out:
/* Leave device stopped if necessary */
- disabled = rc || method == RESET_TYPE_DISABLE;
+ disabled = rc ||
+ method == RESET_TYPE_DISABLE ||
+ method == RESET_TYPE_RECOVER_OR_DISABLE;
rc2 = efx_reset_up(efx, method, !disabled);
if (rc2) {
disabled = true;
@@ -2328,13 +2358,48 @@
return rc;
}
+/* Try recovery mechanisms.
+ * For now only EEH is supported.
+ * Returns 0 if the recovery mechanisms are unsuccessful.
+ * Returns a non-zero value otherwise.
+ */
+static int efx_try_recovery(struct efx_nic *efx)
+{
+#ifdef CONFIG_EEH
+ /* A PCI error can occur and not be seen by EEH because nothing
+ * happens on the PCI bus. In this case the driver may fail and
+ * schedule a 'recover or reset', leading to this recovery handler.
+ * Manually call the eeh failure check function.
+ */
+ struct eeh_dev *eehdev =
+ of_node_to_eeh_dev(pci_device_to_OF_node(efx->pci_dev));
+
+ if (eeh_dev_check_failure(eehdev)) {
+ /* The EEH mechanisms will handle the error and reset the
+ * device if necessary.
+ */
+ return 1;
+ }
+#endif
+ return 0;
+}
+
/* The worker thread exists so that code that cannot sleep can
* schedule a reset for later.
*/
static void efx_reset_work(struct work_struct *data)
{
struct efx_nic *efx = container_of(data, struct efx_nic, reset_work);
- unsigned long pending = ACCESS_ONCE(efx->reset_pending);
+ unsigned long pending;
+ enum reset_type method;
+
+ pending = ACCESS_ONCE(efx->reset_pending);
+ method = fls(pending) - 1;
+
+ if ((method == RESET_TYPE_RECOVER_OR_DISABLE ||
+ method == RESET_TYPE_RECOVER_OR_ALL) &&
+ efx_try_recovery(efx))
+ return;
if (!pending)
return;
@@ -2346,7 +2411,7 @@
* it cannot change again.
*/
if (efx->state == STATE_READY)
- (void)efx_reset(efx, fls(pending) - 1);
+ (void)efx_reset(efx, method);
rtnl_unlock();
}
@@ -2355,11 +2420,20 @@
{
enum reset_type method;
+ if (efx->state == STATE_RECOVERY) {
+ netif_dbg(efx, drv, efx->net_dev,
+ "recovering: skip scheduling %s reset\n",
+ RESET_TYPE(type));
+ return;
+ }
+
switch (type) {
case RESET_TYPE_INVISIBLE:
case RESET_TYPE_ALL:
+ case RESET_TYPE_RECOVER_OR_ALL:
case RESET_TYPE_WORLD:
case RESET_TYPE_DISABLE:
+ case RESET_TYPE_RECOVER_OR_DISABLE:
method = type;
netif_dbg(efx, drv, efx->net_dev, "scheduling %s reset\n",
RESET_TYPE(method));
@@ -2569,6 +2643,8 @@
efx_fini_struct(efx);
pci_set_drvdata(pci_dev, NULL);
free_netdev(efx->net_dev);
+
+ pci_disable_pcie_error_reporting(pci_dev);
};
/* NIC VPD information
@@ -2741,6 +2817,11 @@
netif_warn(efx, probe, efx->net_dev,
"failed to create MTDs (%d)\n", rc);
+ rc = pci_enable_pcie_error_reporting(pci_dev);
+ if (rc && rc != -EINVAL)
+ netif_warn(efx, probe, efx->net_dev,
+ "pci_enable_pcie_error_reporting failed (%d)\n", rc);
+
return 0;
fail4:
@@ -2865,12 +2946,112 @@
.restore = efx_pm_resume,
};
+/* A PCI error affecting this device was detected.
+ * At this point MMIO and DMA may be disabled.
+ * Stop the software path and request a slot reset.
+ */
+static pci_ers_result_t efx_io_error_detected(struct pci_dev *pdev,
+ enum pci_channel_state state)
+{
+ pci_ers_result_t status = PCI_ERS_RESULT_RECOVERED;
+ struct efx_nic *efx = pci_get_drvdata(pdev);
+
+ if (state == pci_channel_io_perm_failure)
+ return PCI_ERS_RESULT_DISCONNECT;
+
+ rtnl_lock();
+
+ if (efx->state != STATE_DISABLED) {
+ efx->state = STATE_RECOVERY;
+ efx->reset_pending = 0;
+
+ efx_device_detach_sync(efx);
+
+ efx_stop_all(efx);
+ efx_stop_interrupts(efx, false);
+
+ status = PCI_ERS_RESULT_NEED_RESET;
+ } else {
+ /* If the interface is disabled we don't want to do anything
+ * with it.
+ */
+ status = PCI_ERS_RESULT_RECOVERED;
+ }
+
+ rtnl_unlock();
+
+ pci_disable_device(pdev);
+
+ return status;
+}
+
+/* Fake a successfull reset, which will be performed later in efx_io_resume. */
+static pci_ers_result_t efx_io_slot_reset(struct pci_dev *pdev)
+{
+ struct efx_nic *efx = pci_get_drvdata(pdev);
+ pci_ers_result_t status = PCI_ERS_RESULT_RECOVERED;
+ int rc;
+
+ if (pci_enable_device(pdev)) {
+ netif_err(efx, hw, efx->net_dev,
+ "Cannot re-enable PCI device after reset.\n");
+ status = PCI_ERS_RESULT_DISCONNECT;
+ }
+
+ rc = pci_cleanup_aer_uncorrect_error_status(pdev);
+ if (rc) {
+ netif_err(efx, hw, efx->net_dev,
+ "pci_cleanup_aer_uncorrect_error_status failed (%d)\n", rc);
+ /* Non-fatal error. Continue. */
+ }
+
+ return status;
+}
+
+/* Perform the actual reset and resume I/O operations. */
+static void efx_io_resume(struct pci_dev *pdev)
+{
+ struct efx_nic *efx = pci_get_drvdata(pdev);
+ int rc;
+
+ rtnl_lock();
+
+ if (efx->state == STATE_DISABLED)
+ goto out;
+
+ rc = efx_reset(efx, RESET_TYPE_ALL);
+ if (rc) {
+ netif_err(efx, hw, efx->net_dev,
+ "efx_reset failed after PCI error (%d)\n", rc);
+ } else {
+ efx->state = STATE_READY;
+ netif_dbg(efx, hw, efx->net_dev,
+ "Done resetting and resuming IO after PCI error.\n");
+ }
+
+out:
+ rtnl_unlock();
+}
+
+/* For simplicity and reliability, we always require a slot reset and try to
+ * reset the hardware when a pci error affecting the device is detected.
+ * We leave both the link_reset and mmio_enabled callback unimplemented:
+ * with our request for slot reset the mmio_enabled callback will never be
+ * called, and the link_reset callback is not used by AER or EEH mechanisms.
+ */
+static struct pci_error_handlers efx_err_handlers = {
+ .error_detected = efx_io_error_detected,
+ .slot_reset = efx_io_slot_reset,
+ .resume = efx_io_resume,
+};
+
static struct pci_driver efx_pci_driver = {
.name = KBUILD_MODNAME,
.id_table = efx_pci_table,
.probe = efx_pci_probe,
.remove = efx_pci_remove,
.driver.pm = &efx_pm_ops,
+ .err_handler = &efx_err_handlers,
};
/**************************************************************************
diff --git a/drivers/net/ethernet/sfc/efx.h b/drivers/net/ethernet/sfc/efx.h
index d2f790d..8372da2 100644
--- a/drivers/net/ethernet/sfc/efx.h
+++ b/drivers/net/ethernet/sfc/efx.h
@@ -33,17 +33,22 @@
extern unsigned int efx_tx_max_skb_descs(struct efx_nic *efx);
/* RX */
+extern void efx_rx_config_page_split(struct efx_nic *efx);
extern int efx_probe_rx_queue(struct efx_rx_queue *rx_queue);
extern void efx_remove_rx_queue(struct efx_rx_queue *rx_queue);
extern void efx_init_rx_queue(struct efx_rx_queue *rx_queue);
extern void efx_fini_rx_queue(struct efx_rx_queue *rx_queue);
-extern void efx_rx_strategy(struct efx_channel *channel);
extern void efx_fast_push_rx_descriptors(struct efx_rx_queue *rx_queue);
extern void efx_rx_slow_fill(unsigned long context);
-extern void __efx_rx_packet(struct efx_channel *channel,
- struct efx_rx_buffer *rx_buf);
-extern void efx_rx_packet(struct efx_rx_queue *rx_queue, unsigned int index,
+extern void __efx_rx_packet(struct efx_channel *channel);
+extern void efx_rx_packet(struct efx_rx_queue *rx_queue,
+ unsigned int index, unsigned int n_frags,
unsigned int len, u16 flags);
+static inline void efx_rx_flush_packet(struct efx_channel *channel)
+{
+ if (channel->rx_pkt_n_frags)
+ __efx_rx_packet(channel);
+}
extern void efx_schedule_slow_fill(struct efx_rx_queue *rx_queue);
#define EFX_MAX_DMAQ_SIZE 4096UL
@@ -67,6 +72,7 @@
extern int efx_probe_filters(struct efx_nic *efx);
extern void efx_restore_filters(struct efx_nic *efx);
extern void efx_remove_filters(struct efx_nic *efx);
+extern void efx_filter_update_rx_scatter(struct efx_nic *efx);
extern s32 efx_filter_insert_filter(struct efx_nic *efx,
struct efx_filter_spec *spec,
bool replace);
diff --git a/drivers/net/ethernet/sfc/enum.h b/drivers/net/ethernet/sfc/enum.h
index 182dbe2..ab8fb58 100644
--- a/drivers/net/ethernet/sfc/enum.h
+++ b/drivers/net/ethernet/sfc/enum.h
@@ -137,8 +137,12 @@
* Reset methods are numbered in order of increasing scope.
*
* @RESET_TYPE_INVISIBLE: Reset datapath and MAC (Falcon only)
+ * @RESET_TYPE_RECOVER_OR_ALL: Try to recover. Apply RESET_TYPE_ALL
+ * if unsuccessful.
* @RESET_TYPE_ALL: Reset datapath, MAC and PHY
* @RESET_TYPE_WORLD: Reset as much as possible
+ * @RESET_TYPE_RECOVER_OR_DISABLE: Try to recover. Apply RESET_TYPE_DISABLE if
+ * unsuccessful.
* @RESET_TYPE_DISABLE: Reset datapath, MAC and PHY; leave NIC disabled
* @RESET_TYPE_TX_WATCHDOG: reset due to TX watchdog
* @RESET_TYPE_INT_ERROR: reset due to internal error
@@ -150,9 +154,11 @@
*/
enum reset_type {
RESET_TYPE_INVISIBLE = 0,
- RESET_TYPE_ALL = 1,
- RESET_TYPE_WORLD = 2,
- RESET_TYPE_DISABLE = 3,
+ RESET_TYPE_RECOVER_OR_ALL = 1,
+ RESET_TYPE_ALL = 2,
+ RESET_TYPE_WORLD = 3,
+ RESET_TYPE_RECOVER_OR_DISABLE = 4,
+ RESET_TYPE_DISABLE = 5,
RESET_TYPE_MAX_METHOD,
RESET_TYPE_TX_WATCHDOG,
RESET_TYPE_INT_ERROR,
diff --git a/drivers/net/ethernet/sfc/ethtool.c b/drivers/net/ethernet/sfc/ethtool.c
index 8e61cd0..6e76817 100644
--- a/drivers/net/ethernet/sfc/ethtool.c
+++ b/drivers/net/ethernet/sfc/ethtool.c
@@ -154,6 +154,7 @@
EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_tcp_udp_chksum_err),
EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_mcast_mismatch),
EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_frm_trunc),
+ EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_nodesc_trunc),
};
/* Number of ethtool statistics */
@@ -978,7 +979,8 @@
rule->m_ext.data[1]))
return -EINVAL;
- efx_filter_init_rx(&spec, EFX_FILTER_PRI_MANUAL, 0,
+ efx_filter_init_rx(&spec, EFX_FILTER_PRI_MANUAL,
+ efx->rx_scatter ? EFX_FILTER_FLAG_RX_SCATTER : 0,
(rule->ring_cookie == RX_CLS_FLOW_DISC) ?
0xfff : rule->ring_cookie);
diff --git a/drivers/net/ethernet/sfc/falcon.c b/drivers/net/ethernet/sfc/falcon.c
index 49bcd19..4486102 100644
--- a/drivers/net/ethernet/sfc/falcon.c
+++ b/drivers/net/ethernet/sfc/falcon.c
@@ -1546,10 +1546,6 @@
static void falcon_init_rx_cfg(struct efx_nic *efx)
{
- /* Prior to Siena the RX DMA engine will split each frame at
- * intervals of RX_USR_BUF_SIZE (32-byte units). We set it to
- * be so large that that never happens. */
- const unsigned huge_buf_size = (3 * 4096) >> 5;
/* RX control FIFO thresholds (32 entries) */
const unsigned ctrl_xon_thr = 20;
const unsigned ctrl_xoff_thr = 25;
@@ -1557,10 +1553,15 @@
efx_reado(efx, ®, FR_AZ_RX_CFG);
if (efx_nic_rev(efx) <= EFX_REV_FALCON_A1) {
- /* Data FIFO size is 5.5K */
+ /* Data FIFO size is 5.5K. The RX DMA engine only
+ * supports scattering for user-mode queues, but will
+ * split DMA writes at intervals of RX_USR_BUF_SIZE
+ * (32-byte units) even for kernel-mode queues. We
+ * set it to be so large that that never happens.
+ */
EFX_SET_OWORD_FIELD(reg, FRF_AA_RX_DESC_PUSH_EN, 0);
EFX_SET_OWORD_FIELD(reg, FRF_AA_RX_USR_BUF_SIZE,
- huge_buf_size);
+ (3 * 4096) >> 5);
EFX_SET_OWORD_FIELD(reg, FRF_AA_RX_XON_MAC_TH, 512 >> 8);
EFX_SET_OWORD_FIELD(reg, FRF_AA_RX_XOFF_MAC_TH, 2048 >> 8);
EFX_SET_OWORD_FIELD(reg, FRF_AA_RX_XON_TX_TH, ctrl_xon_thr);
@@ -1569,7 +1570,7 @@
/* Data FIFO size is 80K; register fields moved */
EFX_SET_OWORD_FIELD(reg, FRF_BZ_RX_DESC_PUSH_EN, 0);
EFX_SET_OWORD_FIELD(reg, FRF_BZ_RX_USR_BUF_SIZE,
- huge_buf_size);
+ EFX_RX_USR_BUF_SIZE >> 5);
/* Send XON and XOFF at ~3 * max MTU away from empty/full */
EFX_SET_OWORD_FIELD(reg, FRF_BZ_RX_XON_MAC_TH, 27648 >> 8);
EFX_SET_OWORD_FIELD(reg, FRF_BZ_RX_XOFF_MAC_TH, 54272 >> 8);
@@ -1815,6 +1816,7 @@
.evq_rptr_tbl_base = FR_AA_EVQ_RPTR_KER,
.max_dma_mask = DMA_BIT_MASK(FSF_AZ_TX_KER_BUF_ADDR_WIDTH),
.rx_buffer_padding = 0x24,
+ .can_rx_scatter = false,
.max_interrupt_mode = EFX_INT_MODE_MSI,
.phys_addr_channels = 4,
.timer_period_max = 1 << FRF_AB_TC_TIMER_VAL_WIDTH,
@@ -1865,6 +1867,7 @@
.max_dma_mask = DMA_BIT_MASK(FSF_AZ_TX_KER_BUF_ADDR_WIDTH),
.rx_buffer_hash_size = 0x10,
.rx_buffer_padding = 0,
+ .can_rx_scatter = true,
.max_interrupt_mode = EFX_INT_MODE_MSIX,
.phys_addr_channels = 32, /* Hardware limit is 64, but the legacy
* interrupt handler only supports 32
diff --git a/drivers/net/ethernet/sfc/filter.c b/drivers/net/ethernet/sfc/filter.c
index 8af42cd..2397f0e 100644
--- a/drivers/net/ethernet/sfc/filter.c
+++ b/drivers/net/ethernet/sfc/filter.c
@@ -66,6 +66,10 @@
#endif
};
+static void efx_filter_table_clear_entry(struct efx_nic *efx,
+ struct efx_filter_table *table,
+ unsigned int filter_idx);
+
/* The filter hash function is LFSR polynomial x^16 + x^3 + 1 of a 32-bit
* key derived from the n-tuple. The initial LFSR state is 0xffff. */
static u16 efx_filter_hash(u32 key)
@@ -168,6 +172,25 @@
filter_ctl, FRF_CZ_MULTICAST_NOMATCH_RSS_ENABLED,
!!(table->spec[EFX_FILTER_INDEX_MC_DEF].flags &
EFX_FILTER_FLAG_RX_RSS));
+
+ /* There is a single bit to enable RX scatter for all
+ * unmatched packets. Only set it if scatter is
+ * enabled in both filter specs.
+ */
+ EFX_SET_OWORD_FIELD(
+ filter_ctl, FRF_BZ_SCATTER_ENBL_NO_MATCH_Q,
+ !!(table->spec[EFX_FILTER_INDEX_UC_DEF].flags &
+ table->spec[EFX_FILTER_INDEX_MC_DEF].flags &
+ EFX_FILTER_FLAG_RX_SCATTER));
+ } else if (efx_nic_rev(efx) >= EFX_REV_FALCON_B0) {
+ /* We don't expose 'default' filters because unmatched
+ * packets always go to the queue number found in the
+ * RSS table. But we still need to set the RX scatter
+ * bit here.
+ */
+ EFX_SET_OWORD_FIELD(
+ filter_ctl, FRF_BZ_SCATTER_ENBL_NO_MATCH_Q,
+ efx->rx_scatter);
}
efx_writeo(efx, &filter_ctl, FR_BZ_RX_FILTER_CTL);
@@ -409,9 +432,18 @@
struct efx_filter_state *state = efx->filter_state;
struct efx_filter_table *table = &state->table[EFX_FILTER_TABLE_RX_DEF];
struct efx_filter_spec *spec = &table->spec[filter_idx];
+ enum efx_filter_flags flags = 0;
- efx_filter_init_rx(spec, EFX_FILTER_PRI_MANUAL,
- EFX_FILTER_FLAG_RX_RSS, 0);
+ /* If there's only one channel then disable RSS for non VF
+ * traffic, thereby allowing VFs to use RSS when the PF can't.
+ */
+ if (efx->n_rx_channels > 1)
+ flags |= EFX_FILTER_FLAG_RX_RSS;
+
+ if (efx->rx_scatter)
+ flags |= EFX_FILTER_FLAG_RX_SCATTER;
+
+ efx_filter_init_rx(spec, EFX_FILTER_PRI_MANUAL, flags, 0);
spec->type = EFX_FILTER_UC_DEF + filter_idx;
table->used_bitmap[0] |= 1 << filter_idx;
}
@@ -463,13 +495,6 @@
break;
}
- case EFX_FILTER_TABLE_RX_DEF:
- /* One filter spec per type */
- BUILD_BUG_ON(EFX_FILTER_INDEX_UC_DEF != 0);
- BUILD_BUG_ON(EFX_FILTER_INDEX_MC_DEF !=
- EFX_FILTER_MC_DEF - EFX_FILTER_UC_DEF);
- return spec->type - EFX_FILTER_UC_DEF;
-
case EFX_FILTER_TABLE_RX_MAC: {
bool is_wild = spec->type == EFX_FILTER_MAC_WILD;
EFX_POPULATE_OWORD_7(
@@ -520,42 +545,6 @@
return true;
}
-static int efx_filter_search(struct efx_filter_table *table,
- struct efx_filter_spec *spec, u32 key,
- bool for_insert, unsigned int *depth_required)
-{
- unsigned hash, incr, filter_idx, depth, depth_max;
-
- hash = efx_filter_hash(key);
- incr = efx_filter_increment(key);
-
- filter_idx = hash & (table->size - 1);
- depth = 1;
- depth_max = (for_insert ?
- (spec->priority <= EFX_FILTER_PRI_HINT ?
- FILTER_CTL_SRCH_HINT_MAX : FILTER_CTL_SRCH_MAX) :
- table->search_depth[spec->type]);
-
- for (;;) {
- /* Return success if entry is used and matches this spec
- * or entry is unused and we are trying to insert.
- */
- if (test_bit(filter_idx, table->used_bitmap) ?
- efx_filter_equal(spec, &table->spec[filter_idx]) :
- for_insert) {
- *depth_required = depth;
- return filter_idx;
- }
-
- /* Return failure if we reached the maximum search depth */
- if (depth == depth_max)
- return for_insert ? -EBUSY : -ENOENT;
-
- filter_idx = (filter_idx + incr) & (table->size - 1);
- ++depth;
- }
-}
-
/*
* Construct/deconstruct external filter IDs. At least the RX filter
* IDs must be ordered by matching priority, for RX NFC semantics.
@@ -650,44 +639,111 @@
* efx_filter_insert_filter - add or replace a filter
* @efx: NIC in which to insert the filter
* @spec: Specification for the filter
- * @replace: Flag for whether the specified filter may replace a filter
- * with an identical match expression and equal or lower priority
+ * @replace_equal: Flag for whether the specified filter may replace an
+ * existing filter with equal priority
*
* On success, return the filter ID.
* On failure, return a negative error code.
+ *
+ * If an existing filter has equal match values to the new filter
+ * spec, then the new filter might replace it, depending on the
+ * relative priorities. If the existing filter has lower priority, or
+ * if @replace_equal is set and it has equal priority, then it is
+ * replaced. Otherwise the function fails, returning -%EPERM if
+ * the existing filter has higher priority or -%EEXIST if it has
+ * equal priority.
*/
s32 efx_filter_insert_filter(struct efx_nic *efx, struct efx_filter_spec *spec,
- bool replace)
+ bool replace_equal)
{
struct efx_filter_state *state = efx->filter_state;
struct efx_filter_table *table = efx_filter_spec_table(state, spec);
- struct efx_filter_spec *saved_spec;
efx_oword_t filter;
- unsigned int filter_idx, depth = 0;
- u32 key;
+ int rep_index, ins_index;
+ unsigned int depth = 0;
int rc;
if (!table || table->size == 0)
return -EINVAL;
- key = efx_filter_build(&filter, spec);
-
netif_vdbg(efx, hw, efx->net_dev,
"%s: type %d search_depth=%d", __func__, spec->type,
table->search_depth[spec->type]);
- spin_lock_bh(&state->lock);
+ if (table->id == EFX_FILTER_TABLE_RX_DEF) {
+ /* One filter spec per type */
+ BUILD_BUG_ON(EFX_FILTER_INDEX_UC_DEF != 0);
+ BUILD_BUG_ON(EFX_FILTER_INDEX_MC_DEF !=
+ EFX_FILTER_MC_DEF - EFX_FILTER_UC_DEF);
+ rep_index = spec->type - EFX_FILTER_INDEX_UC_DEF;
+ ins_index = rep_index;
- rc = efx_filter_search(table, spec, key, true, &depth);
- if (rc < 0)
- goto out;
- filter_idx = rc;
- BUG_ON(filter_idx >= table->size);
- saved_spec = &table->spec[filter_idx];
+ spin_lock_bh(&state->lock);
+ } else {
+ /* Search concurrently for
+ * (1) a filter to be replaced (rep_index): any filter
+ * with the same match values, up to the current
+ * search depth for this type, and
+ * (2) the insertion point (ins_index): (1) or any
+ * free slot before it or up to the maximum search
+ * depth for this priority
+ * We fail if we cannot find (2).
+ *
+ * We can stop once either
+ * (a) we find (1), in which case we have definitely
+ * found (2) as well; or
+ * (b) we have searched exhaustively for (1), and have
+ * either found (2) or searched exhaustively for it
+ */
+ u32 key = efx_filter_build(&filter, spec);
+ unsigned int hash = efx_filter_hash(key);
+ unsigned int incr = efx_filter_increment(key);
+ unsigned int max_rep_depth = table->search_depth[spec->type];
+ unsigned int max_ins_depth =
+ spec->priority <= EFX_FILTER_PRI_HINT ?
+ FILTER_CTL_SRCH_HINT_MAX : FILTER_CTL_SRCH_MAX;
+ unsigned int i = hash & (table->size - 1);
- if (test_bit(filter_idx, table->used_bitmap)) {
- /* Should we replace the existing filter? */
- if (!replace) {
+ ins_index = -1;
+ depth = 1;
+
+ spin_lock_bh(&state->lock);
+
+ for (;;) {
+ if (!test_bit(i, table->used_bitmap)) {
+ if (ins_index < 0)
+ ins_index = i;
+ } else if (efx_filter_equal(spec, &table->spec[i])) {
+ /* Case (a) */
+ if (ins_index < 0)
+ ins_index = i;
+ rep_index = i;
+ break;
+ }
+
+ if (depth >= max_rep_depth &&
+ (ins_index >= 0 || depth >= max_ins_depth)) {
+ /* Case (b) */
+ if (ins_index < 0) {
+ rc = -EBUSY;
+ goto out;
+ }
+ rep_index = -1;
+ break;
+ }
+
+ i = (i + incr) & (table->size - 1);
+ ++depth;
+ }
+ }
+
+ /* If we found a filter to be replaced, check whether we
+ * should do so
+ */
+ if (rep_index >= 0) {
+ struct efx_filter_spec *saved_spec = &table->spec[rep_index];
+
+ if (spec->priority == saved_spec->priority && !replace_equal) {
rc = -EEXIST;
goto out;
}
@@ -695,11 +751,14 @@
rc = -EPERM;
goto out;
}
- } else {
- __set_bit(filter_idx, table->used_bitmap);
+ }
+
+ /* Insert the filter */
+ if (ins_index != rep_index) {
+ __set_bit(ins_index, table->used_bitmap);
++table->used;
}
- *saved_spec = *spec;
+ table->spec[ins_index] = *spec;
if (table->id == EFX_FILTER_TABLE_RX_DEF) {
efx_filter_push_rx_config(efx);
@@ -713,13 +772,19 @@
}
efx_writeo(efx, &filter,
- table->offset + table->step * filter_idx);
+ table->offset + table->step * ins_index);
+
+ /* If we were able to replace a filter by inserting
+ * at a lower depth, clear the replaced filter
+ */
+ if (ins_index != rep_index && rep_index >= 0)
+ efx_filter_table_clear_entry(efx, table, rep_index);
}
netif_vdbg(efx, hw, efx->net_dev,
"%s: filter type %d index %d rxq %u set",
- __func__, spec->type, filter_idx, spec->dmaq_id);
- rc = efx_filter_make_id(spec, filter_idx);
+ __func__, spec->type, ins_index, spec->dmaq_id);
+ rc = efx_filter_make_id(spec, ins_index);
out:
spin_unlock_bh(&state->lock);
@@ -1060,6 +1125,50 @@
kfree(state);
}
+/* Update scatter enable flags for filters pointing to our own RX queues */
+void efx_filter_update_rx_scatter(struct efx_nic *efx)
+{
+ struct efx_filter_state *state = efx->filter_state;
+ enum efx_filter_table_id table_id;
+ struct efx_filter_table *table;
+ efx_oword_t filter;
+ unsigned int filter_idx;
+
+ spin_lock_bh(&state->lock);
+
+ for (table_id = EFX_FILTER_TABLE_RX_IP;
+ table_id <= EFX_FILTER_TABLE_RX_DEF;
+ table_id++) {
+ table = &state->table[table_id];
+
+ for (filter_idx = 0; filter_idx < table->size; filter_idx++) {
+ if (!test_bit(filter_idx, table->used_bitmap) ||
+ table->spec[filter_idx].dmaq_id >=
+ efx->n_rx_channels)
+ continue;
+
+ if (efx->rx_scatter)
+ table->spec[filter_idx].flags |=
+ EFX_FILTER_FLAG_RX_SCATTER;
+ else
+ table->spec[filter_idx].flags &=
+ ~EFX_FILTER_FLAG_RX_SCATTER;
+
+ if (table_id == EFX_FILTER_TABLE_RX_DEF)
+ /* Pushed by efx_filter_push_rx_config() */
+ continue;
+
+ efx_filter_build(&filter, &table->spec[filter_idx]);
+ efx_writeo(efx, &filter,
+ table->offset + table->step * filter_idx);
+ }
+ }
+
+ efx_filter_push_rx_config(efx);
+
+ spin_unlock_bh(&state->lock);
+}
+
#ifdef CONFIG_RFS_ACCEL
int efx_filter_rfs(struct net_device *net_dev, const struct sk_buff *skb,
diff --git a/drivers/net/ethernet/sfc/mcdi_pcol.h b/drivers/net/ethernet/sfc/mcdi_pcol.h
index 9d426d0..c5c9747 100644
--- a/drivers/net/ethernet/sfc/mcdi_pcol.h
+++ b/drivers/net/ethernet/sfc/mcdi_pcol.h
@@ -553,6 +553,7 @@
#define MC_CMD_PTP_MODE_V1_VLAN 0x1 /* enum */
#define MC_CMD_PTP_MODE_V2 0x2 /* enum */
#define MC_CMD_PTP_MODE_V2_VLAN 0x3 /* enum */
+#define MC_CMD_PTP_MODE_V2_ENHANCED 0x4 /* enum */
/* MC_CMD_PTP_IN_DISABLE msgrequest */
#define MC_CMD_PTP_IN_DISABLE_LEN 8
diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h
index 0a90abd..9bd433a 100644
--- a/drivers/net/ethernet/sfc/net_driver.h
+++ b/drivers/net/ethernet/sfc/net_driver.h
@@ -69,6 +69,12 @@
#define EFX_TXQ_TYPES 4
#define EFX_MAX_TX_QUEUES (EFX_TXQ_TYPES * EFX_MAX_CHANNELS)
+/* Maximum possible MTU the driver supports */
+#define EFX_MAX_MTU (9 * 1024)
+
+/* Size of an RX scatter buffer. Small enough to pack 2 into a 4K page. */
+#define EFX_RX_USR_BUF_SIZE 1824
+
/* Forward declare Precision Time Protocol (PTP) support structure. */
struct efx_ptp_data;
@@ -206,25 +212,23 @@
/**
* struct efx_rx_buffer - An Efx RX data buffer
* @dma_addr: DMA base address of the buffer
- * @skb: The associated socket buffer. Valid iff !(@flags & %EFX_RX_BUF_PAGE).
+ * @page: The associated page buffer.
* Will be %NULL if the buffer slot is currently free.
- * @page: The associated page buffer. Valif iff @flags & %EFX_RX_BUF_PAGE.
- * Will be %NULL if the buffer slot is currently free.
- * @page_offset: Offset within page. Valid iff @flags & %EFX_RX_BUF_PAGE.
- * @len: Buffer length, in bytes.
- * @flags: Flags for buffer and packet state.
+ * @page_offset: If pending: offset in @page of DMA base address.
+ * If completed: offset in @page of Ethernet header.
+ * @len: If pending: length for DMA descriptor.
+ * If completed: received length, excluding hash prefix.
+ * @flags: Flags for buffer and packet state. These are only set on the
+ * first buffer of a scattered packet.
*/
struct efx_rx_buffer {
dma_addr_t dma_addr;
- union {
- struct sk_buff *skb;
- struct page *page;
- } u;
+ struct page *page;
u16 page_offset;
u16 len;
u16 flags;
};
-#define EFX_RX_BUF_PAGE 0x0001
+#define EFX_RX_BUF_LAST_IN_PAGE 0x0001
#define EFX_RX_PKT_CSUMMED 0x0002
#define EFX_RX_PKT_DISCARD 0x0004
@@ -260,14 +264,23 @@
* @added_count: Number of buffers added to the receive queue.
* @notified_count: Number of buffers given to NIC (<= @added_count).
* @removed_count: Number of buffers removed from the receive queue.
+ * @scatter_n: Number of buffers used by current packet
+ * @page_ring: The ring to store DMA mapped pages for reuse.
+ * @page_add: Counter to calculate the write pointer for the recycle ring.
+ * @page_remove: Counter to calculate the read pointer for the recycle ring.
+ * @page_recycle_count: The number of pages that have been recycled.
+ * @page_recycle_failed: The number of pages that couldn't be recycled because
+ * the kernel still held a reference to them.
+ * @page_recycle_full: The number of pages that were released because the
+ * recycle ring was full.
+ * @page_ptr_mask: The number of pages in the RX recycle ring minus 1.
* @max_fill: RX descriptor maximum fill level (<= ring size)
* @fast_fill_trigger: RX descriptor fill level that will trigger a fast fill
* (<= @max_fill)
* @min_fill: RX descriptor minimum non-zero fill level.
* This records the minimum fill level observed when a ring
* refill was triggered.
- * @alloc_page_count: RX allocation strategy counter.
- * @alloc_skb_count: RX allocation strategy counter.
+ * @recycle_count: RX buffer recycle counter.
* @slow_fill: Timer used to defer efx_nic_generate_fill_event().
*/
struct efx_rx_queue {
@@ -279,15 +292,22 @@
bool enabled;
bool flush_pending;
- int added_count;
- int notified_count;
- int removed_count;
+ unsigned int added_count;
+ unsigned int notified_count;
+ unsigned int removed_count;
+ unsigned int scatter_n;
+ struct page **page_ring;
+ unsigned int page_add;
+ unsigned int page_remove;
+ unsigned int page_recycle_count;
+ unsigned int page_recycle_failed;
+ unsigned int page_recycle_full;
+ unsigned int page_ptr_mask;
unsigned int max_fill;
unsigned int fast_fill_trigger;
unsigned int min_fill;
unsigned int min_overfill;
- unsigned int alloc_page_count;
- unsigned int alloc_skb_count;
+ unsigned int recycle_count;
struct timer_list slow_fill;
unsigned int slow_fill_count;
};
@@ -336,10 +356,6 @@
* @event_test_cpu: Last CPU to handle interrupt or test event for this channel
* @irq_count: Number of IRQs since last adaptive moderation decision
* @irq_mod_score: IRQ moderation score
- * @rx_alloc_level: Watermark based heuristic counter for pushing descriptors
- * and diagnostic counters
- * @rx_alloc_push_pages: RX allocation method currently in use for pushing
- * descriptors
* @n_rx_tobe_disc: Count of RX_TOBE_DISC errors
* @n_rx_ip_hdr_chksum_err: Count of RX IP header checksum errors
* @n_rx_tcp_udp_chksum_err: Count of RX TCP and UDP checksum errors
@@ -347,6 +363,12 @@
* @n_rx_frm_trunc: Count of RX_FRM_TRUNC errors
* @n_rx_overlength: Count of RX_OVERLENGTH errors
* @n_skbuff_leaks: Count of skbuffs leaked due to RX overrun
+ * @n_rx_nodesc_trunc: Number of RX packets truncated and then dropped due to
+ * lack of descriptors
+ * @rx_pkt_n_frags: Number of fragments in next packet to be delivered by
+ * __efx_rx_packet(), or zero if there is none
+ * @rx_pkt_index: Ring index of first buffer for next packet to be delivered
+ * by __efx_rx_packet(), if @rx_pkt_n_frags != 0
* @rx_queue: RX queue for this channel
* @tx_queue: TX queues for this channel
*/
@@ -371,9 +393,6 @@
unsigned int rfs_filters_added;
#endif
- int rx_alloc_level;
- int rx_alloc_push_pages;
-
unsigned n_rx_tobe_disc;
unsigned n_rx_ip_hdr_chksum_err;
unsigned n_rx_tcp_udp_chksum_err;
@@ -381,11 +400,10 @@
unsigned n_rx_frm_trunc;
unsigned n_rx_overlength;
unsigned n_skbuff_leaks;
+ unsigned int n_rx_nodesc_trunc;
- /* Used to pipeline received packets in order to optimise memory
- * access with prefetches.
- */
- struct efx_rx_buffer *rx_pkt;
+ unsigned int rx_pkt_n_frags;
+ unsigned int rx_pkt_index;
struct efx_rx_queue rx_queue;
struct efx_tx_queue tx_queue[EFX_TXQ_TYPES];
@@ -410,7 +428,7 @@
void (*post_remove)(struct efx_channel *);
void (*get_name)(struct efx_channel *, char *buf, size_t len);
struct efx_channel *(*copy)(const struct efx_channel *);
- void (*receive_skb)(struct efx_channel *, struct sk_buff *);
+ bool (*receive_skb)(struct efx_channel *, struct sk_buff *);
bool keep_eventq;
};
@@ -446,6 +464,7 @@
STATE_UNINIT = 0, /* device being probed/removed or is frozen */
STATE_READY = 1, /* hardware ready and netdev registered */
STATE_DISABLED = 2, /* device disabled due to hardware errors */
+ STATE_RECOVERY = 3, /* device recovering from PCI error */
};
/*
@@ -684,10 +703,13 @@
* @n_channels: Number of channels in use
* @n_rx_channels: Number of channels used for RX (= number of RX queues)
* @n_tx_channels: Number of channels used for TX
- * @rx_buffer_len: RX buffer length
+ * @rx_dma_len: Current maximum RX DMA length
* @rx_buffer_order: Order (log2) of number of pages for each RX buffer
+ * @rx_buffer_truesize: Amortised allocation size of an RX buffer,
+ * for use in sk_buff::truesize
* @rx_hash_key: Toeplitz hash key for RSS
* @rx_indir_table: Indirection table for RSS
+ * @rx_scatter: Scatter mode enabled for receives
* @int_error_count: Number of internal errors seen recently
* @int_error_expire: Time at which error count will be expired
* @irq_status: Interrupt status buffer
@@ -800,10 +822,15 @@
unsigned rss_spread;
unsigned tx_channel_offset;
unsigned n_tx_channels;
- unsigned int rx_buffer_len;
+ unsigned int rx_dma_len;
unsigned int rx_buffer_order;
+ unsigned int rx_buffer_truesize;
+ unsigned int rx_page_buf_step;
+ unsigned int rx_bufs_per_page;
+ unsigned int rx_pages_per_batch;
u8 rx_hash_key[40];
u32 rx_indir_table[128];
+ bool rx_scatter;
unsigned int_error_count;
unsigned long int_error_expire;
@@ -934,8 +961,9 @@
* @evq_ptr_tbl_base: Event queue pointer table base address
* @evq_rptr_tbl_base: Event queue read-pointer table base address
* @max_dma_mask: Maximum possible DMA mask
- * @rx_buffer_hash_size: Size of hash at start of RX buffer
- * @rx_buffer_padding: Size of padding at end of RX buffer
+ * @rx_buffer_hash_size: Size of hash at start of RX packet
+ * @rx_buffer_padding: Size of padding at end of RX packet
+ * @can_rx_scatter: NIC is able to scatter packet to multiple buffers
* @max_interrupt_mode: Highest capability interrupt mode supported
* from &enum efx_init_mode.
* @phys_addr_channels: Number of channels with physically addressed
@@ -983,6 +1011,7 @@
u64 max_dma_mask;
unsigned int rx_buffer_hash_size;
unsigned int rx_buffer_padding;
+ bool can_rx_scatter;
unsigned int max_interrupt_mode;
unsigned int phys_addr_channels;
unsigned int timer_period_max;
diff --git a/drivers/net/ethernet/sfc/nic.c b/drivers/net/ethernet/sfc/nic.c
index eaa8e87..b0503cd 100644
--- a/drivers/net/ethernet/sfc/nic.c
+++ b/drivers/net/ethernet/sfc/nic.c
@@ -305,11 +305,11 @@
unsigned int len)
{
buffer->addr = dma_alloc_coherent(&efx->pci_dev->dev, len,
- &buffer->dma_addr, GFP_ATOMIC);
+ &buffer->dma_addr,
+ GFP_ATOMIC | __GFP_ZERO);
if (!buffer->addr)
return -ENOMEM;
buffer->len = len;
- memset(buffer->addr, 0, len);
return 0;
}
@@ -592,12 +592,22 @@
struct efx_nic *efx = rx_queue->efx;
bool is_b0 = efx_nic_rev(efx) >= EFX_REV_FALCON_B0;
bool iscsi_digest_en = is_b0;
+ bool jumbo_en;
+
+ /* For kernel-mode queues in Falcon A1, the JUMBO flag enables
+ * DMA to continue after a PCIe page boundary (and scattering
+ * is not possible). In Falcon B0 and Siena, it enables
+ * scatter.
+ */
+ jumbo_en = !is_b0 || efx->rx_scatter;
netif_dbg(efx, hw, efx->net_dev,
"RX queue %d ring in special buffers %d-%d\n",
efx_rx_queue_index(rx_queue), rx_queue->rxd.index,
rx_queue->rxd.index + rx_queue->rxd.entries - 1);
+ rx_queue->scatter_n = 0;
+
/* Pin RX descriptor ring */
efx_init_special_buffer(efx, &rx_queue->rxd);
@@ -614,8 +624,7 @@
FRF_AZ_RX_DESCQ_SIZE,
__ffs(rx_queue->rxd.entries),
FRF_AZ_RX_DESCQ_TYPE, 0 /* kernel queue */ ,
- /* For >=B0 this is scatter so disable */
- FRF_AZ_RX_DESCQ_JUMBO, !is_b0,
+ FRF_AZ_RX_DESCQ_JUMBO, jumbo_en,
FRF_AZ_RX_DESCQ_EN, 1);
efx_writeo_table(efx, &rx_desc_ptr, efx->type->rxd_ptr_tbl_base,
efx_rx_queue_index(rx_queue));
@@ -969,13 +978,24 @@
EFX_RX_PKT_DISCARD : 0;
}
-/* Handle receive events that are not in-order. */
-static void
+/* Handle receive events that are not in-order. Return true if this
+ * can be handled as a partial packet discard, false if it's more
+ * serious.
+ */
+static bool
efx_handle_rx_bad_index(struct efx_rx_queue *rx_queue, unsigned index)
{
+ struct efx_channel *channel = efx_rx_queue_channel(rx_queue);
struct efx_nic *efx = rx_queue->efx;
unsigned expected, dropped;
+ if (rx_queue->scatter_n &&
+ index == ((rx_queue->removed_count + rx_queue->scatter_n - 1) &
+ rx_queue->ptr_mask)) {
+ ++channel->n_rx_nodesc_trunc;
+ return true;
+ }
+
expected = rx_queue->removed_count & rx_queue->ptr_mask;
dropped = (index - expected) & rx_queue->ptr_mask;
netif_info(efx, rx_err, efx->net_dev,
@@ -984,6 +1004,7 @@
efx_schedule_reset(efx, EFX_WORKAROUND_5676(efx) ?
RESET_TYPE_RX_RECOVERY : RESET_TYPE_DISABLE);
+ return false;
}
/* Handle a packet received event
@@ -999,7 +1020,7 @@
unsigned int rx_ev_desc_ptr, rx_ev_byte_cnt;
unsigned int rx_ev_hdr_type, rx_ev_mcast_pkt;
unsigned expected_ptr;
- bool rx_ev_pkt_ok;
+ bool rx_ev_pkt_ok, rx_ev_sop, rx_ev_cont;
u16 flags;
struct efx_rx_queue *rx_queue;
struct efx_nic *efx = channel->efx;
@@ -1007,21 +1028,56 @@
if (unlikely(ACCESS_ONCE(efx->reset_pending)))
return;
- /* Basic packet information */
- rx_ev_byte_cnt = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_BYTE_CNT);
- rx_ev_pkt_ok = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_PKT_OK);
- rx_ev_hdr_type = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_HDR_TYPE);
- WARN_ON(EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_JUMBO_CONT));
- WARN_ON(EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_SOP) != 1);
+ rx_ev_cont = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_JUMBO_CONT);
+ rx_ev_sop = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_SOP);
WARN_ON(EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_Q_LABEL) !=
channel->channel);
rx_queue = efx_channel_get_rx_queue(channel);
rx_ev_desc_ptr = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_DESC_PTR);
- expected_ptr = rx_queue->removed_count & rx_queue->ptr_mask;
- if (unlikely(rx_ev_desc_ptr != expected_ptr))
- efx_handle_rx_bad_index(rx_queue, rx_ev_desc_ptr);
+ expected_ptr = ((rx_queue->removed_count + rx_queue->scatter_n) &
+ rx_queue->ptr_mask);
+
+ /* Check for partial drops and other errors */
+ if (unlikely(rx_ev_desc_ptr != expected_ptr) ||
+ unlikely(rx_ev_sop != (rx_queue->scatter_n == 0))) {
+ if (rx_ev_desc_ptr != expected_ptr &&
+ !efx_handle_rx_bad_index(rx_queue, rx_ev_desc_ptr))
+ return;
+
+ /* Discard all pending fragments */
+ if (rx_queue->scatter_n) {
+ efx_rx_packet(
+ rx_queue,
+ rx_queue->removed_count & rx_queue->ptr_mask,
+ rx_queue->scatter_n, 0, EFX_RX_PKT_DISCARD);
+ rx_queue->removed_count += rx_queue->scatter_n;
+ rx_queue->scatter_n = 0;
+ }
+
+ /* Return if there is no new fragment */
+ if (rx_ev_desc_ptr != expected_ptr)
+ return;
+
+ /* Discard new fragment if not SOP */
+ if (!rx_ev_sop) {
+ efx_rx_packet(
+ rx_queue,
+ rx_queue->removed_count & rx_queue->ptr_mask,
+ 1, 0, EFX_RX_PKT_DISCARD);
+ ++rx_queue->removed_count;
+ return;
+ }
+ }
+
+ ++rx_queue->scatter_n;
+ if (rx_ev_cont)
+ return;
+
+ rx_ev_byte_cnt = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_BYTE_CNT);
+ rx_ev_pkt_ok = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_PKT_OK);
+ rx_ev_hdr_type = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_HDR_TYPE);
if (likely(rx_ev_pkt_ok)) {
/* If packet is marked as OK and packet type is TCP/IP or
@@ -1049,7 +1105,11 @@
channel->irq_mod_score += 2;
/* Handle received packet */
- efx_rx_packet(rx_queue, rx_ev_desc_ptr, rx_ev_byte_cnt, flags);
+ efx_rx_packet(rx_queue,
+ rx_queue->removed_count & rx_queue->ptr_mask,
+ rx_queue->scatter_n, rx_ev_byte_cnt, flags);
+ rx_queue->removed_count += rx_queue->scatter_n;
+ rx_queue->scatter_n = 0;
}
/* If this flush done event corresponds to a &struct efx_tx_queue, then
diff --git a/drivers/net/ethernet/sfc/ptp.c b/drivers/net/ethernet/sfc/ptp.c
index 3f93624..07f6baa 100644
--- a/drivers/net/ethernet/sfc/ptp.c
+++ b/drivers/net/ethernet/sfc/ptp.c
@@ -99,6 +99,9 @@
#define PTP_V2_VERSION_LENGTH 1
#define PTP_V2_VERSION_OFFSET 29
+#define PTP_V2_UUID_LENGTH 8
+#define PTP_V2_UUID_OFFSET 48
+
/* Although PTP V2 UUIDs are comprised a ClockIdentity (8) and PortNumber (2),
* the MC only captures the last six bytes of the clock identity. These values
* reflect those, not the ones used in the standard. The standard permits
@@ -429,13 +432,10 @@
unsigned number_readings = (response_length /
MC_CMD_PTP_OUT_SYNCHRONIZE_TIMESET_LEN);
unsigned i;
- unsigned min;
- unsigned min_set = 0;
unsigned total;
unsigned ngood = 0;
unsigned last_good = 0;
struct efx_ptp_data *ptp = efx->ptp_data;
- bool min_valid = false;
u32 last_sec;
u32 start_sec;
struct timespec delta;
@@ -443,35 +443,17 @@
if (number_readings == 0)
return -EAGAIN;
- /* Find minimum value in this set of results, discarding clearly
- * erroneous results.
+ /* Read the set of results and increment stats for any results that
+ * appera to be erroneous.
*/
for (i = 0; i < number_readings; i++) {
efx_ptp_read_timeset(synch_buf, &ptp->timeset[i]);
synch_buf += MC_CMD_PTP_OUT_SYNCHRONIZE_TIMESET_LEN;
- if (ptp->timeset[i].window > SYNCHRONISATION_GRANULARITY_NS) {
- if (min_valid) {
- if (ptp->timeset[i].window < min_set)
- min_set = ptp->timeset[i].window;
- } else {
- min_valid = true;
- min_set = ptp->timeset[i].window;
- }
- }
}
- if (min_valid) {
- if (ptp->base_sync_valid && (min_set > ptp->base_sync_ns))
- min = ptp->base_sync_ns;
- else
- min = min_set;
- } else {
- min = SYNCHRONISATION_GRANULARITY_NS;
- }
-
- /* Discard excessively long synchronise durations. The MC times
- * when it finishes reading the host time so the corrected window
- * time should be fairly constant for a given platform.
+ /* Find the last good host-MC synchronization result. The MC times
+ * when it finishes reading the host time so the corrected window time
+ * should be fairly constant for a given platform.
*/
total = 0;
for (i = 0; i < number_readings; i++)
@@ -489,8 +471,8 @@
if (ngood == 0) {
netif_warn(efx, drv, efx->net_dev,
- "PTP no suitable synchronisations %dns %dns\n",
- ptp->base_sync_ns, min_set);
+ "PTP no suitable synchronisations %dns\n",
+ ptp->base_sync_ns);
return -EAGAIN;
}
@@ -1006,43 +988,53 @@
* the receive timestamp from the MC - this will probably occur after the
* packet arrival because of the processing in the MC.
*/
-static void efx_ptp_rx(struct efx_channel *channel, struct sk_buff *skb)
+static bool efx_ptp_rx(struct efx_channel *channel, struct sk_buff *skb)
{
struct efx_nic *efx = channel->efx;
struct efx_ptp_data *ptp = efx->ptp_data;
struct efx_ptp_match *match = (struct efx_ptp_match *)skb->cb;
- u8 *data;
+ u8 *match_data_012, *match_data_345;
unsigned int version;
match->expiry = jiffies + msecs_to_jiffies(PKT_EVENT_LIFETIME_MS);
/* Correct version? */
if (ptp->mode == MC_CMD_PTP_MODE_V1) {
- if (skb->len < PTP_V1_MIN_LENGTH) {
- netif_receive_skb(skb);
- return;
+ if (!pskb_may_pull(skb, PTP_V1_MIN_LENGTH)) {
+ return false;
}
version = ntohs(*(__be16 *)&skb->data[PTP_V1_VERSION_OFFSET]);
if (version != PTP_VERSION_V1) {
- netif_receive_skb(skb);
- return;
+ return false;
}
+
+ /* PTP V1 uses all six bytes of the UUID to match the packet
+ * to the timestamp
+ */
+ match_data_012 = skb->data + PTP_V1_UUID_OFFSET;
+ match_data_345 = skb->data + PTP_V1_UUID_OFFSET + 3;
} else {
- if (skb->len < PTP_V2_MIN_LENGTH) {
- netif_receive_skb(skb);
- return;
+ if (!pskb_may_pull(skb, PTP_V2_MIN_LENGTH)) {
+ return false;
}
version = skb->data[PTP_V2_VERSION_OFFSET];
-
- BUG_ON(ptp->mode != MC_CMD_PTP_MODE_V2);
- BUILD_BUG_ON(PTP_V1_UUID_OFFSET != PTP_V2_MC_UUID_OFFSET);
- BUILD_BUG_ON(PTP_V1_UUID_LENGTH != PTP_V2_MC_UUID_LENGTH);
- BUILD_BUG_ON(PTP_V1_SEQUENCE_OFFSET != PTP_V2_SEQUENCE_OFFSET);
- BUILD_BUG_ON(PTP_V1_SEQUENCE_LENGTH != PTP_V2_SEQUENCE_LENGTH);
-
if ((version & PTP_VERSION_V2_MASK) != PTP_VERSION_V2) {
- netif_receive_skb(skb);
- return;
+ return false;
+ }
+
+ /* The original V2 implementation uses bytes 2-7 of
+ * the UUID to match the packet to the timestamp. This
+ * discards two of the bytes of the MAC address used
+ * to create the UUID (SF bug 33070). The PTP V2
+ * enhanced mode fixes this issue and uses bytes 0-2
+ * and byte 5-7 of the UUID.
+ */
+ match_data_345 = skb->data + PTP_V2_UUID_OFFSET + 5;
+ if (ptp->mode == MC_CMD_PTP_MODE_V2) {
+ match_data_012 = skb->data + PTP_V2_UUID_OFFSET + 2;
+ } else {
+ match_data_012 = skb->data + PTP_V2_UUID_OFFSET + 0;
+ BUG_ON(ptp->mode != MC_CMD_PTP_MODE_V2_ENHANCED);
}
}
@@ -1056,14 +1048,19 @@
timestamps = skb_hwtstamps(skb);
memset(timestamps, 0, sizeof(*timestamps));
+ /* We expect the sequence number to be in the same position in
+ * the packet for PTP V1 and V2
+ */
+ BUILD_BUG_ON(PTP_V1_SEQUENCE_OFFSET != PTP_V2_SEQUENCE_OFFSET);
+ BUILD_BUG_ON(PTP_V1_SEQUENCE_LENGTH != PTP_V2_SEQUENCE_LENGTH);
+
/* Extract UUID/Sequence information */
- data = skb->data + PTP_V1_UUID_OFFSET;
- match->words[0] = (data[0] |
- (data[1] << 8) |
- (data[2] << 16) |
- (data[3] << 24));
- match->words[1] = (data[4] |
- (data[5] << 8) |
+ match->words[0] = (match_data_012[0] |
+ (match_data_012[1] << 8) |
+ (match_data_012[2] << 16) |
+ (match_data_345[0] << 24));
+ match->words[1] = (match_data_345[1] |
+ (match_data_345[2] << 8) |
(skb->data[PTP_V1_SEQUENCE_OFFSET +
PTP_V1_SEQUENCE_LENGTH - 1] <<
16));
@@ -1073,6 +1070,8 @@
skb_queue_tail(&ptp->rxq, skb);
queue_work(ptp->workwq, &ptp->work);
+
+ return true;
}
/* Transmit a PTP packet. This has to be transmitted by the MC
@@ -1167,7 +1166,7 @@
* timestamped
*/
init->rx_filter = HWTSTAMP_FILTER_PTP_V2_L4_EVENT;
- new_mode = MC_CMD_PTP_MODE_V2;
+ new_mode = MC_CMD_PTP_MODE_V2_ENHANCED;
enable_wanted = true;
break;
case HWTSTAMP_FILTER_PTP_V2_EVENT:
@@ -1186,7 +1185,14 @@
if (init->tx_type != HWTSTAMP_TX_OFF)
enable_wanted = true;
+ /* Old versions of the firmware do not support the improved
+ * UUID filtering option (SF bug 33070). If the firmware does
+ * not accept the enhanced mode, fall back to the standard PTP
+ * v2 UUID filtering.
+ */
rc = efx_ptp_change_mode(efx, enable_wanted, new_mode);
+ if ((rc != 0) && (new_mode == MC_CMD_PTP_MODE_V2_ENHANCED))
+ rc = efx_ptp_change_mode(efx, enable_wanted, MC_CMD_PTP_MODE_V2);
if (rc != 0)
return rc;
diff --git a/drivers/net/ethernet/sfc/rx.c b/drivers/net/ethernet/sfc/rx.c
index bb579a6..e73e30b 100644
--- a/drivers/net/ethernet/sfc/rx.c
+++ b/drivers/net/ethernet/sfc/rx.c
@@ -16,6 +16,7 @@
#include <linux/udp.h>
#include <linux/prefetch.h>
#include <linux/moduleparam.h>
+#include <linux/iommu.h>
#include <net/ip.h>
#include <net/checksum.h>
#include "net_driver.h"
@@ -24,85 +25,39 @@
#include "selftest.h"
#include "workarounds.h"
-/* Number of RX descriptors pushed at once. */
-#define EFX_RX_BATCH 8
+/* Preferred number of descriptors to fill at once */
+#define EFX_RX_PREFERRED_BATCH 8U
-/* Maximum size of a buffer sharing a page */
-#define EFX_RX_HALF_PAGE ((PAGE_SIZE >> 1) - sizeof(struct efx_rx_page_state))
+/* Number of RX buffers to recycle pages for. When creating the RX page recycle
+ * ring, this number is divided by the number of buffers per page to calculate
+ * the number of pages to store in the RX page recycle ring.
+ */
+#define EFX_RECYCLE_RING_SIZE_IOMMU 4096
+#define EFX_RECYCLE_RING_SIZE_NOIOMMU (2 * EFX_RX_PREFERRED_BATCH)
/* Size of buffer allocated for skb header area. */
#define EFX_SKB_HEADERS 64u
-/*
- * rx_alloc_method - RX buffer allocation method
- *
- * This driver supports two methods for allocating and using RX buffers:
- * each RX buffer may be backed by an skb or by an order-n page.
- *
- * When GRO is in use then the second method has a lower overhead,
- * since we don't have to allocate then free skbs on reassembled frames.
- *
- * Values:
- * - RX_ALLOC_METHOD_AUTO = 0
- * - RX_ALLOC_METHOD_SKB = 1
- * - RX_ALLOC_METHOD_PAGE = 2
- *
- * The heuristic for %RX_ALLOC_METHOD_AUTO is a simple hysteresis count
- * controlled by the parameters below.
- *
- * - Since pushing and popping descriptors are separated by the rx_queue
- * size, so the watermarks should be ~rxd_size.
- * - The performance win by using page-based allocation for GRO is less
- * than the performance hit of using page-based allocation of non-GRO,
- * so the watermarks should reflect this.
- *
- * Per channel we maintain a single variable, updated by each channel:
- *
- * rx_alloc_level += (gro_performed ? RX_ALLOC_FACTOR_GRO :
- * RX_ALLOC_FACTOR_SKB)
- * Per NAPI poll interval, we constrain rx_alloc_level to 0..MAX (which
- * limits the hysteresis), and update the allocation strategy:
- *
- * rx_alloc_method = (rx_alloc_level > RX_ALLOC_LEVEL_GRO ?
- * RX_ALLOC_METHOD_PAGE : RX_ALLOC_METHOD_SKB)
- */
-static int rx_alloc_method = RX_ALLOC_METHOD_AUTO;
-
-#define RX_ALLOC_LEVEL_GRO 0x2000
-#define RX_ALLOC_LEVEL_MAX 0x3000
-#define RX_ALLOC_FACTOR_GRO 1
-#define RX_ALLOC_FACTOR_SKB (-2)
-
/* This is the percentage fill level below which new RX descriptors
* will be added to the RX descriptor ring.
*/
static unsigned int rx_refill_threshold;
+/* Each packet can consume up to ceil(max_frame_len / buffer_size) buffers */
+#define EFX_RX_MAX_FRAGS DIV_ROUND_UP(EFX_MAX_FRAME_LEN(EFX_MAX_MTU), \
+ EFX_RX_USR_BUF_SIZE)
+
/*
* RX maximum head room required.
*
- * This must be at least 1 to prevent overflow and at least 2 to allow
- * pipelined receives.
+ * This must be at least 1 to prevent overflow, plus one packet-worth
+ * to allow pipelined receives.
*/
-#define EFX_RXD_HEAD_ROOM 2
+#define EFX_RXD_HEAD_ROOM (1 + EFX_RX_MAX_FRAGS)
-/* Offset of ethernet header within page */
-static inline unsigned int efx_rx_buf_offset(struct efx_nic *efx,
- struct efx_rx_buffer *buf)
+static inline u8 *efx_rx_buf_va(struct efx_rx_buffer *buf)
{
- return buf->page_offset + efx->type->rx_buffer_hash_size;
-}
-static inline unsigned int efx_rx_buf_size(struct efx_nic *efx)
-{
- return PAGE_SIZE << efx->rx_buffer_order;
-}
-
-static u8 *efx_rx_buf_eh(struct efx_nic *efx, struct efx_rx_buffer *buf)
-{
- if (buf->flags & EFX_RX_BUF_PAGE)
- return page_address(buf->u.page) + efx_rx_buf_offset(efx, buf);
- else
- return (u8 *)buf->u.skb->data + efx->type->rx_buffer_hash_size;
+ return page_address(buf->page) + buf->page_offset;
}
static inline u32 efx_rx_buf_hash(const u8 *eh)
@@ -119,66 +74,81 @@
#endif
}
-/**
- * efx_init_rx_buffers_skb - create EFX_RX_BATCH skb-based RX buffers
- *
- * @rx_queue: Efx RX queue
- *
- * This allocates EFX_RX_BATCH skbs, maps them for DMA, and populates a
- * struct efx_rx_buffer for each one. Return a negative error code or 0
- * on success. May fail having only inserted fewer than EFX_RX_BATCH
- * buffers.
- */
-static int efx_init_rx_buffers_skb(struct efx_rx_queue *rx_queue)
+static inline struct efx_rx_buffer *
+efx_rx_buf_next(struct efx_rx_queue *rx_queue, struct efx_rx_buffer *rx_buf)
+{
+ if (unlikely(rx_buf == efx_rx_buffer(rx_queue, rx_queue->ptr_mask)))
+ return efx_rx_buffer(rx_queue, 0);
+ else
+ return rx_buf + 1;
+}
+
+static inline void efx_sync_rx_buffer(struct efx_nic *efx,
+ struct efx_rx_buffer *rx_buf,
+ unsigned int len)
+{
+ dma_sync_single_for_cpu(&efx->pci_dev->dev, rx_buf->dma_addr, len,
+ DMA_FROM_DEVICE);
+}
+
+void efx_rx_config_page_split(struct efx_nic *efx)
+{
+ efx->rx_page_buf_step = ALIGN(efx->rx_dma_len + EFX_PAGE_IP_ALIGN,
+ L1_CACHE_BYTES);
+ efx->rx_bufs_per_page = efx->rx_buffer_order ? 1 :
+ ((PAGE_SIZE - sizeof(struct efx_rx_page_state)) /
+ efx->rx_page_buf_step);
+ efx->rx_buffer_truesize = (PAGE_SIZE << efx->rx_buffer_order) /
+ efx->rx_bufs_per_page;
+ efx->rx_pages_per_batch = DIV_ROUND_UP(EFX_RX_PREFERRED_BATCH,
+ efx->rx_bufs_per_page);
+}
+
+/* Check the RX page recycle ring for a page that can be reused. */
+static struct page *efx_reuse_page(struct efx_rx_queue *rx_queue)
{
struct efx_nic *efx = rx_queue->efx;
- struct net_device *net_dev = efx->net_dev;
- struct efx_rx_buffer *rx_buf;
- struct sk_buff *skb;
- int skb_len = efx->rx_buffer_len;
- unsigned index, count;
+ struct page *page;
+ struct efx_rx_page_state *state;
+ unsigned index;
- for (count = 0; count < EFX_RX_BATCH; ++count) {
- index = rx_queue->added_count & rx_queue->ptr_mask;
- rx_buf = efx_rx_buffer(rx_queue, index);
+ index = rx_queue->page_remove & rx_queue->page_ptr_mask;
+ page = rx_queue->page_ring[index];
+ if (page == NULL)
+ return NULL;
- rx_buf->u.skb = skb = netdev_alloc_skb(net_dev, skb_len);
- if (unlikely(!skb))
- return -ENOMEM;
+ rx_queue->page_ring[index] = NULL;
+ /* page_remove cannot exceed page_add. */
+ if (rx_queue->page_remove != rx_queue->page_add)
+ ++rx_queue->page_remove;
- /* Adjust the SKB for padding */
- skb_reserve(skb, NET_IP_ALIGN);
- rx_buf->len = skb_len - NET_IP_ALIGN;
- rx_buf->flags = 0;
-
- rx_buf->dma_addr = dma_map_single(&efx->pci_dev->dev,
- skb->data, rx_buf->len,
- DMA_FROM_DEVICE);
- if (unlikely(dma_mapping_error(&efx->pci_dev->dev,
- rx_buf->dma_addr))) {
- dev_kfree_skb_any(skb);
- rx_buf->u.skb = NULL;
- return -EIO;
- }
-
- ++rx_queue->added_count;
- ++rx_queue->alloc_skb_count;
+ /* If page_count is 1 then we hold the only reference to this page. */
+ if (page_count(page) == 1) {
+ ++rx_queue->page_recycle_count;
+ return page;
+ } else {
+ state = page_address(page);
+ dma_unmap_page(&efx->pci_dev->dev, state->dma_addr,
+ PAGE_SIZE << efx->rx_buffer_order,
+ DMA_FROM_DEVICE);
+ put_page(page);
+ ++rx_queue->page_recycle_failed;
}
- return 0;
+ return NULL;
}
/**
- * efx_init_rx_buffers_page - create EFX_RX_BATCH page-based RX buffers
+ * efx_init_rx_buffers - create EFX_RX_BATCH page-based RX buffers
*
* @rx_queue: Efx RX queue
*
- * This allocates memory for EFX_RX_BATCH receive buffers, maps them for DMA,
- * and populates struct efx_rx_buffers for each one. Return a negative error
- * code or 0 on success. If a single page can be split between two buffers,
- * then the page will either be inserted fully, or not at at all.
+ * This allocates a batch of pages, maps them for DMA, and populates
+ * struct efx_rx_buffers for each one. Return a negative error code or
+ * 0 on success. If a single page can be used for multiple buffers,
+ * then the page will either be inserted fully, or not at all.
*/
-static int efx_init_rx_buffers_page(struct efx_rx_queue *rx_queue)
+static int efx_init_rx_buffers(struct efx_rx_queue *rx_queue)
{
struct efx_nic *efx = rx_queue->efx;
struct efx_rx_buffer *rx_buf;
@@ -188,150 +158,140 @@
dma_addr_t dma_addr;
unsigned index, count;
- /* We can split a page between two buffers */
- BUILD_BUG_ON(EFX_RX_BATCH & 1);
-
- for (count = 0; count < EFX_RX_BATCH; ++count) {
- page = alloc_pages(__GFP_COLD | __GFP_COMP | GFP_ATOMIC,
- efx->rx_buffer_order);
- if (unlikely(page == NULL))
- return -ENOMEM;
- dma_addr = dma_map_page(&efx->pci_dev->dev, page, 0,
- efx_rx_buf_size(efx),
- DMA_FROM_DEVICE);
- if (unlikely(dma_mapping_error(&efx->pci_dev->dev, dma_addr))) {
- __free_pages(page, efx->rx_buffer_order);
- return -EIO;
+ count = 0;
+ do {
+ page = efx_reuse_page(rx_queue);
+ if (page == NULL) {
+ page = alloc_pages(__GFP_COLD | __GFP_COMP | GFP_ATOMIC,
+ efx->rx_buffer_order);
+ if (unlikely(page == NULL))
+ return -ENOMEM;
+ dma_addr =
+ dma_map_page(&efx->pci_dev->dev, page, 0,
+ PAGE_SIZE << efx->rx_buffer_order,
+ DMA_FROM_DEVICE);
+ if (unlikely(dma_mapping_error(&efx->pci_dev->dev,
+ dma_addr))) {
+ __free_pages(page, efx->rx_buffer_order);
+ return -EIO;
+ }
+ state = page_address(page);
+ state->dma_addr = dma_addr;
+ } else {
+ state = page_address(page);
+ dma_addr = state->dma_addr;
}
- state = page_address(page);
- state->refcnt = 0;
- state->dma_addr = dma_addr;
dma_addr += sizeof(struct efx_rx_page_state);
page_offset = sizeof(struct efx_rx_page_state);
- split:
- index = rx_queue->added_count & rx_queue->ptr_mask;
- rx_buf = efx_rx_buffer(rx_queue, index);
- rx_buf->dma_addr = dma_addr + EFX_PAGE_IP_ALIGN;
- rx_buf->u.page = page;
- rx_buf->page_offset = page_offset + EFX_PAGE_IP_ALIGN;
- rx_buf->len = efx->rx_buffer_len - EFX_PAGE_IP_ALIGN;
- rx_buf->flags = EFX_RX_BUF_PAGE;
- ++rx_queue->added_count;
- ++rx_queue->alloc_page_count;
- ++state->refcnt;
-
- if ((~count & 1) && (efx->rx_buffer_len <= EFX_RX_HALF_PAGE)) {
- /* Use the second half of the page */
+ do {
+ index = rx_queue->added_count & rx_queue->ptr_mask;
+ rx_buf = efx_rx_buffer(rx_queue, index);
+ rx_buf->dma_addr = dma_addr + EFX_PAGE_IP_ALIGN;
+ rx_buf->page = page;
+ rx_buf->page_offset = page_offset + EFX_PAGE_IP_ALIGN;
+ rx_buf->len = efx->rx_dma_len;
+ rx_buf->flags = 0;
+ ++rx_queue->added_count;
get_page(page);
- dma_addr += (PAGE_SIZE >> 1);
- page_offset += (PAGE_SIZE >> 1);
- ++count;
- goto split;
- }
- }
+ dma_addr += efx->rx_page_buf_step;
+ page_offset += efx->rx_page_buf_step;
+ } while (page_offset + efx->rx_page_buf_step <= PAGE_SIZE);
+
+ rx_buf->flags = EFX_RX_BUF_LAST_IN_PAGE;
+ } while (++count < efx->rx_pages_per_batch);
return 0;
}
+/* Unmap a DMA-mapped page. This function is only called for the final RX
+ * buffer in a page.
+ */
static void efx_unmap_rx_buffer(struct efx_nic *efx,
- struct efx_rx_buffer *rx_buf,
- unsigned int used_len)
+ struct efx_rx_buffer *rx_buf)
{
- if ((rx_buf->flags & EFX_RX_BUF_PAGE) && rx_buf->u.page) {
- struct efx_rx_page_state *state;
+ struct page *page = rx_buf->page;
- state = page_address(rx_buf->u.page);
- if (--state->refcnt == 0) {
- dma_unmap_page(&efx->pci_dev->dev,
- state->dma_addr,
- efx_rx_buf_size(efx),
- DMA_FROM_DEVICE);
- } else if (used_len) {
- dma_sync_single_for_cpu(&efx->pci_dev->dev,
- rx_buf->dma_addr, used_len,
- DMA_FROM_DEVICE);
- }
- } else if (!(rx_buf->flags & EFX_RX_BUF_PAGE) && rx_buf->u.skb) {
- dma_unmap_single(&efx->pci_dev->dev, rx_buf->dma_addr,
- rx_buf->len, DMA_FROM_DEVICE);
+ if (page) {
+ struct efx_rx_page_state *state = page_address(page);
+ dma_unmap_page(&efx->pci_dev->dev,
+ state->dma_addr,
+ PAGE_SIZE << efx->rx_buffer_order,
+ DMA_FROM_DEVICE);
}
}
-static void efx_free_rx_buffer(struct efx_nic *efx,
- struct efx_rx_buffer *rx_buf)
+static void efx_free_rx_buffer(struct efx_rx_buffer *rx_buf)
{
- if ((rx_buf->flags & EFX_RX_BUF_PAGE) && rx_buf->u.page) {
- __free_pages(rx_buf->u.page, efx->rx_buffer_order);
- rx_buf->u.page = NULL;
- } else if (!(rx_buf->flags & EFX_RX_BUF_PAGE) && rx_buf->u.skb) {
- dev_kfree_skb_any(rx_buf->u.skb);
- rx_buf->u.skb = NULL;
+ if (rx_buf->page) {
+ put_page(rx_buf->page);
+ rx_buf->page = NULL;
}
}
+/* Attempt to recycle the page if there is an RX recycle ring; the page can
+ * only be added if this is the final RX buffer, to prevent pages being used in
+ * the descriptor ring and appearing in the recycle ring simultaneously.
+ */
+static void efx_recycle_rx_page(struct efx_channel *channel,
+ struct efx_rx_buffer *rx_buf)
+{
+ struct page *page = rx_buf->page;
+ struct efx_rx_queue *rx_queue = efx_channel_get_rx_queue(channel);
+ struct efx_nic *efx = rx_queue->efx;
+ unsigned index;
+
+ /* Only recycle the page after processing the final buffer. */
+ if (!(rx_buf->flags & EFX_RX_BUF_LAST_IN_PAGE))
+ return;
+
+ index = rx_queue->page_add & rx_queue->page_ptr_mask;
+ if (rx_queue->page_ring[index] == NULL) {
+ unsigned read_index = rx_queue->page_remove &
+ rx_queue->page_ptr_mask;
+
+ /* The next slot in the recycle ring is available, but
+ * increment page_remove if the read pointer currently
+ * points here.
+ */
+ if (read_index == index)
+ ++rx_queue->page_remove;
+ rx_queue->page_ring[index] = page;
+ ++rx_queue->page_add;
+ return;
+ }
+ ++rx_queue->page_recycle_full;
+ efx_unmap_rx_buffer(efx, rx_buf);
+ put_page(rx_buf->page);
+}
+
static void efx_fini_rx_buffer(struct efx_rx_queue *rx_queue,
struct efx_rx_buffer *rx_buf)
{
- efx_unmap_rx_buffer(rx_queue->efx, rx_buf, 0);
- efx_free_rx_buffer(rx_queue->efx, rx_buf);
-}
+ /* Release the page reference we hold for the buffer. */
+ if (rx_buf->page)
+ put_page(rx_buf->page);
-/* Attempt to resurrect the other receive buffer that used to share this page,
- * which had previously been passed up to the kernel and freed. */
-static void efx_resurrect_rx_buffer(struct efx_rx_queue *rx_queue,
- struct efx_rx_buffer *rx_buf)
-{
- struct efx_rx_page_state *state = page_address(rx_buf->u.page);
- struct efx_rx_buffer *new_buf;
- unsigned fill_level, index;
-
- /* +1 because efx_rx_packet() incremented removed_count. +1 because
- * we'd like to insert an additional descriptor whilst leaving
- * EFX_RXD_HEAD_ROOM for the non-recycle path */
- fill_level = (rx_queue->added_count - rx_queue->removed_count + 2);
- if (unlikely(fill_level > rx_queue->max_fill)) {
- /* We could place "state" on a list, and drain the list in
- * efx_fast_push_rx_descriptors(). For now, this will do. */
- return;
+ /* If this is the last buffer in a page, unmap and free it. */
+ if (rx_buf->flags & EFX_RX_BUF_LAST_IN_PAGE) {
+ efx_unmap_rx_buffer(rx_queue->efx, rx_buf);
+ efx_free_rx_buffer(rx_buf);
}
-
- ++state->refcnt;
- get_page(rx_buf->u.page);
-
- index = rx_queue->added_count & rx_queue->ptr_mask;
- new_buf = efx_rx_buffer(rx_queue, index);
- new_buf->dma_addr = rx_buf->dma_addr ^ (PAGE_SIZE >> 1);
- new_buf->u.page = rx_buf->u.page;
- new_buf->len = rx_buf->len;
- new_buf->flags = EFX_RX_BUF_PAGE;
- ++rx_queue->added_count;
+ rx_buf->page = NULL;
}
-/* Recycle the given rx buffer directly back into the rx_queue. There is
- * always room to add this buffer, because we've just popped a buffer. */
-static void efx_recycle_rx_buffer(struct efx_channel *channel,
- struct efx_rx_buffer *rx_buf)
+/* Recycle the pages that are used by buffers that have just been received. */
+static void efx_recycle_rx_buffers(struct efx_channel *channel,
+ struct efx_rx_buffer *rx_buf,
+ unsigned int n_frags)
{
- struct efx_nic *efx = channel->efx;
struct efx_rx_queue *rx_queue = efx_channel_get_rx_queue(channel);
- struct efx_rx_buffer *new_buf;
- unsigned index;
- rx_buf->flags &= EFX_RX_BUF_PAGE;
-
- if ((rx_buf->flags & EFX_RX_BUF_PAGE) &&
- efx->rx_buffer_len <= EFX_RX_HALF_PAGE &&
- page_count(rx_buf->u.page) == 1)
- efx_resurrect_rx_buffer(rx_queue, rx_buf);
-
- index = rx_queue->added_count & rx_queue->ptr_mask;
- new_buf = efx_rx_buffer(rx_queue, index);
-
- memcpy(new_buf, rx_buf, sizeof(*new_buf));
- rx_buf->u.page = NULL;
- ++rx_queue->added_count;
+ do {
+ efx_recycle_rx_page(channel, rx_buf);
+ rx_buf = efx_rx_buf_next(rx_queue, rx_buf);
+ } while (--n_frags);
}
/**
@@ -348,8 +308,8 @@
*/
void efx_fast_push_rx_descriptors(struct efx_rx_queue *rx_queue)
{
- struct efx_channel *channel = efx_rx_queue_channel(rx_queue);
- unsigned fill_level;
+ struct efx_nic *efx = rx_queue->efx;
+ unsigned int fill_level, batch_size;
int space, rc = 0;
/* Calculate current fill level, and exit if we don't need to fill */
@@ -364,28 +324,26 @@
rx_queue->min_fill = fill_level;
}
+ batch_size = efx->rx_pages_per_batch * efx->rx_bufs_per_page;
space = rx_queue->max_fill - fill_level;
- EFX_BUG_ON_PARANOID(space < EFX_RX_BATCH);
+ EFX_BUG_ON_PARANOID(space < batch_size);
netif_vdbg(rx_queue->efx, rx_status, rx_queue->efx->net_dev,
"RX queue %d fast-filling descriptor ring from"
- " level %d to level %d using %s allocation\n",
+ " level %d to level %d\n",
efx_rx_queue_index(rx_queue), fill_level,
- rx_queue->max_fill,
- channel->rx_alloc_push_pages ? "page" : "skb");
+ rx_queue->max_fill);
+
do {
- if (channel->rx_alloc_push_pages)
- rc = efx_init_rx_buffers_page(rx_queue);
- else
- rc = efx_init_rx_buffers_skb(rx_queue);
+ rc = efx_init_rx_buffers(rx_queue);
if (unlikely(rc)) {
/* Ensure that we don't leave the rx queue empty */
if (rx_queue->added_count == rx_queue->removed_count)
efx_schedule_slow_fill(rx_queue);
goto out;
}
- } while ((space -= EFX_RX_BATCH) >= EFX_RX_BATCH);
+ } while ((space -= batch_size) >= batch_size);
netif_vdbg(rx_queue->efx, rx_status, rx_queue->efx->net_dev,
"RX queue %d fast-filled descriptor ring "
@@ -408,7 +366,7 @@
static void efx_rx_packet__check_len(struct efx_rx_queue *rx_queue,
struct efx_rx_buffer *rx_buf,
- int len, bool *leak_packet)
+ int len)
{
struct efx_nic *efx = rx_queue->efx;
unsigned max_len = rx_buf->len - efx->type->rx_buffer_padding;
@@ -428,11 +386,6 @@
"RX event (0x%x > 0x%x+0x%x). Leaking\n",
efx_rx_queue_index(rx_queue), len, max_len,
efx->type->rx_buffer_padding);
- /* If this buffer was skb-allocated, then the meta
- * data at the end of the skb will be trashed. So
- * we have no choice but to leak the fragment.
- */
- *leak_packet = !(rx_buf->flags & EFX_RX_BUF_PAGE);
efx_schedule_reset(efx, RESET_TYPE_RX_RECOVERY);
} else {
if (net_ratelimit())
@@ -448,212 +401,238 @@
/* Pass a received packet up through GRO. GRO can handle pages
* regardless of checksum state and skbs with a good checksum.
*/
-static void efx_rx_packet_gro(struct efx_channel *channel,
- struct efx_rx_buffer *rx_buf,
- const u8 *eh)
+static void
+efx_rx_packet_gro(struct efx_channel *channel, struct efx_rx_buffer *rx_buf,
+ unsigned int n_frags, u8 *eh)
{
struct napi_struct *napi = &channel->napi_str;
gro_result_t gro_result;
+ struct efx_nic *efx = channel->efx;
+ struct sk_buff *skb;
- if (rx_buf->flags & EFX_RX_BUF_PAGE) {
- struct efx_nic *efx = channel->efx;
- struct page *page = rx_buf->u.page;
- struct sk_buff *skb;
-
- rx_buf->u.page = NULL;
-
- skb = napi_get_frags(napi);
- if (!skb) {
- put_page(page);
- return;
+ skb = napi_get_frags(napi);
+ if (unlikely(!skb)) {
+ while (n_frags--) {
+ put_page(rx_buf->page);
+ rx_buf->page = NULL;
+ rx_buf = efx_rx_buf_next(&channel->rx_queue, rx_buf);
}
-
- if (efx->net_dev->features & NETIF_F_RXHASH)
- skb->rxhash = efx_rx_buf_hash(eh);
-
- skb_fill_page_desc(skb, 0, page,
- efx_rx_buf_offset(efx, rx_buf), rx_buf->len);
-
- skb->len = rx_buf->len;
- skb->data_len = rx_buf->len;
- skb->truesize += rx_buf->len;
- skb->ip_summed = ((rx_buf->flags & EFX_RX_PKT_CSUMMED) ?
- CHECKSUM_UNNECESSARY : CHECKSUM_NONE);
-
- skb_record_rx_queue(skb, channel->rx_queue.core_index);
-
- gro_result = napi_gro_frags(napi);
- } else {
- struct sk_buff *skb = rx_buf->u.skb;
-
- EFX_BUG_ON_PARANOID(!(rx_buf->flags & EFX_RX_PKT_CSUMMED));
- rx_buf->u.skb = NULL;
- skb->ip_summed = CHECKSUM_UNNECESSARY;
-
- gro_result = napi_gro_receive(napi, skb);
+ return;
}
- if (gro_result == GRO_NORMAL) {
- channel->rx_alloc_level += RX_ALLOC_FACTOR_SKB;
- } else if (gro_result != GRO_DROP) {
- channel->rx_alloc_level += RX_ALLOC_FACTOR_GRO;
+ if (efx->net_dev->features & NETIF_F_RXHASH)
+ skb->rxhash = efx_rx_buf_hash(eh);
+ skb->ip_summed = ((rx_buf->flags & EFX_RX_PKT_CSUMMED) ?
+ CHECKSUM_UNNECESSARY : CHECKSUM_NONE);
+
+ for (;;) {
+ skb_fill_page_desc(skb, skb_shinfo(skb)->nr_frags,
+ rx_buf->page, rx_buf->page_offset,
+ rx_buf->len);
+ rx_buf->page = NULL;
+ skb->len += rx_buf->len;
+ if (skb_shinfo(skb)->nr_frags == n_frags)
+ break;
+
+ rx_buf = efx_rx_buf_next(&channel->rx_queue, rx_buf);
+ }
+
+ skb->data_len = skb->len;
+ skb->truesize += n_frags * efx->rx_buffer_truesize;
+
+ skb_record_rx_queue(skb, channel->rx_queue.core_index);
+
+ gro_result = napi_gro_frags(napi);
+ if (gro_result != GRO_DROP)
channel->irq_mod_score += 2;
+}
+
+/* Allocate and construct an SKB around page fragments */
+static struct sk_buff *efx_rx_mk_skb(struct efx_channel *channel,
+ struct efx_rx_buffer *rx_buf,
+ unsigned int n_frags,
+ u8 *eh, int hdr_len)
+{
+ struct efx_nic *efx = channel->efx;
+ struct sk_buff *skb;
+
+ /* Allocate an SKB to store the headers */
+ skb = netdev_alloc_skb(efx->net_dev, hdr_len + EFX_PAGE_SKB_ALIGN);
+ if (unlikely(skb == NULL))
+ return NULL;
+
+ EFX_BUG_ON_PARANOID(rx_buf->len < hdr_len);
+
+ skb_reserve(skb, EFX_PAGE_SKB_ALIGN);
+ memcpy(__skb_put(skb, hdr_len), eh, hdr_len);
+
+ /* Append the remaining page(s) onto the frag list */
+ if (rx_buf->len > hdr_len) {
+ rx_buf->page_offset += hdr_len;
+ rx_buf->len -= hdr_len;
+
+ for (;;) {
+ skb_fill_page_desc(skb, skb_shinfo(skb)->nr_frags,
+ rx_buf->page, rx_buf->page_offset,
+ rx_buf->len);
+ rx_buf->page = NULL;
+ skb->len += rx_buf->len;
+ skb->data_len += rx_buf->len;
+ if (skb_shinfo(skb)->nr_frags == n_frags)
+ break;
+
+ rx_buf = efx_rx_buf_next(&channel->rx_queue, rx_buf);
+ }
+ } else {
+ __free_pages(rx_buf->page, efx->rx_buffer_order);
+ rx_buf->page = NULL;
+ n_frags = 0;
}
+
+ skb->truesize += n_frags * efx->rx_buffer_truesize;
+
+ /* Move past the ethernet header */
+ skb->protocol = eth_type_trans(skb, efx->net_dev);
+
+ return skb;
}
void efx_rx_packet(struct efx_rx_queue *rx_queue, unsigned int index,
- unsigned int len, u16 flags)
+ unsigned int n_frags, unsigned int len, u16 flags)
{
struct efx_nic *efx = rx_queue->efx;
struct efx_channel *channel = efx_rx_queue_channel(rx_queue);
struct efx_rx_buffer *rx_buf;
- bool leak_packet = false;
rx_buf = efx_rx_buffer(rx_queue, index);
rx_buf->flags |= flags;
- /* This allows the refill path to post another buffer.
- * EFX_RXD_HEAD_ROOM ensures that the slot we are using
- * isn't overwritten yet.
- */
- rx_queue->removed_count++;
-
- /* Validate the length encoded in the event vs the descriptor pushed */
- efx_rx_packet__check_len(rx_queue, rx_buf, len, &leak_packet);
+ /* Validate the number of fragments and completed length */
+ if (n_frags == 1) {
+ efx_rx_packet__check_len(rx_queue, rx_buf, len);
+ } else if (unlikely(n_frags > EFX_RX_MAX_FRAGS) ||
+ unlikely(len <= (n_frags - 1) * EFX_RX_USR_BUF_SIZE) ||
+ unlikely(len > n_frags * EFX_RX_USR_BUF_SIZE) ||
+ unlikely(!efx->rx_scatter)) {
+ /* If this isn't an explicit discard request, either
+ * the hardware or the driver is broken.
+ */
+ WARN_ON(!(len == 0 && rx_buf->flags & EFX_RX_PKT_DISCARD));
+ rx_buf->flags |= EFX_RX_PKT_DISCARD;
+ }
netif_vdbg(efx, rx_status, efx->net_dev,
- "RX queue %d received id %x at %llx+%x %s%s\n",
+ "RX queue %d received ids %x-%x len %d %s%s\n",
efx_rx_queue_index(rx_queue), index,
- (unsigned long long)rx_buf->dma_addr, len,
+ (index + n_frags - 1) & rx_queue->ptr_mask, len,
(rx_buf->flags & EFX_RX_PKT_CSUMMED) ? " [SUMMED]" : "",
(rx_buf->flags & EFX_RX_PKT_DISCARD) ? " [DISCARD]" : "");
- /* Discard packet, if instructed to do so */
+ /* Discard packet, if instructed to do so. Process the
+ * previous receive first.
+ */
if (unlikely(rx_buf->flags & EFX_RX_PKT_DISCARD)) {
- if (unlikely(leak_packet))
- channel->n_skbuff_leaks++;
- else
- efx_recycle_rx_buffer(channel, rx_buf);
-
- /* Don't hold off the previous receive */
- rx_buf = NULL;
- goto out;
+ efx_rx_flush_packet(channel);
+ put_page(rx_buf->page);
+ efx_recycle_rx_buffers(channel, rx_buf, n_frags);
+ return;
}
- /* Release and/or sync DMA mapping - assumes all RX buffers
- * consumed in-order per RX queue
+ if (n_frags == 1)
+ rx_buf->len = len;
+
+ /* Release and/or sync the DMA mapping - assumes all RX buffers
+ * consumed in-order per RX queue.
*/
- efx_unmap_rx_buffer(efx, rx_buf, len);
+ efx_sync_rx_buffer(efx, rx_buf, rx_buf->len);
/* Prefetch nice and early so data will (hopefully) be in cache by
* the time we look at it.
*/
- prefetch(efx_rx_buf_eh(efx, rx_buf));
+ prefetch(efx_rx_buf_va(rx_buf));
+
+ rx_buf->page_offset += efx->type->rx_buffer_hash_size;
+ rx_buf->len -= efx->type->rx_buffer_hash_size;
+
+ if (n_frags > 1) {
+ /* Release/sync DMA mapping for additional fragments.
+ * Fix length for last fragment.
+ */
+ unsigned int tail_frags = n_frags - 1;
+
+ for (;;) {
+ rx_buf = efx_rx_buf_next(rx_queue, rx_buf);
+ if (--tail_frags == 0)
+ break;
+ efx_sync_rx_buffer(efx, rx_buf, EFX_RX_USR_BUF_SIZE);
+ }
+ rx_buf->len = len - (n_frags - 1) * EFX_RX_USR_BUF_SIZE;
+ efx_sync_rx_buffer(efx, rx_buf, rx_buf->len);
+ }
+
+ /* All fragments have been DMA-synced, so recycle buffers and pages. */
+ rx_buf = efx_rx_buffer(rx_queue, index);
+ efx_recycle_rx_buffers(channel, rx_buf, n_frags);
/* Pipeline receives so that we give time for packet headers to be
* prefetched into cache.
*/
- rx_buf->len = len - efx->type->rx_buffer_hash_size;
-out:
- if (channel->rx_pkt)
- __efx_rx_packet(channel, channel->rx_pkt);
- channel->rx_pkt = rx_buf;
+ efx_rx_flush_packet(channel);
+ channel->rx_pkt_n_frags = n_frags;
+ channel->rx_pkt_index = index;
}
-static void efx_rx_deliver(struct efx_channel *channel,
- struct efx_rx_buffer *rx_buf)
+static void efx_rx_deliver(struct efx_channel *channel, u8 *eh,
+ struct efx_rx_buffer *rx_buf,
+ unsigned int n_frags)
{
struct sk_buff *skb;
+ u16 hdr_len = min_t(u16, rx_buf->len, EFX_SKB_HEADERS);
- /* We now own the SKB */
- skb = rx_buf->u.skb;
- rx_buf->u.skb = NULL;
+ skb = efx_rx_mk_skb(channel, rx_buf, n_frags, eh, hdr_len);
+ if (unlikely(skb == NULL)) {
+ efx_free_rx_buffer(rx_buf);
+ return;
+ }
+ skb_record_rx_queue(skb, channel->rx_queue.core_index);
/* Set the SKB flags */
skb_checksum_none_assert(skb);
- /* Record the rx_queue */
- skb_record_rx_queue(skb, channel->rx_queue.core_index);
+ if (channel->type->receive_skb)
+ if (channel->type->receive_skb(channel, skb))
+ return;
/* Pass the packet up */
- if (channel->type->receive_skb)
- channel->type->receive_skb(channel, skb);
- else
- netif_receive_skb(skb);
-
- /* Update allocation strategy method */
- channel->rx_alloc_level += RX_ALLOC_FACTOR_SKB;
+ netif_receive_skb(skb);
}
/* Handle a received packet. Second half: Touches packet payload. */
-void __efx_rx_packet(struct efx_channel *channel, struct efx_rx_buffer *rx_buf)
+void __efx_rx_packet(struct efx_channel *channel)
{
struct efx_nic *efx = channel->efx;
- u8 *eh = efx_rx_buf_eh(efx, rx_buf);
+ struct efx_rx_buffer *rx_buf =
+ efx_rx_buffer(&channel->rx_queue, channel->rx_pkt_index);
+ u8 *eh = efx_rx_buf_va(rx_buf);
/* If we're in loopback test, then pass the packet directly to the
* loopback layer, and free the rx_buf here
*/
if (unlikely(efx->loopback_selftest)) {
efx_loopback_rx_packet(efx, eh, rx_buf->len);
- efx_free_rx_buffer(efx, rx_buf);
- return;
- }
-
- if (!(rx_buf->flags & EFX_RX_BUF_PAGE)) {
- struct sk_buff *skb = rx_buf->u.skb;
-
- prefetch(skb_shinfo(skb));
-
- skb_reserve(skb, efx->type->rx_buffer_hash_size);
- skb_put(skb, rx_buf->len);
-
- if (efx->net_dev->features & NETIF_F_RXHASH)
- skb->rxhash = efx_rx_buf_hash(eh);
-
- /* Move past the ethernet header. rx_buf->data still points
- * at the ethernet header */
- skb->protocol = eth_type_trans(skb, efx->net_dev);
-
- skb_record_rx_queue(skb, channel->rx_queue.core_index);
+ efx_free_rx_buffer(rx_buf);
+ goto out;
}
if (unlikely(!(efx->net_dev->features & NETIF_F_RXCSUM)))
rx_buf->flags &= ~EFX_RX_PKT_CSUMMED;
- if (likely(rx_buf->flags & (EFX_RX_BUF_PAGE | EFX_RX_PKT_CSUMMED)) &&
- !channel->type->receive_skb)
- efx_rx_packet_gro(channel, rx_buf, eh);
+ if (!channel->type->receive_skb)
+ efx_rx_packet_gro(channel, rx_buf, channel->rx_pkt_n_frags, eh);
else
- efx_rx_deliver(channel, rx_buf);
-}
-
-void efx_rx_strategy(struct efx_channel *channel)
-{
- enum efx_rx_alloc_method method = rx_alloc_method;
-
- if (channel->type->receive_skb) {
- channel->rx_alloc_push_pages = false;
- return;
- }
-
- /* Only makes sense to use page based allocation if GRO is enabled */
- if (!(channel->efx->net_dev->features & NETIF_F_GRO)) {
- method = RX_ALLOC_METHOD_SKB;
- } else if (method == RX_ALLOC_METHOD_AUTO) {
- /* Constrain the rx_alloc_level */
- if (channel->rx_alloc_level < 0)
- channel->rx_alloc_level = 0;
- else if (channel->rx_alloc_level > RX_ALLOC_LEVEL_MAX)
- channel->rx_alloc_level = RX_ALLOC_LEVEL_MAX;
-
- /* Decide on the allocation method */
- method = ((channel->rx_alloc_level > RX_ALLOC_LEVEL_GRO) ?
- RX_ALLOC_METHOD_PAGE : RX_ALLOC_METHOD_SKB);
- }
-
- /* Push the option */
- channel->rx_alloc_push_pages = (method == RX_ALLOC_METHOD_PAGE);
+ efx_rx_deliver(channel, eh, rx_buf, channel->rx_pkt_n_frags);
+out:
+ channel->rx_pkt_n_frags = 0;
}
int efx_probe_rx_queue(struct efx_rx_queue *rx_queue)
@@ -683,9 +662,32 @@
kfree(rx_queue->buffer);
rx_queue->buffer = NULL;
}
+
return rc;
}
+static void efx_init_rx_recycle_ring(struct efx_nic *efx,
+ struct efx_rx_queue *rx_queue)
+{
+ unsigned int bufs_in_recycle_ring, page_ring_size;
+
+ /* Set the RX recycle ring size */
+#ifdef CONFIG_PPC64
+ bufs_in_recycle_ring = EFX_RECYCLE_RING_SIZE_IOMMU;
+#else
+ if (efx->pci_dev->dev.iommu_group)
+ bufs_in_recycle_ring = EFX_RECYCLE_RING_SIZE_IOMMU;
+ else
+ bufs_in_recycle_ring = EFX_RECYCLE_RING_SIZE_NOIOMMU;
+#endif /* CONFIG_PPC64 */
+
+ page_ring_size = roundup_pow_of_two(bufs_in_recycle_ring /
+ efx->rx_bufs_per_page);
+ rx_queue->page_ring = kcalloc(page_ring_size,
+ sizeof(*rx_queue->page_ring), GFP_KERNEL);
+ rx_queue->page_ptr_mask = page_ring_size - 1;
+}
+
void efx_init_rx_queue(struct efx_rx_queue *rx_queue)
{
struct efx_nic *efx = rx_queue->efx;
@@ -699,10 +701,18 @@
rx_queue->notified_count = 0;
rx_queue->removed_count = 0;
rx_queue->min_fill = -1U;
+ efx_init_rx_recycle_ring(efx, rx_queue);
+
+ rx_queue->page_remove = 0;
+ rx_queue->page_add = rx_queue->page_ptr_mask + 1;
+ rx_queue->page_recycle_count = 0;
+ rx_queue->page_recycle_failed = 0;
+ rx_queue->page_recycle_full = 0;
/* Initialise limit fields */
max_fill = efx->rxq_entries - EFX_RXD_HEAD_ROOM;
- max_trigger = max_fill - EFX_RX_BATCH;
+ max_trigger =
+ max_fill - efx->rx_pages_per_batch * efx->rx_bufs_per_page;
if (rx_refill_threshold != 0) {
trigger = max_fill * min(rx_refill_threshold, 100U) / 100U;
if (trigger > max_trigger)
@@ -722,6 +732,7 @@
void efx_fini_rx_queue(struct efx_rx_queue *rx_queue)
{
int i;
+ struct efx_nic *efx = rx_queue->efx;
struct efx_rx_buffer *rx_buf;
netif_dbg(rx_queue->efx, drv, rx_queue->efx->net_dev,
@@ -733,13 +744,32 @@
del_timer_sync(&rx_queue->slow_fill);
efx_nic_fini_rx(rx_queue);
- /* Release RX buffers NB start at index 0 not current HW ptr */
+ /* Release RX buffers from the current read ptr to the write ptr */
if (rx_queue->buffer) {
- for (i = 0; i <= rx_queue->ptr_mask; i++) {
- rx_buf = efx_rx_buffer(rx_queue, i);
+ for (i = rx_queue->removed_count; i < rx_queue->added_count;
+ i++) {
+ unsigned index = i & rx_queue->ptr_mask;
+ rx_buf = efx_rx_buffer(rx_queue, index);
efx_fini_rx_buffer(rx_queue, rx_buf);
}
}
+
+ /* Unmap and release the pages in the recycle ring. Remove the ring. */
+ for (i = 0; i <= rx_queue->page_ptr_mask; i++) {
+ struct page *page = rx_queue->page_ring[i];
+ struct efx_rx_page_state *state;
+
+ if (page == NULL)
+ continue;
+
+ state = page_address(page);
+ dma_unmap_page(&efx->pci_dev->dev, state->dma_addr,
+ PAGE_SIZE << efx->rx_buffer_order,
+ DMA_FROM_DEVICE);
+ put_page(page);
+ }
+ kfree(rx_queue->page_ring);
+ rx_queue->page_ring = NULL;
}
void efx_remove_rx_queue(struct efx_rx_queue *rx_queue)
@@ -754,9 +784,6 @@
}
-module_param(rx_alloc_method, int, 0644);
-MODULE_PARM_DESC(rx_alloc_method, "Allocation method used for RX buffers");
-
module_param(rx_refill_threshold, uint, 0444);
MODULE_PARM_DESC(rx_refill_threshold,
"RX descriptor ring refill threshold (%)");
diff --git a/drivers/net/ethernet/sfc/siena.c b/drivers/net/ethernet/sfc/siena.c
index ba40f67..5166924 100644
--- a/drivers/net/ethernet/sfc/siena.c
+++ b/drivers/net/ethernet/sfc/siena.c
@@ -202,7 +202,7 @@
static enum reset_type siena_map_reset_reason(enum reset_type reason)
{
- return RESET_TYPE_ALL;
+ return RESET_TYPE_RECOVER_OR_ALL;
}
static int siena_map_reset_flags(u32 *flags)
@@ -245,6 +245,22 @@
return efx_mcdi_reset_port(efx);
}
+#ifdef CONFIG_EEH
+/* When a PCI device is isolated from the bus, a subsequent MMIO read is
+ * required for the kernel EEH mechanisms to notice. As the Solarflare driver
+ * was written to minimise MMIO read (for latency) then a periodic call to check
+ * the EEH status of the device is required so that device recovery can happen
+ * in a timely fashion.
+ */
+static void siena_monitor(struct efx_nic *efx)
+{
+ struct eeh_dev *eehdev =
+ of_node_to_eeh_dev(pci_device_to_OF_node(efx->pci_dev));
+
+ eeh_dev_check_failure(eehdev);
+}
+#endif
+
static int siena_probe_nvconfig(struct efx_nic *efx)
{
u32 caps = 0;
@@ -398,6 +414,8 @@
EFX_SET_OWORD_FIELD(temp, FRF_BZ_RX_HASH_INSRT_HDR, 1);
EFX_SET_OWORD_FIELD(temp, FRF_BZ_RX_HASH_ALG, 1);
EFX_SET_OWORD_FIELD(temp, FRF_BZ_RX_IP_HASH, 1);
+ EFX_SET_OWORD_FIELD(temp, FRF_BZ_RX_USR_BUF_SIZE,
+ EFX_RX_USR_BUF_SIZE >> 5);
efx_writeo(efx, &temp, FR_AZ_RX_CFG);
/* Set hash key for IPv4 */
@@ -665,7 +683,11 @@
.init = siena_init_nic,
.dimension_resources = siena_dimension_resources,
.fini = efx_port_dummy_op_void,
+#ifdef CONFIG_EEH
+ .monitor = siena_monitor,
+#else
.monitor = NULL,
+#endif
.map_reset_reason = siena_map_reset_reason,
.map_reset_flags = siena_map_reset_flags,
.reset = siena_reset_hw,
@@ -698,6 +720,7 @@
.max_dma_mask = DMA_BIT_MASK(FSF_AZ_TX_KER_BUF_ADDR_WIDTH),
.rx_buffer_hash_size = 0x10,
.rx_buffer_padding = 0,
+ .can_rx_scatter = true,
.max_interrupt_mode = EFX_INT_MODE_MSIX,
.phys_addr_channels = 32, /* Hardware limit is 64, but the legacy
* interrupt handler only supports 32
diff --git a/drivers/net/ethernet/sgi/meth.c b/drivers/net/ethernet/sgi/meth.c
index 79ad9c9..4bdbaad9 100644
--- a/drivers/net/ethernet/sgi/meth.c
+++ b/drivers/net/ethernet/sgi/meth.c
@@ -213,10 +213,11 @@
{
/* Init TX ring */
priv->tx_ring = dma_alloc_coherent(NULL, TX_RING_BUFFER_SIZE,
- &priv->tx_ring_dma, GFP_ATOMIC);
+ &priv->tx_ring_dma,
+ GFP_ATOMIC | __GFP_ZERO);
if (!priv->tx_ring)
return -ENOMEM;
- memset(priv->tx_ring, 0, TX_RING_BUFFER_SIZE);
+
priv->tx_count = priv->tx_read = priv->tx_write = 0;
mace->eth.tx_ring_base = priv->tx_ring_dma;
/* Now init skb save area */
diff --git a/drivers/net/ethernet/sis/sis900.c b/drivers/net/ethernet/sis/sis900.c
index efca14e..e458296 100644
--- a/drivers/net/ethernet/sis/sis900.c
+++ b/drivers/net/ethernet/sis/sis900.c
@@ -1841,15 +1841,12 @@
entry = sis_priv->dirty_rx % NUM_RX_DESC;
if (sis_priv->rx_skbuff[entry] == NULL) {
- if ((skb = netdev_alloc_skb(net_dev, RX_BUF_SIZE)) == NULL) {
+ skb = netdev_alloc_skb(net_dev, RX_BUF_SIZE);
+ if (skb == NULL) {
/* not enough memory for skbuff, this makes a
* "hole" on the buffer ring, it is not clear
* how the hardware will react to this kind
* of degenerated buffer */
- if (netif_msg_rx_err(sis_priv))
- printk(KERN_INFO "%s: Memory squeeze, "
- "deferring packet.\n",
- net_dev->name);
net_dev->stats.rx_dropped++;
break;
}
diff --git a/drivers/net/ethernet/smsc/smc9194.c b/drivers/net/ethernet/smsc/smc9194.c
index 50823da..e85c2e7 100644
--- a/drivers/net/ethernet/smsc/smc9194.c
+++ b/drivers/net/ethernet/smsc/smc9194.c
@@ -1223,9 +1223,7 @@
dev->stats.multicast++;
skb = netdev_alloc_skb(dev, packet_length + 5);
-
if ( skb == NULL ) {
- printk(KERN_NOTICE CARDNAME ": Low memory, packet dropped.\n");
dev->stats.rx_dropped++;
goto done;
}
diff --git a/drivers/net/ethernet/smsc/smc91x.c b/drivers/net/ethernet/smsc/smc91x.c
index 591650a..dfbf978 100644
--- a/drivers/net/ethernet/smsc/smc91x.c
+++ b/drivers/net/ethernet/smsc/smc91x.c
@@ -465,8 +465,6 @@
*/
skb = netdev_alloc_skb(dev, packet_len);
if (unlikely(skb == NULL)) {
- printk(KERN_NOTICE "%s: Low memory, packet dropped.\n",
- dev->name);
SMC_WAIT_MMU_BUSY(lp);
SMC_SET_MMU_CMD(lp, MC_RELEASE);
dev->stats.rx_dropped++;
diff --git a/drivers/net/ethernet/smsc/smsc911x.c b/drivers/net/ethernet/smsc/smsc911x.c
index da5cc9a..48e2b99 100644
--- a/drivers/net/ethernet/smsc/smsc911x.c
+++ b/drivers/net/ethernet/smsc/smsc911x.c
@@ -2115,7 +2115,7 @@
spin_lock_init(&pdata->dev_lock);
spin_lock_init(&pdata->mac_lock);
- if (pdata->ioaddr == 0) {
+ if (pdata->ioaddr == NULL) {
SMSC_WARN(pdata, probe, "pdata->ioaddr: 0x00000000");
return -ENODEV;
}
diff --git a/drivers/net/ethernet/smsc/smsc9420.c b/drivers/net/ethernet/smsc/smsc9420.c
index d457fa2..ffa5c4a 100644
--- a/drivers/net/ethernet/smsc/smsc9420.c
+++ b/drivers/net/ethernet/smsc/smsc9420.c
@@ -848,10 +848,8 @@
BUG_ON(pd->rx_buffers[index].skb);
BUG_ON(pd->rx_buffers[index].mapping);
- if (unlikely(!skb)) {
- smsc_warn(RX_ERR, "Failed to allocate new skb!");
+ if (unlikely(!skb))
return -ENOMEM;
- }
mapping = pci_map_single(pd->pdev, skb_tail_pointer(skb),
PKT_BUF_SZ, PCI_DMA_FROMDEVICE);
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 39c6c55..d02b446 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -534,25 +534,17 @@
GFP_KERNEL);
priv->rx_skbuff = kmalloc_array(rxsize, sizeof(struct sk_buff *),
GFP_KERNEL);
- priv->dma_rx =
- (struct dma_desc *)dma_alloc_coherent(priv->device,
- rxsize *
- sizeof(struct dma_desc),
- &priv->dma_rx_phy,
- GFP_KERNEL);
+ priv->dma_rx = dma_alloc_coherent(priv->device,
+ rxsize * sizeof(struct dma_desc),
+ &priv->dma_rx_phy, GFP_KERNEL);
priv->tx_skbuff = kmalloc_array(txsize, sizeof(struct sk_buff *),
GFP_KERNEL);
- priv->dma_tx =
- (struct dma_desc *)dma_alloc_coherent(priv->device,
- txsize *
- sizeof(struct dma_desc),
- &priv->dma_tx_phy,
- GFP_KERNEL);
+ priv->dma_tx = dma_alloc_coherent(priv->device,
+ txsize * sizeof(struct dma_desc),
+ &priv->dma_tx_phy, GFP_KERNEL);
- if ((priv->dma_rx == NULL) || (priv->dma_tx == NULL)) {
- pr_err("%s:ERROR allocating the DMA Tx/Rx desc\n", __func__);
+ if ((priv->dma_rx == NULL) || (priv->dma_tx == NULL))
return;
- }
DBG(probe, INFO, "stmmac (%s) DMA desc: virt addr (Rx %p, "
"Tx %p)\n\tDMA phy addr (Rx 0x%08x, Tx 0x%08x)\n",
diff --git a/drivers/net/ethernet/sun/sunbmac.c b/drivers/net/ethernet/sun/sunbmac.c
index 5fafca0..0549759 100644
--- a/drivers/net/ethernet/sun/sunbmac.c
+++ b/drivers/net/ethernet/sun/sunbmac.c
@@ -1169,10 +1169,8 @@
bp->bmac_block = dma_alloc_coherent(&bp->bigmac_op->dev,
PAGE_SIZE,
&bp->bblock_dvma, GFP_ATOMIC);
- if (bp->bmac_block == NULL || bp->bblock_dvma == 0) {
- printk(KERN_ERR "BIGMAC: Cannot allocate consistent DMA.\n");
+ if (bp->bmac_block == NULL || bp->bblock_dvma == 0)
goto fail_and_cleanup;
- }
/* Get the board revision of this BigMAC. */
bp->board_rev = of_getintprop_default(bp->bigmac_op->dev.of_node,
diff --git a/drivers/net/ethernet/sun/sunhme.c b/drivers/net/ethernet/sun/sunhme.c
index a1bff49..436fa9d 100644
--- a/drivers/net/ethernet/sun/sunhme.c
+++ b/drivers/net/ethernet/sun/sunhme.c
@@ -2752,10 +2752,8 @@
&hp->hblock_dvma,
GFP_ATOMIC);
err = -ENOMEM;
- if (!hp->happy_block) {
- printk(KERN_ERR "happymeal: Cannot allocate descriptors.\n");
+ if (!hp->happy_block)
goto err_out_iounmap;
- }
/* Force check of the link first time we are brought up. */
hp->linkcheck = 0;
@@ -3068,14 +3066,11 @@
hp->happy_bursts = DMA_BURSTBITS;
#endif
- hp->happy_block = (struct hmeal_init_block *)
- dma_alloc_coherent(&pdev->dev, PAGE_SIZE, &hp->hblock_dvma, GFP_KERNEL);
-
+ hp->happy_block = dma_alloc_coherent(&pdev->dev, PAGE_SIZE,
+ &hp->hblock_dvma, GFP_KERNEL);
err = -ENODEV;
- if (!hp->happy_block) {
- printk(KERN_ERR "happymeal(PCI): Cannot get hme init block.\n");
+ if (!hp->happy_block)
goto err_out_iounmap;
- }
hp->linkcheck = 0;
hp->timer_state = asleep;
diff --git a/drivers/net/ethernet/sun/sunqe.c b/drivers/net/ethernet/sun/sunqe.c
index 49bf3e2..8182591b 100644
--- a/drivers/net/ethernet/sun/sunqe.c
+++ b/drivers/net/ethernet/sun/sunqe.c
@@ -414,7 +414,7 @@
struct qe_rxd *this;
struct sunqe_buffers *qbufs = qep->buffers;
__u32 qbufs_dvma = qep->buffers_dvma;
- int elem = qep->rx_new, drops = 0;
+ int elem = qep->rx_new;
u32 flags;
this = &rxbase[elem];
@@ -436,7 +436,6 @@
} else {
skb = netdev_alloc_skb(dev, len + 2);
if (skb == NULL) {
- drops++;
dev->stats.rx_dropped++;
} else {
skb_reserve(skb, 2);
@@ -456,8 +455,6 @@
this = &rxbase[elem];
}
qep->rx_new = elem;
- if (drops)
- printk(KERN_NOTICE "%s: Memory squeeze, deferring packet.\n", qep->dev->name);
}
static void qe_tx_reclaim(struct sunqe *qep);
diff --git a/drivers/net/ethernet/tehuti/tehuti.c b/drivers/net/ethernet/tehuti/tehuti.c
index e15cc71..e8824ce 100644
--- a/drivers/net/ethernet/tehuti/tehuti.c
+++ b/drivers/net/ethernet/tehuti/tehuti.c
@@ -1102,10 +1102,9 @@
dno = bdx_rxdb_available(db) - 1;
while (dno > 0) {
skb = netdev_alloc_skb(priv->ndev, f->m.pktsz + NET_IP_ALIGN);
- if (!skb) {
- pr_err("NO MEM: netdev_alloc_skb failed\n");
+ if (!skb)
break;
- }
+
skb_reserve(skb, NET_IP_ALIGN);
idx = bdx_rxdb_alloc_elem(db);
diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c
index 75c4855..5ceaa4c3 100644
--- a/drivers/net/ethernet/ti/cpsw.c
+++ b/drivers/net/ethernet/ti/cpsw.c
@@ -126,6 +126,13 @@
#define CPSW_FIFO_DUAL_MAC_MODE (1 << 15)
#define CPSW_FIFO_RATE_LIMIT_MODE (2 << 15)
+#define CPSW_INTPACEEN (0x3f << 16)
+#define CPSW_INTPRESCALE_MASK (0x7FF << 0)
+#define CPSW_CMINTMAX_CNT 63
+#define CPSW_CMINTMIN_CNT 2
+#define CPSW_CMINTMAX_INTVL (1000 / CPSW_CMINTMIN_CNT)
+#define CPSW_CMINTMIN_INTVL ((1000 / CPSW_CMINTMAX_CNT) + 1)
+
#define cpsw_enable_irq(priv) \
do { \
u32 i; \
@@ -139,6 +146,10 @@
disable_irq_nosync(priv->irqs_table[i]); \
} while (0);
+#define cpsw_slave_index(priv) \
+ ((priv->data.dual_emac) ? priv->emac_port : \
+ priv->data.active_slave)
+
static int debug_level;
module_param(debug_level, int, 0);
MODULE_PARM_DESC(debug_level, "cpsw debug level (NETIF_MSG bits)");
@@ -160,6 +171,15 @@
u32 rx_en;
u32 tx_en;
u32 misc_en;
+ u32 mem_allign1[8];
+ u32 rx_thresh_stat;
+ u32 rx_stat;
+ u32 tx_stat;
+ u32 misc_stat;
+ u32 mem_allign2[8];
+ u32 rx_imax;
+ u32 tx_imax;
+
};
struct cpsw_ss_regs {
@@ -314,6 +334,8 @@
struct cpsw_host_regs __iomem *host_port_regs;
u32 msg_enable;
u32 version;
+ u32 coal_intvl;
+ u32 bus_freq_mhz;
struct net_device_stats stats;
int rx_packet_max;
int host_port;
@@ -612,6 +634,77 @@
}
}
+static int cpsw_get_coalesce(struct net_device *ndev,
+ struct ethtool_coalesce *coal)
+{
+ struct cpsw_priv *priv = netdev_priv(ndev);
+
+ coal->rx_coalesce_usecs = priv->coal_intvl;
+ return 0;
+}
+
+static int cpsw_set_coalesce(struct net_device *ndev,
+ struct ethtool_coalesce *coal)
+{
+ struct cpsw_priv *priv = netdev_priv(ndev);
+ u32 int_ctrl;
+ u32 num_interrupts = 0;
+ u32 prescale = 0;
+ u32 addnl_dvdr = 1;
+ u32 coal_intvl = 0;
+
+ if (!coal->rx_coalesce_usecs)
+ return -EINVAL;
+
+ coal_intvl = coal->rx_coalesce_usecs;
+
+ int_ctrl = readl(&priv->wr_regs->int_control);
+ prescale = priv->bus_freq_mhz * 4;
+
+ if (coal_intvl < CPSW_CMINTMIN_INTVL)
+ coal_intvl = CPSW_CMINTMIN_INTVL;
+
+ if (coal_intvl > CPSW_CMINTMAX_INTVL) {
+ /* Interrupt pacer works with 4us Pulse, we can
+ * throttle further by dilating the 4us pulse.
+ */
+ addnl_dvdr = CPSW_INTPRESCALE_MASK / prescale;
+
+ if (addnl_dvdr > 1) {
+ prescale *= addnl_dvdr;
+ if (coal_intvl > (CPSW_CMINTMAX_INTVL * addnl_dvdr))
+ coal_intvl = (CPSW_CMINTMAX_INTVL
+ * addnl_dvdr);
+ } else {
+ addnl_dvdr = 1;
+ coal_intvl = CPSW_CMINTMAX_INTVL;
+ }
+ }
+
+ num_interrupts = (1000 * addnl_dvdr) / coal_intvl;
+ writel(num_interrupts, &priv->wr_regs->rx_imax);
+ writel(num_interrupts, &priv->wr_regs->tx_imax);
+
+ int_ctrl |= CPSW_INTPACEEN;
+ int_ctrl &= (~CPSW_INTPRESCALE_MASK);
+ int_ctrl |= (prescale & CPSW_INTPRESCALE_MASK);
+ writel(int_ctrl, &priv->wr_regs->int_control);
+
+ cpsw_notice(priv, timer, "Set coalesce to %d usecs.\n", coal_intvl);
+ if (priv->data.dual_emac) {
+ int i;
+
+ for (i = 0; i < priv->data.slaves; i++) {
+ priv = netdev_priv(priv->slaves[i].ndev);
+ priv->coal_intvl = coal_intvl;
+ }
+ } else {
+ priv->coal_intvl = coal_intvl;
+ }
+
+ return 0;
+}
+
static inline int __show_stat(char *buf, int maxlen, const char *name, u32 val)
{
static char *leader = "........................................";
@@ -834,6 +927,14 @@
cpsw_info(priv, ifup, "submitted %d rx descriptors\n", i);
}
+ /* Enable Interrupt pacing if configured */
+ if (priv->coal_intvl != 0) {
+ struct ethtool_coalesce coal;
+
+ coal.rx_coalesce_usecs = (priv->coal_intvl << 4);
+ cpsw_set_coalesce(ndev, &coal);
+ }
+
cpdma_ctlr_start(priv->dma);
cpsw_intr_enable(priv);
napi_enable(&priv->napi);
@@ -942,7 +1043,7 @@
static void cpsw_hwtstamp_v1(struct cpsw_priv *priv)
{
- struct cpsw_slave *slave = &priv->slaves[priv->data.cpts_active_slave];
+ struct cpsw_slave *slave = &priv->slaves[priv->data.active_slave];
u32 ts_en, seq_id;
if (!priv->cpts->tx_enable && !priv->cpts->rx_enable) {
@@ -971,7 +1072,7 @@
if (priv->data.dual_emac)
slave = &priv->slaves[priv->emac_port];
else
- slave = &priv->slaves[priv->data.cpts_active_slave];
+ slave = &priv->slaves[priv->data.active_slave];
ctrl = slave_read(slave, CPSW2_CONTROL);
ctrl &= ~CTRL_ALL_TS_MASK;
@@ -1056,14 +1157,26 @@
static int cpsw_ndo_ioctl(struct net_device *dev, struct ifreq *req, int cmd)
{
+ struct cpsw_priv *priv = netdev_priv(dev);
+ struct mii_ioctl_data *data = if_mii(req);
+ int slave_no = cpsw_slave_index(priv);
+
if (!netif_running(dev))
return -EINVAL;
+ switch (cmd) {
#ifdef CONFIG_TI_CPTS
- if (cmd == SIOCSHWTSTAMP)
+ case SIOCSHWTSTAMP:
return cpsw_hwtstamp_ioctl(dev, req);
#endif
- return -ENOTSUPP;
+ case SIOCGMIIPHY:
+ data->phy_id = priv->slaves[slave_no].phy->addr;
+ break;
+ default:
+ return -ENOTSUPP;
+ }
+
+ return 0;
}
static void cpsw_ndo_tx_timeout(struct net_device *ndev)
@@ -1244,12 +1357,39 @@
return 0;
}
+static int cpsw_get_settings(struct net_device *ndev,
+ struct ethtool_cmd *ecmd)
+{
+ struct cpsw_priv *priv = netdev_priv(ndev);
+ int slave_no = cpsw_slave_index(priv);
+
+ if (priv->slaves[slave_no].phy)
+ return phy_ethtool_gset(priv->slaves[slave_no].phy, ecmd);
+ else
+ return -EOPNOTSUPP;
+}
+
+static int cpsw_set_settings(struct net_device *ndev, struct ethtool_cmd *ecmd)
+{
+ struct cpsw_priv *priv = netdev_priv(ndev);
+ int slave_no = cpsw_slave_index(priv);
+
+ if (priv->slaves[slave_no].phy)
+ return phy_ethtool_sset(priv->slaves[slave_no].phy, ecmd);
+ else
+ return -EOPNOTSUPP;
+}
+
static const struct ethtool_ops cpsw_ethtool_ops = {
.get_drvinfo = cpsw_get_drvinfo,
.get_msglevel = cpsw_get_msglevel,
.set_msglevel = cpsw_set_msglevel,
.get_link = ethtool_op_get_link,
.get_ts_info = cpsw_get_ts_info,
+ .get_settings = cpsw_get_settings,
+ .set_settings = cpsw_set_settings,
+ .get_coalesce = cpsw_get_coalesce,
+ .set_coalesce = cpsw_set_coalesce,
};
static void cpsw_slave_init(struct cpsw_slave *slave, struct cpsw_priv *priv,
@@ -1282,12 +1422,12 @@
}
data->slaves = prop;
- if (of_property_read_u32(node, "cpts_active_slave", &prop)) {
- pr_err("Missing cpts_active_slave property in the DT.\n");
+ if (of_property_read_u32(node, "active_slave", &prop)) {
+ pr_err("Missing active_slave property in the DT.\n");
ret = -EINVAL;
goto error_ret;
}
- data->cpts_active_slave = prop;
+ data->active_slave = prop;
if (of_property_read_u32(node, "cpts_clock_mult", &prop)) {
pr_err("Missing cpts_clock_mult property in the DT.\n");
@@ -1437,6 +1577,9 @@
priv_sl2->slaves = priv->slaves;
priv_sl2->clk = priv->clk;
+ priv_sl2->coal_intvl = 0;
+ priv_sl2->bus_freq_mhz = priv->bus_freq_mhz;
+
priv_sl2->cpsw_res = priv->cpsw_res;
priv_sl2->regs = priv->regs;
priv_sl2->host_port = priv->host_port;
@@ -1546,6 +1689,8 @@
ret = -ENODEV;
goto clean_slave_ret;
}
+ priv->coal_intvl = 0;
+ priv->bus_freq_mhz = clk_get_rate(priv->clk) / 1000000;
priv->cpsw_res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
if (!priv->cpsw_res) {
diff --git a/drivers/net/ethernet/ti/tlan.c b/drivers/net/ethernet/ti/tlan.c
index 2272538..bdda36f 100644
--- a/drivers/net/ethernet/ti/tlan.c
+++ b/drivers/net/ethernet/ti/tlan.c
@@ -1911,10 +1911,8 @@
list->frame_size = TLAN_MAX_FRAME_SIZE;
list->buffer[0].count = TLAN_MAX_FRAME_SIZE | TLAN_LAST_BUFFER;
skb = netdev_alloc_skb_ip_align(dev, TLAN_MAX_FRAME_SIZE + 5);
- if (!skb) {
- netdev_err(dev, "Out of memory for received data\n");
+ if (!skb)
break;
- }
list->buffer[0].address = pci_map_single(priv->pci_dev,
skb->data,
diff --git a/drivers/net/ethernet/toshiba/spider_net.c b/drivers/net/ethernet/toshiba/spider_net.c
index f1b91fd..fef6b59 100644
--- a/drivers/net/ethernet/toshiba/spider_net.c
+++ b/drivers/net/ethernet/toshiba/spider_net.c
@@ -352,8 +352,7 @@
alloc_size = chain->num_desc * sizeof(struct spider_net_hw_descr);
chain->hwring = dma_alloc_coherent(&card->pdev->dev, alloc_size,
- &chain->dma_addr, GFP_KERNEL);
-
+ &chain->dma_addr, GFP_KERNEL);
if (!chain->hwring)
return -ENOMEM;
diff --git a/drivers/net/ethernet/tundra/tsi108_eth.c b/drivers/net/ethernet/tundra/tsi108_eth.c
index 8fa947a..3c69a04 100644
--- a/drivers/net/ethernet/tundra/tsi108_eth.c
+++ b/drivers/net/ethernet/tundra/tsi108_eth.c
@@ -1308,27 +1308,16 @@
data->id, dev->irq, dev->name);
}
- data->rxring = dma_alloc_coherent(NULL, rxring_size,
- &data->rxdma, GFP_KERNEL);
-
- if (!data->rxring) {
- printk(KERN_DEBUG
- "TSI108_ETH: failed to allocate memory for rxring!\n");
+ data->rxring = dma_alloc_coherent(NULL, rxring_size, &data->rxdma,
+ GFP_KERNEL | __GFP_ZERO);
+ if (!data->rxring)
return -ENOMEM;
- } else {
- memset(data->rxring, 0, rxring_size);
- }
- data->txring = dma_alloc_coherent(NULL, txring_size,
- &data->txdma, GFP_KERNEL);
-
+ data->txring = dma_alloc_coherent(NULL, txring_size, &data->txdma,
+ GFP_KERNEL | __GFP_ZERO);
if (!data->txring) {
- printk(KERN_DEBUG
- "TSI108_ETH: failed to allocate memory for txring!\n");
pci_free_consistent(0, rxring_size, data->rxring, data->rxdma);
return -ENOMEM;
- } else {
- memset(data->txring, 0, txring_size);
}
for (i = 0; i < TSI108_RXRING_LEN; i++) {
diff --git a/drivers/net/ethernet/xilinx/ll_temac_main.c b/drivers/net/ethernet/xilinx/ll_temac_main.c
index 9fc2ada..4a7c60f 100644
--- a/drivers/net/ethernet/xilinx/ll_temac_main.c
+++ b/drivers/net/ethernet/xilinx/ll_temac_main.c
@@ -245,39 +245,30 @@
/* returns a virtual address and a physical address. */
lp->tx_bd_v = dma_alloc_coherent(ndev->dev.parent,
sizeof(*lp->tx_bd_v) * TX_BD_NUM,
- &lp->tx_bd_p, GFP_KERNEL);
- if (!lp->tx_bd_v) {
- dev_err(&ndev->dev,
- "unable to allocate DMA TX buffer descriptors");
+ &lp->tx_bd_p, GFP_KERNEL | __GFP_ZERO);
+ if (!lp->tx_bd_v)
goto out;
- }
+
lp->rx_bd_v = dma_alloc_coherent(ndev->dev.parent,
sizeof(*lp->rx_bd_v) * RX_BD_NUM,
- &lp->rx_bd_p, GFP_KERNEL);
- if (!lp->rx_bd_v) {
- dev_err(&ndev->dev,
- "unable to allocate DMA RX buffer descriptors");
+ &lp->rx_bd_p, GFP_KERNEL | __GFP_ZERO);
+ if (!lp->rx_bd_v)
goto out;
- }
- memset(lp->tx_bd_v, 0, sizeof(*lp->tx_bd_v) * TX_BD_NUM);
for (i = 0; i < TX_BD_NUM; i++) {
lp->tx_bd_v[i].next = lp->tx_bd_p +
sizeof(*lp->tx_bd_v) * ((i + 1) % TX_BD_NUM);
}
- memset(lp->rx_bd_v, 0, sizeof(*lp->rx_bd_v) * RX_BD_NUM);
for (i = 0; i < RX_BD_NUM; i++) {
lp->rx_bd_v[i].next = lp->rx_bd_p +
sizeof(*lp->rx_bd_v) * ((i + 1) % RX_BD_NUM);
skb = netdev_alloc_skb_ip_align(ndev,
XTE_MAX_JUMBO_FRAME_SIZE);
-
- if (skb == 0) {
- dev_err(&ndev->dev, "alloc_skb error %d\n", i);
+ if (!skb)
goto out;
- }
+
lp->rx_skb[i] = skb;
/* returns physical address of skb->data */
lp->rx_bd_v[i].phys = dma_map_single(ndev->dev.parent,
@@ -789,9 +780,7 @@
new_skb = netdev_alloc_skb_ip_align(ndev,
XTE_MAX_JUMBO_FRAME_SIZE);
-
- if (new_skb == 0) {
- dev_err(&ndev->dev, "no memory for new sk_buff\n");
+ if (!new_skb) {
spin_unlock_irqrestore(&lp->rx_lock, flags);
return;
}
diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
index 278c9db..24748e8 100644
--- a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
+++ b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
@@ -204,41 +204,31 @@
lp->tx_bd_v = dma_alloc_coherent(ndev->dev.parent,
sizeof(*lp->tx_bd_v) * TX_BD_NUM,
&lp->tx_bd_p,
- GFP_KERNEL);
- if (!lp->tx_bd_v) {
- dev_err(&ndev->dev, "unable to allocate DMA Tx buffer "
- "descriptors");
+ GFP_KERNEL | __GFP_ZERO);
+ if (!lp->tx_bd_v)
goto out;
- }
lp->rx_bd_v = dma_alloc_coherent(ndev->dev.parent,
sizeof(*lp->rx_bd_v) * RX_BD_NUM,
&lp->rx_bd_p,
- GFP_KERNEL);
- if (!lp->rx_bd_v) {
- dev_err(&ndev->dev, "unable to allocate DMA Rx buffer "
- "descriptors");
+ GFP_KERNEL | __GFP_ZERO);
+ if (!lp->rx_bd_v)
goto out;
- }
- memset(lp->tx_bd_v, 0, sizeof(*lp->tx_bd_v) * TX_BD_NUM);
for (i = 0; i < TX_BD_NUM; i++) {
lp->tx_bd_v[i].next = lp->tx_bd_p +
sizeof(*lp->tx_bd_v) *
((i + 1) % TX_BD_NUM);
}
- memset(lp->rx_bd_v, 0, sizeof(*lp->rx_bd_v) * RX_BD_NUM);
for (i = 0; i < RX_BD_NUM; i++) {
lp->rx_bd_v[i].next = lp->rx_bd_p +
sizeof(*lp->rx_bd_v) *
((i + 1) % RX_BD_NUM);
skb = netdev_alloc_skb_ip_align(ndev, lp->max_frm_size);
- if (!skb) {
- dev_err(&ndev->dev, "alloc_skb error %d\n", i);
+ if (!skb)
goto out;
- }
lp->rx_bd_v[i].sw_id_offset = (u32) skb;
lp->rx_bd_v[i].phys = dma_map_single(ndev->dev.parent,
@@ -777,10 +767,9 @@
packets++;
new_skb = netdev_alloc_skb_ip_align(ndev, lp->max_frm_size);
- if (!new_skb) {
- dev_err(&ndev->dev, "no memory for new sk_buff\n");
+ if (!new_skb)
return;
- }
+
cur_p->phys = dma_map_single(ndev->dev.parent, new_skb->data,
lp->max_frm_size,
DMA_FROM_DEVICE);
diff --git a/drivers/net/ethernet/xircom/xirc2ps_cs.c b/drivers/net/ethernet/xircom/xirc2ps_cs.c
index 98e09d0..76210ab 100644
--- a/drivers/net/ethernet/xircom/xirc2ps_cs.c
+++ b/drivers/net/ethernet/xircom/xirc2ps_cs.c
@@ -1041,7 +1041,6 @@
/* 1 extra so we can use insw */
skb = netdev_alloc_skb(dev, pktlen + 3);
if (!skb) {
- pr_notice("low memory, packet dropped (size=%u)\n", pktlen);
dev->stats.rx_dropped++;
} else { /* okay get the packet */
skb_reserve(skb, 2);
diff --git a/drivers/net/fddi/defxx.c b/drivers/net/fddi/defxx.c
index 502c8ff..4c8ddc9 100644
--- a/drivers/net/fddi/defxx.c
+++ b/drivers/net/fddi/defxx.c
@@ -1070,13 +1070,10 @@
(PI_ALIGN_K_DESC_BLK - 1);
bp->kmalloced = top_v = dma_alloc_coherent(bp->bus_dev, alloc_size,
&bp->kmalloced_dma,
- GFP_ATOMIC);
- if (top_v == NULL) {
- printk("%s: Could not allocate memory for host buffers "
- "and structures!\n", print_name);
+ GFP_ATOMIC | __GFP_ZERO);
+ if (top_v == NULL)
return DFX_K_FAILURE;
- }
- memset(top_v, 0, alloc_size); /* zero out memory before continuing */
+
top_p = bp->kmalloced_dma; /* get physical address of buffer */
/*
diff --git a/drivers/net/ieee802154/mrf24j40.c b/drivers/net/ieee802154/mrf24j40.c
index 3f2c7aa..0ca8f88 100644
--- a/drivers/net/ieee802154/mrf24j40.c
+++ b/drivers/net/ieee802154/mrf24j40.c
@@ -22,6 +22,7 @@
#include <linux/spi/spi.h>
#include <linux/interrupt.h>
#include <linux/module.h>
+#include <linux/pinctrl/consumer.h>
#include <net/wpan-phy.h>
#include <net/mac802154.h>
@@ -91,9 +92,8 @@
#define MRF24J40_READLONG(reg) (1 << 15 | (reg) << 5)
#define MRF24J40_WRITELONG(reg) (1 << 15 | (reg) << 5 | 1 << 4)
-/* Maximum speed to run the device at. TODO: Get the real max value from
- * someone at Microchip since it isn't in the datasheet. */
-#define MAX_SPI_SPEED_HZ 1000000
+/* The datasheet indicates the theoretical maximum for SCK to be 10MHz */
+#define MAX_SPI_SPEED_HZ 10000000
#define printdev(X) (&X->spi->dev)
@@ -361,6 +361,7 @@
if (ret == -ERESTARTSYS)
goto err;
if (ret == 0) {
+ dev_warn(printdev(devrec), "Timeout waiting for TX interrupt\n");
ret = -ETIMEDOUT;
goto err;
}
@@ -477,7 +478,7 @@
int i;
for (i = 0; i < 8; i++)
write_short_reg(devrec, REG_EADR0+i,
- filt->ieee_addr[i]);
+ filt->ieee_addr[7-i]);
#ifdef DEBUG
printk(KERN_DEBUG "Set long addr to: ");
@@ -623,6 +624,7 @@
int ret = -ENOMEM;
u8 val;
struct mrf24j40 *devrec;
+ struct pinctrl *pinctrl;
printk(KERN_INFO "mrf24j40: probe(). IRQ: %d\n", spi->irq);
@@ -633,6 +635,11 @@
if (!devrec->buf)
goto err_buf;
+ pinctrl = devm_pinctrl_get_select_default(&spi->dev);
+ if (IS_ERR(pinctrl))
+ dev_warn(&spi->dev,
+ "pinctrl pins are not configured from the driver");
+
spi->mode = SPI_MODE_0; /* TODO: Is this appropriate for right here? */
if (spi->max_speed_hz > MAX_SPI_SPEED_HZ)
spi->max_speed_hz = MAX_SPI_SPEED_HZ;
diff --git a/drivers/net/irda/ali-ircc.c b/drivers/net/irda/ali-ircc.c
index 9cea451..3adb43c 100644
--- a/drivers/net/irda/ali-ircc.c
+++ b/drivers/net/irda/ali-ircc.c
@@ -352,21 +352,19 @@
/* Allocate memory if needed */
self->rx_buff.head =
dma_alloc_coherent(NULL, self->rx_buff.truesize,
- &self->rx_buff_dma, GFP_KERNEL);
+ &self->rx_buff_dma, GFP_KERNEL | __GFP_ZERO);
if (self->rx_buff.head == NULL) {
err = -ENOMEM;
goto err_out2;
}
- memset(self->rx_buff.head, 0, self->rx_buff.truesize);
self->tx_buff.head =
dma_alloc_coherent(NULL, self->tx_buff.truesize,
- &self->tx_buff_dma, GFP_KERNEL);
+ &self->tx_buff_dma, GFP_KERNEL | __GFP_ZERO);
if (self->tx_buff.head == NULL) {
err = -ENOMEM;
goto err_out3;
}
- memset(self->tx_buff.head, 0, self->tx_buff.truesize);
self->rx_buff.in_frame = FALSE;
self->rx_buff.state = OUTSIDE_FRAME;
diff --git a/drivers/net/irda/au1k_ir.c b/drivers/net/irda/au1k_ir.c
index b5151e4..56f1e6d 100644
--- a/drivers/net/irda/au1k_ir.c
+++ b/drivers/net/irda/au1k_ir.c
@@ -27,6 +27,7 @@
#include <linux/slab.h>
#include <linux/time.h>
#include <linux/types.h>
+#include <linux/ioport.h>
#include <net/irda/irda.h>
#include <net/irda/irmod.h>
@@ -882,12 +883,12 @@
goto out;
err = -EBUSY;
- aup->ioarea = request_mem_region(r->start, r->end - r->start + 1,
+ aup->ioarea = request_mem_region(r->start, resource_size(r),
pdev->name);
if (!aup->ioarea)
goto out;
- aup->iobase = ioremap_nocache(r->start, r->end - r->start + 1);
+ aup->iobase = ioremap_nocache(r->start, resource_size(r));
if (!aup->iobase)
goto out2;
diff --git a/drivers/net/irda/bfin_sir.c b/drivers/net/irda/bfin_sir.c
index fed4a05..a06fca6 100644
--- a/drivers/net/irda/bfin_sir.c
+++ b/drivers/net/irda/bfin_sir.c
@@ -389,7 +389,8 @@
set_dma_callback(port->rx_dma_channel, bfin_sir_dma_rx_int, dev);
set_dma_callback(port->tx_dma_channel, bfin_sir_dma_tx_int, dev);
- port->rx_dma_buf.buf = (unsigned char *)dma_alloc_coherent(NULL, PAGE_SIZE, &dma_handle, GFP_DMA);
+ port->rx_dma_buf.buf = dma_alloc_coherent(NULL, PAGE_SIZE,
+ &dma_handle, GFP_DMA);
port->rx_dma_buf.head = 0;
port->rx_dma_buf.tail = 0;
port->rx_dma_nrows = 0;
diff --git a/drivers/net/irda/nsc-ircc.c b/drivers/net/irda/nsc-ircc.c
index 2a4f2f1..9cf836b5 100644
--- a/drivers/net/irda/nsc-ircc.c
+++ b/drivers/net/irda/nsc-ircc.c
@@ -431,22 +431,20 @@
/* Allocate memory if needed */
self->rx_buff.head =
dma_alloc_coherent(NULL, self->rx_buff.truesize,
- &self->rx_buff_dma, GFP_KERNEL);
+ &self->rx_buff_dma, GFP_KERNEL | __GFP_ZERO);
if (self->rx_buff.head == NULL) {
err = -ENOMEM;
goto out2;
}
- memset(self->rx_buff.head, 0, self->rx_buff.truesize);
self->tx_buff.head =
dma_alloc_coherent(NULL, self->tx_buff.truesize,
- &self->tx_buff_dma, GFP_KERNEL);
+ &self->tx_buff_dma, GFP_KERNEL | __GFP_ZERO);
if (self->tx_buff.head == NULL) {
err = -ENOMEM;
goto out3;
}
- memset(self->tx_buff.head, 0, self->tx_buff.truesize);
self->rx_buff.in_frame = FALSE;
self->rx_buff.state = OUTSIDE_FRAME;
diff --git a/drivers/net/irda/pxaficp_ir.c b/drivers/net/irda/pxaficp_ir.c
index 858de05..964b116 100644
--- a/drivers/net/irda/pxaficp_ir.c
+++ b/drivers/net/irda/pxaficp_ir.c
@@ -700,12 +700,12 @@
err = -ENOMEM;
si->dma_rx_buff = dma_alloc_coherent(si->dev, IRDA_FRAME_SIZE_LIMIT,
- &si->dma_rx_buff_phy, GFP_KERNEL );
+ &si->dma_rx_buff_phy, GFP_KERNEL);
if (!si->dma_rx_buff)
goto err_dma_rx_buff;
si->dma_tx_buff = dma_alloc_coherent(si->dev, IRDA_FRAME_SIZE_LIMIT,
- &si->dma_tx_buff_phy, GFP_KERNEL );
+ &si->dma_tx_buff_phy, GFP_KERNEL);
if (!si->dma_tx_buff)
goto err_dma_tx_buff;
diff --git a/drivers/net/irda/smsc-ircc2.c b/drivers/net/irda/smsc-ircc2.c
index 5290952..aa05dad 100644
--- a/drivers/net/irda/smsc-ircc2.c
+++ b/drivers/net/irda/smsc-ircc2.c
@@ -563,24 +563,15 @@
self->rx_buff.head =
dma_alloc_coherent(NULL, self->rx_buff.truesize,
- &self->rx_buff_dma, GFP_KERNEL);
- if (self->rx_buff.head == NULL) {
- IRDA_ERROR("%s, Can't allocate memory for receive buffer!\n",
- driver_name);
+ &self->rx_buff_dma, GFP_KERNEL | __GFP_ZERO);
+ if (self->rx_buff.head == NULL)
goto err_out2;
- }
self->tx_buff.head =
dma_alloc_coherent(NULL, self->tx_buff.truesize,
- &self->tx_buff_dma, GFP_KERNEL);
- if (self->tx_buff.head == NULL) {
- IRDA_ERROR("%s, Can't allocate memory for transmit buffer!\n",
- driver_name);
+ &self->tx_buff_dma, GFP_KERNEL | __GFP_ZERO);
+ if (self->tx_buff.head == NULL)
goto err_out3;
- }
-
- memset(self->rx_buff.head, 0, self->rx_buff.truesize);
- memset(self->tx_buff.head, 0, self->tx_buff.truesize);
self->rx_buff.in_frame = FALSE;
self->rx_buff.state = OUTSIDE_FRAME;
diff --git a/drivers/net/irda/via-ircc.c b/drivers/net/irda/via-ircc.c
index f9033c6..51f2bc3 100644
--- a/drivers/net/irda/via-ircc.c
+++ b/drivers/net/irda/via-ircc.c
@@ -364,21 +364,19 @@
/* Allocate memory if needed */
self->rx_buff.head =
dma_alloc_coherent(&pdev->dev, self->rx_buff.truesize,
- &self->rx_buff_dma, GFP_KERNEL);
+ &self->rx_buff_dma, GFP_KERNEL | __GFP_ZERO);
if (self->rx_buff.head == NULL) {
err = -ENOMEM;
goto err_out2;
}
- memset(self->rx_buff.head, 0, self->rx_buff.truesize);
self->tx_buff.head =
dma_alloc_coherent(&pdev->dev, self->tx_buff.truesize,
- &self->tx_buff_dma, GFP_KERNEL);
+ &self->tx_buff_dma, GFP_KERNEL | __GFP_ZERO);
if (self->tx_buff.head == NULL) {
err = -ENOMEM;
goto err_out3;
}
- memset(self->tx_buff.head, 0, self->tx_buff.truesize);
self->rx_buff.in_frame = FALSE;
self->rx_buff.state = OUTSIDE_FRAME;
diff --git a/drivers/net/irda/w83977af_ir.c b/drivers/net/irda/w83977af_ir.c
index f5bb92f..bb8857a 100644
--- a/drivers/net/irda/w83977af_ir.c
+++ b/drivers/net/irda/w83977af_ir.c
@@ -216,22 +216,19 @@
/* Allocate memory if needed */
self->rx_buff.head =
dma_alloc_coherent(NULL, self->rx_buff.truesize,
- &self->rx_buff_dma, GFP_KERNEL);
+ &self->rx_buff_dma, GFP_KERNEL | __GFP_ZERO);
if (self->rx_buff.head == NULL) {
err = -ENOMEM;
goto err_out1;
}
- memset(self->rx_buff.head, 0, self->rx_buff.truesize);
-
self->tx_buff.head =
dma_alloc_coherent(NULL, self->tx_buff.truesize,
- &self->tx_buff_dma, GFP_KERNEL);
+ &self->tx_buff_dma, GFP_KERNEL | __GFP_ZERO);
if (self->tx_buff.head == NULL) {
err = -ENOMEM;
goto err_out2;
}
- memset(self->tx_buff.head, 0, self->tx_buff.truesize);
self->rx_buff.in_frame = FALSE;
self->rx_buff.state = OUTSIDE_FRAME;
diff --git a/drivers/net/phy/lxt.c b/drivers/net/phy/lxt.c
index ec40ba8..ff2e45e 100644
--- a/drivers/net/phy/lxt.c
+++ b/drivers/net/phy/lxt.c
@@ -159,7 +159,7 @@
return 0;
}
-int lxt973a2_read_status(struct phy_device *phydev)
+static int lxt973a2_read_status(struct phy_device *phydev)
{
int adv;
int err;
diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c
index 22dec9c..202fe1f 100644
--- a/drivers/net/phy/marvell.c
+++ b/drivers/net/phy/marvell.c
@@ -7,6 +7,8 @@
*
* Copyright (c) 2004 Freescale Semiconductor, Inc.
*
+ * Copyright (c) 2013 Michael Stapelberg <michael@stapelberg.de>
+ *
* This program is free software; you can redistribute it and/or modify it
* under the terms of the GNU General Public License as published by the
* Free Software Foundation; either version 2 of the License, or (at your
@@ -80,6 +82,28 @@
#define MII_88E1318S_PHY_MSCR1_REG 16
#define MII_88E1318S_PHY_MSCR1_PAD_ODD BIT(6)
+/* Copper Specific Interrupt Enable Register */
+#define MII_88E1318S_PHY_CSIER 0x12
+/* WOL Event Interrupt Enable */
+#define MII_88E1318S_PHY_CSIER_WOL_EIE BIT(7)
+
+/* LED Timer Control Register */
+#define MII_88E1318S_PHY_LED_PAGE 0x03
+#define MII_88E1318S_PHY_LED_TCR 0x12
+#define MII_88E1318S_PHY_LED_TCR_FORCE_INT BIT(15)
+#define MII_88E1318S_PHY_LED_TCR_INTn_ENABLE BIT(7)
+#define MII_88E1318S_PHY_LED_TCR_INT_ACTIVE_LOW BIT(11)
+
+/* Magic Packet MAC address registers */
+#define MII_88E1318S_PHY_MAGIC_PACKET_WORD2 0x17
+#define MII_88E1318S_PHY_MAGIC_PACKET_WORD1 0x18
+#define MII_88E1318S_PHY_MAGIC_PACKET_WORD0 0x19
+
+#define MII_88E1318S_PHY_WOL_PAGE 0x11
+#define MII_88E1318S_PHY_WOL_CTRL 0x10
+#define MII_88E1318S_PHY_WOL_CTRL_CLEAR_WOL_STATUS BIT(12)
+#define MII_88E1318S_PHY_WOL_CTRL_MAGIC_PACKET_MATCH_ENABLE BIT(14)
+
#define MII_88E1121_PHY_LED_CTRL 16
#define MII_88E1121_PHY_LED_PAGE 3
#define MII_88E1121_PHY_LED_DEF 0x0030
@@ -696,6 +720,107 @@
return 0;
}
+static void m88e1318_get_wol(struct phy_device *phydev, struct ethtool_wolinfo *wol)
+{
+ wol->supported = WAKE_MAGIC;
+ wol->wolopts = 0;
+
+ if (phy_write(phydev, MII_MARVELL_PHY_PAGE,
+ MII_88E1318S_PHY_WOL_PAGE) < 0)
+ return;
+
+ if (phy_read(phydev, MII_88E1318S_PHY_WOL_CTRL) &
+ MII_88E1318S_PHY_WOL_CTRL_MAGIC_PACKET_MATCH_ENABLE)
+ wol->wolopts |= WAKE_MAGIC;
+
+ if (phy_write(phydev, MII_MARVELL_PHY_PAGE, 0x00) < 0)
+ return;
+}
+
+static int m88e1318_set_wol(struct phy_device *phydev, struct ethtool_wolinfo *wol)
+{
+ int err, oldpage, temp;
+
+ oldpage = phy_read(phydev, MII_MARVELL_PHY_PAGE);
+
+ if (wol->wolopts & WAKE_MAGIC) {
+ /* Explicitly switch to page 0x00, just to be sure */
+ err = phy_write(phydev, MII_MARVELL_PHY_PAGE, 0x00);
+ if (err < 0)
+ return err;
+
+ /* Enable the WOL interrupt */
+ temp = phy_read(phydev, MII_88E1318S_PHY_CSIER);
+ temp |= MII_88E1318S_PHY_CSIER_WOL_EIE;
+ err = phy_write(phydev, MII_88E1318S_PHY_CSIER, temp);
+ if (err < 0)
+ return err;
+
+ err = phy_write(phydev, MII_MARVELL_PHY_PAGE,
+ MII_88E1318S_PHY_LED_PAGE);
+ if (err < 0)
+ return err;
+
+ /* Setup LED[2] as interrupt pin (active low) */
+ temp = phy_read(phydev, MII_88E1318S_PHY_LED_TCR);
+ temp &= ~MII_88E1318S_PHY_LED_TCR_FORCE_INT;
+ temp |= MII_88E1318S_PHY_LED_TCR_INTn_ENABLE;
+ temp |= MII_88E1318S_PHY_LED_TCR_INT_ACTIVE_LOW;
+ err = phy_write(phydev, MII_88E1318S_PHY_LED_TCR, temp);
+ if (err < 0)
+ return err;
+
+ err = phy_write(phydev, MII_MARVELL_PHY_PAGE,
+ MII_88E1318S_PHY_WOL_PAGE);
+ if (err < 0)
+ return err;
+
+ /* Store the device address for the magic packet */
+ err = phy_write(phydev, MII_88E1318S_PHY_MAGIC_PACKET_WORD2,
+ ((phydev->attached_dev->dev_addr[5] << 8) |
+ phydev->attached_dev->dev_addr[4]));
+ if (err < 0)
+ return err;
+ err = phy_write(phydev, MII_88E1318S_PHY_MAGIC_PACKET_WORD1,
+ ((phydev->attached_dev->dev_addr[3] << 8) |
+ phydev->attached_dev->dev_addr[2]));
+ if (err < 0)
+ return err;
+ err = phy_write(phydev, MII_88E1318S_PHY_MAGIC_PACKET_WORD0,
+ ((phydev->attached_dev->dev_addr[1] << 8) |
+ phydev->attached_dev->dev_addr[0]));
+ if (err < 0)
+ return err;
+
+ /* Clear WOL status and enable magic packet matching */
+ temp = phy_read(phydev, MII_88E1318S_PHY_WOL_CTRL);
+ temp |= MII_88E1318S_PHY_WOL_CTRL_CLEAR_WOL_STATUS;
+ temp |= MII_88E1318S_PHY_WOL_CTRL_MAGIC_PACKET_MATCH_ENABLE;
+ err = phy_write(phydev, MII_88E1318S_PHY_WOL_CTRL, temp);
+ if (err < 0)
+ return err;
+ } else {
+ err = phy_write(phydev, MII_MARVELL_PHY_PAGE,
+ MII_88E1318S_PHY_WOL_PAGE);
+ if (err < 0)
+ return err;
+
+ /* Clear WOL status and disable magic packet matching */
+ temp = phy_read(phydev, MII_88E1318S_PHY_WOL_CTRL);
+ temp |= MII_88E1318S_PHY_WOL_CTRL_CLEAR_WOL_STATUS;
+ temp &= ~MII_88E1318S_PHY_WOL_CTRL_MAGIC_PACKET_MATCH_ENABLE;
+ err = phy_write(phydev, MII_88E1318S_PHY_WOL_CTRL, temp);
+ if (err < 0)
+ return err;
+ }
+
+ err = phy_write(phydev, MII_MARVELL_PHY_PAGE, oldpage);
+ if (err < 0)
+ return err;
+
+ return 0;
+}
+
static struct phy_driver marvell_drivers[] = {
{
.phy_id = MARVELL_PHY_ID_88E1101,
@@ -772,6 +897,8 @@
.ack_interrupt = &marvell_ack_interrupt,
.config_intr = &marvell_config_intr,
.did_interrupt = &m88e1121_did_interrupt,
+ .get_wol = &m88e1318_get_wol,
+ .set_wol = &m88e1318_set_wol,
.driver = { .owner = THIS_MODULE },
},
{
diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c
index abf7b61..2510435 100644
--- a/drivers/net/phy/micrel.c
+++ b/drivers/net/phy/micrel.c
@@ -53,6 +53,18 @@
#define KS8737_CTRL_INT_ACTIVE_HIGH (1 << 14)
#define KSZ8051_RMII_50MHZ_CLK (1 << 7)
+static int ksz_config_flags(struct phy_device *phydev)
+{
+ int regval;
+
+ if (phydev->dev_flags & MICREL_PHY_50MHZ_CLK) {
+ regval = phy_read(phydev, MII_KSZPHY_CTRL);
+ regval |= KSZ8051_RMII_50MHZ_CLK;
+ return phy_write(phydev, MII_KSZPHY_CTRL, regval);
+ }
+ return 0;
+}
+
static int kszphy_ack_interrupt(struct phy_device *phydev)
{
/* bit[7..0] int status, which is a read and clear register. */
@@ -114,22 +126,19 @@
static int ksz8021_config_init(struct phy_device *phydev)
{
+ int rc;
const u16 val = KSZPHY_OMSO_B_CAST_OFF | KSZPHY_OMSO_RMII_OVERRIDE;
phy_write(phydev, MII_KSZPHY_OMSO, val);
- return 0;
+ rc = ksz_config_flags(phydev);
+ return rc < 0 ? rc : 0;
}
static int ks8051_config_init(struct phy_device *phydev)
{
- int regval;
+ int rc;
- if (phydev->dev_flags & MICREL_PHY_50MHZ_CLK) {
- regval = phy_read(phydev, MII_KSZPHY_CTRL);
- regval |= KSZ8051_RMII_50MHZ_CLK;
- phy_write(phydev, MII_KSZPHY_CTRL, regval);
- }
-
- return 0;
+ rc = ksz_config_flags(phydev);
+ return rc < 0 ? rc : 0;
}
#define KSZ8873MLL_GLOBAL_CONTROL_4 0x06
@@ -192,6 +201,19 @@
.config_intr = kszphy_config_intr,
.driver = { .owner = THIS_MODULE,},
}, {
+ .phy_id = PHY_ID_KSZ8031,
+ .phy_id_mask = 0x00ffffff,
+ .name = "Micrel KSZ8031",
+ .features = (PHY_BASIC_FEATURES | SUPPORTED_Pause |
+ SUPPORTED_Asym_Pause),
+ .flags = PHY_HAS_MAGICANEG | PHY_HAS_INTERRUPT,
+ .config_init = ksz8021_config_init,
+ .config_aneg = genphy_config_aneg,
+ .read_status = genphy_read_status,
+ .ack_interrupt = kszphy_ack_interrupt,
+ .config_intr = kszphy_config_intr,
+ .driver = { .owner = THIS_MODULE,},
+}, {
.phy_id = PHY_ID_KSZ8041,
.phy_id_mask = 0x00fffff0,
.name = "Micrel KSZ8041",
@@ -325,6 +347,7 @@
{ PHY_ID_KSZ8001, 0x00ffffff },
{ PHY_ID_KS8737, 0x00fffff0 },
{ PHY_ID_KSZ8021, 0x00ffffff },
+ { PHY_ID_KSZ8031, 0x00ffffff },
{ PHY_ID_KSZ8041, 0x00fffff0 },
{ PHY_ID_KSZ8051, 0x00fffff0 },
{ PHY_ID_KSZ8061, 0x00fffff0 },
diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c
index ef9ea92..298b4c2 100644
--- a/drivers/net/phy/phy.c
+++ b/drivers/net/phy/phy.c
@@ -1188,3 +1188,19 @@
return 0;
}
EXPORT_SYMBOL(phy_ethtool_set_eee);
+
+int phy_ethtool_set_wol(struct phy_device *phydev, struct ethtool_wolinfo *wol)
+{
+ if (phydev->drv->set_wol)
+ return phydev->drv->set_wol(phydev, wol);
+
+ return -EOPNOTSUPP;
+}
+EXPORT_SYMBOL(phy_ethtool_set_wol);
+
+void phy_ethtool_get_wol(struct phy_device *phydev, struct ethtool_wolinfo *wol)
+{
+ if (phydev->drv->get_wol)
+ phydev->drv->get_wol(phydev, wol);
+}
+EXPORT_SYMBOL(phy_ethtool_get_wol);
diff --git a/drivers/net/phy/vitesse.c b/drivers/net/phy/vitesse.c
index 2585c38..3492b53 100644
--- a/drivers/net/phy/vitesse.c
+++ b/drivers/net/phy/vitesse.c
@@ -61,7 +61,7 @@
MODULE_AUTHOR("Kriston Carson");
MODULE_LICENSE("GPL");
-int vsc824x_add_skew(struct phy_device *phydev)
+static int vsc824x_add_skew(struct phy_device *phydev)
{
int err;
int extcon;
@@ -81,7 +81,6 @@
return err;
}
-EXPORT_SYMBOL(vsc824x_add_skew);
static int vsc824x_config_init(struct phy_device *phydev)
{
diff --git a/drivers/net/team/Kconfig b/drivers/net/team/Kconfig
index c3011af..c853d84 100644
--- a/drivers/net/team/Kconfig
+++ b/drivers/net/team/Kconfig
@@ -37,6 +37,18 @@
To compile this team mode as a module, choose M here: the module
will be called team_mode_roundrobin.
+config NET_TEAM_MODE_RANDOM
+ tristate "Random mode support"
+ depends on NET_TEAM
+ ---help---
+ Basic mode where port used for transmitting packets is selected
+ randomly.
+
+ All added ports are setup to have team's device address.
+
+ To compile this team mode as a module, choose M here: the module
+ will be called team_mode_random.
+
config NET_TEAM_MODE_ACTIVEBACKUP
tristate "Active-backup mode support"
depends on NET_TEAM
diff --git a/drivers/net/team/Makefile b/drivers/net/team/Makefile
index 9757630..c57e858 100644
--- a/drivers/net/team/Makefile
+++ b/drivers/net/team/Makefile
@@ -5,5 +5,6 @@
obj-$(CONFIG_NET_TEAM) += team.o
obj-$(CONFIG_NET_TEAM_MODE_BROADCAST) += team_mode_broadcast.o
obj-$(CONFIG_NET_TEAM_MODE_ROUNDROBIN) += team_mode_roundrobin.o
+obj-$(CONFIG_NET_TEAM_MODE_RANDOM) += team_mode_random.o
obj-$(CONFIG_NET_TEAM_MODE_ACTIVEBACKUP) += team_mode_activebackup.o
obj-$(CONFIG_NET_TEAM_MODE_LOADBALANCE) += team_mode_loadbalance.o
diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c
index bf34192..621c1bd 100644
--- a/drivers/net/team/team.c
+++ b/drivers/net/team/team.c
@@ -73,11 +73,24 @@
return __set_port_dev_addr(port->dev, port->orig.dev_addr);
}
-int team_port_set_team_dev_addr(struct team_port *port)
+static int team_port_set_team_dev_addr(struct team *team,
+ struct team_port *port)
{
- return __set_port_dev_addr(port->dev, port->team->dev->dev_addr);
+ return __set_port_dev_addr(port->dev, team->dev->dev_addr);
}
-EXPORT_SYMBOL(team_port_set_team_dev_addr);
+
+int team_modeop_port_enter(struct team *team, struct team_port *port)
+{
+ return team_port_set_team_dev_addr(team, port);
+}
+EXPORT_SYMBOL(team_modeop_port_enter);
+
+void team_modeop_port_change_dev_addr(struct team *team,
+ struct team_port *port)
+{
+ team_port_set_team_dev_addr(team, port);
+}
+EXPORT_SYMBOL(team_modeop_port_change_dev_addr);
static void team_refresh_port_linkup(struct team_port *port)
{
@@ -490,9 +503,9 @@
return false;
}
-rx_handler_result_t team_dummy_receive(struct team *team,
- struct team_port *port,
- struct sk_buff *skb)
+static rx_handler_result_t team_dummy_receive(struct team *team,
+ struct team_port *port,
+ struct sk_buff *skb)
{
return RX_HANDLER_ANOTHER;
}
diff --git a/drivers/net/team/team_mode_broadcast.c b/drivers/net/team/team_mode_broadcast.c
index c5db428..c366cd2 100644
--- a/drivers/net/team/team_mode_broadcast.c
+++ b/drivers/net/team/team_mode_broadcast.c
@@ -46,20 +46,10 @@
return sum_ret;
}
-static int bc_port_enter(struct team *team, struct team_port *port)
-{
- return team_port_set_team_dev_addr(port);
-}
-
-static void bc_port_change_dev_addr(struct team *team, struct team_port *port)
-{
- team_port_set_team_dev_addr(port);
-}
-
static const struct team_mode_ops bc_mode_ops = {
.transmit = bc_transmit,
- .port_enter = bc_port_enter,
- .port_change_dev_addr = bc_port_change_dev_addr,
+ .port_enter = team_modeop_port_enter,
+ .port_change_dev_addr = team_modeop_port_change_dev_addr,
};
static const struct team_mode bc_mode = {
diff --git a/drivers/net/team/team_mode_random.c b/drivers/net/team/team_mode_random.c
new file mode 100644
index 0000000..9eabfaa
--- /dev/null
+++ b/drivers/net/team/team_mode_random.c
@@ -0,0 +1,71 @@
+/*
+ * drivers/net/team/team_mode_random.c - Random mode for team
+ * Copyright (c) 2013 Jiri Pirko <jiri@resnulli.us>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/skbuff.h>
+#include <linux/reciprocal_div.h>
+#include <linux/if_team.h>
+
+static u32 random_N(unsigned int N)
+{
+ return reciprocal_divide(random32(), N);
+}
+
+static bool rnd_transmit(struct team *team, struct sk_buff *skb)
+{
+ struct team_port *port;
+ int port_index;
+
+ port_index = random_N(team->en_port_count);
+ port = team_get_port_by_index_rcu(team, port_index);
+ port = team_get_first_port_txable_rcu(team, port);
+ if (unlikely(!port))
+ goto drop;
+ if (team_dev_queue_xmit(team, port, skb))
+ return false;
+ return true;
+
+drop:
+ dev_kfree_skb_any(skb);
+ return false;
+}
+
+static const struct team_mode_ops rnd_mode_ops = {
+ .transmit = rnd_transmit,
+ .port_enter = team_modeop_port_enter,
+ .port_change_dev_addr = team_modeop_port_change_dev_addr,
+};
+
+static const struct team_mode rnd_mode = {
+ .kind = "random",
+ .owner = THIS_MODULE,
+ .ops = &rnd_mode_ops,
+};
+
+static int __init rnd_init_module(void)
+{
+ return team_mode_register(&rnd_mode);
+}
+
+static void __exit rnd_cleanup_module(void)
+{
+ team_mode_unregister(&rnd_mode);
+}
+
+module_init(rnd_init_module);
+module_exit(rnd_cleanup_module);
+
+MODULE_LICENSE("GPL v2");
+MODULE_AUTHOR("Jiri Pirko <jiri@resnulli.us>");
+MODULE_DESCRIPTION("Random mode for team");
+MODULE_ALIAS("team-mode-random");
diff --git a/drivers/net/team/team_mode_roundrobin.c b/drivers/net/team/team_mode_roundrobin.c
index 105135aa..d268e4d 100644
--- a/drivers/net/team/team_mode_roundrobin.c
+++ b/drivers/net/team/team_mode_roundrobin.c
@@ -25,26 +25,6 @@
return (struct rr_priv *) &team->mode_priv;
}
-static struct team_port *__get_first_port_up(struct team *team,
- struct team_port *port)
-{
- struct team_port *cur;
-
- if (team_port_txable(port))
- return port;
- cur = port;
- list_for_each_entry_continue_rcu(cur, &team->port_list, list)
- if (team_port_txable(port))
- return cur;
- list_for_each_entry_rcu(cur, &team->port_list, list) {
- if (cur == port)
- break;
- if (team_port_txable(port))
- return cur;
- }
- return NULL;
-}
-
static bool rr_transmit(struct team *team, struct sk_buff *skb)
{
struct team_port *port;
@@ -52,7 +32,7 @@
port_index = rr_priv(team)->sent_packets++ % team->en_port_count;
port = team_get_port_by_index_rcu(team, port_index);
- port = __get_first_port_up(team, port);
+ port = team_get_first_port_txable_rcu(team, port);
if (unlikely(!port))
goto drop;
if (team_dev_queue_xmit(team, port, skb))
@@ -64,20 +44,10 @@
return false;
}
-static int rr_port_enter(struct team *team, struct team_port *port)
-{
- return team_port_set_team_dev_addr(port);
-}
-
-static void rr_port_change_dev_addr(struct team *team, struct team_port *port)
-{
- team_port_set_team_dev_addr(port);
-}
-
static const struct team_mode_ops rr_mode_ops = {
.transmit = rr_transmit,
- .port_enter = rr_port_enter,
- .port_change_dev_addr = rr_port_change_dev_addr,
+ .port_enter = team_modeop_port_enter,
+ .port_change_dev_addr = team_modeop_port_change_dev_addr,
};
static const struct team_mode rr_mode = {
diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index b7c457a..95837c1 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -409,14 +409,12 @@
{
struct tun_file *ntfile;
struct tun_struct *tun;
- struct net_device *dev;
tun = rtnl_dereference(tfile->tun);
if (tun && !tfile->detached) {
u16 index = tfile->queue_index;
BUG_ON(index >= tun->numqueues);
- dev = tun->dev;
rcu_assign_pointer(tun->tfiles[index],
tun->tfiles[tun->numqueues - 1]);
diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 7cee7a3..33427fd 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -81,13 +81,22 @@
struct hlist_head vni_list[VNI_HASH_SIZE];
};
+struct vxlan_rdst {
+ struct rcu_head rcu;
+ __be32 remote_ip;
+ __be16 remote_port;
+ u32 remote_vni;
+ u32 remote_ifindex;
+ struct vxlan_rdst *remote_next;
+};
+
/* Forwarding table entry */
struct vxlan_fdb {
struct hlist_node hlist; /* linked list of entries */
struct rcu_head rcu;
unsigned long updated; /* jiffies */
unsigned long used;
- __be32 remote_ip;
+ struct vxlan_rdst remote;
u16 state; /* see ndm_state */
u8 eth_addr[ETH_ALEN];
};
@@ -157,7 +166,8 @@
/* Fill in neighbour message in skbuff. */
static int vxlan_fdb_info(struct sk_buff *skb, struct vxlan_dev *vxlan,
const struct vxlan_fdb *fdb,
- u32 portid, u32 seq, int type, unsigned int flags)
+ u32 portid, u32 seq, int type, unsigned int flags,
+ const struct vxlan_rdst *rdst)
{
unsigned long now = jiffies;
struct nda_cacheinfo ci;
@@ -176,7 +186,7 @@
if (type == RTM_GETNEIGH) {
ndm->ndm_family = AF_INET;
- send_ip = fdb->remote_ip != 0;
+ send_ip = rdst->remote_ip != htonl(INADDR_ANY);
send_eth = !is_zero_ether_addr(fdb->eth_addr);
} else
ndm->ndm_family = AF_BRIDGE;
@@ -188,7 +198,17 @@
if (send_eth && nla_put(skb, NDA_LLADDR, ETH_ALEN, &fdb->eth_addr))
goto nla_put_failure;
- if (send_ip && nla_put_be32(skb, NDA_DST, fdb->remote_ip))
+ if (send_ip && nla_put_be32(skb, NDA_DST, rdst->remote_ip))
+ goto nla_put_failure;
+
+ if (rdst->remote_port && rdst->remote_port != vxlan_port &&
+ nla_put_be16(skb, NDA_PORT, rdst->remote_port))
+ goto nla_put_failure;
+ if (rdst->remote_vni != vxlan->vni &&
+ nla_put_be32(skb, NDA_VNI, rdst->remote_vni))
+ goto nla_put_failure;
+ if (rdst->remote_ifindex &&
+ nla_put_u32(skb, NDA_IFINDEX, rdst->remote_ifindex))
goto nla_put_failure;
ci.ndm_used = jiffies_to_clock_t(now - fdb->used);
@@ -211,6 +231,9 @@
return NLMSG_ALIGN(sizeof(struct ndmsg))
+ nla_total_size(ETH_ALEN) /* NDA_LLADDR */
+ nla_total_size(sizeof(__be32)) /* NDA_DST */
+ + nla_total_size(sizeof(__be32)) /* NDA_PORT */
+ + nla_total_size(sizeof(__be32)) /* NDA_VNI */
+ + nla_total_size(sizeof(__u32)) /* NDA_IFINDEX */
+ nla_total_size(sizeof(struct nda_cacheinfo));
}
@@ -225,7 +248,7 @@
if (skb == NULL)
goto errout;
- err = vxlan_fdb_info(skb, vxlan, fdb, 0, 0, type, 0);
+ err = vxlan_fdb_info(skb, vxlan, fdb, 0, 0, type, 0, &fdb->remote);
if (err < 0) {
/* -EMSGSIZE implies BUG in vxlan_nlmsg_size() */
WARN_ON(err == -EMSGSIZE);
@@ -247,7 +270,8 @@
memset(&f, 0, sizeof f);
f.state = NUD_STALE;
- f.remote_ip = ipa; /* goes to NDA_DST */
+ f.remote.remote_ip = ipa; /* goes to NDA_DST */
+ f.remote.remote_vni = VXLAN_N_VID;
vxlan_fdb_notify(vxlan, &f, RTM_GETNEIGH);
}
@@ -300,10 +324,38 @@
return NULL;
}
+/* Add/update destinations for multicast */
+static int vxlan_fdb_append(struct vxlan_fdb *f,
+ __be32 ip, __u32 port, __u32 vni, __u32 ifindex)
+{
+ struct vxlan_rdst *rd_prev, *rd;
+
+ rd_prev = NULL;
+ for (rd = &f->remote; rd; rd = rd->remote_next) {
+ if (rd->remote_ip == ip &&
+ rd->remote_port == port &&
+ rd->remote_vni == vni &&
+ rd->remote_ifindex == ifindex)
+ return 0;
+ rd_prev = rd;
+ }
+ rd = kmalloc(sizeof(*rd), GFP_ATOMIC);
+ if (rd == NULL)
+ return -ENOBUFS;
+ rd->remote_ip = ip;
+ rd->remote_port = port;
+ rd->remote_vni = vni;
+ rd->remote_ifindex = ifindex;
+ rd->remote_next = NULL;
+ rd_prev->remote_next = rd;
+ return 1;
+}
+
/* Add new entry to forwarding table -- assumes lock held */
static int vxlan_fdb_create(struct vxlan_dev *vxlan,
const u8 *mac, __be32 ip,
- __u16 state, __u16 flags)
+ __u16 state, __u16 flags,
+ __u32 port, __u32 vni, __u32 ifindex)
{
struct vxlan_fdb *f;
int notify = 0;
@@ -320,6 +372,14 @@
f->updated = jiffies;
notify = 1;
}
+ if ((flags & NLM_F_APPEND) &&
+ is_multicast_ether_addr(f->eth_addr)) {
+ int rc = vxlan_fdb_append(f, ip, port, vni, ifindex);
+
+ if (rc < 0)
+ return rc;
+ notify |= rc;
+ }
} else {
if (!(flags & NLM_F_CREATE))
return -ENOENT;
@@ -333,7 +393,11 @@
return -ENOMEM;
notify = 1;
- f->remote_ip = ip;
+ f->remote.remote_ip = ip;
+ f->remote.remote_port = port;
+ f->remote.remote_vni = vni;
+ f->remote.remote_ifindex = ifindex;
+ f->remote.remote_next = NULL;
f->state = state;
f->updated = f->used = jiffies;
memcpy(f->eth_addr, mac, ETH_ALEN);
@@ -349,6 +413,19 @@
return 0;
}
+void vxlan_fdb_free(struct rcu_head *head)
+{
+ struct vxlan_fdb *f = container_of(head, struct vxlan_fdb, rcu);
+
+ while (f->remote.remote_next) {
+ struct vxlan_rdst *rd = f->remote.remote_next;
+
+ f->remote.remote_next = rd->remote_next;
+ kfree(rd);
+ }
+ kfree(f);
+}
+
static void vxlan_fdb_destroy(struct vxlan_dev *vxlan, struct vxlan_fdb *f)
{
netdev_dbg(vxlan->dev,
@@ -358,7 +435,7 @@
vxlan_fdb_notify(vxlan, f, RTM_DELNEIGH);
hlist_del_rcu(&f->hlist);
- kfree_rcu(f, rcu);
+ call_rcu(&f->rcu, vxlan_fdb_free);
}
/* Add static entry (via netlink) */
@@ -367,7 +444,9 @@
const unsigned char *addr, u16 flags)
{
struct vxlan_dev *vxlan = netdev_priv(dev);
+ struct net *net = dev_net(vxlan->dev);
__be32 ip;
+ u32 port, vni, ifindex;
int err;
if (!(ndm->ndm_state & (NUD_PERMANENT|NUD_REACHABLE))) {
@@ -384,8 +463,36 @@
ip = nla_get_be32(tb[NDA_DST]);
+ if (tb[NDA_PORT]) {
+ if (nla_len(tb[NDA_PORT]) != sizeof(u32))
+ return -EINVAL;
+ port = nla_get_u32(tb[NDA_PORT]);
+ } else
+ port = vxlan_port;
+
+ if (tb[NDA_VNI]) {
+ if (nla_len(tb[NDA_VNI]) != sizeof(u32))
+ return -EINVAL;
+ vni = nla_get_u32(tb[NDA_VNI]);
+ } else
+ vni = vxlan->vni;
+
+ if (tb[NDA_IFINDEX]) {
+ struct net_device *dev;
+
+ if (nla_len(tb[NDA_IFINDEX]) != sizeof(u32))
+ return -EINVAL;
+ ifindex = nla_get_u32(tb[NDA_IFINDEX]);
+ dev = dev_get_by_index(net, ifindex);
+ if (!dev)
+ return -EADDRNOTAVAIL;
+ dev_put(dev);
+ } else
+ ifindex = 0;
+
spin_lock_bh(&vxlan->hash_lock);
- err = vxlan_fdb_create(vxlan, addr, ip, ndm->ndm_state, flags);
+ err = vxlan_fdb_create(vxlan, addr, ip, ndm->ndm_state, flags, port,
+ vni, ifindex);
spin_unlock_bh(&vxlan->hash_lock);
return err;
@@ -423,18 +530,21 @@
int err;
hlist_for_each_entry_rcu(f, &vxlan->fdb_head[h], hlist) {
- if (idx < cb->args[0])
- goto skip;
+ struct vxlan_rdst *rd;
+ for (rd = &f->remote; rd; rd = rd->remote_next) {
+ if (idx < cb->args[0])
+ goto skip;
- err = vxlan_fdb_info(skb, vxlan, f,
- NETLINK_CB(cb->skb).portid,
- cb->nlh->nlmsg_seq,
- RTM_NEWNEIGH,
- NLM_F_MULTI);
- if (err < 0)
- break;
+ err = vxlan_fdb_info(skb, vxlan, f,
+ NETLINK_CB(cb->skb).portid,
+ cb->nlh->nlmsg_seq,
+ RTM_NEWNEIGH,
+ NLM_F_MULTI, rd);
+ if (err < 0)
+ break;
skip:
- ++idx;
+ ++idx;
+ }
}
}
@@ -454,22 +564,23 @@
f = vxlan_find_mac(vxlan, src_mac);
if (likely(f)) {
f->used = jiffies;
- if (likely(f->remote_ip == src_ip))
+ if (likely(f->remote.remote_ip == src_ip))
return;
if (net_ratelimit())
netdev_info(dev,
"%pM migrated from %pI4 to %pI4\n",
- src_mac, &f->remote_ip, &src_ip);
+ src_mac, &f->remote.remote_ip, &src_ip);
- f->remote_ip = src_ip;
+ f->remote.remote_ip = src_ip;
f->updated = jiffies;
} else {
/* learned new entry */
spin_lock(&vxlan->hash_lock);
err = vxlan_fdb_create(vxlan, src_mac, src_ip,
NUD_REACHABLE,
- NLM_F_EXCL|NLM_F_CREATE);
+ NLM_F_EXCL|NLM_F_CREATE,
+ vxlan_port, vxlan->vni, 0);
spin_unlock(&vxlan->hash_lock);
}
}
@@ -701,7 +812,7 @@
}
f = vxlan_find_mac(vxlan, n->ha);
- if (f && f->remote_ip == 0) {
+ if (f && f->remote.remote_ip == htonl(INADDR_ANY)) {
/* bridge-local neighbor */
neigh_release(n);
goto out;
@@ -820,48 +931,40 @@
return (((u64) hash * range) >> 32) + vxlan->port_min;
}
-/* Transmit local packets over Vxlan
- *
- * Outer IP header inherits ECN and DF from inner header.
- * Outer UDP destination is the VXLAN assigned port.
- * source port is based on hash of flow
- */
-static netdev_tx_t vxlan_xmit(struct sk_buff *skb, struct net_device *dev)
+static int handle_offloads(struct sk_buff *skb)
+{
+ if (skb_is_gso(skb)) {
+ int err = skb_unclone(skb, GFP_ATOMIC);
+ if (unlikely(err))
+ return err;
+
+ skb_shinfo(skb)->gso_type |= (SKB_GSO_UDP_TUNNEL | SKB_GSO_UDP);
+ } else if (skb->ip_summed != CHECKSUM_PARTIAL)
+ skb->ip_summed = CHECKSUM_NONE;
+
+ return 0;
+}
+
+static netdev_tx_t vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
+ struct vxlan_rdst *rdst, bool did_rsc)
{
struct vxlan_dev *vxlan = netdev_priv(dev);
struct rtable *rt;
const struct iphdr *old_iph;
- struct ethhdr *eth;
struct iphdr *iph;
struct vxlanhdr *vxh;
struct udphdr *uh;
struct flowi4 fl4;
unsigned int pkt_len = skb->len;
__be32 dst;
- __u16 src_port;
+ __u16 src_port, dst_port;
+ u32 vni;
__be16 df = 0;
__u8 tos, ttl;
- int err;
- bool did_rsc = false;
- const struct vxlan_fdb *f;
- skb_reset_mac_header(skb);
- eth = eth_hdr(skb);
-
- if ((vxlan->flags & VXLAN_F_PROXY) && ntohs(eth->h_proto) == ETH_P_ARP)
- return arp_reduce(dev, skb);
- else if ((vxlan->flags&VXLAN_F_RSC) && ntohs(eth->h_proto) == ETH_P_IP)
- did_rsc = route_shortcircuit(dev, skb);
-
- f = vxlan_find_mac(vxlan, eth->h_dest);
- if (f == NULL) {
- did_rsc = false;
- dst = vxlan->gaddr;
- if (!dst && (vxlan->flags & VXLAN_F_L2MISS) &&
- !is_multicast_ether_addr(eth->h_dest))
- vxlan_fdb_miss(vxlan, eth->h_dest);
- } else
- dst = f->remote_ip;
+ dst_port = rdst->remote_port ? rdst->remote_port : vxlan_port;
+ vni = rdst->remote_vni;
+ dst = rdst->remote_ip;
if (!dst) {
if (did_rsc) {
@@ -909,7 +1012,7 @@
src_port = vxlan_src_port(vxlan, skb);
memset(&fl4, 0, sizeof(fl4));
- fl4.flowi4_oif = vxlan->link;
+ fl4.flowi4_oif = rdst->remote_ifindex;
fl4.flowi4_tos = RT_TOS(tos);
fl4.daddr = dst;
fl4.saddr = vxlan->saddr;
@@ -936,13 +1039,13 @@
vxh = (struct vxlanhdr *) __skb_push(skb, sizeof(*vxh));
vxh->vx_flags = htonl(VXLAN_FLAGS);
- vxh->vx_vni = htonl(vxlan->vni << 8);
+ vxh->vx_vni = htonl(vni << 8);
__skb_push(skb, sizeof(*uh));
skb_reset_transport_header(skb);
uh = udp_hdr(skb);
- uh->dest = htons(vxlan_port);
+ uh->dest = htons(dst_port);
uh->source = htons(src_port);
uh->len = htons(skb->len);
@@ -965,22 +1068,10 @@
vxlan_set_owner(dev, skb);
- /* See iptunnel_xmit() */
- if (skb->ip_summed != CHECKSUM_PARTIAL)
- skb->ip_summed = CHECKSUM_NONE;
+ if (handle_offloads(skb))
+ goto drop;
- err = ip_local_out(skb);
- if (likely(net_xmit_eval(err) == 0)) {
- struct vxlan_stats *stats = this_cpu_ptr(vxlan->stats);
-
- u64_stats_update_begin(&stats->syncp);
- stats->tx_packets++;
- stats->tx_bytes += pkt_len;
- u64_stats_update_end(&stats->syncp);
- } else {
- dev->stats.tx_errors++;
- dev->stats.tx_aborted_errors++;
- }
+ iptunnel_xmit(skb, dev);
return NETDEV_TX_OK;
drop:
@@ -994,6 +1085,64 @@
return NETDEV_TX_OK;
}
+/* Transmit local packets over Vxlan
+ *
+ * Outer IP header inherits ECN and DF from inner header.
+ * Outer UDP destination is the VXLAN assigned port.
+ * source port is based on hash of flow
+ */
+static netdev_tx_t vxlan_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+ struct vxlan_dev *vxlan = netdev_priv(dev);
+ struct ethhdr *eth;
+ bool did_rsc = false;
+ struct vxlan_rdst group, *rdst0, *rdst;
+ struct vxlan_fdb *f;
+ int rc1, rc;
+
+ skb_reset_mac_header(skb);
+ eth = eth_hdr(skb);
+
+ if ((vxlan->flags & VXLAN_F_PROXY) && ntohs(eth->h_proto) == ETH_P_ARP)
+ return arp_reduce(dev, skb);
+ else if ((vxlan->flags&VXLAN_F_RSC) && ntohs(eth->h_proto) == ETH_P_IP)
+ did_rsc = route_shortcircuit(dev, skb);
+
+ f = vxlan_find_mac(vxlan, eth->h_dest);
+ if (f == NULL) {
+ did_rsc = false;
+ group.remote_port = vxlan_port;
+ group.remote_vni = vxlan->vni;
+ group.remote_ip = vxlan->gaddr;
+ group.remote_ifindex = vxlan->link;
+ group.remote_next = 0;
+ rdst0 = &group;
+
+ if (group.remote_ip == htonl(INADDR_ANY) &&
+ (vxlan->flags & VXLAN_F_L2MISS) &&
+ !is_multicast_ether_addr(eth->h_dest))
+ vxlan_fdb_miss(vxlan, eth->h_dest);
+ } else
+ rdst0 = &f->remote;
+
+ rc = NETDEV_TX_OK;
+
+ /* if there are multiple destinations, send copies */
+ for (rdst = rdst0->remote_next; rdst; rdst = rdst->remote_next) {
+ struct sk_buff *skb1;
+
+ skb1 = skb_clone(skb, GFP_ATOMIC);
+ rc1 = vxlan_xmit_one(skb1, dev, rdst, did_rsc);
+ if (rc == NETDEV_TX_OK)
+ rc = rc1;
+ }
+
+ rc1 = vxlan_xmit_one(skb, dev, rdst0, did_rsc);
+ if (rc == NETDEV_TX_OK)
+ rc = rc1;
+ return rc;
+}
+
/* Walk the forwarding table and purge stale entries */
static void vxlan_cleanup(unsigned long arg)
{
@@ -1189,8 +1338,10 @@
dev->features |= NETIF_F_NETNS_LOCAL;
dev->features |= NETIF_F_SG | NETIF_F_HW_CSUM;
dev->features |= NETIF_F_RXCSUM;
+ dev->features |= NETIF_F_GSO_SOFTWARE;
dev->hw_features |= NETIF_F_SG | NETIF_F_HW_CSUM | NETIF_F_RXCSUM;
+ dev->hw_features |= NETIF_F_GSO_SOFTWARE;
dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
dev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
@@ -1555,6 +1706,7 @@
{
rtnl_link_unregister(&vxlan_link_ops);
unregister_pernet_device(&vxlan_net_ops);
+ rcu_barrier();
}
module_exit(vxlan_cleanup_module);
diff --git a/drivers/net/wireless/ath/wil6210/txrx.c b/drivers/net/wireless/ath/wil6210/txrx.c
index d1315b4..55dd95f 100644
--- a/drivers/net/wireless/ath/wil6210/txrx.c
+++ b/drivers/net/wireless/ath/wil6210/txrx.c
@@ -83,8 +83,6 @@
*/
vring->va = dma_alloc_coherent(dev, sz, &vring->pa, GFP_KERNEL);
if (!vring->va) {
- wil_err(wil, "vring_alloc [%d] failed to alloc DMA mem\n",
- vring->size);
kfree(vring->ctx);
vring->ctx = NULL;
return -ENOMEM;
diff --git a/drivers/net/wireless/b43/dma.c b/drivers/net/wireless/b43/dma.c
index 38bc5a7..f73cbb5 100644
--- a/drivers/net/wireless/b43/dma.c
+++ b/drivers/net/wireless/b43/dma.c
@@ -419,8 +419,6 @@
static int alloc_ringmemory(struct b43_dmaring *ring)
{
- gfp_t flags = GFP_KERNEL;
-
/* The specs call for 4K buffers for 30- and 32-bit DMA with 4K
* alignment and 8K buffers for 64-bit DMA with 8K alignment.
* In practice we could use smaller buffers for the latter, but the
@@ -435,12 +433,9 @@
ring->descbase = dma_alloc_coherent(ring->dev->dev->dma_dev,
ring_mem_size, &(ring->dmabase),
- flags);
- if (!ring->descbase) {
- b43err(ring->dev->wl, "DMA ringmemory allocation failed\n");
+ GFP_KERNEL | __GFP_ZERO);
+ if (!ring->descbase)
return -ENOMEM;
- }
- memset(ring->descbase, 0, ring_mem_size);
return 0;
}
diff --git a/drivers/net/wireless/b43legacy/dma.c b/drivers/net/wireless/b43legacy/dma.c
index 2d3c664..faeafe2 100644
--- a/drivers/net/wireless/b43legacy/dma.c
+++ b/drivers/net/wireless/b43legacy/dma.c
@@ -334,13 +334,9 @@
ring->descbase = dma_alloc_coherent(ring->dev->dev->dma_dev,
B43legacy_DMA_RINGMEMSIZE,
&(ring->dmabase),
- GFP_KERNEL);
- if (!ring->descbase) {
- b43legacyerr(ring->dev->wl, "DMA ringmemory allocation"
- " failed\n");
+ GFP_KERNEL | __GFP_ZERO);
+ if (!ring->descbase)
return -ENOMEM;
- }
- memset(ring->descbase, 0, B43legacy_DMA_RINGMEMSIZE);
return 0;
}
diff --git a/drivers/net/wireless/iwlegacy/3945.c b/drivers/net/wireless/iwlegacy/3945.c
index e0b9d7f..dc1e6da 100644
--- a/drivers/net/wireless/iwlegacy/3945.c
+++ b/drivers/net/wireless/iwlegacy/3945.c
@@ -2379,10 +2379,8 @@
il->_3945.shared_virt =
dma_alloc_coherent(&il->pci_dev->dev, sizeof(struct il3945_shared),
&il->_3945.shared_phys, GFP_KERNEL);
- if (!il->_3945.shared_virt) {
- IL_ERR("failed to allocate pci memory\n");
+ if (!il->_3945.shared_virt)
return -ENOMEM;
- }
il->hw_params.bcast_id = IL3945_BROADCAST_ID;
diff --git a/drivers/net/wireless/iwlegacy/4965-mac.c b/drivers/net/wireless/iwlegacy/4965-mac.c
index 7941eb3..238f528 100644
--- a/drivers/net/wireless/iwlegacy/4965-mac.c
+++ b/drivers/net/wireless/iwlegacy/4965-mac.c
@@ -1921,8 +1921,8 @@
static inline int
il4965_alloc_dma_ptr(struct il_priv *il, struct il_dma_ptr *ptr, size_t size)
{
- ptr->addr =
- dma_alloc_coherent(&il->pci_dev->dev, size, &ptr->dma, GFP_KERNEL);
+ ptr->addr = dma_alloc_coherent(&il->pci_dev->dev, size, &ptr->dma,
+ GFP_KERNEL);
if (!ptr->addr)
return -ENOMEM;
ptr->size = size;
diff --git a/drivers/net/wireless/iwlegacy/common.c b/drivers/net/wireless/iwlegacy/common.c
index e006ea8..db21871 100644
--- a/drivers/net/wireless/iwlegacy/common.c
+++ b/drivers/net/wireless/iwlegacy/common.c
@@ -2566,15 +2566,13 @@
INIT_LIST_HEAD(&rxq->rx_used);
/* Alloc the circular buffer of Read Buffer Descriptors (RBDs) */
- rxq->bd =
- dma_alloc_coherent(dev, 4 * RX_QUEUE_SIZE, &rxq->bd_dma,
- GFP_KERNEL);
+ rxq->bd = dma_alloc_coherent(dev, 4 * RX_QUEUE_SIZE, &rxq->bd_dma,
+ GFP_KERNEL);
if (!rxq->bd)
goto err_bd;
- rxq->rb_stts =
- dma_alloc_coherent(dev, sizeof(struct il_rb_status),
- &rxq->rb_stts_dma, GFP_KERNEL);
+ rxq->rb_stts = dma_alloc_coherent(dev, sizeof(struct il_rb_status),
+ &rxq->rb_stts_dma, GFP_KERNEL);
if (!rxq->rb_stts)
goto err_rb;
@@ -2941,10 +2939,9 @@
* shared with device */
txq->tfds =
dma_alloc_coherent(dev, tfd_sz, &txq->q.dma_addr, GFP_KERNEL);
- if (!txq->tfds) {
- IL_ERR("Fail to alloc TFDs\n");
+ if (!txq->tfds)
goto error;
- }
+
txq->q.id = id;
return 0;
diff --git a/drivers/net/wireless/iwlegacy/common.h b/drivers/net/wireless/iwlegacy/common.h
index 96f2025..458e699 100644
--- a/drivers/net/wireless/iwlegacy/common.h
+++ b/drivers/net/wireless/iwlegacy/common.h
@@ -2235,9 +2235,8 @@
return -EINVAL;
}
- desc->v_addr =
- dma_alloc_coherent(&pci_dev->dev, desc->len, &desc->p_addr,
- GFP_KERNEL);
+ desc->v_addr = dma_alloc_coherent(&pci_dev->dev, desc->len,
+ &desc->p_addr, GFP_KERNEL);
return (desc->v_addr != NULL) ? 0 : -ENOMEM;
}
diff --git a/drivers/net/wireless/iwlwifi/pcie/tx.c b/drivers/net/wireless/iwlwifi/pcie/tx.c
index 8595c16..7a508d8 100644
--- a/drivers/net/wireless/iwlwifi/pcie/tx.c
+++ b/drivers/net/wireless/iwlwifi/pcie/tx.c
@@ -501,10 +501,8 @@
* shared with device */
txq->tfds = dma_alloc_coherent(trans->dev, tfd_sz,
&txq->q.dma_addr, GFP_KERNEL);
- if (!txq->tfds) {
- IWL_ERR(trans, "dma_alloc_coherent(%zd) failed\n", tfd_sz);
+ if (!txq->tfds)
goto error;
- }
BUILD_BUG_ON(IWL_HCMD_SCRATCHBUF_SIZE != sizeof(*txq->scratchbufs));
BUILD_BUG_ON(offsetof(struct iwl_pcie_txq_scratch_buf, scratch) !=
diff --git a/drivers/net/wireless/rt2x00/rt2x00pci.c b/drivers/net/wireless/rt2x00/rt2x00pci.c
index a0c8cae..696abed 100644
--- a/drivers/net/wireless/rt2x00/rt2x00pci.c
+++ b/drivers/net/wireless/rt2x00/rt2x00pci.c
@@ -124,12 +124,10 @@
*/
addr = dma_alloc_coherent(rt2x00dev->dev,
queue->limit * queue->desc_size,
- &dma, GFP_KERNEL);
+ &dma, GFP_KERNEL | __GFP_ZERO);
if (!addr)
return -ENOMEM;
- memset(addr, 0, queue->limit * queue->desc_size);
-
/*
* Initialize all queue entries to contain valid addresses.
*/
diff --git a/drivers/scsi/csiostor/Makefile b/drivers/scsi/csiostor/Makefile
index b581966..913b9a9 100644
--- a/drivers/scsi/csiostor/Makefile
+++ b/drivers/scsi/csiostor/Makefile
@@ -8,4 +8,5 @@
obj-$(CONFIG_SCSI_CHELSIO_FCOE) += csiostor.o
csiostor-objs := csio_attr.o csio_init.o csio_lnode.o csio_scsi.o \
- csio_hw.o csio_isr.o csio_mb.o csio_rnode.o csio_wr.o
+ csio_hw.o csio_hw_t4.o csio_hw_t5.o csio_isr.o \
+ csio_mb.o csio_rnode.o csio_wr.o
diff --git a/drivers/scsi/csiostor/csio_hw.c b/drivers/scsi/csiostor/csio_hw.c
index bdd78fb..a0b4c89 100644
--- a/drivers/scsi/csiostor/csio_hw.c
+++ b/drivers/scsi/csiostor/csio_hw.c
@@ -61,7 +61,7 @@
static int dev_num;
/* FCoE Adapter types & its description */
-static const struct csio_adap_desc csio_fcoe_adapters[] = {
+static const struct csio_adap_desc csio_t4_fcoe_adapters[] = {
{"T440-Dbg 10G", "Chelsio T440-Dbg 10G [FCoE]"},
{"T420-CR 10G", "Chelsio T420-CR 10G [FCoE]"},
{"T422-CR 10G/1G", "Chelsio T422-CR 10G/1G [FCoE]"},
@@ -77,7 +77,38 @@
{"B404-BT 1G", "Chelsio B404-BT 1G [FCoE]"},
{"T480-CR 10G", "Chelsio T480-CR 10G [FCoE]"},
{"T440-LP-CR 10G", "Chelsio T440-LP-CR 10G [FCoE]"},
- {"T4 FPGA", "Chelsio T4 FPGA [FCoE]"}
+ {"AMSTERDAM 10G", "Chelsio AMSTERDAM 10G [FCoE]"},
+ {"HUAWEI T480 10G", "Chelsio HUAWEI T480 10G [FCoE]"},
+ {"HUAWEI T440 10G", "Chelsio HUAWEI T440 10G [FCoE]"},
+ {"HUAWEI STG 10G", "Chelsio HUAWEI STG 10G [FCoE]"},
+ {"ACROMAG XAUI 10G", "Chelsio ACROMAG XAUI 10G [FCoE]"},
+ {"ACROMAG SFP+ 10G", "Chelsio ACROMAG SFP+ 10G [FCoE]"},
+ {"QUANTA SFP+ 10G", "Chelsio QUANTA SFP+ 10G [FCoE]"},
+ {"HUAWEI 10Gbase-T", "Chelsio HUAWEI 10Gbase-T [FCoE]"},
+ {"HUAWEI T4TOE 10G", "Chelsio HUAWEI T4TOE 10G [FCoE]"}
+};
+
+static const struct csio_adap_desc csio_t5_fcoe_adapters[] = {
+ {"T580-Dbg 10G", "Chelsio T580-Dbg 10G [FCoE]"},
+ {"T520-CR 10G", "Chelsio T520-CR 10G [FCoE]"},
+ {"T522-CR 10G/1G", "Chelsio T452-CR 10G/1G [FCoE]"},
+ {"T540-CR 10G", "Chelsio T540-CR 10G [FCoE]"},
+ {"T520-BCH 10G", "Chelsio T520-BCH 10G [FCoE]"},
+ {"T540-BCH 10G", "Chelsio T540-BCH 10G [FCoE]"},
+ {"T540-CH 10G", "Chelsio T540-CH 10G [FCoE]"},
+ {"T520-SO 10G", "Chelsio T520-SO 10G [FCoE]"},
+ {"T520-CX4 10G", "Chelsio T520-CX4 10G [FCoE]"},
+ {"T520-BT 10G", "Chelsio T520-BT 10G [FCoE]"},
+ {"T504-BT 1G", "Chelsio T504-BT 1G [FCoE]"},
+ {"B520-SR 10G", "Chelsio B520-SR 10G [FCoE]"},
+ {"B504-BT 1G", "Chelsio B504-BT 1G [FCoE]"},
+ {"T580-CR 10G", "Chelsio T580-CR 10G [FCoE]"},
+ {"T540-LP-CR 10G", "Chelsio T540-LP-CR 10G [FCoE]"},
+ {"AMSTERDAM 10G", "Chelsio AMSTERDAM 10G [FCoE]"},
+ {"T580-LP-CR 40G", "Chelsio T580-LP-CR 40G [FCoE]"},
+ {"T520-LL-CR 10G", "Chelsio T520-LL-CR 10G [FCoE]"},
+ {"T560-CR 40G", "Chelsio T560-CR 40G [FCoE]"},
+ {"T580-CR 40G", "Chelsio T580-CR 40G [FCoE]"}
};
static void csio_mgmtm_cleanup(struct csio_mgmtm *);
@@ -124,7 +155,7 @@
* at the time it indicated completion is stored there. Returns 0 if the
* operation completes and -EAGAIN otherwise.
*/
-static int
+int
csio_hw_wait_op_done_val(struct csio_hw *hw, int reg, uint32_t mask,
int polarity, int attempts, int delay, uint32_t *valp)
{
@@ -145,6 +176,24 @@
}
}
+/*
+ * csio_hw_tp_wr_bits_indirect - set/clear bits in an indirect TP register
+ * @hw: the adapter
+ * @addr: the indirect TP register address
+ * @mask: specifies the field within the register to modify
+ * @val: new value for the field
+ *
+ * Sets a field of an indirect TP register to the given value.
+ */
+void
+csio_hw_tp_wr_bits_indirect(struct csio_hw *hw, unsigned int addr,
+ unsigned int mask, unsigned int val)
+{
+ csio_wr_reg32(hw, addr, TP_PIO_ADDR);
+ val |= csio_rd_reg32(hw, TP_PIO_DATA) & ~mask;
+ csio_wr_reg32(hw, val, TP_PIO_DATA);
+}
+
void
csio_set_reg_field(struct csio_hw *hw, uint32_t reg, uint32_t mask,
uint32_t value)
@@ -157,242 +206,22 @@
}
-/*
- * csio_hw_mc_read - read from MC through backdoor accesses
- * @hw: the hw module
- * @addr: address of first byte requested
- * @data: 64 bytes of data containing the requested address
- * @ecc: where to store the corresponding 64-bit ECC word
- *
- * Read 64 bytes of data from MC starting at a 64-byte-aligned address
- * that covers the requested address @addr. If @parity is not %NULL it
- * is assigned the 64-bit ECC word for the read data.
- */
-int
-csio_hw_mc_read(struct csio_hw *hw, uint32_t addr, __be32 *data,
- uint64_t *ecc)
-{
- int i;
-
- if (csio_rd_reg32(hw, MC_BIST_CMD) & START_BIST)
- return -EBUSY;
- csio_wr_reg32(hw, addr & ~0x3fU, MC_BIST_CMD_ADDR);
- csio_wr_reg32(hw, 64, MC_BIST_CMD_LEN);
- csio_wr_reg32(hw, 0xc, MC_BIST_DATA_PATTERN);
- csio_wr_reg32(hw, BIST_OPCODE(1) | START_BIST | BIST_CMD_GAP(1),
- MC_BIST_CMD);
- i = csio_hw_wait_op_done_val(hw, MC_BIST_CMD, START_BIST,
- 0, 10, 1, NULL);
- if (i)
- return i;
-
-#define MC_DATA(i) MC_BIST_STATUS_REG(MC_BIST_STATUS_RDATA, i)
-
- for (i = 15; i >= 0; i--)
- *data++ = htonl(csio_rd_reg32(hw, MC_DATA(i)));
- if (ecc)
- *ecc = csio_rd_reg64(hw, MC_DATA(16));
-#undef MC_DATA
- return 0;
-}
-
-/*
- * csio_hw_edc_read - read from EDC through backdoor accesses
- * @hw: the hw module
- * @idx: which EDC to access
- * @addr: address of first byte requested
- * @data: 64 bytes of data containing the requested address
- * @ecc: where to store the corresponding 64-bit ECC word
- *
- * Read 64 bytes of data from EDC starting at a 64-byte-aligned address
- * that covers the requested address @addr. If @parity is not %NULL it
- * is assigned the 64-bit ECC word for the read data.
- */
-int
-csio_hw_edc_read(struct csio_hw *hw, int idx, uint32_t addr, __be32 *data,
- uint64_t *ecc)
-{
- int i;
-
- idx *= EDC_STRIDE;
- if (csio_rd_reg32(hw, EDC_BIST_CMD + idx) & START_BIST)
- return -EBUSY;
- csio_wr_reg32(hw, addr & ~0x3fU, EDC_BIST_CMD_ADDR + idx);
- csio_wr_reg32(hw, 64, EDC_BIST_CMD_LEN + idx);
- csio_wr_reg32(hw, 0xc, EDC_BIST_DATA_PATTERN + idx);
- csio_wr_reg32(hw, BIST_OPCODE(1) | BIST_CMD_GAP(1) | START_BIST,
- EDC_BIST_CMD + idx);
- i = csio_hw_wait_op_done_val(hw, EDC_BIST_CMD + idx, START_BIST,
- 0, 10, 1, NULL);
- if (i)
- return i;
-
-#define EDC_DATA(i) (EDC_BIST_STATUS_REG(EDC_BIST_STATUS_RDATA, i) + idx)
-
- for (i = 15; i >= 0; i--)
- *data++ = htonl(csio_rd_reg32(hw, EDC_DATA(i)));
- if (ecc)
- *ecc = csio_rd_reg64(hw, EDC_DATA(16));
-#undef EDC_DATA
- return 0;
-}
-
-/*
- * csio_mem_win_rw - read/write memory through PCIE memory window
- * @hw: the adapter
- * @addr: address of first byte requested
- * @data: MEMWIN0_APERTURE bytes of data containing the requested address
- * @dir: direction of transfer 1 => read, 0 => write
- *
- * Read/write MEMWIN0_APERTURE bytes of data from MC starting at a
- * MEMWIN0_APERTURE-byte-aligned address that covers the requested
- * address @addr.
- */
-static int
-csio_mem_win_rw(struct csio_hw *hw, u32 addr, u32 *data, int dir)
-{
- int i;
-
- /*
- * Setup offset into PCIE memory window. Address must be a
- * MEMWIN0_APERTURE-byte-aligned address. (Read back MA register to
- * ensure that changes propagate before we attempt to use the new
- * values.)
- */
- csio_wr_reg32(hw, addr & ~(MEMWIN0_APERTURE - 1),
- PCIE_MEM_ACCESS_OFFSET);
- csio_rd_reg32(hw, PCIE_MEM_ACCESS_OFFSET);
-
- /* Collecting data 4 bytes at a time upto MEMWIN0_APERTURE */
- for (i = 0; i < MEMWIN0_APERTURE; i = i + sizeof(__be32)) {
- if (dir)
- *data++ = csio_rd_reg32(hw, (MEMWIN0_BASE + i));
- else
- csio_wr_reg32(hw, *data++, (MEMWIN0_BASE + i));
- }
-
- return 0;
-}
-
-/*
- * csio_memory_rw - read/write EDC 0, EDC 1 or MC via PCIE memory window
- * @hw: the csio_hw
- * @mtype: memory type: MEM_EDC0, MEM_EDC1 or MEM_MC
- * @addr: address within indicated memory type
- * @len: amount of memory to transfer
- * @buf: host memory buffer
- * @dir: direction of transfer 1 => read, 0 => write
- *
- * Reads/writes an [almost] arbitrary memory region in the firmware: the
- * firmware memory address, length and host buffer must be aligned on
- * 32-bit boudaries. The memory is transferred as a raw byte sequence
- * from/to the firmware's memory. If this memory contains data
- * structures which contain multi-byte integers, it's the callers
- * responsibility to perform appropriate byte order conversions.
- */
-static int
-csio_memory_rw(struct csio_hw *hw, int mtype, u32 addr, u32 len,
- uint32_t *buf, int dir)
-{
- uint32_t pos, start, end, offset, memoffset;
- int ret;
- uint32_t *data;
-
- /*
- * Argument sanity checks ...
- */
- if ((addr & 0x3) || (len & 0x3))
- return -EINVAL;
-
- data = kzalloc(MEMWIN0_APERTURE, GFP_KERNEL);
- if (!data)
- return -ENOMEM;
-
- /* Offset into the region of memory which is being accessed
- * MEM_EDC0 = 0
- * MEM_EDC1 = 1
- * MEM_MC = 2
- */
- memoffset = (mtype * (5 * 1024 * 1024));
-
- /* Determine the PCIE_MEM_ACCESS_OFFSET */
- addr = addr + memoffset;
-
- /*
- * The underlaying EDC/MC read routines read MEMWIN0_APERTURE bytes
- * at a time so we need to round down the start and round up the end.
- * We'll start copying out of the first line at (addr - start) a word
- * at a time.
- */
- start = addr & ~(MEMWIN0_APERTURE-1);
- end = (addr + len + MEMWIN0_APERTURE-1) & ~(MEMWIN0_APERTURE-1);
- offset = (addr - start)/sizeof(__be32);
-
- for (pos = start; pos < end; pos += MEMWIN0_APERTURE, offset = 0) {
- /*
- * If we're writing, copy the data from the caller's memory
- * buffer
- */
- if (!dir) {
- /*
- * If we're doing a partial write, then we need to do
- * a read-modify-write ...
- */
- if (offset || len < MEMWIN0_APERTURE) {
- ret = csio_mem_win_rw(hw, pos, data, 1);
- if (ret) {
- kfree(data);
- return ret;
- }
- }
- while (offset < (MEMWIN0_APERTURE/sizeof(__be32)) &&
- len > 0) {
- data[offset++] = *buf++;
- len -= sizeof(__be32);
- }
- }
-
- /*
- * Transfer a block of memory and bail if there's an error.
- */
- ret = csio_mem_win_rw(hw, pos, data, dir);
- if (ret) {
- kfree(data);
- return ret;
- }
-
- /*
- * If we're reading, copy the data into the caller's memory
- * buffer.
- */
- if (dir)
- while (offset < (MEMWIN0_APERTURE/sizeof(__be32)) &&
- len > 0) {
- *buf++ = data[offset++];
- len -= sizeof(__be32);
- }
- }
-
- kfree(data);
-
- return 0;
-}
-
static int
csio_memory_write(struct csio_hw *hw, int mtype, u32 addr, u32 len, u32 *buf)
{
- return csio_memory_rw(hw, mtype, addr, len, buf, 0);
+ return hw->chip_ops->chip_memory_rw(hw, MEMWIN_CSIOSTOR, mtype,
+ addr, len, buf, 0);
}
/*
* EEPROM reads take a few tens of us while writes can take a bit over 5 ms.
*/
-#define EEPROM_MAX_RD_POLL 40
-#define EEPROM_MAX_WR_POLL 6
-#define EEPROM_STAT_ADDR 0x7bfc
-#define VPD_BASE 0x400
-#define VPD_BASE_OLD 0
-#define VPD_LEN 512
+#define EEPROM_MAX_RD_POLL 40
+#define EEPROM_MAX_WR_POLL 6
+#define EEPROM_STAT_ADDR 0x7bfc
+#define VPD_BASE 0x400
+#define VPD_BASE_OLD 0
+#define VPD_LEN 1024
#define VPD_INFO_FLD_HDR_SIZE 3
/*
@@ -817,23 +646,6 @@
return 0;
}
-/*
- * csio_hw_flash_cfg_addr - return the address of the flash
- * configuration file
- * @hw: the HW module
- *
- * Return the address within the flash where the Firmware Configuration
- * File is stored.
- */
-static unsigned int
-csio_hw_flash_cfg_addr(struct csio_hw *hw)
-{
- if (hw->params.sf_size == 0x100000)
- return FPGA_FLASH_CFG_OFFSET;
- else
- return FLASH_CFG_OFFSET;
-}
-
static void
csio_hw_print_fw_version(struct csio_hw *hw, char *str)
{
@@ -898,13 +710,13 @@
minor = FW_HDR_FW_VER_MINOR_GET(hw->fwrev);
micro = FW_HDR_FW_VER_MICRO_GET(hw->fwrev);
- if (major != FW_VERSION_MAJOR) { /* major mismatch - fail */
+ if (major != FW_VERSION_MAJOR(hw)) { /* major mismatch - fail */
csio_err(hw, "card FW has major version %u, driver wants %u\n",
- major, FW_VERSION_MAJOR);
+ major, FW_VERSION_MAJOR(hw));
return -EINVAL;
}
- if (minor == FW_VERSION_MINOR && micro == FW_VERSION_MICRO)
+ if (minor == FW_VERSION_MINOR(hw) && micro == FW_VERSION_MICRO(hw))
return 0; /* perfect match */
/* Minor/micro version mismatch */
@@ -1044,7 +856,7 @@
csio_set_pcie_completion_timeout(struct csio_hw *hw, u8 range)
{
uint16_t val;
- uint32_t pcie_cap;
+ int pcie_cap;
if (!csio_pci_capability(hw->pdev, PCI_CAP_ID_EXP, &pcie_cap)) {
pci_read_config_word(hw->pdev,
@@ -1056,84 +868,6 @@
}
}
-
-/*
- * Return the specified PCI-E Configuration Space register from our Physical
- * Function. We try first via a Firmware LDST Command since we prefer to let
- * the firmware own all of these registers, but if that fails we go for it
- * directly ourselves.
- */
-static uint32_t
-csio_read_pcie_cfg4(struct csio_hw *hw, int reg)
-{
- u32 val = 0;
- struct csio_mb *mbp;
- int rv;
- struct fw_ldst_cmd *ldst_cmd;
-
- mbp = mempool_alloc(hw->mb_mempool, GFP_ATOMIC);
- if (!mbp) {
- CSIO_INC_STATS(hw, n_err_nomem);
- pci_read_config_dword(hw->pdev, reg, &val);
- return val;
- }
-
- csio_mb_ldst(hw, mbp, CSIO_MB_DEFAULT_TMO, reg);
-
- rv = csio_mb_issue(hw, mbp);
-
- /*
- * If the LDST Command suucceeded, exctract the returned register
- * value. Otherwise read it directly ourself.
- */
- if (rv == 0) {
- ldst_cmd = (struct fw_ldst_cmd *)(mbp->mb);
- val = ntohl(ldst_cmd->u.pcie.data[0]);
- } else
- pci_read_config_dword(hw->pdev, reg, &val);
-
- mempool_free(mbp, hw->mb_mempool);
-
- return val;
-} /* csio_read_pcie_cfg4 */
-
-static int
-csio_hw_set_mem_win(struct csio_hw *hw)
-{
- u32 bar0;
-
- /*
- * Truncation intentional: we only read the bottom 32-bits of the
- * 64-bit BAR0/BAR1 ... We use the hardware backdoor mechanism to
- * read BAR0 instead of using pci_resource_start() because we could be
- * operating from within a Virtual Machine which is trapping our
- * accesses to our Configuration Space and we need to set up the PCI-E
- * Memory Window decoders with the actual addresses which will be
- * coming across the PCI-E link.
- */
- bar0 = csio_read_pcie_cfg4(hw, PCI_BASE_ADDRESS_0);
- bar0 &= PCI_BASE_ADDRESS_MEM_MASK;
-
- /*
- * Set up memory window for accessing adapter memory ranges. (Read
- * back MA register to ensure that changes propagate before we attempt
- * to use the new values.)
- */
- csio_wr_reg32(hw, (bar0 + MEMWIN0_BASE) | BIR(0) |
- WINDOW(ilog2(MEMWIN0_APERTURE) - 10),
- PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_BASE_WIN, 0));
- csio_wr_reg32(hw, (bar0 + MEMWIN1_BASE) | BIR(0) |
- WINDOW(ilog2(MEMWIN1_APERTURE) - 10),
- PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_BASE_WIN, 1));
- csio_wr_reg32(hw, (bar0 + MEMWIN2_BASE) | BIR(0) |
- WINDOW(ilog2(MEMWIN2_APERTURE) - 10),
- PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_BASE_WIN, 2));
- csio_rd_reg32(hw, PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_BASE_WIN, 2));
- return 0;
-} /* csio_hw_set_mem_win */
-
-
-
/*****************************************************************************/
/* HW State machine assists */
/*****************************************************************************/
@@ -1234,7 +968,9 @@
for (;;) {
uint32_t pcie_fw;
+ spin_unlock_irq(&hw->lock);
msleep(50);
+ spin_lock_irq(&hw->lock);
waiting -= 50;
/*
@@ -2121,9 +1857,9 @@
uint32_t *cfg_data;
int value_to_add = 0;
- if (request_firmware(&cf, CSIO_CF_FNAME, dev) < 0) {
- csio_err(hw, "could not find config file " CSIO_CF_FNAME
- ",err: %d\n", ret);
+ if (request_firmware(&cf, CSIO_CF_FNAME(hw), dev) < 0) {
+ csio_err(hw, "could not find config file %s, err: %d\n",
+ CSIO_CF_FNAME(hw), ret);
return -ENOENT;
}
@@ -2147,9 +1883,24 @@
ret = csio_memory_write(hw, mtype, maddr,
cf->size + value_to_add, cfg_data);
+
+ if ((ret == 0) && (value_to_add != 0)) {
+ union {
+ u32 word;
+ char buf[4];
+ } last;
+ size_t size = cf->size & ~0x3;
+ int i;
+
+ last.word = cfg_data[size >> 2];
+ for (i = value_to_add; i < 4; i++)
+ last.buf[i] = 0;
+ ret = csio_memory_write(hw, mtype, maddr + size, 4, &last.word);
+ }
if (ret == 0) {
- csio_info(hw, "config file upgraded to " CSIO_CF_FNAME "\n");
- strncpy(path, "/lib/firmware/" CSIO_CF_FNAME, 64);
+ csio_info(hw, "config file upgraded to %s\n",
+ CSIO_CF_FNAME(hw));
+ snprintf(path, 64, "%s%s", "/lib/firmware/", CSIO_CF_FNAME(hw));
}
leave:
@@ -2179,7 +1930,7 @@
{
unsigned int mtype, maddr;
int rv;
- uint32_t finiver, finicsum, cfcsum;
+ uint32_t finiver = 0, finicsum = 0, cfcsum = 0;
int using_flash;
char path[64];
@@ -2207,7 +1958,7 @@
* config file from flash.
*/
mtype = FW_MEMTYPE_CF_FLASH;
- maddr = csio_hw_flash_cfg_addr(hw);
+ maddr = hw->chip_ops->chip_flash_cfg_addr(hw);
using_flash = 1;
} else {
/*
@@ -2346,30 +2097,32 @@
struct pci_dev *pci_dev = hw->pdev;
struct device *dev = &pci_dev->dev ;
- if (request_firmware(&fw, CSIO_FW_FNAME, dev) < 0) {
- csio_err(hw, "could not find firmware image " CSIO_FW_FNAME
- ",err: %d\n", ret);
+ if (request_firmware(&fw, CSIO_FW_FNAME(hw), dev) < 0) {
+ csio_err(hw, "could not find firmware image %s, err: %d\n",
+ CSIO_FW_FNAME(hw), ret);
return -EINVAL;
}
hdr = (const struct fw_hdr *)fw->data;
fw_ver = ntohl(hdr->fw_ver);
- if (FW_HDR_FW_VER_MAJOR_GET(fw_ver) != FW_VERSION_MAJOR)
+ if (FW_HDR_FW_VER_MAJOR_GET(fw_ver) != FW_VERSION_MAJOR(hw))
return -EINVAL; /* wrong major version, won't do */
/*
* If the flash FW is unusable or we found something newer, load it.
*/
- if (FW_HDR_FW_VER_MAJOR_GET(hw->fwrev) != FW_VERSION_MAJOR ||
+ if (FW_HDR_FW_VER_MAJOR_GET(hw->fwrev) != FW_VERSION_MAJOR(hw) ||
fw_ver > hw->fwrev) {
ret = csio_hw_fw_upgrade(hw, hw->pfn, fw->data, fw->size,
/*force=*/false);
if (!ret)
- csio_info(hw, "firmware upgraded to version %pI4 from "
- CSIO_FW_FNAME "\n", &hdr->fw_ver);
+ csio_info(hw,
+ "firmware upgraded to version %pI4 from %s\n",
+ &hdr->fw_ver, CSIO_FW_FNAME(hw));
else
csio_err(hw, "firmware upgrade failed! err=%d\n", ret);
- }
+ } else
+ ret = -EINVAL;
release_firmware(fw);
@@ -2410,7 +2163,7 @@
/* Set pci completion timeout value to 4 seconds. */
csio_set_pcie_completion_timeout(hw, 0xd);
- csio_hw_set_mem_win(hw);
+ hw->chip_ops->chip_set_mem_win(hw, MEMWIN_CSIOSTOR);
rv = csio_hw_get_fw_version(hw, &hw->fwrev);
if (rv != 0)
@@ -2478,6 +2231,8 @@
} else {
if (hw->fw_state == CSIO_DEV_STATE_INIT) {
+ hw->flags |= CSIO_HWF_USING_SOFT_PARAMS;
+
/* device parameters */
rv = csio_get_device_params(hw);
if (rv != 0)
@@ -2651,7 +2406,7 @@
}
-static void
+void
csio_hw_fatal_err(struct csio_hw *hw)
{
csio_set_reg_field(hw, SGE_CONTROL, GLOBALENABLE, 0);
@@ -2990,14 +2745,6 @@
/* END: HW SM */
/*****************************************************************************/
-/* Slow path handlers */
-struct intr_info {
- unsigned int mask; /* bits to check in interrupt status */
- const char *msg; /* message to print or NULL */
- short stat_idx; /* stat counter to increment or -1 */
- unsigned short fatal; /* whether the condition reported is fatal */
-};
-
/*
* csio_handle_intr_status - table driven interrupt handler
* @hw: HW instance
@@ -3011,7 +2758,7 @@
* by an entry specifying mask 0. Returns the number of fatal interrupt
* conditions.
*/
-static int
+int
csio_handle_intr_status(struct csio_hw *hw, unsigned int reg,
const struct intr_info *acts)
{
@@ -3038,80 +2785,6 @@
}
/*
- * Interrupt handler for the PCIE module.
- */
-static void
-csio_pcie_intr_handler(struct csio_hw *hw)
-{
- static struct intr_info sysbus_intr_info[] = {
- { RNPP, "RXNP array parity error", -1, 1 },
- { RPCP, "RXPC array parity error", -1, 1 },
- { RCIP, "RXCIF array parity error", -1, 1 },
- { RCCP, "Rx completions control array parity error", -1, 1 },
- { RFTP, "RXFT array parity error", -1, 1 },
- { 0, NULL, 0, 0 }
- };
- static struct intr_info pcie_port_intr_info[] = {
- { TPCP, "TXPC array parity error", -1, 1 },
- { TNPP, "TXNP array parity error", -1, 1 },
- { TFTP, "TXFT array parity error", -1, 1 },
- { TCAP, "TXCA array parity error", -1, 1 },
- { TCIP, "TXCIF array parity error", -1, 1 },
- { RCAP, "RXCA array parity error", -1, 1 },
- { OTDD, "outbound request TLP discarded", -1, 1 },
- { RDPE, "Rx data parity error", -1, 1 },
- { TDUE, "Tx uncorrectable data error", -1, 1 },
- { 0, NULL, 0, 0 }
- };
- static struct intr_info pcie_intr_info[] = {
- { MSIADDRLPERR, "MSI AddrL parity error", -1, 1 },
- { MSIADDRHPERR, "MSI AddrH parity error", -1, 1 },
- { MSIDATAPERR, "MSI data parity error", -1, 1 },
- { MSIXADDRLPERR, "MSI-X AddrL parity error", -1, 1 },
- { MSIXADDRHPERR, "MSI-X AddrH parity error", -1, 1 },
- { MSIXDATAPERR, "MSI-X data parity error", -1, 1 },
- { MSIXDIPERR, "MSI-X DI parity error", -1, 1 },
- { PIOCPLPERR, "PCI PIO completion FIFO parity error", -1, 1 },
- { PIOREQPERR, "PCI PIO request FIFO parity error", -1, 1 },
- { TARTAGPERR, "PCI PCI target tag FIFO parity error", -1, 1 },
- { CCNTPERR, "PCI CMD channel count parity error", -1, 1 },
- { CREQPERR, "PCI CMD channel request parity error", -1, 1 },
- { CRSPPERR, "PCI CMD channel response parity error", -1, 1 },
- { DCNTPERR, "PCI DMA channel count parity error", -1, 1 },
- { DREQPERR, "PCI DMA channel request parity error", -1, 1 },
- { DRSPPERR, "PCI DMA channel response parity error", -1, 1 },
- { HCNTPERR, "PCI HMA channel count parity error", -1, 1 },
- { HREQPERR, "PCI HMA channel request parity error", -1, 1 },
- { HRSPPERR, "PCI HMA channel response parity error", -1, 1 },
- { CFGSNPPERR, "PCI config snoop FIFO parity error", -1, 1 },
- { FIDPERR, "PCI FID parity error", -1, 1 },
- { INTXCLRPERR, "PCI INTx clear parity error", -1, 1 },
- { MATAGPERR, "PCI MA tag parity error", -1, 1 },
- { PIOTAGPERR, "PCI PIO tag parity error", -1, 1 },
- { RXCPLPERR, "PCI Rx completion parity error", -1, 1 },
- { RXWRPERR, "PCI Rx write parity error", -1, 1 },
- { RPLPERR, "PCI replay buffer parity error", -1, 1 },
- { PCIESINT, "PCI core secondary fault", -1, 1 },
- { PCIEPINT, "PCI core primary fault", -1, 1 },
- { UNXSPLCPLERR, "PCI unexpected split completion error", -1,
- 0 },
- { 0, NULL, 0, 0 }
- };
-
- int fat;
-
- fat = csio_handle_intr_status(hw,
- PCIE_CORE_UTL_SYSTEM_BUS_AGENT_STATUS,
- sysbus_intr_info) +
- csio_handle_intr_status(hw,
- PCIE_CORE_UTL_PCI_EXPRESS_PORT_STATUS,
- pcie_port_intr_info) +
- csio_handle_intr_status(hw, PCIE_INT_CAUSE, pcie_intr_info);
- if (fat)
- csio_hw_fatal_err(hw);
-}
-
-/*
* TP interrupt handler.
*/
static void csio_tp_intr_handler(struct csio_hw *hw)
@@ -3517,7 +3190,7 @@
*/
static void csio_xgmac_intr_handler(struct csio_hw *hw, int port)
{
- uint32_t v = csio_rd_reg32(hw, PORT_REG(port, XGMAC_PORT_INT_CAUSE));
+ uint32_t v = csio_rd_reg32(hw, CSIO_MAC_INT_CAUSE_REG(hw, port));
v &= TXFIFO_PRTY_ERR | RXFIFO_PRTY_ERR;
if (!v)
@@ -3527,7 +3200,7 @@
csio_fatal(hw, "XGMAC %d Tx FIFO parity error\n", port);
if (v & RXFIFO_PRTY_ERR)
csio_fatal(hw, "XGMAC %d Rx FIFO parity error\n", port);
- csio_wr_reg32(hw, v, PORT_REG(port, XGMAC_PORT_INT_CAUSE));
+ csio_wr_reg32(hw, v, CSIO_MAC_INT_CAUSE_REG(hw, port));
csio_hw_fatal_err(hw);
}
@@ -3596,7 +3269,7 @@
csio_xgmac_intr_handler(hw, 3);
if (cause & PCIE)
- csio_pcie_intr_handler(hw);
+ hw->chip_ops->chip_pcie_intr_handler(hw);
if (cause & MC)
csio_mem_intr_handler(hw, MEM_MC);
@@ -4262,6 +3935,7 @@
&hw->params.pci.device_id);
csio_dev_id_cached(hw);
+ hw->chip_id = (hw->params.pci.device_id & CSIO_HW_CHIP_MASK);
} /* csio_hw_get_device_id */
@@ -4280,19 +3954,21 @@
prot_type = (dev_id & CSIO_ASIC_DEVID_PROTO_MASK);
adap_type = (dev_id & CSIO_ASIC_DEVID_TYPE_MASK);
- if (prot_type == CSIO_FPGA) {
- memcpy(hw->model_desc,
- csio_fcoe_adapters[13].description, 32);
- } else if (prot_type == CSIO_T4_FCOE_ASIC) {
+ if (prot_type == CSIO_T4_FCOE_ASIC) {
memcpy(hw->hw_ver,
- csio_fcoe_adapters[adap_type].model_no, 16);
+ csio_t4_fcoe_adapters[adap_type].model_no, 16);
memcpy(hw->model_desc,
- csio_fcoe_adapters[adap_type].description, 32);
+ csio_t4_fcoe_adapters[adap_type].description,
+ 32);
+ } else if (prot_type == CSIO_T5_FCOE_ASIC) {
+ memcpy(hw->hw_ver,
+ csio_t5_fcoe_adapters[adap_type].model_no, 16);
+ memcpy(hw->model_desc,
+ csio_t5_fcoe_adapters[adap_type].description,
+ 32);
} else {
char tempName[32] = "Chelsio FCoE Controller";
memcpy(hw->model_desc, tempName, 32);
-
- CSIO_DB_ASSERT(0);
}
}
} /* csio_hw_set_description */
@@ -4321,6 +3997,9 @@
strcpy(hw->name, CSIO_HW_NAME);
+ /* Initialize the HW chip ops with T4/T5 specific ops */
+ hw->chip_ops = csio_is_t4(hw->chip_id) ? &t4_ops : &t5_ops;
+
/* Set the model & its description */
ven_id = hw->params.pci.vendor_id;
diff --git a/drivers/scsi/csiostor/csio_hw.h b/drivers/scsi/csiostor/csio_hw.h
index 9edcca4..489fc09 100644
--- a/drivers/scsi/csiostor/csio_hw.h
+++ b/drivers/scsi/csiostor/csio_hw.h
@@ -48,6 +48,7 @@
#include <scsi/scsi_device.h>
#include <scsi/scsi_transport_fc.h>
+#include "csio_hw_chip.h"
#include "csio_wr.h"
#include "csio_mb.h"
#include "csio_scsi.h"
@@ -60,13 +61,6 @@
*/
#define FW_HOSTERROR 255
-#define CSIO_FW_FNAME "cxgb4/t4fw.bin"
-#define CSIO_CF_FNAME "cxgb4/t4-config.txt"
-
-#define FW_VERSION_MAJOR 1
-#define FW_VERSION_MINOR 2
-#define FW_VERSION_MICRO 8
-
#define CSIO_HW_NAME "Chelsio FCoE Adapter"
#define CSIO_MAX_PFN 8
#define CSIO_MAX_PPORTS 4
@@ -123,8 +117,6 @@
#define CSIO_VENDOR_ID 0x1425
#define CSIO_ASIC_DEVID_PROTO_MASK 0xFF00
#define CSIO_ASIC_DEVID_TYPE_MASK 0x00FF
-#define CSIO_FPGA 0xA000
-#define CSIO_T4_FCOE_ASIC 0x4600
#define CSIO_GLBL_INTR_MASK (CIM | MPS | PL | PCIE | MC | EDC0 | \
EDC1 | LE | TP | MA | PM_TX | PM_RX | \
@@ -207,17 +199,6 @@
SF_SIZE = SF_SEC_SIZE * 16, /* serial flash size */
};
-enum { MEM_EDC0, MEM_EDC1, MEM_MC };
-
-enum {
- MEMWIN0_APERTURE = 2048,
- MEMWIN0_BASE = 0x1b800,
- MEMWIN1_APERTURE = 32768,
- MEMWIN1_BASE = 0x28000,
- MEMWIN2_APERTURE = 65536,
- MEMWIN2_BASE = 0x30000,
-};
-
/* serial flash and firmware constants */
enum {
SF_ATTEMPTS = 10, /* max retries for SF operations */
@@ -239,9 +220,6 @@
FLASH_CFG_MAX_SIZE = 0x10000 , /* max size of the flash config file*/
FLASH_CFG_OFFSET = 0x1f0000,
FLASH_CFG_START_SEC = FLASH_CFG_OFFSET / SF_SEC_SIZE,
- FPGA_FLASH_CFG_OFFSET = 0xf0000 , /* if FPGA mode, then cfg file is
- * at 1MB - 64KB */
- FPGA_FLASH_CFG_START_SEC = FPGA_FLASH_CFG_OFFSET / SF_SEC_SIZE,
};
/*
@@ -259,6 +237,8 @@
FLASH_FW_START = FLASH_START(FLASH_FW_START_SEC),
FLASH_FW_MAX_SIZE = FLASH_MAX_SIZE(FLASH_FW_NSECS),
+ /* Location of Firmware Configuration File in FLASH. */
+ FLASH_CFG_START = FLASH_START(FLASH_CFG_START_SEC),
};
#undef FLASH_START
@@ -310,7 +290,7 @@
struct pci_params {
uint16_t vendor_id;
uint16_t device_id;
- uint32_t vpd_cap_addr;
+ int vpd_cap_addr;
uint16_t speed;
uint8_t width;
};
@@ -513,6 +493,7 @@
uint32_t fwrev;
uint32_t tp_vers;
char chip_ver;
+ uint16_t chip_id; /* Tells T4/T5 chip */
uint32_t cfg_finiver;
uint32_t cfg_finicsum;
uint32_t cfg_cfcsum;
@@ -556,6 +537,9 @@
*/
struct csio_fcoe_res_info fres_info; /* Fcoe resource info */
+ struct csio_hw_chip_ops *chip_ops; /* T4/T5 Chip specific
+ * Operations
+ */
/* MSIX vectors */
struct csio_msix_entries msix_entries[CSIO_MAX_MSIX_VECS];
@@ -636,9 +620,16 @@
#define csio_dbg(__hw, __fmt, ...)
#endif
+int csio_hw_wait_op_done_val(struct csio_hw *, int, uint32_t, int,
+ int, int, uint32_t *);
+void csio_hw_tp_wr_bits_indirect(struct csio_hw *, unsigned int,
+ unsigned int, unsigned int);
int csio_mgmt_req_lookup(struct csio_mgmtm *, struct csio_ioreq *);
void csio_hw_intr_disable(struct csio_hw *);
-int csio_hw_slow_intr_handler(struct csio_hw *hw);
+int csio_hw_slow_intr_handler(struct csio_hw *);
+int csio_handle_intr_status(struct csio_hw *, unsigned int,
+ const struct intr_info *);
+
int csio_hw_start(struct csio_hw *);
int csio_hw_stop(struct csio_hw *);
int csio_hw_reset(struct csio_hw *);
@@ -647,19 +638,17 @@
int csio_fwevtq_handler(struct csio_hw *);
void csio_evtq_worker(struct work_struct *);
-int csio_enqueue_evt(struct csio_hw *hw, enum csio_evt type,
- void *evt_msg, uint16_t len);
+int csio_enqueue_evt(struct csio_hw *, enum csio_evt, void *, uint16_t);
void csio_evtq_flush(struct csio_hw *hw);
int csio_request_irqs(struct csio_hw *);
void csio_intr_enable(struct csio_hw *);
void csio_intr_disable(struct csio_hw *, bool);
+void csio_hw_fatal_err(struct csio_hw *);
struct csio_lnode *csio_lnode_alloc(struct csio_hw *);
int csio_config_queues(struct csio_hw *);
-int csio_hw_mc_read(struct csio_hw *, uint32_t, __be32 *, uint64_t *);
-int csio_hw_edc_read(struct csio_hw *, int, uint32_t, __be32 *, uint64_t *);
int csio_hw_init(struct csio_hw *);
void csio_hw_exit(struct csio_hw *);
#endif /* ifndef __CSIO_HW_H__ */
diff --git a/drivers/scsi/csiostor/csio_hw_chip.h b/drivers/scsi/csiostor/csio_hw_chip.h
new file mode 100644
index 0000000..bca0de6
--- /dev/null
+++ b/drivers/scsi/csiostor/csio_hw_chip.h
@@ -0,0 +1,175 @@
+/*
+ * This file is part of the Chelsio FCoE driver for Linux.
+ *
+ * Copyright (c) 2008-2013 Chelsio Communications, Inc. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#ifndef __CSIO_HW_CHIP_H__
+#define __CSIO_HW_CHIP_H__
+
+#include "csio_defs.h"
+
+/* FCoE device IDs for T4 */
+#define CSIO_DEVID_T440DBG_FCOE 0x4600
+#define CSIO_DEVID_T420CR_FCOE 0x4601
+#define CSIO_DEVID_T422CR_FCOE 0x4602
+#define CSIO_DEVID_T440CR_FCOE 0x4603
+#define CSIO_DEVID_T420BCH_FCOE 0x4604
+#define CSIO_DEVID_T440BCH_FCOE 0x4605
+#define CSIO_DEVID_T440CH_FCOE 0x4606
+#define CSIO_DEVID_T420SO_FCOE 0x4607
+#define CSIO_DEVID_T420CX_FCOE 0x4608
+#define CSIO_DEVID_T420BT_FCOE 0x4609
+#define CSIO_DEVID_T404BT_FCOE 0x460A
+#define CSIO_DEVID_B420_FCOE 0x460B
+#define CSIO_DEVID_B404_FCOE 0x460C
+#define CSIO_DEVID_T480CR_FCOE 0x460D
+#define CSIO_DEVID_T440LPCR_FCOE 0x460E
+#define CSIO_DEVID_AMSTERDAM_T4_FCOE 0x460F
+#define CSIO_DEVID_HUAWEI_T480_FCOE 0x4680
+#define CSIO_DEVID_HUAWEI_T440_FCOE 0x4681
+#define CSIO_DEVID_HUAWEI_STG310_FCOE 0x4682
+#define CSIO_DEVID_ACROMAG_XMC_XAUI 0x4683
+#define CSIO_DEVID_ACROMAG_XMC_SFP_FCOE 0x4684
+#define CSIO_DEVID_QUANTA_MEZZ_SFP_FCOE 0x4685
+#define CSIO_DEVID_HUAWEI_10GT_FCOE 0x4686
+#define CSIO_DEVID_HUAWEI_T440_TOE_FCOE 0x4687
+
+/* FCoE device IDs for T5 */
+#define CSIO_DEVID_T580DBG_FCOE 0x5600
+#define CSIO_DEVID_T520CR_FCOE 0x5601
+#define CSIO_DEVID_T522CR_FCOE 0x5602
+#define CSIO_DEVID_T540CR_FCOE 0x5603
+#define CSIO_DEVID_T520BCH_FCOE 0x5604
+#define CSIO_DEVID_T540BCH_FCOE 0x5605
+#define CSIO_DEVID_T540CH_FCOE 0x5606
+#define CSIO_DEVID_T520SO_FCOE 0x5607
+#define CSIO_DEVID_T520CX_FCOE 0x5608
+#define CSIO_DEVID_T520BT_FCOE 0x5609
+#define CSIO_DEVID_T504BT_FCOE 0x560A
+#define CSIO_DEVID_B520_FCOE 0x560B
+#define CSIO_DEVID_B504_FCOE 0x560C
+#define CSIO_DEVID_T580CR2_FCOE 0x560D
+#define CSIO_DEVID_T540LPCR_FCOE 0x560E
+#define CSIO_DEVID_AMSTERDAM_T5_FCOE 0x560F
+#define CSIO_DEVID_T580LPCR_FCOE 0x5610
+#define CSIO_DEVID_T520LLCR_FCOE 0x5611
+#define CSIO_DEVID_T560CR_FCOE 0x5612
+#define CSIO_DEVID_T580CR_FCOE 0x5613
+
+/* Define MACRO values */
+#define CSIO_HW_T4 0x4000
+#define CSIO_T4_FCOE_ASIC 0x4600
+#define CSIO_HW_T5 0x5000
+#define CSIO_T5_FCOE_ASIC 0x5600
+#define CSIO_HW_CHIP_MASK 0xF000
+#define T4_REGMAP_SIZE (160 * 1024)
+#define T5_REGMAP_SIZE (332 * 1024)
+#define FW_FNAME_T4 "cxgb4/t4fw.bin"
+#define FW_FNAME_T5 "cxgb4/t5fw.bin"
+#define FW_CFG_NAME_T4 "cxgb4/t4-config.txt"
+#define FW_CFG_NAME_T5 "cxgb4/t5-config.txt"
+
+/* Define static functions */
+static inline int csio_is_t4(uint16_t chip)
+{
+ return (chip == CSIO_HW_T4);
+}
+
+static inline int csio_is_t5(uint16_t chip)
+{
+ return (chip == CSIO_HW_T5);
+}
+
+/* Define MACRO DEFINITIONS */
+#define CSIO_DEVICE(devid, idx) \
+ { PCI_VENDOR_ID_CHELSIO, (devid), PCI_ANY_ID, PCI_ANY_ID, 0, 0, (idx) }
+
+#define CSIO_HW_PIDX(hw, index) \
+ (csio_is_t4(hw->chip_id) ? (PIDX(index)) : \
+ (PIDX_T5(index) | DBTYPE(1U)))
+
+#define CSIO_HW_LP_INT_THRESH(hw, val) \
+ (csio_is_t4(hw->chip_id) ? (LP_INT_THRESH(val)) : \
+ (V_LP_INT_THRESH_T5(val)))
+
+#define CSIO_HW_M_LP_INT_THRESH(hw) \
+ (csio_is_t4(hw->chip_id) ? (LP_INT_THRESH_MASK) : (M_LP_INT_THRESH_T5))
+
+#define CSIO_MAC_INT_CAUSE_REG(hw, port) \
+ (csio_is_t4(hw->chip_id) ? (PORT_REG(port, XGMAC_PORT_INT_CAUSE)) : \
+ (T5_PORT_REG(port, MAC_PORT_INT_CAUSE)))
+
+#define FW_VERSION_MAJOR(hw) (csio_is_t4(hw->chip_id) ? 1 : 0)
+#define FW_VERSION_MINOR(hw) (csio_is_t4(hw->chip_id) ? 2 : 0)
+#define FW_VERSION_MICRO(hw) (csio_is_t4(hw->chip_id) ? 8 : 0)
+
+#define CSIO_FW_FNAME(hw) \
+ (csio_is_t4(hw->chip_id) ? FW_FNAME_T4 : FW_FNAME_T5)
+
+#define CSIO_CF_FNAME(hw) \
+ (csio_is_t4(hw->chip_id) ? FW_CFG_NAME_T4 : FW_CFG_NAME_T5)
+
+/* Declare ENUMS */
+enum { MEM_EDC0, MEM_EDC1, MEM_MC, MEM_MC0 = MEM_MC, MEM_MC1 };
+
+enum {
+ MEMWIN_APERTURE = 2048,
+ MEMWIN_BASE = 0x1b800,
+ MEMWIN_CSIOSTOR = 6, /* PCI-e Memory Window access */
+};
+
+/* Slow path handlers */
+struct intr_info {
+ unsigned int mask; /* bits to check in interrupt status */
+ const char *msg; /* message to print or NULL */
+ short stat_idx; /* stat counter to increment or -1 */
+ unsigned short fatal; /* whether the condition reported is fatal */
+};
+
+/* T4/T5 Chip specific ops */
+struct csio_hw;
+struct csio_hw_chip_ops {
+ int (*chip_set_mem_win)(struct csio_hw *, uint32_t);
+ void (*chip_pcie_intr_handler)(struct csio_hw *);
+ uint32_t (*chip_flash_cfg_addr)(struct csio_hw *);
+ int (*chip_mc_read)(struct csio_hw *, int, uint32_t,
+ __be32 *, uint64_t *);
+ int (*chip_edc_read)(struct csio_hw *, int, uint32_t,
+ __be32 *, uint64_t *);
+ int (*chip_memory_rw)(struct csio_hw *, u32, int, u32,
+ u32, uint32_t *, int);
+ void (*chip_dfs_create_ext_mem)(struct csio_hw *);
+};
+
+extern struct csio_hw_chip_ops t4_ops;
+extern struct csio_hw_chip_ops t5_ops;
+
+#endif /* #ifndef __CSIO_HW_CHIP_H__ */
diff --git a/drivers/scsi/csiostor/csio_hw_t4.c b/drivers/scsi/csiostor/csio_hw_t4.c
new file mode 100644
index 0000000..89ecbac
--- /dev/null
+++ b/drivers/scsi/csiostor/csio_hw_t4.c
@@ -0,0 +1,403 @@
+/*
+ * This file is part of the Chelsio FCoE driver for Linux.
+ *
+ * Copyright (c) 2008-2013 Chelsio Communications, Inc. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include "csio_hw.h"
+#include "csio_init.h"
+
+/*
+ * Return the specified PCI-E Configuration Space register from our Physical
+ * Function. We try first via a Firmware LDST Command since we prefer to let
+ * the firmware own all of these registers, but if that fails we go for it
+ * directly ourselves.
+ */
+static uint32_t
+csio_t4_read_pcie_cfg4(struct csio_hw *hw, int reg)
+{
+ u32 val = 0;
+ struct csio_mb *mbp;
+ int rv;
+ struct fw_ldst_cmd *ldst_cmd;
+
+ mbp = mempool_alloc(hw->mb_mempool, GFP_ATOMIC);
+ if (!mbp) {
+ CSIO_INC_STATS(hw, n_err_nomem);
+ pci_read_config_dword(hw->pdev, reg, &val);
+ return val;
+ }
+
+ csio_mb_ldst(hw, mbp, CSIO_MB_DEFAULT_TMO, reg);
+ rv = csio_mb_issue(hw, mbp);
+
+ /*
+ * If the LDST Command suucceeded, exctract the returned register
+ * value. Otherwise read it directly ourself.
+ */
+ if (rv == 0) {
+ ldst_cmd = (struct fw_ldst_cmd *)(mbp->mb);
+ val = ntohl(ldst_cmd->u.pcie.data[0]);
+ } else
+ pci_read_config_dword(hw->pdev, reg, &val);
+
+ mempool_free(mbp, hw->mb_mempool);
+
+ return val;
+}
+
+static int
+csio_t4_set_mem_win(struct csio_hw *hw, uint32_t win)
+{
+ u32 bar0;
+ u32 mem_win_base;
+
+ /*
+ * Truncation intentional: we only read the bottom 32-bits of the
+ * 64-bit BAR0/BAR1 ... We use the hardware backdoor mechanism to
+ * read BAR0 instead of using pci_resource_start() because we could be
+ * operating from within a Virtual Machine which is trapping our
+ * accesses to our Configuration Space and we need to set up the PCI-E
+ * Memory Window decoders with the actual addresses which will be
+ * coming across the PCI-E link.
+ */
+ bar0 = csio_t4_read_pcie_cfg4(hw, PCI_BASE_ADDRESS_0);
+ bar0 &= PCI_BASE_ADDRESS_MEM_MASK;
+
+ mem_win_base = bar0 + MEMWIN_BASE;
+
+ /*
+ * Set up memory window for accessing adapter memory ranges. (Read
+ * back MA register to ensure that changes propagate before we attempt
+ * to use the new values.)
+ */
+ csio_wr_reg32(hw, mem_win_base | BIR(0) |
+ WINDOW(ilog2(MEMWIN_APERTURE) - 10),
+ PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_BASE_WIN, win));
+ csio_rd_reg32(hw,
+ PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_BASE_WIN, win));
+ return 0;
+}
+
+/*
+ * Interrupt handler for the PCIE module.
+ */
+static void
+csio_t4_pcie_intr_handler(struct csio_hw *hw)
+{
+ static struct intr_info sysbus_intr_info[] = {
+ { RNPP, "RXNP array parity error", -1, 1 },
+ { RPCP, "RXPC array parity error", -1, 1 },
+ { RCIP, "RXCIF array parity error", -1, 1 },
+ { RCCP, "Rx completions control array parity error", -1, 1 },
+ { RFTP, "RXFT array parity error", -1, 1 },
+ { 0, NULL, 0, 0 }
+ };
+ static struct intr_info pcie_port_intr_info[] = {
+ { TPCP, "TXPC array parity error", -1, 1 },
+ { TNPP, "TXNP array parity error", -1, 1 },
+ { TFTP, "TXFT array parity error", -1, 1 },
+ { TCAP, "TXCA array parity error", -1, 1 },
+ { TCIP, "TXCIF array parity error", -1, 1 },
+ { RCAP, "RXCA array parity error", -1, 1 },
+ { OTDD, "outbound request TLP discarded", -1, 1 },
+ { RDPE, "Rx data parity error", -1, 1 },
+ { TDUE, "Tx uncorrectable data error", -1, 1 },
+ { 0, NULL, 0, 0 }
+ };
+
+ static struct intr_info pcie_intr_info[] = {
+ { MSIADDRLPERR, "MSI AddrL parity error", -1, 1 },
+ { MSIADDRHPERR, "MSI AddrH parity error", -1, 1 },
+ { MSIDATAPERR, "MSI data parity error", -1, 1 },
+ { MSIXADDRLPERR, "MSI-X AddrL parity error", -1, 1 },
+ { MSIXADDRHPERR, "MSI-X AddrH parity error", -1, 1 },
+ { MSIXDATAPERR, "MSI-X data parity error", -1, 1 },
+ { MSIXDIPERR, "MSI-X DI parity error", -1, 1 },
+ { PIOCPLPERR, "PCI PIO completion FIFO parity error", -1, 1 },
+ { PIOREQPERR, "PCI PIO request FIFO parity error", -1, 1 },
+ { TARTAGPERR, "PCI PCI target tag FIFO parity error", -1, 1 },
+ { CCNTPERR, "PCI CMD channel count parity error", -1, 1 },
+ { CREQPERR, "PCI CMD channel request parity error", -1, 1 },
+ { CRSPPERR, "PCI CMD channel response parity error", -1, 1 },
+ { DCNTPERR, "PCI DMA channel count parity error", -1, 1 },
+ { DREQPERR, "PCI DMA channel request parity error", -1, 1 },
+ { DRSPPERR, "PCI DMA channel response parity error", -1, 1 },
+ { HCNTPERR, "PCI HMA channel count parity error", -1, 1 },
+ { HREQPERR, "PCI HMA channel request parity error", -1, 1 },
+ { HRSPPERR, "PCI HMA channel response parity error", -1, 1 },
+ { CFGSNPPERR, "PCI config snoop FIFO parity error", -1, 1 },
+ { FIDPERR, "PCI FID parity error", -1, 1 },
+ { INTXCLRPERR, "PCI INTx clear parity error", -1, 1 },
+ { MATAGPERR, "PCI MA tag parity error", -1, 1 },
+ { PIOTAGPERR, "PCI PIO tag parity error", -1, 1 },
+ { RXCPLPERR, "PCI Rx completion parity error", -1, 1 },
+ { RXWRPERR, "PCI Rx write parity error", -1, 1 },
+ { RPLPERR, "PCI replay buffer parity error", -1, 1 },
+ { PCIESINT, "PCI core secondary fault", -1, 1 },
+ { PCIEPINT, "PCI core primary fault", -1, 1 },
+ { UNXSPLCPLERR, "PCI unexpected split completion error", -1,
+ 0 },
+ { 0, NULL, 0, 0 }
+ };
+
+ int fat;
+ fat = csio_handle_intr_status(hw,
+ PCIE_CORE_UTL_SYSTEM_BUS_AGENT_STATUS,
+ sysbus_intr_info) +
+ csio_handle_intr_status(hw,
+ PCIE_CORE_UTL_PCI_EXPRESS_PORT_STATUS,
+ pcie_port_intr_info) +
+ csio_handle_intr_status(hw, PCIE_INT_CAUSE, pcie_intr_info);
+ if (fat)
+ csio_hw_fatal_err(hw);
+}
+
+/*
+ * csio_t4_flash_cfg_addr - return the address of the flash configuration file
+ * @hw: the HW module
+ *
+ * Return the address within the flash where the Firmware Configuration
+ * File is stored.
+ */
+static unsigned int
+csio_t4_flash_cfg_addr(struct csio_hw *hw)
+{
+ return FLASH_CFG_OFFSET;
+}
+
+/*
+ * csio_t4_mc_read - read from MC through backdoor accesses
+ * @hw: the hw module
+ * @idx: not used for T4 adapter
+ * @addr: address of first byte requested
+ * @data: 64 bytes of data containing the requested address
+ * @ecc: where to store the corresponding 64-bit ECC word
+ *
+ * Read 64 bytes of data from MC starting at a 64-byte-aligned address
+ * that covers the requested address @addr. If @parity is not %NULL it
+ * is assigned the 64-bit ECC word for the read data.
+ */
+static int
+csio_t4_mc_read(struct csio_hw *hw, int idx, uint32_t addr, __be32 *data,
+ uint64_t *ecc)
+{
+ int i;
+
+ if (csio_rd_reg32(hw, MC_BIST_CMD) & START_BIST)
+ return -EBUSY;
+ csio_wr_reg32(hw, addr & ~0x3fU, MC_BIST_CMD_ADDR);
+ csio_wr_reg32(hw, 64, MC_BIST_CMD_LEN);
+ csio_wr_reg32(hw, 0xc, MC_BIST_DATA_PATTERN);
+ csio_wr_reg32(hw, BIST_OPCODE(1) | START_BIST | BIST_CMD_GAP(1),
+ MC_BIST_CMD);
+ i = csio_hw_wait_op_done_val(hw, MC_BIST_CMD, START_BIST,
+ 0, 10, 1, NULL);
+ if (i)
+ return i;
+
+#define MC_DATA(i) MC_BIST_STATUS_REG(MC_BIST_STATUS_RDATA, i)
+
+ for (i = 15; i >= 0; i--)
+ *data++ = htonl(csio_rd_reg32(hw, MC_DATA(i)));
+ if (ecc)
+ *ecc = csio_rd_reg64(hw, MC_DATA(16));
+#undef MC_DATA
+ return 0;
+}
+
+/*
+ * csio_t4_edc_read - read from EDC through backdoor accesses
+ * @hw: the hw module
+ * @idx: which EDC to access
+ * @addr: address of first byte requested
+ * @data: 64 bytes of data containing the requested address
+ * @ecc: where to store the corresponding 64-bit ECC word
+ *
+ * Read 64 bytes of data from EDC starting at a 64-byte-aligned address
+ * that covers the requested address @addr. If @parity is not %NULL it
+ * is assigned the 64-bit ECC word for the read data.
+ */
+static int
+csio_t4_edc_read(struct csio_hw *hw, int idx, uint32_t addr, __be32 *data,
+ uint64_t *ecc)
+{
+ int i;
+
+ idx *= EDC_STRIDE;
+ if (csio_rd_reg32(hw, EDC_BIST_CMD + idx) & START_BIST)
+ return -EBUSY;
+ csio_wr_reg32(hw, addr & ~0x3fU, EDC_BIST_CMD_ADDR + idx);
+ csio_wr_reg32(hw, 64, EDC_BIST_CMD_LEN + idx);
+ csio_wr_reg32(hw, 0xc, EDC_BIST_DATA_PATTERN + idx);
+ csio_wr_reg32(hw, BIST_OPCODE(1) | BIST_CMD_GAP(1) | START_BIST,
+ EDC_BIST_CMD + idx);
+ i = csio_hw_wait_op_done_val(hw, EDC_BIST_CMD + idx, START_BIST,
+ 0, 10, 1, NULL);
+ if (i)
+ return i;
+
+#define EDC_DATA(i) (EDC_BIST_STATUS_REG(EDC_BIST_STATUS_RDATA, i) + idx)
+
+ for (i = 15; i >= 0; i--)
+ *data++ = htonl(csio_rd_reg32(hw, EDC_DATA(i)));
+ if (ecc)
+ *ecc = csio_rd_reg64(hw, EDC_DATA(16));
+#undef EDC_DATA
+ return 0;
+}
+
+/*
+ * csio_t4_memory_rw - read/write EDC 0, EDC 1 or MC via PCIE memory window
+ * @hw: the csio_hw
+ * @win: PCI-E memory Window to use
+ * @mtype: memory type: MEM_EDC0, MEM_EDC1, MEM_MC0 (or MEM_MC) or MEM_MC1
+ * @addr: address within indicated memory type
+ * @len: amount of memory to transfer
+ * @buf: host memory buffer
+ * @dir: direction of transfer 1 => read, 0 => write
+ *
+ * Reads/writes an [almost] arbitrary memory region in the firmware: the
+ * firmware memory address, length and host buffer must be aligned on
+ * 32-bit boudaries. The memory is transferred as a raw byte sequence
+ * from/to the firmware's memory. If this memory contains data
+ * structures which contain multi-byte integers, it's the callers
+ * responsibility to perform appropriate byte order conversions.
+ */
+static int
+csio_t4_memory_rw(struct csio_hw *hw, u32 win, int mtype, u32 addr,
+ u32 len, uint32_t *buf, int dir)
+{
+ u32 pos, start, offset, memoffset, bar0;
+ u32 edc_size, mc_size, mem_reg, mem_aperture, mem_base;
+
+ /*
+ * Argument sanity checks ...
+ */
+ if ((addr & 0x3) || (len & 0x3))
+ return -EINVAL;
+
+ /* Offset into the region of memory which is being accessed
+ * MEM_EDC0 = 0
+ * MEM_EDC1 = 1
+ * MEM_MC = 2 -- T4
+ */
+ edc_size = EDRAM_SIZE_GET(csio_rd_reg32(hw, MA_EDRAM0_BAR));
+ if (mtype != MEM_MC1)
+ memoffset = (mtype * (edc_size * 1024 * 1024));
+ else {
+ mc_size = EXT_MEM_SIZE_GET(csio_rd_reg32(hw,
+ MA_EXT_MEMORY_BAR));
+ memoffset = (MEM_MC0 * edc_size + mc_size) * 1024 * 1024;
+ }
+
+ /* Determine the PCIE_MEM_ACCESS_OFFSET */
+ addr = addr + memoffset;
+
+ /*
+ * Each PCI-E Memory Window is programmed with a window size -- or
+ * "aperture" -- which controls the granularity of its mapping onto
+ * adapter memory. We need to grab that aperture in order to know
+ * how to use the specified window. The window is also programmed
+ * with the base address of the Memory Window in BAR0's address
+ * space. For T4 this is an absolute PCI-E Bus Address. For T5
+ * the address is relative to BAR0.
+ */
+ mem_reg = csio_rd_reg32(hw,
+ PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_BASE_WIN, win));
+ mem_aperture = 1 << (WINDOW(mem_reg) + 10);
+ mem_base = GET_PCIEOFST(mem_reg) << 10;
+
+ bar0 = csio_t4_read_pcie_cfg4(hw, PCI_BASE_ADDRESS_0);
+ bar0 &= PCI_BASE_ADDRESS_MEM_MASK;
+ mem_base -= bar0;
+
+ start = addr & ~(mem_aperture-1);
+ offset = addr - start;
+
+ csio_dbg(hw, "csio_t4_memory_rw: mem_reg: 0x%x, mem_aperture: 0x%x\n",
+ mem_reg, mem_aperture);
+ csio_dbg(hw, "csio_t4_memory_rw: mem_base: 0x%x, mem_offset: 0x%x\n",
+ mem_base, memoffset);
+ csio_dbg(hw, "csio_t4_memory_rw: bar0: 0x%x, start:0x%x, offset:0x%x\n",
+ bar0, start, offset);
+ csio_dbg(hw, "csio_t4_memory_rw: mtype: %d, addr: 0x%x, len: %d\n",
+ mtype, addr, len);
+
+ for (pos = start; len > 0; pos += mem_aperture, offset = 0) {
+ /*
+ * Move PCI-E Memory Window to our current transfer
+ * position. Read it back to ensure that changes propagate
+ * before we attempt to use the new value.
+ */
+ csio_wr_reg32(hw, pos,
+ PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_OFFSET, win));
+ csio_rd_reg32(hw,
+ PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_OFFSET, win));
+
+ while (offset < mem_aperture && len > 0) {
+ if (dir)
+ *buf++ = csio_rd_reg32(hw, mem_base + offset);
+ else
+ csio_wr_reg32(hw, *buf++, mem_base + offset);
+
+ offset += sizeof(__be32);
+ len -= sizeof(__be32);
+ }
+ }
+ return 0;
+}
+
+/*
+ * csio_t4_dfs_create_ext_mem - setup debugfs for MC to read the values
+ * @hw: the csio_hw
+ *
+ * This function creates files in the debugfs with external memory region MC.
+ */
+static void
+csio_t4_dfs_create_ext_mem(struct csio_hw *hw)
+{
+ u32 size;
+ int i = csio_rd_reg32(hw, MA_TARGET_MEM_ENABLE);
+ if (i & EXT_MEM_ENABLE) {
+ size = csio_rd_reg32(hw, MA_EXT_MEMORY_BAR);
+ csio_add_debugfs_mem(hw, "mc", MEM_MC,
+ EXT_MEM_SIZE_GET(size));
+ }
+}
+
+/* T4 adapter specific function */
+struct csio_hw_chip_ops t4_ops = {
+ .chip_set_mem_win = csio_t4_set_mem_win,
+ .chip_pcie_intr_handler = csio_t4_pcie_intr_handler,
+ .chip_flash_cfg_addr = csio_t4_flash_cfg_addr,
+ .chip_mc_read = csio_t4_mc_read,
+ .chip_edc_read = csio_t4_edc_read,
+ .chip_memory_rw = csio_t4_memory_rw,
+ .chip_dfs_create_ext_mem = csio_t4_dfs_create_ext_mem,
+};
diff --git a/drivers/scsi/csiostor/csio_hw_t5.c b/drivers/scsi/csiostor/csio_hw_t5.c
new file mode 100644
index 0000000..27745c1
--- /dev/null
+++ b/drivers/scsi/csiostor/csio_hw_t5.c
@@ -0,0 +1,397 @@
+/*
+ * This file is part of the Chelsio FCoE driver for Linux.
+ *
+ * Copyright (c) 2008-2013 Chelsio Communications, Inc. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include "csio_hw.h"
+#include "csio_init.h"
+
+static int
+csio_t5_set_mem_win(struct csio_hw *hw, uint32_t win)
+{
+ u32 mem_win_base;
+ /*
+ * Truncation intentional: we only read the bottom 32-bits of the
+ * 64-bit BAR0/BAR1 ... We use the hardware backdoor mechanism to
+ * read BAR0 instead of using pci_resource_start() because we could be
+ * operating from within a Virtual Machine which is trapping our
+ * accesses to our Configuration Space and we need to set up the PCI-E
+ * Memory Window decoders with the actual addresses which will be
+ * coming across the PCI-E link.
+ */
+
+ /* For T5, only relative offset inside the PCIe BAR is passed */
+ mem_win_base = MEMWIN_BASE;
+
+ /*
+ * Set up memory window for accessing adapter memory ranges. (Read
+ * back MA register to ensure that changes propagate before we attempt
+ * to use the new values.)
+ */
+ csio_wr_reg32(hw, mem_win_base | BIR(0) |
+ WINDOW(ilog2(MEMWIN_APERTURE) - 10),
+ PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_BASE_WIN, win));
+ csio_rd_reg32(hw,
+ PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_BASE_WIN, win));
+
+ return 0;
+}
+
+/*
+ * Interrupt handler for the PCIE module.
+ */
+static void
+csio_t5_pcie_intr_handler(struct csio_hw *hw)
+{
+ static struct intr_info sysbus_intr_info[] = {
+ { RNPP, "RXNP array parity error", -1, 1 },
+ { RPCP, "RXPC array parity error", -1, 1 },
+ { RCIP, "RXCIF array parity error", -1, 1 },
+ { RCCP, "Rx completions control array parity error", -1, 1 },
+ { RFTP, "RXFT array parity error", -1, 1 },
+ { 0, NULL, 0, 0 }
+ };
+ static struct intr_info pcie_port_intr_info[] = {
+ { TPCP, "TXPC array parity error", -1, 1 },
+ { TNPP, "TXNP array parity error", -1, 1 },
+ { TFTP, "TXFT array parity error", -1, 1 },
+ { TCAP, "TXCA array parity error", -1, 1 },
+ { TCIP, "TXCIF array parity error", -1, 1 },
+ { RCAP, "RXCA array parity error", -1, 1 },
+ { OTDD, "outbound request TLP discarded", -1, 1 },
+ { RDPE, "Rx data parity error", -1, 1 },
+ { TDUE, "Tx uncorrectable data error", -1, 1 },
+ { 0, NULL, 0, 0 }
+ };
+
+ static struct intr_info pcie_intr_info[] = {
+ { MSTGRPPERR, "Master Response Read Queue parity error",
+ -1, 1 },
+ { MSTTIMEOUTPERR, "Master Timeout FIFO parity error", -1, 1 },
+ { MSIXSTIPERR, "MSI-X STI SRAM parity error", -1, 1 },
+ { MSIXADDRLPERR, "MSI-X AddrL parity error", -1, 1 },
+ { MSIXADDRHPERR, "MSI-X AddrH parity error", -1, 1 },
+ { MSIXDATAPERR, "MSI-X data parity error", -1, 1 },
+ { MSIXDIPERR, "MSI-X DI parity error", -1, 1 },
+ { PIOCPLGRPPERR, "PCI PIO completion Group FIFO parity error",
+ -1, 1 },
+ { PIOREQGRPPERR, "PCI PIO request Group FIFO parity error",
+ -1, 1 },
+ { TARTAGPERR, "PCI PCI target tag FIFO parity error", -1, 1 },
+ { MSTTAGQPERR, "PCI master tag queue parity error", -1, 1 },
+ { CREQPERR, "PCI CMD channel request parity error", -1, 1 },
+ { CRSPPERR, "PCI CMD channel response parity error", -1, 1 },
+ { DREQWRPERR, "PCI DMA channel write request parity error",
+ -1, 1 },
+ { DREQPERR, "PCI DMA channel request parity error", -1, 1 },
+ { DRSPPERR, "PCI DMA channel response parity error", -1, 1 },
+ { HREQWRPERR, "PCI HMA channel count parity error", -1, 1 },
+ { HREQPERR, "PCI HMA channel request parity error", -1, 1 },
+ { HRSPPERR, "PCI HMA channel response parity error", -1, 1 },
+ { CFGSNPPERR, "PCI config snoop FIFO parity error", -1, 1 },
+ { FIDPERR, "PCI FID parity error", -1, 1 },
+ { VFIDPERR, "PCI INTx clear parity error", -1, 1 },
+ { MAGRPPERR, "PCI MA group FIFO parity error", -1, 1 },
+ { PIOTAGPERR, "PCI PIO tag parity error", -1, 1 },
+ { IPRXHDRGRPPERR, "PCI IP Rx header group parity error",
+ -1, 1 },
+ { IPRXDATAGRPPERR, "PCI IP Rx data group parity error",
+ -1, 1 },
+ { RPLPERR, "PCI IP replay buffer parity error", -1, 1 },
+ { IPSOTPERR, "PCI IP SOT buffer parity error", -1, 1 },
+ { TRGT1GRPPERR, "PCI TRGT1 group FIFOs parity error", -1, 1 },
+ { READRSPERR, "Outbound read error", -1, 0 },
+ { 0, NULL, 0, 0 }
+ };
+
+ int fat;
+ fat = csio_handle_intr_status(hw,
+ PCIE_CORE_UTL_SYSTEM_BUS_AGENT_STATUS,
+ sysbus_intr_info) +
+ csio_handle_intr_status(hw,
+ PCIE_CORE_UTL_PCI_EXPRESS_PORT_STATUS,
+ pcie_port_intr_info) +
+ csio_handle_intr_status(hw, PCIE_INT_CAUSE, pcie_intr_info);
+ if (fat)
+ csio_hw_fatal_err(hw);
+}
+
+/*
+ * csio_t5_flash_cfg_addr - return the address of the flash configuration file
+ * @hw: the HW module
+ *
+ * Return the address within the flash where the Firmware Configuration
+ * File is stored.
+ */
+static unsigned int
+csio_t5_flash_cfg_addr(struct csio_hw *hw)
+{
+ return FLASH_CFG_START;
+}
+
+/*
+ * csio_t5_mc_read - read from MC through backdoor accesses
+ * @hw: the hw module
+ * @idx: index to the register
+ * @addr: address of first byte requested
+ * @data: 64 bytes of data containing the requested address
+ * @ecc: where to store the corresponding 64-bit ECC word
+ *
+ * Read 64 bytes of data from MC starting at a 64-byte-aligned address
+ * that covers the requested address @addr. If @parity is not %NULL it
+ * is assigned the 64-bit ECC word for the read data.
+ */
+static int
+csio_t5_mc_read(struct csio_hw *hw, int idx, uint32_t addr, __be32 *data,
+ uint64_t *ecc)
+{
+ int i;
+ uint32_t mc_bist_cmd_reg, mc_bist_cmd_addr_reg, mc_bist_cmd_len_reg;
+ uint32_t mc_bist_status_rdata_reg, mc_bist_data_pattern_reg;
+
+ mc_bist_cmd_reg = MC_REG(MC_P_BIST_CMD, idx);
+ mc_bist_cmd_addr_reg = MC_REG(MC_P_BIST_CMD_ADDR, idx);
+ mc_bist_cmd_len_reg = MC_REG(MC_P_BIST_CMD_LEN, idx);
+ mc_bist_status_rdata_reg = MC_REG(MC_P_BIST_STATUS_RDATA, idx);
+ mc_bist_data_pattern_reg = MC_REG(MC_P_BIST_DATA_PATTERN, idx);
+
+ if (csio_rd_reg32(hw, mc_bist_cmd_reg) & START_BIST)
+ return -EBUSY;
+ csio_wr_reg32(hw, addr & ~0x3fU, mc_bist_cmd_addr_reg);
+ csio_wr_reg32(hw, 64, mc_bist_cmd_len_reg);
+ csio_wr_reg32(hw, 0xc, mc_bist_data_pattern_reg);
+ csio_wr_reg32(hw, BIST_OPCODE(1) | START_BIST | BIST_CMD_GAP(1),
+ mc_bist_cmd_reg);
+ i = csio_hw_wait_op_done_val(hw, mc_bist_cmd_reg, START_BIST,
+ 0, 10, 1, NULL);
+ if (i)
+ return i;
+
+#define MC_DATA(i) MC_BIST_STATUS_REG(MC_BIST_STATUS_RDATA, i)
+
+ for (i = 15; i >= 0; i--)
+ *data++ = htonl(csio_rd_reg32(hw, MC_DATA(i)));
+ if (ecc)
+ *ecc = csio_rd_reg64(hw, MC_DATA(16));
+#undef MC_DATA
+ return 0;
+}
+
+/*
+ * csio_t5_edc_read - read from EDC through backdoor accesses
+ * @hw: the hw module
+ * @idx: which EDC to access
+ * @addr: address of first byte requested
+ * @data: 64 bytes of data containing the requested address
+ * @ecc: where to store the corresponding 64-bit ECC word
+ *
+ * Read 64 bytes of data from EDC starting at a 64-byte-aligned address
+ * that covers the requested address @addr. If @parity is not %NULL it
+ * is assigned the 64-bit ECC word for the read data.
+ */
+static int
+csio_t5_edc_read(struct csio_hw *hw, int idx, uint32_t addr, __be32 *data,
+ uint64_t *ecc)
+{
+ int i;
+ uint32_t edc_bist_cmd_reg, edc_bist_cmd_addr_reg, edc_bist_cmd_len_reg;
+ uint32_t edc_bist_cmd_data_pattern, edc_bist_status_rdata_reg;
+
+/*
+ * These macro are missing in t4_regs.h file.
+ */
+#define EDC_STRIDE_T5 (EDC_T51_BASE_ADDR - EDC_T50_BASE_ADDR)
+#define EDC_REG_T5(reg, idx) (reg + EDC_STRIDE_T5 * idx)
+
+ edc_bist_cmd_reg = EDC_REG_T5(EDC_H_BIST_CMD, idx);
+ edc_bist_cmd_addr_reg = EDC_REG_T5(EDC_H_BIST_CMD_ADDR, idx);
+ edc_bist_cmd_len_reg = EDC_REG_T5(EDC_H_BIST_CMD_LEN, idx);
+ edc_bist_cmd_data_pattern = EDC_REG_T5(EDC_H_BIST_DATA_PATTERN, idx);
+ edc_bist_status_rdata_reg = EDC_REG_T5(EDC_H_BIST_STATUS_RDATA, idx);
+#undef EDC_REG_T5
+#undef EDC_STRIDE_T5
+
+ if (csio_rd_reg32(hw, edc_bist_cmd_reg) & START_BIST)
+ return -EBUSY;
+ csio_wr_reg32(hw, addr & ~0x3fU, edc_bist_cmd_addr_reg);
+ csio_wr_reg32(hw, 64, edc_bist_cmd_len_reg);
+ csio_wr_reg32(hw, 0xc, edc_bist_cmd_data_pattern);
+ csio_wr_reg32(hw, BIST_OPCODE(1) | START_BIST | BIST_CMD_GAP(1),
+ edc_bist_cmd_reg);
+ i = csio_hw_wait_op_done_val(hw, edc_bist_cmd_reg, START_BIST,
+ 0, 10, 1, NULL);
+ if (i)
+ return i;
+
+#define EDC_DATA(i) (EDC_BIST_STATUS_REG(EDC_BIST_STATUS_RDATA, i) + idx)
+
+ for (i = 15; i >= 0; i--)
+ *data++ = htonl(csio_rd_reg32(hw, EDC_DATA(i)));
+ if (ecc)
+ *ecc = csio_rd_reg64(hw, EDC_DATA(16));
+#undef EDC_DATA
+ return 0;
+}
+
+/*
+ * csio_t5_memory_rw - read/write EDC 0, EDC 1 or MC via PCIE memory window
+ * @hw: the csio_hw
+ * @win: PCI-E memory Window to use
+ * @mtype: memory type: MEM_EDC0, MEM_EDC1, MEM_MC0 (or MEM_MC) or MEM_MC1
+ * @addr: address within indicated memory type
+ * @len: amount of memory to transfer
+ * @buf: host memory buffer
+ * @dir: direction of transfer 1 => read, 0 => write
+ *
+ * Reads/writes an [almost] arbitrary memory region in the firmware: the
+ * firmware memory address, length and host buffer must be aligned on
+ * 32-bit boudaries. The memory is transferred as a raw byte sequence
+ * from/to the firmware's memory. If this memory contains data
+ * structures which contain multi-byte integers, it's the callers
+ * responsibility to perform appropriate byte order conversions.
+ */
+static int
+csio_t5_memory_rw(struct csio_hw *hw, u32 win, int mtype, u32 addr,
+ u32 len, uint32_t *buf, int dir)
+{
+ u32 pos, start, offset, memoffset;
+ u32 edc_size, mc_size, win_pf, mem_reg, mem_aperture, mem_base;
+
+ /*
+ * Argument sanity checks ...
+ */
+ if ((addr & 0x3) || (len & 0x3))
+ return -EINVAL;
+
+ /* Offset into the region of memory which is being accessed
+ * MEM_EDC0 = 0
+ * MEM_EDC1 = 1
+ * MEM_MC = 2 -- T4
+ * MEM_MC0 = 2 -- For T5
+ * MEM_MC1 = 3 -- For T5
+ */
+ edc_size = EDRAM_SIZE_GET(csio_rd_reg32(hw, MA_EDRAM0_BAR));
+ if (mtype != MEM_MC1)
+ memoffset = (mtype * (edc_size * 1024 * 1024));
+ else {
+ mc_size = EXT_MEM_SIZE_GET(csio_rd_reg32(hw,
+ MA_EXT_MEMORY_BAR));
+ memoffset = (MEM_MC0 * edc_size + mc_size) * 1024 * 1024;
+ }
+
+ /* Determine the PCIE_MEM_ACCESS_OFFSET */
+ addr = addr + memoffset;
+
+ /*
+ * Each PCI-E Memory Window is programmed with a window size -- or
+ * "aperture" -- which controls the granularity of its mapping onto
+ * adapter memory. We need to grab that aperture in order to know
+ * how to use the specified window. The window is also programmed
+ * with the base address of the Memory Window in BAR0's address
+ * space. For T4 this is an absolute PCI-E Bus Address. For T5
+ * the address is relative to BAR0.
+ */
+ mem_reg = csio_rd_reg32(hw,
+ PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_BASE_WIN, win));
+ mem_aperture = 1 << (WINDOW(mem_reg) + 10);
+ mem_base = GET_PCIEOFST(mem_reg) << 10;
+
+ start = addr & ~(mem_aperture-1);
+ offset = addr - start;
+ win_pf = V_PFNUM(hw->pfn);
+
+ csio_dbg(hw, "csio_t5_memory_rw: mem_reg: 0x%x, mem_aperture: 0x%x\n",
+ mem_reg, mem_aperture);
+ csio_dbg(hw, "csio_t5_memory_rw: mem_base: 0x%x, mem_offset: 0x%x\n",
+ mem_base, memoffset);
+ csio_dbg(hw, "csio_t5_memory_rw: start:0x%x, offset:0x%x, win_pf:%d\n",
+ start, offset, win_pf);
+ csio_dbg(hw, "csio_t5_memory_rw: mtype: %d, addr: 0x%x, len: %d\n",
+ mtype, addr, len);
+
+ for (pos = start; len > 0; pos += mem_aperture, offset = 0) {
+ /*
+ * Move PCI-E Memory Window to our current transfer
+ * position. Read it back to ensure that changes propagate
+ * before we attempt to use the new value.
+ */
+ csio_wr_reg32(hw, pos | win_pf,
+ PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_OFFSET, win));
+ csio_rd_reg32(hw,
+ PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_OFFSET, win));
+
+ while (offset < mem_aperture && len > 0) {
+ if (dir)
+ *buf++ = csio_rd_reg32(hw, mem_base + offset);
+ else
+ csio_wr_reg32(hw, *buf++, mem_base + offset);
+
+ offset += sizeof(__be32);
+ len -= sizeof(__be32);
+ }
+ }
+ return 0;
+}
+
+/*
+ * csio_t5_dfs_create_ext_mem - setup debugfs for MC0 or MC1 to read the values
+ * @hw: the csio_hw
+ *
+ * This function creates files in the debugfs with external memory region
+ * MC0 & MC1.
+ */
+static void
+csio_t5_dfs_create_ext_mem(struct csio_hw *hw)
+{
+ u32 size;
+ int i = csio_rd_reg32(hw, MA_TARGET_MEM_ENABLE);
+ if (i & EXT_MEM_ENABLE) {
+ size = csio_rd_reg32(hw, MA_EXT_MEMORY_BAR);
+ csio_add_debugfs_mem(hw, "mc0", MEM_MC0,
+ EXT_MEM_SIZE_GET(size));
+ }
+ if (i & EXT_MEM1_ENABLE) {
+ size = csio_rd_reg32(hw, MA_EXT_MEMORY1_BAR);
+ csio_add_debugfs_mem(hw, "mc1", MEM_MC1,
+ EXT_MEM_SIZE_GET(size));
+ }
+}
+
+/* T5 adapter specific function */
+struct csio_hw_chip_ops t5_ops = {
+ .chip_set_mem_win = csio_t5_set_mem_win,
+ .chip_pcie_intr_handler = csio_t5_pcie_intr_handler,
+ .chip_flash_cfg_addr = csio_t5_flash_cfg_addr,
+ .chip_mc_read = csio_t5_mc_read,
+ .chip_edc_read = csio_t5_edc_read,
+ .chip_memory_rw = csio_t5_memory_rw,
+ .chip_dfs_create_ext_mem = csio_t5_dfs_create_ext_mem,
+};
diff --git a/drivers/scsi/csiostor/csio_init.c b/drivers/scsi/csiostor/csio_init.c
index 0604b5f..00346fe 100644
--- a/drivers/scsi/csiostor/csio_init.c
+++ b/drivers/scsi/csiostor/csio_init.c
@@ -81,9 +81,11 @@
__be32 data[16];
if (mem == MEM_MC)
- ret = csio_hw_mc_read(hw, pos, data, NULL);
+ ret = hw->chip_ops->chip_mc_read(hw, 0, pos,
+ data, NULL);
else
- ret = csio_hw_edc_read(hw, mem, pos, data, NULL);
+ ret = hw->chip_ops->chip_edc_read(hw, mem, pos,
+ data, NULL);
if (ret)
return ret;
@@ -108,7 +110,7 @@
.llseek = default_llseek,
};
-static void csio_add_debugfs_mem(struct csio_hw *hw, const char *name,
+void csio_add_debugfs_mem(struct csio_hw *hw, const char *name,
unsigned int idx, unsigned int size_mb)
{
struct dentry *de;
@@ -131,9 +133,8 @@
csio_add_debugfs_mem(hw, "edc0", MEM_EDC0, 5);
if (i & EDRAM1_ENABLE)
csio_add_debugfs_mem(hw, "edc1", MEM_EDC1, 5);
- if (i & EXT_MEM_ENABLE)
- csio_add_debugfs_mem(hw, "mc", MEM_MC,
- EXT_MEM_SIZE_GET(csio_rd_reg32(hw, MA_EXT_MEMORY_BAR)));
+
+ hw->chip_ops->chip_dfs_create_ext_mem(hw);
return 0;
}
@@ -1169,7 +1170,7 @@
};
static DEFINE_PCI_DEVICE_TABLE(csio_pci_tbl) = {
- CSIO_DEVICE(CSIO_DEVID_T440DBG_FCOE, 0), /* T440DBG FCOE */
+ CSIO_DEVICE(CSIO_DEVID_T440DBG_FCOE, 0), /* T4 DEBUG FCOE */
CSIO_DEVICE(CSIO_DEVID_T420CR_FCOE, 0), /* T420CR FCOE */
CSIO_DEVICE(CSIO_DEVID_T422CR_FCOE, 0), /* T422CR FCOE */
CSIO_DEVICE(CSIO_DEVID_T440CR_FCOE, 0), /* T440CR FCOE */
@@ -1184,8 +1185,34 @@
CSIO_DEVICE(CSIO_DEVID_B404_FCOE, 0), /* B404 FCOE */
CSIO_DEVICE(CSIO_DEVID_T480CR_FCOE, 0), /* T480 CR FCOE */
CSIO_DEVICE(CSIO_DEVID_T440LPCR_FCOE, 0), /* T440 LP-CR FCOE */
- CSIO_DEVICE(CSIO_DEVID_PE10K, 0), /* PE10K FCOE */
- CSIO_DEVICE(CSIO_DEVID_PE10K_PF1, 0), /* PE10K FCOE on PF1 */
+ CSIO_DEVICE(CSIO_DEVID_AMSTERDAM_T4_FCOE, 0), /* AMSTERDAM T4 FCOE */
+ CSIO_DEVICE(CSIO_DEVID_HUAWEI_T480_FCOE, 0), /* HUAWEI T480 FCOE */
+ CSIO_DEVICE(CSIO_DEVID_HUAWEI_T440_FCOE, 0), /* HUAWEI T440 FCOE */
+ CSIO_DEVICE(CSIO_DEVID_HUAWEI_STG310_FCOE, 0), /* HUAWEI STG FCOE */
+ CSIO_DEVICE(CSIO_DEVID_ACROMAG_XMC_XAUI, 0), /* ACROMAG XAUI FCOE */
+ CSIO_DEVICE(CSIO_DEVID_QUANTA_MEZZ_SFP_FCOE, 0),/* QUANTA MEZZ FCOE */
+ CSIO_DEVICE(CSIO_DEVID_HUAWEI_10GT_FCOE, 0), /* HUAWEI 10GT FCOE */
+ CSIO_DEVICE(CSIO_DEVID_HUAWEI_T440_TOE_FCOE, 0),/* HUAWEI T4 TOE FCOE */
+ CSIO_DEVICE(CSIO_DEVID_T580DBG_FCOE, 0), /* T5 DEBUG FCOE */
+ CSIO_DEVICE(CSIO_DEVID_T520CR_FCOE, 0), /* T520CR FCOE */
+ CSIO_DEVICE(CSIO_DEVID_T522CR_FCOE, 0), /* T522CR FCOE */
+ CSIO_DEVICE(CSIO_DEVID_T540CR_FCOE, 0), /* T540CR FCOE */
+ CSIO_DEVICE(CSIO_DEVID_T520BCH_FCOE, 0), /* T520BCH FCOE */
+ CSIO_DEVICE(CSIO_DEVID_T540BCH_FCOE, 0), /* T540BCH FCOE */
+ CSIO_DEVICE(CSIO_DEVID_T540CH_FCOE, 0), /* T540CH FCOE */
+ CSIO_DEVICE(CSIO_DEVID_T520SO_FCOE, 0), /* T520SO FCOE */
+ CSIO_DEVICE(CSIO_DEVID_T520CX_FCOE, 0), /* T520CX FCOE */
+ CSIO_DEVICE(CSIO_DEVID_T520BT_FCOE, 0), /* T520BT FCOE */
+ CSIO_DEVICE(CSIO_DEVID_T504BT_FCOE, 0), /* T504BT FCOE */
+ CSIO_DEVICE(CSIO_DEVID_B520_FCOE, 0), /* B520 FCOE */
+ CSIO_DEVICE(CSIO_DEVID_B504_FCOE, 0), /* B504 FCOE */
+ CSIO_DEVICE(CSIO_DEVID_T580CR2_FCOE, 0), /* T580 CR FCOE */
+ CSIO_DEVICE(CSIO_DEVID_T540LPCR_FCOE, 0), /* T540 LP-CR FCOE */
+ CSIO_DEVICE(CSIO_DEVID_AMSTERDAM_T5_FCOE, 0), /* AMSTERDAM T5 FCOE */
+ CSIO_DEVICE(CSIO_DEVID_T580LPCR_FCOE, 0), /* T580 LP-CR FCOE */
+ CSIO_DEVICE(CSIO_DEVID_T520LLCR_FCOE, 0), /* T520 LL-CR FCOE */
+ CSIO_DEVICE(CSIO_DEVID_T560CR_FCOE, 0), /* T560 CR FCOE */
+ CSIO_DEVICE(CSIO_DEVID_T580CR_FCOE, 0), /* T580 CR FCOE */
{ 0, 0, 0, 0, 0, 0, 0 }
};
@@ -1259,4 +1286,5 @@
MODULE_LICENSE(CSIO_DRV_LICENSE);
MODULE_DEVICE_TABLE(pci, csio_pci_tbl);
MODULE_VERSION(CSIO_DRV_VERSION);
-MODULE_FIRMWARE(CSIO_FW_FNAME);
+MODULE_FIRMWARE(FW_FNAME_T4);
+MODULE_FIRMWARE(FW_FNAME_T5);
diff --git a/drivers/scsi/csiostor/csio_init.h b/drivers/scsi/csiostor/csio_init.h
index 0838fd7..5cc5d31 100644
--- a/drivers/scsi/csiostor/csio_init.h
+++ b/drivers/scsi/csiostor/csio_init.h
@@ -52,31 +52,6 @@
#define CSIO_DRV_DESC "Chelsio FCoE driver"
#define CSIO_DRV_VERSION "1.0.0"
-#define CSIO_DEVICE(devid, idx) \
-{ PCI_VENDOR_ID_CHELSIO, (devid), PCI_ANY_ID, PCI_ANY_ID, 0, 0, (idx) }
-
-#define CSIO_IS_T4_FPGA(_dev) (((_dev) == CSIO_DEVID_PE10K) ||\
- ((_dev) == CSIO_DEVID_PE10K_PF1))
-
-/* FCoE device IDs */
-#define CSIO_DEVID_PE10K 0xA000
-#define CSIO_DEVID_PE10K_PF1 0xA001
-#define CSIO_DEVID_T440DBG_FCOE 0x4600
-#define CSIO_DEVID_T420CR_FCOE 0x4601
-#define CSIO_DEVID_T422CR_FCOE 0x4602
-#define CSIO_DEVID_T440CR_FCOE 0x4603
-#define CSIO_DEVID_T420BCH_FCOE 0x4604
-#define CSIO_DEVID_T440BCH_FCOE 0x4605
-#define CSIO_DEVID_T440CH_FCOE 0x4606
-#define CSIO_DEVID_T420SO_FCOE 0x4607
-#define CSIO_DEVID_T420CX_FCOE 0x4608
-#define CSIO_DEVID_T420BT_FCOE 0x4609
-#define CSIO_DEVID_T404BT_FCOE 0x460A
-#define CSIO_DEVID_B420_FCOE 0x460B
-#define CSIO_DEVID_B404_FCOE 0x460C
-#define CSIO_DEVID_T480CR_FCOE 0x460D
-#define CSIO_DEVID_T440LPCR_FCOE 0x460E
-
extern struct fc_function_template csio_fc_transport_funcs;
extern struct fc_function_template csio_fc_transport_vport_funcs;
@@ -100,6 +75,10 @@
void csio_shost_exit(struct csio_lnode *);
void csio_lnodes_exit(struct csio_hw *, bool);
+/* DebugFS helper routines */
+void csio_add_debugfs_mem(struct csio_hw *, const char *,
+ unsigned int, unsigned int);
+
static inline struct Scsi_Host *
csio_ln_to_shost(struct csio_lnode *ln)
{
diff --git a/drivers/scsi/csiostor/csio_lnode.h b/drivers/scsi/csiostor/csio_lnode.h
index 8d84988..0f9c041 100644
--- a/drivers/scsi/csiostor/csio_lnode.h
+++ b/drivers/scsi/csiostor/csio_lnode.h
@@ -114,7 +114,7 @@
uint32_t n_rnode_match; /* matched rnode */
uint32_t n_dev_loss_tmo; /* Device loss timeout */
uint32_t n_fdmi_err; /* fdmi err */
- uint32_t n_evt_fw[RSCN_DEV_LOST]; /* fw events */
+ uint32_t n_evt_fw[PROTO_ERR_IMPL_LOGO]; /* fw events */
enum csio_ln_ev n_evt_sm[CSIO_LNE_MAX_EVENT]; /* State m/c events */
uint32_t n_rnode_alloc; /* rnode allocated */
uint32_t n_rnode_free; /* rnode freed */
diff --git a/drivers/scsi/csiostor/csio_rnode.c b/drivers/scsi/csiostor/csio_rnode.c
index 51c6a38..e9c3b04 100644
--- a/drivers/scsi/csiostor/csio_rnode.c
+++ b/drivers/scsi/csiostor/csio_rnode.c
@@ -302,7 +302,7 @@
{
uint8_t rport_type;
struct csio_rnode *rn, *match_rn;
- uint32_t vnp_flowid;
+ uint32_t vnp_flowid = 0;
__be32 *port_id;
port_id = (__be32 *)&rdevp->r_id[0];
@@ -350,6 +350,14 @@
* Else, go ahead and alloc a new rnode.
*/
if (!memcmp(csio_rn_wwpn(match_rn), rdevp->wwpn, 8)) {
+ if (rn == match_rn)
+ goto found_rnode;
+ csio_ln_dbg(ln,
+ "nport_id:x%x and wwpn:%llx"
+ " match for ssni:x%x\n",
+ rn->nport_id,
+ wwn_to_u64(rdevp->wwpn),
+ rdev_flowid);
if (csio_is_rnode_ready(rn)) {
csio_ln_warn(ln,
"rnode is already"
diff --git a/drivers/scsi/csiostor/csio_rnode.h b/drivers/scsi/csiostor/csio_rnode.h
index a3b434c..6594009 100644
--- a/drivers/scsi/csiostor/csio_rnode.h
+++ b/drivers/scsi/csiostor/csio_rnode.h
@@ -63,7 +63,7 @@
uint32_t n_err_nomem; /* error nomem */
uint32_t n_evt_unexp; /* unexpected event */
uint32_t n_evt_drop; /* unexpected event */
- uint32_t n_evt_fw[RSCN_DEV_LOST]; /* fw events */
+ uint32_t n_evt_fw[PROTO_ERR_IMPL_LOGO]; /* fw events */
enum csio_rn_ev n_evt_sm[CSIO_RNFE_MAX_EVENT]; /* State m/c events */
uint32_t n_lun_rst; /* Number of resets of
* of LUNs under this
diff --git a/drivers/scsi/csiostor/csio_wr.c b/drivers/scsi/csiostor/csio_wr.c
index c32df1b..4255ce2 100644
--- a/drivers/scsi/csiostor/csio_wr.c
+++ b/drivers/scsi/csiostor/csio_wr.c
@@ -85,8 +85,8 @@
*/
if (flq->inc_idx >= 8) {
csio_wr_reg32(hw, DBPRIO(1) | QID(flq->un.fl.flid) |
- PIDX(flq->inc_idx / 8),
- MYPF_REG(SGE_PF_KDOORBELL));
+ CSIO_HW_PIDX(hw, flq->inc_idx / 8),
+ MYPF_REG(SGE_PF_KDOORBELL));
flq->inc_idx &= 7;
}
}
@@ -989,7 +989,8 @@
wmb();
/* Ring SGE Doorbell writing q->pidx into it */
csio_wr_reg32(hw, DBPRIO(prio) | QID(q->un.eq.physeqid) |
- PIDX(q->inc_idx), MYPF_REG(SGE_PF_KDOORBELL));
+ CSIO_HW_PIDX(hw, q->inc_idx),
+ MYPF_REG(SGE_PF_KDOORBELL));
q->inc_idx = 0;
return 0;
@@ -1331,20 +1332,30 @@
/* FL BUFFER SIZE#0 is Page size i,e already aligned to cache line */
csio_wr_reg32(hw, PAGE_SIZE, SGE_FL_BUFFER_SIZE0);
- csio_wr_reg32(hw,
- (csio_rd_reg32(hw, SGE_FL_BUFFER_SIZE2) +
- sge->csio_fl_align - 1) & ~(sge->csio_fl_align - 1),
- SGE_FL_BUFFER_SIZE2);
- csio_wr_reg32(hw,
- (csio_rd_reg32(hw, SGE_FL_BUFFER_SIZE3) +
- sge->csio_fl_align - 1) & ~(sge->csio_fl_align - 1),
- SGE_FL_BUFFER_SIZE3);
+
+ /*
+ * If using hard params, the following will get set correctly
+ * in csio_wr_set_sge().
+ */
+ if (hw->flags & CSIO_HWF_USING_SOFT_PARAMS) {
+ csio_wr_reg32(hw,
+ (csio_rd_reg32(hw, SGE_FL_BUFFER_SIZE2) +
+ sge->csio_fl_align - 1) & ~(sge->csio_fl_align - 1),
+ SGE_FL_BUFFER_SIZE2);
+ csio_wr_reg32(hw,
+ (csio_rd_reg32(hw, SGE_FL_BUFFER_SIZE3) +
+ sge->csio_fl_align - 1) & ~(sge->csio_fl_align - 1),
+ SGE_FL_BUFFER_SIZE3);
+ }
csio_wr_reg32(hw, HPZ0(PAGE_SHIFT - 12), ULP_RX_TDDP_PSZ);
/* default value of rx_dma_offset of the NIC driver */
csio_set_reg_field(hw, SGE_CONTROL, PKTSHIFT_MASK,
PKTSHIFT(CSIO_SGE_RX_DMA_OFFSET));
+
+ csio_hw_tp_wr_bits_indirect(hw, TP_INGRESS_CONFIG,
+ CSUM_HAS_PSEUDO_HDR, 0);
}
static void
@@ -1460,18 +1471,21 @@
* and generate an interrupt when this occurs so we can recover.
*/
csio_set_reg_field(hw, SGE_DBFIFO_STATUS,
- HP_INT_THRESH(HP_INT_THRESH_MASK) |
- LP_INT_THRESH(LP_INT_THRESH_MASK),
- HP_INT_THRESH(CSIO_SGE_DBFIFO_INT_THRESH) |
- LP_INT_THRESH(CSIO_SGE_DBFIFO_INT_THRESH));
+ HP_INT_THRESH(HP_INT_THRESH_MASK) |
+ CSIO_HW_LP_INT_THRESH(hw, CSIO_HW_M_LP_INT_THRESH(hw)),
+ HP_INT_THRESH(CSIO_SGE_DBFIFO_INT_THRESH) |
+ CSIO_HW_LP_INT_THRESH(hw, CSIO_SGE_DBFIFO_INT_THRESH));
+
csio_set_reg_field(hw, SGE_DOORBELL_CONTROL, ENABLE_DROP,
ENABLE_DROP);
/* SGE_FL_BUFFER_SIZE0 is set up by csio_wr_fixup_host_params(). */
CSIO_SET_FLBUF_SIZE(hw, 1, CSIO_SGE_FLBUF_SIZE1);
- CSIO_SET_FLBUF_SIZE(hw, 2, CSIO_SGE_FLBUF_SIZE2);
- CSIO_SET_FLBUF_SIZE(hw, 3, CSIO_SGE_FLBUF_SIZE3);
+ csio_wr_reg32(hw, (CSIO_SGE_FLBUF_SIZE2 + sge->csio_fl_align - 1)
+ & ~(sge->csio_fl_align - 1), SGE_FL_BUFFER_SIZE2);
+ csio_wr_reg32(hw, (CSIO_SGE_FLBUF_SIZE3 + sge->csio_fl_align - 1)
+ & ~(sge->csio_fl_align - 1), SGE_FL_BUFFER_SIZE3);
CSIO_SET_FLBUF_SIZE(hw, 4, CSIO_SGE_FLBUF_SIZE4);
CSIO_SET_FLBUF_SIZE(hw, 5, CSIO_SGE_FLBUF_SIZE5);
CSIO_SET_FLBUF_SIZE(hw, 6, CSIO_SGE_FLBUF_SIZE6);
@@ -1522,22 +1536,24 @@
csio_wr_sge_init(struct csio_hw *hw)
{
/*
- * If we are master:
+ * If we are master and chip is not initialized:
* - If we plan to use the config file, we need to fixup some
* host specific registers, and read the rest of the SGE
* configuration.
* - If we dont plan to use the config file, we need to initialize
* SGE entirely, including fixing the host specific registers.
+ * If we are master and chip is initialized, just read and work off of
+ * the already initialized SGE values.
* If we arent the master, we are only allowed to read and work off of
* the already initialized SGE values.
*
* Therefore, before calling this function, we assume that the master-
- * ship of the card, and whether to use config file or not, have
- * already been decided. In other words, CSIO_HWF_USING_SOFT_PARAMS and
- * CSIO_HWF_MASTER should be set/unset.
+ * ship of the card, state and whether to use config file or not, have
+ * already been decided.
*/
if (csio_is_hw_master(hw)) {
- csio_wr_fixup_host_params(hw);
+ if (hw->fw_state != CSIO_DEV_STATE_INIT)
+ csio_wr_fixup_host_params(hw);
if (hw->flags & CSIO_HWF_USING_SOFT_PARAMS)
csio_wr_get_sge(hw);
diff --git a/include/linux/if_team.h b/include/linux/if_team.h
index cfd21e3..4474557 100644
--- a/include/linux/if_team.h
+++ b/include/linux/if_team.h
@@ -112,6 +112,10 @@
void (*port_disabled)(struct team *team, struct team_port *port);
};
+extern int team_modeop_port_enter(struct team *team, struct team_port *port);
+extern void team_modeop_port_change_dev_addr(struct team *team,
+ struct team_port *port);
+
enum team_option_type {
TEAM_OPTION_TYPE_U32,
TEAM_OPTION_TYPE_STRING,
@@ -236,7 +240,26 @@
return NULL;
}
-extern int team_port_set_team_dev_addr(struct team_port *port);
+static inline struct team_port *
+team_get_first_port_txable_rcu(struct team *team, struct team_port *port)
+{
+ struct team_port *cur;
+
+ if (likely(team_port_txable(port)))
+ return port;
+ cur = port;
+ list_for_each_entry_continue_rcu(cur, &team->port_list, list)
+ if (team_port_txable(port))
+ return cur;
+ list_for_each_entry_rcu(cur, &team->port_list, list) {
+ if (cur == port)
+ break;
+ if (team_port_txable(port))
+ return cur;
+ }
+ return NULL;
+}
+
extern int team_options_register(struct team *team,
const struct team_option *option,
size_t option_count);
diff --git a/include/linux/micrel_phy.h b/include/linux/micrel_phy.h
index 9dbb41a..8752dbb 100644
--- a/include/linux/micrel_phy.h
+++ b/include/linux/micrel_phy.h
@@ -19,6 +19,7 @@
#define PHY_ID_KSZ9021 0x00221610
#define PHY_ID_KS8737 0x00221720
#define PHY_ID_KSZ8021 0x00221555
+#define PHY_ID_KSZ8031 0x00221556
#define PHY_ID_KSZ8041 0x00221510
#define PHY_ID_KSZ8051 0x00221550
/* same id: ks8001 Rev. A/B, and ks8721 Rev 3. */
diff --git a/include/linux/netdev_features.h b/include/linux/netdev_features.h
index 3dd3934..d6ee2d0 100644
--- a/include/linux/netdev_features.h
+++ b/include/linux/netdev_features.h
@@ -42,9 +42,9 @@
NETIF_F_TSO6_BIT, /* ... TCPv6 segmentation */
NETIF_F_FSO_BIT, /* ... FCoE segmentation */
NETIF_F_GSO_GRE_BIT, /* ... GRE with TSO */
- /**/NETIF_F_GSO_LAST, /* [can't be last bit, see GSO_MASK] */
- NETIF_F_GSO_RESERVED2 /* ... free (fill GSO_MASK to 8 bits) */
- = NETIF_F_GSO_LAST,
+ NETIF_F_GSO_UDP_TUNNEL_BIT, /* ... UDP TUNNEL with TSO */
+ /**/NETIF_F_GSO_LAST = /* last bit, see GSO_MASK */
+ NETIF_F_GSO_UDP_TUNNEL_BIT,
NETIF_F_FCOE_CRC_BIT, /* FCoE CRC32 */
NETIF_F_SCTP_CSUM_BIT, /* SCTP checksum offload */
@@ -102,7 +102,8 @@
#define NETIF_F_VLAN_CHALLENGED __NETIF_F(VLAN_CHALLENGED)
#define NETIF_F_RXFCS __NETIF_F(RXFCS)
#define NETIF_F_RXALL __NETIF_F(RXALL)
-#define NETIF_F_GRE_GSO __NETIF_F(GSO_GRE)
+#define NETIF_F_GSO_GRE __NETIF_F(GSO_GRE)
+#define NETIF_F_GSO_UDP_TUNNEL __NETIF_F(GSO_UDP_TUNNEL)
/* Features valid for ethtool to change */
/* = all defined minus driver/device-class-related */
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index b3d00fa..56e3e06 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1475,6 +1475,11 @@
*/
#define SET_NETDEV_DEVTYPE(net, devtype) ((net)->dev.type = (devtype))
+/* Default NAPI poll() weight
+ * Device drivers are strongly advised to not use bigger value
+ */
+#define NAPI_POLL_WEIGHT 64
+
/**
* netif_napi_add - initialize a napi context
* @dev: network device
@@ -1612,6 +1617,9 @@
list_for_each_entry_continue(d, &(net)->dev_base_head, dev_list)
#define for_each_netdev_continue_rcu(net, d) \
list_for_each_entry_continue_rcu(d, &(net)->dev_base_head, dev_list)
+#define for_each_netdev_in_bond_rcu(bond, slave) \
+ for_each_netdev_rcu(&init_net, slave) \
+ if (netdev_master_upper_dev_get_rcu(slave) == bond)
#define net_device_entry(lh) list_entry(lh, struct net_device, dev_list)
static inline struct net_device *next_net_device(struct net_device *dev)
@@ -1684,7 +1692,6 @@
extern void free_netdev(struct net_device *dev);
extern void synchronize_net(void);
extern int init_dummy_netdev(struct net_device *dev);
-extern void netdev_resync_ops(struct net_device *dev);
extern struct net_device *dev_get_by_index(struct net *net, int ifindex);
extern struct net_device *__dev_get_by_index(struct net *net, int ifindex);
@@ -2678,6 +2685,19 @@
{
return __skb_gso_segment(skb, features, true);
}
+__be16 skb_network_protocol(struct sk_buff *skb);
+
+static inline bool can_checksum_protocol(netdev_features_t features,
+ __be16 protocol)
+{
+ return ((features & NETIF_F_GEN_CSUM) ||
+ ((features & NETIF_F_V4_CSUM) &&
+ protocol == htons(ETH_P_IP)) ||
+ ((features & NETIF_F_V6_CSUM) &&
+ protocol == htons(ETH_P_IPV6)) ||
+ ((features & NETIF_F_FCOE_CRC) &&
+ protocol == htons(ETH_P_FCOE)));
+}
#ifdef CONFIG_BUG
extern void netdev_rx_csum_fault(struct net_device *dev);
@@ -2756,6 +2776,11 @@
dev->gso_max_size = size;
}
+static inline bool netif_is_bond_master(struct net_device *dev)
+{
+ return dev->flags & IFF_MASTER && dev->priv_flags & IFF_BONDING;
+}
+
static inline bool netif_is_bond_slave(struct net_device *dev)
{
return dev->flags & IFF_SLAVE && dev->priv_flags & IFF_BONDING;
diff --git a/include/linux/openvswitch.h b/include/linux/openvswitch.h
index d42e174..67d6c7b 100644
--- a/include/linux/openvswitch.h
+++ b/include/linux/openvswitch.h
@@ -94,7 +94,7 @@
};
/* Fixed logical ports. */
-#define OVSP_LOCAL ((__u16)0)
+#define OVSP_LOCAL ((__u32)0)
/* Packet transfer. */
@@ -127,7 +127,8 @@
* for %OVS_PACKET_CMD_EXECUTE. It has nested %OVS_ACTION_ATTR_* attributes.
* @OVS_PACKET_ATTR_USERDATA: Present for an %OVS_PACKET_CMD_ACTION
* notification if the %OVS_ACTION_ATTR_USERSPACE action specified an
- * %OVS_USERSPACE_ATTR_USERDATA attribute.
+ * %OVS_USERSPACE_ATTR_USERDATA attribute, with the same length and content
+ * specified there.
*
* These attributes follow the &struct ovs_header within the Generic Netlink
* payload for %OVS_PACKET_* commands.
@@ -137,7 +138,7 @@
OVS_PACKET_ATTR_PACKET, /* Packet data. */
OVS_PACKET_ATTR_KEY, /* Nested OVS_KEY_ATTR_* attributes. */
OVS_PACKET_ATTR_ACTIONS, /* Nested OVS_ACTION_ATTR_* attributes. */
- OVS_PACKET_ATTR_USERDATA, /* u64 OVS_ACTION_ATTR_USERSPACE arg. */
+ OVS_PACKET_ATTR_USERDATA, /* OVS_ACTION_ATTR_USERSPACE arg. */
__OVS_PACKET_ATTR_MAX
};
@@ -389,13 +390,13 @@
* enum ovs_userspace_attr - Attributes for %OVS_ACTION_ATTR_USERSPACE action.
* @OVS_USERSPACE_ATTR_PID: u32 Netlink PID to which the %OVS_PACKET_CMD_ACTION
* message should be sent. Required.
- * @OVS_USERSPACE_ATTR_USERDATA: If present, its u64 argument is copied to the
- * %OVS_PACKET_CMD_ACTION message as %OVS_PACKET_ATTR_USERDATA,
+ * @OVS_USERSPACE_ATTR_USERDATA: If present, its variable-length argument is
+ * copied to the %OVS_PACKET_CMD_ACTION message as %OVS_PACKET_ATTR_USERDATA.
*/
enum ovs_userspace_attr {
OVS_USERSPACE_ATTR_UNSPEC,
OVS_USERSPACE_ATTR_PID, /* u32 Netlink PID to receive upcalls. */
- OVS_USERSPACE_ATTR_USERDATA, /* u64 optional user-specified cookie. */
+ OVS_USERSPACE_ATTR_USERDATA, /* Optional user-specified cookie. */
__OVS_USERSPACE_ATTR_MAX
};
diff --git a/include/linux/phy.h b/include/linux/phy.h
index 33999ad..9e11039 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -455,6 +455,14 @@
*/
void (*txtstamp)(struct phy_device *dev, struct sk_buff *skb, int type);
+ /* Some devices (e.g. qnap TS-119P II) require PHY register changes to
+ * enable Wake on LAN, so set_wol is provided to be called in the
+ * ethernet driver's set_wol function. */
+ int (*set_wol)(struct phy_device *dev, struct ethtool_wolinfo *wol);
+
+ /* See set_wol, but for checking whether Wake on LAN is enabled. */
+ void (*get_wol)(struct phy_device *dev, struct ethtool_wolinfo *wol);
+
struct device_driver driver;
};
#define to_phy_driver(d) container_of(d, struct phy_driver, driver)
@@ -560,6 +568,8 @@
int phy_get_eee_err(struct phy_device *phydev);
int phy_ethtool_set_eee(struct phy_device *phydev, struct ethtool_eee *data);
int phy_ethtool_get_eee(struct phy_device *phydev, struct ethtool_eee *data);
+int phy_ethtool_set_wol(struct phy_device *phydev, struct ethtool_wolinfo *wol);
+void phy_ethtool_get_wol(struct phy_device *phydev, struct ethtool_wolinfo *wol);
int __init mdio_bus_init(void);
void mdio_bus_exit(void);
diff --git a/include/linux/platform_data/cpsw.h b/include/linux/platform_data/cpsw.h
index 798fb80..bb3cd58 100644
--- a/include/linux/platform_data/cpsw.h
+++ b/include/linux/platform_data/cpsw.h
@@ -30,7 +30,7 @@
u32 channels; /* number of cpdma channels (symmetric) */
u32 slaves; /* number of slave cpgmac ports */
struct cpsw_slave_data *slave_data;
- u32 cpts_active_slave; /* time stamping slave */
+ u32 active_slave; /* time stamping, ethtool and SIOCGMIIPHY slave */
u32 cpts_clock_mult; /* convert input clock ticks to nanoseconds */
u32 cpts_clock_shift; /* convert input clock ticks to nanoseconds */
u32 ale_entries; /* ale table size */
diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index 489dd7bb..f28544b 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -69,6 +69,15 @@
struct netlink_callback *cb,
struct net_device *dev,
int idx);
+extern int ndo_dflt_fdb_add(struct ndmsg *ndm,
+ struct nlattr *tb[],
+ struct net_device *dev,
+ const unsigned char *addr,
+ u16 flags);
+extern int ndo_dflt_fdb_del(struct ndmsg *ndm,
+ struct nlattr *tb[],
+ struct net_device *dev,
+ const unsigned char *addr);
extern int ndo_dflt_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq,
struct net_device *dev, u16 mode);
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 441f5bf..b66ecc6 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -316,6 +316,8 @@
SKB_GSO_FCOE = 1 << 5,
SKB_GSO_GRE = 1 << 6,
+
+ SKB_GSO_UDP_TUNNEL = 1 << 7,
};
#if BITS_PER_LONG > 32
@@ -387,6 +389,7 @@
* @vlan_tci: vlan tag control information
* @inner_transport_header: Inner transport layer header (encapsulation)
* @inner_network_header: Network layer header (encapsulation)
+ * @inner_mac_header: Link layer header (encapsulation)
* @transport_header: Transport layer header
* @network_header: Network layer header
* @mac_header: Link layer header
@@ -505,6 +508,7 @@
sk_buff_data_t inner_transport_header;
sk_buff_data_t inner_network_header;
+ sk_buff_data_t inner_mac_header;
sk_buff_data_t transport_header;
sk_buff_data_t network_header;
sk_buff_data_t mac_header;
@@ -1471,6 +1475,7 @@
static inline void skb_reset_inner_headers(struct sk_buff *skb)
{
+ skb->inner_mac_header = skb->mac_header;
skb->inner_network_header = skb->network_header;
skb->inner_transport_header = skb->transport_header;
}
@@ -1516,6 +1521,22 @@
skb->inner_network_header += offset;
}
+static inline unsigned char *skb_inner_mac_header(const struct sk_buff *skb)
+{
+ return skb->head + skb->inner_mac_header;
+}
+
+static inline void skb_reset_inner_mac_header(struct sk_buff *skb)
+{
+ skb->inner_mac_header = skb->data - skb->head;
+}
+
+static inline void skb_set_inner_mac_header(struct sk_buff *skb,
+ const int offset)
+{
+ skb_reset_inner_mac_header(skb);
+ skb->inner_mac_header += offset;
+}
static inline bool skb_transport_header_was_set(const struct sk_buff *skb)
{
return skb->transport_header != ~0U;
@@ -1609,6 +1630,21 @@
skb->inner_network_header = skb->data + offset;
}
+static inline unsigned char *skb_inner_mac_header(const struct sk_buff *skb)
+{
+ return skb->inner_mac_header;
+}
+
+static inline void skb_reset_inner_mac_header(struct sk_buff *skb)
+{
+ skb->inner_mac_header = skb->data;
+}
+
+static inline void skb_set_inner_mac_header(struct sk_buff *skb,
+ const int offset)
+{
+ skb->inner_mac_header = skb->data + offset;
+}
static inline bool skb_transport_header_was_set(const struct sk_buff *skb)
{
return skb->transport_header != NULL;
diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index f28408c..ed6a745 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -90,9 +90,6 @@
sack_ok : 4, /* SACK seen on SYN packet */
snd_wscale : 4, /* Window scaling received from sender */
rcv_wscale : 4; /* Window scaling to send to receiver */
- u8 cookie_plus:6, /* bytes in authenticator/cookie option */
- cookie_out_never:1,
- cookie_in_always:1;
u8 num_sacks; /* Number of SACK blocks */
u16 user_mss; /* mss requested by user in ioctl */
u16 mss_clamp; /* Maximal mss, negotiated at connection setup */
@@ -102,7 +99,6 @@
{
rx_opt->tstamp_ok = rx_opt->sack_ok = 0;
rx_opt->wscale_ok = rx_opt->snd_wscale = 0;
- rx_opt->cookie_plus = 0;
}
/* This is the max number of SACKS that we'll generate and process. It's safe
@@ -201,10 +197,10 @@
unused : 1;
u8 repair_queue;
u8 do_early_retrans:1,/* Enable RFC5827 early-retransmit */
- early_retrans_delayed:1, /* Delayed ER timer installed */
syn_data:1, /* SYN includes data */
syn_fastopen:1, /* SYN includes Fast Open option */
syn_data_acked:1;/* data in SYN is acked by SYN-ACK */
+ u32 tlp_high_seq; /* snd_nxt at the time of TLP retransmit. */
/* RTT measurement */
u32 srtt; /* smoothed round trip time << 3 */
@@ -320,12 +316,6 @@
struct tcp_md5sig_info __rcu *md5sig_info;
#endif
- /* When the cookie options are generated and exchanged, then this
- * object holds a reference to them (cookie_values->kref). Also
- * contains related tcp_cookie_transactions fields.
- */
- struct tcp_cookie_values *cookie_values;
-
/* TCP fastopen related information */
struct tcp_fastopen_request *fastopen_req;
/* fastopen_rsk points to request_sock that resulted in this big
@@ -361,10 +351,6 @@
#ifdef CONFIG_TCP_MD5SIG
struct tcp_md5sig_key *tw_md5_key;
#endif
- /* Few sockets in timewait have cookies; in that case, then this
- * object holds a reference to them (tw_cookie_values->kref).
- */
- struct tcp_cookie_values *tw_cookie_values;
};
static inline struct tcp_timewait_sock *tcp_twsk(const struct sock *sk)
diff --git a/include/linux/vm_sockets.h b/include/linux/vm_sockets.h
new file mode 100644
index 0000000..0805eec
--- /dev/null
+++ b/include/linux/vm_sockets.h
@@ -0,0 +1,23 @@
+/*
+ * VMware vSockets Driver
+ *
+ * Copyright (C) 2007-2013 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ */
+
+#ifndef _VM_SOCKETS_H
+#define _VM_SOCKETS_H
+
+#include <uapi/linux/vm_sockets.h>
+
+int vm_sockets_get_local_cid(void);
+
+#endif /* _VM_SOCKETS_H */
diff --git a/include/net/caif/caif_shm.h b/include/net/caif/caif_shm.h
deleted file mode 100644
index 5bcce55..0000000
--- a/include/net/caif/caif_shm.h
+++ /dev/null
@@ -1,26 +0,0 @@
-/*
- * Copyright (C) ST-Ericsson AB 2010
- * Contact: Sjur Brendeland / sjur.brandeland@stericsson.com
- * Author: Amarnath Revanna / amarnath.bangalore.revanna@stericsson.com
- * License terms: GNU General Public License (GPL) version 2
- */
-
-#ifndef CAIF_SHM_H_
-#define CAIF_SHM_H_
-
-struct shmdev_layer {
- u32 shm_base_addr;
- u32 shm_total_sz;
- u32 shm_id;
- u32 shm_loopback;
- void *hmbx;
- int (*pshmdev_mbxsend) (u32 shm_id, u32 mbx_msg);
- int (*pshmdev_mbxsetup) (void *pshmdrv_cb,
- struct shmdev_layer *pshm_dev, void *pshm_drv);
- struct net_device *pshm_netdev;
-};
-
-extern int caif_shmcore_probe(struct shmdev_layer *pshm_dev);
-extern void caif_shmcore_remove(struct net_device *pshm_netdev);
-
-#endif
diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h
index 1832927..de2c785 100644
--- a/include/net/inet_connection_sock.h
+++ b/include/net/inet_connection_sock.h
@@ -133,6 +133,8 @@
#define ICSK_TIME_RETRANS 1 /* Retransmit timer */
#define ICSK_TIME_DACK 2 /* Delayed ack timer */
#define ICSK_TIME_PROBE0 3 /* Zero window probe timer */
+#define ICSK_TIME_EARLY_RETRANS 4 /* Early retransmit timer */
+#define ICSK_TIME_LOSS_PROBE 5 /* Tail loss probe timer */
static inline struct inet_connection_sock *inet_csk(const struct sock *sk)
{
@@ -222,7 +224,8 @@
when = max_when;
}
- if (what == ICSK_TIME_RETRANS || what == ICSK_TIME_PROBE0) {
+ if (what == ICSK_TIME_RETRANS || what == ICSK_TIME_PROBE0 ||
+ what == ICSK_TIME_EARLY_RETRANS || what == ICSK_TIME_LOSS_PROBE) {
icsk->icsk_pending = what;
icsk->icsk_timeout = jiffies + when;
sk_reset_timer(sk, &icsk->icsk_retransmit_timer, icsk->icsk_timeout);
diff --git a/include/net/ip6_tunnel.h b/include/net/ip6_tunnel.h
index e03047f..ebdef7f 100644
--- a/include/net/ip6_tunnel.h
+++ b/include/net/ip6_tunnel.h
@@ -68,4 +68,24 @@
__u32 ip6_tnl_get_cap(struct ip6_tnl *t, const struct in6_addr *laddr,
const struct in6_addr *raddr);
+static inline void ip6tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+ struct net_device_stats *stats = &dev->stats;
+ int pkt_len, err;
+
+ nf_reset(skb);
+ pkt_len = skb->len;
+ err = ip6_local_out(skb);
+
+ if (net_xmit_eval(err) == 0) {
+ struct pcpu_tstats *tstats = this_cpu_ptr(dev->tstats);
+ u64_stats_update_begin(&tstats->syncp);
+ tstats->tx_bytes += pkt_len;
+ tstats->tx_packets++;
+ u64_stats_update_end(&tstats->syncp);
+ } else {
+ stats->tx_errors++;
+ stats->tx_aborted_errors++;
+ }
+}
#endif
diff --git a/include/net/ipip.h b/include/net/ipip.h
index fd19625..0c046e3 100644
--- a/include/net/ipip.h
+++ b/include/net/ipip.h
@@ -51,13 +51,10 @@
static inline void iptunnel_xmit(struct sk_buff *skb, struct net_device *dev)
{
int err;
- struct iphdr *iph = ip_hdr(skb);
int pkt_len = skb->len - skb_transport_offset(skb);
struct pcpu_tstats *tstats = this_cpu_ptr(dev->tstats);
nf_reset(skb);
- skb->ip_summed = CHECKSUM_NONE;
- ip_select_ident(iph, skb_dst(skb), NULL);
err = ip_local_out(skb);
if (likely(net_xmit_eval(err) == 0)) {
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 64d12e7..42ef6ab 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -217,7 +217,7 @@
};
struct ip6_flowlabel {
- struct ip6_flowlabel *next;
+ struct ip6_flowlabel __rcu *next;
__be32 label;
atomic_t users;
struct in6_addr dst;
@@ -238,9 +238,9 @@
#define IPV6_FLOWLABEL_MASK cpu_to_be32(0x000FFFFF)
struct ipv6_fl_socklist {
- struct ipv6_fl_socklist *next;
- struct ip6_flowlabel *fl;
- struct rcu_head rcu;
+ struct ipv6_fl_socklist __rcu *next;
+ struct ip6_flowlabel *fl;
+ struct rcu_head rcu;
};
extern struct ip6_flowlabel *fl6_sock_lookup(struct sock *sk, __be32 label);
@@ -320,6 +320,18 @@
return __ipv6_addr_src_scope(__ipv6_addr_type(addr));
}
+static inline bool __ipv6_addr_needs_scope_id(int type)
+{
+ return type & IPV6_ADDR_LINKLOCAL ||
+ (type & IPV6_ADDR_MULTICAST &&
+ (type & (IPV6_ADDR_LOOPBACK|IPV6_ADDR_LINKLOCAL)));
+}
+
+static inline __u32 ipv6_iface_scope_id(const struct in6_addr *addr, int iface)
+{
+ return __ipv6_addr_needs_scope_id(__ipv6_addr_type(addr)) ? iface : 0;
+}
+
static inline int ipv6_addr_cmp(const struct in6_addr *a1, const struct in6_addr *a2)
{
return memcmp(a1, a2, sizeof(struct in6_addr));
diff --git a/include/net/request_sock.h b/include/net/request_sock.h
index a51dbd1..9069e65 100644
--- a/include/net/request_sock.h
+++ b/include/net/request_sock.h
@@ -27,19 +27,13 @@
struct dst_entry;
struct proto;
-/* empty to "strongly type" an otherwise void parameter.
- */
-struct request_values {
-};
-
struct request_sock_ops {
int family;
int obj_size;
struct kmem_cache *slab;
char *slab_name;
int (*rtx_syn_ack)(struct sock *sk,
- struct request_sock *req,
- struct request_values *rvp);
+ struct request_sock *req);
void (*send_ack)(struct sock *sk, struct sk_buff *skb,
struct request_sock *req);
void (*send_reset)(struct sock *sk,
diff --git a/include/net/tcp.h b/include/net/tcp.h
index cf0694d..7f2f171 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -179,7 +179,6 @@
#define TCPOPT_SACK 5 /* SACK Block */
#define TCPOPT_TIMESTAMP 8 /* Better RTT estimations/PAWS */
#define TCPOPT_MD5SIG 19 /* MD5 Signature (RFC2385) */
-#define TCPOPT_COOKIE 253 /* Cookie extension (experimental) */
#define TCPOPT_EXP 254 /* Experimental */
/* Magic number to be after the option value for sharing TCP
* experimental options. See draft-ietf-tcpm-experimental-options-00.txt
@@ -454,7 +453,7 @@
extern int tcp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
size_t len, int nonblock, int flags, int *addr_len);
extern void tcp_parse_options(const struct sk_buff *skb,
- struct tcp_options_received *opt_rx, const u8 **hvpp,
+ struct tcp_options_received *opt_rx,
int estab, struct tcp_fastopen_cookie *foc);
extern const u8 *tcp_parse_md5sig_option(const struct tcphdr *th);
@@ -476,7 +475,6 @@
extern int tcp_connect(struct sock *sk);
extern struct sk_buff * tcp_make_synack(struct sock *sk, struct dst_entry *dst,
struct request_sock *req,
- struct request_values *rvp,
struct tcp_fastopen_cookie *foc);
extern int tcp_disconnect(struct sock *sk, int flags);
@@ -543,6 +541,8 @@
extern void tcp_push_one(struct sock *, unsigned int mss_now);
extern void tcp_send_ack(struct sock *sk);
extern void tcp_send_delayed_ack(struct sock *sk);
+extern void tcp_send_loss_probe(struct sock *sk);
+extern bool tcp_schedule_loss_probe(struct sock *sk);
/* tcp_input.c */
extern void tcp_cwnd_application_limited(struct sock *sk);
@@ -873,8 +873,8 @@
static inline void tcp_enable_early_retrans(struct tcp_sock *tp)
{
tp->do_early_retrans = sysctl_tcp_early_retrans &&
- !sysctl_tcp_thin_dupack && sysctl_tcp_reordering == 3;
- tp->early_retrans_delayed = 0;
+ sysctl_tcp_early_retrans < 4 && !sysctl_tcp_thin_dupack &&
+ sysctl_tcp_reordering == 3;
}
static inline void tcp_disable_early_retrans(struct tcp_sock *tp)
@@ -1030,50 +1030,7 @@
#endif
}
-/* Packet is added to VJ-style prequeue for processing in process
- * context, if a reader task is waiting. Apparently, this exciting
- * idea (VJ's mail "Re: query about TCP header on tcp-ip" of 07 Sep 93)
- * failed somewhere. Latency? Burstiness? Well, at least now we will
- * see, why it failed. 8)8) --ANK
- *
- * NOTE: is this not too big to inline?
- */
-static inline bool tcp_prequeue(struct sock *sk, struct sk_buff *skb)
-{
- struct tcp_sock *tp = tcp_sk(sk);
-
- if (sysctl_tcp_low_latency || !tp->ucopy.task)
- return false;
-
- if (skb->len <= tcp_hdrlen(skb) &&
- skb_queue_len(&tp->ucopy.prequeue) == 0)
- return false;
-
- __skb_queue_tail(&tp->ucopy.prequeue, skb);
- tp->ucopy.memory += skb->truesize;
- if (tp->ucopy.memory > sk->sk_rcvbuf) {
- struct sk_buff *skb1;
-
- BUG_ON(sock_owned_by_user(sk));
-
- while ((skb1 = __skb_dequeue(&tp->ucopy.prequeue)) != NULL) {
- sk_backlog_rcv(sk, skb1);
- NET_INC_STATS_BH(sock_net(sk),
- LINUX_MIB_TCPPREQUEUEDROPPED);
- }
-
- tp->ucopy.memory = 0;
- } else if (skb_queue_len(&tp->ucopy.prequeue) == 1) {
- wake_up_interruptible_sync_poll(sk_sleep(sk),
- POLLIN | POLLRDNORM | POLLRDBAND);
- if (!inet_csk_ack_scheduled(sk))
- inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK,
- (3 * tcp_rto_min(sk)) / 4,
- TCP_RTO_MAX);
- }
- return true;
-}
-
+extern bool tcp_prequeue(struct sock *sk, struct sk_buff *skb);
#undef STATE_TRACE
@@ -1630,91 +1587,6 @@
#endif
};
-/* Using SHA1 for now, define some constants.
- */
-#define COOKIE_DIGEST_WORDS (SHA_DIGEST_WORDS)
-#define COOKIE_MESSAGE_WORDS (SHA_MESSAGE_BYTES / 4)
-#define COOKIE_WORKSPACE_WORDS (COOKIE_DIGEST_WORDS + COOKIE_MESSAGE_WORDS)
-
-extern int tcp_cookie_generator(u32 *bakery);
-
-/**
- * struct tcp_cookie_values - each socket needs extra space for the
- * cookies, together with (optional) space for any SYN data.
- *
- * A tcp_sock contains a pointer to the current value, and this is
- * cloned to the tcp_timewait_sock.
- *
- * @cookie_pair: variable data from the option exchange.
- *
- * @cookie_desired: user specified tcpct_cookie_desired. Zero
- * indicates default (sysctl_tcp_cookie_size).
- * After cookie sent, remembers size of cookie.
- * Range 0, TCP_COOKIE_MIN to TCP_COOKIE_MAX.
- *
- * @s_data_desired: user specified tcpct_s_data_desired. When the
- * constant payload is specified (@s_data_constant),
- * holds its length instead.
- * Range 0 to TCP_MSS_DESIRED.
- *
- * @s_data_payload: constant data that is to be included in the
- * payload of SYN or SYNACK segments when the
- * cookie option is present.
- */
-struct tcp_cookie_values {
- struct kref kref;
- u8 cookie_pair[TCP_COOKIE_PAIR_SIZE];
- u8 cookie_pair_size;
- u8 cookie_desired;
- u16 s_data_desired:11,
- s_data_constant:1,
- s_data_in:1,
- s_data_out:1,
- s_data_unused:2;
- u8 s_data_payload[0];
-};
-
-static inline void tcp_cookie_values_release(struct kref *kref)
-{
- kfree(container_of(kref, struct tcp_cookie_values, kref));
-}
-
-/* The length of constant payload data. Note that s_data_desired is
- * overloaded, depending on s_data_constant: either the length of constant
- * data (returned here) or the limit on variable data.
- */
-static inline int tcp_s_data_size(const struct tcp_sock *tp)
-{
- return (tp->cookie_values != NULL && tp->cookie_values->s_data_constant)
- ? tp->cookie_values->s_data_desired
- : 0;
-}
-
-/**
- * struct tcp_extend_values - tcp_ipv?.c to tcp_output.c workspace.
- *
- * As tcp_request_sock has already been extended in other places, the
- * only remaining method is to pass stack values along as function
- * parameters. These parameters are not needed after sending SYNACK.
- *
- * @cookie_bakery: cryptographic secret and message workspace.
- *
- * @cookie_plus: bytes in authenticator/cookie option, copied from
- * struct tcp_options_received (above).
- */
-struct tcp_extend_values {
- struct request_values rv;
- u32 cookie_bakery[COOKIE_WORKSPACE_WORDS];
- u8 cookie_plus:6,
- cookie_out_never:1,
- cookie_in_always:1;
-};
-
-static inline struct tcp_extend_values *tcp_xv(struct request_values *rvp)
-{
- return (struct tcp_extend_values *)rvp;
-}
-
extern void tcp_v4_init(void);
extern void tcp_init(void);
diff --git a/include/uapi/linux/if_packet.h b/include/uapi/linux/if_packet.h
index f9a6037..8136658 100644
--- a/include/uapi/linux/if_packet.h
+++ b/include/uapi/linux/if_packet.h
@@ -55,6 +55,8 @@
#define PACKET_FANOUT_HASH 0
#define PACKET_FANOUT_LB 1
#define PACKET_FANOUT_CPU 2
+#define PACKET_FANOUT_ROLLOVER 3
+#define PACKET_FANOUT_FLAG_ROLLOVER 0x1000
#define PACKET_FANOUT_FLAG_DEFRAG 0x8000
struct tpacket_stats {
diff --git a/include/uapi/linux/neighbour.h b/include/uapi/linux/neighbour.h
index adb068c..f175212 100644
--- a/include/uapi/linux/neighbour.h
+++ b/include/uapi/linux/neighbour.h
@@ -21,6 +21,9 @@
NDA_CACHEINFO,
NDA_PROBES,
NDA_VLAN,
+ NDA_PORT,
+ NDA_VNI,
+ NDA_IFINDEX,
__NDA_MAX
};
diff --git a/include/uapi/linux/pkt_sched.h b/include/uapi/linux/pkt_sched.h
index 32aef0a..dbd71b0 100644
--- a/include/uapi/linux/pkt_sched.h
+++ b/include/uapi/linux/pkt_sched.h
@@ -348,6 +348,7 @@
TCA_HTB_INIT,
TCA_HTB_CTAB,
TCA_HTB_RTAB,
+ TCA_HTB_DIRECT_QLEN,
__TCA_HTB_MAX,
};
diff --git a/include/uapi/linux/snmp.h b/include/uapi/linux/snmp.h
index b49eab8..e00013a 100644
--- a/include/uapi/linux/snmp.h
+++ b/include/uapi/linux/snmp.h
@@ -202,6 +202,8 @@
LINUX_MIB_TCPFORWARDRETRANS, /* TCPForwardRetrans */
LINUX_MIB_TCPSLOWSTARTRETRANS, /* TCPSlowStartRetrans */
LINUX_MIB_TCPTIMEOUTS, /* TCPTimeouts */
+ LINUX_MIB_TCPLOSSPROBES, /* TCPLossProbes */
+ LINUX_MIB_TCPLOSSPROBERECOVERY, /* TCPLossProbeRecovery */
LINUX_MIB_TCPRENORECOVERYFAIL, /* TCPRenoRecoveryFail */
LINUX_MIB_TCPSACKRECOVERYFAIL, /* TCPSackRecoveryFail */
LINUX_MIB_TCPSCHEDULERFAILED, /* TCPSchedulerFailed */
diff --git a/include/uapi/linux/tcp.h b/include/uapi/linux/tcp.h
index 6b1ead0..8d776eb 100644
--- a/include/uapi/linux/tcp.h
+++ b/include/uapi/linux/tcp.h
@@ -102,7 +102,6 @@
#define TCP_QUICKACK 12 /* Block/reenable quick acks */
#define TCP_CONGESTION 13 /* Congestion control algorithm */
#define TCP_MD5SIG 14 /* TCP MD5 Signature (RFC2385) */
-#define TCP_COOKIE_TRANSACTIONS 15 /* TCP Cookie Transactions */
#define TCP_THIN_LINEAR_TIMEOUTS 16 /* Use linear timeouts for thin streams*/
#define TCP_THIN_DUPACK 17 /* Fast retrans. after 1 dupack */
#define TCP_USER_TIMEOUT 18 /* How long for loss retry before timeout */
@@ -199,29 +198,4 @@
__u8 tcpm_key[TCP_MD5SIG_MAXKEYLEN]; /* key (binary) */
};
-/* for TCP_COOKIE_TRANSACTIONS (TCPCT) socket option */
-#define TCP_COOKIE_MIN 8 /* 64-bits */
-#define TCP_COOKIE_MAX 16 /* 128-bits */
-#define TCP_COOKIE_PAIR_SIZE (2*TCP_COOKIE_MAX)
-
-/* Flags for both getsockopt and setsockopt */
-#define TCP_COOKIE_IN_ALWAYS (1 << 0) /* Discard SYN without cookie */
-#define TCP_COOKIE_OUT_NEVER (1 << 1) /* Prohibit outgoing cookies,
- * supercedes everything. */
-
-/* Flags for getsockopt */
-#define TCP_S_DATA_IN (1 << 2) /* Was data received? */
-#define TCP_S_DATA_OUT (1 << 3) /* Was data sent? */
-
-/* TCP_COOKIE_TRANSACTIONS data */
-struct tcp_cookie_transactions {
- __u16 tcpct_flags; /* see above */
- __u8 __tcpct_pad1; /* zero */
- __u8 tcpct_cookie_desired; /* bytes */
- __u16 tcpct_s_data_desired; /* bytes of variable data */
- __u16 tcpct_used; /* bytes in value */
- __u8 tcpct_value[TCP_MSS_DEFAULT];
-};
-
-
#endif /* _UAPI_LINUX_TCP_H */
diff --git a/include/uapi/linux/vm_sockets.h b/include/uapi/linux/vm_sockets.h
index df91301..b4ed5d8 100644
--- a/include/uapi/linux/vm_sockets.h
+++ b/include/uapi/linux/vm_sockets.h
@@ -13,12 +13,10 @@
* more details.
*/
-#ifndef _VM_SOCKETS_H_
-#define _VM_SOCKETS_H_
+#ifndef _UAPI_VM_SOCKETS_H
+#define _UAPI_VM_SOCKETS_H
-#if !defined(__KERNEL__)
-#include <sys/socket.h>
-#endif
+#include <linux/socket.h>
/* Option name for STREAM socket buffer size. Use as the option name in
* setsockopt(3) or getsockopt(3) to set or get an unsigned long long that
@@ -137,14 +135,13 @@
#define VM_SOCKETS_VERSION_MINOR(_v) (((_v) & 0x0000FFFF))
/* Address structure for vSockets. The address family should be set to
- * whatever vmci_sock_get_af_value_fd() returns. The structure members should
- * all align on their natural boundaries without resorting to compiler packing
- * directives. The total size of this structure should be exactly the same as
- * that of struct sockaddr.
+ * AF_VSOCK. The structure members should all align on their natural
+ * boundaries without resorting to compiler packing directives. The total size
+ * of this structure should be exactly the same as that of struct sockaddr.
*/
struct sockaddr_vm {
- sa_family_t svm_family;
+ __kernel_sa_family_t svm_family;
unsigned short svm_reserved1;
unsigned int svm_port;
unsigned int svm_cid;
@@ -156,8 +153,4 @@
#define IOCTL_VM_SOCKETS_GET_LOCAL_CID _IO(7, 0xb9)
-#if defined(__KERNEL__)
-int vm_sockets_get_local_cid(void);
-#endif
-
-#endif
+#endif /* _UAPI_VM_SOCKETS_H */
diff --git a/net/batman-adv/Kconfig b/net/batman-adv/Kconfig
index 8d8afb1..fa780b7 100644
--- a/net/batman-adv/Kconfig
+++ b/net/batman-adv/Kconfig
@@ -36,6 +36,20 @@
mesh networks. If you think that your network does not need
this option you can safely remove it and save some space.
+config BATMAN_ADV_NC
+ bool "Network Coding"
+ depends on BATMAN_ADV
+ default n
+ help
+ This option enables network coding, a mechanism that aims to
+ increase the overall network throughput by fusing multiple
+ packets in one transmission.
+ Note that interfaces controlled by batman-adv must be manually
+ configured to have promiscuous mode enabled in order to make
+ network coding work.
+ If you think that your network does not need this feature you
+ can safely disable it and save some space.
+
config BATMAN_ADV_DEBUG
bool "B.A.T.M.A.N. debugging"
depends on BATMAN_ADV
diff --git a/net/batman-adv/Makefile b/net/batman-adv/Makefile
index e45e3b4..4b8f192 100644
--- a/net/batman-adv/Makefile
+++ b/net/batman-adv/Makefile
@@ -30,6 +30,7 @@
batman-adv-y += hash.o
batman-adv-y += icmp_socket.o
batman-adv-y += main.o
+batman-adv-$(CONFIG_BATMAN_ADV_NC) += network-coding.o
batman-adv-y += originator.o
batman-adv-y += ring_buffer.o
batman-adv-y += routing.o
diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index a5bb0a76..071f288 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -27,6 +27,7 @@
#include "hard-interface.h"
#include "send.h"
#include "bat_algo.h"
+#include "network-coding.h"
static struct batadv_neigh_node *
batadv_iv_ogm_neigh_new(struct batadv_hard_iface *hard_iface,
@@ -1185,6 +1186,10 @@
if (!orig_neigh_node)
goto out;
+ /* Update nc_nodes of the originator */
+ batadv_nc_update_nc_node(bat_priv, orig_node, orig_neigh_node,
+ batadv_ogm_packet, is_single_hop_neigh);
+
orig_neigh_router = batadv_orig_node_get_router(orig_neigh_node);
/* drop packet if sender is not a direct neighbor and if we
diff --git a/net/batman-adv/debugfs.c b/net/batman-adv/debugfs.c
index 6ae8651..f186a55 100644
--- a/net/batman-adv/debugfs.c
+++ b/net/batman-adv/debugfs.c
@@ -32,6 +32,7 @@
#include "icmp_socket.h"
#include "bridge_loop_avoidance.h"
#include "distributed-arp-table.h"
+#include "network-coding.h"
static struct dentry *batadv_debugfs;
@@ -310,6 +311,14 @@
const struct file_operations fops;
};
+#ifdef CONFIG_BATMAN_ADV_NC
+static int batadv_nc_nodes_open(struct inode *inode, struct file *file)
+{
+ struct net_device *net_dev = (struct net_device *)inode->i_private;
+ return single_open(file, batadv_nc_nodes_seq_print_text, net_dev);
+}
+#endif
+
#define BATADV_DEBUGINFO(_name, _mode, _open) \
struct batadv_debuginfo batadv_debuginfo_##_name = { \
.attr = { .name = __stringify(_name), \
@@ -348,6 +357,9 @@
static BATADV_DEBUGINFO(transtable_local, S_IRUGO,
batadv_transtable_local_open);
static BATADV_DEBUGINFO(vis_data, S_IRUGO, batadv_vis_data_open);
+#ifdef CONFIG_BATMAN_ADV_NC
+static BATADV_DEBUGINFO(nc_nodes, S_IRUGO, batadv_nc_nodes_open);
+#endif
static struct batadv_debuginfo *batadv_mesh_debuginfos[] = {
&batadv_debuginfo_originators,
@@ -362,6 +374,9 @@
#endif
&batadv_debuginfo_transtable_local,
&batadv_debuginfo_vis_data,
+#ifdef CONFIG_BATMAN_ADV_NC
+ &batadv_debuginfo_nc_nodes,
+#endif
NULL,
};
@@ -431,6 +446,9 @@
}
}
+ if (batadv_nc_init_debugfs(bat_priv) < 0)
+ goto rem_attr;
+
return 0;
rem_attr:
debugfs_remove_recursive(bat_priv->debug_dir);
diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
index d54188a..8e15d96 100644
--- a/net/batman-adv/distributed-arp-table.c
+++ b/net/batman-adv/distributed-arp-table.c
@@ -816,7 +816,6 @@
bool ret = false;
struct batadv_dat_entry *dat_entry = NULL;
struct sk_buff *skb_new;
- struct batadv_hard_iface *primary_if = NULL;
if (!atomic_read(&bat_priv->distributed_arp_table))
goto out;
@@ -838,22 +837,18 @@
dat_entry = batadv_dat_entry_hash_find(bat_priv, ip_dst);
if (dat_entry) {
- primary_if = batadv_primary_if_get_selected(bat_priv);
- if (!primary_if)
- goto out;
-
skb_new = arp_create(ARPOP_REPLY, ETH_P_ARP, ip_src,
- primary_if->soft_iface, ip_dst, hw_src,
+ bat_priv->soft_iface, ip_dst, hw_src,
dat_entry->mac_addr, hw_src);
if (!skb_new)
goto out;
skb_reset_mac_header(skb_new);
skb_new->protocol = eth_type_trans(skb_new,
- primary_if->soft_iface);
+ bat_priv->soft_iface);
bat_priv->stats.rx_packets++;
bat_priv->stats.rx_bytes += skb->len + ETH_HLEN;
- primary_if->soft_iface->last_rx = jiffies;
+ bat_priv->soft_iface->last_rx = jiffies;
netif_rx(skb_new);
batadv_dbg(BATADV_DBG_DAT, bat_priv, "ARP request replied locally\n");
@@ -866,8 +861,6 @@
out:
if (dat_entry)
batadv_dat_entry_free_ref(dat_entry);
- if (primary_if)
- batadv_hardif_free_ref(primary_if);
return ret;
}
@@ -887,7 +880,6 @@
__be32 ip_src, ip_dst;
uint8_t *hw_src;
struct sk_buff *skb_new;
- struct batadv_hard_iface *primary_if = NULL;
struct batadv_dat_entry *dat_entry = NULL;
bool ret = false;
int err;
@@ -912,12 +904,8 @@
if (!dat_entry)
goto out;
- primary_if = batadv_primary_if_get_selected(bat_priv);
- if (!primary_if)
- goto out;
-
skb_new = arp_create(ARPOP_REPLY, ETH_P_ARP, ip_src,
- primary_if->soft_iface, ip_dst, hw_src,
+ bat_priv->soft_iface, ip_dst, hw_src,
dat_entry->mac_addr, hw_src);
if (!skb_new)
@@ -941,8 +929,6 @@
out:
if (dat_entry)
batadv_dat_entry_free_ref(dat_entry);
- if (primary_if)
- batadv_hardif_free_ref(primary_if);
if (ret)
kfree_skb(skb);
return ret;
diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index 0488d70..0495a7d 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -35,6 +35,7 @@
#include "vis.h"
#include "hash.h"
#include "bat_algo.h"
+#include "network-coding.h"
/* List manipulations on hardif_list have to be rtnl_lock()'ed,
@@ -135,6 +136,10 @@
if (ret < 0)
goto err;
+ ret = batadv_nc_init(bat_priv);
+ if (ret < 0)
+ goto err;
+
atomic_set(&bat_priv->gw.reselect, 0);
atomic_set(&bat_priv->mesh_state, BATADV_MESH_ACTIVE);
@@ -157,6 +162,7 @@
batadv_gw_node_purge(bat_priv);
batadv_originator_free(bat_priv);
+ batadv_nc_free(bat_priv);
batadv_tt_free(bat_priv);
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index ced08b9..0afd4ee 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -105,6 +105,8 @@
#define BATADV_RESET_PROTECTION_MS 30000
#define BATADV_EXPECTED_SEQNO_RANGE 65536
+#define BATADV_NC_NODE_TIMEOUT 10000 /* Milliseconds */
+
enum batadv_mesh_state {
BATADV_MESH_INACTIVE,
BATADV_MESH_ACTIVE,
@@ -185,6 +187,7 @@
* @BATADV_DBG_TT: translation table messages
* @BATADV_DBG_BLA: bridge loop avoidance messages
* @BATADV_DBG_DAT: ARP snooping and DAT related messages
+ * @BATADV_DBG_NC: network coding related messages
* @BATADV_DBG_ALL: the union of all the above log levels
*/
enum batadv_dbg_level {
@@ -193,7 +196,8 @@
BATADV_DBG_TT = BIT(2),
BATADV_DBG_BLA = BIT(3),
BATADV_DBG_DAT = BIT(4),
- BATADV_DBG_ALL = 31,
+ BATADV_DBG_NC = BIT(5),
+ BATADV_DBG_ALL = 63,
};
#ifdef CONFIG_BATMAN_ADV_DEBUG
@@ -298,4 +302,10 @@
return sum;
}
+/* Define a macro to reach the control buffer of the skb. The members of the
+ * control buffer are defined in struct batadv_skb_cb in types.h.
+ * The macro is inspired by the similar macro TCP_SKB_CB() in tcp.h.
+ */
+#define BATADV_SKB_CB(__skb) ((struct batadv_skb_cb *)&((__skb)->cb[0]))
+
#endif /* _NET_BATMAN_ADV_MAIN_H_ */
diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c
new file mode 100644
index 0000000..5728079
--- /dev/null
+++ b/net/batman-adv/network-coding.c
@@ -0,0 +1,1821 @@
+/* Copyright (C) 2012-2013 B.A.T.M.A.N. contributors:
+ *
+ * Martin Hundebøll, Jeppe Ledet-Pedersen
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA
+ */
+
+#include <linux/debugfs.h>
+
+#include "main.h"
+#include "hash.h"
+#include "network-coding.h"
+#include "send.h"
+#include "originator.h"
+#include "hard-interface.h"
+#include "routing.h"
+
+static struct lock_class_key batadv_nc_coding_hash_lock_class_key;
+static struct lock_class_key batadv_nc_decoding_hash_lock_class_key;
+
+static void batadv_nc_worker(struct work_struct *work);
+static int batadv_nc_recv_coded_packet(struct sk_buff *skb,
+ struct batadv_hard_iface *recv_if);
+
+/**
+ * batadv_nc_start_timer - initialise the nc periodic worker
+ * @bat_priv: the bat priv with all the soft interface information
+ */
+static void batadv_nc_start_timer(struct batadv_priv *bat_priv)
+{
+ queue_delayed_work(batadv_event_workqueue, &bat_priv->nc.work,
+ msecs_to_jiffies(10));
+}
+
+/**
+ * batadv_nc_init - initialise coding hash table and start house keeping
+ * @bat_priv: the bat priv with all the soft interface information
+ */
+int batadv_nc_init(struct batadv_priv *bat_priv)
+{
+ bat_priv->nc.timestamp_fwd_flush = jiffies;
+ bat_priv->nc.timestamp_sniffed_purge = jiffies;
+
+ if (bat_priv->nc.coding_hash || bat_priv->nc.decoding_hash)
+ return 0;
+
+ bat_priv->nc.coding_hash = batadv_hash_new(128);
+ if (!bat_priv->nc.coding_hash)
+ goto err;
+
+ batadv_hash_set_lock_class(bat_priv->nc.coding_hash,
+ &batadv_nc_coding_hash_lock_class_key);
+
+ bat_priv->nc.decoding_hash = batadv_hash_new(128);
+ if (!bat_priv->nc.decoding_hash)
+ goto err;
+
+ batadv_hash_set_lock_class(bat_priv->nc.coding_hash,
+ &batadv_nc_decoding_hash_lock_class_key);
+
+ /* Register our packet type */
+ if (batadv_recv_handler_register(BATADV_CODED,
+ batadv_nc_recv_coded_packet) < 0)
+ goto err;
+
+ INIT_DELAYED_WORK(&bat_priv->nc.work, batadv_nc_worker);
+ batadv_nc_start_timer(bat_priv);
+
+ return 0;
+
+err:
+ return -ENOMEM;
+}
+
+/**
+ * batadv_nc_init_bat_priv - initialise the nc specific bat_priv variables
+ * @bat_priv: the bat priv with all the soft interface information
+ */
+void batadv_nc_init_bat_priv(struct batadv_priv *bat_priv)
+{
+ atomic_set(&bat_priv->network_coding, 1);
+ bat_priv->nc.min_tq = 200;
+ bat_priv->nc.max_fwd_delay = 10;
+ bat_priv->nc.max_buffer_time = 200;
+}
+
+/**
+ * batadv_nc_init_orig - initialise the nc fields of an orig_node
+ * @orig_node: the orig_node which is going to be initialised
+ */
+void batadv_nc_init_orig(struct batadv_orig_node *orig_node)
+{
+ INIT_LIST_HEAD(&orig_node->in_coding_list);
+ INIT_LIST_HEAD(&orig_node->out_coding_list);
+ spin_lock_init(&orig_node->in_coding_list_lock);
+ spin_lock_init(&orig_node->out_coding_list_lock);
+}
+
+/**
+ * batadv_nc_node_free_rcu - rcu callback to free an nc node and remove
+ * its refcount on the orig_node
+ * @rcu: rcu pointer of the nc node
+ */
+static void batadv_nc_node_free_rcu(struct rcu_head *rcu)
+{
+ struct batadv_nc_node *nc_node;
+
+ nc_node = container_of(rcu, struct batadv_nc_node, rcu);
+ batadv_orig_node_free_ref(nc_node->orig_node);
+ kfree(nc_node);
+}
+
+/**
+ * batadv_nc_node_free_ref - decrements the nc node refcounter and possibly
+ * frees it
+ * @nc_node: the nc node to free
+ */
+static void batadv_nc_node_free_ref(struct batadv_nc_node *nc_node)
+{
+ if (atomic_dec_and_test(&nc_node->refcount))
+ call_rcu(&nc_node->rcu, batadv_nc_node_free_rcu);
+}
+
+/**
+ * batadv_nc_path_free_ref - decrements the nc path refcounter and possibly
+ * frees it
+ * @nc_path: the nc node to free
+ */
+static void batadv_nc_path_free_ref(struct batadv_nc_path *nc_path)
+{
+ if (atomic_dec_and_test(&nc_path->refcount))
+ kfree_rcu(nc_path, rcu);
+}
+
+/**
+ * batadv_nc_packet_free - frees nc packet
+ * @nc_packet: the nc packet to free
+ */
+static void batadv_nc_packet_free(struct batadv_nc_packet *nc_packet)
+{
+ if (nc_packet->skb)
+ kfree_skb(nc_packet->skb);
+
+ batadv_nc_path_free_ref(nc_packet->nc_path);
+ kfree(nc_packet);
+}
+
+/**
+ * batadv_nc_to_purge_nc_node - checks whether an nc node has to be purged
+ * @bat_priv: the bat priv with all the soft interface information
+ * @nc_node: the nc node to check
+ *
+ * Returns true if the entry has to be purged now, false otherwise
+ */
+static bool batadv_nc_to_purge_nc_node(struct batadv_priv *bat_priv,
+ struct batadv_nc_node *nc_node)
+{
+ if (atomic_read(&bat_priv->mesh_state) != BATADV_MESH_ACTIVE)
+ return true;
+
+ return batadv_has_timed_out(nc_node->last_seen, BATADV_NC_NODE_TIMEOUT);
+}
+
+/**
+ * batadv_nc_to_purge_nc_path_coding - checks whether an nc path has timed out
+ * @bat_priv: the bat priv with all the soft interface information
+ * @nc_path: the nc path to check
+ *
+ * Returns true if the entry has to be purged now, false otherwise
+ */
+static bool batadv_nc_to_purge_nc_path_coding(struct batadv_priv *bat_priv,
+ struct batadv_nc_path *nc_path)
+{
+ if (atomic_read(&bat_priv->mesh_state) != BATADV_MESH_ACTIVE)
+ return true;
+
+ /* purge the path when no packets has been added for 10 times the
+ * max_fwd_delay time
+ */
+ return batadv_has_timed_out(nc_path->last_valid,
+ bat_priv->nc.max_fwd_delay * 10);
+}
+
+/**
+ * batadv_nc_to_purge_nc_path_decoding - checks whether an nc path has timed out
+ * @bat_priv: the bat priv with all the soft interface information
+ * @nc_path: the nc path to check
+ *
+ * Returns true if the entry has to be purged now, false otherwise
+ */
+static bool batadv_nc_to_purge_nc_path_decoding(struct batadv_priv *bat_priv,
+ struct batadv_nc_path *nc_path)
+{
+ if (atomic_read(&bat_priv->mesh_state) != BATADV_MESH_ACTIVE)
+ return true;
+
+ /* purge the path when no packets has been added for 10 times the
+ * max_buffer time
+ */
+ return batadv_has_timed_out(nc_path->last_valid,
+ bat_priv->nc.max_buffer_time*10);
+}
+
+/**
+ * batadv_nc_purge_orig_nc_nodes - go through list of nc nodes and purge stale
+ * entries
+ * @bat_priv: the bat priv with all the soft interface information
+ * @list: list of nc nodes
+ * @lock: nc node list lock
+ * @to_purge: function in charge to decide whether an entry has to be purged or
+ * not. This function takes the nc node as argument and has to return
+ * a boolean value: true if the entry has to be deleted, false
+ * otherwise
+ */
+static void
+batadv_nc_purge_orig_nc_nodes(struct batadv_priv *bat_priv,
+ struct list_head *list,
+ spinlock_t *lock,
+ bool (*to_purge)(struct batadv_priv *,
+ struct batadv_nc_node *))
+{
+ struct batadv_nc_node *nc_node, *nc_node_tmp;
+
+ /* For each nc_node in list */
+ spin_lock_bh(lock);
+ list_for_each_entry_safe(nc_node, nc_node_tmp, list, list) {
+ /* if an helper function has been passed as parameter,
+ * ask it if the entry has to be purged or not
+ */
+ if (to_purge && !to_purge(bat_priv, nc_node))
+ continue;
+
+ batadv_dbg(BATADV_DBG_NC, bat_priv,
+ "Removing nc_node %pM -> %pM\n",
+ nc_node->addr, nc_node->orig_node->orig);
+ list_del_rcu(&nc_node->list);
+ batadv_nc_node_free_ref(nc_node);
+ }
+ spin_unlock_bh(lock);
+}
+
+/**
+ * batadv_nc_purge_orig - purges all nc node data attached of the given
+ * originator
+ * @bat_priv: the bat priv with all the soft interface information
+ * @orig_node: orig_node with the nc node entries to be purged
+ * @to_purge: function in charge to decide whether an entry has to be purged or
+ * not. This function takes the nc node as argument and has to return
+ * a boolean value: true is the entry has to be deleted, false
+ * otherwise
+ */
+void batadv_nc_purge_orig(struct batadv_priv *bat_priv,
+ struct batadv_orig_node *orig_node,
+ bool (*to_purge)(struct batadv_priv *,
+ struct batadv_nc_node *))
+{
+ /* Check ingoing nc_node's of this orig_node */
+ batadv_nc_purge_orig_nc_nodes(bat_priv, &orig_node->in_coding_list,
+ &orig_node->in_coding_list_lock,
+ to_purge);
+
+ /* Check outgoing nc_node's of this orig_node */
+ batadv_nc_purge_orig_nc_nodes(bat_priv, &orig_node->out_coding_list,
+ &orig_node->out_coding_list_lock,
+ to_purge);
+}
+
+/**
+ * batadv_nc_purge_orig_hash - traverse entire originator hash to check if they
+ * have timed out nc nodes
+ * @bat_priv: the bat priv with all the soft interface information
+ */
+static void batadv_nc_purge_orig_hash(struct batadv_priv *bat_priv)
+{
+ struct batadv_hashtable *hash = bat_priv->orig_hash;
+ struct hlist_head *head;
+ struct batadv_orig_node *orig_node;
+ uint32_t i;
+
+ if (!hash)
+ return;
+
+ /* For each orig_node */
+ for (i = 0; i < hash->size; i++) {
+ head = &hash->table[i];
+
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(orig_node, head, hash_entry)
+ batadv_nc_purge_orig(bat_priv, orig_node,
+ batadv_nc_to_purge_nc_node);
+ rcu_read_unlock();
+ }
+}
+
+/**
+ * batadv_nc_purge_paths - traverse all nc paths part of the hash and remove
+ * unused ones
+ * @bat_priv: the bat priv with all the soft interface information
+ * @hash: hash table containing the nc paths to check
+ * @to_purge: function in charge to decide whether an entry has to be purged or
+ * not. This function takes the nc node as argument and has to return
+ * a boolean value: true is the entry has to be deleted, false
+ * otherwise
+ */
+static void batadv_nc_purge_paths(struct batadv_priv *bat_priv,
+ struct batadv_hashtable *hash,
+ bool (*to_purge)(struct batadv_priv *,
+ struct batadv_nc_path *))
+{
+ struct hlist_head *head;
+ struct hlist_node *node_tmp;
+ struct batadv_nc_path *nc_path;
+ spinlock_t *lock; /* Protects lists in hash */
+ uint32_t i;
+
+ for (i = 0; i < hash->size; i++) {
+ head = &hash->table[i];
+ lock = &hash->list_locks[i];
+
+ /* For each nc_path in this bin */
+ spin_lock_bh(lock);
+ hlist_for_each_entry_safe(nc_path, node_tmp, head, hash_entry) {
+ /* if an helper function has been passed as parameter,
+ * ask it if the entry has to be purged or not
+ */
+ if (to_purge && !to_purge(bat_priv, nc_path))
+ continue;
+
+ /* purging an non-empty nc_path should never happen, but
+ * is observed under high CPU load. Delay the purging
+ * until next iteration to allow the packet_list to be
+ * emptied first.
+ */
+ if (!unlikely(list_empty(&nc_path->packet_list))) {
+ net_ratelimited_function(printk,
+ KERN_WARNING
+ "Skipping free of non-empty nc_path (%pM -> %pM)!\n",
+ nc_path->prev_hop,
+ nc_path->next_hop);
+ continue;
+ }
+
+ /* nc_path is unused, so remove it */
+ batadv_dbg(BATADV_DBG_NC, bat_priv,
+ "Remove nc_path %pM -> %pM\n",
+ nc_path->prev_hop, nc_path->next_hop);
+ hlist_del_rcu(&nc_path->hash_entry);
+ batadv_nc_path_free_ref(nc_path);
+ }
+ spin_unlock_bh(lock);
+ }
+}
+
+/**
+ * batadv_nc_hash_key_gen - computes the nc_path hash key
+ * @key: buffer to hold the final hash key
+ * @src: source ethernet mac address going into the hash key
+ * @dst: destination ethernet mac address going into the hash key
+ */
+static void batadv_nc_hash_key_gen(struct batadv_nc_path *key, const char *src,
+ const char *dst)
+{
+ memcpy(key->prev_hop, src, sizeof(key->prev_hop));
+ memcpy(key->next_hop, dst, sizeof(key->next_hop));
+}
+
+/**
+ * batadv_nc_hash_choose - compute the hash value for an nc path
+ * @data: data to hash
+ * @size: size of the hash table
+ *
+ * Returns the selected index in the hash table for the given data.
+ */
+static uint32_t batadv_nc_hash_choose(const void *data, uint32_t size)
+{
+ const struct batadv_nc_path *nc_path = data;
+ uint32_t hash = 0;
+
+ hash = batadv_hash_bytes(hash, &nc_path->prev_hop,
+ sizeof(nc_path->prev_hop));
+ hash = batadv_hash_bytes(hash, &nc_path->next_hop,
+ sizeof(nc_path->next_hop));
+
+ hash += (hash << 3);
+ hash ^= (hash >> 11);
+ hash += (hash << 15);
+
+ return hash % size;
+}
+
+/**
+ * batadv_nc_hash_compare - comparing function used in the network coding hash
+ * tables
+ * @node: node in the local table
+ * @data2: second object to compare the node to
+ *
+ * Returns 1 if the two entry are the same, 0 otherwise
+ */
+static int batadv_nc_hash_compare(const struct hlist_node *node,
+ const void *data2)
+{
+ const struct batadv_nc_path *nc_path1, *nc_path2;
+
+ nc_path1 = container_of(node, struct batadv_nc_path, hash_entry);
+ nc_path2 = data2;
+
+ /* Return 1 if the two keys are identical */
+ if (memcmp(nc_path1->prev_hop, nc_path2->prev_hop,
+ sizeof(nc_path1->prev_hop)) != 0)
+ return 0;
+
+ if (memcmp(nc_path1->next_hop, nc_path2->next_hop,
+ sizeof(nc_path1->next_hop)) != 0)
+ return 0;
+
+ return 1;
+}
+
+/**
+ * batadv_nc_hash_find - search for an existing nc path and return it
+ * @hash: hash table containing the nc path
+ * @data: search key
+ *
+ * Returns the nc_path if found, NULL otherwise.
+ */
+static struct batadv_nc_path *
+batadv_nc_hash_find(struct batadv_hashtable *hash,
+ void *data)
+{
+ struct hlist_head *head;
+ struct batadv_nc_path *nc_path, *nc_path_tmp = NULL;
+ int index;
+
+ if (!hash)
+ return NULL;
+
+ index = batadv_nc_hash_choose(data, hash->size);
+ head = &hash->table[index];
+
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(nc_path, head, hash_entry) {
+ if (!batadv_nc_hash_compare(&nc_path->hash_entry, data))
+ continue;
+
+ if (!atomic_inc_not_zero(&nc_path->refcount))
+ continue;
+
+ nc_path_tmp = nc_path;
+ break;
+ }
+ rcu_read_unlock();
+
+ return nc_path_tmp;
+}
+
+/**
+ * batadv_nc_send_packet - send non-coded packet and free nc_packet struct
+ * @nc_packet: the nc packet to send
+ */
+static void batadv_nc_send_packet(struct batadv_nc_packet *nc_packet)
+{
+ batadv_send_skb_packet(nc_packet->skb,
+ nc_packet->neigh_node->if_incoming,
+ nc_packet->nc_path->next_hop);
+ nc_packet->skb = NULL;
+ batadv_nc_packet_free(nc_packet);
+}
+
+/**
+ * batadv_nc_sniffed_purge - Checks timestamp of given sniffed nc_packet.
+ * @bat_priv: the bat priv with all the soft interface information
+ * @nc_path: the nc path the packet belongs to
+ * @nc_packet: the nc packet to be checked
+ *
+ * Checks whether the given sniffed (overheard) nc_packet has hit its buffering
+ * timeout. If so, the packet is no longer kept and the entry deleted from the
+ * queue. Has to be called with the appropriate locks.
+ *
+ * Returns false as soon as the entry in the fifo queue has not been timed out
+ * yet and true otherwise.
+ */
+static bool batadv_nc_sniffed_purge(struct batadv_priv *bat_priv,
+ struct batadv_nc_path *nc_path,
+ struct batadv_nc_packet *nc_packet)
+{
+ unsigned long timeout = bat_priv->nc.max_buffer_time;
+ bool res = false;
+
+ /* Packets are added to tail, so the remaining packets did not time
+ * out and we can stop processing the current queue
+ */
+ if (atomic_read(&bat_priv->mesh_state) == BATADV_MESH_ACTIVE &&
+ !batadv_has_timed_out(nc_packet->timestamp, timeout))
+ goto out;
+
+ /* purge nc packet */
+ list_del(&nc_packet->list);
+ batadv_nc_packet_free(nc_packet);
+
+ res = true;
+
+out:
+ return res;
+}
+
+/**
+ * batadv_nc_fwd_flush - Checks the timestamp of the given nc packet.
+ * @bat_priv: the bat priv with all the soft interface information
+ * @nc_path: the nc path the packet belongs to
+ * @nc_packet: the nc packet to be checked
+ *
+ * Checks whether the given nc packet has hit its forward timeout. If so, the
+ * packet is no longer delayed, immediately sent and the entry deleted from the
+ * queue. Has to be called with the appropriate locks.
+ *
+ * Returns false as soon as the entry in the fifo queue has not been timed out
+ * yet and true otherwise.
+ */
+static bool batadv_nc_fwd_flush(struct batadv_priv *bat_priv,
+ struct batadv_nc_path *nc_path,
+ struct batadv_nc_packet *nc_packet)
+{
+ unsigned long timeout = bat_priv->nc.max_fwd_delay;
+
+ /* Packets are added to tail, so the remaining packets did not time
+ * out and we can stop processing the current queue
+ */
+ if (atomic_read(&bat_priv->mesh_state) == BATADV_MESH_ACTIVE &&
+ !batadv_has_timed_out(nc_packet->timestamp, timeout))
+ return false;
+
+ /* Send packet */
+ batadv_inc_counter(bat_priv, BATADV_CNT_FORWARD);
+ batadv_add_counter(bat_priv, BATADV_CNT_FORWARD_BYTES,
+ nc_packet->skb->len + ETH_HLEN);
+ list_del(&nc_packet->list);
+ batadv_nc_send_packet(nc_packet);
+
+ return true;
+}
+
+/**
+ * batadv_nc_process_nc_paths - traverse given nc packet pool and free timed out
+ * nc packets
+ * @bat_priv: the bat priv with all the soft interface information
+ * @hash: to be processed hash table
+ * @process_fn: Function called to process given nc packet. Should return true
+ * to encourage this function to proceed with the next packet.
+ * Otherwise the rest of the current queue is skipped.
+ */
+static void
+batadv_nc_process_nc_paths(struct batadv_priv *bat_priv,
+ struct batadv_hashtable *hash,
+ bool (*process_fn)(struct batadv_priv *,
+ struct batadv_nc_path *,
+ struct batadv_nc_packet *))
+{
+ struct hlist_head *head;
+ struct batadv_nc_packet *nc_packet, *nc_packet_tmp;
+ struct batadv_nc_path *nc_path;
+ bool ret;
+ int i;
+
+ if (!hash)
+ return;
+
+ /* Loop hash table bins */
+ for (i = 0; i < hash->size; i++) {
+ head = &hash->table[i];
+
+ /* Loop coding paths */
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(nc_path, head, hash_entry) {
+ /* Loop packets */
+ spin_lock_bh(&nc_path->packet_list_lock);
+ list_for_each_entry_safe(nc_packet, nc_packet_tmp,
+ &nc_path->packet_list, list) {
+ ret = process_fn(bat_priv, nc_path, nc_packet);
+ if (!ret)
+ break;
+ }
+ spin_unlock_bh(&nc_path->packet_list_lock);
+ }
+ rcu_read_unlock();
+ }
+}
+
+/**
+ * batadv_nc_worker - periodic task for house keeping related to network coding
+ * @work: kernel work struct
+ */
+static void batadv_nc_worker(struct work_struct *work)
+{
+ struct delayed_work *delayed_work;
+ struct batadv_priv_nc *priv_nc;
+ struct batadv_priv *bat_priv;
+ unsigned long timeout;
+
+ delayed_work = container_of(work, struct delayed_work, work);
+ priv_nc = container_of(delayed_work, struct batadv_priv_nc, work);
+ bat_priv = container_of(priv_nc, struct batadv_priv, nc);
+
+ batadv_nc_purge_orig_hash(bat_priv);
+ batadv_nc_purge_paths(bat_priv, bat_priv->nc.coding_hash,
+ batadv_nc_to_purge_nc_path_coding);
+ batadv_nc_purge_paths(bat_priv, bat_priv->nc.decoding_hash,
+ batadv_nc_to_purge_nc_path_decoding);
+
+ timeout = bat_priv->nc.max_fwd_delay;
+
+ if (batadv_has_timed_out(bat_priv->nc.timestamp_fwd_flush, timeout)) {
+ batadv_nc_process_nc_paths(bat_priv, bat_priv->nc.coding_hash,
+ batadv_nc_fwd_flush);
+ bat_priv->nc.timestamp_fwd_flush = jiffies;
+ }
+
+ if (batadv_has_timed_out(bat_priv->nc.timestamp_sniffed_purge,
+ bat_priv->nc.max_buffer_time)) {
+ batadv_nc_process_nc_paths(bat_priv, bat_priv->nc.decoding_hash,
+ batadv_nc_sniffed_purge);
+ bat_priv->nc.timestamp_sniffed_purge = jiffies;
+ }
+
+ /* Schedule a new check */
+ batadv_nc_start_timer(bat_priv);
+}
+
+/**
+ * batadv_can_nc_with_orig - checks whether the given orig node is suitable for
+ * coding or not
+ * @bat_priv: the bat priv with all the soft interface information
+ * @orig_node: neighboring orig node which may be used as nc candidate
+ * @ogm_packet: incoming ogm packet also used for the checks
+ *
+ * Returns true if:
+ * 1) The OGM must have the most recent sequence number.
+ * 2) The TTL must be decremented by one and only one.
+ * 3) The OGM must be received from the first hop from orig_node.
+ * 4) The TQ value of the OGM must be above bat_priv->nc.min_tq.
+ */
+static bool batadv_can_nc_with_orig(struct batadv_priv *bat_priv,
+ struct batadv_orig_node *orig_node,
+ struct batadv_ogm_packet *ogm_packet)
+{
+ if (orig_node->last_real_seqno != ogm_packet->seqno)
+ return false;
+ if (orig_node->last_ttl != ogm_packet->header.ttl + 1)
+ return false;
+ if (!batadv_compare_eth(ogm_packet->orig, ogm_packet->prev_sender))
+ return false;
+ if (ogm_packet->tq < bat_priv->nc.min_tq)
+ return false;
+
+ return true;
+}
+
+/**
+ * batadv_nc_find_nc_node - search for an existing nc node and return it
+ * @orig_node: orig node originating the ogm packet
+ * @orig_neigh_node: neighboring orig node from which we received the ogm packet
+ * (can be equal to orig_node)
+ * @in_coding: traverse incoming or outgoing network coding list
+ *
+ * Returns the nc_node if found, NULL otherwise.
+ */
+static struct batadv_nc_node
+*batadv_nc_find_nc_node(struct batadv_orig_node *orig_node,
+ struct batadv_orig_node *orig_neigh_node,
+ bool in_coding)
+{
+ struct batadv_nc_node *nc_node, *nc_node_out = NULL;
+ struct list_head *list;
+
+ if (in_coding)
+ list = &orig_neigh_node->in_coding_list;
+ else
+ list = &orig_neigh_node->out_coding_list;
+
+ /* Traverse list of nc_nodes to orig_node */
+ rcu_read_lock();
+ list_for_each_entry_rcu(nc_node, list, list) {
+ if (!batadv_compare_eth(nc_node->addr, orig_node->orig))
+ continue;
+
+ if (!atomic_inc_not_zero(&nc_node->refcount))
+ continue;
+
+ /* Found a match */
+ nc_node_out = nc_node;
+ break;
+ }
+ rcu_read_unlock();
+
+ return nc_node_out;
+}
+
+/**
+ * batadv_nc_get_nc_node - retrieves an nc node or creates the entry if it was
+ * not found
+ * @bat_priv: the bat priv with all the soft interface information
+ * @orig_node: orig node originating the ogm packet
+ * @orig_neigh_node: neighboring orig node from which we received the ogm packet
+ * (can be equal to orig_node)
+ * @in_coding: traverse incoming or outgoing network coding list
+ *
+ * Returns the nc_node if found or created, NULL in case of an error.
+ */
+static struct batadv_nc_node
+*batadv_nc_get_nc_node(struct batadv_priv *bat_priv,
+ struct batadv_orig_node *orig_node,
+ struct batadv_orig_node *orig_neigh_node,
+ bool in_coding)
+{
+ struct batadv_nc_node *nc_node;
+ spinlock_t *lock; /* Used to lock list selected by "int in_coding" */
+ struct list_head *list;
+
+ /* Check if nc_node is already added */
+ nc_node = batadv_nc_find_nc_node(orig_node, orig_neigh_node, in_coding);
+
+ /* Node found */
+ if (nc_node)
+ return nc_node;
+
+ nc_node = kzalloc(sizeof(*nc_node), GFP_ATOMIC);
+ if (!nc_node)
+ return NULL;
+
+ if (!atomic_inc_not_zero(&orig_neigh_node->refcount))
+ goto free;
+
+ /* Initialize nc_node */
+ INIT_LIST_HEAD(&nc_node->list);
+ memcpy(nc_node->addr, orig_node->orig, ETH_ALEN);
+ nc_node->orig_node = orig_neigh_node;
+ atomic_set(&nc_node->refcount, 2);
+
+ /* Select ingoing or outgoing coding node */
+ if (in_coding) {
+ lock = &orig_neigh_node->in_coding_list_lock;
+ list = &orig_neigh_node->in_coding_list;
+ } else {
+ lock = &orig_neigh_node->out_coding_list_lock;
+ list = &orig_neigh_node->out_coding_list;
+ }
+
+ batadv_dbg(BATADV_DBG_NC, bat_priv, "Adding nc_node %pM -> %pM\n",
+ nc_node->addr, nc_node->orig_node->orig);
+
+ /* Add nc_node to orig_node */
+ spin_lock_bh(lock);
+ list_add_tail_rcu(&nc_node->list, list);
+ spin_unlock_bh(lock);
+
+ return nc_node;
+
+free:
+ kfree(nc_node);
+ return NULL;
+}
+
+/**
+ * batadv_nc_update_nc_node - updates stored incoming and outgoing nc node structs
+ * (best called on incoming OGMs)
+ * @bat_priv: the bat priv with all the soft interface information
+ * @orig_node: orig node originating the ogm packet
+ * @orig_neigh_node: neighboring orig node from which we received the ogm packet
+ * (can be equal to orig_node)
+ * @ogm_packet: incoming ogm packet
+ * @is_single_hop_neigh: orig_node is a single hop neighbor
+ */
+void batadv_nc_update_nc_node(struct batadv_priv *bat_priv,
+ struct batadv_orig_node *orig_node,
+ struct batadv_orig_node *orig_neigh_node,
+ struct batadv_ogm_packet *ogm_packet,
+ int is_single_hop_neigh)
+{
+ struct batadv_nc_node *in_nc_node = NULL, *out_nc_node = NULL;
+
+ /* Check if network coding is enabled */
+ if (!atomic_read(&bat_priv->network_coding))
+ goto out;
+
+ /* accept ogms from 'good' neighbors and single hop neighbors */
+ if (!batadv_can_nc_with_orig(bat_priv, orig_node, ogm_packet) &&
+ !is_single_hop_neigh)
+ goto out;
+
+ /* Add orig_node as in_nc_node on hop */
+ in_nc_node = batadv_nc_get_nc_node(bat_priv, orig_node,
+ orig_neigh_node, true);
+ if (!in_nc_node)
+ goto out;
+
+ in_nc_node->last_seen = jiffies;
+
+ /* Add hop as out_nc_node on orig_node */
+ out_nc_node = batadv_nc_get_nc_node(bat_priv, orig_neigh_node,
+ orig_node, false);
+ if (!out_nc_node)
+ goto out;
+
+ out_nc_node->last_seen = jiffies;
+
+out:
+ if (in_nc_node)
+ batadv_nc_node_free_ref(in_nc_node);
+ if (out_nc_node)
+ batadv_nc_node_free_ref(out_nc_node);
+}
+
+/**
+ * batadv_nc_get_path - get existing nc_path or allocate a new one
+ * @bat_priv: the bat priv with all the soft interface information
+ * @hash: hash table containing the nc path
+ * @src: ethernet source address - first half of the nc path search key
+ * @dst: ethernet destination address - second half of the nc path search key
+ *
+ * Returns pointer to nc_path if the path was found or created, returns NULL
+ * on error.
+ */
+static struct batadv_nc_path *batadv_nc_get_path(struct batadv_priv *bat_priv,
+ struct batadv_hashtable *hash,
+ uint8_t *src,
+ uint8_t *dst)
+{
+ int hash_added;
+ struct batadv_nc_path *nc_path, nc_path_key;
+
+ batadv_nc_hash_key_gen(&nc_path_key, src, dst);
+
+ /* Search for existing nc_path */
+ nc_path = batadv_nc_hash_find(hash, (void *)&nc_path_key);
+
+ if (nc_path) {
+ /* Set timestamp to delay removal of nc_path */
+ nc_path->last_valid = jiffies;
+ return nc_path;
+ }
+
+ /* No existing nc_path was found; create a new */
+ nc_path = kzalloc(sizeof(*nc_path), GFP_ATOMIC);
+
+ if (!nc_path)
+ return NULL;
+
+ /* Initialize nc_path */
+ INIT_LIST_HEAD(&nc_path->packet_list);
+ spin_lock_init(&nc_path->packet_list_lock);
+ atomic_set(&nc_path->refcount, 2);
+ nc_path->last_valid = jiffies;
+ memcpy(nc_path->next_hop, dst, ETH_ALEN);
+ memcpy(nc_path->prev_hop, src, ETH_ALEN);
+
+ batadv_dbg(BATADV_DBG_NC, bat_priv, "Adding nc_path %pM -> %pM\n",
+ nc_path->prev_hop,
+ nc_path->next_hop);
+
+ /* Add nc_path to hash table */
+ hash_added = batadv_hash_add(hash, batadv_nc_hash_compare,
+ batadv_nc_hash_choose, &nc_path_key,
+ &nc_path->hash_entry);
+
+ if (hash_added < 0) {
+ kfree(nc_path);
+ return NULL;
+ }
+
+ return nc_path;
+}
+
+/**
+ * batadv_nc_random_weight_tq - scale the receivers TQ-value to avoid unfair
+ * selection of a receiver with slightly lower TQ than the other
+ * @tq: to be weighted tq value
+ */
+static uint8_t batadv_nc_random_weight_tq(uint8_t tq)
+{
+ uint8_t rand_val, rand_tq;
+
+ get_random_bytes(&rand_val, sizeof(rand_val));
+
+ /* randomize the estimated packet loss (max TQ - estimated TQ) */
+ rand_tq = rand_val * (BATADV_TQ_MAX_VALUE - tq);
+
+ /* normalize the randomized packet loss */
+ rand_tq /= BATADV_TQ_MAX_VALUE;
+
+ /* convert to (randomized) estimated tq again */
+ return BATADV_TQ_MAX_VALUE - rand_tq;
+}
+
+/**
+ * batadv_nc_memxor - XOR destination with source
+ * @dst: byte array to XOR into
+ * @src: byte array to XOR from
+ * @len: length of destination array
+ */
+static void batadv_nc_memxor(char *dst, const char *src, unsigned int len)
+{
+ unsigned int i;
+
+ for (i = 0; i < len; ++i)
+ dst[i] ^= src[i];
+}
+
+/**
+ * batadv_nc_code_packets - code a received unicast_packet with an nc packet
+ * into a coded_packet and send it
+ * @bat_priv: the bat priv with all the soft interface information
+ * @skb: data skb to forward
+ * @ethhdr: pointer to the ethernet header inside the skb
+ * @nc_packet: structure containing the packet to the skb can be coded with
+ * @neigh_node: next hop to forward packet to
+ *
+ * Returns true if both packets are consumed, false otherwise.
+ */
+static bool batadv_nc_code_packets(struct batadv_priv *bat_priv,
+ struct sk_buff *skb,
+ struct ethhdr *ethhdr,
+ struct batadv_nc_packet *nc_packet,
+ struct batadv_neigh_node *neigh_node)
+{
+ uint8_t tq_weighted_neigh, tq_weighted_coding;
+ struct sk_buff *skb_dest, *skb_src;
+ struct batadv_unicast_packet *packet1;
+ struct batadv_unicast_packet *packet2;
+ struct batadv_coded_packet *coded_packet;
+ struct batadv_neigh_node *neigh_tmp, *router_neigh;
+ struct batadv_neigh_node *router_coding = NULL;
+ uint8_t *first_source, *first_dest, *second_source, *second_dest;
+ __be32 packet_id1, packet_id2;
+ size_t count;
+ bool res = false;
+ int coding_len;
+ int unicast_size = sizeof(*packet1);
+ int coded_size = sizeof(*coded_packet);
+ int header_add = coded_size - unicast_size;
+
+ router_neigh = batadv_orig_node_get_router(neigh_node->orig_node);
+ if (!router_neigh)
+ goto out;
+
+ neigh_tmp = nc_packet->neigh_node;
+ router_coding = batadv_orig_node_get_router(neigh_tmp->orig_node);
+ if (!router_coding)
+ goto out;
+
+ tq_weighted_neigh = batadv_nc_random_weight_tq(router_neigh->tq_avg);
+ tq_weighted_coding = batadv_nc_random_weight_tq(router_coding->tq_avg);
+
+ /* Select one destination for the MAC-header dst-field based on
+ * weighted TQ-values.
+ */
+ if (tq_weighted_neigh >= tq_weighted_coding) {
+ /* Destination from nc_packet is selected for MAC-header */
+ first_dest = nc_packet->nc_path->next_hop;
+ first_source = nc_packet->nc_path->prev_hop;
+ second_dest = neigh_node->addr;
+ second_source = ethhdr->h_source;
+ packet1 = (struct batadv_unicast_packet *)nc_packet->skb->data;
+ packet2 = (struct batadv_unicast_packet *)skb->data;
+ packet_id1 = nc_packet->packet_id;
+ packet_id2 = batadv_skb_crc32(skb,
+ skb->data + sizeof(*packet2));
+ } else {
+ /* Destination for skb is selected for MAC-header */
+ first_dest = neigh_node->addr;
+ first_source = ethhdr->h_source;
+ second_dest = nc_packet->nc_path->next_hop;
+ second_source = nc_packet->nc_path->prev_hop;
+ packet1 = (struct batadv_unicast_packet *)skb->data;
+ packet2 = (struct batadv_unicast_packet *)nc_packet->skb->data;
+ packet_id1 = batadv_skb_crc32(skb,
+ skb->data + sizeof(*packet1));
+ packet_id2 = nc_packet->packet_id;
+ }
+
+ /* Instead of zero padding the smallest data buffer, we
+ * code into the largest.
+ */
+ if (skb->len <= nc_packet->skb->len) {
+ skb_dest = nc_packet->skb;
+ skb_src = skb;
+ } else {
+ skb_dest = skb;
+ skb_src = nc_packet->skb;
+ }
+
+ /* coding_len is used when decoding the packet shorter packet */
+ coding_len = skb_src->len - unicast_size;
+
+ if (skb_linearize(skb_dest) < 0 || skb_linearize(skb_src) < 0)
+ goto out;
+
+ skb_push(skb_dest, header_add);
+
+ coded_packet = (struct batadv_coded_packet *)skb_dest->data;
+ skb_reset_mac_header(skb_dest);
+
+ coded_packet->header.packet_type = BATADV_CODED;
+ coded_packet->header.version = BATADV_COMPAT_VERSION;
+ coded_packet->header.ttl = packet1->header.ttl;
+
+ /* Info about first unicast packet */
+ memcpy(coded_packet->first_source, first_source, ETH_ALEN);
+ memcpy(coded_packet->first_orig_dest, packet1->dest, ETH_ALEN);
+ coded_packet->first_crc = packet_id1;
+ coded_packet->first_ttvn = packet1->ttvn;
+
+ /* Info about second unicast packet */
+ memcpy(coded_packet->second_dest, second_dest, ETH_ALEN);
+ memcpy(coded_packet->second_source, second_source, ETH_ALEN);
+ memcpy(coded_packet->second_orig_dest, packet2->dest, ETH_ALEN);
+ coded_packet->second_crc = packet_id2;
+ coded_packet->second_ttl = packet2->header.ttl;
+ coded_packet->second_ttvn = packet2->ttvn;
+ coded_packet->coded_len = htons(coding_len);
+
+ /* This is where the magic happens: Code skb_src into skb_dest */
+ batadv_nc_memxor(skb_dest->data + coded_size,
+ skb_src->data + unicast_size, coding_len);
+
+ /* Update counters accordingly */
+ if (BATADV_SKB_CB(skb_src)->decoded &&
+ BATADV_SKB_CB(skb_dest)->decoded) {
+ /* Both packets are recoded */
+ count = skb_src->len + ETH_HLEN;
+ count += skb_dest->len + ETH_HLEN;
+ batadv_add_counter(bat_priv, BATADV_CNT_NC_RECODE, 2);
+ batadv_add_counter(bat_priv, BATADV_CNT_NC_RECODE_BYTES, count);
+ } else if (!BATADV_SKB_CB(skb_src)->decoded &&
+ !BATADV_SKB_CB(skb_dest)->decoded) {
+ /* Both packets are newly coded */
+ count = skb_src->len + ETH_HLEN;
+ count += skb_dest->len + ETH_HLEN;
+ batadv_add_counter(bat_priv, BATADV_CNT_NC_CODE, 2);
+ batadv_add_counter(bat_priv, BATADV_CNT_NC_CODE_BYTES, count);
+ } else if (BATADV_SKB_CB(skb_src)->decoded &&
+ !BATADV_SKB_CB(skb_dest)->decoded) {
+ /* skb_src recoded and skb_dest is newly coded */
+ batadv_inc_counter(bat_priv, BATADV_CNT_NC_RECODE);
+ batadv_add_counter(bat_priv, BATADV_CNT_NC_RECODE_BYTES,
+ skb_src->len + ETH_HLEN);
+ batadv_inc_counter(bat_priv, BATADV_CNT_NC_CODE);
+ batadv_add_counter(bat_priv, BATADV_CNT_NC_CODE_BYTES,
+ skb_dest->len + ETH_HLEN);
+ } else if (!BATADV_SKB_CB(skb_src)->decoded &&
+ BATADV_SKB_CB(skb_dest)->decoded) {
+ /* skb_src is newly coded and skb_dest is recoded */
+ batadv_inc_counter(bat_priv, BATADV_CNT_NC_CODE);
+ batadv_add_counter(bat_priv, BATADV_CNT_NC_CODE_BYTES,
+ skb_src->len + ETH_HLEN);
+ batadv_inc_counter(bat_priv, BATADV_CNT_NC_RECODE);
+ batadv_add_counter(bat_priv, BATADV_CNT_NC_RECODE_BYTES,
+ skb_dest->len + ETH_HLEN);
+ }
+
+ /* skb_src is now coded into skb_dest, so free it */
+ kfree_skb(skb_src);
+
+ /* avoid duplicate free of skb from nc_packet */
+ nc_packet->skb = NULL;
+ batadv_nc_packet_free(nc_packet);
+
+ /* Send the coded packet and return true */
+ batadv_send_skb_packet(skb_dest, neigh_node->if_incoming, first_dest);
+ res = true;
+out:
+ if (router_neigh)
+ batadv_neigh_node_free_ref(router_neigh);
+ if (router_coding)
+ batadv_neigh_node_free_ref(router_coding);
+ return res;
+}
+
+/**
+ * batadv_nc_skb_coding_possible - true if a decoded skb is available at dst.
+ * @skb: data skb to forward
+ * @dst: destination mac address of the other skb to code with
+ * @src: source mac address of skb
+ *
+ * Whenever we network code a packet we have to check whether we received it in
+ * a network coded form. If so, we may not be able to use it for coding because
+ * some neighbors may also have received (overheard) the packet in the network
+ * coded form without being able to decode it. It is hard to know which of the
+ * neighboring nodes was able to decode the packet, therefore we can only
+ * re-code the packet if the source of the previous encoded packet is involved.
+ * Since the source encoded the packet we can be certain it has all necessary
+ * decode information.
+ *
+ * Returns true if coding of a decoded packet is allowed.
+ */
+static bool batadv_nc_skb_coding_possible(struct sk_buff *skb,
+ uint8_t *dst, uint8_t *src)
+{
+ if (BATADV_SKB_CB(skb)->decoded && !batadv_compare_eth(dst, src))
+ return false;
+ else
+ return true;
+}
+
+/**
+ * batadv_nc_path_search - Find the coding path matching in_nc_node and
+ * out_nc_node to retrieve a buffered packet that can be used for coding.
+ * @bat_priv: the bat priv with all the soft interface information
+ * @in_nc_node: pointer to skb next hop's neighbor nc node
+ * @out_nc_node: pointer to skb source's neighbor nc node
+ * @skb: data skb to forward
+ * @eth_dst: next hop mac address of skb
+ *
+ * Returns true if coding of a decoded skb is allowed.
+ */
+static struct batadv_nc_packet *
+batadv_nc_path_search(struct batadv_priv *bat_priv,
+ struct batadv_nc_node *in_nc_node,
+ struct batadv_nc_node *out_nc_node,
+ struct sk_buff *skb,
+ uint8_t *eth_dst)
+{
+ struct batadv_nc_path *nc_path, nc_path_key;
+ struct batadv_nc_packet *nc_packet_out = NULL;
+ struct batadv_nc_packet *nc_packet, *nc_packet_tmp;
+ struct batadv_hashtable *hash = bat_priv->nc.coding_hash;
+ int idx;
+
+ if (!hash)
+ return NULL;
+
+ /* Create almost path key */
+ batadv_nc_hash_key_gen(&nc_path_key, in_nc_node->addr,
+ out_nc_node->addr);
+ idx = batadv_nc_hash_choose(&nc_path_key, hash->size);
+
+ /* Check for coding opportunities in this nc_path */
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(nc_path, &hash->table[idx], hash_entry) {
+ if (!batadv_compare_eth(nc_path->prev_hop, in_nc_node->addr))
+ continue;
+
+ if (!batadv_compare_eth(nc_path->next_hop, out_nc_node->addr))
+ continue;
+
+ spin_lock_bh(&nc_path->packet_list_lock);
+ if (list_empty(&nc_path->packet_list)) {
+ spin_unlock_bh(&nc_path->packet_list_lock);
+ continue;
+ }
+
+ list_for_each_entry_safe(nc_packet, nc_packet_tmp,
+ &nc_path->packet_list, list) {
+ if (!batadv_nc_skb_coding_possible(nc_packet->skb,
+ eth_dst,
+ in_nc_node->addr))
+ continue;
+
+ /* Coding opportunity is found! */
+ list_del(&nc_packet->list);
+ nc_packet_out = nc_packet;
+ break;
+ }
+
+ spin_unlock_bh(&nc_path->packet_list_lock);
+ break;
+ }
+ rcu_read_unlock();
+
+ return nc_packet_out;
+}
+
+/**
+ * batadv_nc_skb_src_search - Loops through the list of neighoring nodes of the
+ * skb's sender (may be equal to the originator).
+ * @bat_priv: the bat priv with all the soft interface information
+ * @skb: data skb to forward
+ * @eth_dst: next hop mac address of skb
+ * @eth_src: source mac address of skb
+ * @in_nc_node: pointer to skb next hop's neighbor nc node
+ *
+ * Returns an nc packet if a suitable coding packet was found, NULL otherwise.
+ */
+static struct batadv_nc_packet *
+batadv_nc_skb_src_search(struct batadv_priv *bat_priv,
+ struct sk_buff *skb,
+ uint8_t *eth_dst,
+ uint8_t *eth_src,
+ struct batadv_nc_node *in_nc_node)
+{
+ struct batadv_orig_node *orig_node;
+ struct batadv_nc_node *out_nc_node;
+ struct batadv_nc_packet *nc_packet = NULL;
+
+ orig_node = batadv_orig_hash_find(bat_priv, eth_src);
+ if (!orig_node)
+ return NULL;
+
+ rcu_read_lock();
+ list_for_each_entry_rcu(out_nc_node,
+ &orig_node->out_coding_list, list) {
+ /* Check if the skb is decoded and if recoding is possible */
+ if (!batadv_nc_skb_coding_possible(skb,
+ out_nc_node->addr, eth_src))
+ continue;
+
+ /* Search for an opportunity in this nc_path */
+ nc_packet = batadv_nc_path_search(bat_priv, in_nc_node,
+ out_nc_node, skb, eth_dst);
+ if (nc_packet)
+ break;
+ }
+ rcu_read_unlock();
+
+ batadv_orig_node_free_ref(orig_node);
+ return nc_packet;
+}
+
+/**
+ * batadv_nc_skb_store_before_coding - set the ethernet src and dst of the
+ * unicast skb before it is stored for use in later decoding
+ * @bat_priv: the bat priv with all the soft interface information
+ * @skb: data skb to store
+ * @eth_dst_new: new destination mac address of skb
+ */
+static void batadv_nc_skb_store_before_coding(struct batadv_priv *bat_priv,
+ struct sk_buff *skb,
+ uint8_t *eth_dst_new)
+{
+ struct ethhdr *ethhdr;
+
+ /* Copy skb header to change the mac header */
+ skb = pskb_copy(skb, GFP_ATOMIC);
+ if (!skb)
+ return;
+
+ /* Set the mac header as if we actually sent the packet uncoded */
+ ethhdr = (struct ethhdr *)skb_mac_header(skb);
+ memcpy(ethhdr->h_source, ethhdr->h_dest, ETH_ALEN);
+ memcpy(ethhdr->h_dest, eth_dst_new, ETH_ALEN);
+
+ /* Set data pointer to MAC header to mimic packets from our tx path */
+ skb_push(skb, ETH_HLEN);
+
+ /* Add the packet to the decoding packet pool */
+ batadv_nc_skb_store_for_decoding(bat_priv, skb);
+
+ /* batadv_nc_skb_store_for_decoding() clones the skb, so we must free
+ * our ref
+ */
+ kfree_skb(skb);
+}
+
+/**
+ * batadv_nc_skb_dst_search - Loops through list of neighboring nodes to dst.
+ * @skb: data skb to forward
+ * @neigh_node: next hop to forward packet to
+ * @ethhdr: pointer to the ethernet header inside the skb
+ *
+ * Loops through list of neighboring nodes the next hop has a good connection to
+ * (receives OGMs with a sufficient quality). We need to find a neighbor of our
+ * next hop that potentially sent a packet which our next hop also received
+ * (overheard) and has stored for later decoding.
+ *
+ * Returns true if the skb was consumed (encoded packet sent) or false otherwise
+ */
+static bool batadv_nc_skb_dst_search(struct sk_buff *skb,
+ struct batadv_neigh_node *neigh_node,
+ struct ethhdr *ethhdr)
+{
+ struct net_device *netdev = neigh_node->if_incoming->soft_iface;
+ struct batadv_priv *bat_priv = netdev_priv(netdev);
+ struct batadv_orig_node *orig_node = neigh_node->orig_node;
+ struct batadv_nc_node *nc_node;
+ struct batadv_nc_packet *nc_packet = NULL;
+
+ rcu_read_lock();
+ list_for_each_entry_rcu(nc_node, &orig_node->in_coding_list, list) {
+ /* Search for coding opportunity with this in_nc_node */
+ nc_packet = batadv_nc_skb_src_search(bat_priv, skb,
+ neigh_node->addr,
+ ethhdr->h_source, nc_node);
+
+ /* Opportunity was found, so stop searching */
+ if (nc_packet)
+ break;
+ }
+ rcu_read_unlock();
+
+ if (!nc_packet)
+ return false;
+
+ /* Save packets for later decoding */
+ batadv_nc_skb_store_before_coding(bat_priv, skb,
+ neigh_node->addr);
+ batadv_nc_skb_store_before_coding(bat_priv, nc_packet->skb,
+ nc_packet->neigh_node->addr);
+
+ /* Code and send packets */
+ if (batadv_nc_code_packets(bat_priv, skb, ethhdr, nc_packet,
+ neigh_node))
+ return true;
+
+ /* out of mem ? Coding failed - we have to free the buffered packet
+ * to avoid memleaks. The skb passed as argument will be dealt with
+ * by the calling function.
+ */
+ batadv_nc_send_packet(nc_packet);
+ return false;
+}
+
+/**
+ * batadv_nc_skb_add_to_path - buffer skb for later encoding / decoding
+ * @skb: skb to add to path
+ * @nc_path: path to add skb to
+ * @neigh_node: next hop to forward packet to
+ * @packet_id: checksum to identify packet
+ *
+ * Returns true if the packet was buffered or false in case of an error.
+ */
+static bool batadv_nc_skb_add_to_path(struct sk_buff *skb,
+ struct batadv_nc_path *nc_path,
+ struct batadv_neigh_node *neigh_node,
+ __be32 packet_id)
+{
+ struct batadv_nc_packet *nc_packet;
+
+ nc_packet = kzalloc(sizeof(*nc_packet), GFP_ATOMIC);
+ if (!nc_packet)
+ return false;
+
+ /* Initialize nc_packet */
+ nc_packet->timestamp = jiffies;
+ nc_packet->packet_id = packet_id;
+ nc_packet->skb = skb;
+ nc_packet->neigh_node = neigh_node;
+ nc_packet->nc_path = nc_path;
+
+ /* Add coding packet to list */
+ spin_lock_bh(&nc_path->packet_list_lock);
+ list_add_tail(&nc_packet->list, &nc_path->packet_list);
+ spin_unlock_bh(&nc_path->packet_list_lock);
+
+ return true;
+}
+
+/**
+ * batadv_nc_skb_forward - try to code a packet or add it to the coding packet
+ * buffer
+ * @skb: data skb to forward
+ * @neigh_node: next hop to forward packet to
+ * @ethhdr: pointer to the ethernet header inside the skb
+ *
+ * Returns true if the skb was consumed (encoded packet sent) or false otherwise
+ */
+bool batadv_nc_skb_forward(struct sk_buff *skb,
+ struct batadv_neigh_node *neigh_node,
+ struct ethhdr *ethhdr)
+{
+ const struct net_device *netdev = neigh_node->if_incoming->soft_iface;
+ struct batadv_priv *bat_priv = netdev_priv(netdev);
+ struct batadv_unicast_packet *packet;
+ struct batadv_nc_path *nc_path;
+ __be32 packet_id;
+ u8 *payload;
+
+ /* Check if network coding is enabled */
+ if (!atomic_read(&bat_priv->network_coding))
+ goto out;
+
+ /* We only handle unicast packets */
+ payload = skb_network_header(skb);
+ packet = (struct batadv_unicast_packet *)payload;
+ if (packet->header.packet_type != BATADV_UNICAST)
+ goto out;
+
+ /* Try to find a coding opportunity and send the skb if one is found */
+ if (batadv_nc_skb_dst_search(skb, neigh_node, ethhdr))
+ return true;
+
+ /* Find or create a nc_path for this src-dst pair */
+ nc_path = batadv_nc_get_path(bat_priv,
+ bat_priv->nc.coding_hash,
+ ethhdr->h_source,
+ neigh_node->addr);
+
+ if (!nc_path)
+ goto out;
+
+ /* Add skb to nc_path */
+ packet_id = batadv_skb_crc32(skb, payload + sizeof(*packet));
+ if (!batadv_nc_skb_add_to_path(skb, nc_path, neigh_node, packet_id))
+ goto free_nc_path;
+
+ /* Packet is consumed */
+ return true;
+
+free_nc_path:
+ batadv_nc_path_free_ref(nc_path);
+out:
+ /* Packet is not consumed */
+ return false;
+}
+
+/**
+ * batadv_nc_skb_store_for_decoding - save a clone of the skb which can be used
+ * when decoding coded packets
+ * @bat_priv: the bat priv with all the soft interface information
+ * @skb: data skb to store
+ */
+void batadv_nc_skb_store_for_decoding(struct batadv_priv *bat_priv,
+ struct sk_buff *skb)
+{
+ struct batadv_unicast_packet *packet;
+ struct batadv_nc_path *nc_path;
+ struct ethhdr *ethhdr = (struct ethhdr *)skb_mac_header(skb);
+ __be32 packet_id;
+ u8 *payload;
+
+ /* Check if network coding is enabled */
+ if (!atomic_read(&bat_priv->network_coding))
+ goto out;
+
+ /* Check for supported packet type */
+ payload = skb_network_header(skb);
+ packet = (struct batadv_unicast_packet *)payload;
+ if (packet->header.packet_type != BATADV_UNICAST)
+ goto out;
+
+ /* Find existing nc_path or create a new */
+ nc_path = batadv_nc_get_path(bat_priv,
+ bat_priv->nc.decoding_hash,
+ ethhdr->h_source,
+ ethhdr->h_dest);
+
+ if (!nc_path)
+ goto out;
+
+ /* Clone skb and adjust skb->data to point at batman header */
+ skb = skb_clone(skb, GFP_ATOMIC);
+ if (unlikely(!skb))
+ goto free_nc_path;
+
+ if (unlikely(!pskb_may_pull(skb, ETH_HLEN)))
+ goto free_skb;
+
+ if (unlikely(!skb_pull_rcsum(skb, ETH_HLEN)))
+ goto free_skb;
+
+ /* Add skb to nc_path */
+ packet_id = batadv_skb_crc32(skb, payload + sizeof(*packet));
+ if (!batadv_nc_skb_add_to_path(skb, nc_path, NULL, packet_id))
+ goto free_skb;
+
+ batadv_inc_counter(bat_priv, BATADV_CNT_NC_BUFFER);
+ return;
+
+free_skb:
+ kfree_skb(skb);
+free_nc_path:
+ batadv_nc_path_free_ref(nc_path);
+out:
+ return;
+}
+
+/**
+ * batadv_nc_skb_store_sniffed_unicast - check if a received unicast packet
+ * should be saved in the decoding buffer and, if so, store it there
+ * @bat_priv: the bat priv with all the soft interface information
+ * @skb: unicast skb to store
+ */
+void batadv_nc_skb_store_sniffed_unicast(struct batadv_priv *bat_priv,
+ struct sk_buff *skb)
+{
+ struct ethhdr *ethhdr = (struct ethhdr *)skb_mac_header(skb);
+
+ if (batadv_is_my_mac(ethhdr->h_dest))
+ return;
+
+ /* Set data pointer to MAC header to mimic packets from our tx path */
+ skb_push(skb, ETH_HLEN);
+
+ batadv_nc_skb_store_for_decoding(bat_priv, skb);
+}
+
+/**
+ * batadv_nc_skb_decode_packet - decode given skb using the decode data stored
+ * in nc_packet
+ * @skb: unicast skb to decode
+ * @nc_packet: decode data needed to decode the skb
+ *
+ * Returns pointer to decoded unicast packet if the packet was decoded or NULL
+ * in case of an error.
+ */
+static struct batadv_unicast_packet *
+batadv_nc_skb_decode_packet(struct sk_buff *skb,
+ struct batadv_nc_packet *nc_packet)
+{
+ const int h_size = sizeof(struct batadv_unicast_packet);
+ const int h_diff = sizeof(struct batadv_coded_packet) - h_size;
+ struct batadv_unicast_packet *unicast_packet;
+ struct batadv_coded_packet coded_packet_tmp;
+ struct ethhdr *ethhdr, ethhdr_tmp;
+ uint8_t *orig_dest, ttl, ttvn;
+ unsigned int coding_len;
+
+ /* Save headers temporarily */
+ memcpy(&coded_packet_tmp, skb->data, sizeof(coded_packet_tmp));
+ memcpy(ðhdr_tmp, skb_mac_header(skb), sizeof(ethhdr_tmp));
+
+ if (skb_cow(skb, 0) < 0)
+ return NULL;
+
+ if (unlikely(!skb_pull_rcsum(skb, h_diff)))
+ return NULL;
+
+ /* Data points to batman header, so set mac header 14 bytes before
+ * and network to data
+ */
+ skb_set_mac_header(skb, -ETH_HLEN);
+ skb_reset_network_header(skb);
+
+ /* Reconstruct original mac header */
+ ethhdr = (struct ethhdr *)skb_mac_header(skb);
+ memcpy(ethhdr, ðhdr_tmp, sizeof(*ethhdr));
+
+ /* Select the correct unicast header information based on the location
+ * of our mac address in the coded_packet header
+ */
+ if (batadv_is_my_mac(coded_packet_tmp.second_dest)) {
+ /* If we are the second destination the packet was overheard,
+ * so the Ethernet address must be copied to h_dest and
+ * pkt_type changed from PACKET_OTHERHOST to PACKET_HOST
+ */
+ memcpy(ethhdr->h_dest, coded_packet_tmp.second_dest, ETH_ALEN);
+ skb->pkt_type = PACKET_HOST;
+
+ orig_dest = coded_packet_tmp.second_orig_dest;
+ ttl = coded_packet_tmp.second_ttl;
+ ttvn = coded_packet_tmp.second_ttvn;
+ } else {
+ orig_dest = coded_packet_tmp.first_orig_dest;
+ ttl = coded_packet_tmp.header.ttl;
+ ttvn = coded_packet_tmp.first_ttvn;
+ }
+
+ coding_len = ntohs(coded_packet_tmp.coded_len);
+
+ if (coding_len > skb->len)
+ return NULL;
+
+ /* Here the magic is reversed:
+ * extract the missing packet from the received coded packet
+ */
+ batadv_nc_memxor(skb->data + h_size,
+ nc_packet->skb->data + h_size,
+ coding_len);
+
+ /* Resize decoded skb if decoded with larger packet */
+ if (nc_packet->skb->len > coding_len + h_size)
+ pskb_trim_rcsum(skb, coding_len + h_size);
+
+ /* Create decoded unicast packet */
+ unicast_packet = (struct batadv_unicast_packet *)skb->data;
+ unicast_packet->header.packet_type = BATADV_UNICAST;
+ unicast_packet->header.version = BATADV_COMPAT_VERSION;
+ unicast_packet->header.ttl = ttl;
+ memcpy(unicast_packet->dest, orig_dest, ETH_ALEN);
+ unicast_packet->ttvn = ttvn;
+
+ batadv_nc_packet_free(nc_packet);
+ return unicast_packet;
+}
+
+/**
+ * batadv_nc_find_decoding_packet - search through buffered decoding data to
+ * find the data needed to decode the coded packet
+ * @bat_priv: the bat priv with all the soft interface information
+ * @ethhdr: pointer to the ethernet header inside the coded packet
+ * @coded: coded packet we try to find decode data for
+ *
+ * Returns pointer to nc packet if the needed data was found or NULL otherwise.
+ */
+static struct batadv_nc_packet *
+batadv_nc_find_decoding_packet(struct batadv_priv *bat_priv,
+ struct ethhdr *ethhdr,
+ struct batadv_coded_packet *coded)
+{
+ struct batadv_hashtable *hash = bat_priv->nc.decoding_hash;
+ struct batadv_nc_packet *tmp_nc_packet, *nc_packet = NULL;
+ struct batadv_nc_path *nc_path, nc_path_key;
+ uint8_t *dest, *source;
+ __be32 packet_id;
+ int index;
+
+ if (!hash)
+ return NULL;
+
+ /* Select the correct packet id based on the location of our mac-addr */
+ dest = ethhdr->h_source;
+ if (!batadv_is_my_mac(coded->second_dest)) {
+ source = coded->second_source;
+ packet_id = coded->second_crc;
+ } else {
+ source = coded->first_source;
+ packet_id = coded->first_crc;
+ }
+
+ batadv_nc_hash_key_gen(&nc_path_key, source, dest);
+ index = batadv_nc_hash_choose(&nc_path_key, hash->size);
+
+ /* Search for matching coding path */
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(nc_path, &hash->table[index], hash_entry) {
+ /* Find matching nc_packet */
+ spin_lock_bh(&nc_path->packet_list_lock);
+ list_for_each_entry(tmp_nc_packet,
+ &nc_path->packet_list, list) {
+ if (packet_id == tmp_nc_packet->packet_id) {
+ list_del(&tmp_nc_packet->list);
+
+ nc_packet = tmp_nc_packet;
+ break;
+ }
+ }
+ spin_unlock_bh(&nc_path->packet_list_lock);
+
+ if (nc_packet)
+ break;
+ }
+ rcu_read_unlock();
+
+ if (!nc_packet)
+ batadv_dbg(BATADV_DBG_NC, bat_priv,
+ "No decoding packet found for %u\n", packet_id);
+
+ return nc_packet;
+}
+
+/**
+ * batadv_nc_recv_coded_packet - try to decode coded packet and enqueue the
+ * resulting unicast packet
+ * @skb: incoming coded packet
+ * @recv_if: pointer to interface this packet was received on
+ */
+static int batadv_nc_recv_coded_packet(struct sk_buff *skb,
+ struct batadv_hard_iface *recv_if)
+{
+ struct batadv_priv *bat_priv = netdev_priv(recv_if->soft_iface);
+ struct batadv_unicast_packet *unicast_packet;
+ struct batadv_coded_packet *coded_packet;
+ struct batadv_nc_packet *nc_packet;
+ struct ethhdr *ethhdr;
+ int hdr_size = sizeof(*coded_packet);
+
+ /* Check if network coding is enabled */
+ if (!atomic_read(&bat_priv->network_coding))
+ return NET_RX_DROP;
+
+ /* Make sure we can access (and remove) header */
+ if (unlikely(!pskb_may_pull(skb, hdr_size)))
+ return NET_RX_DROP;
+
+ coded_packet = (struct batadv_coded_packet *)skb->data;
+ ethhdr = (struct ethhdr *)skb_mac_header(skb);
+
+ /* Verify frame is destined for us */
+ if (!batadv_is_my_mac(ethhdr->h_dest) &&
+ !batadv_is_my_mac(coded_packet->second_dest))
+ return NET_RX_DROP;
+
+ /* Update stat counter */
+ if (batadv_is_my_mac(coded_packet->second_dest))
+ batadv_inc_counter(bat_priv, BATADV_CNT_NC_SNIFFED);
+
+ nc_packet = batadv_nc_find_decoding_packet(bat_priv, ethhdr,
+ coded_packet);
+ if (!nc_packet) {
+ batadv_inc_counter(bat_priv, BATADV_CNT_NC_DECODE_FAILED);
+ return NET_RX_DROP;
+ }
+
+ /* Make skb's linear, because decoding accesses the entire buffer */
+ if (skb_linearize(skb) < 0)
+ goto free_nc_packet;
+
+ if (skb_linearize(nc_packet->skb) < 0)
+ goto free_nc_packet;
+
+ /* Decode the packet */
+ unicast_packet = batadv_nc_skb_decode_packet(skb, nc_packet);
+ if (!unicast_packet) {
+ batadv_inc_counter(bat_priv, BATADV_CNT_NC_DECODE_FAILED);
+ goto free_nc_packet;
+ }
+
+ /* Mark packet as decoded to do correct recoding when forwarding */
+ BATADV_SKB_CB(skb)->decoded = true;
+ batadv_inc_counter(bat_priv, BATADV_CNT_NC_DECODE);
+ batadv_add_counter(bat_priv, BATADV_CNT_NC_DECODE_BYTES,
+ skb->len + ETH_HLEN);
+ return batadv_recv_unicast_packet(skb, recv_if);
+
+free_nc_packet:
+ batadv_nc_packet_free(nc_packet);
+ return NET_RX_DROP;
+}
+
+/**
+ * batadv_nc_free - clean up network coding memory
+ * @bat_priv: the bat priv with all the soft interface information
+ */
+void batadv_nc_free(struct batadv_priv *bat_priv)
+{
+ batadv_recv_handler_unregister(BATADV_CODED);
+ cancel_delayed_work_sync(&bat_priv->nc.work);
+
+ batadv_nc_purge_paths(bat_priv, bat_priv->nc.coding_hash, NULL);
+ batadv_hash_destroy(bat_priv->nc.coding_hash);
+ batadv_nc_purge_paths(bat_priv, bat_priv->nc.decoding_hash, NULL);
+ batadv_hash_destroy(bat_priv->nc.decoding_hash);
+}
+
+/**
+ * batadv_nc_nodes_seq_print_text - print the nc node information
+ * @seq: seq file to print on
+ * @offset: not used
+ */
+int batadv_nc_nodes_seq_print_text(struct seq_file *seq, void *offset)
+{
+ struct net_device *net_dev = (struct net_device *)seq->private;
+ struct batadv_priv *bat_priv = netdev_priv(net_dev);
+ struct batadv_hashtable *hash = bat_priv->orig_hash;
+ struct batadv_hard_iface *primary_if;
+ struct hlist_head *head;
+ struct batadv_orig_node *orig_node;
+ struct batadv_nc_node *nc_node;
+ int i;
+
+ primary_if = batadv_seq_print_text_primary_if_get(seq);
+ if (!primary_if)
+ goto out;
+
+ /* Traverse list of originators */
+ for (i = 0; i < hash->size; i++) {
+ head = &hash->table[i];
+
+ /* For each orig_node in this bin */
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(orig_node, head, hash_entry) {
+ seq_printf(seq, "Node: %pM\n", orig_node->orig);
+
+ seq_printf(seq, " Ingoing: ");
+ /* For each in_nc_node to this orig_node */
+ list_for_each_entry_rcu(nc_node,
+ &orig_node->in_coding_list,
+ list)
+ seq_printf(seq, "%pM ",
+ nc_node->addr);
+ seq_printf(seq, "\n");
+
+ seq_printf(seq, " Outgoing: ");
+ /* For out_nc_node to this orig_node */
+ list_for_each_entry_rcu(nc_node,
+ &orig_node->out_coding_list,
+ list)
+ seq_printf(seq, "%pM ",
+ nc_node->addr);
+ seq_printf(seq, "\n\n");
+ }
+ rcu_read_unlock();
+ }
+
+out:
+ if (primary_if)
+ batadv_hardif_free_ref(primary_if);
+ return 0;
+}
+
+/**
+ * batadv_nc_init_debugfs - create nc folder and related files in debugfs
+ * @bat_priv: the bat priv with all the soft interface information
+ */
+int batadv_nc_init_debugfs(struct batadv_priv *bat_priv)
+{
+ struct dentry *nc_dir, *file;
+
+ nc_dir = debugfs_create_dir("nc", bat_priv->debug_dir);
+ if (!nc_dir)
+ goto out;
+
+ file = debugfs_create_u8("min_tq", S_IRUGO | S_IWUSR, nc_dir,
+ &bat_priv->nc.min_tq);
+ if (!file)
+ goto out;
+
+ file = debugfs_create_u32("max_fwd_delay", S_IRUGO | S_IWUSR, nc_dir,
+ &bat_priv->nc.max_fwd_delay);
+ if (!file)
+ goto out;
+
+ file = debugfs_create_u32("max_buffer_time", S_IRUGO | S_IWUSR, nc_dir,
+ &bat_priv->nc.max_buffer_time);
+ if (!file)
+ goto out;
+
+ return 0;
+
+out:
+ return -ENOMEM;
+}
diff --git a/net/batman-adv/network-coding.h b/net/batman-adv/network-coding.h
new file mode 100644
index 0000000..4fa6d0c
--- /dev/null
+++ b/net/batman-adv/network-coding.h
@@ -0,0 +1,123 @@
+/* Copyright (C) 2012-2013 B.A.T.M.A.N. contributors:
+ *
+ * Martin Hundebøll, Jeppe Ledet-Pedersen
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA
+ */
+
+#ifndef _NET_BATMAN_ADV_NETWORK_CODING_H_
+#define _NET_BATMAN_ADV_NETWORK_CODING_H_
+
+#ifdef CONFIG_BATMAN_ADV_NC
+
+int batadv_nc_init(struct batadv_priv *bat_priv);
+void batadv_nc_free(struct batadv_priv *bat_priv);
+void batadv_nc_update_nc_node(struct batadv_priv *bat_priv,
+ struct batadv_orig_node *orig_node,
+ struct batadv_orig_node *orig_neigh_node,
+ struct batadv_ogm_packet *ogm_packet,
+ int is_single_hop_neigh);
+void batadv_nc_purge_orig(struct batadv_priv *bat_priv,
+ struct batadv_orig_node *orig_node,
+ bool (*to_purge)(struct batadv_priv *,
+ struct batadv_nc_node *));
+void batadv_nc_init_bat_priv(struct batadv_priv *bat_priv);
+void batadv_nc_init_orig(struct batadv_orig_node *orig_node);
+bool batadv_nc_skb_forward(struct sk_buff *skb,
+ struct batadv_neigh_node *neigh_node,
+ struct ethhdr *ethhdr);
+void batadv_nc_skb_store_for_decoding(struct batadv_priv *bat_priv,
+ struct sk_buff *skb);
+void batadv_nc_skb_store_sniffed_unicast(struct batadv_priv *bat_priv,
+ struct sk_buff *skb);
+int batadv_nc_nodes_seq_print_text(struct seq_file *seq, void *offset);
+int batadv_nc_init_debugfs(struct batadv_priv *bat_priv);
+
+#else /* ifdef CONFIG_BATMAN_ADV_NC */
+
+static inline int batadv_nc_init(struct batadv_priv *bat_priv)
+{
+ return 0;
+}
+
+static inline void batadv_nc_free(struct batadv_priv *bat_priv)
+{
+ return;
+}
+
+static inline void
+batadv_nc_update_nc_node(struct batadv_priv *bat_priv,
+ struct batadv_orig_node *orig_node,
+ struct batadv_orig_node *orig_neigh_node,
+ struct batadv_ogm_packet *ogm_packet,
+ int is_single_hop_neigh)
+{
+ return;
+}
+
+static inline void
+batadv_nc_purge_orig(struct batadv_priv *bat_priv,
+ struct batadv_orig_node *orig_node,
+ bool (*to_purge)(struct batadv_priv *,
+ struct batadv_nc_node *))
+{
+ return;
+}
+
+static inline void batadv_nc_init_bat_priv(struct batadv_priv *bat_priv)
+{
+ return;
+}
+
+static inline void batadv_nc_init_orig(struct batadv_orig_node *orig_node)
+{
+ return;
+}
+
+static inline bool batadv_nc_skb_forward(struct sk_buff *skb,
+ struct batadv_neigh_node *neigh_node,
+ struct ethhdr *ethhdr)
+{
+ return false;
+}
+
+static inline void
+batadv_nc_skb_store_for_decoding(struct batadv_priv *bat_priv,
+ struct sk_buff *skb)
+{
+ return;
+}
+
+static inline void
+batadv_nc_skb_store_sniffed_unicast(struct batadv_priv *bat_priv,
+ struct sk_buff *skb)
+{
+ return;
+}
+
+static inline int batadv_nc_nodes_seq_print_text(struct seq_file *seq,
+ void *offset)
+{
+ return 0;
+}
+
+static inline int batadv_nc_init_debugfs(struct batadv_priv *bat_priv)
+{
+ return 0;
+}
+
+#endif /* ifdef CONFIG_BATMAN_ADV_NC */
+
+#endif /* _NET_BATMAN_ADV_NETWORK_CODING_H_ */
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index 96fb80b..585e684 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -28,6 +28,7 @@
#include "unicast.h"
#include "soft-interface.h"
#include "bridge_loop_avoidance.h"
+#include "network-coding.h"
/* hash class keys */
static struct lock_class_key batadv_orig_hash_lock_class_key;
@@ -142,6 +143,9 @@
spin_unlock_bh(&orig_node->neigh_list_lock);
+ /* Free nc_nodes */
+ batadv_nc_purge_orig(orig_node->bat_priv, orig_node, NULL);
+
batadv_frag_list_free(&orig_node->frag_list);
batadv_tt_global_del_orig(orig_node->bat_priv, orig_node,
"originator timed out");
@@ -219,6 +223,8 @@
spin_lock_init(&orig_node->neigh_list_lock);
spin_lock_init(&orig_node->tt_buff_lock);
+ batadv_nc_init_orig(orig_node);
+
/* extra reference for return */
atomic_set(&orig_node->refcount, 2);
diff --git a/net/batman-adv/packet.h b/net/batman-adv/packet.h
index ed0aa89..a079958 100644
--- a/net/batman-adv/packet.h
+++ b/net/batman-adv/packet.h
@@ -30,6 +30,7 @@
BATADV_TT_QUERY = 0x07,
BATADV_ROAM_ADV = 0x08,
BATADV_UNICAST_4ADDR = 0x09,
+ BATADV_CODED = 0x0a,
};
/**
@@ -278,4 +279,36 @@
uint8_t addr[ETH_ALEN];
} __packed;
+/**
+ * struct batadv_coded_packet - network coded packet
+ * @header: common batman packet header and ttl of first included packet
+ * @reserved: Align following fields to 2-byte boundaries
+ * @first_source: original source of first included packet
+ * @first_orig_dest: original destinal of first included packet
+ * @first_crc: checksum of first included packet
+ * @first_ttvn: tt-version number of first included packet
+ * @second_ttl: ttl of second packet
+ * @second_dest: second receiver of this coded packet
+ * @second_source: original source of second included packet
+ * @second_orig_dest: original destination of second included packet
+ * @second_crc: checksum of second included packet
+ * @second_ttvn: tt version number of second included packet
+ * @coded_len: length of network coded part of the payload
+ */
+struct batadv_coded_packet {
+ struct batadv_header header;
+ uint8_t first_ttvn;
+ /* uint8_t first_dest[ETH_ALEN]; - saved in mac header destination */
+ uint8_t first_source[ETH_ALEN];
+ uint8_t first_orig_dest[ETH_ALEN];
+ __be32 first_crc;
+ uint8_t second_ttl;
+ uint8_t second_ttvn;
+ uint8_t second_dest[ETH_ALEN];
+ uint8_t second_source[ETH_ALEN];
+ uint8_t second_orig_dest[ETH_ALEN];
+ __be32 second_crc;
+ uint16_t coded_len;
+};
+
#endif /* _NET_BATMAN_ADV_PACKET_H_ */
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index 5ee21ce..8f88967 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -29,6 +29,7 @@
#include "unicast.h"
#include "bridge_loop_avoidance.h"
#include "distributed-arp-table.h"
+#include "network-coding.h"
static int batadv_route_unicast_packet(struct sk_buff *skb,
struct batadv_hard_iface *recv_if);
@@ -548,27 +549,37 @@
return router;
}
+/**
+ * batadv_check_unicast_packet - Check for malformed unicast packets
+ * @skb: packet to check
+ * @hdr_size: size of header to pull
+ *
+ * Check for short header and bad addresses in given packet. Returns negative
+ * value when check fails and 0 otherwise. The negative value depends on the
+ * reason: -ENODATA for bad header, -EBADR for broadcast destination or source,
+ * and -EREMOTE for non-local (other host) destination.
+ */
static int batadv_check_unicast_packet(struct sk_buff *skb, int hdr_size)
{
struct ethhdr *ethhdr;
/* drop packet if it has not necessary minimum size */
if (unlikely(!pskb_may_pull(skb, hdr_size)))
- return -1;
+ return -ENODATA;
ethhdr = (struct ethhdr *)skb_mac_header(skb);
/* packet with unicast indication but broadcast recipient */
if (is_broadcast_ether_addr(ethhdr->h_dest))
- return -1;
+ return -EBADR;
/* packet with broadcast sender address */
if (is_broadcast_ether_addr(ethhdr->h_source))
- return -1;
+ return -EBADR;
/* not for me */
if (!batadv_is_my_mac(ethhdr->h_dest))
- return -1;
+ return -EREMOTE;
return 0;
}
@@ -850,14 +861,17 @@
/* decrement ttl */
unicast_packet->header.ttl--;
- /* Update stats counter */
- batadv_inc_counter(bat_priv, BATADV_CNT_FORWARD);
- batadv_add_counter(bat_priv, BATADV_CNT_FORWARD_BYTES,
- skb->len + ETH_HLEN);
-
- /* route it */
- if (batadv_send_skb_to_orig(skb, orig_node, recv_if))
+ /* network code packet if possible */
+ if (batadv_nc_skb_forward(skb, neigh_node, ethhdr)) {
ret = NET_RX_SUCCESS;
+ } else if (batadv_send_skb_to_orig(skb, orig_node, recv_if)) {
+ ret = NET_RX_SUCCESS;
+
+ /* Update stats counter */
+ batadv_inc_counter(bat_priv, BATADV_CNT_FORWARD);
+ batadv_add_counter(bat_priv, BATADV_CNT_FORWARD_BYTES,
+ skb->len + ETH_HLEN);
+ }
out:
if (neigh_node)
@@ -1033,7 +1047,7 @@
struct batadv_unicast_4addr_packet *unicast_4addr_packet;
uint8_t *orig_addr;
struct batadv_orig_node *orig_node = NULL;
- int hdr_size = sizeof(*unicast_packet);
+ int check, hdr_size = sizeof(*unicast_packet);
bool is4addr;
unicast_packet = (struct batadv_unicast_packet *)skb->data;
@@ -1044,7 +1058,16 @@
if (is4addr)
hdr_size = sizeof(*unicast_4addr_packet);
- if (batadv_check_unicast_packet(skb, hdr_size) < 0)
+ /* function returns -EREMOTE for promiscuous packets */
+ check = batadv_check_unicast_packet(skb, hdr_size);
+
+ /* Even though the packet is not for us, we might save it to use for
+ * decoding a later received coded packet
+ */
+ if (check == -EREMOTE)
+ batadv_nc_skb_store_sniffed_unicast(bat_priv, skb);
+
+ if (check < 0)
return NET_RX_DROP;
if (!batadv_check_unicast_ttvn(bat_priv, skb))
diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c
index a67cffd..263cfd1 100644
--- a/net/batman-adv/send.c
+++ b/net/batman-adv/send.c
@@ -27,6 +27,7 @@
#include "vis.h"
#include "gateway_common.h"
#include "originator.h"
+#include "network-coding.h"
#include <linux/if_ether.h>
@@ -39,6 +40,7 @@
struct batadv_hard_iface *hard_iface,
const uint8_t *dst_addr)
{
+ struct batadv_priv *bat_priv = netdev_priv(hard_iface->soft_iface);
struct ethhdr *ethhdr;
if (hard_iface->if_status != BATADV_IF_ACTIVE)
@@ -70,6 +72,9 @@
skb->dev = hard_iface->net_dev;
+ /* Save a clone of the skb to use when decoding coded packets */
+ batadv_nc_skb_store_for_decoding(bat_priv, skb);
+
/* dev_queue_xmit() returns a negative result on error. However on
* congestion and traffic shaping, it drops and returns NET_XMIT_DROP
* (which is > 0). This will not be treated as an error.
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index 2711e87..f93ae42 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -37,6 +37,7 @@
#include <linux/if_ether.h>
#include "unicast.h"
#include "bridge_loop_avoidance.h"
+#include "network-coding.h"
static int batadv_get_settings(struct net_device *dev, struct ethtool_cmd *cmd);
@@ -544,6 +545,8 @@
if (ret < 0)
goto unreg_soft_iface;
+ batadv_nc_init_bat_priv(bat_priv);
+
ret = batadv_sysfs_add_meshif(soft_iface);
if (ret < 0)
goto unreg_soft_iface;
@@ -662,6 +665,17 @@
{ "dat_put_rx" },
{ "dat_cached_reply_tx" },
#endif
+#ifdef CONFIG_BATMAN_ADV_NC
+ { "nc_code" },
+ { "nc_code_bytes" },
+ { "nc_recode" },
+ { "nc_recode_bytes" },
+ { "nc_buffer" },
+ { "nc_decode" },
+ { "nc_decode_bytes" },
+ { "nc_decode_failed" },
+ { "nc_sniffed" },
+#endif
};
static void batadv_get_strings(struct net_device *dev, uint32_t stringset,
diff --git a/net/batman-adv/sysfs.c b/net/batman-adv/sysfs.c
index afbba31..ce39f62 100644
--- a/net/batman-adv/sysfs.c
+++ b/net/batman-adv/sysfs.c
@@ -442,6 +442,9 @@
#ifdef CONFIG_BATMAN_ADV_DEBUG
BATADV_ATTR_SIF_UINT(log_level, S_IRUGO | S_IWUSR, 0, BATADV_DBG_ALL, NULL);
#endif
+#ifdef CONFIG_BATMAN_ADV_NC
+BATADV_ATTR_SIF_BOOL(network_coding, S_IRUGO | S_IWUSR, NULL);
+#endif
static struct batadv_attribute *batadv_mesh_attrs[] = {
&batadv_attr_aggregated_ogms,
@@ -464,6 +467,9 @@
#ifdef CONFIG_BATMAN_ADV_DEBUG
&batadv_attr_log_level,
#endif
+#ifdef CONFIG_BATMAN_ADV_NC
+ &batadv_attr_network_coding,
+#endif
NULL,
};
@@ -688,15 +694,10 @@
enum batadv_uev_action action, const char *data)
{
int ret = -ENOMEM;
- struct batadv_hard_iface *primary_if;
struct kobject *bat_kobj;
char *uevent_env[4] = { NULL, NULL, NULL, NULL };
- primary_if = batadv_primary_if_get_selected(bat_priv);
- if (!primary_if)
- goto out;
-
- bat_kobj = &primary_if->soft_iface->dev.kobj;
+ bat_kobj = &bat_priv->soft_iface->dev.kobj;
uevent_env[0] = kmalloc(strlen(BATADV_UEV_TYPE_VAR) +
strlen(batadv_uev_type_str[type]) + 1,
@@ -732,9 +733,6 @@
kfree(uevent_env[1]);
kfree(uevent_env[2]);
- if (primary_if)
- batadv_hardif_free_ref(primary_if);
-
if (ret)
batadv_dbg(BATADV_DBG_BATMAN, bat_priv,
"Impossible to send uevent for (%s,%s,%s) event (err: %d)\n",
diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index 98a66a0..9322320 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -385,25 +385,19 @@
int *packet_buff_len,
int min_packet_len)
{
- struct batadv_hard_iface *primary_if;
int req_len;
- primary_if = batadv_primary_if_get_selected(bat_priv);
-
req_len = min_packet_len;
req_len += batadv_tt_len(atomic_read(&bat_priv->tt.local_changes));
/* if we have too many changes for one packet don't send any
* and wait for the tt table request which will be fragmented
*/
- if ((!primary_if) || (req_len > primary_if->soft_iface->mtu))
+ if (req_len > bat_priv->soft_iface->mtu)
req_len = min_packet_len;
batadv_tt_realloc_packet_buff(packet_buff, packet_buff_len,
min_packet_len, req_len);
-
- if (primary_if)
- batadv_hardif_free_ref(primary_if);
}
static int batadv_tt_changes_fill_buff(struct batadv_priv *bat_priv,
@@ -908,7 +902,7 @@
/* remove address from local hash if present */
local_flags = batadv_tt_local_remove(bat_priv, tt_addr,
"global tt received",
- !!(flags & BATADV_TT_CLIENT_ROAM));
+ flags & BATADV_TT_CLIENT_ROAM);
tt_global_entry->common.flags |= local_flags & BATADV_TT_CLIENT_WIFI;
if (!(flags & BATADV_TT_CLIENT_ROAM))
@@ -1580,7 +1574,7 @@
static struct sk_buff *
batadv_tt_response_fill_table(uint16_t tt_len, uint8_t ttvn,
struct batadv_hashtable *hash,
- struct batadv_hard_iface *primary_if,
+ struct batadv_priv *bat_priv,
int (*valid_cb)(const void *, const void *),
void *cb_data)
{
@@ -1594,8 +1588,8 @@
uint32_t i;
size_t len;
- if (tt_query_size + tt_len > primary_if->soft_iface->mtu) {
- tt_len = primary_if->soft_iface->mtu - tt_query_size;
+ if (tt_query_size + tt_len > bat_priv->soft_iface->mtu) {
+ tt_len = bat_priv->soft_iface->mtu - tt_query_size;
tt_len -= tt_len % sizeof(struct batadv_tt_change);
}
tt_tot = tt_len / sizeof(struct batadv_tt_change);
@@ -1715,7 +1709,6 @@
{
struct batadv_orig_node *req_dst_orig_node;
struct batadv_orig_node *res_dst_orig_node = NULL;
- struct batadv_hard_iface *primary_if = NULL;
uint8_t orig_ttvn, req_ttvn, ttvn;
int ret = false;
unsigned char *tt_buff;
@@ -1740,10 +1733,6 @@
if (!res_dst_orig_node)
goto out;
- primary_if = batadv_primary_if_get_selected(bat_priv);
- if (!primary_if)
- goto out;
-
orig_ttvn = (uint8_t)atomic_read(&req_dst_orig_node->last_ttvn);
req_ttvn = tt_request->ttvn;
@@ -1791,7 +1780,7 @@
skb = batadv_tt_response_fill_table(tt_len, ttvn,
bat_priv->tt.global_hash,
- primary_if,
+ bat_priv,
batadv_tt_global_valid,
req_dst_orig_node);
if (!skb)
@@ -1828,8 +1817,6 @@
batadv_orig_node_free_ref(res_dst_orig_node);
if (req_dst_orig_node)
batadv_orig_node_free_ref(req_dst_orig_node);
- if (primary_if)
- batadv_hardif_free_ref(primary_if);
if (!ret)
kfree_skb(skb);
return ret;
@@ -1907,7 +1894,7 @@
skb = batadv_tt_response_fill_table(tt_len, ttvn,
bat_priv->tt.local_hash,
- primary_if,
+ bat_priv,
batadv_tt_local_valid_entry,
NULL);
if (!skb)
@@ -2528,7 +2515,7 @@
if (!tt_global_entry)
goto out;
- ret = !!(tt_global_entry->common.flags & BATADV_TT_CLIENT_ROAM);
+ ret = tt_global_entry->common.flags & BATADV_TT_CLIENT_ROAM;
batadv_tt_global_entry_free_ref(tt_global_entry);
out:
return ret;
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 4cd87a0..aba8364 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -128,6 +128,10 @@
* @bond_list: list of bonding candidates
* @refcount: number of contexts the object is used
* @rcu: struct used for freeing in an RCU-safe manner
+ * @in_coding_list: list of nodes this orig can hear
+ * @out_coding_list: list of nodes that can hear this orig
+ * @in_coding_list_lock: protects in_coding_list
+ * @out_coding_list_lock: protects out_coding_list
*/
struct batadv_orig_node {
uint8_t orig[ETH_ALEN];
@@ -171,6 +175,12 @@
struct list_head bond_list;
atomic_t refcount;
struct rcu_head rcu;
+#ifdef CONFIG_BATMAN_ADV_NC
+ struct list_head in_coding_list;
+ struct list_head out_coding_list;
+ spinlock_t in_coding_list_lock; /* Protects in_coding_list */
+ spinlock_t out_coding_list_lock; /* Protects out_coding_list */
+#endif
};
/**
@@ -265,6 +275,17 @@
* @BATADV_CNT_DAT_PUT_RX: received dht PUT traffic packet counter
* @BATADV_CNT_DAT_CACHED_REPLY_TX: transmitted dat cache reply traffic packet
* counter
+ * @BATADV_CNT_NC_CODE: transmitted nc-combined traffic packet counter
+ * @BATADV_CNT_NC_CODE_BYTES: transmitted nc-combined traffic bytes counter
+ * @BATADV_CNT_NC_RECODE: transmitted nc-recombined traffic packet counter
+ * @BATADV_CNT_NC_RECODE_BYTES: transmitted nc-recombined traffic bytes counter
+ * @BATADV_CNT_NC_BUFFER: counter for packets buffered for later nc decoding
+ * @BATADV_CNT_NC_DECODE: received and nc-decoded traffic packet counter
+ * @BATADV_CNT_NC_DECODE_BYTES: received and nc-decoded traffic bytes counter
+ * @BATADV_CNT_NC_DECODE_FAILED: received and decode-failed traffic packet
+ * counter
+ * @BATADV_CNT_NC_SNIFFED: counter for nc-decoded packets received in promisc
+ * mode.
* @BATADV_CNT_NUM: number of traffic counters
*/
enum batadv_counters {
@@ -292,6 +313,17 @@
BATADV_CNT_DAT_PUT_RX,
BATADV_CNT_DAT_CACHED_REPLY_TX,
#endif
+#ifdef CONFIG_BATMAN_ADV_NC
+ BATADV_CNT_NC_CODE,
+ BATADV_CNT_NC_CODE_BYTES,
+ BATADV_CNT_NC_RECODE,
+ BATADV_CNT_NC_RECODE_BYTES,
+ BATADV_CNT_NC_BUFFER,
+ BATADV_CNT_NC_DECODE,
+ BATADV_CNT_NC_DECODE_BYTES,
+ BATADV_CNT_NC_DECODE_FAILED,
+ BATADV_CNT_NC_SNIFFED,
+#endif
BATADV_CNT_NUM,
};
@@ -428,6 +460,35 @@
#endif
/**
+ * struct batadv_priv_nc - per mesh interface network coding private data
+ * @work: work queue callback item for cleanup
+ * @debug_dir: dentry for nc subdir in batman-adv directory in debugfs
+ * @min_tq: only consider neighbors for encoding if neigh_tq > min_tq
+ * @max_fwd_delay: maximum packet forward delay to allow coding of packets
+ * @max_buffer_time: buffer time for sniffed packets used to decoding
+ * @timestamp_fwd_flush: timestamp of last forward packet queue flush
+ * @timestamp_sniffed_purge: timestamp of last sniffed packet queue purge
+ * @coding_hash: Hash table used to buffer skbs while waiting for another
+ * incoming skb to code it with. Skbs are added to the buffer just before being
+ * forwarded in routing.c
+ * @decoding_hash: Hash table used to buffer skbs that might be needed to decode
+ * a received coded skb. The buffer is used for 1) skbs arriving on the
+ * soft-interface; 2) skbs overheard on the hard-interface; and 3) skbs
+ * forwarded by batman-adv.
+ */
+struct batadv_priv_nc {
+ struct delayed_work work;
+ struct dentry *debug_dir;
+ u8 min_tq;
+ u32 max_fwd_delay;
+ u32 max_buffer_time;
+ unsigned long timestamp_fwd_flush;
+ unsigned long timestamp_sniffed_purge;
+ struct batadv_hashtable *coding_hash;
+ struct batadv_hashtable *decoding_hash;
+};
+
+/**
* struct batadv_priv - per mesh interface data
* @mesh_state: current status of the mesh (inactive/active/deactivating)
* @soft_iface: net device which holds this struct as private data
@@ -470,6 +531,8 @@
* @tt: translation table data
* @vis: vis data
* @dat: distributed arp table data
+ * @network_coding: bool indicating whether network coding is enabled
+ * @batadv_priv_nc: network coding data
*/
struct batadv_priv {
atomic_t mesh_state;
@@ -522,6 +585,10 @@
#ifdef CONFIG_BATMAN_ADV_DAT
struct batadv_priv_dat dat;
#endif
+#ifdef CONFIG_BATMAN_ADV_NC
+ atomic_t network_coding;
+ struct batadv_priv_nc nc;
+#endif /* CONFIG_BATMAN_ADV_NC */
};
/**
@@ -702,6 +769,75 @@
};
/**
+ * struct batadv_nc_node - network coding node
+ * @list: next and prev pointer for the list handling
+ * @addr: the node's mac address
+ * @refcount: number of contexts the object is used by
+ * @rcu: struct used for freeing in an RCU-safe manner
+ * @orig_node: pointer to corresponding orig node struct
+ * @last_seen: timestamp of last ogm received from this node
+ */
+struct batadv_nc_node {
+ struct list_head list;
+ uint8_t addr[ETH_ALEN];
+ atomic_t refcount;
+ struct rcu_head rcu;
+ struct batadv_orig_node *orig_node;
+ unsigned long last_seen;
+};
+
+/**
+ * struct batadv_nc_path - network coding path
+ * @hash_entry: next and prev pointer for the list handling
+ * @rcu: struct used for freeing in an RCU-safe manner
+ * @refcount: number of contexts the object is used by
+ * @packet_list: list of buffered packets for this path
+ * @packet_list_lock: access lock for packet list
+ * @next_hop: next hop (destination) of path
+ * @prev_hop: previous hop (source) of path
+ * @last_valid: timestamp for last validation of path
+ */
+struct batadv_nc_path {
+ struct hlist_node hash_entry;
+ struct rcu_head rcu;
+ atomic_t refcount;
+ struct list_head packet_list;
+ spinlock_t packet_list_lock; /* Protects packet_list */
+ uint8_t next_hop[ETH_ALEN];
+ uint8_t prev_hop[ETH_ALEN];
+ unsigned long last_valid;
+};
+
+/**
+ * struct batadv_nc_packet - network coding packet used when coding and
+ * decoding packets
+ * @list: next and prev pointer for the list handling
+ * @packet_id: crc32 checksum of skb data
+ * @timestamp: field containing the info when the packet was added to path
+ * @neigh_node: pointer to original next hop neighbor of skb
+ * @skb: skb which can be encoded or used for decoding
+ * @nc_path: pointer to path this nc packet is attached to
+ */
+struct batadv_nc_packet {
+ struct list_head list;
+ __be32 packet_id;
+ unsigned long timestamp;
+ struct batadv_neigh_node *neigh_node;
+ struct sk_buff *skb;
+ struct batadv_nc_path *nc_path;
+};
+
+/**
+ * batadv_skb_cb - control buffer structure used to store private data relevant
+ * to batman-adv in the skb->cb buffer in skbs.
+ * @decoded: Marks a skb as decoded, which is checked when searching for coding
+ * opportunities in network-coding.c
+ */
+struct batadv_skb_cb {
+ bool decoded;
+};
+
+/**
* struct batadv_forw_packet - structure for bcast packets to be sent/forwarded
* @list: list node for batadv_socket_client::queue_list
* @send_time: execution time for delayed_work (packet sending)
diff --git a/net/batman-adv/unicast.c b/net/batman-adv/unicast.c
index 50e079f..0bb3b59 100644
--- a/net/batman-adv/unicast.c
+++ b/net/batman-adv/unicast.c
@@ -122,7 +122,7 @@
{
struct batadv_frag_packet_list_entry *tfp;
struct batadv_unicast_frag_packet *tmp_up = NULL;
- int is_head_tmp, is_head;
+ bool is_head_tmp, is_head;
uint16_t search_seqno;
if (up->flags & BATADV_UNI_FRAG_HEAD)
@@ -130,7 +130,7 @@
else
search_seqno = ntohs(up->seqno)-1;
- is_head = !!(up->flags & BATADV_UNI_FRAG_HEAD);
+ is_head = up->flags & BATADV_UNI_FRAG_HEAD;
list_for_each_entry(tfp, head, list) {
if (!tfp->skb)
@@ -142,7 +142,7 @@
tmp_up = (struct batadv_unicast_frag_packet *)tfp->skb->data;
if (tfp->seqno == search_seqno) {
- is_head_tmp = !!(tmp_up->flags & BATADV_UNI_FRAG_HEAD);
+ is_head_tmp = tmp_up->flags & BATADV_UNI_FRAG_HEAD;
if (is_head_tmp != is_head)
return tfp;
else
diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c
index b0812c9..10b47d4 100644
--- a/net/bridge/br_fdb.c
+++ b/net/bridge/br_fdb.c
@@ -161,9 +161,7 @@
if (!pv)
return;
- for (vid = find_next_bit(pv->vlan_bitmap, BR_VLAN_BITMAP_LEN, vid);
- vid < BR_VLAN_BITMAP_LEN;
- vid = find_next_bit(pv->vlan_bitmap, BR_VLAN_BITMAP_LEN, vid+1)) {
+ for_each_set_bit_from(vid, pv->vlan_bitmap, BR_VLAN_BITMAP_LEN) {
f = __br_fdb_get(br, br->dev->dev_addr, vid);
if (f && f->is_local && !f->dst)
fdb_delete(br, f);
@@ -724,13 +722,10 @@
* specify a VLAN. To be nice, add/update entry for every
* vlan on this port.
*/
- vid = find_first_bit(pv->vlan_bitmap, BR_VLAN_BITMAP_LEN);
- while (vid < BR_VLAN_BITMAP_LEN) {
+ for_each_set_bit(vid, pv->vlan_bitmap, BR_VLAN_BITMAP_LEN) {
err = __br_fdb_add(ndm, p, addr, nlh_flags, vid);
if (err)
goto out;
- vid = find_next_bit(pv->vlan_bitmap,
- BR_VLAN_BITMAP_LEN, vid+1);
}
}
@@ -815,11 +810,8 @@
* vlan on this port.
*/
err = -ENOENT;
- vid = find_first_bit(pv->vlan_bitmap, BR_VLAN_BITMAP_LEN);
- while (vid < BR_VLAN_BITMAP_LEN) {
+ for_each_set_bit(vid, pv->vlan_bitmap, BR_VLAN_BITMAP_LEN) {
err &= __br_fdb_delete(p, addr, vid);
- vid = find_next_bit(pv->vlan_bitmap,
- BR_VLAN_BITMAP_LEN, vid+1);
}
}
out:
diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c
index 923fbea..81f2389 100644
--- a/net/bridge/br_multicast.c
+++ b/net/bridge/br_multicast.c
@@ -1369,7 +1369,7 @@
return -EINVAL;
if (iph->protocol != IPPROTO_IGMP) {
- if ((iph->daddr & IGMP_LOCAL_GROUP_MASK) != IGMP_LOCAL_GROUP)
+ if (!ipv4_is_local_multicast(iph->daddr))
BR_INPUT_SKB_CB(skb)->mrouters_only = 1;
return 0;
}
diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c
index 299fc5f..84c3b7d 100644
--- a/net/bridge/br_netlink.c
+++ b/net/bridge/br_netlink.c
@@ -136,10 +136,7 @@
goto nla_put_failure;
pvid = br_get_pvid(pv);
- for (vid = find_first_bit(pv->vlan_bitmap, BR_VLAN_BITMAP_LEN);
- vid < BR_VLAN_BITMAP_LEN;
- vid = find_next_bit(pv->vlan_bitmap,
- BR_VLAN_BITMAP_LEN, vid+1)) {
+ for_each_set_bit(vid, pv->vlan_bitmap, BR_VLAN_BITMAP_LEN) {
vinfo.vid = vid;
vinfo.flags = 0;
if (vid == pvid)
diff --git a/net/caif/caif_dev.c b/net/caif/caif_dev.c
index 21760f0..df6d56d 100644
--- a/net/caif/caif_dev.c
+++ b/net/caif/caif_dev.c
@@ -301,10 +301,11 @@
}
void caif_enroll_dev(struct net_device *dev, struct caif_dev_common *caifdev,
- struct cflayer *link_support, int head_room,
- struct cflayer **layer, int (**rcv_func)(
- struct sk_buff *, struct net_device *,
- struct packet_type *, struct net_device *))
+ struct cflayer *link_support, int head_room,
+ struct cflayer **layer,
+ int (**rcv_func)(struct sk_buff *, struct net_device *,
+ struct packet_type *,
+ struct net_device *))
{
struct caif_device_entry *caifd;
enum cfcnfg_phy_preference pref;
diff --git a/net/caif/caif_socket.c b/net/caif/caif_socket.c
index 095259f..1d337e0 100644
--- a/net/caif/caif_socket.c
+++ b/net/caif/caif_socket.c
@@ -197,8 +197,8 @@
/* Packet Control Callback function called from CAIF */
static void caif_ctrl_cb(struct cflayer *layr,
- enum caif_ctrlcmd flow,
- int phyid)
+ enum caif_ctrlcmd flow,
+ int phyid)
{
struct caifsock *cf_sk = container_of(layr, struct caifsock, layer);
switch (flow) {
@@ -274,7 +274,7 @@
* changed locking, address handling and added MSG_TRUNC.
*/
static int caif_seqpkt_recvmsg(struct kiocb *iocb, struct socket *sock,
- struct msghdr *m, size_t len, int flags)
+ struct msghdr *m, size_t len, int flags)
{
struct sock *sk = sock->sk;
@@ -346,8 +346,8 @@
* changed locking calls, changed address handling.
*/
static int caif_stream_recvmsg(struct kiocb *iocb, struct socket *sock,
- struct msghdr *msg, size_t size,
- int flags)
+ struct msghdr *msg, size_t size,
+ int flags)
{
struct sock *sk = sock->sk;
int copied = 0;
@@ -462,7 +462,7 @@
* CAIF flow-on and sock_writable.
*/
static long caif_wait_for_flow_on(struct caifsock *cf_sk,
- int wait_writeable, long timeo, int *err)
+ int wait_writeable, long timeo, int *err)
{
struct sock *sk = &cf_sk->sk;
DEFINE_WAIT(wait);
@@ -516,7 +516,7 @@
/* Copied from af_unix:unix_dgram_sendmsg, and adapted to CAIF */
static int caif_seqpkt_sendmsg(struct kiocb *kiocb, struct socket *sock,
- struct msghdr *msg, size_t len)
+ struct msghdr *msg, size_t len)
{
struct sock *sk = sock->sk;
struct caifsock *cf_sk = container_of(sk, struct caifsock, sk);
@@ -591,7 +591,7 @@
* and other minor adaptations.
*/
static int caif_stream_sendmsg(struct kiocb *kiocb, struct socket *sock,
- struct msghdr *msg, size_t len)
+ struct msghdr *msg, size_t len)
{
struct sock *sk = sock->sk;
struct caifsock *cf_sk = container_of(sk, struct caifsock, sk);
@@ -670,7 +670,7 @@
}
static int setsockopt(struct socket *sock,
- int lvl, int opt, char __user *ov, unsigned int ol)
+ int lvl, int opt, char __user *ov, unsigned int ol)
{
struct sock *sk = sock->sk;
struct caifsock *cf_sk = container_of(sk, struct caifsock, sk);
@@ -932,7 +932,7 @@
/* Copied from af_unix.c:unix_poll(), added CAIF tx_flow handling */
static unsigned int caif_poll(struct file *file,
- struct socket *sock, poll_table *wait)
+ struct socket *sock, poll_table *wait)
{
struct sock *sk = sock->sk;
unsigned int mask;
@@ -1022,7 +1022,7 @@
}
static int caif_create(struct net *net, struct socket *sock, int protocol,
- int kern)
+ int kern)
{
struct sock *sk = NULL;
struct caifsock *cf_sk = NULL;
diff --git a/net/caif/caif_usb.c b/net/caif/caif_usb.c
index ef8ebaa..d76278d 100644
--- a/net/caif/caif_usb.c
+++ b/net/caif/caif_usb.c
@@ -75,7 +75,7 @@
}
static void cfusbl_ctrlcmd(struct cflayer *layr, enum caif_ctrlcmd ctrl,
- int phyid)
+ int phyid)
{
if (layr->up && layr->up->ctrlcmd)
layr->up->ctrlcmd(layr->up, ctrl, layr->id);
@@ -121,7 +121,7 @@
};
static int cfusbl_device_notify(struct notifier_block *me, unsigned long what,
- void *arg)
+ void *arg)
{
struct net_device *dev = arg;
struct caif_dev_common common;
diff --git a/net/caif/cfcnfg.c b/net/caif/cfcnfg.c
index f1dbddb..246ac3a 100644
--- a/net/caif/cfcnfg.c
+++ b/net/caif/cfcnfg.c
@@ -61,11 +61,11 @@
};
static void cfcnfg_linkup_rsp(struct cflayer *layer, u8 channel_id,
- enum cfctrl_srv serv, u8 phyid,
- struct cflayer *adapt_layer);
+ enum cfctrl_srv serv, u8 phyid,
+ struct cflayer *adapt_layer);
static void cfcnfg_linkdestroy_rsp(struct cflayer *layer, u8 channel_id);
static void cfcnfg_reject_rsp(struct cflayer *layer, u8 channel_id,
- struct cflayer *adapt_layer);
+ struct cflayer *adapt_layer);
static void cfctrl_resp_func(void);
static void cfctrl_enum_resp(void);
@@ -131,7 +131,7 @@
}
static struct cfcnfg_phyinfo *cfcnfg_get_phyinfo_rcu(struct cfcnfg *cnfg,
- u8 phyid)
+ u8 phyid)
{
struct cfcnfg_phyinfo *phy;
@@ -216,8 +216,8 @@
static int caif_connect_req_to_link_param(struct cfcnfg *cnfg,
- struct caif_connect_request *s,
- struct cfctrl_link_param *l)
+ struct caif_connect_request *s,
+ struct cfctrl_link_param *l)
{
struct dev_info *dev_info;
enum cfcnfg_phy_preference pref;
@@ -301,8 +301,7 @@
int caif_connect_client(struct net *net, struct caif_connect_request *conn_req,
struct cflayer *adap_layer, int *ifindex,
- int *proto_head,
- int *proto_tail)
+ int *proto_head, int *proto_tail)
{
struct cflayer *frml;
struct cfcnfg_phyinfo *phy;
@@ -364,7 +363,7 @@
EXPORT_SYMBOL(caif_connect_client);
static void cfcnfg_reject_rsp(struct cflayer *layer, u8 channel_id,
- struct cflayer *adapt_layer)
+ struct cflayer *adapt_layer)
{
if (adapt_layer != NULL && adapt_layer->ctrlcmd != NULL)
adapt_layer->ctrlcmd(adapt_layer,
@@ -526,7 +525,7 @@
EXPORT_SYMBOL(cfcnfg_add_phy_layer);
int cfcnfg_set_phy_state(struct cfcnfg *cnfg, struct cflayer *phy_layer,
- bool up)
+ bool up)
{
struct cfcnfg_phyinfo *phyinfo;
diff --git a/net/caif/cfctrl.c b/net/caif/cfctrl.c
index a376ec1..9cd057c 100644
--- a/net/caif/cfctrl.c
+++ b/net/caif/cfctrl.c
@@ -20,12 +20,12 @@
#ifdef CAIF_NO_LOOP
static int handle_loop(struct cfctrl *ctrl,
- int cmd, struct cfpkt *pkt){
+ int cmd, struct cfpkt *pkt){
return -1;
}
#else
static int handle_loop(struct cfctrl *ctrl,
- int cmd, struct cfpkt *pkt);
+ int cmd, struct cfpkt *pkt);
#endif
static int cfctrl_recv(struct cflayer *layr, struct cfpkt *pkt);
static void cfctrl_ctrlcmd(struct cflayer *layr, enum caif_ctrlcmd ctrl,
@@ -72,7 +72,7 @@
}
static bool param_eq(const struct cfctrl_link_param *p1,
- const struct cfctrl_link_param *p2)
+ const struct cfctrl_link_param *p2)
{
bool eq =
p1->linktype == p2->linktype &&
@@ -197,8 +197,8 @@
}
int cfctrl_linkup_request(struct cflayer *layer,
- struct cfctrl_link_param *param,
- struct cflayer *user_layer)
+ struct cfctrl_link_param *param,
+ struct cflayer *user_layer)
{
struct cfctrl *cfctrl = container_obj(layer);
u32 tmp32;
@@ -301,7 +301,7 @@
}
int cfctrl_linkdown_req(struct cflayer *layer, u8 channelid,
- struct cflayer *client)
+ struct cflayer *client)
{
int ret;
struct cfpkt *pkt;
@@ -555,7 +555,7 @@
}
static void cfctrl_ctrlcmd(struct cflayer *layr, enum caif_ctrlcmd ctrl,
- int phyid)
+ int phyid)
{
struct cfctrl *this = container_obj(layr);
switch (ctrl) {
diff --git a/net/caif/cffrml.c b/net/caif/cffrml.c
index 0a7df7e..204c5e2 100644
--- a/net/caif/cffrml.c
+++ b/net/caif/cffrml.c
@@ -28,7 +28,7 @@
static int cffrml_receive(struct cflayer *layr, struct cfpkt *pkt);
static int cffrml_transmit(struct cflayer *layr, struct cfpkt *pkt);
static void cffrml_ctrlcmd(struct cflayer *layr, enum caif_ctrlcmd ctrl,
- int phyid);
+ int phyid);
static u32 cffrml_rcv_error;
static u32 cffrml_rcv_checsum_error;
@@ -167,7 +167,7 @@
}
static void cffrml_ctrlcmd(struct cflayer *layr, enum caif_ctrlcmd ctrl,
- int phyid)
+ int phyid)
{
if (layr->up && layr->up->ctrlcmd)
layr->up->ctrlcmd(layr->up, ctrl, layr->id);
diff --git a/net/caif/cfmuxl.c b/net/caif/cfmuxl.c
index 94b0861..154d9f8 100644
--- a/net/caif/cfmuxl.c
+++ b/net/caif/cfmuxl.c
@@ -42,7 +42,7 @@
static int cfmuxl_receive(struct cflayer *layr, struct cfpkt *pkt);
static int cfmuxl_transmit(struct cflayer *layr, struct cfpkt *pkt);
static void cfmuxl_ctrlcmd(struct cflayer *layr, enum caif_ctrlcmd ctrl,
- int phyid);
+ int phyid);
static struct cflayer *get_up(struct cfmuxl *muxl, u16 id);
struct cflayer *cfmuxl_create(void)
@@ -244,7 +244,7 @@
}
static void cfmuxl_ctrlcmd(struct cflayer *layr, enum caif_ctrlcmd ctrl,
- int phyid)
+ int phyid)
{
struct cfmuxl *muxl = container_obj(layr);
struct cflayer *layer;
diff --git a/net/caif/cfpkt_skbuff.c b/net/caif/cfpkt_skbuff.c
index 863dedd..e8f9c14 100644
--- a/net/caif/cfpkt_skbuff.c
+++ b/net/caif/cfpkt_skbuff.c
@@ -266,8 +266,8 @@
}
inline u16 cfpkt_iterate(struct cfpkt *pkt,
- u16 (*iter_func)(u16, void *, u16),
- u16 data)
+ u16 (*iter_func)(u16, void *, u16),
+ u16 data)
{
/*
* Don't care about the performance hit of linearizing,
@@ -307,8 +307,8 @@
}
struct cfpkt *cfpkt_append(struct cfpkt *dstpkt,
- struct cfpkt *addpkt,
- u16 expectlen)
+ struct cfpkt *addpkt,
+ u16 expectlen)
{
struct sk_buff *dst = pkt_to_skb(dstpkt);
struct sk_buff *add = pkt_to_skb(addpkt);
diff --git a/net/caif/cfrfml.c b/net/caif/cfrfml.c
index 2b563ad..db51830c 100644
--- a/net/caif/cfrfml.c
+++ b/net/caif/cfrfml.c
@@ -43,7 +43,7 @@
}
struct cflayer *cfrfml_create(u8 channel_id, struct dev_info *dev_info,
- int mtu_size)
+ int mtu_size)
{
int tmp;
struct cfrfml *this = kzalloc(sizeof(struct cfrfml), GFP_ATOMIC);
@@ -69,7 +69,7 @@
}
static struct cfpkt *rfm_append(struct cfrfml *rfml, char *seghead,
- struct cfpkt *pkt, int *err)
+ struct cfpkt *pkt, int *err)
{
struct cfpkt *tmppkt;
*err = -EPROTO;
diff --git a/net/caif/cfserl.c b/net/caif/cfserl.c
index 8e68b97..147c232 100644
--- a/net/caif/cfserl.c
+++ b/net/caif/cfserl.c
@@ -29,7 +29,7 @@
static int cfserl_receive(struct cflayer *layr, struct cfpkt *pkt);
static int cfserl_transmit(struct cflayer *layr, struct cfpkt *pkt);
static void cfserl_ctrlcmd(struct cflayer *layr, enum caif_ctrlcmd ctrl,
- int phyid);
+ int phyid);
struct cflayer *cfserl_create(int instance, bool use_stx)
{
@@ -182,7 +182,7 @@
}
static void cfserl_ctrlcmd(struct cflayer *layr, enum caif_ctrlcmd ctrl,
- int phyid)
+ int phyid)
{
layr->up->ctrlcmd(layr->up, ctrl, phyid);
}
diff --git a/net/caif/cfsrvl.c b/net/caif/cfsrvl.c
index ba217e9..95f7f5e 100644
--- a/net/caif/cfsrvl.c
+++ b/net/caif/cfsrvl.c
@@ -25,7 +25,7 @@
#define container_obj(layr) container_of(layr, struct cfsrvl, layer)
static void cfservl_ctrlcmd(struct cflayer *layr, enum caif_ctrlcmd ctrl,
- int phyid)
+ int phyid)
{
struct cfsrvl *service = container_obj(layr);
@@ -158,10 +158,9 @@
}
void cfsrvl_init(struct cfsrvl *service,
- u8 channel_id,
- struct dev_info *dev_info,
- bool supports_flowctrl
- )
+ u8 channel_id,
+ struct dev_info *dev_info,
+ bool supports_flowctrl)
{
caif_assert(offsetof(struct cfsrvl, layer) == 0);
service->open = false;
@@ -207,8 +206,8 @@
EXPORT_SYMBOL(caif_free_client);
void caif_client_register_refcnt(struct cflayer *adapt_layer,
- void (*hold)(struct cflayer *lyr),
- void (*put)(struct cflayer *lyr))
+ void (*hold)(struct cflayer *lyr),
+ void (*put)(struct cflayer *lyr))
{
struct cfsrvl *service;
diff --git a/net/caif/chnl_net.c b/net/caif/chnl_net.c
index e597733..26a4e4e 100644
--- a/net/caif/chnl_net.c
+++ b/net/caif/chnl_net.c
@@ -167,7 +167,7 @@
}
static void chnl_flowctrl_cb(struct cflayer *layr, enum caif_ctrlcmd flow,
- int phyid)
+ int phyid)
{
struct chnl_net *priv = container_of(layr, struct chnl_net, chnl);
pr_debug("NET flowctrl func called flow: %s\n",
@@ -443,7 +443,7 @@
}
static void caif_netlink_parms(struct nlattr *data[],
- struct caif_connect_request *conn_req)
+ struct caif_connect_request *conn_req)
{
if (!data) {
pr_warn("no params data found\n");
@@ -488,7 +488,7 @@
}
static int ipcaif_changelink(struct net_device *dev, struct nlattr *tb[],
- struct nlattr *data[])
+ struct nlattr *data[])
{
struct chnl_net *caifdev;
ASSERT_RTNL();
diff --git a/net/can/af_can.c b/net/can/af_can.c
index c48e522..c4e5085 100644
--- a/net/can/af_can.c
+++ b/net/can/af_can.c
@@ -525,7 +525,7 @@
d = find_dev_rcv_lists(dev);
if (!d) {
- printk(KERN_ERR "BUG: receive list not found for "
+ pr_err("BUG: receive list not found for "
"dev %s, id %03X, mask %03X\n",
DNAME(dev), can_id, mask);
goto out;
@@ -546,16 +546,13 @@
}
/*
- * Check for bugs in CAN protocol implementations:
- * If no matching list item was found, the list cursor variable next
- * will be NULL, while r will point to the last item of the list.
+ * Check for bugs in CAN protocol implementations using af_can.c:
+ * 'r' will be NULL if no matching list item was found for removal.
*/
if (!r) {
- printk(KERN_ERR "BUG: receive list entry not found for "
- "dev %s, id %03X, mask %03X\n",
- DNAME(dev), can_id, mask);
- r = NULL;
+ WARN(1, "BUG: receive list entry not found for dev %s, "
+ "id %03X, mask %03X\n", DNAME(dev), can_id, mask);
goto out;
}
@@ -749,8 +746,7 @@
int err = 0;
if (proto < 0 || proto >= CAN_NPROTO) {
- printk(KERN_ERR "can: protocol number %d out of range\n",
- proto);
+ pr_err("can: protocol number %d out of range\n", proto);
return -EINVAL;
}
@@ -761,8 +757,7 @@
mutex_lock(&proto_tab_lock);
if (proto_tab[proto]) {
- printk(KERN_ERR "can: protocol %d already registered\n",
- proto);
+ pr_err("can: protocol %d already registered\n", proto);
err = -EBUSY;
} else
RCU_INIT_POINTER(proto_tab[proto], cp);
@@ -816,11 +811,8 @@
/* create new dev_rcv_lists for this device */
d = kzalloc(sizeof(*d), GFP_KERNEL);
- if (!d) {
- printk(KERN_ERR
- "can: allocation of receive list failed\n");
+ if (!d)
return NOTIFY_DONE;
- }
BUG_ON(dev->ml_priv);
dev->ml_priv = d;
@@ -838,8 +830,8 @@
dev->ml_priv = NULL;
}
} else
- printk(KERN_ERR "can: notifier: receive list not "
- "found for dev %s\n", dev->name);
+ pr_err("can: notifier: receive list not found for dev "
+ "%s\n", dev->name);
spin_unlock(&can_rcvlists_lock);
@@ -927,7 +919,7 @@
/* remove created dev_rcv_lists from still registered CAN devices */
rcu_read_lock();
for_each_netdev_rcu(&init_net, dev) {
- if (dev->type == ARPHRD_CAN && dev->ml_priv){
+ if (dev->type == ARPHRD_CAN && dev->ml_priv) {
struct dev_rcv_lists *d = dev->ml_priv;
diff --git a/net/core/dev.c b/net/core/dev.c
index d540ced..de930b7 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2208,6 +2208,25 @@
}
EXPORT_SYMBOL(skb_checksum_help);
+__be16 skb_network_protocol(struct sk_buff *skb)
+{
+ __be16 type = skb->protocol;
+ int vlan_depth = ETH_HLEN;
+
+ while (type == htons(ETH_P_8021Q)) {
+ struct vlan_hdr *vh;
+
+ if (unlikely(!pskb_may_pull(skb, vlan_depth + VLAN_HLEN)))
+ return 0;
+
+ vh = (struct vlan_hdr *)(skb->data + vlan_depth);
+ type = vh->h_vlan_encapsulated_proto;
+ vlan_depth += VLAN_HLEN;
+ }
+
+ return type;
+}
+
/**
* skb_mac_gso_segment - mac layer segmentation handler.
* @skb: buffer to segment
@@ -2218,19 +2237,10 @@
{
struct sk_buff *segs = ERR_PTR(-EPROTONOSUPPORT);
struct packet_offload *ptype;
- __be16 type = skb->protocol;
- int vlan_depth = ETH_HLEN;
+ __be16 type = skb_network_protocol(skb);
- while (type == htons(ETH_P_8021Q)) {
- struct vlan_hdr *vh;
-
- if (unlikely(!pskb_may_pull(skb, vlan_depth + VLAN_HLEN)))
- return ERR_PTR(-EINVAL);
-
- vh = (struct vlan_hdr *)(skb->data + vlan_depth);
- type = vh->h_vlan_encapsulated_proto;
- vlan_depth += VLAN_HLEN;
- }
+ if (unlikely(!type))
+ return ERR_PTR(-EINVAL);
__skb_pull(skb, skb->mac_len);
@@ -2398,24 +2408,12 @@
return 0;
}
-static bool can_checksum_protocol(netdev_features_t features, __be16 protocol)
-{
- return ((features & NETIF_F_GEN_CSUM) ||
- ((features & NETIF_F_V4_CSUM) &&
- protocol == htons(ETH_P_IP)) ||
- ((features & NETIF_F_V6_CSUM) &&
- protocol == htons(ETH_P_IPV6)) ||
- ((features & NETIF_F_FCOE_CRC) &&
- protocol == htons(ETH_P_FCOE)));
-}
-
static netdev_features_t harmonize_features(struct sk_buff *skb,
__be16 protocol, netdev_features_t features)
{
if (skb->ip_summed != CHECKSUM_NONE &&
!can_checksum_protocol(features, protocol)) {
features &= ~NETIF_F_ALL_CSUM;
- features &= ~NETIF_F_SG;
} else if (illegal_highdma(skb->dev, skb)) {
features &= ~NETIF_F_SG;
}
@@ -3326,7 +3324,7 @@
* netdev_rx_handler_unregister - unregister receive handler
* @dev: device to unregister a handler from
*
- * Unregister a receive hander from a device.
+ * Unregister a receive handler from a device.
*
* The caller must hold the rtnl_mutex.
*/
@@ -4058,6 +4056,9 @@
napi->gro_list = NULL;
napi->skb = NULL;
napi->poll = poll;
+ if (weight > NAPI_POLL_WEIGHT)
+ pr_err_once("netif_napi_add() called with weight %d on device %s\n",
+ weight, dev->name);
napi->weight = weight;
list_add(&napi->dev_list, &dev->napi_list);
napi->dev = dev;
@@ -4919,20 +4920,25 @@
features &= ~(NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM);
}
- /* Fix illegal SG+CSUM combinations. */
- if ((features & NETIF_F_SG) &&
- !(features & NETIF_F_ALL_CSUM)) {
- netdev_dbg(dev,
- "Dropping NETIF_F_SG since no checksum feature.\n");
- features &= ~NETIF_F_SG;
- }
-
/* TSO requires that SG is present as well. */
if ((features & NETIF_F_ALL_TSO) && !(features & NETIF_F_SG)) {
netdev_dbg(dev, "Dropping TSO features since no SG feature.\n");
features &= ~NETIF_F_ALL_TSO;
}
+ if ((features & NETIF_F_TSO) && !(features & NETIF_F_HW_CSUM) &&
+ !(features & NETIF_F_IP_CSUM)) {
+ netdev_dbg(dev, "Dropping TSO features since no CSUM feature.\n");
+ features &= ~NETIF_F_TSO;
+ features &= ~NETIF_F_TSO_ECN;
+ }
+
+ if ((features & NETIF_F_TSO6) && !(features & NETIF_F_HW_CSUM) &&
+ !(features & NETIF_F_IPV6_CSUM)) {
+ netdev_dbg(dev, "Dropping TSO6 features since no CSUM feature.\n");
+ features &= ~NETIF_F_TSO6;
+ }
+
/* TSO ECN requires that TSO is present as well. */
if ((features & NETIF_F_ALL_TSO) == NETIF_F_TSO_ECN)
features &= ~NETIF_F_TSO_ECN;
@@ -5200,6 +5206,10 @@
*/
dev->vlan_features |= NETIF_F_HIGHDMA;
+ /* Make NETIF_F_SG inheritable to tunnel devices.
+ */
+ dev->hw_enc_features |= NETIF_F_SG;
+
ret = call_netdevice_notifiers(NETDEV_POST_INIT, dev);
ret = notifier_to_errno(ret);
if (ret)
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index 3e9b2c3..adc1351 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -78,6 +78,7 @@
[NETIF_F_TSO6_BIT] = "tx-tcp6-segmentation",
[NETIF_F_FSO_BIT] = "tx-fcoe-segmentation",
[NETIF_F_GSO_GRE_BIT] = "tx-gre-segmentation",
+ [NETIF_F_GSO_UDP_TUNNEL_BIT] = "tx-udp_tnl-segmentation",
[NETIF_F_FCOE_CRC_BIT] = "tx-checksum-fcoe-crc",
[NETIF_F_SCTP_CSUM_BIT] = "tx-checksum-sctp",
diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index e187bf0..f4be293 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -119,6 +119,17 @@
nhoff += 4;
if (hdr->flags & GRE_SEQ)
nhoff += 4;
+ if (proto == htons(ETH_P_TEB)) {
+ const struct ethhdr *eth;
+ struct ethhdr _eth;
+
+ eth = skb_header_pointer(skb, nhoff,
+ sizeof(_eth), &_eth);
+ if (!eth)
+ return false;
+ proto = eth->h_proto;
+ nhoff += sizeof(*eth);
+ }
goto again;
}
break;
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index fa32899..a3a17ae 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -47,7 +47,7 @@
static atomic_t trapped;
-static struct srcu_struct netpoll_srcu;
+DEFINE_STATIC_SRCU(netpoll_srcu);
#define USEC_PER_POLL 50
#define NETPOLL_RX_ENABLED 1
@@ -1212,7 +1212,6 @@
static int __init netpoll_init(void)
{
skb_queue_head_init(&skb_pool);
- init_srcu_struct(&netpoll_srcu);
return 0;
}
core_initcall(netpoll_init);
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 5fb8d7e..9a9b99e 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -2049,6 +2049,38 @@
rtnl_set_sk_err(net, RTNLGRP_NEIGH, err);
}
+/**
+ * ndo_dflt_fdb_add - default netdevice operation to add an FDB entry
+ */
+int ndo_dflt_fdb_add(struct ndmsg *ndm,
+ struct nlattr *tb[],
+ struct net_device *dev,
+ const unsigned char *addr,
+ u16 flags)
+{
+ int err = -EINVAL;
+
+ /* If aging addresses are supported device will need to
+ * implement its own handler for this.
+ */
+ if (ndm->ndm_state && !(ndm->ndm_state & NUD_PERMANENT)) {
+ pr_info("%s: FDB only supports static addresses\n", dev->name);
+ return err;
+ }
+
+ if (is_unicast_ether_addr(addr) || is_link_local_ether_addr(addr))
+ err = dev_uc_add_excl(dev, addr);
+ else if (is_multicast_ether_addr(addr))
+ err = dev_mc_add_excl(dev, addr);
+
+ /* Only return duplicate errors if NLM_F_EXCL is set */
+ if (err == -EEXIST && !(flags & NLM_F_EXCL))
+ err = 0;
+
+ return err;
+}
+EXPORT_SYMBOL(ndo_dflt_fdb_add);
+
static int rtnl_fdb_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
{
struct net *net = sock_net(skb->sk);
@@ -2080,7 +2112,7 @@
}
addr = nla_data(tb[NDA_LLADDR]);
- if (!is_valid_ether_addr(addr)) {
+ if (is_zero_ether_addr(addr)) {
pr_info("PF_BRIDGE: RTM_NEWNEIGH with invalid ether address\n");
return -EINVAL;
}
@@ -2101,10 +2133,13 @@
}
/* Embedded bridge, macvlan, and any other device support */
- if ((ndm->ndm_flags & NTF_SELF) && dev->netdev_ops->ndo_fdb_add) {
- err = dev->netdev_ops->ndo_fdb_add(ndm, tb,
- dev, addr,
- nlh->nlmsg_flags);
+ if ((ndm->ndm_flags & NTF_SELF)) {
+ if (dev->netdev_ops->ndo_fdb_add)
+ err = dev->netdev_ops->ndo_fdb_add(ndm, tb, dev, addr,
+ nlh->nlmsg_flags);
+ else
+ err = ndo_dflt_fdb_add(ndm, tb, dev, addr,
+ nlh->nlmsg_flags);
if (!err) {
rtnl_fdb_notify(dev, addr, RTM_NEWNEIGH);
@@ -2115,6 +2150,35 @@
return err;
}
+/**
+ * ndo_dflt_fdb_del - default netdevice operation to delete an FDB entry
+ */
+int ndo_dflt_fdb_del(struct ndmsg *ndm,
+ struct nlattr *tb[],
+ struct net_device *dev,
+ const unsigned char *addr)
+{
+ int err = -EOPNOTSUPP;
+
+ /* If aging addresses are supported device will need to
+ * implement its own handler for this.
+ */
+ if (ndm->ndm_state & NUD_PERMANENT) {
+ pr_info("%s: FDB only supports static addresses\n", dev->name);
+ return -EINVAL;
+ }
+
+ if (is_unicast_ether_addr(addr) || is_link_local_ether_addr(addr))
+ err = dev_uc_del(dev, addr);
+ else if (is_multicast_ether_addr(addr))
+ err = dev_mc_del(dev, addr);
+ else
+ err = -EINVAL;
+
+ return err;
+}
+EXPORT_SYMBOL(ndo_dflt_fdb_del);
+
static int rtnl_fdb_del(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
{
struct net *net = sock_net(skb->sk);
@@ -2172,8 +2236,11 @@
}
/* Embedded bridge, macvlan, and any other device support */
- if ((ndm->ndm_flags & NTF_SELF) && dev->netdev_ops->ndo_fdb_del) {
- err = dev->netdev_ops->ndo_fdb_del(ndm, tb, dev, addr);
+ if (ndm->ndm_flags & NTF_SELF) {
+ if (dev->netdev_ops->ndo_fdb_del)
+ err = dev->netdev_ops->ndo_fdb_del(ndm, tb, dev, addr);
+ else
+ err = ndo_dflt_fdb_del(ndm, tb, dev, addr);
if (!err) {
rtnl_fdb_notify(dev, addr, RTM_DELNEIGH);
@@ -2258,6 +2325,8 @@
if (dev->netdev_ops->ndo_fdb_dump)
idx = dev->netdev_ops->ndo_fdb_dump(skb, cb, dev, idx);
+ else
+ ndo_dflt_fdb_dump(skb, cb, dev, idx);
}
rcu_read_unlock();
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 33245ef..31c6737 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -673,6 +673,7 @@
new->mac_header = old->mac_header;
new->inner_transport_header = old->inner_transport_header;
new->inner_network_header = old->inner_network_header;
+ new->inner_mac_header = old->inner_mac_header;
skb_dst_copy(new, old);
new->rxhash = old->rxhash;
new->ooo_okay = old->ooo_okay;
@@ -867,6 +868,18 @@
}
EXPORT_SYMBOL(skb_clone);
+static void skb_headers_offset_update(struct sk_buff *skb, int off)
+{
+ /* {transport,network,mac}_header and tail are relative to skb->head */
+ skb->transport_header += off;
+ skb->network_header += off;
+ if (skb_mac_header_was_set(skb))
+ skb->mac_header += off;
+ skb->inner_transport_header += off;
+ skb->inner_network_header += off;
+ skb->inner_mac_header += off;
+}
+
static void copy_skb_header(struct sk_buff *new, const struct sk_buff *old)
{
#ifndef NET_SKBUFF_DATA_USES_OFFSET
@@ -879,13 +892,7 @@
__copy_skb_header(new, old);
#ifndef NET_SKBUFF_DATA_USES_OFFSET
- /* {transport,network,mac}_header are relative to skb->head */
- new->transport_header += offset;
- new->network_header += offset;
- if (skb_mac_header_was_set(new))
- new->mac_header += offset;
- new->inner_transport_header += offset;
- new->inner_network_header += offset;
+ skb_headers_offset_update(new, offset);
#endif
skb_shinfo(new)->gso_size = skb_shinfo(old)->gso_size;
skb_shinfo(new)->gso_segs = skb_shinfo(old)->gso_segs;
@@ -1077,14 +1084,8 @@
#else
skb->end = skb->head + size;
#endif
- /* {transport,network,mac}_header and tail are relative to skb->head */
skb->tail += off;
- skb->transport_header += off;
- skb->network_header += off;
- if (skb_mac_header_was_set(skb))
- skb->mac_header += off;
- skb->inner_transport_header += off;
- skb->inner_network_header += off;
+ skb_headers_offset_update(skb, off);
/* Only adjust this if it actually is csum_start rather than csum */
if (skb->ip_summed == CHECKSUM_PARTIAL)
skb->csum_start += nhead;
@@ -1180,12 +1181,7 @@
if (n->ip_summed == CHECKSUM_PARTIAL)
n->csum_start += off;
#ifdef NET_SKBUFF_DATA_USES_OFFSET
- n->transport_header += off;
- n->network_header += off;
- if (skb_mac_header_was_set(skb))
- n->mac_header += off;
- n->inner_transport_header += off;
- n->inner_network_header += off;
+ skb_headers_offset_update(n, off);
#endif
return n;
@@ -2741,12 +2737,19 @@
unsigned int tnl_hlen = skb_tnl_header_len(skb);
unsigned int headroom;
unsigned int len;
+ __be16 proto;
+ bool csum;
int sg = !!(features & NETIF_F_SG);
int nfrags = skb_shinfo(skb)->nr_frags;
int err = -ENOMEM;
int i = 0;
int pos;
+ proto = skb_network_protocol(skb);
+ if (unlikely(!proto))
+ return ERR_PTR(-EINVAL);
+
+ csum = !!can_checksum_protocol(features, proto);
__skb_push(skb, doffset);
headroom = skb_headroom(skb);
pos = skb_headlen(skb);
@@ -2884,6 +2887,12 @@
nskb->data_len = len - hsize;
nskb->len += nskb->data_len;
nskb->truesize += nskb->data_len;
+
+ if (!csum) {
+ nskb->csum = skb_checksum(nskb, doffset,
+ nskb->len - doffset, 0);
+ nskb->ip_summed = CHECKSUM_NONE;
+ }
} while ((offset += len) < skb->len);
return segs;
diff --git a/net/dcb/dcbevent.c b/net/dcb/dcbevent.c
index 1d9eb7c..4f72fc4 100644
--- a/net/dcb/dcbevent.c
+++ b/net/dcb/dcbevent.c
@@ -20,6 +20,7 @@
#include <linux/rtnetlink.h>
#include <linux/notifier.h>
#include <linux/export.h>
+#include <net/dcbevent.h>
static ATOMIC_NOTIFIER_HEAD(dcbevent_notif_chain);
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 4f9f5eb..ebc54fe 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -500,8 +500,7 @@
return &rt->dst;
}
-static int dccp_v4_send_response(struct sock *sk, struct request_sock *req,
- struct request_values *rv_unused)
+static int dccp_v4_send_response(struct sock *sk, struct request_sock *req)
{
int err = -1;
struct sk_buff *skb;
@@ -658,7 +657,7 @@
dreq->dreq_gss = dreq->dreq_iss;
dreq->dreq_service = service;
- if (dccp_v4_send_response(sk, req, NULL))
+ if (dccp_v4_send_response(sk, req))
goto drop_and_free;
inet_csk_reqsk_queue_hash_add(sk, req, DCCP_TIMEOUT_INIT);
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index 6e05981..9c61f9c 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -213,8 +213,7 @@
}
-static int dccp_v6_send_response(struct sock *sk, struct request_sock *req,
- struct request_values *rv_unused)
+static int dccp_v6_send_response(struct sock *sk, struct request_sock *req)
{
struct inet6_request_sock *ireq6 = inet6_rsk(req);
struct ipv6_pinfo *np = inet6_sk(sk);
@@ -428,7 +427,7 @@
dreq->dreq_gss = dreq->dreq_iss;
dreq->dreq_service = service;
- if (dccp_v6_send_response(sk, req, NULL))
+ if (dccp_v6_send_response(sk, req))
goto drop_and_free;
inet6_csk_reqsk_queue_hash_add(sk, req, DCCP_TIMEOUT_INIT);
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 68f6a94..9e5882c 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1283,9 +1283,7 @@
int ihl;
int id;
unsigned int offset = 0;
-
- if (!(features & NETIF_F_V4_CSUM))
- features &= ~NETIF_F_SG;
+ bool tunnel;
if (unlikely(skb_shinfo(skb)->gso_type &
~(SKB_GSO_TCPV4 |
@@ -1293,6 +1291,7 @@
SKB_GSO_DODGY |
SKB_GSO_TCP_ECN |
SKB_GSO_GRE |
+ SKB_GSO_UDP_TUNNEL |
0)))
goto out;
@@ -1307,6 +1306,8 @@
if (unlikely(!pskb_may_pull(skb, ihl)))
goto out;
+ tunnel = !!skb->encapsulation;
+
__skb_pull(skb, ihl);
skb_reset_transport_header(skb);
iph = ip_hdr(skb);
@@ -1326,7 +1327,7 @@
skb = segs;
do {
iph = ip_hdr(skb);
- if (proto == IPPROTO_UDP) {
+ if (!tunnel && proto == IPPROTO_UDP) {
iph->id = htons(id);
iph->frag_off = htons(offset >> 3);
if (skb->next != NULL)
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index f678507..af57bba 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -1791,6 +1791,74 @@
return err;
}
+static int inet_netconf_dump_devconf(struct sk_buff *skb,
+ struct netlink_callback *cb)
+{
+ struct net *net = sock_net(skb->sk);
+ int h, s_h;
+ int idx, s_idx;
+ struct net_device *dev;
+ struct in_device *in_dev;
+ struct hlist_head *head;
+
+ s_h = cb->args[0];
+ s_idx = idx = cb->args[1];
+
+ for (h = s_h; h < NETDEV_HASHENTRIES; h++, s_idx = 0) {
+ idx = 0;
+ head = &net->dev_index_head[h];
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(dev, head, index_hlist) {
+ if (idx < s_idx)
+ goto cont;
+ in_dev = __in_dev_get_rcu(dev);
+ if (!in_dev)
+ goto cont;
+
+ if (inet_netconf_fill_devconf(skb, dev->ifindex,
+ &in_dev->cnf,
+ NETLINK_CB(cb->skb).portid,
+ cb->nlh->nlmsg_seq,
+ RTM_NEWNETCONF,
+ NLM_F_MULTI,
+ -1) <= 0) {
+ rcu_read_unlock();
+ goto done;
+ }
+cont:
+ idx++;
+ }
+ rcu_read_unlock();
+ }
+ if (h == NETDEV_HASHENTRIES) {
+ if (inet_netconf_fill_devconf(skb, NETCONFA_IFINDEX_ALL,
+ net->ipv4.devconf_all,
+ NETLINK_CB(cb->skb).portid,
+ cb->nlh->nlmsg_seq,
+ RTM_NEWNETCONF, NLM_F_MULTI,
+ -1) <= 0)
+ goto done;
+ else
+ h++;
+ }
+ if (h == NETDEV_HASHENTRIES + 1) {
+ if (inet_netconf_fill_devconf(skb, NETCONFA_IFINDEX_DEFAULT,
+ net->ipv4.devconf_dflt,
+ NETLINK_CB(cb->skb).portid,
+ cb->nlh->nlmsg_seq,
+ RTM_NEWNETCONF, NLM_F_MULTI,
+ -1) <= 0)
+ goto done;
+ else
+ h++;
+ }
+done:
+ cb->args[0] = h;
+ cb->args[1] = idx;
+
+ return skb->len;
+}
+
#ifdef CONFIG_SYSCTL
static void devinet_copy_dflt_conf(struct net *net, int i)
@@ -2195,6 +2263,6 @@
rtnl_register(PF_INET, RTM_DELADDR, inet_rtm_deladdr, NULL, NULL);
rtnl_register(PF_INET, RTM_GETADDR, NULL, inet_dump_ifaddr, NULL);
rtnl_register(PF_INET, RTM_GETNETCONF, inet_netconf_get_devconf,
- NULL, NULL);
+ inet_netconf_dump_devconf, NULL);
}
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 786d97a..6acb541 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -559,7 +559,7 @@
int inet_rtx_syn_ack(struct sock *parent, struct request_sock *req)
{
- int err = req->rsk_ops->rtx_syn_ack(parent, req, NULL);
+ int err = req->rsk_ops->rtx_syn_ack(parent, req);
if (!err)
req->num_retrans++;
diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c
index 7afa2c3..8620408 100644
--- a/net/ipv4/inet_diag.c
+++ b/net/ipv4/inet_diag.c
@@ -158,7 +158,9 @@
#define EXPIRES_IN_MS(tmo) DIV_ROUND_UP((tmo - jiffies) * 1000, HZ)
- if (icsk->icsk_pending == ICSK_TIME_RETRANS) {
+ if (icsk->icsk_pending == ICSK_TIME_RETRANS ||
+ icsk->icsk_pending == ICSK_TIME_EARLY_RETRANS ||
+ icsk->icsk_pending == ICSK_TIME_LOSS_PROBE) {
r->idiag_timer = 1;
r->idiag_retrans = icsk->icsk_retransmits;
r->idiag_expires = EXPIRES_IN_MS(icsk->icsk_timeout);
diff --git a/net/ipv4/inet_lro.c b/net/ipv4/inet_lro.c
index cc280a3..1975f52 100644
--- a/net/ipv4/inet_lro.c
+++ b/net/ipv4/inet_lro.c
@@ -29,6 +29,7 @@
#include <linux/module.h>
#include <linux/if_vlan.h>
#include <linux/inet_lro.h>
+#include <net/checksum.h>
MODULE_LICENSE("GPL");
MODULE_AUTHOR("Jan-Bernd Themann <themann@de.ibm.com>");
@@ -114,11 +115,9 @@
*(p+2) = lro_desc->tcp_rcv_tsecr;
}
+ csum_replace2(&iph->check, iph->tot_len, htons(lro_desc->ip_tot_len));
iph->tot_len = htons(lro_desc->ip_tot_len);
- iph->check = 0;
- iph->check = ip_fast_csum((u8 *)lro_desc->iph, iph->ihl);
-
tcph->check = 0;
tcp_hdr_csum = csum_partial(tcph, TCP_HDR_LEN(tcph), 0);
lro_desc->data_csum = csum_add(lro_desc->data_csum, tcp_hdr_csum);
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index 91d66db..2e94289 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -762,7 +762,6 @@
static netdev_tx_t ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
{
- struct pcpu_tstats *tstats = this_cpu_ptr(dev->tstats);
struct ip_tunnel *tunnel = netdev_priv(dev);
const struct iphdr *old_iph;
const struct iphdr *tiph;
@@ -778,7 +777,6 @@
int mtu;
u8 ttl;
int err;
- int pkt_len;
skb = handle_offloads(tunnel, skb);
if (IS_ERR(skb)) {
@@ -1019,19 +1017,7 @@
}
}
- nf_reset(skb);
-
- pkt_len = skb->len - skb_transport_offset(skb);
- err = ip_local_out(skb);
- if (likely(net_xmit_eval(err) == 0)) {
- u64_stats_update_begin(&tstats->syncp);
- tstats->tx_bytes += pkt_len;
- tstats->tx_packets++;
- u64_stats_update_end(&tstats->syncp);
- } else {
- dev->stats.tx_errors++;
- dev->stats.tx_aborted_errors++;
- }
+ iptunnel_xmit(skb, dev);
return NETDEV_TX_OK;
#if IS_ENABLED(CONFIG_IPV6)
diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c
index 8f024d4..34e006f 100644
--- a/net/ipv4/ipip.c
+++ b/net/ipv4/ipip.c
@@ -481,11 +481,6 @@
if (skb->protocol != htons(ETH_P_IP))
goto tx_error;
-
- if (skb->ip_summed == CHECKSUM_PARTIAL &&
- skb_checksum_help(skb))
- goto tx_error;
-
old_iph = ip_hdr(skb);
if (tos & 1)
@@ -570,6 +565,13 @@
old_iph = ip_hdr(skb);
}
+ if (!skb->encapsulation) {
+ skb_reset_inner_headers(skb);
+ skb->encapsulation = 1;
+ }
+ if (skb->ip_summed != CHECKSUM_PARTIAL)
+ skb->ip_summed = CHECKSUM_NONE;
+
skb->transport_header = skb->network_header;
skb_push(skb, sizeof(struct iphdr));
skb_reset_network_header(skb);
@@ -591,11 +593,13 @@
iph->tos = INET_ECN_encapsulate(tos, old_iph->tos);
iph->daddr = fl4.daddr;
iph->saddr = fl4.saddr;
+ tunnel_ip_select_ident(skb, old_iph, &rt->dst);
if ((iph->ttl = tiph->ttl) == 0)
iph->ttl = old_iph->ttl;
iptunnel_xmit(skb, dev);
+
return NETDEV_TX_OK;
tx_error_icmp:
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index 32030a2..b6f2ea1 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -224,6 +224,8 @@
SNMP_MIB_ITEM("TCPForwardRetrans", LINUX_MIB_TCPFORWARDRETRANS),
SNMP_MIB_ITEM("TCPSlowStartRetrans", LINUX_MIB_TCPSLOWSTARTRETRANS),
SNMP_MIB_ITEM("TCPTimeouts", LINUX_MIB_TCPTIMEOUTS),
+ SNMP_MIB_ITEM("TCPLossProbes", LINUX_MIB_TCPLOSSPROBES),
+ SNMP_MIB_ITEM("TCPLossProbeRecovery", LINUX_MIB_TCPLOSSPROBERECOVERY),
SNMP_MIB_ITEM("TCPRenoRecoveryFail", LINUX_MIB_TCPRENORECOVERYFAIL),
SNMP_MIB_ITEM("TCPSackRecoveryFail", LINUX_MIB_TCPSACKRECOVERYFAIL),
SNMP_MIB_ITEM("TCPSchedulerFailed", LINUX_MIB_TCPSCHEDULERFAILED),
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index ef54377..7f4a5cb 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -267,7 +267,6 @@
struct ip_options *opt)
{
struct tcp_options_received tcp_opt;
- const u8 *hash_location;
struct inet_request_sock *ireq;
struct tcp_request_sock *treq;
struct tcp_sock *tp = tcp_sk(sk);
@@ -294,7 +293,7 @@
/* check for timestamp cookie support */
memset(&tcp_opt, 0, sizeof(tcp_opt));
- tcp_parse_options(skb, &tcp_opt, &hash_location, 0, NULL);
+ tcp_parse_options(skb, &tcp_opt, 0, NULL);
if (!cookie_check_timestamp(&tcp_opt, sock_net(sk), &ecn_ok))
goto out;
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 960fd29..cb45062 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -28,7 +28,7 @@
static int zero;
static int one = 1;
-static int two = 2;
+static int four = 4;
static int tcp_retr1_max = 255;
static int ip_local_port_range_min[] = { 1, 1 };
static int ip_local_port_range_max[] = { 65535, 65535 };
@@ -733,13 +733,6 @@
.proc_handler = proc_dointvec,
},
{
- .procname = "tcp_cookie_size",
- .data = &sysctl_tcp_cookie_size,
- .maxlen = sizeof(int),
- .mode = 0644,
- .proc_handler = proc_dointvec
- },
- {
.procname = "tcp_thin_linear_timeouts",
.data = &sysctl_tcp_thin_linear_timeouts,
.maxlen = sizeof(int),
@@ -760,7 +753,7 @@
.mode = 0644,
.proc_handler = proc_dointvec_minmax,
.extra1 = &zero,
- .extra2 = &two,
+ .extra2 = &four,
},
{
.procname = "udp_mem",
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index e220207..a96f7b5 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -409,15 +409,6 @@
icsk->icsk_sync_mss = tcp_sync_mss;
- /* TCP Cookie Transactions */
- if (sysctl_tcp_cookie_size > 0) {
- /* Default, cookies without s_data_payload. */
- tp->cookie_values =
- kzalloc(sizeof(*tp->cookie_values),
- sk->sk_allocation);
- if (tp->cookie_values != NULL)
- kref_init(&tp->cookie_values->kref);
- }
/* Presumed zeroed, in order of appearance:
* cookie_in_always, cookie_out_never,
* s_data_constant, s_data_in, s_data_out
@@ -2397,92 +2388,6 @@
release_sock(sk);
return err;
}
- case TCP_COOKIE_TRANSACTIONS: {
- struct tcp_cookie_transactions ctd;
- struct tcp_cookie_values *cvp = NULL;
-
- if (sizeof(ctd) > optlen)
- return -EINVAL;
- if (copy_from_user(&ctd, optval, sizeof(ctd)))
- return -EFAULT;
-
- if (ctd.tcpct_used > sizeof(ctd.tcpct_value) ||
- ctd.tcpct_s_data_desired > TCP_MSS_DESIRED)
- return -EINVAL;
-
- if (ctd.tcpct_cookie_desired == 0) {
- /* default to global value */
- } else if ((0x1 & ctd.tcpct_cookie_desired) ||
- ctd.tcpct_cookie_desired > TCP_COOKIE_MAX ||
- ctd.tcpct_cookie_desired < TCP_COOKIE_MIN) {
- return -EINVAL;
- }
-
- if (TCP_COOKIE_OUT_NEVER & ctd.tcpct_flags) {
- /* Supercedes all other values */
- lock_sock(sk);
- if (tp->cookie_values != NULL) {
- kref_put(&tp->cookie_values->kref,
- tcp_cookie_values_release);
- tp->cookie_values = NULL;
- }
- tp->rx_opt.cookie_in_always = 0; /* false */
- tp->rx_opt.cookie_out_never = 1; /* true */
- release_sock(sk);
- return err;
- }
-
- /* Allocate ancillary memory before locking.
- */
- if (ctd.tcpct_used > 0 ||
- (tp->cookie_values == NULL &&
- (sysctl_tcp_cookie_size > 0 ||
- ctd.tcpct_cookie_desired > 0 ||
- ctd.tcpct_s_data_desired > 0))) {
- cvp = kzalloc(sizeof(*cvp) + ctd.tcpct_used,
- GFP_KERNEL);
- if (cvp == NULL)
- return -ENOMEM;
-
- kref_init(&cvp->kref);
- }
- lock_sock(sk);
- tp->rx_opt.cookie_in_always =
- (TCP_COOKIE_IN_ALWAYS & ctd.tcpct_flags);
- tp->rx_opt.cookie_out_never = 0; /* false */
-
- if (tp->cookie_values != NULL) {
- if (cvp != NULL) {
- /* Changed values are recorded by a changed
- * pointer, ensuring the cookie will differ,
- * without separately hashing each value later.
- */
- kref_put(&tp->cookie_values->kref,
- tcp_cookie_values_release);
- } else {
- cvp = tp->cookie_values;
- }
- }
-
- if (cvp != NULL) {
- cvp->cookie_desired = ctd.tcpct_cookie_desired;
-
- if (ctd.tcpct_used > 0) {
- memcpy(cvp->s_data_payload, ctd.tcpct_value,
- ctd.tcpct_used);
- cvp->s_data_desired = ctd.tcpct_used;
- cvp->s_data_constant = 1; /* true */
- } else {
- /* No constant payload data. */
- cvp->s_data_desired = ctd.tcpct_s_data_desired;
- cvp->s_data_constant = 0; /* false */
- }
-
- tp->cookie_values = cvp;
- }
- release_sock(sk);
- return err;
- }
default:
/* fallthru */
break;
@@ -2902,41 +2807,6 @@
return -EFAULT;
return 0;
- case TCP_COOKIE_TRANSACTIONS: {
- struct tcp_cookie_transactions ctd;
- struct tcp_cookie_values *cvp = tp->cookie_values;
-
- if (get_user(len, optlen))
- return -EFAULT;
- if (len < sizeof(ctd))
- return -EINVAL;
-
- memset(&ctd, 0, sizeof(ctd));
- ctd.tcpct_flags = (tp->rx_opt.cookie_in_always ?
- TCP_COOKIE_IN_ALWAYS : 0)
- | (tp->rx_opt.cookie_out_never ?
- TCP_COOKIE_OUT_NEVER : 0);
-
- if (cvp != NULL) {
- ctd.tcpct_flags |= (cvp->s_data_in ?
- TCP_S_DATA_IN : 0)
- | (cvp->s_data_out ?
- TCP_S_DATA_OUT : 0);
-
- ctd.tcpct_cookie_desired = cvp->cookie_desired;
- ctd.tcpct_s_data_desired = cvp->s_data_desired;
-
- memcpy(&ctd.tcpct_value[0], &cvp->cookie_pair[0],
- cvp->cookie_pair_size);
- ctd.tcpct_used = cvp->cookie_pair_size;
- }
-
- if (put_user(sizeof(ctd), optlen))
- return -EFAULT;
- if (copy_to_user(optval, &ctd, sizeof(ctd)))
- return -EFAULT;
- return 0;
- }
case TCP_THIN_LINEAR_TIMEOUTS:
val = tp->thin_lto;
break;
@@ -3044,6 +2914,7 @@
SKB_GSO_TCP_ECN |
SKB_GSO_TCPV6 |
SKB_GSO_GRE |
+ SKB_GSO_UDP_TUNNEL |
0) ||
!(type & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6))))
goto out;
@@ -3408,134 +3279,6 @@
#endif
-/* Each Responder maintains up to two secret values concurrently for
- * efficient secret rollover. Each secret value has 4 states:
- *
- * Generating. (tcp_secret_generating != tcp_secret_primary)
- * Generates new Responder-Cookies, but not yet used for primary
- * verification. This is a short-term state, typically lasting only
- * one round trip time (RTT).
- *
- * Primary. (tcp_secret_generating == tcp_secret_primary)
- * Used both for generation and primary verification.
- *
- * Retiring. (tcp_secret_retiring != tcp_secret_secondary)
- * Used for verification, until the first failure that can be
- * verified by the newer Generating secret. At that time, this
- * cookie's state is changed to Secondary, and the Generating
- * cookie's state is changed to Primary. This is a short-term state,
- * typically lasting only one round trip time (RTT).
- *
- * Secondary. (tcp_secret_retiring == tcp_secret_secondary)
- * Used for secondary verification, after primary verification
- * failures. This state lasts no more than twice the Maximum Segment
- * Lifetime (2MSL). Then, the secret is discarded.
- */
-struct tcp_cookie_secret {
- /* The secret is divided into two parts. The digest part is the
- * equivalent of previously hashing a secret and saving the state,
- * and serves as an initialization vector (IV). The message part
- * serves as the trailing secret.
- */
- u32 secrets[COOKIE_WORKSPACE_WORDS];
- unsigned long expires;
-};
-
-#define TCP_SECRET_1MSL (HZ * TCP_PAWS_MSL)
-#define TCP_SECRET_2MSL (HZ * TCP_PAWS_MSL * 2)
-#define TCP_SECRET_LIFE (HZ * 600)
-
-static struct tcp_cookie_secret tcp_secret_one;
-static struct tcp_cookie_secret tcp_secret_two;
-
-/* Essentially a circular list, without dynamic allocation. */
-static struct tcp_cookie_secret *tcp_secret_generating;
-static struct tcp_cookie_secret *tcp_secret_primary;
-static struct tcp_cookie_secret *tcp_secret_retiring;
-static struct tcp_cookie_secret *tcp_secret_secondary;
-
-static DEFINE_SPINLOCK(tcp_secret_locker);
-
-/* Select a pseudo-random word in the cookie workspace.
- */
-static inline u32 tcp_cookie_work(const u32 *ws, const int n)
-{
- return ws[COOKIE_DIGEST_WORDS + ((COOKIE_MESSAGE_WORDS-1) & ws[n])];
-}
-
-/* Fill bakery[COOKIE_WORKSPACE_WORDS] with generator, updating as needed.
- * Called in softirq context.
- * Returns: 0 for success.
- */
-int tcp_cookie_generator(u32 *bakery)
-{
- unsigned long jiffy = jiffies;
-
- if (unlikely(time_after_eq(jiffy, tcp_secret_generating->expires))) {
- spin_lock_bh(&tcp_secret_locker);
- if (!time_after_eq(jiffy, tcp_secret_generating->expires)) {
- /* refreshed by another */
- memcpy(bakery,
- &tcp_secret_generating->secrets[0],
- COOKIE_WORKSPACE_WORDS);
- } else {
- /* still needs refreshing */
- get_random_bytes(bakery, COOKIE_WORKSPACE_WORDS);
-
- /* The first time, paranoia assumes that the
- * randomization function isn't as strong. But,
- * this secret initialization is delayed until
- * the last possible moment (packet arrival).
- * Although that time is observable, it is
- * unpredictably variable. Mash in the most
- * volatile clock bits available, and expire the
- * secret extra quickly.
- */
- if (unlikely(tcp_secret_primary->expires ==
- tcp_secret_secondary->expires)) {
- struct timespec tv;
-
- getnstimeofday(&tv);
- bakery[COOKIE_DIGEST_WORDS+0] ^=
- (u32)tv.tv_nsec;
-
- tcp_secret_secondary->expires = jiffy
- + TCP_SECRET_1MSL
- + (0x0f & tcp_cookie_work(bakery, 0));
- } else {
- tcp_secret_secondary->expires = jiffy
- + TCP_SECRET_LIFE
- + (0xff & tcp_cookie_work(bakery, 1));
- tcp_secret_primary->expires = jiffy
- + TCP_SECRET_2MSL
- + (0x1f & tcp_cookie_work(bakery, 2));
- }
- memcpy(&tcp_secret_secondary->secrets[0],
- bakery, COOKIE_WORKSPACE_WORDS);
-
- rcu_assign_pointer(tcp_secret_generating,
- tcp_secret_secondary);
- rcu_assign_pointer(tcp_secret_retiring,
- tcp_secret_primary);
- /*
- * Neither call_rcu() nor synchronize_rcu() needed.
- * Retiring data is not freed. It is replaced after
- * further (locked) pointer updates, and a quiet time
- * (minimum 1MSL, maximum LIFE - 2MSL).
- */
- }
- spin_unlock_bh(&tcp_secret_locker);
- } else {
- rcu_read_lock_bh();
- memcpy(bakery,
- &rcu_dereference(tcp_secret_generating)->secrets[0],
- COOKIE_WORKSPACE_WORDS);
- rcu_read_unlock_bh();
- }
- return 0;
-}
-EXPORT_SYMBOL(tcp_cookie_generator);
-
void tcp_done(struct sock *sk)
{
struct request_sock *req = tcp_sk(sk)->fastopen_rsk;
@@ -3590,7 +3333,6 @@
unsigned long limit;
int max_rshare, max_wshare, cnt;
unsigned int i;
- unsigned long jiffy = jiffies;
BUILD_BUG_ON(sizeof(struct tcp_skb_cb) > sizeof(skb->cb));
@@ -3666,13 +3408,5 @@
tcp_register_congestion_control(&tcp_reno);
- memset(&tcp_secret_one.secrets[0], 0, sizeof(tcp_secret_one.secrets));
- memset(&tcp_secret_two.secrets[0], 0, sizeof(tcp_secret_two.secrets));
- tcp_secret_one.expires = jiffy; /* past due */
- tcp_secret_two.expires = jiffy; /* past due */
- tcp_secret_generating = &tcp_secret_one;
- tcp_secret_primary = &tcp_secret_one;
- tcp_secret_retiring = &tcp_secret_two;
- tcp_secret_secondary = &tcp_secret_two;
tcp_tasklet_init();
}
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 0d9bdac..19f0149 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -98,7 +98,7 @@
int sysctl_tcp_thin_dupack __read_mostly;
int sysctl_tcp_moderate_rcvbuf __read_mostly = 1;
-int sysctl_tcp_early_retrans __read_mostly = 2;
+int sysctl_tcp_early_retrans __read_mostly = 3;
#define FLAG_DATA 0x01 /* Incoming frame contained data. */
#define FLAG_WIN_UPDATE 0x02 /* Incoming ACK was a window update. */
@@ -2150,15 +2150,16 @@
* max(RTT/4, 2msec) unless ack has ECE mark, no RTT samples
* available, or RTO is scheduled to fire first.
*/
- if (sysctl_tcp_early_retrans < 2 || (flag & FLAG_ECE) || !tp->srtt)
+ if (sysctl_tcp_early_retrans < 2 || sysctl_tcp_early_retrans > 3 ||
+ (flag & FLAG_ECE) || !tp->srtt)
return false;
delay = max_t(unsigned long, (tp->srtt >> 5), msecs_to_jiffies(2));
if (!time_after(inet_csk(sk)->icsk_timeout, (jiffies + delay)))
return false;
- inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, delay, TCP_RTO_MAX);
- tp->early_retrans_delayed = 1;
+ inet_csk_reset_xmit_timer(sk, ICSK_TIME_EARLY_RETRANS, delay,
+ TCP_RTO_MAX);
return true;
}
@@ -2321,7 +2322,7 @@
* interval if appropriate.
*/
if (tp->do_early_retrans && !tp->retrans_out && tp->sacked_out &&
- (tp->packets_out == (tp->sacked_out + 1) && tp->packets_out < 4) &&
+ (tp->packets_out >= (tp->sacked_out + 1) && tp->packets_out < 4) &&
!tcp_may_send_now(sk))
return !tcp_pause_early_retransmit(sk, flag);
@@ -2681,6 +2682,7 @@
struct tcp_sock *tp = tcp_sk(sk);
tp->high_seq = tp->snd_nxt;
+ tp->tlp_high_seq = 0;
tp->snd_cwnd_cnt = 0;
tp->prior_cwnd = tp->snd_cwnd;
tp->prr_delivered = 0;
@@ -3081,6 +3083,7 @@
*/
void tcp_rearm_rto(struct sock *sk)
{
+ const struct inet_connection_sock *icsk = inet_csk(sk);
struct tcp_sock *tp = tcp_sk(sk);
/* If the retrans timer is currently being used by Fast Open
@@ -3094,12 +3097,13 @@
} else {
u32 rto = inet_csk(sk)->icsk_rto;
/* Offset the time elapsed after installing regular RTO */
- if (tp->early_retrans_delayed) {
+ if (icsk->icsk_pending == ICSK_TIME_EARLY_RETRANS ||
+ icsk->icsk_pending == ICSK_TIME_LOSS_PROBE) {
struct sk_buff *skb = tcp_write_queue_head(sk);
const u32 rto_time_stamp = TCP_SKB_CB(skb)->when + rto;
s32 delta = (s32)(rto_time_stamp - tcp_time_stamp);
/* delta may not be positive if the socket is locked
- * when the delayed ER timer fires and is rescheduled.
+ * when the retrans timer fires and is rescheduled.
*/
if (delta > 0)
rto = delta;
@@ -3107,7 +3111,6 @@
inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, rto,
TCP_RTO_MAX);
}
- tp->early_retrans_delayed = 0;
}
/* This function is called when the delayed ER timer fires. TCP enters
@@ -3567,6 +3570,38 @@
}
}
+/* This routine deals with acks during a TLP episode.
+ * Ref: loss detection algorithm in draft-dukkipati-tcpm-tcp-loss-probe.
+ */
+static void tcp_process_tlp_ack(struct sock *sk, u32 ack, int flag)
+{
+ struct tcp_sock *tp = tcp_sk(sk);
+ bool is_tlp_dupack = (ack == tp->tlp_high_seq) &&
+ !(flag & (FLAG_SND_UNA_ADVANCED |
+ FLAG_NOT_DUP | FLAG_DATA_SACKED));
+
+ /* Mark the end of TLP episode on receiving TLP dupack or when
+ * ack is after tlp_high_seq.
+ */
+ if (is_tlp_dupack) {
+ tp->tlp_high_seq = 0;
+ return;
+ }
+
+ if (after(ack, tp->tlp_high_seq)) {
+ tp->tlp_high_seq = 0;
+ /* Don't reduce cwnd if DSACK arrives for TLP retrans. */
+ if (!(flag & FLAG_DSACKING_ACK)) {
+ tcp_init_cwnd_reduction(sk, true);
+ tcp_set_ca_state(sk, TCP_CA_CWR);
+ tcp_end_cwnd_reduction(sk);
+ tcp_set_ca_state(sk, TCP_CA_Open);
+ NET_INC_STATS_BH(sock_net(sk),
+ LINUX_MIB_TCPLOSSPROBERECOVERY);
+ }
+ }
+}
+
/* This routine deals with incoming acks, but not outgoing ones. */
static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
{
@@ -3601,7 +3636,8 @@
if (after(ack, tp->snd_nxt))
goto invalid_ack;
- if (tp->early_retrans_delayed)
+ if (icsk->icsk_pending == ICSK_TIME_EARLY_RETRANS ||
+ icsk->icsk_pending == ICSK_TIME_LOSS_PROBE)
tcp_rearm_rto(sk);
if (after(ack, prior_snd_una))
@@ -3673,11 +3709,17 @@
tcp_cong_avoid(sk, ack, prior_in_flight);
}
+ if (tp->tlp_high_seq)
+ tcp_process_tlp_ack(sk, ack, flag);
+
if ((flag & FLAG_FORWARD_PROGRESS) || !(flag & FLAG_NOT_DUP)) {
struct dst_entry *dst = __sk_dst_get(sk);
if (dst)
dst_confirm(dst);
}
+
+ if (icsk->icsk_pending == ICSK_TIME_RETRANS)
+ tcp_schedule_loss_probe(sk);
return 1;
no_queue:
@@ -3691,6 +3733,9 @@
*/
if (tcp_send_head(sk))
tcp_ack_probe(sk);
+
+ if (tp->tlp_high_seq)
+ tcp_process_tlp_ack(sk, ack, flag);
return 1;
invalid_ack:
@@ -3715,8 +3760,8 @@
* But, this can also be called on packets in the established flow when
* the fast version below fails.
*/
-void tcp_parse_options(const struct sk_buff *skb, struct tcp_options_received *opt_rx,
- const u8 **hvpp, int estab,
+void tcp_parse_options(const struct sk_buff *skb,
+ struct tcp_options_received *opt_rx, int estab,
struct tcp_fastopen_cookie *foc)
{
const unsigned char *ptr;
@@ -3800,31 +3845,6 @@
*/
break;
#endif
- case TCPOPT_COOKIE:
- /* This option is variable length.
- */
- switch (opsize) {
- case TCPOLEN_COOKIE_BASE:
- /* not yet implemented */
- break;
- case TCPOLEN_COOKIE_PAIR:
- /* not yet implemented */
- break;
- case TCPOLEN_COOKIE_MIN+0:
- case TCPOLEN_COOKIE_MIN+2:
- case TCPOLEN_COOKIE_MIN+4:
- case TCPOLEN_COOKIE_MIN+6:
- case TCPOLEN_COOKIE_MAX:
- /* 16-bit multiple */
- opt_rx->cookie_plus = opsize;
- *hvpp = ptr;
- break;
- default:
- /* ignore option */
- break;
- }
- break;
-
case TCPOPT_EXP:
/* Fast Open option shares code 254 using a
* 16 bits magic number. It's valid only in
@@ -3870,8 +3890,7 @@
* If it is wrong it falls back on tcp_parse_options().
*/
static bool tcp_fast_parse_options(const struct sk_buff *skb,
- const struct tcphdr *th,
- struct tcp_sock *tp, const u8 **hvpp)
+ const struct tcphdr *th, struct tcp_sock *tp)
{
/* In the spirit of fast parsing, compare doff directly to constant
* values. Because equality is used, short doff can be ignored here.
@@ -3885,7 +3904,7 @@
return true;
}
- tcp_parse_options(skb, &tp->rx_opt, hvpp, 1, NULL);
+ tcp_parse_options(skb, &tp->rx_opt, 1, NULL);
if (tp->rx_opt.saw_tstamp)
tp->rx_opt.rcv_tsecr -= tp->tsoffset;
@@ -5266,12 +5285,10 @@
static bool tcp_validate_incoming(struct sock *sk, struct sk_buff *skb,
const struct tcphdr *th, int syn_inerr)
{
- const u8 *hash_location;
struct tcp_sock *tp = tcp_sk(sk);
/* RFC1323: H1. Apply PAWS check first. */
- if (tcp_fast_parse_options(skb, th, tp, &hash_location) &&
- tp->rx_opt.saw_tstamp &&
+ if (tcp_fast_parse_options(skb, th, tp) && tp->rx_opt.saw_tstamp &&
tcp_paws_discard(sk, skb)) {
if (!th->rst) {
NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_PAWSESTABREJECTED);
@@ -5625,12 +5642,11 @@
if (mss == tp->rx_opt.user_mss) {
struct tcp_options_received opt;
- const u8 *hash_location;
/* Get original SYNACK MSS value if user MSS sets mss_clamp */
tcp_clear_options(&opt);
opt.user_mss = opt.mss_clamp = 0;
- tcp_parse_options(synack, &opt, &hash_location, 0, NULL);
+ tcp_parse_options(synack, &opt, 0, NULL);
mss = opt.mss_clamp;
}
@@ -5661,14 +5677,12 @@
static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb,
const struct tcphdr *th, unsigned int len)
{
- const u8 *hash_location;
struct inet_connection_sock *icsk = inet_csk(sk);
struct tcp_sock *tp = tcp_sk(sk);
- struct tcp_cookie_values *cvp = tp->cookie_values;
struct tcp_fastopen_cookie foc = { .len = -1 };
int saved_clamp = tp->rx_opt.mss_clamp;
- tcp_parse_options(skb, &tp->rx_opt, &hash_location, 0, &foc);
+ tcp_parse_options(skb, &tp->rx_opt, 0, &foc);
if (tp->rx_opt.saw_tstamp)
tp->rx_opt.rcv_tsecr -= tp->tsoffset;
@@ -5765,30 +5779,6 @@
* is initialized. */
tp->copied_seq = tp->rcv_nxt;
- if (cvp != NULL &&
- cvp->cookie_pair_size > 0 &&
- tp->rx_opt.cookie_plus > 0) {
- int cookie_size = tp->rx_opt.cookie_plus
- - TCPOLEN_COOKIE_BASE;
- int cookie_pair_size = cookie_size
- + cvp->cookie_desired;
-
- /* A cookie extension option was sent and returned.
- * Note that each incoming SYNACK replaces the
- * Responder cookie. The initial exchange is most
- * fragile, as protection against spoofing relies
- * entirely upon the sequence and timestamp (above).
- * This replacement strategy allows the correct pair to
- * pass through, while any others will be filtered via
- * Responder verification later.
- */
- if (sizeof(cvp->cookie_pair) >= cookie_pair_size) {
- memcpy(&cvp->cookie_pair[cvp->cookie_desired],
- hash_location, cookie_size);
- cvp->cookie_pair_size = cookie_pair_size;
- }
- }
-
smp_mb();
tcp_finish_connect(sk, skb);
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index d09203c..2278669 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -838,7 +838,6 @@
*/
static int tcp_v4_send_synack(struct sock *sk, struct dst_entry *dst,
struct request_sock *req,
- struct request_values *rvp,
u16 queue_mapping,
bool nocache)
{
@@ -851,7 +850,7 @@
if (!dst && (dst = inet_csk_route_req(sk, &fl4, req)) == NULL)
return -1;
- skb = tcp_make_synack(sk, dst, req, rvp, NULL);
+ skb = tcp_make_synack(sk, dst, req, NULL);
if (skb) {
__tcp_v4_send_check(skb, ireq->loc_addr, ireq->rmt_addr);
@@ -868,10 +867,9 @@
return err;
}
-static int tcp_v4_rtx_synack(struct sock *sk, struct request_sock *req,
- struct request_values *rvp)
+static int tcp_v4_rtx_synack(struct sock *sk, struct request_sock *req)
{
- int res = tcp_v4_send_synack(sk, NULL, req, rvp, 0, false);
+ int res = tcp_v4_send_synack(sk, NULL, req, 0, false);
if (!res)
TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_RETRANSSEGS);
@@ -1371,8 +1369,7 @@
static int tcp_v4_conn_req_fastopen(struct sock *sk,
struct sk_buff *skb,
struct sk_buff *skb_synack,
- struct request_sock *req,
- struct request_values *rvp)
+ struct request_sock *req)
{
struct tcp_sock *tp = tcp_sk(sk);
struct request_sock_queue *queue = &inet_csk(sk)->icsk_accept_queue;
@@ -1467,9 +1464,7 @@
int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
{
- struct tcp_extend_values tmp_ext;
struct tcp_options_received tmp_opt;
- const u8 *hash_location;
struct request_sock *req;
struct inet_request_sock *ireq;
struct tcp_sock *tp = tcp_sk(sk);
@@ -1519,42 +1514,7 @@
tcp_clear_options(&tmp_opt);
tmp_opt.mss_clamp = TCP_MSS_DEFAULT;
tmp_opt.user_mss = tp->rx_opt.user_mss;
- tcp_parse_options(skb, &tmp_opt, &hash_location, 0,
- want_cookie ? NULL : &foc);
-
- if (tmp_opt.cookie_plus > 0 &&
- tmp_opt.saw_tstamp &&
- !tp->rx_opt.cookie_out_never &&
- (sysctl_tcp_cookie_size > 0 ||
- (tp->cookie_values != NULL &&
- tp->cookie_values->cookie_desired > 0))) {
- u8 *c;
- u32 *mess = &tmp_ext.cookie_bakery[COOKIE_DIGEST_WORDS];
- int l = tmp_opt.cookie_plus - TCPOLEN_COOKIE_BASE;
-
- if (tcp_cookie_generator(&tmp_ext.cookie_bakery[0]) != 0)
- goto drop_and_release;
-
- /* Secret recipe starts with IP addresses */
- *mess++ ^= (__force u32)daddr;
- *mess++ ^= (__force u32)saddr;
-
- /* plus variable length Initiator Cookie */
- c = (u8 *)mess;
- while (l-- > 0)
- *c++ ^= *hash_location++;
-
- want_cookie = false; /* not our kind of cookie */
- tmp_ext.cookie_out_never = 0; /* false */
- tmp_ext.cookie_plus = tmp_opt.cookie_plus;
- } else if (!tp->rx_opt.cookie_in_always) {
- /* redundant indications, but ensure initialization. */
- tmp_ext.cookie_out_never = 1; /* true */
- tmp_ext.cookie_plus = 0;
- } else {
- goto drop_and_release;
- }
- tmp_ext.cookie_in_always = tp->rx_opt.cookie_in_always;
+ tcp_parse_options(skb, &tmp_opt, 0, want_cookie ? NULL : &foc);
if (want_cookie && !tmp_opt.saw_tstamp)
tcp_clear_options(&tmp_opt);
@@ -1636,7 +1596,6 @@
* of tcp_v4_send_synack()->tcp_select_initial_window().
*/
skb_synack = tcp_make_synack(sk, dst, req,
- (struct request_values *)&tmp_ext,
fastopen_cookie_present(&valid_foc) ? &valid_foc : NULL);
if (skb_synack) {
@@ -1660,8 +1619,7 @@
if (fastopen_cookie_present(&foc) && foc.len != 0)
NET_INC_STATS_BH(sock_net(sk),
LINUX_MIB_TCPFASTOPENPASSIVEFAIL);
- } else if (tcp_v4_conn_req_fastopen(sk, skb, skb_synack, req,
- (struct request_values *)&tmp_ext))
+ } else if (tcp_v4_conn_req_fastopen(sk, skb, skb_synack, req))
goto drop_and_free;
return 0;
@@ -1950,6 +1908,50 @@
}
}
+/* Packet is added to VJ-style prequeue for processing in process
+ * context, if a reader task is waiting. Apparently, this exciting
+ * idea (VJ's mail "Re: query about TCP header on tcp-ip" of 07 Sep 93)
+ * failed somewhere. Latency? Burstiness? Well, at least now we will
+ * see, why it failed. 8)8) --ANK
+ *
+ */
+bool tcp_prequeue(struct sock *sk, struct sk_buff *skb)
+{
+ struct tcp_sock *tp = tcp_sk(sk);
+
+ if (sysctl_tcp_low_latency || !tp->ucopy.task)
+ return false;
+
+ if (skb->len <= tcp_hdrlen(skb) &&
+ skb_queue_len(&tp->ucopy.prequeue) == 0)
+ return false;
+
+ __skb_queue_tail(&tp->ucopy.prequeue, skb);
+ tp->ucopy.memory += skb->truesize;
+ if (tp->ucopy.memory > sk->sk_rcvbuf) {
+ struct sk_buff *skb1;
+
+ BUG_ON(sock_owned_by_user(sk));
+
+ while ((skb1 = __skb_dequeue(&tp->ucopy.prequeue)) != NULL) {
+ sk_backlog_rcv(sk, skb1);
+ NET_INC_STATS_BH(sock_net(sk),
+ LINUX_MIB_TCPPREQUEUEDROPPED);
+ }
+
+ tp->ucopy.memory = 0;
+ } else if (skb_queue_len(&tp->ucopy.prequeue) == 1) {
+ wake_up_interruptible_sync_poll(sk_sleep(sk),
+ POLLIN | POLLRDNORM | POLLRDBAND);
+ if (!inet_csk_ack_scheduled(sk))
+ inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK,
+ (3 * tcp_rto_min(sk)) / 4,
+ TCP_RTO_MAX);
+ }
+ return true;
+}
+EXPORT_SYMBOL(tcp_prequeue);
+
/*
* From tcp_input.c
*/
@@ -2197,12 +2199,6 @@
if (inet_csk(sk)->icsk_bind_hash)
inet_put_port(sk);
- /* TCP Cookie Transactions */
- if (tp->cookie_values != NULL) {
- kref_put(&tp->cookie_values->kref,
- tcp_cookie_values_release);
- tp->cookie_values = NULL;
- }
BUG_ON(tp->fastopen_rsk != NULL);
/* If socket is aborted during connect operation */
@@ -2659,7 +2655,9 @@
__u16 srcp = ntohs(inet->inet_sport);
int rx_queue;
- if (icsk->icsk_pending == ICSK_TIME_RETRANS) {
+ if (icsk->icsk_pending == ICSK_TIME_RETRANS ||
+ icsk->icsk_pending == ICSK_TIME_EARLY_RETRANS ||
+ icsk->icsk_pending == ICSK_TIME_LOSS_PROBE) {
timer_active = 1;
timer_expires = icsk->icsk_timeout;
} else if (icsk->icsk_pending == ICSK_TIME_PROBE0) {
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index b83a49c..8f0234f 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -93,13 +93,12 @@
const struct tcphdr *th)
{
struct tcp_options_received tmp_opt;
- const u8 *hash_location;
struct tcp_timewait_sock *tcptw = tcp_twsk((struct sock *)tw);
bool paws_reject = false;
tmp_opt.saw_tstamp = 0;
if (th->doff > (sizeof(*th) >> 2) && tcptw->tw_ts_recent_stamp) {
- tcp_parse_options(skb, &tmp_opt, &hash_location, 0, NULL);
+ tcp_parse_options(skb, &tmp_opt, 0, NULL);
if (tmp_opt.saw_tstamp) {
tmp_opt.rcv_tsecr -= tcptw->tw_ts_offset;
@@ -388,32 +387,6 @@
struct tcp_request_sock *treq = tcp_rsk(req);
struct inet_connection_sock *newicsk = inet_csk(newsk);
struct tcp_sock *newtp = tcp_sk(newsk);
- struct tcp_sock *oldtp = tcp_sk(sk);
- struct tcp_cookie_values *oldcvp = oldtp->cookie_values;
-
- /* TCP Cookie Transactions require space for the cookie pair,
- * as it differs for each connection. There is no need to
- * copy any s_data_payload stored at the original socket.
- * Failure will prevent resuming the connection.
- *
- * Presumed copied, in order of appearance:
- * cookie_in_always, cookie_out_never
- */
- if (oldcvp != NULL) {
- struct tcp_cookie_values *newcvp =
- kzalloc(sizeof(*newtp->cookie_values),
- GFP_ATOMIC);
-
- if (newcvp != NULL) {
- kref_init(&newcvp->kref);
- newcvp->cookie_desired =
- oldcvp->cookie_desired;
- newtp->cookie_values = newcvp;
- } else {
- /* Not Yet Implemented */
- newtp->cookie_values = NULL;
- }
- }
/* Now setup tcp_sock */
newtp->pred_flags = 0;
@@ -422,8 +395,7 @@
newtp->rcv_nxt = treq->rcv_isn + 1;
newtp->snd_sml = newtp->snd_una =
- newtp->snd_nxt = newtp->snd_up =
- treq->snt_isn + 1 + tcp_s_data_size(oldtp);
+ newtp->snd_nxt = newtp->snd_up = treq->snt_isn + 1;
tcp_prequeue_init(newtp);
INIT_LIST_HEAD(&newtp->tsq_node);
@@ -440,6 +412,7 @@
newtp->fackets_out = 0;
newtp->snd_ssthresh = TCP_INFINITE_SSTHRESH;
tcp_enable_early_retrans(newtp);
+ newtp->tlp_high_seq = 0;
/* So many TCP implementations out there (incorrectly) count the
* initial SYN frame in their delayed-ACK and congestion control
@@ -459,8 +432,7 @@
tcp_set_ca_state(newsk, TCP_CA_Open);
tcp_init_xmit_timers(newsk);
skb_queue_head_init(&newtp->out_of_order_queue);
- newtp->write_seq = newtp->pushed_seq =
- treq->snt_isn + 1 + tcp_s_data_size(oldtp);
+ newtp->write_seq = newtp->pushed_seq = treq->snt_isn + 1;
newtp->rx_opt.saw_tstamp = 0;
@@ -537,7 +509,6 @@
bool fastopen)
{
struct tcp_options_received tmp_opt;
- const u8 *hash_location;
struct sock *child;
const struct tcphdr *th = tcp_hdr(skb);
__be32 flg = tcp_flag_word(th) & (TCP_FLAG_RST|TCP_FLAG_SYN|TCP_FLAG_ACK);
@@ -547,7 +518,7 @@
tmp_opt.saw_tstamp = 0;
if (th->doff > (sizeof(struct tcphdr)>>2)) {
- tcp_parse_options(skb, &tmp_opt, &hash_location, 0, NULL);
+ tcp_parse_options(skb, &tmp_opt, 0, NULL);
if (tmp_opt.saw_tstamp) {
tmp_opt.ts_recent = req->ts_recent;
@@ -647,7 +618,7 @@
*/
if ((flg & TCP_FLAG_ACK) && !fastopen &&
(TCP_SKB_CB(skb)->ack_seq !=
- tcp_rsk(req)->snt_isn + 1 + tcp_s_data_size(tcp_sk(sk))))
+ tcp_rsk(req)->snt_isn + 1))
return sk;
/* Also, it would be not so bad idea to check rcv_tsecr, which
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 817fbb3..e787ece 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -65,15 +65,13 @@
/* By default, RFC2861 behavior. */
int sysctl_tcp_slow_start_after_idle __read_mostly = 1;
-int sysctl_tcp_cookie_size __read_mostly = 0; /* TCP_COOKIE_MAX */
-EXPORT_SYMBOL_GPL(sysctl_tcp_cookie_size);
-
static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle,
int push_one, gfp_t gfp);
/* Account for new data that has been sent to the network. */
static void tcp_event_new_data_sent(struct sock *sk, const struct sk_buff *skb)
{
+ struct inet_connection_sock *icsk = inet_csk(sk);
struct tcp_sock *tp = tcp_sk(sk);
unsigned int prior_packets = tp->packets_out;
@@ -85,7 +83,8 @@
tp->frto_counter = 3;
tp->packets_out += tcp_skb_pcount(skb);
- if (!prior_packets || tp->early_retrans_delayed)
+ if (!prior_packets || icsk->icsk_pending == ICSK_TIME_EARLY_RETRANS ||
+ icsk->icsk_pending == ICSK_TIME_LOSS_PROBE)
tcp_rearm_rto(sk);
}
@@ -384,7 +383,6 @@
#define OPTION_TS (1 << 1)
#define OPTION_MD5 (1 << 2)
#define OPTION_WSCALE (1 << 3)
-#define OPTION_COOKIE_EXTENSION (1 << 4)
#define OPTION_FAST_OPEN_COOKIE (1 << 8)
struct tcp_out_options {
@@ -398,36 +396,6 @@
struct tcp_fastopen_cookie *fastopen_cookie; /* Fast open cookie */
};
-/* The sysctl int routines are generic, so check consistency here.
- */
-static u8 tcp_cookie_size_check(u8 desired)
-{
- int cookie_size;
-
- if (desired > 0)
- /* previously specified */
- return desired;
-
- cookie_size = ACCESS_ONCE(sysctl_tcp_cookie_size);
- if (cookie_size <= 0)
- /* no default specified */
- return 0;
-
- if (cookie_size <= TCP_COOKIE_MIN)
- /* value too small, specify minimum */
- return TCP_COOKIE_MIN;
-
- if (cookie_size >= TCP_COOKIE_MAX)
- /* value too large, specify maximum */
- return TCP_COOKIE_MAX;
-
- if (cookie_size & 1)
- /* 8-bit multiple, illegal, fix it */
- cookie_size++;
-
- return (u8)cookie_size;
-}
-
/* Write previously computed TCP options to the packet.
*
* Beware: Something in the Internet is very sensitive to the ordering of
@@ -446,27 +414,9 @@
{
u16 options = opts->options; /* mungable copy */
- /* Having both authentication and cookies for security is redundant,
- * and there's certainly not enough room. Instead, the cookie-less
- * extension variant is proposed.
- *
- * Consider the pessimal case with authentication. The options
- * could look like:
- * COOKIE|MD5(20) + MSS(4) + SACK|TS(12) + WSCALE(4) == 40
- */
if (unlikely(OPTION_MD5 & options)) {
- if (unlikely(OPTION_COOKIE_EXTENSION & options)) {
- *ptr++ = htonl((TCPOPT_COOKIE << 24) |
- (TCPOLEN_COOKIE_BASE << 16) |
- (TCPOPT_MD5SIG << 8) |
- TCPOLEN_MD5SIG);
- } else {
- *ptr++ = htonl((TCPOPT_NOP << 24) |
- (TCPOPT_NOP << 16) |
- (TCPOPT_MD5SIG << 8) |
- TCPOLEN_MD5SIG);
- }
- options &= ~OPTION_COOKIE_EXTENSION;
+ *ptr++ = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) |
+ (TCPOPT_MD5SIG << 8) | TCPOLEN_MD5SIG);
/* overload cookie hash location */
opts->hash_location = (__u8 *)ptr;
ptr += 4;
@@ -495,44 +445,6 @@
*ptr++ = htonl(opts->tsecr);
}
- /* Specification requires after timestamp, so do it now.
- *
- * Consider the pessimal case without authentication. The options
- * could look like:
- * MSS(4) + SACK|TS(12) + COOKIE(20) + WSCALE(4) == 40
- */
- if (unlikely(OPTION_COOKIE_EXTENSION & options)) {
- __u8 *cookie_copy = opts->hash_location;
- u8 cookie_size = opts->hash_size;
-
- /* 8-bit multiple handled in tcp_cookie_size_check() above,
- * and elsewhere.
- */
- if (0x2 & cookie_size) {
- __u8 *p = (__u8 *)ptr;
-
- /* 16-bit multiple */
- *p++ = TCPOPT_COOKIE;
- *p++ = TCPOLEN_COOKIE_BASE + cookie_size;
- *p++ = *cookie_copy++;
- *p++ = *cookie_copy++;
- ptr++;
- cookie_size -= 2;
- } else {
- /* 32-bit multiple */
- *ptr++ = htonl(((TCPOPT_NOP << 24) |
- (TCPOPT_NOP << 16) |
- (TCPOPT_COOKIE << 8) |
- TCPOLEN_COOKIE_BASE) +
- cookie_size);
- }
-
- if (cookie_size > 0) {
- memcpy(ptr, cookie_copy, cookie_size);
- ptr += (cookie_size / 4);
- }
- }
-
if (unlikely(OPTION_SACK_ADVERTISE & options)) {
*ptr++ = htonl((TCPOPT_NOP << 24) |
(TCPOPT_NOP << 16) |
@@ -591,11 +503,7 @@
struct tcp_md5sig_key **md5)
{
struct tcp_sock *tp = tcp_sk(sk);
- struct tcp_cookie_values *cvp = tp->cookie_values;
unsigned int remaining = MAX_TCP_OPTION_SPACE;
- u8 cookie_size = (!tp->rx_opt.cookie_out_never && cvp != NULL) ?
- tcp_cookie_size_check(cvp->cookie_desired) :
- 0;
struct tcp_fastopen_request *fastopen = tp->fastopen_req;
#ifdef CONFIG_TCP_MD5SIG
@@ -647,52 +555,7 @@
tp->syn_fastopen = 1;
}
}
- /* Note that timestamps are required by the specification.
- *
- * Odd numbers of bytes are prohibited by the specification, ensuring
- * that the cookie is 16-bit aligned, and the resulting cookie pair is
- * 32-bit aligned.
- */
- if (*md5 == NULL &&
- (OPTION_TS & opts->options) &&
- cookie_size > 0) {
- int need = TCPOLEN_COOKIE_BASE + cookie_size;
- if (0x2 & need) {
- /* 32-bit multiple */
- need += 2; /* NOPs */
-
- if (need > remaining) {
- /* try shrinking cookie to fit */
- cookie_size -= 2;
- need -= 4;
- }
- }
- while (need > remaining && TCP_COOKIE_MIN <= cookie_size) {
- cookie_size -= 4;
- need -= 4;
- }
- if (TCP_COOKIE_MIN <= cookie_size) {
- opts->options |= OPTION_COOKIE_EXTENSION;
- opts->hash_location = (__u8 *)&cvp->cookie_pair[0];
- opts->hash_size = cookie_size;
-
- /* Remember for future incarnations. */
- cvp->cookie_desired = cookie_size;
-
- if (cvp->cookie_desired != cvp->cookie_pair_size) {
- /* Currently use random bytes as a nonce,
- * assuming these are completely unpredictable
- * by hostile users of the same system.
- */
- get_random_bytes(&cvp->cookie_pair[0],
- cookie_size);
- cvp->cookie_pair_size = cookie_size;
- }
-
- remaining -= need;
- }
- }
return MAX_TCP_OPTION_SPACE - remaining;
}
@@ -702,14 +565,10 @@
unsigned int mss, struct sk_buff *skb,
struct tcp_out_options *opts,
struct tcp_md5sig_key **md5,
- struct tcp_extend_values *xvp,
struct tcp_fastopen_cookie *foc)
{
struct inet_request_sock *ireq = inet_rsk(req);
unsigned int remaining = MAX_TCP_OPTION_SPACE;
- u8 cookie_plus = (xvp != NULL && !xvp->cookie_out_never) ?
- xvp->cookie_plus :
- 0;
#ifdef CONFIG_TCP_MD5SIG
*md5 = tcp_rsk(req)->af_specific->md5_lookup(sk, req);
@@ -757,28 +616,7 @@
remaining -= need;
}
}
- /* Similar rationale to tcp_syn_options() applies here, too.
- * If the <SYN> options fit, the same options should fit now!
- */
- if (*md5 == NULL &&
- ireq->tstamp_ok &&
- cookie_plus > TCPOLEN_COOKIE_BASE) {
- int need = cookie_plus; /* has TCPOLEN_COOKIE_BASE */
- if (0x2 & need) {
- /* 32-bit multiple */
- need += 2; /* NOPs */
- }
- if (need <= remaining) {
- opts->options |= OPTION_COOKIE_EXTENSION;
- opts->hash_size = cookie_plus - TCPOLEN_COOKIE_BASE;
- remaining -= need;
- } else {
- /* There's no error return, so flag it. */
- xvp->cookie_out_never = 1; /* true */
- opts->hash_size = 0;
- }
- }
return MAX_TCP_OPTION_SPACE - remaining;
}
@@ -1958,6 +1796,9 @@
* snd_up-64k-mss .. snd_up cannot be large. However, taking into
* account rare use of URG, this is not a big flaw.
*
+ * Send at most one packet when push_one > 0. Temporarily ignore
+ * cwnd limit to force at most one packet out when push_one == 2.
+
* Returns true, if no segments are in flight and we have queued segments,
* but cannot send anything now because of SWS or another problem.
*/
@@ -1993,8 +1834,13 @@
goto repair; /* Skip network transmission */
cwnd_quota = tcp_cwnd_test(tp, skb);
- if (!cwnd_quota)
- break;
+ if (!cwnd_quota) {
+ if (push_one == 2)
+ /* Force out a loss probe pkt. */
+ cwnd_quota = 1;
+ else
+ break;
+ }
if (unlikely(!tcp_snd_wnd_test(tp, skb, mss_now)))
break;
@@ -2048,10 +1894,129 @@
if (likely(sent_pkts)) {
if (tcp_in_cwnd_reduction(sk))
tp->prr_out += sent_pkts;
+
+ /* Send one loss probe per tail loss episode. */
+ if (push_one != 2)
+ tcp_schedule_loss_probe(sk);
tcp_cwnd_validate(sk);
return false;
}
- return !tp->packets_out && tcp_send_head(sk);
+ return (push_one == 2) || (!tp->packets_out && tcp_send_head(sk));
+}
+
+bool tcp_schedule_loss_probe(struct sock *sk)
+{
+ struct inet_connection_sock *icsk = inet_csk(sk);
+ struct tcp_sock *tp = tcp_sk(sk);
+ u32 timeout, tlp_time_stamp, rto_time_stamp;
+ u32 rtt = tp->srtt >> 3;
+
+ if (WARN_ON(icsk->icsk_pending == ICSK_TIME_EARLY_RETRANS))
+ return false;
+ /* No consecutive loss probes. */
+ if (WARN_ON(icsk->icsk_pending == ICSK_TIME_LOSS_PROBE)) {
+ tcp_rearm_rto(sk);
+ return false;
+ }
+ /* Don't do any loss probe on a Fast Open connection before 3WHS
+ * finishes.
+ */
+ if (sk->sk_state == TCP_SYN_RECV)
+ return false;
+
+ /* TLP is only scheduled when next timer event is RTO. */
+ if (icsk->icsk_pending != ICSK_TIME_RETRANS)
+ return false;
+
+ /* Schedule a loss probe in 2*RTT for SACK capable connections
+ * in Open state, that are either limited by cwnd or application.
+ */
+ if (sysctl_tcp_early_retrans < 3 || !rtt || !tp->packets_out ||
+ !tcp_is_sack(tp) || inet_csk(sk)->icsk_ca_state != TCP_CA_Open)
+ return false;
+
+ if ((tp->snd_cwnd > tcp_packets_in_flight(tp)) &&
+ tcp_send_head(sk))
+ return false;
+
+ /* Probe timeout is at least 1.5*rtt + TCP_DELACK_MAX to account
+ * for delayed ack when there's one outstanding packet.
+ */
+ timeout = rtt << 1;
+ if (tp->packets_out == 1)
+ timeout = max_t(u32, timeout,
+ (rtt + (rtt >> 1) + TCP_DELACK_MAX));
+ timeout = max_t(u32, timeout, msecs_to_jiffies(10));
+
+ /* If RTO is shorter, just schedule TLP in its place. */
+ tlp_time_stamp = tcp_time_stamp + timeout;
+ rto_time_stamp = (u32)inet_csk(sk)->icsk_timeout;
+ if ((s32)(tlp_time_stamp - rto_time_stamp) > 0) {
+ s32 delta = rto_time_stamp - tcp_time_stamp;
+ if (delta > 0)
+ timeout = delta;
+ }
+
+ inet_csk_reset_xmit_timer(sk, ICSK_TIME_LOSS_PROBE, timeout,
+ TCP_RTO_MAX);
+ return true;
+}
+
+/* When probe timeout (PTO) fires, send a new segment if one exists, else
+ * retransmit the last segment.
+ */
+void tcp_send_loss_probe(struct sock *sk)
+{
+ struct tcp_sock *tp = tcp_sk(sk);
+ struct sk_buff *skb;
+ int pcount;
+ int mss = tcp_current_mss(sk);
+ int err = -1;
+
+ if (tcp_send_head(sk) != NULL) {
+ err = tcp_write_xmit(sk, mss, TCP_NAGLE_OFF, 2, GFP_ATOMIC);
+ goto rearm_timer;
+ }
+
+ /* At most one outstanding TLP retransmission. */
+ if (tp->tlp_high_seq)
+ goto rearm_timer;
+
+ /* Retransmit last segment. */
+ skb = tcp_write_queue_tail(sk);
+ if (WARN_ON(!skb))
+ goto rearm_timer;
+
+ pcount = tcp_skb_pcount(skb);
+ if (WARN_ON(!pcount))
+ goto rearm_timer;
+
+ if ((pcount > 1) && (skb->len > (pcount - 1) * mss)) {
+ if (unlikely(tcp_fragment(sk, skb, (pcount - 1) * mss, mss)))
+ goto rearm_timer;
+ skb = tcp_write_queue_tail(sk);
+ }
+
+ if (WARN_ON(!skb || !tcp_skb_pcount(skb)))
+ goto rearm_timer;
+
+ /* Probe with zero data doesn't trigger fast recovery. */
+ if (skb->len > 0)
+ err = __tcp_retransmit_skb(sk, skb);
+
+ /* Record snd_nxt for loss detection. */
+ if (likely(!err))
+ tp->tlp_high_seq = tp->snd_nxt;
+
+rearm_timer:
+ inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS,
+ inet_csk(sk)->icsk_rto,
+ TCP_RTO_MAX);
+
+ if (likely(!err))
+ NET_INC_STATS_BH(sock_net(sk),
+ LINUX_MIB_TCPLOSSPROBES);
+ return;
}
/* Push out any pending frames which were held back due to
@@ -2672,32 +2637,24 @@
* sk: listener socket
* dst: dst entry attached to the SYNACK
* req: request_sock pointer
- * rvp: request_values pointer
*
* Allocate one skb and build a SYNACK packet.
* @dst is consumed : Caller should not use it again.
*/
struct sk_buff *tcp_make_synack(struct sock *sk, struct dst_entry *dst,
struct request_sock *req,
- struct request_values *rvp,
struct tcp_fastopen_cookie *foc)
{
struct tcp_out_options opts;
- struct tcp_extend_values *xvp = tcp_xv(rvp);
struct inet_request_sock *ireq = inet_rsk(req);
struct tcp_sock *tp = tcp_sk(sk);
- const struct tcp_cookie_values *cvp = tp->cookie_values;
struct tcphdr *th;
struct sk_buff *skb;
struct tcp_md5sig_key *md5;
int tcp_header_size;
int mss;
- int s_data_desired = 0;
- if (cvp != NULL && cvp->s_data_constant && cvp->s_data_desired)
- s_data_desired = cvp->s_data_desired;
- skb = alloc_skb(MAX_TCP_HEADER + 15 + s_data_desired,
- sk_gfp_atomic(sk, GFP_ATOMIC));
+ skb = alloc_skb(MAX_TCP_HEADER + 15, sk_gfp_atomic(sk, GFP_ATOMIC));
if (unlikely(!skb)) {
dst_release(dst);
return NULL;
@@ -2739,9 +2696,8 @@
else
#endif
TCP_SKB_CB(skb)->when = tcp_time_stamp;
- tcp_header_size = tcp_synack_options(sk, req, mss,
- skb, &opts, &md5, xvp, foc)
- + sizeof(*th);
+ tcp_header_size = tcp_synack_options(sk, req, mss, skb, &opts, &md5,
+ foc) + sizeof(*th);
skb_push(skb, tcp_header_size);
skb_reset_transport_header(skb);
@@ -2759,40 +2715,6 @@
tcp_init_nondata_skb(skb, tcp_rsk(req)->snt_isn,
TCPHDR_SYN | TCPHDR_ACK);
- if (OPTION_COOKIE_EXTENSION & opts.options) {
- if (s_data_desired) {
- u8 *buf = skb_put(skb, s_data_desired);
-
- /* copy data directly from the listening socket. */
- memcpy(buf, cvp->s_data_payload, s_data_desired);
- TCP_SKB_CB(skb)->end_seq += s_data_desired;
- }
-
- if (opts.hash_size > 0) {
- __u32 workspace[SHA_WORKSPACE_WORDS];
- u32 *mess = &xvp->cookie_bakery[COOKIE_DIGEST_WORDS];
- u32 *tail = &mess[COOKIE_MESSAGE_WORDS-1];
-
- /* Secret recipe depends on the Timestamp, (future)
- * Sequence and Acknowledgment Numbers, Initiator
- * Cookie, and others handled by IP variant caller.
- */
- *tail-- ^= opts.tsval;
- *tail-- ^= tcp_rsk(req)->rcv_isn + 1;
- *tail-- ^= TCP_SKB_CB(skb)->seq + 1;
-
- /* recommended */
- *tail-- ^= (((__force u32)th->dest << 16) | (__force u32)th->source);
- *tail-- ^= (u32)(unsigned long)cvp; /* per sockopt */
-
- sha_transform((__u32 *)&xvp->cookie_bakery[0],
- (char *)mess,
- &workspace[0]);
- opts.hash_location =
- (__u8 *)&xvp->cookie_bakery[0];
- }
- }
-
th->seq = htonl(TCP_SKB_CB(skb)->seq);
/* XXX data is queued and acked as is. No buffer/window check */
th->ack_seq = htonl(tcp_rsk(req)->rcv_nxt);
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index b78aac3..eeccf79 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -342,10 +342,6 @@
struct tcp_sock *tp = tcp_sk(sk);
struct inet_connection_sock *icsk = inet_csk(sk);
- if (tp->early_retrans_delayed) {
- tcp_resume_early_retransmit(sk);
- return;
- }
if (tp->fastopen_rsk) {
WARN_ON_ONCE(sk->sk_state != TCP_SYN_RECV &&
sk->sk_state != TCP_FIN_WAIT1);
@@ -360,6 +356,8 @@
WARN_ON(tcp_write_queue_empty(sk));
+ tp->tlp_high_seq = 0;
+
if (!tp->snd_wnd && !sock_flag(sk, SOCK_DEAD) &&
!((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV))) {
/* Receiver dastardly shrinks window. Our retransmits
@@ -495,13 +493,20 @@
}
event = icsk->icsk_pending;
- icsk->icsk_pending = 0;
switch (event) {
+ case ICSK_TIME_EARLY_RETRANS:
+ tcp_resume_early_retransmit(sk);
+ break;
+ case ICSK_TIME_LOSS_PROBE:
+ tcp_send_loss_probe(sk);
+ break;
case ICSK_TIME_RETRANS:
+ icsk->icsk_pending = 0;
tcp_retransmit_timer(sk);
break;
case ICSK_TIME_PROBE0:
+ icsk->icsk_pending = 0;
tcp_probe_timer(sk);
break;
}
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 0a073a2..7117d14 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -2279,31 +2279,88 @@
int udp4_ufo_send_check(struct sk_buff *skb)
{
- const struct iphdr *iph;
- struct udphdr *uh;
-
- if (!pskb_may_pull(skb, sizeof(*uh)))
+ if (!pskb_may_pull(skb, sizeof(struct udphdr)))
return -EINVAL;
- iph = ip_hdr(skb);
- uh = udp_hdr(skb);
+ if (likely(!skb->encapsulation)) {
+ const struct iphdr *iph;
+ struct udphdr *uh;
- uh->check = ~csum_tcpudp_magic(iph->saddr, iph->daddr, skb->len,
- IPPROTO_UDP, 0);
- skb->csum_start = skb_transport_header(skb) - skb->head;
- skb->csum_offset = offsetof(struct udphdr, check);
- skb->ip_summed = CHECKSUM_PARTIAL;
+ iph = ip_hdr(skb);
+ uh = udp_hdr(skb);
+
+ uh->check = ~csum_tcpudp_magic(iph->saddr, iph->daddr, skb->len,
+ IPPROTO_UDP, 0);
+ skb->csum_start = skb_transport_header(skb) - skb->head;
+ skb->csum_offset = offsetof(struct udphdr, check);
+ skb->ip_summed = CHECKSUM_PARTIAL;
+ }
return 0;
}
+static struct sk_buff *skb_udp_tunnel_segment(struct sk_buff *skb,
+ netdev_features_t features)
+{
+ struct sk_buff *segs = ERR_PTR(-EINVAL);
+ int mac_len = skb->mac_len;
+ int tnl_hlen = skb_inner_mac_header(skb) - skb_transport_header(skb);
+ int outer_hlen;
+ netdev_features_t enc_features;
+
+ if (unlikely(!pskb_may_pull(skb, tnl_hlen)))
+ goto out;
+
+ skb->encapsulation = 0;
+ __skb_pull(skb, tnl_hlen);
+ skb_reset_mac_header(skb);
+ skb_set_network_header(skb, skb_inner_network_offset(skb));
+ skb->mac_len = skb_inner_network_offset(skb);
+
+ /* segment inner packet. */
+ enc_features = skb->dev->hw_enc_features & netif_skb_features(skb);
+ segs = skb_mac_gso_segment(skb, enc_features);
+ if (!segs || IS_ERR(segs))
+ goto out;
+
+ outer_hlen = skb_tnl_header_len(skb);
+ skb = segs;
+ do {
+ struct udphdr *uh;
+ int udp_offset = outer_hlen - tnl_hlen;
+
+ skb->mac_len = mac_len;
+
+ skb_push(skb, outer_hlen);
+ skb_reset_mac_header(skb);
+ skb_set_network_header(skb, mac_len);
+ skb_set_transport_header(skb, udp_offset);
+ uh = udp_hdr(skb);
+ uh->len = htons(skb->len - udp_offset);
+
+ /* csum segment if tunnel sets skb with csum. */
+ if (unlikely(uh->check)) {
+ struct iphdr *iph = ip_hdr(skb);
+
+ uh->check = ~csum_tcpudp_magic(iph->saddr, iph->daddr,
+ skb->len - udp_offset,
+ IPPROTO_UDP, 0);
+ uh->check = csum_fold(skb_checksum(skb, udp_offset,
+ skb->len - udp_offset, 0));
+ if (uh->check == 0)
+ uh->check = CSUM_MANGLED_0;
+
+ }
+ skb->ip_summed = CHECKSUM_NONE;
+ } while ((skb = skb->next));
+out:
+ return segs;
+}
+
struct sk_buff *udp4_ufo_fragment(struct sk_buff *skb,
netdev_features_t features)
{
struct sk_buff *segs = ERR_PTR(-EINVAL);
unsigned int mss;
- int offset;
- __wsum csum;
-
mss = skb_shinfo(skb)->gso_size;
if (unlikely(skb->len <= mss))
goto out;
@@ -2313,6 +2370,7 @@
int type = skb_shinfo(skb)->gso_type;
if (unlikely(type & ~(SKB_GSO_UDP | SKB_GSO_DODGY |
+ SKB_GSO_UDP_TUNNEL |
SKB_GSO_GRE) ||
!(type & (SKB_GSO_UDP))))
goto out;
@@ -2323,20 +2381,27 @@
goto out;
}
- /* Do software UFO. Complete and fill in the UDP checksum as HW cannot
- * do checksum of UDP packets sent as multiple IP fragments.
- */
- offset = skb_checksum_start_offset(skb);
- csum = skb_checksum(skb, offset, skb->len - offset, 0);
- offset += skb->csum_offset;
- *(__sum16 *)(skb->data + offset) = csum_fold(csum);
- skb->ip_summed = CHECKSUM_NONE;
-
/* Fragment the skb. IP headers of the fragments are updated in
* inet_gso_segment()
*/
- segs = skb_segment(skb, features);
+ if (skb->encapsulation && skb_shinfo(skb)->gso_type & SKB_GSO_UDP_TUNNEL)
+ segs = skb_udp_tunnel_segment(skb, features);
+ else {
+ int offset;
+ __wsum csum;
+
+ /* Do software UFO. Complete and fill in the UDP checksum as
+ * HW cannot do checksum of UDP packets sent as multiple
+ * IP fragments.
+ */
+ offset = skb_checksum_start_offset(skb);
+ csum = skb_checksum(skb, offset, skb->len - offset, 0);
+ offset += skb->csum_offset;
+ *(__sum16 *)(skb->data + offset) = csum_fold(csum);
+ skb->ip_summed = CHECKSUM_NONE;
+
+ segs = skb_segment(skb, features);
+ }
out:
return segs;
}
-
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index f2c7e61..fa36a67 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -605,6 +605,74 @@
return err;
}
+static int inet6_netconf_dump_devconf(struct sk_buff *skb,
+ struct netlink_callback *cb)
+{
+ struct net *net = sock_net(skb->sk);
+ int h, s_h;
+ int idx, s_idx;
+ struct net_device *dev;
+ struct inet6_dev *idev;
+ struct hlist_head *head;
+
+ s_h = cb->args[0];
+ s_idx = idx = cb->args[1];
+
+ for (h = s_h; h < NETDEV_HASHENTRIES; h++, s_idx = 0) {
+ idx = 0;
+ head = &net->dev_index_head[h];
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(dev, head, index_hlist) {
+ if (idx < s_idx)
+ goto cont;
+ idev = __in6_dev_get(dev);
+ if (!idev)
+ goto cont;
+
+ if (inet6_netconf_fill_devconf(skb, dev->ifindex,
+ &idev->cnf,
+ NETLINK_CB(cb->skb).portid,
+ cb->nlh->nlmsg_seq,
+ RTM_NEWNETCONF,
+ NLM_F_MULTI,
+ -1) <= 0) {
+ rcu_read_unlock();
+ goto done;
+ }
+cont:
+ idx++;
+ }
+ rcu_read_unlock();
+ }
+ if (h == NETDEV_HASHENTRIES) {
+ if (inet6_netconf_fill_devconf(skb, NETCONFA_IFINDEX_ALL,
+ net->ipv6.devconf_all,
+ NETLINK_CB(cb->skb).portid,
+ cb->nlh->nlmsg_seq,
+ RTM_NEWNETCONF, NLM_F_MULTI,
+ -1) <= 0)
+ goto done;
+ else
+ h++;
+ }
+ if (h == NETDEV_HASHENTRIES + 1) {
+ if (inet6_netconf_fill_devconf(skb, NETCONFA_IFINDEX_DEFAULT,
+ net->ipv6.devconf_dflt,
+ NETLINK_CB(cb->skb).portid,
+ cb->nlh->nlmsg_seq,
+ RTM_NEWNETCONF, NLM_F_MULTI,
+ -1) <= 0)
+ goto done;
+ else
+ h++;
+ }
+done:
+ cb->args[0] = h;
+ cb->args[1] = idx;
+
+ return skb->len;
+}
+
#ifdef CONFIG_SYSCTL
static void dev_forward_change(struct inet6_dev *idev)
{
@@ -4940,7 +5008,7 @@
__rtnl_register(PF_INET6, RTM_GETANYCAST, NULL,
inet6_dump_ifacaddr, NULL);
__rtnl_register(PF_INET6, RTM_GETNETCONF, inet6_netconf_get_devconf,
- NULL, NULL);
+ inet6_netconf_dump_devconf, NULL);
ipv6_addr_label_rtnl_register();
diff --git a/net/ipv6/addrlabel.c b/net/ipv6/addrlabel.c
index aad6435..6f226c8 100644
--- a/net/ipv6/addrlabel.c
+++ b/net/ipv6/addrlabel.c
@@ -436,10 +436,7 @@
if (!tb[IFAL_ADDRESS])
return -EINVAL;
-
pfx = nla_data(tb[IFAL_ADDRESS]);
- if (!pfx)
- return -EINVAL;
if (!tb[IFAL_LABEL])
return -EINVAL;
@@ -561,10 +558,7 @@
if (!tb[IFAL_ADDRESS])
return -EINVAL;
-
addr = nla_data(tb[IFAL_ADDRESS]);
- if (!addr)
- return -EINVAL;
rcu_read_lock();
p = __ipv6_addr_label(net, addr, ipv6_addr_type(addr), ifal->ifal_index);
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 6b793bf..f56277f 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -323,7 +323,7 @@
struct net_device *dev = NULL;
rcu_read_lock();
- if (addr_type & IPV6_ADDR_LINKLOCAL) {
+ if (__ipv6_addr_needs_scope_id(addr_type)) {
if (addr_len >= sizeof(struct sockaddr_in6) &&
addr->sin6_scope_id) {
/* Override any existing binding, if another one
@@ -471,8 +471,8 @@
sin->sin6_port = inet->inet_sport;
}
- if (ipv6_addr_type(&sin->sin6_addr) & IPV6_ADDR_LINKLOCAL)
- sin->sin6_scope_id = sk->sk_bound_dev_if;
+ sin->sin6_scope_id = ipv6_iface_scope_id(&sin->sin6_addr,
+ sk->sk_bound_dev_if);
*uaddr_len = sizeof(*sin);
return 0;
}
diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c
index f5a5478..4b56cbb 100644
--- a/net/ipv6/datagram.c
+++ b/net/ipv6/datagram.c
@@ -124,7 +124,7 @@
goto out;
}
- if (addr_type&IPV6_ADDR_LINKLOCAL) {
+ if (__ipv6_addr_needs_scope_id(addr_type)) {
if (addr_len >= sizeof(struct sockaddr_in6) &&
usin->sin6_scope_id) {
if (sk->sk_bound_dev_if &&
@@ -355,18 +355,19 @@
sin->sin6_family = AF_INET6;
sin->sin6_flowinfo = 0;
sin->sin6_port = serr->port;
- sin->sin6_scope_id = 0;
if (skb->protocol == htons(ETH_P_IPV6)) {
const struct ipv6hdr *ip6h = container_of((struct in6_addr *)(nh + serr->addr_offset),
struct ipv6hdr, daddr);
sin->sin6_addr = ip6h->daddr;
if (np->sndflow)
sin->sin6_flowinfo = ip6_flowinfo(ip6h);
- if (ipv6_addr_type(&sin->sin6_addr) & IPV6_ADDR_LINKLOCAL)
- sin->sin6_scope_id = IP6CB(skb)->iif;
+ sin->sin6_scope_id =
+ ipv6_iface_scope_id(&sin->sin6_addr,
+ IP6CB(skb)->iif);
} else {
ipv6_addr_set_v4mapped(*(__be32 *)(nh + serr->addr_offset),
&sin->sin6_addr);
+ sin->sin6_scope_id = 0;
}
}
@@ -376,18 +377,19 @@
if (serr->ee.ee_origin != SO_EE_ORIGIN_LOCAL) {
sin->sin6_family = AF_INET6;
sin->sin6_flowinfo = 0;
- sin->sin6_scope_id = 0;
if (skb->protocol == htons(ETH_P_IPV6)) {
sin->sin6_addr = ipv6_hdr(skb)->saddr;
if (np->rxopt.all)
ip6_datagram_recv_ctl(sk, msg, skb);
- if (ipv6_addr_type(&sin->sin6_addr) & IPV6_ADDR_LINKLOCAL)
- sin->sin6_scope_id = IP6CB(skb)->iif;
+ sin->sin6_scope_id =
+ ipv6_iface_scope_id(&sin->sin6_addr,
+ IP6CB(skb)->iif);
} else {
struct inet_sock *inet = inet_sk(sk);
ipv6_addr_set_v4mapped(ip_hdr(skb)->saddr,
&sin->sin6_addr);
+ sin->sin6_scope_id = 0;
if (inet->cmsg_flags)
ip_cmsg_recv(msg, skb);
}
@@ -592,7 +594,9 @@
sin6.sin6_addr = ipv6_hdr(skb)->daddr;
sin6.sin6_port = ports[1];
sin6.sin6_flowinfo = 0;
- sin6.sin6_scope_id = 0;
+ sin6.sin6_scope_id =
+ ipv6_iface_scope_id(&ipv6_hdr(skb)->daddr,
+ opt->iif);
put_cmsg(msg, SOL_IPV6, IPV6_ORIGDSTADDR, sizeof(sin6), &sin6);
}
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index fff5bdd..71b900c 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -434,7 +434,7 @@
* Source addr check
*/
- if (addr_type & IPV6_ADDR_LINKLOCAL)
+ if (__ipv6_addr_needs_scope_id(addr_type))
iif = skb->dev->ifindex;
/*
diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c
index 9bfab19..e4311cb 100644
--- a/net/ipv6/inet6_connection_sock.c
+++ b/net/ipv6/inet6_connection_sock.c
@@ -54,6 +54,10 @@
if (ipv6_rcv_saddr_equal(sk, sk2))
break;
}
+ if (!relax && reuse && sk2->sk_reuse &&
+ sk2->sk_state != TCP_LISTEN &&
+ ipv6_rcv_saddr_equal(sk, sk2))
+ break;
}
}
@@ -169,10 +173,8 @@
sin6->sin6_port = inet_sk(sk)->inet_dport;
/* We do not store received flowlabel for TCP */
sin6->sin6_flowinfo = 0;
- sin6->sin6_scope_id = 0;
- if (sk->sk_bound_dev_if &&
- ipv6_addr_type(&sin6->sin6_addr) & IPV6_ADDR_LINKLOCAL)
- sin6->sin6_scope_id = sk->sk_bound_dev_if;
+ sin6->sin6_scope_id = ipv6_iface_scope_id(&sin6->sin6_addr,
+ sk->sk_bound_dev_if);
}
EXPORT_SYMBOL_GPL(inet6_csk_addr2sockaddr);
diff --git a/net/ipv6/ip6_flowlabel.c b/net/ipv6/ip6_flowlabel.c
index b973ed3..46e8843 100644
--- a/net/ipv6/ip6_flowlabel.c
+++ b/net/ipv6/ip6_flowlabel.c
@@ -144,7 +144,9 @@
spin_lock(&ip6_fl_lock);
for (i=0; i<=FL_HASH_MASK; i++) {
- struct ip6_flowlabel *fl, **flp;
+ struct ip6_flowlabel *fl;
+ struct ip6_flowlabel __rcu **flp;
+
flp = &fl_ht[i];
while ((fl = rcu_dereference_protected(*flp,
lockdep_is_held(&ip6_fl_lock))) != NULL) {
@@ -179,7 +181,9 @@
spin_lock(&ip6_fl_lock);
for (i = 0; i <= FL_HASH_MASK; i++) {
- struct ip6_flowlabel *fl, **flp;
+ struct ip6_flowlabel *fl;
+ struct ip6_flowlabel __rcu **flp;
+
flp = &fl_ht[i];
while ((fl = rcu_dereference_protected(*flp,
lockdep_is_held(&ip6_fl_lock))) != NULL) {
@@ -506,7 +510,8 @@
struct ipv6_pinfo *np = inet6_sk(sk);
struct in6_flowlabel_req freq;
struct ipv6_fl_socklist *sfl1=NULL;
- struct ipv6_fl_socklist *sfl, **sflp;
+ struct ipv6_fl_socklist *sfl;
+ struct ipv6_fl_socklist __rcu **sflp;
struct ip6_flowlabel *fl, *fl1 = NULL;
diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c
index e4efffe2..6a6ba73 100644
--- a/net/ipv6/ip6_gre.c
+++ b/net/ipv6/ip6_gre.c
@@ -667,7 +667,6 @@
struct net_device_stats *stats = &tunnel->dev->stats;
int err = -1;
u8 proto;
- int pkt_len;
struct sk_buff *new_skb;
if (dev->type == ARPHRD_ETHER)
@@ -801,23 +800,9 @@
}
}
- nf_reset(skb);
- pkt_len = skb->len;
- err = ip6_local_out(skb);
-
- if (net_xmit_eval(err) == 0) {
- struct pcpu_tstats *tstats = this_cpu_ptr(tunnel->dev->tstats);
-
- tstats->tx_bytes += pkt_len;
- tstats->tx_packets++;
- } else {
- stats->tx_errors++;
- stats->tx_aborted_errors++;
- }
-
+ ip6tunnel_xmit(skb, dev);
if (ndst)
ip6_tnl_dst_store(tunnel, ndst);
-
return 0;
tx_err_link_failure:
stats->tx_carrier_errors++;
diff --git a/net/ipv6/ip6_offload.c b/net/ipv6/ip6_offload.c
index 8234c1d..71b766e 100644
--- a/net/ipv6/ip6_offload.c
+++ b/net/ipv6/ip6_offload.c
@@ -92,14 +92,12 @@
u8 *prevhdr;
int offset = 0;
- if (!(features & NETIF_F_V6_CSUM))
- features &= ~NETIF_F_SG;
-
if (unlikely(skb_shinfo(skb)->gso_type &
~(SKB_GSO_UDP |
SKB_GSO_DODGY |
SKB_GSO_TCP_ECN |
SKB_GSO_GRE |
+ SKB_GSO_UDP_TUNNEL |
SKB_GSO_TCPV6 |
0)))
goto out;
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index fff83cb..bef3fed 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -955,7 +955,6 @@
unsigned int max_headroom = sizeof(struct ipv6hdr);
u8 proto;
int err = -1;
- int pkt_len;
if (!fl6->flowi6_mark)
dst = ip6_tnl_dst_check(t);
@@ -1035,19 +1034,7 @@
ipv6h->nexthdr = proto;
ipv6h->saddr = fl6->saddr;
ipv6h->daddr = fl6->daddr;
- nf_reset(skb);
- pkt_len = skb->len;
- err = ip6_local_out(skb);
-
- if (net_xmit_eval(err) == 0) {
- struct pcpu_tstats *tstats = this_cpu_ptr(t->dev->tstats);
-
- tstats->tx_bytes += pkt_len;
- tstats->tx_packets++;
- } else {
- stats->tx_errors++;
- stats->tx_aborted_errors++;
- }
+ ip6tunnel_xmit(skb, dev);
if (ndst)
ip6_tnl_dst_store(t, ndst);
return 0;
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 76ef435..2712ab2 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -610,8 +610,6 @@
}
}
#endif
- if (!dev->addr_len)
- send_sllao = 0;
if (send_sllao)
optlen += ndisc_opt_addr_space(dev);
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 330b5e7..eedff8c 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -263,7 +263,7 @@
if (addr_type != IPV6_ADDR_ANY) {
struct net_device *dev = NULL;
- if (addr_type & IPV6_ADDR_LINKLOCAL) {
+ if (__ipv6_addr_needs_scope_id(addr_type)) {
if (addr_len >= sizeof(struct sockaddr_in6) &&
addr->sin6_scope_id) {
/* Override any existing binding, if another
@@ -498,9 +498,8 @@
sin6->sin6_port = 0;
sin6->sin6_addr = ipv6_hdr(skb)->saddr;
sin6->sin6_flowinfo = 0;
- sin6->sin6_scope_id = 0;
- if (ipv6_addr_type(&sin6->sin6_addr) & IPV6_ADDR_LINKLOCAL)
- sin6->sin6_scope_id = IP6CB(skb)->iif;
+ sin6->sin6_scope_id = ipv6_iface_scope_id(&sin6->sin6_addr,
+ IP6CB(skb)->iif);
}
sock_recv_ts_and_drops(msg, sk, skb);
@@ -802,7 +801,7 @@
if (addr_len >= sizeof(struct sockaddr_in6) &&
sin6->sin6_scope_id &&
- ipv6_addr_type(daddr)&IPV6_ADDR_LINKLOCAL)
+ __ipv6_addr_needs_scope_id(__ipv6_addr_type(daddr)))
fl6.flowi6_oif = sin6->sin6_scope_id;
} else {
if (sk->sk_state != TCP_ESTABLISHED)
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index 02f96dc..898e671 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -899,6 +899,8 @@
if ((iph->ttl = tiph->ttl) == 0)
iph->ttl = iph6->hop_limit;
+ skb->ip_summed = CHECKSUM_NONE;
+ ip_select_ident(iph, skb_dst(skb), NULL);
iptunnel_xmit(skb, dev);
return NETDEV_TX_OK;
diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c
index 8a0848b..d5dda20 100644
--- a/net/ipv6/syncookies.c
+++ b/net/ipv6/syncookies.c
@@ -149,7 +149,6 @@
struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
{
struct tcp_options_received tcp_opt;
- const u8 *hash_location;
struct inet_request_sock *ireq;
struct inet6_request_sock *ireq6;
struct tcp_request_sock *treq;
@@ -177,7 +176,7 @@
/* check for timestamp cookie support */
memset(&tcp_opt, 0, sizeof(tcp_opt));
- tcp_parse_options(skb, &tcp_opt, &hash_location, 0, NULL);
+ tcp_parse_options(skb, &tcp_opt, 0, NULL);
if (!cookie_check_timestamp(&tcp_opt, sock_net(sk), &ecn_ok))
goto out;
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index f6d629f..1033d2b 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -461,7 +461,6 @@
static int tcp_v6_send_synack(struct sock *sk, struct dst_entry *dst,
struct flowi6 *fl6,
struct request_sock *req,
- struct request_values *rvp,
u16 queue_mapping)
{
struct inet6_request_sock *treq = inet6_rsk(req);
@@ -473,7 +472,7 @@
if (!dst && (dst = inet6_csk_route_req(sk, fl6, req)) == NULL)
goto done;
- skb = tcp_make_synack(sk, dst, req, rvp, NULL);
+ skb = tcp_make_synack(sk, dst, req, NULL);
if (skb) {
__tcp_v6_send_check(skb, &treq->loc_addr, &treq->rmt_addr);
@@ -488,13 +487,12 @@
return err;
}
-static int tcp_v6_rtx_synack(struct sock *sk, struct request_sock *req,
- struct request_values *rvp)
+static int tcp_v6_rtx_synack(struct sock *sk, struct request_sock *req)
{
struct flowi6 fl6;
int res;
- res = tcp_v6_send_synack(sk, NULL, &fl6, req, rvp, 0);
+ res = tcp_v6_send_synack(sk, NULL, &fl6, req, 0);
if (!res)
TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_RETRANSSEGS);
return res;
@@ -947,9 +945,7 @@
*/
static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb)
{
- struct tcp_extend_values tmp_ext;
struct tcp_options_received tmp_opt;
- const u8 *hash_location;
struct request_sock *req;
struct inet6_request_sock *treq;
struct ipv6_pinfo *np = inet6_sk(sk);
@@ -987,50 +983,7 @@
tcp_clear_options(&tmp_opt);
tmp_opt.mss_clamp = IPV6_MIN_MTU - sizeof(struct tcphdr) - sizeof(struct ipv6hdr);
tmp_opt.user_mss = tp->rx_opt.user_mss;
- tcp_parse_options(skb, &tmp_opt, &hash_location, 0, NULL);
-
- if (tmp_opt.cookie_plus > 0 &&
- tmp_opt.saw_tstamp &&
- !tp->rx_opt.cookie_out_never &&
- (sysctl_tcp_cookie_size > 0 ||
- (tp->cookie_values != NULL &&
- tp->cookie_values->cookie_desired > 0))) {
- u8 *c;
- u32 *d;
- u32 *mess = &tmp_ext.cookie_bakery[COOKIE_DIGEST_WORDS];
- int l = tmp_opt.cookie_plus - TCPOLEN_COOKIE_BASE;
-
- if (tcp_cookie_generator(&tmp_ext.cookie_bakery[0]) != 0)
- goto drop_and_free;
-
- /* Secret recipe starts with IP addresses */
- d = (__force u32 *)&ipv6_hdr(skb)->daddr.s6_addr32[0];
- *mess++ ^= *d++;
- *mess++ ^= *d++;
- *mess++ ^= *d++;
- *mess++ ^= *d++;
- d = (__force u32 *)&ipv6_hdr(skb)->saddr.s6_addr32[0];
- *mess++ ^= *d++;
- *mess++ ^= *d++;
- *mess++ ^= *d++;
- *mess++ ^= *d++;
-
- /* plus variable length Initiator Cookie */
- c = (u8 *)mess;
- while (l-- > 0)
- *c++ ^= *hash_location++;
-
- want_cookie = false; /* not our kind of cookie */
- tmp_ext.cookie_out_never = 0; /* false */
- tmp_ext.cookie_plus = tmp_opt.cookie_plus;
- } else if (!tp->rx_opt.cookie_in_always) {
- /* redundant indications, but ensure initialization. */
- tmp_ext.cookie_out_never = 1; /* true */
- tmp_ext.cookie_plus = 0;
- } else {
- goto drop_and_free;
- }
- tmp_ext.cookie_in_always = tp->rx_opt.cookie_in_always;
+ tcp_parse_options(skb, &tmp_opt, 0, NULL);
if (want_cookie && !tmp_opt.saw_tstamp)
tcp_clear_options(&tmp_opt);
@@ -1108,7 +1061,6 @@
goto drop_and_release;
if (tcp_v6_send_synack(sk, dst, &fl6, req,
- (struct request_values *)&tmp_ext,
skb_get_queue_mapping(skb)) ||
want_cookie)
goto drop_and_free;
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index d8e5e85..da6019b 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -450,15 +450,16 @@
sin6->sin6_family = AF_INET6;
sin6->sin6_port = udp_hdr(skb)->source;
sin6->sin6_flowinfo = 0;
- sin6->sin6_scope_id = 0;
- if (is_udp4)
+ if (is_udp4) {
ipv6_addr_set_v4mapped(ip_hdr(skb)->saddr,
&sin6->sin6_addr);
- else {
+ sin6->sin6_scope_id = 0;
+ } else {
sin6->sin6_addr = ipv6_hdr(skb)->saddr;
- if (ipv6_addr_type(&sin6->sin6_addr) & IPV6_ADDR_LINKLOCAL)
- sin6->sin6_scope_id = IP6CB(skb)->iif;
+ sin6->sin6_scope_id =
+ ipv6_iface_scope_id(&sin6->sin6_addr,
+ IP6CB(skb)->iif);
}
}
@@ -1118,7 +1119,7 @@
if (addr_len >= sizeof(struct sockaddr_in6) &&
sin6->sin6_scope_id &&
- ipv6_addr_type(daddr)&IPV6_ADDR_LINKLOCAL)
+ __ipv6_addr_needs_scope_id(__ipv6_addr_type(daddr)))
fl6.flowi6_oif = sin6->sin6_scope_id;
} else {
if (sk->sk_state != TCP_ESTABLISHED)
diff --git a/net/ipv6/udp_offload.c b/net/ipv6/udp_offload.c
index cf05cf0..3bb3a89 100644
--- a/net/ipv6/udp_offload.c
+++ b/net/ipv6/udp_offload.c
@@ -21,6 +21,10 @@
const struct ipv6hdr *ipv6h;
struct udphdr *uh;
+ /* UDP Tunnel offload on ipv6 is not yet supported. */
+ if (skb->encapsulation)
+ return -EINVAL;
+
if (!pskb_may_pull(skb, sizeof(*uh)))
return -EINVAL;
@@ -56,7 +60,9 @@
/* Packet is from an untrusted source, reset gso_segs. */
int type = skb_shinfo(skb)->gso_type;
- if (unlikely(type & ~(SKB_GSO_UDP | SKB_GSO_DODGY |
+ if (unlikely(type & ~(SKB_GSO_UDP |
+ SKB_GSO_DODGY |
+ SKB_GSO_UDP_TUNNEL |
SKB_GSO_GRE) ||
!(type & (SKB_GSO_UDP))))
goto out;
diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index a4b7247..d61cd99 100644
--- a/net/openvswitch/datapath.c
+++ b/net/openvswitch/datapath.c
@@ -369,8 +369,8 @@
len = sizeof(struct ovs_header);
len += nla_total_size(skb->len);
len += nla_total_size(FLOW_BUFSIZE);
- if (upcall_info->cmd == OVS_PACKET_CMD_ACTION)
- len += nla_total_size(8);
+ if (upcall_info->userdata)
+ len += NLA_ALIGN(upcall_info->userdata->nla_len);
user_skb = genlmsg_new(len, GFP_ATOMIC);
if (!user_skb) {
@@ -387,8 +387,9 @@
nla_nest_end(user_skb, nla);
if (upcall_info->userdata)
- nla_put_u64(user_skb, OVS_PACKET_ATTR_USERDATA,
- nla_get_u64(upcall_info->userdata));
+ __nla_put(user_skb, OVS_PACKET_ATTR_USERDATA,
+ nla_len(upcall_info->userdata),
+ nla_data(upcall_info->userdata));
nla = __nla_reserve(user_skb, OVS_PACKET_ATTR_PACKET, skb->len);
@@ -544,7 +545,7 @@
{
static const struct nla_policy userspace_policy[OVS_USERSPACE_ATTR_MAX + 1] = {
[OVS_USERSPACE_ATTR_PID] = {.type = NLA_U32 },
- [OVS_USERSPACE_ATTR_USERDATA] = {.type = NLA_U64 },
+ [OVS_USERSPACE_ATTR_USERDATA] = {.type = NLA_UNSPEC },
};
struct nlattr *a[OVS_USERSPACE_ATTR_MAX + 1];
int error;
@@ -1628,7 +1629,7 @@
vport = ovs_vport_rtnl_rcu(dp, port_no);
if (!vport)
- return ERR_PTR(-ENOENT);
+ return ERR_PTR(-ENODEV);
return vport;
} else
return ERR_PTR(-EINVAL);
diff --git a/net/openvswitch/datapath.h b/net/openvswitch/datapath.h
index 031dfbf..9125ad5 100644
--- a/net/openvswitch/datapath.h
+++ b/net/openvswitch/datapath.h
@@ -119,7 +119,7 @@
* struct dp_upcall - metadata to include with a packet to send to userspace
* @cmd: One of %OVS_PACKET_CMD_*.
* @key: Becomes %OVS_PACKET_ATTR_KEY. Must be nonnull.
- * @userdata: If nonnull, its u64 value is extracted and passed to userspace as
+ * @userdata: If nonnull, its variable-length value is passed to userspace as
* %OVS_PACKET_ATTR_USERDATA.
* @pid: Netlink PID to which packet should be sent. If @pid is 0 then no
* packet is sent and the packet is accounted in the datapath's @n_lost
diff --git a/net/openvswitch/vport-internal_dev.c b/net/openvswitch/vport-internal_dev.c
index 0531de6..40f8a24 100644
--- a/net/openvswitch/vport-internal_dev.c
+++ b/net/openvswitch/vport-internal_dev.c
@@ -63,16 +63,6 @@
return stats;
}
-static int internal_dev_mac_addr(struct net_device *dev, void *p)
-{
- struct sockaddr *addr = p;
-
- if (!is_valid_ether_addr(addr->sa_data))
- return -EADDRNOTAVAIL;
- memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
- return 0;
-}
-
/* Called with rcu_read_lock_bh. */
static int internal_dev_xmit(struct sk_buff *skb, struct net_device *netdev)
{
@@ -126,7 +116,7 @@
.ndo_open = internal_dev_open,
.ndo_stop = internal_dev_stop,
.ndo_start_xmit = internal_dev_xmit,
- .ndo_set_mac_address = internal_dev_mac_addr,
+ .ndo_set_mac_address = eth_mac_addr,
.ndo_change_mtu = internal_dev_change_mtu,
.ndo_get_stats64 = internal_dev_get_stats,
};
@@ -138,6 +128,7 @@
netdev->netdev_ops = &internal_dev_netdev_ops;
netdev->priv_flags &= ~IFF_TX_SKB_SHARING;
+ netdev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
netdev->destructor = internal_dev_destructor;
SET_ETHTOOL_OPS(netdev, &internal_dev_ethtool_ops);
netdev->tx_queue_len = 0;
diff --git a/net/openvswitch/vport.h b/net/openvswitch/vport.h
index 3f7961e..aee7d43 100644
--- a/net/openvswitch/vport.h
+++ b/net/openvswitch/vport.h
@@ -68,10 +68,10 @@
/**
* struct vport - one port within a datapath
* @rcu: RCU callback head for deferred destruction.
- * @port_no: Index into @dp's @ports array.
* @dp: Datapath to which this port belongs.
* @upcall_portid: The Netlink port to use for packets received on this port that
* miss the flow table.
+ * @port_no: Index into @dp's @ports array.
* @hash_node: Element in @dev_table hash table in vport.c.
* @dp_hash_node: Element in @datapath->ports hash table in datapath.c.
* @ops: Class structure.
@@ -81,9 +81,9 @@
*/
struct vport {
struct rcu_head rcu;
- u16 port_no;
struct datapath *dp;
u32 upcall_portid;
+ u16 port_no;
struct hlist_node hash_node;
struct hlist_node dp_hash_node;
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 1d6793d..bd0d14c 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -181,6 +181,8 @@
struct packet_sock;
static int tpacket_snd(struct packet_sock *po, struct msghdr *msg);
+static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev,
+ struct packet_type *pt, struct net_device *orig_dev);
static void *packet_previous_frame(struct packet_sock *po,
struct packet_ring_buffer *rb,
@@ -973,11 +975,11 @@
static void *prb_lookup_block(struct packet_sock *po,
struct packet_ring_buffer *rb,
- unsigned int previous,
+ unsigned int idx,
int status)
{
struct tpacket_kbdq_core *pkc = GET_PBDQC_FROM_RB(rb);
- struct tpacket_block_desc *pbd = GET_PBLOCK_DESC(pkc, previous);
+ struct tpacket_block_desc *pbd = GET_PBLOCK_DESC(pkc, idx);
if (status != BLOCK_STATUS(pbd))
return NULL;
@@ -1041,6 +1043,29 @@
buff->head = buff->head != buff->frame_max ? buff->head+1 : 0;
}
+static bool packet_rcv_has_room(struct packet_sock *po, struct sk_buff *skb)
+{
+ struct sock *sk = &po->sk;
+ bool has_room;
+
+ if (po->prot_hook.func != tpacket_rcv)
+ return (atomic_read(&sk->sk_rmem_alloc) + skb->truesize)
+ <= sk->sk_rcvbuf;
+
+ spin_lock(&sk->sk_receive_queue.lock);
+ if (po->tp_version == TPACKET_V3)
+ has_room = prb_lookup_block(po, &po->rx_ring,
+ po->rx_ring.prb_bdqc.kactive_blk_num,
+ TP_STATUS_KERNEL);
+ else
+ has_room = packet_lookup_frame(po, &po->rx_ring,
+ po->rx_ring.head,
+ TP_STATUS_KERNEL);
+ spin_unlock(&sk->sk_receive_queue.lock);
+
+ return has_room;
+}
+
static void packet_sock_destruct(struct sock *sk)
{
skb_queue_purge(&sk->sk_error_queue);
@@ -1066,16 +1091,16 @@
return x;
}
-static struct sock *fanout_demux_hash(struct packet_fanout *f, struct sk_buff *skb, unsigned int num)
+static unsigned int fanout_demux_hash(struct packet_fanout *f,
+ struct sk_buff *skb,
+ unsigned int num)
{
- u32 idx, hash = skb->rxhash;
-
- idx = ((u64)hash * num) >> 32;
-
- return f->arr[idx];
+ return (((u64)skb->rxhash) * num) >> 32;
}
-static struct sock *fanout_demux_lb(struct packet_fanout *f, struct sk_buff *skb, unsigned int num)
+static unsigned int fanout_demux_lb(struct packet_fanout *f,
+ struct sk_buff *skb,
+ unsigned int num)
{
int cur, old;
@@ -1083,14 +1108,40 @@
while ((old = atomic_cmpxchg(&f->rr_cur, cur,
fanout_rr_next(f, num))) != cur)
cur = old;
- return f->arr[cur];
+ return cur;
}
-static struct sock *fanout_demux_cpu(struct packet_fanout *f, struct sk_buff *skb, unsigned int num)
+static unsigned int fanout_demux_cpu(struct packet_fanout *f,
+ struct sk_buff *skb,
+ unsigned int num)
{
- unsigned int cpu = smp_processor_id();
+ return smp_processor_id() % num;
+}
- return f->arr[cpu % num];
+static unsigned int fanout_demux_rollover(struct packet_fanout *f,
+ struct sk_buff *skb,
+ unsigned int idx, unsigned int skip,
+ unsigned int num)
+{
+ unsigned int i, j;
+
+ i = j = min_t(int, f->next[idx], num - 1);
+ do {
+ if (i != skip && packet_rcv_has_room(pkt_sk(f->arr[i]), skb)) {
+ if (i != j)
+ f->next[idx] = i;
+ return i;
+ }
+ if (++i == num)
+ i = 0;
+ } while (i != j);
+
+ return idx;
+}
+
+static bool fanout_has_flag(struct packet_fanout *f, u16 flag)
+{
+ return f->flags & (flag >> 8);
}
static int packet_rcv_fanout(struct sk_buff *skb, struct net_device *dev,
@@ -1099,7 +1150,7 @@
struct packet_fanout *f = pt->af_packet_priv;
unsigned int num = f->num_members;
struct packet_sock *po;
- struct sock *sk;
+ unsigned int idx;
if (!net_eq(dev_net(dev), read_pnet(&f->net)) ||
!num) {
@@ -1110,23 +1161,31 @@
switch (f->type) {
case PACKET_FANOUT_HASH:
default:
- if (f->defrag) {
+ if (fanout_has_flag(f, PACKET_FANOUT_FLAG_DEFRAG)) {
skb = ip_check_defrag(skb, IP_DEFRAG_AF_PACKET);
if (!skb)
return 0;
}
skb_get_rxhash(skb);
- sk = fanout_demux_hash(f, skb, num);
+ idx = fanout_demux_hash(f, skb, num);
break;
case PACKET_FANOUT_LB:
- sk = fanout_demux_lb(f, skb, num);
+ idx = fanout_demux_lb(f, skb, num);
break;
case PACKET_FANOUT_CPU:
- sk = fanout_demux_cpu(f, skb, num);
+ idx = fanout_demux_cpu(f, skb, num);
+ break;
+ case PACKET_FANOUT_ROLLOVER:
+ idx = fanout_demux_rollover(f, skb, 0, (unsigned int) -1, num);
break;
}
- po = pkt_sk(sk);
+ po = pkt_sk(f->arr[idx]);
+ if (fanout_has_flag(f, PACKET_FANOUT_FLAG_ROLLOVER) &&
+ unlikely(!packet_rcv_has_room(po, skb))) {
+ idx = fanout_demux_rollover(f, skb, idx, idx, num);
+ po = pkt_sk(f->arr[idx]);
+ }
return po->prot_hook.func(skb, dev, &po->prot_hook, orig_dev);
}
@@ -1175,10 +1234,13 @@
struct packet_sock *po = pkt_sk(sk);
struct packet_fanout *f, *match;
u8 type = type_flags & 0xff;
- u8 defrag = (type_flags & PACKET_FANOUT_FLAG_DEFRAG) ? 1 : 0;
+ u8 flags = type_flags >> 8;
int err;
switch (type) {
+ case PACKET_FANOUT_ROLLOVER:
+ if (type_flags & PACKET_FANOUT_FLAG_ROLLOVER)
+ return -EINVAL;
case PACKET_FANOUT_HASH:
case PACKET_FANOUT_LB:
case PACKET_FANOUT_CPU:
@@ -1203,7 +1265,7 @@
}
}
err = -EINVAL;
- if (match && match->defrag != defrag)
+ if (match && match->flags != flags)
goto out;
if (!match) {
err = -ENOMEM;
@@ -1213,7 +1275,7 @@
write_pnet(&match->net, sock_net(sk));
match->id = id;
match->type = type;
- match->defrag = defrag;
+ match->flags = flags;
atomic_set(&match->rr_cur, 0);
INIT_LIST_HEAD(&match->list);
spin_lock_init(&match->lock);
@@ -3240,7 +3302,8 @@
case PACKET_FANOUT:
val = (po->fanout ?
((u32)po->fanout->id |
- ((u32)po->fanout->type << 16)) :
+ ((u32)po->fanout->type << 16) |
+ ((u32)po->fanout->flags << 24)) :
0);
break;
case PACKET_TX_HAS_OFF:
diff --git a/net/packet/internal.h b/net/packet/internal.h
index e84cab8..e891f02 100644
--- a/net/packet/internal.h
+++ b/net/packet/internal.h
@@ -77,10 +77,11 @@
unsigned int num_members;
u16 id;
u8 type;
- u8 defrag;
+ u8 flags;
atomic_t rr_cur;
struct list_head list;
struct sock *arr[PACKET_FANOUT_MAX];
+ int next[PACKET_FANOUT_MAX];
spinlock_t lock;
atomic_t sk_ref;
struct packet_type prot_hook ____cacheline_aligned_in_smp;
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 571f1d2..79b1876 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -981,6 +981,7 @@
[TCA_HTB_INIT] = { .len = sizeof(struct tc_htb_glob) },
[TCA_HTB_CTAB] = { .type = NLA_BINARY, .len = TC_RTAB_SIZE },
[TCA_HTB_RTAB] = { .type = NLA_BINARY, .len = TC_RTAB_SIZE },
+ [TCA_HTB_DIRECT_QLEN] = { .type = NLA_U32 },
};
static void htb_work_func(struct work_struct *work)
@@ -994,7 +995,7 @@
static int htb_init(struct Qdisc *sch, struct nlattr *opt)
{
struct htb_sched *q = qdisc_priv(sch);
- struct nlattr *tb[TCA_HTB_INIT + 1];
+ struct nlattr *tb[TCA_HTB_MAX + 1];
struct tc_htb_glob *gopt;
int err;
int i;
@@ -1002,20 +1003,16 @@
if (!opt)
return -EINVAL;
- err = nla_parse_nested(tb, TCA_HTB_INIT, opt, htb_policy);
+ err = nla_parse_nested(tb, TCA_HTB_MAX, opt, htb_policy);
if (err < 0)
return err;
- if (tb[TCA_HTB_INIT] == NULL) {
- pr_err("HTB: hey probably you have bad tc tool ?\n");
+ if (!tb[TCA_HTB_INIT])
return -EINVAL;
- }
+
gopt = nla_data(tb[TCA_HTB_INIT]);
- if (gopt->version != HTB_VER >> 16) {
- pr_err("HTB: need tc/htb version %d (minor is %d), you have %d\n",
- HTB_VER >> 16, HTB_VER & 0xffff, gopt->version);
+ if (gopt->version != HTB_VER >> 16)
return -EINVAL;
- }
err = qdisc_class_hash_init(&q->clhash);
if (err < 0)
@@ -1027,10 +1024,13 @@
INIT_WORK(&q->work, htb_work_func);
skb_queue_head_init(&q->direct_queue);
- q->direct_qlen = qdisc_dev(sch)->tx_queue_len;
- if (q->direct_qlen < 2) /* some devices have zero tx_queue_len */
- q->direct_qlen = 2;
-
+ if (tb[TCA_HTB_DIRECT_QLEN])
+ q->direct_qlen = nla_get_u32(tb[TCA_HTB_DIRECT_QLEN]);
+ else {
+ q->direct_qlen = qdisc_dev(sch)->tx_queue_len;
+ if (q->direct_qlen < 2) /* some devices have zero tx_queue_len */
+ q->direct_qlen = 2;
+ }
if ((q->rate2quantum = gopt->rate2quantum) < 1)
q->rate2quantum = 1;
q->defcls = gopt->defcls;
@@ -1056,7 +1056,8 @@
nest = nla_nest_start(skb, TCA_OPTIONS);
if (nest == NULL)
goto nla_put_failure;
- if (nla_put(skb, TCA_HTB_INIT, sizeof(gopt), &gopt))
+ if (nla_put(skb, TCA_HTB_INIT, sizeof(gopt), &gopt) ||
+ nla_put_u32(skb, TCA_HTB_DIRECT_QLEN, q->direct_qlen))
goto nla_put_failure;
nla_nest_end(skb, nest);
@@ -1311,7 +1312,7 @@
struct htb_sched *q = qdisc_priv(sch);
struct htb_class *cl = (struct htb_class *)*arg, *parent;
struct nlattr *opt = tca[TCA_OPTIONS];
- struct nlattr *tb[__TCA_HTB_MAX];
+ struct nlattr *tb[TCA_HTB_MAX + 1];
struct tc_htb_opt *hopt;
/* extract all subattrs from opt attr */
diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c
index a70ace8..faf81b8 100644
--- a/net/vmw_vsock/vmci_transport.c
+++ b/net/vmw_vsock/vmci_transport.c
@@ -123,6 +123,14 @@
return err > 0 ? -err : err;
}
+static u32 vmci_transport_peer_rid(u32 peer_cid)
+{
+ if (VMADDR_CID_HYPERVISOR == peer_cid)
+ return VMCI_TRANSPORT_HYPERVISOR_PACKET_RID;
+
+ return VMCI_TRANSPORT_PACKET_RID;
+}
+
static inline void
vmci_transport_packet_init(struct vmci_transport_packet *pkt,
struct sockaddr_vm *src,
@@ -140,7 +148,7 @@
pkt->dg.src = vmci_make_handle(VMADDR_CID_ANY,
VMCI_TRANSPORT_PACKET_RID);
pkt->dg.dst = vmci_make_handle(dst->svm_cid,
- VMCI_TRANSPORT_PACKET_RID);
+ vmci_transport_peer_rid(dst->svm_cid));
pkt->dg.payload_size = sizeof(*pkt) - sizeof(pkt->dg);
pkt->version = VMCI_TRANSPORT_PACKET_VERSION;
pkt->type = type;
@@ -511,6 +519,9 @@
static bool vmci_transport_allow_dgram(struct vsock_sock *vsock, u32 peer_cid)
{
+ if (VMADDR_CID_HYPERVISOR == peer_cid)
+ return true;
+
if (vsock->cached_peer != peer_cid) {
vsock->cached_peer = peer_cid;
if (!vmci_transport_is_trusted(vsock, peer_cid) &&
@@ -631,7 +642,6 @@
static bool vmci_transport_stream_allow(u32 cid, u32 port)
{
static const u32 non_socket_contexts[] = {
- VMADDR_CID_HYPERVISOR,
VMADDR_CID_RESERVED,
};
int i;
@@ -670,7 +680,7 @@
*/
if (!vmci_transport_stream_allow(dg->src.context, -1)
- || VMCI_TRANSPORT_PACKET_RID != dg->src.resource)
+ || vmci_transport_peer_rid(dg->src.context) != dg->src.resource)
return VMCI_ERROR_NO_ACCESS;
if (VMCI_DG_SIZE(dg) < sizeof(*pkt))
diff --git a/net/vmw_vsock/vmci_transport.h b/net/vmw_vsock/vmci_transport.h
index 1bf9918..fd88ea8 100644
--- a/net/vmw_vsock/vmci_transport.h
+++ b/net/vmw_vsock/vmci_transport.h
@@ -28,6 +28,9 @@
/* The resource ID on which control packets are sent. */
#define VMCI_TRANSPORT_PACKET_RID 1
+/* The resource ID on which control packets are sent to the hypervisor. */
+#define VMCI_TRANSPORT_HYPERVISOR_PACKET_RID 15
+
#define VSOCK_PROTO_INVALID 0
#define VSOCK_PROTO_PKT_ON_NOTIFY (1 << 0)
#define VSOCK_PROTO_ALL_SUPPORTED (VSOCK_PROTO_PKT_ON_NOTIFY)
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 167c67d..23cea0f 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -1037,6 +1037,24 @@
return xfrm_policy_lookup_bytype(net, XFRM_POLICY_TYPE_MAIN, fl, family, dir);
}
+static int flow_to_policy_dir(int dir)
+{
+ if (XFRM_POLICY_IN == FLOW_DIR_IN &&
+ XFRM_POLICY_OUT == FLOW_DIR_OUT &&
+ XFRM_POLICY_FWD == FLOW_DIR_FWD)
+ return dir;
+
+ switch (dir) {
+ default:
+ case FLOW_DIR_IN:
+ return XFRM_POLICY_IN;
+ case FLOW_DIR_OUT:
+ return XFRM_POLICY_OUT;
+ case FLOW_DIR_FWD:
+ return XFRM_POLICY_FWD;
+ }
+}
+
static struct flow_cache_object *
xfrm_policy_lookup(struct net *net, const struct flowi *fl, u16 family,
u8 dir, struct flow_cache_object *old_obj, void *ctx)
@@ -1046,7 +1064,7 @@
if (old_obj)
xfrm_pol_put(container_of(old_obj, struct xfrm_policy, flo));
- pol = __xfrm_policy_lookup(net, fl, family, dir);
+ pol = __xfrm_policy_lookup(net, fl, family, flow_to_policy_dir(dir));
if (IS_ERR_OR_NULL(pol))
return ERR_CAST(pol);
@@ -1932,7 +1950,8 @@
* previous cache entry */
if (xdst == NULL) {
num_pols = 1;
- pols[0] = __xfrm_policy_lookup(net, fl, family, dir);
+ pols[0] = __xfrm_policy_lookup(net, fl, family,
+ flow_to_policy_dir(dir));
err = xfrm_expand_policies(fl, family, pols,
&num_pols, &num_xfrms);
if (err < 0)
diff --git a/tools/testing/selftests/Makefile b/tools/testing/selftests/Makefile
index 3cc0ad7..7f50078 100644
--- a/tools/testing/selftests/Makefile
+++ b/tools/testing/selftests/Makefile
@@ -5,6 +5,8 @@
TARGETS += cpu-hotplug
TARGETS += memory-hotplug
TARGETS += efivarfs
+TARGETS += net-socket
+TARGETS += net-afpacket
all:
for TARGET in $(TARGETS); do \
diff --git a/tools/testing/selftests/net-afpacket/Makefile b/tools/testing/selftests/net-afpacket/Makefile
new file mode 100644
index 0000000..45f2ffb7
--- /dev/null
+++ b/tools/testing/selftests/net-afpacket/Makefile
@@ -0,0 +1,18 @@
+# Makefile for net-socket selftests
+
+CC = $(CROSS_COMPILE)gcc
+CFLAGS = -Wall
+
+CFLAGS += -I../../../../usr/include/
+
+AF_PACKET_PROGS = psock_fanout
+
+all: $(AF_PACKET_PROGS)
+%: %.c
+ $(CC) $(CFLAGS) -o $@ $^
+
+run_tests: all
+ @/bin/sh ./run_afpackettests || echo "afpackettests: [FAIL]"
+
+clean:
+ $(RM) $(AF_PACKET_PROGS)
diff --git a/tools/testing/selftests/net-afpacket/psock_fanout.c b/tools/testing/selftests/net-afpacket/psock_fanout.c
new file mode 100644
index 0000000..f765f09
--- /dev/null
+++ b/tools/testing/selftests/net-afpacket/psock_fanout.c
@@ -0,0 +1,388 @@
+/*
+ * Copyright 2013 Google Inc.
+ * Author: Willem de Bruijn (willemb@google.com)
+ *
+ * A basic test of packet socket fanout behavior.
+ *
+ * Control:
+ * - create fanout fails as expected with illegal flag combinations
+ * - join fanout fails as expected with diverging types or flags
+ *
+ * Datapath:
+ * Open a pair of packet sockets and a pair of INET sockets, send a known
+ * number of packets across the two INET sockets and count the number of
+ * packets enqueued onto the two packet sockets.
+ *
+ * The test currently runs for
+ * - PACKET_FANOUT_HASH
+ * - PACKET_FANOUT_HASH with PACKET_FANOUT_FLAG_ROLLOVER
+ * - PACKET_FANOUT_LB
+ * - PACKET_FANOUT_CPU
+ * - PACKET_FANOUT_ROLLOVER
+ *
+ * Todo:
+ * - functionality: PACKET_FANOUT_FLAG_DEFRAG
+ *
+ * License (GPLv2):
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. * See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc.,
+ * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+ */
+
+#define _GNU_SOURCE /* for sched_setaffinity */
+
+#include <arpa/inet.h>
+#include <errno.h>
+#include <fcntl.h>
+#include <linux/filter.h>
+#include <linux/if_packet.h>
+#include <net/ethernet.h>
+#include <netinet/ip.h>
+#include <netinet/udp.h>
+#include <poll.h>
+#include <sched.h>
+#include <stdint.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/mman.h>
+#include <sys/socket.h>
+#include <sys/stat.h>
+#include <sys/types.h>
+#include <unistd.h>
+
+#define DATA_LEN 100
+#define DATA_CHAR 'a'
+#define RING_NUM_FRAMES 20
+#define PORT_BASE 8000
+
+static void pair_udp_open(int fds[], uint16_t port)
+{
+ struct sockaddr_in saddr, daddr;
+
+ fds[0] = socket(PF_INET, SOCK_DGRAM, 0);
+ fds[1] = socket(PF_INET, SOCK_DGRAM, 0);
+ if (fds[0] == -1 || fds[1] == -1) {
+ fprintf(stderr, "ERROR: socket dgram\n");
+ exit(1);
+ }
+
+ memset(&saddr, 0, sizeof(saddr));
+ saddr.sin_family = AF_INET;
+ saddr.sin_port = htons(port);
+ saddr.sin_addr.s_addr = htonl(INADDR_LOOPBACK);
+
+ memset(&daddr, 0, sizeof(daddr));
+ daddr.sin_family = AF_INET;
+ daddr.sin_port = htons(port + 1);
+ daddr.sin_addr.s_addr = htonl(INADDR_LOOPBACK);
+
+ /* must bind both to get consistent hash result */
+ if (bind(fds[1], (void *) &daddr, sizeof(daddr))) {
+ perror("bind");
+ exit(1);
+ }
+ if (bind(fds[0], (void *) &saddr, sizeof(saddr))) {
+ perror("bind");
+ exit(1);
+ }
+ if (connect(fds[0], (void *) &daddr, sizeof(daddr))) {
+ perror("bind");
+ exit(1);
+ }
+}
+
+static void pair_udp_send(int fds[], int num)
+{
+ char buf[DATA_LEN], rbuf[DATA_LEN];
+
+ memset(buf, DATA_CHAR, sizeof(buf));
+ while (num--) {
+ /* Should really handle EINTR and EAGAIN */
+ if (write(fds[0], buf, sizeof(buf)) != sizeof(buf)) {
+ fprintf(stderr, "ERROR: send failed left=%d\n", num);
+ exit(1);
+ }
+ if (read(fds[1], rbuf, sizeof(rbuf)) != sizeof(rbuf)) {
+ fprintf(stderr, "ERROR: recv failed left=%d\n", num);
+ exit(1);
+ }
+ if (memcmp(buf, rbuf, sizeof(buf))) {
+ fprintf(stderr, "ERROR: data failed left=%d\n", num);
+ exit(1);
+ }
+ }
+}
+
+static void sock_fanout_setfilter(int fd)
+{
+ struct sock_filter bpf_filter[] = {
+ { 0x80, 0, 0, 0x00000000 }, /* LD pktlen */
+ { 0x35, 0, 5, DATA_LEN }, /* JGE DATA_LEN [f goto nomatch]*/
+ { 0x30, 0, 0, 0x00000050 }, /* LD ip[80] */
+ { 0x15, 0, 3, DATA_CHAR }, /* JEQ DATA_CHAR [f goto nomatch]*/
+ { 0x30, 0, 0, 0x00000051 }, /* LD ip[81] */
+ { 0x15, 0, 1, DATA_CHAR }, /* JEQ DATA_CHAR [f goto nomatch]*/
+ { 0x6, 0, 0, 0x00000060 }, /* RET match */
+/* nomatch */ { 0x6, 0, 0, 0x00000000 }, /* RET no match */
+ };
+ struct sock_fprog bpf_prog;
+
+ bpf_prog.filter = bpf_filter;
+ bpf_prog.len = sizeof(bpf_filter) / sizeof(struct sock_filter);
+ if (setsockopt(fd, SOL_SOCKET, SO_ATTACH_FILTER, &bpf_prog,
+ sizeof(bpf_prog))) {
+ perror("setsockopt SO_ATTACH_FILTER");
+ exit(1);
+ }
+}
+
+/* Open a socket in a given fanout mode.
+ * @return -1 if mode is bad, a valid socket otherwise */
+static int sock_fanout_open(uint16_t typeflags, int num_packets)
+{
+ int fd, val;
+
+ fd = socket(PF_PACKET, SOCK_DGRAM, htons(ETH_P_IP));
+ if (fd < 0) {
+ perror("socket packet");
+ exit(1);
+ }
+
+ /* fanout group ID is always 0: tests whether old groups are deleted */
+ val = ((int) typeflags) << 16;
+ if (setsockopt(fd, SOL_PACKET, PACKET_FANOUT, &val, sizeof(val))) {
+ if (close(fd)) {
+ perror("close packet");
+ exit(1);
+ }
+ return -1;
+ }
+
+ sock_fanout_setfilter(fd);
+ return fd;
+}
+
+static char *sock_fanout_open_ring(int fd)
+{
+ struct tpacket_req req = {
+ .tp_block_size = getpagesize(),
+ .tp_frame_size = getpagesize(),
+ .tp_block_nr = RING_NUM_FRAMES,
+ .tp_frame_nr = RING_NUM_FRAMES,
+ };
+ char *ring;
+
+ if (setsockopt(fd, SOL_PACKET, PACKET_RX_RING, (void *) &req,
+ sizeof(req))) {
+ perror("packetsock ring setsockopt");
+ exit(1);
+ }
+
+ ring = mmap(0, req.tp_block_size * req.tp_block_nr,
+ PROT_READ | PROT_WRITE, MAP_SHARED, fd, 0);
+ if (!ring) {
+ fprintf(stderr, "packetsock ring mmap\n");
+ exit(1);
+ }
+
+ return ring;
+}
+
+static int sock_fanout_read_ring(int fd, void *ring)
+{
+ struct tpacket_hdr *header = ring;
+ int count = 0;
+
+ while (header->tp_status & TP_STATUS_USER && count < RING_NUM_FRAMES) {
+ count++;
+ header = ring + (count * getpagesize());
+ }
+
+ return count;
+}
+
+static int sock_fanout_read(int fds[], char *rings[], const int expect[])
+{
+ int ret[2];
+
+ ret[0] = sock_fanout_read_ring(fds[0], rings[0]);
+ ret[1] = sock_fanout_read_ring(fds[1], rings[1]);
+
+ fprintf(stderr, "info: count=%d,%d, expect=%d,%d\n",
+ ret[0], ret[1], expect[0], expect[1]);
+
+ if ((!(ret[0] == expect[0] && ret[1] == expect[1])) &&
+ (!(ret[0] == expect[1] && ret[1] == expect[0]))) {
+ fprintf(stderr, "ERROR: incorrect queue lengths\n");
+ return 1;
+ }
+
+ return 0;
+}
+
+/* Test illegal mode + flag combination */
+static void test_control_single(void)
+{
+ fprintf(stderr, "test: control single socket\n");
+
+ if (sock_fanout_open(PACKET_FANOUT_ROLLOVER |
+ PACKET_FANOUT_FLAG_ROLLOVER, 0) != -1) {
+ fprintf(stderr, "ERROR: opened socket with dual rollover\n");
+ exit(1);
+ }
+}
+
+/* Test illegal group with different modes or flags */
+static void test_control_group(void)
+{
+ int fds[2];
+
+ fprintf(stderr, "test: control multiple sockets\n");
+
+ fds[0] = sock_fanout_open(PACKET_FANOUT_HASH, 20);
+ if (fds[0] == -1) {
+ fprintf(stderr, "ERROR: failed to open HASH socket\n");
+ exit(1);
+ }
+ if (sock_fanout_open(PACKET_FANOUT_HASH |
+ PACKET_FANOUT_FLAG_DEFRAG, 10) != -1) {
+ fprintf(stderr, "ERROR: joined group with wrong flag defrag\n");
+ exit(1);
+ }
+ if (sock_fanout_open(PACKET_FANOUT_HASH |
+ PACKET_FANOUT_FLAG_ROLLOVER, 10) != -1) {
+ fprintf(stderr, "ERROR: joined group with wrong flag ro\n");
+ exit(1);
+ }
+ if (sock_fanout_open(PACKET_FANOUT_CPU, 10) != -1) {
+ fprintf(stderr, "ERROR: joined group with wrong mode\n");
+ exit(1);
+ }
+ fds[1] = sock_fanout_open(PACKET_FANOUT_HASH, 20);
+ if (fds[1] == -1) {
+ fprintf(stderr, "ERROR: failed to join group\n");
+ exit(1);
+ }
+ if (close(fds[1]) || close(fds[0])) {
+ fprintf(stderr, "ERROR: closing sockets\n");
+ exit(1);
+ }
+}
+
+static int test_datapath(uint16_t typeflags, int port_off,
+ const int expect1[], const int expect2[])
+{
+ const int expect0[] = { 0, 0 };
+ char *rings[2];
+ int fds[2], fds_udp[2][2], ret;
+
+ fprintf(stderr, "test: datapath 0x%hx\n", typeflags);
+
+ fds[0] = sock_fanout_open(typeflags, 20);
+ fds[1] = sock_fanout_open(typeflags, 20);
+ if (fds[0] == -1 || fds[1] == -1) {
+ fprintf(stderr, "ERROR: failed open\n");
+ exit(1);
+ }
+ rings[0] = sock_fanout_open_ring(fds[0]);
+ rings[1] = sock_fanout_open_ring(fds[1]);
+ pair_udp_open(fds_udp[0], PORT_BASE);
+ pair_udp_open(fds_udp[1], PORT_BASE + port_off);
+ sock_fanout_read(fds, rings, expect0);
+
+ /* Send data, but not enough to overflow a queue */
+ pair_udp_send(fds_udp[0], 15);
+ pair_udp_send(fds_udp[1], 5);
+ ret = sock_fanout_read(fds, rings, expect1);
+
+ /* Send more data, overflow the queue */
+ pair_udp_send(fds_udp[0], 15);
+ /* TODO: ensure consistent order between expect1 and expect2 */
+ ret |= sock_fanout_read(fds, rings, expect2);
+
+ if (munmap(rings[1], RING_NUM_FRAMES * getpagesize()) ||
+ munmap(rings[0], RING_NUM_FRAMES * getpagesize())) {
+ fprintf(stderr, "close rings\n");
+ exit(1);
+ }
+ if (close(fds_udp[1][1]) || close(fds_udp[1][0]) ||
+ close(fds_udp[0][1]) || close(fds_udp[0][0]) ||
+ close(fds[1]) || close(fds[0])) {
+ fprintf(stderr, "close datapath\n");
+ exit(1);
+ }
+
+ return ret;
+}
+
+static int set_cpuaffinity(int cpuid)
+{
+ cpu_set_t mask;
+
+ CPU_ZERO(&mask);
+ CPU_SET(cpuid, &mask);
+ if (sched_setaffinity(0, sizeof(mask), &mask)) {
+ if (errno != EINVAL) {
+ fprintf(stderr, "setaffinity %d\n", cpuid);
+ exit(1);
+ }
+ return 1;
+ }
+
+ return 0;
+}
+
+int main(int argc, char **argv)
+{
+ const int expect_hash[2][2] = { { 15, 5 }, { 20, 5 } };
+ const int expect_hash_rb[2][2] = { { 15, 5 }, { 20, 15 } };
+ const int expect_lb[2][2] = { { 10, 10 }, { 18, 17 } };
+ const int expect_rb[2][2] = { { 20, 0 }, { 20, 15 } };
+ const int expect_cpu0[2][2] = { { 20, 0 }, { 20, 0 } };
+ const int expect_cpu1[2][2] = { { 0, 20 }, { 0, 20 } };
+ int port_off = 2, tries = 5, ret;
+
+ test_control_single();
+ test_control_group();
+
+ /* find a set of ports that do not collide onto the same socket */
+ ret = test_datapath(PACKET_FANOUT_HASH, port_off,
+ expect_hash[0], expect_hash[1]);
+ while (ret && tries--) {
+ fprintf(stderr, "info: trying alternate ports (%d)\n", tries);
+ ret = test_datapath(PACKET_FANOUT_HASH, ++port_off,
+ expect_hash[0], expect_hash[1]);
+ }
+
+ ret |= test_datapath(PACKET_FANOUT_HASH | PACKET_FANOUT_FLAG_ROLLOVER,
+ port_off, expect_hash_rb[0], expect_hash_rb[1]);
+ ret |= test_datapath(PACKET_FANOUT_LB,
+ port_off, expect_lb[0], expect_lb[1]);
+ ret |= test_datapath(PACKET_FANOUT_ROLLOVER,
+ port_off, expect_rb[0], expect_rb[1]);
+
+ set_cpuaffinity(0);
+ ret |= test_datapath(PACKET_FANOUT_CPU, port_off,
+ expect_cpu0[0], expect_cpu0[1]);
+ if (!set_cpuaffinity(1))
+ /* TODO: test that choice alternates with previous */
+ ret |= test_datapath(PACKET_FANOUT_CPU, port_off,
+ expect_cpu1[0], expect_cpu1[1]);
+
+ if (ret)
+ return 1;
+
+ printf("OK. All tests passed\n");
+ return 0;
+}
diff --git a/tools/testing/selftests/net-afpacket/run_afpackettests b/tools/testing/selftests/net-afpacket/run_afpackettests
new file mode 100644
index 0000000..7907824
--- /dev/null
+++ b/tools/testing/selftests/net-afpacket/run_afpackettests
@@ -0,0 +1,16 @@
+#!/bin/sh
+
+if [ $(id -u) != 0 ]; then
+ echo $msg must be run as root >&2
+ exit 0
+fi
+
+echo "--------------------"
+echo "running psock_fanout test"
+echo "--------------------"
+./psock_fanout
+if [ $? -ne 0 ]; then
+ echo "[FAIL]"
+else
+ echo "[PASS]"
+fi
diff --git a/tools/testing/selftests/net-socket/Makefile b/tools/testing/selftests/net-socket/Makefile
new file mode 100644
index 0000000..2450fd8
--- /dev/null
+++ b/tools/testing/selftests/net-socket/Makefile
@@ -0,0 +1,16 @@
+# Makefile for net-socket selftests
+
+CC = $(CROSS_COMPILE)gcc
+CFLAGS = -Wall
+
+NET_SOCK_PROGS = socket
+
+all: $(NET_SOCK_PROGS)
+%: %.c
+ $(CC) $(CFLAGS) -o $@ $^
+
+run_tests: all
+ @/bin/sh ./run_netsocktests || echo "sockettests: [FAIL]"
+
+clean:
+ $(RM) $(NET_SOCK_PROGS)
diff --git a/tools/testing/selftests/net-socket/run_netsocktests b/tools/testing/selftests/net-socket/run_netsocktests
new file mode 100644
index 0000000..c09a682
--- /dev/null
+++ b/tools/testing/selftests/net-socket/run_netsocktests
@@ -0,0 +1,12 @@
+#!/bin/bash
+
+echo "--------------------"
+echo "running socket test"
+echo "--------------------"
+./socket
+if [ $? -ne 0 ]; then
+ echo "[FAIL]"
+else
+ echo "[PASS]"
+fi
+
diff --git a/tools/testing/selftests/net-socket/socket.c b/tools/testing/selftests/net-socket/socket.c
new file mode 100644
index 0000000..0f227f2
--- /dev/null
+++ b/tools/testing/selftests/net-socket/socket.c
@@ -0,0 +1,92 @@
+#include <stdio.h>
+#include <errno.h>
+#include <unistd.h>
+#include <string.h>
+#include <sys/types.h>
+#include <sys/socket.h>
+#include <netinet/in.h>
+
+struct socket_testcase {
+ int domain;
+ int type;
+ int protocol;
+
+ /* 0 = valid file descriptor
+ * -foo = error foo
+ */
+ int expect;
+
+ /* If non-zero, accept EAFNOSUPPORT to handle the case
+ * of the protocol not being configured into the kernel.
+ */
+ int nosupport_ok;
+};
+
+static struct socket_testcase tests[] = {
+ { AF_MAX, 0, 0, -EAFNOSUPPORT, 0 },
+ { AF_INET, SOCK_STREAM, IPPROTO_TCP, 0, 1 },
+ { AF_INET, SOCK_DGRAM, IPPROTO_TCP, -EPROTONOSUPPORT, 1 },
+ { AF_INET, SOCK_DGRAM, IPPROTO_UDP, 0, 1 },
+ { AF_INET, SOCK_STREAM, IPPROTO_UDP, -EPROTONOSUPPORT, 1 },
+};
+
+#define ARRAY_SIZE(arr) (sizeof(arr) / sizeof((arr)[0]))
+#define ERR_STRING_SZ 64
+
+static int run_tests(void)
+{
+ char err_string1[ERR_STRING_SZ];
+ char err_string2[ERR_STRING_SZ];
+ int i, err;
+
+ err = 0;
+ for (i = 0; i < ARRAY_SIZE(tests); i++) {
+ struct socket_testcase *s = &tests[i];
+ int fd;
+
+ fd = socket(s->domain, s->type, s->protocol);
+ if (fd < 0) {
+ if (s->nosupport_ok &&
+ errno == EAFNOSUPPORT)
+ continue;
+
+ if (s->expect < 0 &&
+ errno == -s->expect)
+ continue;
+
+ strerror_r(-s->expect, err_string1, ERR_STRING_SZ);
+ strerror_r(errno, err_string2, ERR_STRING_SZ);
+
+ fprintf(stderr, "socket(%d, %d, %d) expected "
+ "err (%s) got (%s)\n",
+ s->domain, s->type, s->protocol,
+ err_string1, err_string2);
+
+ err = -1;
+ break;
+ } else {
+ close(fd);
+
+ if (s->expect < 0) {
+ strerror_r(errno, err_string1, ERR_STRING_SZ);
+
+ fprintf(stderr, "socket(%d, %d, %d) expected "
+ "success got err (%s)\n",
+ s->domain, s->type, s->protocol,
+ err_string1);
+
+ err = -1;
+ break;
+ }
+ }
+ }
+
+ return err;
+}
+
+int main(void)
+{
+ int err = run_tests();
+
+ return err;
+}