Auto merge with /home/aegl/GIT/linus
diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index 2e08942..cbb3e0c 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -220,13 +220,6 @@
depends on !IA64_HP_SIM
default y
-config IA64_SGI_SN_SIM
- bool "SGI Medusa Simulator Support"
- depends on IA64_SGI_SN2 || IA64_GENERIC
- help
- If you are compiling a kernel that will run under SGI's IA-64
- simulator (Medusa) then say Y, otherwise say N.
-
config IA64_SGI_SN_XP
tristate "Support communication between SGI SSIs"
select IA64_UNCACHED_ALLOCATOR
diff --git a/arch/ia64/configs/sn2_defconfig b/arch/ia64/configs/sn2_defconfig
index c056139..04d0b00 100644
--- a/arch/ia64/configs/sn2_defconfig
+++ b/arch/ia64/configs/sn2_defconfig
@@ -81,7 +81,6 @@
CONFIG_ARCH_DISCONTIGMEM_ENABLE=y
# CONFIG_IA64_CYCLONE is not set
CONFIG_IOSAPIC=y
-CONFIG_IA64_SGI_SN_SIM=y
CONFIG_FORCE_MAX_ZONEORDER=18
CONFIG_SMP=y
CONFIG_NR_CPUS=512
diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c
index 5c7c957..84f89da 100644
--- a/arch/ia64/kernel/setup.c
+++ b/arch/ia64/kernel/setup.c
@@ -20,6 +20,7 @@
* 02/01/00 R.Seth fixed get_cpuinfo for SMP
* 01/07/99 S.Eranian added the support for command line argument
* 06/24/99 W.Drummond added boot_cpu_data.
+ * 05/28/05 Z. Menyhart Dynamic stride size for "flush_icache_range()"
*/
#include <linux/config.h>
#include <linux/module.h>
@@ -85,6 +86,13 @@
unsigned int num_io_spaces;
/*
+ * "flush_icache_range()" needs to know what processor dependent stride size to use
+ * when it makes i-cache(s) coherent with d-caches.
+ */
+#define I_CACHE_STRIDE_SHIFT 5 /* Safest way to go: 32 bytes by 32 bytes */
+unsigned long ia64_i_cache_stride_shift = ~0;
+
+/*
* The merge_mask variable needs to be set to (max(iommu_page_size(iommu)) - 1). This
* mask specifies a mask of address bits that must be 0 in order for two buffers to be
* mergeable by the I/O MMU (i.e., the end address of the first buffer and the start
@@ -628,6 +636,12 @@
/* start_kernel() requires this... */
}
+/*
+ * Calculate the max. cache line size.
+ *
+ * In addition, the minimum of the i-cache stride sizes is calculated for
+ * "flush_icache_range()".
+ */
static void
get_max_cacheline_size (void)
{
@@ -641,6 +655,8 @@
printk(KERN_ERR "%s: ia64_pal_cache_summary() failed (status=%ld)\n",
__FUNCTION__, status);
max = SMP_CACHE_BYTES;
+ /* Safest setup for "flush_icache_range()" */
+ ia64_i_cache_stride_shift = I_CACHE_STRIDE_SHIFT;
goto out;
}
@@ -649,14 +665,31 @@
&cci);
if (status != 0) {
printk(KERN_ERR
- "%s: ia64_pal_cache_config_info(l=%lu) failed (status=%ld)\n",
+ "%s: ia64_pal_cache_config_info(l=%lu, 2) failed (status=%ld)\n",
__FUNCTION__, l, status);
max = SMP_CACHE_BYTES;
+ /* The safest setup for "flush_icache_range()" */
+ cci.pcci_stride = I_CACHE_STRIDE_SHIFT;
+ cci.pcci_unified = 1;
}
line_size = 1 << cci.pcci_line_size;
if (line_size > max)
max = line_size;
- }
+ if (!cci.pcci_unified) {
+ status = ia64_pal_cache_config_info(l,
+ /* cache_type (instruction)= */ 1,
+ &cci);
+ if (status != 0) {
+ printk(KERN_ERR
+ "%s: ia64_pal_cache_config_info(l=%lu, 1) failed (status=%ld)\n",
+ __FUNCTION__, l, status);
+ /* The safest setup for "flush_icache_range()" */
+ cci.pcci_stride = I_CACHE_STRIDE_SHIFT;
+ }
+ }
+ if (cci.pcci_stride < ia64_i_cache_stride_shift)
+ ia64_i_cache_stride_shift = cci.pcci_stride;
+ }
out:
if (max > ia64_max_cacheline_size)
ia64_max_cacheline_size = max;
diff --git a/arch/ia64/lib/flush.S b/arch/ia64/lib/flush.S
index a1af914..3e2cfa2 100644
--- a/arch/ia64/lib/flush.S
+++ b/arch/ia64/lib/flush.S
@@ -3,37 +3,59 @@
*
* Copyright (C) 1999-2001, 2005 Hewlett-Packard Co
* David Mosberger-Tang <davidm@hpl.hp.com>
+ *
+ * 05/28/05 Zoltan Menyhart Dynamic stride size
*/
+
#include <asm/asmmacro.h>
-#include <asm/page.h>
+
/*
* flush_icache_range(start,end)
- * Must flush range from start to end-1 but nothing else (need to
+ *
+ * Make i-cache(s) coherent with d-caches.
+ *
+ * Must deal with range from start to end-1 but nothing else (need to
* be careful not to touch addresses that may be unmapped).
+ *
+ * Note: "in0" and "in1" are preserved for debugging purposes.
*/
GLOBAL_ENTRY(flush_icache_range)
+
.prologue
- alloc r2=ar.pfs,2,0,0,0
- sub r8=in1,in0,1
+ alloc r2=ar.pfs,2,0,0,0
+ movl r3=ia64_i_cache_stride_shift
+ mov r21=1
;;
- shr.u r8=r8,5 // we flush 32 bytes per iteration
- .save ar.lc, r3
- mov r3=ar.lc // save ar.lc
+ ld8 r20=[r3] // r20: stride shift
+ sub r22=in1,r0,1 // last byte address
+ ;;
+ shr.u r23=in0,r20 // start / (stride size)
+ shr.u r22=r22,r20 // (last byte address) / (stride size)
+ shl r21=r21,r20 // r21: stride size of the i-cache(s)
+ ;;
+ sub r8=r22,r23 // number of strides - 1
+ shl r24=r23,r20 // r24: addresses for "fc.i" =
+ // "start" rounded down to stride boundary
+ .save ar.lc,r3
+ mov r3=ar.lc // save ar.lc
;;
.body
-
- mov ar.lc=r8
+ mov ar.lc=r8
;;
-.Loop: fc.i in0 // issuable on M2 only
- add in0=32,in0
+ /*
+ * 32 byte aligned loop, even number of (actually 2) bundles
+ */
+.Loop: fc.i r24 // issuable on M0 only
+ add r24=r21,r24 // we flush "stride size" bytes per iteration
+ nop.i 0
br.cloop.sptk.few .Loop
;;
sync.i
;;
srlz.i
;;
- mov ar.lc=r3 // restore ar.lc
+ mov ar.lc=r3 // restore ar.lc
br.ret.sptk.many rp
END(flush_icache_range)
diff --git a/arch/ia64/pci/pci.c b/arch/ia64/pci/pci.c
index 720a861..54d9ed4 100644
--- a/arch/ia64/pci/pci.c
+++ b/arch/ia64/pci/pci.c
@@ -157,6 +157,7 @@
memset(controller, 0, sizeof(*controller));
controller->segment = seg;
+ controller->node = -1;
return controller;
}
@@ -288,6 +289,7 @@
unsigned int windows = 0;
struct pci_bus *pbus;
char *name;
+ int pxm;
controller = alloc_pci_controller(domain);
if (!controller)
@@ -295,10 +297,16 @@
controller->acpi_handle = device->handle;
+ pxm = acpi_get_pxm(controller->acpi_handle);
+#ifdef CONFIG_NUMA
+ if (pxm >= 0)
+ controller->node = pxm_to_nid_map[pxm];
+#endif
+
acpi_walk_resources(device->handle, METHOD_NAME__CRS, count_window,
&windows);
- controller->window = kmalloc(sizeof(*controller->window) * windows,
- GFP_KERNEL);
+ controller->window = kmalloc_node(sizeof(*controller->window) * windows,
+ GFP_KERNEL, controller->node);
if (!controller->window)
goto out2;
diff --git a/arch/ia64/sn/kernel/io_init.c b/arch/ia64/sn/kernel/io_init.c
index a67f39e..a6649ba 100644
--- a/arch/ia64/sn/kernel/io_init.c
+++ b/arch/ia64/sn/kernel/io_init.c
@@ -61,7 +61,7 @@
}
static void *
-sn_default_pci_bus_fixup(struct pcibus_bussoft *soft)
+sn_default_pci_bus_fixup(struct pcibus_bussoft *soft, struct pci_controller *controller)
{
return NULL;
}
@@ -362,7 +362,7 @@
provider_soft = NULL;
if (provider->bus_fixup)
- provider_soft = (*provider->bus_fixup) (prom_bussoft_ptr);
+ provider_soft = (*provider->bus_fixup) (prom_bussoft_ptr, controller);
if (provider_soft == NULL)
return; /* fixup failed or not applicable */
@@ -380,6 +380,22 @@
SN_PCIBUS_BUSSOFT(bus)->bs_xwidget_info =
&(hubdev_info->hdi_xwidget_info[SN_PCIBUS_BUSSOFT(bus)->bs_xid]);
+ /*
+ * If the node information we obtained during the fixup phase is invalid
+ * then set controller->node to -1 (undetermined)
+ */
+ if (controller->node >= num_online_nodes()) {
+ struct pcibus_bussoft *b = SN_PCIBUS_BUSSOFT(bus);
+
+ printk(KERN_WARNING "Device ASIC=%u XID=%u PBUSNUM=%lu"
+ "L_IO=%lx L_MEM=%lx BASE=%lx\n",
+ b->bs_asic_type, b->bs_xid, b->bs_persist_busnum,
+ b->bs_legacy_io, b->bs_legacy_mem, b->bs_base);
+ printk(KERN_WARNING "on node %d but only %d nodes online."
+ "Association set to undetermined.\n",
+ controller->node, num_online_nodes());
+ controller->node = -1;
+ }
return;
error_return:
diff --git a/arch/ia64/sn/kernel/xpc_channel.c b/arch/ia64/sn/kernel/xpc_channel.c
index 6d02dac..94698be 100644
--- a/arch/ia64/sn/kernel/xpc_channel.c
+++ b/arch/ia64/sn/kernel/xpc_channel.c
@@ -72,7 +72,7 @@
enum xpc_retval
xpc_setup_infrastructure(struct xpc_partition *part)
{
- int ret;
+ int ret, cpuid;
struct timer_list *timer;
partid_t partid = XPC_PARTID(part);
@@ -223,9 +223,9 @@
xpc_vars_part[partid].openclose_args_pa =
__pa(part->local_openclose_args);
xpc_vars_part[partid].IPI_amo_pa = __pa(part->local_IPI_amo_va);
- xpc_vars_part[partid].IPI_nasid = cpuid_to_nasid(smp_processor_id());
- xpc_vars_part[partid].IPI_phys_cpuid =
- cpu_physical_id(smp_processor_id());
+ cpuid = raw_smp_processor_id(); /* any CPU in this partition will do */
+ xpc_vars_part[partid].IPI_nasid = cpuid_to_nasid(cpuid);
+ xpc_vars_part[partid].IPI_phys_cpuid = cpu_physical_id(cpuid);
xpc_vars_part[partid].nchannels = part->nchannels;
xpc_vars_part[partid].magic = XPC_VP_MAGIC1;
diff --git a/arch/ia64/sn/pci/pci_dma.c b/arch/ia64/sn/pci/pci_dma.c
index a2f7a88..0e4b9ad 100644
--- a/arch/ia64/sn/pci/pci_dma.c
+++ b/arch/ia64/sn/pci/pci_dma.c
@@ -79,6 +79,7 @@
{
void *cpuaddr;
unsigned long phys_addr;
+ int node;
struct pci_dev *pdev = to_pci_dev(dev);
struct sn_pcibus_provider *provider = SN_PCIDEV_BUSPROVIDER(pdev);
@@ -86,10 +87,19 @@
/*
* Allocate the memory.
- * FIXME: We should be doing alloc_pages_node for the node closest
- * to the PCI device.
*/
- if (!(cpuaddr = (void *)__get_free_pages(GFP_ATOMIC, get_order(size))))
+ node = pcibus_to_node(pdev->bus);
+ if (likely(node >=0)) {
+ struct page *p = alloc_pages_node(node, GFP_ATOMIC, get_order(size));
+
+ if (likely(p))
+ cpuaddr = page_address(p);
+ else
+ return NULL;
+ } else
+ cpuaddr = (void *)__get_free_pages(GFP_ATOMIC, get_order(size));
+
+ if (unlikely(!cpuaddr))
return NULL;
memset(cpuaddr, 0x0, size);
diff --git a/arch/ia64/sn/pci/pcibr/pcibr_provider.c b/arch/ia64/sn/pci/pcibr/pcibr_provider.c
index 9813da5..b95e9286 100644
--- a/arch/ia64/sn/pci/pcibr/pcibr_provider.c
+++ b/arch/ia64/sn/pci/pcibr/pcibr_provider.c
@@ -85,7 +85,7 @@
}
void *
-pcibr_bus_fixup(struct pcibus_bussoft *prom_bussoft)
+pcibr_bus_fixup(struct pcibus_bussoft *prom_bussoft, struct pci_controller *controller)
{
int nasid, cnode, j;
struct hubdev_info *hubdev_info;
@@ -158,6 +158,14 @@
memset(soft->pbi_int_ate_resource.ate, 0,
(soft->pbi_int_ate_size * sizeof(uint64_t)));
+ if (prom_bussoft->bs_asic_type == PCIIO_ASIC_TYPE_TIOCP)
+ /*
+ * TIO PCI Bridge with no closest node information.
+ * FIXME: Find another way to determine the closest node
+ */
+ controller->node = -1;
+ else
+ controller->node = cnode;
return soft;
}
diff --git a/arch/ia64/sn/pci/tioca_provider.c b/arch/ia64/sn/pci/tioca_provider.c
index 51cc4e6..5d76a75 100644
--- a/arch/ia64/sn/pci/tioca_provider.c
+++ b/arch/ia64/sn/pci/tioca_provider.c
@@ -581,7 +581,7 @@
* the caller.
*/
static void *
-tioca_bus_fixup(struct pcibus_bussoft *prom_bussoft)
+tioca_bus_fixup(struct pcibus_bussoft *prom_bussoft, struct pci_controller *controller)
{
struct tioca_common *tioca_common;
struct tioca_kernel *tioca_kern;
@@ -646,6 +646,8 @@
__FUNCTION__, SGI_TIOCA_ERROR,
(int)tioca_common->ca_common.bs_persist_busnum);
+ /* Setup locality information */
+ controller->node = tioca_kern->ca_closest_node;
return tioca_common;
}
diff --git a/drivers/firmware/pcdp.h b/drivers/firmware/pcdp.h
index e72cc47..ce910d6 100644
--- a/drivers/firmware/pcdp.h
+++ b/drivers/firmware/pcdp.h
@@ -52,6 +52,8 @@
u32 clock_rate;
u8 pci_prog_intfc;
u8 flags;
+ u16 conout_index;
+ u32 reserved;
} __attribute__((packed));
#define PCDP_IF_PCI 1
diff --git a/include/asm-ia64/pci.h b/include/asm-ia64/pci.h
index f11771e..dba9f22 100644
--- a/include/asm-ia64/pci.h
+++ b/include/asm-ia64/pci.h
@@ -128,6 +128,7 @@
void *acpi_handle;
void *iommu;
int segment;
+ int node; /* nearest node with memory or -1 for global allocation */
unsigned int windows;
struct pci_window *window;
diff --git a/include/asm-ia64/sn/pcibr_provider.h b/include/asm-ia64/sn/pcibr_provider.h
index f9b8d21..2b42d9e 100644
--- a/include/asm-ia64/sn/pcibr_provider.h
+++ b/include/asm-ia64/sn/pcibr_provider.h
@@ -128,7 +128,7 @@
#define pcibr_unlock(pcibus_info, flag) spin_unlock_irqrestore(&pcibus_info->pbi_lock, flag)
extern int pcibr_init_provider(void);
-extern void *pcibr_bus_fixup(struct pcibus_bussoft *);
+extern void *pcibr_bus_fixup(struct pcibus_bussoft *, struct pci_controller *);
extern dma_addr_t pcibr_dma_map(struct pci_dev *, unsigned long, size_t);
extern dma_addr_t pcibr_dma_map_consistent(struct pci_dev *, unsigned long, size_t);
extern void pcibr_dma_unmap(struct pci_dev *, dma_addr_t, int);
diff --git a/include/asm-ia64/sn/pcibus_provider_defs.h b/include/asm-ia64/sn/pcibus_provider_defs.h
index 04e27d5..976f5ef 100644
--- a/include/asm-ia64/sn/pcibus_provider_defs.h
+++ b/include/asm-ia64/sn/pcibus_provider_defs.h
@@ -37,6 +37,7 @@
struct xwidget_info *bs_xwidget_info;
};
+struct pci_controller;
/*
* SN pci bus indirection
*/
@@ -45,7 +46,7 @@
dma_addr_t (*dma_map)(struct pci_dev *, unsigned long, size_t);
dma_addr_t (*dma_map_consistent)(struct pci_dev *, unsigned long, size_t);
void (*dma_unmap)(struct pci_dev *, dma_addr_t, int);
- void * (*bus_fixup)(struct pcibus_bussoft *);
+ void * (*bus_fixup)(struct pcibus_bussoft *, struct pci_controller *);
};
extern struct sn_pcibus_provider *sn_pci_provider[];
diff --git a/include/asm-ia64/sn/simulator.h b/include/asm-ia64/sn/simulator.h
index cf770e2..16a48b5 100644
--- a/include/asm-ia64/sn/simulator.h
+++ b/include/asm-ia64/sn/simulator.h
@@ -13,16 +13,9 @@
#define SNMAGIC 0xaeeeeeee8badbeefL
#define IS_MEDUSA() ({long sn; asm("mov %0=cpuid[%1]" : "=r"(sn) : "r"(2)); sn == SNMAGIC;})
-#ifdef CONFIG_IA64_SGI_SN_SIM
#define SIMULATOR_SLEEP() asm("nop.i 0x8beef")
-#define IS_RUNNING_ON_SIMULATOR() (sn_prom_type)
+#define IS_RUNNING_ON_SIMULATOR() (sn_prom_type)
#define IS_RUNNING_ON_FAKE_PROM() (sn_prom_type == 2)
extern int sn_prom_type; /* 0=hardware, 1=medusa/realprom, 2=medusa/fakeprom */
-#else
-#define IS_RUNNING_ON_SIMULATOR() (0)
-#define IS_RUNNING_ON_FAKE_PROM() (0)
-#define SIMULATOR_SLEEP()
-
-#endif
#endif /* _ASM_IA64_SN_SIMULATOR_H */
diff --git a/include/asm-ia64/topology.h b/include/asm-ia64/topology.h
index 4e64c2a..399bc29 100644
--- a/include/asm-ia64/topology.h
+++ b/include/asm-ia64/topology.h
@@ -40,6 +40,11 @@
*/
#define node_to_first_cpu(node) (__ffs(node_to_cpumask(node)))
+/*
+ * Determines the node for a given pci bus
+ */
+#define pcibus_to_node(bus) PCI_CONTROLLER(bus)->node
+
void build_cpu_to_node_map(void);
#define SD_CPU_INIT (struct sched_domain) { \