hyperv: Enable sendbuf mechanism on the send path
We send packets using a copy-free mechanism (this is the Guest to Host transport
via VMBUS). While this is obviously optimal for large packets,
it may not be optimal for small packets. Hyper-V host supports
a second mechanism for sending packets that is "copy based". We implement that
mechanism in this patch.
In this version of the patch I have addressed a comment from David Miller.
With this patch (and all of the other offload and VRSS patches), we are now able
to almost saturate a 10G interface between Linux VMs on Hyper-V
on different hosts - close to 9 Gbps as measured via iperf.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/drivers/net/hyperv/hyperv_net.h b/drivers/net/hyperv/hyperv_net.h
index d1f7826..4b7df5a 100644
--- a/drivers/net/hyperv/hyperv_net.h
+++ b/drivers/net/hyperv/hyperv_net.h
@@ -140,6 +140,8 @@
void *send_completion_ctx;
void (*send_completion)(void *context);
+ u32 send_buf_index;
+
/* This points to the memory after page_buf */
struct rndis_message *rndis_msg;
@@ -582,6 +584,9 @@
#define NETVSC_RECEIVE_BUFFER_SIZE (1024*1024*16) /* 16MB */
#define NETVSC_RECEIVE_BUFFER_SIZE_LEGACY (1024*1024*15) /* 15MB */
+#define NETVSC_SEND_BUFFER_SIZE (1024 * 1024) /* 1MB */
+#define NETVSC_INVALID_INDEX -1
+
#define NETVSC_RECEIVE_BUFFER_ID 0xcafe
@@ -607,6 +612,15 @@
u32 recv_section_cnt;
struct nvsp_1_receive_buffer_section *recv_section;
+ /* Send buffer allocated by us */
+ void *send_buf;
+ u32 send_buf_size;
+ u32 send_buf_gpadl_handle;
+ u32 send_section_cnt;
+ u32 send_section_size;
+ unsigned long *send_section_map;
+ int map_words;
+
/* Used for NetVSP initialization protocol */
struct completion channel_init_wait;
struct nvsp_message channel_init_pkt;