aboutsummaryrefslogtreecommitdiff
path: root/drivers/infiniband/ulp
diff options
context:
space:
mode:
authorEli Cohen <eli@dev.mellanox.co.il>2008-07-14 23:48:52 -0700
committerRoland Dreier <rolandd@cisco.com>2008-07-14 23:48:52 -0700
commitbc3a290b51aaefc6a6af2d6e6d52ed32387c416c (patch)
tree94b3358177c589796224c818b80ca4fdc0944719 /drivers/infiniband/ulp
parente112373fd6aa280bd2cbc0d5cc3809115325a1be (diff)
IPoIB: Double default RX/TX ring sizes
Increase IPoIB ring sizes to twice their original sizes (RX: 128->256, TX: 64->128) to act as a shock absorber for high traffic peaks. With the current settings, we have seen cases that there are many calls to netif_stop_queue(), which causes degradation in throughput. Also, larger receive buffer sizes help IPoIB in CM mode to avoid experiencing RNR NAK conditions due to insufficient receive buffers at the SRQ. Signed-off-by: Eli Cohen <eli@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
Diffstat (limited to 'drivers/infiniband/ulp')
-rw-r--r--drivers/infiniband/ulp/ipoib/ipoib.h4
1 files changed, 2 insertions, 2 deletions
diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h
index 0281c8fecc9..b0ffc9abe8c 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib.h
+++ b/drivers/infiniband/ulp/ipoib/ipoib.h
@@ -70,8 +70,8 @@ enum {
IPOIB_CM_BUF_SIZE = IPOIB_CM_MTU + IPOIB_ENCAP_LEN,
IPOIB_CM_HEAD_SIZE = IPOIB_CM_BUF_SIZE % PAGE_SIZE,
IPOIB_CM_RX_SG = ALIGN(IPOIB_CM_BUF_SIZE, PAGE_SIZE) / PAGE_SIZE,
- IPOIB_RX_RING_SIZE = 128,
- IPOIB_TX_RING_SIZE = 64,
+ IPOIB_RX_RING_SIZE = 256,
+ IPOIB_TX_RING_SIZE = 128,
IPOIB_MAX_QUEUE_SIZE = 8192,
IPOIB_MIN_QUEUE_SIZE = 2,
IPOIB_CM_MAX_CONN_QP = 4096,