Home
last modified time | relevance | path

Searched hist:ec56dc0b7f6c3fec20bbc2e98ff1a06edf2fc9b9 (Results 1 – 2 of 2) sorted by relevance

/linux/drivers/infiniband/ulp/ipoib/
H A Dipoib.hdiff ec56dc0b7f6c3fec20bbc2e98ff1a06edf2fc9b9 Mon May 28 13:37:27 CEST 2007 Michael S. Tsirkin <mst@dev.mellanox.co.il> IPoIB/cm: Fix performance regression on Mellanox

commit 518b1646 ("IPoIB/cm: Fix SRQ WR leak") introduced a severe
performance regression on Mellanox cards, because keeping a QP in the
error state for extended periods of time moves hardware to the slow
path (until the QP is destroyed). For example, MPI latency goes from
~3 usecs to ~7 usecs.

Fix this by posting a send WR on one of the QPs that are being
flushed, instead of using a separate drain QP that is kept in the
error state.

This fixes bug <https://bugs.openfabrics.org/show_bug.cgi?id=636>,
reported and bisected by Scott Weitzenkamp at Cisco and debugged by
Sasha Mikheev at Voltaire.

Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
H A Dipoib_cm.cdiff ec56dc0b7f6c3fec20bbc2e98ff1a06edf2fc9b9 Mon May 28 13:37:27 CEST 2007 Michael S. Tsirkin <mst@dev.mellanox.co.il> IPoIB/cm: Fix performance regression on Mellanox

commit 518b1646 ("IPoIB/cm: Fix SRQ WR leak") introduced a severe
performance regression on Mellanox cards, because keeping a QP in the
error state for extended periods of time moves hardware to the slow
path (until the QP is destroyed). For example, MPI latency goes from
~3 usecs to ~7 usecs.

Fix this by posting a send WR on one of the QPs that are being
flushed, instead of using a separate drain QP that is kept in the
error state.

This fixes bug <https://bugs.openfabrics.org/show_bug.cgi?id=636>,
reported and bisected by Scott Weitzenkamp at Cisco and debugged by
Sasha Mikheev at Voltaire.

Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>