From mboxrd@z Thu Jan 1 00:00:00 1970 From: Or Gerlitz Subject: Re: [PATCH] IB/mlx5: Reduce mlx5_ib_wq cacheline bouncing Date: Tue, 12 Jan 2016 17:05:24 +0200 Message-ID: <56951634.5030307@mellanox.com> References: <1452594732-9573-1-git-send-email-sagig@mellanox.com> <5694EEAA.3050600@mellanox.com> <56951130.60802@dev.mellanox.co.il> <56951375.1050704@mellanox.com> <569514A5.1050705@dev.mellanox.co.il> Mime-Version: 1.0 Content-Type: text/plain; charset="windows-1252"; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <569514A5.1050705-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org> Sender: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Sagi Grimberg , Sagi Grimberg Cc: linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Matan Barak , Leon Romanovsky List-Id: linux-rdma@vger.kernel.org On 1/12/2016 4:58 PM, Sagi Grimberg wrote: > >>> I didn't took the time to measure cache hit/miss. I just noticed it >>> a while ago and it's been bugging me for some time so I figured I'd >>> send it out... >> >> The thing is that for data-path changes on high performance network >> drivers, we @ least need to know that the perf is as good as it was >> before the change. So you could run your iser perf IOPS test >> before/after the change and post 1-2 lines with results as part of the >> change-log. > > I tested iser perf and it seems to sustain. I didn't see any major > difference but I really don't think block storage iops/latency is not > the correct way to evaluate this change. maybe one nice option would be to take a look on how things are organized in libmlx5 around that corner -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html