From mboxrd@z Thu Jan 1 00:00:00 1970 From: yizhan@redhat.com (Yi Zhang) Date: Thu, 15 Jun 2017 16:45:37 +0800 Subject: mlx4_core 0000:07:00.0: swiotlb buffer is full and OOM observed during stress test on reset_controller In-Reply-To: <6bf26cbc-71e4-a030-628b-a2ee1d1de94b@grimberg.me> References: <2013049462.31187009.1488542111040.JavaMail.zimbra@redhat.com> <56e8ccd3-8116-89a1-2f65-eb61a91c5f84@mellanox.com> <860db62d-ae93-d94c-e5fb-88e7b643f737@redhat.com> <0a825b18-df06-9a6d-38c9-402f4ee121f7@mellanox.com> <7496c68a-15f3-d8cb-b17f-20f5a59a24d2@redhat.com> <31678a43-f76c-a921-e40c-470b0de1a86c@grimberg.me> <20170319070115.GP2079@mtr-leonro.local> <136275928.8307994.1495126919829.JavaMail.zimbra@redhat.com> <358169046.8629042.1495210672801.JavaMail.zimbra@redhat.com> <6bf26cbc-71e4-a030-628b-a2ee1d1de94b@grimberg.me> Message-ID: <6f5249b4-e8fe-5def-bbc9-0079b7457df9@redhat.com> On 06/04/2017 11:49 PM, Sagi Grimberg wrote: > Hi Yi, > >> Finally found below patch [1] that fixed this issue. >> With [1], I can see the speed of reset_controller operation[2] is >> obviously slow than before. >> >> >> [1] >> commit b7363e67b23e04c23c2a99437feefac7292a88bc >> Author: Sagi Grimberg >> Date: Wed Mar 8 22:03:17 2017 +0200 >> >> IB/device: Convert ib-comp-wq to be CPU-bound > > This is very unlikely. > > I think that what made this go away is: > > commit 777dc82395de6e04b3a5fedcf153eb99bf5f1241 > Author: Sagi Grimberg > Date: Tue Mar 21 16:29:49 2017 +0200 > > nvmet-rdma: occasionally flush ongoing controller teardown > > If we are attacked with establishments/teradowns we need to > make sure we do not consume too much system memory. Thus > let ongoing controller teardowns complete before accepting > new controller establishments. > Hi Sagi This patch fixed the issue, thanks again. Yi > > Cheers, > Sagi. > > _______________________________________________ > Linux-nvme mailing list > Linux-nvme at lists.infradead.org > http://lists.infradead.org/mailman/listinfo/linux-nvme