From mboxrd@z Thu Jan 1 00:00:00 1970 From: Laurence Oberman Subject: Re: [PATCH 11/11] IB/srp: Prevent mapping failures Date: Thu, 5 May 2016 16:55:56 -0400 (EDT) Message-ID: <804753248.34153510.1462481756460.JavaMail.zimbra@redhat.com> References: <571A936F.7040409@sandisk.com> <571A94AF.7000609@sandisk.com> <20160503093307.GL19931@lst.de> <572A8975.9060606@sandisk.com> <20160505140616.GA15000@lst.de> <2102017158.34060891.1462478304470.JavaMail.zimbra@redhat.com> <572BA98F.4090109@sandisk.com> <572BAA6F.1060607@sandisk.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <572BAA6F.1060607-XdAiOPVOjttBDgjK7y7TUQ@public.gmane.org> Sender: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Bart Van Assche Cc: Christoph Hellwig , Doug Ledford , Sagi Grimberg , linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: linux-rdma@vger.kernel.org ----- Original Message ----- > From: "Bart Van Assche" > To: "Laurence Oberman" , "Christoph Hellwig" > Cc: "Doug Ledford" , "Sagi Grimberg" , linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org > Sent: Thursday, May 5, 2016 4:17:51 PM > Subject: Re: [PATCH 11/11] IB/srp: Prevent mapping failures > > On 05/05/2016 01:14 PM, Bart Van Assche wrote: > > On 05/05/2016 12:58 PM, Laurence Oberman wrote: > >> I added and tested this patch. > >> I left the warn message in as I need it. > >> > >> [root@srptest ~]# dmesg | grep Reducing > >> [ 551.526015] scsi host4: ib_srp: Reducing max_sectors from 8192 to 4080 > >> [ 559.367609] scsi host5: ib_srp: Reducing max_sectors from 8192 to 4080 > >> [ 570.449121] scsi host5: ib_srp: Reducing max_sectors from 8192 to 4080 > >> [ 577.831703] scsi host6: ib_srp: Reducing max_sectors from 8192 to 4080 > >> > >> Its stable but only because we constrain the max_sectors. > >> I am writing 4MB buffered to 10 file systems (ext3) however the > >> max_sectors is constrained to 2040k and its stable. > >> > >> I have been told that RHEL6 and the Mofed stack happily does 4MB buffered > >> I/O with mapping failures but I need to check that out. > >> That would make no sense to me unless the Mofed stack has code that > >> differs enough to what we have upstream to make a difference. > >> > >> I am about to install RHEL6 and Mofed to compare and see what happens with > >> 4MB buffered. > >> > >> Tested-by Laurence Oberman > > > > Thanks for the help with testing :-) > > > > Can you try to increase SRP_MAX_PAGES_PER_MR in ib_srp.h from 512 to > > 1024 and see whether that allows 4MB I/O? > > (replying to my own e-mail) > > That was a typo - I meant 1025 instead of 1024. > > Bart. > > Set at 1025, rebuilt ib_srp module, rebooted client and: Still get: [root@srptest ~]# dmesg | grep Reducing [ 227.076550] scsi host4: ib_srp: Reducing max_sectors from 8192 to 4080 [ 235.316995] scsi host5: ib_srp: Reducing max_sectors from 8192 to 4080 [ 245.726223] scsi host5: ib_srp: Reducing max_sectors from 8192 to 4080 [ 253.197573] scsi host6: ib_srp: Reducing max_sectors from 8192 to 4080 And then of course we are constrained [root@srptest queue]# cat max_sectors_kb 2040 [root@srptest queue]# echo 4096 > max_sectors_kb -bash: echo: write error: Invalid argument -- 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