| Message ID | 56C662F1.8070407@ce.jp.nec.com (mailing list archive) |
|---|---|
| State | Not Applicable, archived |
| Headers | show |
On 02/19/16 09:33, Nomura Junichi wrote: > On 02/19/16 02:17, Mike Snitzer wrote: >> What is the last kernel version that your scripts have worked on? > > v4.4 worked fine. I'll check with v4.5-rc4 when I get a machine. v4.5-rc4 works fine, too. So if all tests fail for you, it might be due to difference in environment or kernel config. I'm running tests on RHEL7.2 and kernel config is based on RHEL7 kernel. # rpm -q targetcli targetcli-2.1.fb41-3.el7.noarch # cp /boot/config-3.10.0-327.el7.x86_64 .config # make olddefconfig ...
On Fri, Feb 19 2016 at 3:37am -0500, Junichi Nomura <j-nomura@ce.jp.nec.com> wrote: > On 02/19/16 09:33, Nomura Junichi wrote: > > On 02/19/16 02:17, Mike Snitzer wrote: > >> What is the last kernel version that your scripts have worked on? > > > > v4.4 worked fine. I'll check with v4.5-rc4 when I get a machine. > > v4.5-rc4 works fine, too. Have you been running with blk-mq? Either by setting CONFIG_DM_MQ_DEFAULT or: echo Y > /sys/module/dm_mod/parameters/use_blk_mq I'm seeing test_02_sdev_delete fail with blk-mq enabled. (and with my latest DM mpath code I'm testing for Linux 4.6: I'm seeing a nasty deadlock from test_01_sdev_offline... so that's "fun") > So if all tests fail for you, it might be due to difference in > environment or kernel config. After upgrading fio to latest fio.git I no longer see fio validate errors. > I'm running tests on RHEL7.2 and kernel config is based on RHEL7 kernel. > > # rpm -q targetcli > targetcli-2.1.fb41-3.el7.noarch > # cp /boot/config-3.10.0-327.el7.x86_64 .config > # make olddefconfig > ... Yes, I'm using a comparable setup. But with localmodconfig. -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Fri, Feb 19 2016 at 2:42pm -0500, Mike Snitzer <snitzer@redhat.com> wrote: > On Fri, Feb 19 2016 at 3:37am -0500, > Junichi Nomura <j-nomura@ce.jp.nec.com> wrote: > > > On 02/19/16 09:33, Nomura Junichi wrote: > > > On 02/19/16 02:17, Mike Snitzer wrote: > > >> What is the last kernel version that your scripts have worked on? > > > > > > v4.4 worked fine. I'll check with v4.5-rc4 when I get a machine. > > > > v4.5-rc4 works fine, too. > > Have you been running with blk-mq? > Either by setting CONFIG_DM_MQ_DEFAULT or: > echo Y > /sys/module/dm_mod/parameters/use_blk_mq > > I'm seeing test_02_sdev_delete fail with blk-mq enabled. I only see failure if I stack dm-mq ontop of old non-mq scsi devices with: echo N > /sys/module/scsi_mod/parameters/use_blk_mq echo Y > /sys/module/dm_mod/parameters/use_blk_mq If I use scsi-mq for the underlying devices all works fine (been testing the latest dm-4.6 branch though, I'll go back and try stock 4.5-rc4 just to double check). But this makes me think the novelty of having dm-mq support stacking on non-blk-mq devices was misplaced. It is a senseless config. I'll probably remove support for such stacking soon (next week). -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 02/20/2016 07:12 AM, Mike Snitzer wrote: > On Fri, Feb 19 2016 at 2:42pm -0500, > Mike Snitzer <snitzer@redhat.com> wrote: > >> On Fri, Feb 19 2016 at 3:37am -0500, >> Junichi Nomura <j-nomura@ce.jp.nec.com> wrote: >> >>> On 02/19/16 09:33, Nomura Junichi wrote: >>>> On 02/19/16 02:17, Mike Snitzer wrote: >>>>> What is the last kernel version that your scripts have worked on? >>>> >>>> v4.4 worked fine. I'll check with v4.5-rc4 when I get a machine. >>> >>> v4.5-rc4 works fine, too. >> >> Have you been running with blk-mq? >> Either by setting CONFIG_DM_MQ_DEFAULT or: >> echo Y > /sys/module/dm_mod/parameters/use_blk_mq >> >> I'm seeing test_02_sdev_delete fail with blk-mq enabled. > > I only see failure if I stack dm-mq ontop of old non-mq scsi devices with: > > echo N > /sys/module/scsi_mod/parameters/use_blk_mq > echo Y > /sys/module/dm_mod/parameters/use_blk_mq > > If I use scsi-mq for the underlying devices all works fine (been testing > the latest dm-4.6 branch though, I'll go back and try stock 4.5-rc4 just > to double check). > > But this makes me think the novelty of having dm-mq support stacking on > non-blk-mq devices was misplaced. It is a senseless config. I'll > probably remove support for such stacking soon (next week). Hmm. I must admit I really, really don't like these 'once-and-for-all' parameter. ATM the only SCSI driver to support SCSI-mq properly are lpfc, virtio, and fnic. None of the other driver have been modified, and I suspect the performance might be less than stellar. So there will be configurations where one might want to run scsi-mq alongside non-mq HBAs. I would really love to see to make that more granular so that these configurations can run efficiently. I know Christoph is violently against it, but I don't really see any solution presenting itself at the moment. Maybe a good topic for LSF ... Cheers, Hannes
On 02/20/16 01:42, Hannes Reinecke wrote: > ATM the only SCSI driver to support SCSI-mq properly are lpfc, virtio, > and fnic. None of the other driver have been modified, and I suspect the > performance might be less than stellar. Hello Hannes, Before scsi-mq support was added to the above drivers it was added to the ib_srp driver. Apparently today there are multiple drivers that support multiple hardware queues: $ git grep -nH 'nr_hw_queues = [^1]' | grep -vE 'block/blk-mq|scsi/scsi_lib' drivers/block/null_blk.c:667: nullb->tag_set.nr_hw_queues = submit_queues; drivers/block/virtio_blk.c:638: vblk->tag_set.nr_hw_queues = vblk->num_vqs; drivers/block/xen-blkfront.c:927: info->tag_set.nr_hw_queues = info->nr_rings; drivers/infiniband/ulp/srp/ib_srp.c:3337: target->scsi_host->nr_hw_queues = target->ch_count; drivers/nvme/host/pci.c:1680: dev->tagset.nr_hw_queues = dev->online_queues - 1; drivers/scsi/lpfc/lpfc_init.c:3318: shost->nr_hw_queues = phba->cfg_fcp_io_channel; drivers/scsi/virtio_scsi.c:1008: shost->nr_hw_queues = num_queues; Performance data is available in http://events.linuxfoundation.org/sites/events/files/slides/Vault%20-%20scsi-mq%20v2.pdf. Bart. -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Sat, Feb 20 2016 at 4:42am -0500, Hannes Reinecke <hare@suse.de> wrote: > On 02/20/2016 07:12 AM, Mike Snitzer wrote: > >On Fri, Feb 19 2016 at 2:42pm -0500, > >Mike Snitzer <snitzer@redhat.com> wrote: > > > >>On Fri, Feb 19 2016 at 3:37am -0500, > >>Junichi Nomura <j-nomura@ce.jp.nec.com> wrote: > >> > >>>On 02/19/16 09:33, Nomura Junichi wrote: > >>>>On 02/19/16 02:17, Mike Snitzer wrote: > >>>>>What is the last kernel version that your scripts have worked on? > >>>> > >>>>v4.4 worked fine. I'll check with v4.5-rc4 when I get a machine. > >>> > >>>v4.5-rc4 works fine, too. > >> > >>Have you been running with blk-mq? > >>Either by setting CONFIG_DM_MQ_DEFAULT or: > >>echo Y > /sys/module/dm_mod/parameters/use_blk_mq > >> > >>I'm seeing test_02_sdev_delete fail with blk-mq enabled. > > > >I only see failure if I stack dm-mq ontop of old non-mq scsi devices with: > > > >echo N > /sys/module/scsi_mod/parameters/use_blk_mq > >echo Y > /sys/module/dm_mod/parameters/use_blk_mq > > > >If I use scsi-mq for the underlying devices all works fine (been testing > >the latest dm-4.6 branch though, I'll go back and try stock 4.5-rc4 just > >to double check). > > > >But this makes me think the novelty of having dm-mq support stacking on > >non-blk-mq devices was misplaced. It is a senseless config. I'll > >probably remove support for such stacking soon (next week). > > Hmm. I must admit I really, really don't like these 'once-and-for-all' > parameter. > > ATM the only SCSI driver to support SCSI-mq properly are lpfc, > virtio, and fnic. None of the other driver have been modified, and I > suspect the performance might be less than stellar. > > So there will be configurations where one might want to run scsi-mq > alongside non-mq HBAs. dm-mq already disallows such a mix though. From dm_table_set_type(): if (use_blk_mq) { /* verify _all_ devices in the table are blk-mq devices */ list_for_each_entry(dd, devices, list) if (!bdev_get_queue(dd->dm_dev->bdev)->mq_ops) { DMERR("table load rejected: not all devices" " are blk-mq request-stackable"); return -EINVAL; } t->type = DM_TYPE_MQ_REQUEST_BASED; } But I was talking about removing support for dm-mq stacked on _all_ old .request_fn devices. > I would really love to see to make that more granular so that these > configurations can run efficiently. > I know Christoph is violently against it, but I don't really see any > solution presenting itself at the moment. I'm with Christoph on this. Supporting such elaborate mixing is too fragile. Time would be much better spent converting drivers to properly support scsi-mq and/or fixing scsi-mq to perform better. > Maybe a good topic for LSF ... Unlikely.. but don't let me stop you! ;) -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Thu, Feb 18 2016 at 7:33pm -0500, Junichi Nomura <j-nomura@ce.jp.nec.com> wrote: > Hi Mike, > > On 02/19/16 02:17, Mike Snitzer wrote: > > > Taking a step back: > > These scripts don't belong in Documentation/device-mapper/mptest/ (or > > anywhere in the kernel tree for that matter). > > > > I'd really prefer it if we could port your scripts over to the > > device-mapper-test-suite, see: > > https://github.com/jthornber/device-mapper-test-suite > > Yes, I agree such a project is better place for this to live. I was going to attempt porting your scripts to device-mapper-test-suite but I'll have to come back to that (I have more important tasks at this time). So I've created a guthub repo for your scripts: https://github.com/snitm/mptest I'll let you know once I've ported to device-mapper-test-suite. But in the meantime I'll take any changes you or others have to 'mptest'. Thanks, Mike -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/lib/failpath_dm_message b/lib/failpath_dm_message index 1a3bcf8..5b8f28a 100755 --- a/lib/failpath_dm_message +++ b/lib/failpath_dm_message @@ -30,9 +30,11 @@ start_failpath_dm_message () { for m in $majs; do dmsetup message $MPNAME 0 "fail_path $m" done + sleep 1 for m in $majs; do dmsetup message $MPNAME 0 "reinstate_path $m" done + sleep 1 done & } --