Re: [RFC PATCH 0/4] dm mpath: vastly improve blk-mq IO performance

On Fri, Apr 08 2016 at 7:42am -0400, Johannes Thumshirn wrote: > Ladies and Gentlemen, > To show off some numbers from our testing: > > All tests are performed against the cache of the Array, not the disks as we > wanted to test the Linux stack not the Disk Array. > > All

Re: [RFC PATCH 0/4] dm mpath: vastly improve blk-mq IO performance

On Thu, Apr 07 2016 at 10:58am -0400, Hannes Reinecke <h...@suse.de> wrote: > On 03/31/2016 10:04 PM, Mike Snitzer wrote: > > I developed these changes some weeks ago but have since focused on > > regression and performance testing on larger NUMA systems. > > > >

Re: [RFC PATCH 0/4] dm mpath: vastly improve blk-mq IO performance

On Fri, Apr 01 2016 at 9:37am -0400, Johannes Thumshirn <jthumsh...@suse.de> wrote: > [ +Cc Hannes ] > > On 2016-04-01 15:22, Mike Snitzer wrote: > >On Fri, Apr 01 2016 at 4:12am -0400, > >Johannes Thumshirn <jthumsh...@suse.de> wrote: > > > >>

Re: [RFC PATCH 0/4] dm mpath: vastly improve blk-mq IO performance

On Fri, Apr 01 2016 at 4:12am -0400, Johannes Thumshirn <jthumsh...@suse.de> wrote: > On 2016-03-31 22:04, Mike Snitzer wrote: > >I developed these changes some weeks ago but have since focused on > >regression and performance testing on larger NUMA systems. > > >

[RFC PATCH 3/4] dm mpath: move trigger_event member to the end of 'struct multipath'

Allows the 'work_mutex' member to no longer cross a cacheline. Signed-off-by: Mike Snitzer <snit...@redhat.com> --- drivers/md/dm-mpath.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c index 780e5d0..54daf96

[RFC PATCH 1/4] dm mpath: switch to using bitops for state flags

Mechanical change that doesn't make any real effort to reduce the use of m->lock; that will come later (once atomics are used for counters, etc). Suggested-by: Hannes Reinecke <h...@suse.de> Signed-off-by: Mike Snitzer <snit...@redhat.com> --- drivers/md/d

[RFC PATCH 2/4] dm mpath: use atomic_t for counting members of 'struct multipath'

The use of atomic_t for nr_valid_paths, pg_init_in_progress and pg_init_count will allow relaxing the use of the m->lock spinlock. Suggested-by: Hannes Reinecke <h...@suse.de> Signed-off-by: Mike Snitzer <snit...@redhat.com> --- drivers/md/

[RFC PATCH 0/4] dm mpath: vastly improve blk-mq IO performance

ow to raise awareness about how I think DM multipath will be changing (for inclussion during the Linux 4.7 merge window). Mike Snitzer (4): dm mpath: switch to using bitops for state flags dm mpath: use atomic_t for counting members of 'struct multipath' dm mpath: move trigger_event member to

[RFC PATCH 4/4] dm mpath: eliminate use of spinlock in IO fast-paths

<jmo...@redhat.com> Signed-off-by: Mike Snitzer <snit...@redhat.com> --- drivers/md/dm-mpath.c | 170 +++--- 1 file changed, 93 insertions(+), 77 deletions(-) diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c index 54daf96..52ba

Re: scsi-debug regression with 4.5-rc?

On Wed, Jan 20 2016 at 10:57am -0500, Martin K. Petersen wrote: > > "Ewan" == Ewan Milne writes: > > Ewan> So I have a report from our test people that the optimal_io_size > Ewan> sysfs value is now different by a factor of 512 from what it

Re: dm-multipath test scripts

On Thu, Feb 18 2016 at 7:33pm -0500, Junichi Nomura <j-nom...@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 > > anywher

why is blk-mq requeue foricbly kicking stopped queues? [was: Re: dm-multipath test scripts]

On Mon, Feb 22 2016 at 4:51am -0500, Junichi Nomura <j-nom...@ce.jp.nec.com> wrote: > On 02/20/16 15:12, Mike Snitzer wrote: > > On Fri, Feb 19 2016 at 2:42pm -0500, Mike Snitzer <snit...@redhat.com> > > wrote: > >> Have you been running with blk-mq? > &g

Re: dm-multipath test scripts

On Sat, Feb 20 2016 at 4:42am -0500, Hannes Reinecke <h...@suse.de> wrote: > On 02/20/2016 07:12 AM, Mike Snitzer wrote: > >On Fri, Feb 19 2016 at 2:42pm -0500, > >Mike Snitzer <snit...@redhat.com> wrote: > > > >>On Fri, Feb 19 2016 at 3:37am -0500,

Re: dm-multipath test scripts

On Fri, Feb 19 2016 at 2:42pm -0500, Mike Snitzer <snit...@redhat.com> wrote: > On Fri, Feb 19 2016 at 3:37am -0500, > Junichi Nomura <j-nom...@ce.jp.nec.com> wrote: > > > On 02/19/16 09:33, Nomura Junichi wrote: > > > On 02/19/16 02:17, Mike Snitzer w

Re: dm-multipath test scripts

On Fri, Feb 19 2016 at 3:37am -0500, Junichi Nomura <j-nom...@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 w

Re: dm-multipath test scripts

On Wed, Oct 07 2015 at 1:39am -0400, Junichi Nomura wrote: > This is a set of scripts for kernel-side dm-multipath testing. Current > set of scripts are stress testing of extreme situation and its coverage > is limited. But recently found dm-mpath regressions should be

Re: dm-mpath: fix a tiny case which can cause an infinite loop

On Wed, Feb 03 2016 at 9:08pm -0500, jiangyiwen wrote: > When two processes submit WRTIE SAME bio simultaneously and > first IO return failed because of INVALID FIELD IN CDB, and > then second IO can enter into an infinite loop. > The problem can be described as follows:

Re: block: Always check queue limits for cloned requests

On Sun, Nov 29 2015 at 11:15am -0500, Markus Trippelsdorf wrote: > On 2015.11.29 at 16:43 +0100, Hannes Reinecke wrote: > > On 11/29/2015 12:49 PM, Markus Trippelsdorf wrote: > > > > > > I'm still seeing the issue (BUG at drivers/scsi/scsi_lib.c:1096!) even > > > with

Re: block: Always check queue limits for cloned requests

leading to a crash in scsi_init_sgtable(). > > To clarify this the patch renames blk_rq_check_limits() > to blk_cloned_rq_check_limits() and removes the symbol > export, as the new function should only be used for > cloned requests and never exported. > > Cc: Mike Snitzer <

Re: kernel BUG at drivers/scsi/scsi_lib.c:1096!

On Wed, Nov 25 2015 at 4:04am -0500, Hannes Reinecke wrote: > On 11/20/2015 04:28 PM, Ewan Milne wrote: > > On Fri, 2015-11-20 at 15:55 +0100, Hannes Reinecke wrote: > >> Can't we have a joint effort here? > >> I've been spending a _LOT_ of time trying to debug things here, but >

Re: kernel BUG at drivers/scsi/scsi_lib.c:1096!

On Wed, Nov 25 2015 at 3:23pm -0500, Mike Snitzer <snit...@redhat.com> wrote: > On Wed, Nov 25 2015 at 2:24pm -0500, > Jens Axboe <ax...@fb.com> wrote: > > > On 11/25/2015 12:10 PM, Hannes Reinecke wrote: > > >The problem is that NOMERGE does to

Re: kernel BUG at drivers/scsi/scsi_lib.c:1096!

On Wed, Nov 25 2015 at 2:24pm -0500, Jens Axboe wrote: > On 11/25/2015 12:10 PM, Hannes Reinecke wrote: > >On 11/25/2015 06:56 PM, Jens Axboe wrote: > >>On 11/25/2015 02:04 AM, Hannes Reinecke wrote: > >>>On 11/20/2015 04:28 PM, Ewan Milne wrote: > On Fri, 2015-11-20 at 15:55

Re: dm-multipath test scripts

On Fri, Nov 13 2015 at 5:55am -0500, Johannes Thumshirn wrote: > On Wed, 2015-10-07 at 05:39 +, Junichi Nomura wrote: > > This is a set of scripts for kernel-side dm-multipath testing. > > Current > > set of scripts are stress testing of extreme situation and its > >

Re: [RFC PATCH 00/32] separate operations from flags in the bio/request structs

On Wed, Nov 11 2015 at 6:28am -0500, Christoph Hellwig wrote: > On Wed, Nov 11, 2015 at 01:53:24AM -0600, Mike Christie wrote: > > We no longer have the bvec merge functions so the original reason given > > in the thread/patch Bart referenced is no longer valid. > > > >

Re: IBM request to allow unprivledged ioctls [Was: Revert "dm mpath: fix stalls when handling invalid ioctls"]

On Mon, Nov 02 2015 at 10:45am -0500, Paolo Bonzini <pbonz...@redhat.com> wrote: > > > On 02/11/2015 16:05, Mike Snitzer wrote: > > > In any case, if we don't start path activation we should return > > > ENOTCONN, not ENOTTY. > > > > Currently, if w

Re: IBM request to allow unprivledged ioctls [Was: Revert "dm mpath: fix stalls when handling invalid ioctls"]

On Mon, Nov 02 2015 at 2:28am -0500, Hannes Reinecke <h...@suse.de> wrote: > On 10/31/2015 11:47 PM, Mike Snitzer wrote: > > > > For Hannes, and in my head, it didn't matter if a future bdev satisfies > > the length condition. I don't think Hannes was trying to

Re: IBM request to allow unprivledged ioctls [Was: Revert "dm mpath: fix stalls when handling invalid ioctls"]

On Mon, Nov 02 2015 at 9:52am -0500, Paolo Bonzini wrote: > > > On 02/11/2015 14:56, Hannes Reinecke wrote: > > But then the real question remains: > > > > What is the 'correct' behaviour for ioctls when no path retry > > is active (or when no paths are present)? > > >

Re: IBM request to allow unprivledged ioctls [Was: Revert "dm mpath: fix stalls when handling invalid ioctls"]

On Mon, Nov 02 2015 at 8:56am -0500, Hannes Reinecke <h...@suse.de> wrote: > On 11/02/2015 02:31 PM, Mike Snitzer wrote: > > On Mon, Nov 02 2015 at 2:28am -0500, > > Hannes Reinecke <h...@suse.de> wrote: > > > >> On 10/31/2015 11:47 PM, Mike Snitze

Re: IBM request to allow unprivledged ioctls [Was: Revert "dm mpath: fix stalls when handling invalid ioctls"]

On Mon, Nov 02 2015 at 9:36am -0500, Hannes Reinecke <h...@suse.de> wrote: > On 11/02/2015 03:12 PM, Mike Snitzer wrote: > > On Mon, Nov 02 2015 at 8:56am -0500, > > Hannes Reinecke <h...@suse.de> wrote: > > > >> On 11/02/2015 02:31 PM, Mike Snitzer w

Re: IBM request to allow unprivledged ioctls [Was: Revert "dm mpath: fix stalls when handling invalid ioctls"]

On Sat, Oct 31 2015 at 11:33am -0400, Paolo Bonzini <pbonz...@redhat.com> wrote: > > > On 29/10/2015 14:18, Mike Snitzer wrote: > > > 4) dmesg shows that scsi_verify_blk_ioctl() failed for SG_IO (0x2285); > > >it returns -ENOIOCTLCMD, later r

Re: IBM request to allow unprivledged ioctls [Was: Revert "dm mpath: fix stalls when handling invalid ioctls"]

On Sat, Oct 31 2015 at 2:13P -0400, Mike Snitzer <snit...@redhat.com> wrote: > On Sat, Oct 31 2015 at 11:33am -0400, > Paolo Bonzini <pbonz...@redhat.com> wrote: > > > > > > > On 29/10/2015 14:18, Mike Snitzer wrote: > > > > 4) dmesg shows th

Re: IBM request to allow unprivledged ioctls [Was: Revert "dm mpath: fix stalls when handling invalid ioctls"]

On Sat, Oct 31 2015 at 3:07pm -0400, Paolo Bonzini <pbonz...@redhat.com> wrote: > > > On 31/10/2015 19:13, Mike Snitzer wrote: > > > But that's wrong, I think. It's a false positive in > > > scsi_verify_blk_ioctl(). > > > > > &g

IBM request to allow unprivledged ioctls [Was: Revert "dm mpath: fix stalls when handling invalid ioctls"]

On Thu, Oct 29 2015 at 8:24am -0400, Mauricio Faria de Oliveira wrote: > This reverts commit a1989b330093578ea5470bea0a00f940c444c466. > > That commit introduced a regression at least for the case of the SG_IO ioctl() > running without CAP_SYS_RAWIO capability

Re: Persistent Reservation API V4

On Thu, Oct 15 2015 at 8:10am -0400, Christoph Hellwig wrote: > This series adds support for a simplified Persistent Reservation API > to the block layer. The intent is that both in-kernel and userspace > consumers can use the API instead of having to hand craft SCSI or NVMe >

Re: Persistent Reservation API V4

On Wed, Oct 28 2015 at 2:43pm -0400, Mike Snitzer <snit...@redhat.com> wrote: > On Thu, Oct 15 2015 at 8:10am -0400, > Christoph Hellwig <h...@lst.de> wrote: > > > This series adds support for a simplified Persistent Reservation API > > to the block layer.

Re: Bugs in multipath scsi in 4.3-rc2

On Mon, Oct 12, 2015 at 10:39 AM, Christoph Hellwig wrote: > On Mon, Oct 12, 2015 at 02:45:38PM +0200, Hannes Reinecke wrote: >> You cannot easily use the driver model here as the scsi_device is >> already (potentially) bound to the ULDs. >> If you were to go with the driver model

Re: req->nr_phys_segments > queue_max_segments (was Re: kernel BUG at drivers/block/virtio_blk.c:172!)

On Thu, Oct 1, 2015 at 5:20 AM, Hannes Reinecke wrote: > On 10/01/2015 11:00 AM, Michael S. Tsirkin wrote: >> On Thu, Oct 01, 2015 at 03:10:14AM +0200, Thomas D. wrote: >>> Hi, >>> >>> I have a virtual machine which fails to boot linux-4.1.8 while mounting >>> file systems: >>>

Re: [PATCH 05/17] Update the md driver to use idr helper functions.

Subject should really be: "dm: update to use idr helper functions" -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: randconfig build error with next-20150908, in drivers/md/dm-mpath.c

On Tue, Sep 08 2015 at 1:34pm -0400, James Bottomley <jbottom...@odin.com> wrote: > On Tue, 2015-09-08 at 13:23 -0400, Mike Snitzer wrote: > > On Tue, Sep 08 2015 at 1:10pm -0400, > > Jim Davis <jim.ep...@gmail.com> wrote: > > > > > Building w

Re: Persistent Reservation API

On Tue, Aug 04 2015 at 3:11am -0400, Christoph Hellwig h...@lst.de wrote: This series adds support for a simplified persistent reservation API to the block layer. The intent is that both in-kernel and userspace consumers can use the API instead of having to hand craft SCSI or NVMe command

Re: dm-mpath: always return reservation conflict

On Wed, Jul 15 2015 at 8:15am -0400, Hannes Reinecke h...@suse.de wrote: On 07/15/2015 02:01 PM, James Bottomley wrote: On Wed, 2015-07-15 at 13:52 +0200, Hannes Reinecke wrote: On 07/15/2015 01:35 PM, James Bottomley wrote: On Wed, 2015-07-15 at 13:23 +0200, Hannes Reinecke wrote: If

Re: [PATCH 4/9] dm-mpath, scsi_dh: request scsi_dh modules in scsi_dh, not dm-mpath

On Thu, Apr 30 2015 at 1:32pm -0400, Christoph Hellwig h...@lst.de wrote: This way we can reused the same code any attachment method, not just those requested from dm-mpath. Signed-off-by: Christoph Hellwig h...@lst.de Acked-by: Mike Snitzer snit...@redhat.com -- To unsubscribe from

Re: [PATCH 1/7] blk-mq: fix use of incorrect goto label in blk_mq_init_queue error path

On Thu, Mar 12 2015 at 3:48am -0400, Ming Lei tom.leim...@gmail.com wrote: On Thu, Mar 12, 2015 at 11:56 AM, Mike Snitzer snit...@redhat.com wrote: If percpu_ref_init() fails the 'err_hctxs' label should be used instead of 'err_map'. Signed-off-by: Mike Snitzer snit...@redhat.com

[PATCH 1/7 v3] blk-mq: fix use of incorrect goto label in blk_mq_init_queue error path

If percpu_ref_init() fails the allocated q and hctxs must get cleaned up; using 'err_map' doesn't allow that to happen. Signed-off-by: Mike Snitzer snit...@redhat.com Reviewed-by: Ming Lei ming@canonical.com --- block/blk-mq.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff

Re: [PATCH 1/7] blk-mq: fix use of incorrect goto label in blk_mq_init_queue error path

On Thu, Mar 12 2015 at 9:51am -0400, Mike Snitzer snit...@redhat.com wrote: On Thu, Mar 12 2015 at 3:48am -0400, Ming Lei tom.leim...@gmail.com wrote: On Thu, Mar 12, 2015 at 11:56 AM, Mike Snitzer snit...@redhat.com wrote: If percpu_ref_init() fails the 'err_hctxs' label should

[PATCH 2/7 v3] blk-mq: add blk_mq_init_allocated_queue and export blk_mq_register_disk

of the request_queue type the call to blk_mq_register_disk() doesn't happen during alloc_dev(). Must export blk_mq_register_disk() so that DM can backfill the 'mq' dir once the blk-mq queue is fully allocated. Signed-off-by: Mike Snitzer snit...@redhat.com Reviewed-by: Ming Lei ming@canonical.com --- block

[PATCH 4/7] blk-mq: don't wait in blk_mq_queue_enter() if __GFP_WAIT isn't set

From: Keith Busch keith.bu...@intel.com Return -EBUSY if we're unable to enter a queue immediately when allocating a blk-mq request without __GFP_WAIT. Signed-off-by: Keith Busch keith.bu...@intel.com Signed-off-by: Mike Snitzer snit...@redhat.com --- block/blk-mq.c | 9 ++--- 1 file

[PATCH 6/7] dm: optimize dm_mq_queue_rq to _not_ use kthread if using pure blk-mq

). But the kthread will still be used to queue work if blk-mq is used ontop of old-style request_fn device(s). Also prepare for supporting DM blk-mq ontop of old-style request_fn device(s) if a new dm-mod 'use_blk_mq' parameter is set. Signed-off-by: Mike Snitzer snit...@redhat.com --- drivers/md

[PATCH 3/7] blk-mq: export blk_mq_run_hw_queues

Rename blk_mq_run_queues to blk_mq_run_hw_queues, add async argument, and export it. DM's suspend support must be able to run the queue without starting stopped hw queues. Signed-off-by: Mike Snitzer snit...@redhat.com --- block/blk-mq.c | 8 include/linux/blk-mq.h | 1 + 2

[PATCH 0/7] dm: add full blk-mq support to request-based DM

for your help answering more blk-mq questions than I'd have hoped would be needed while at LSF/MM. Keith Busch (1): blk-mq: don't wait in blk_mq_queue_enter() if __GFP_WAIT isn't set Mike Snitzer (6): blk-mq: fix use of incorrect goto label in blk_mq_init_queue error path blk-mq: add

[PATCH 5/7] dm: add full blk-mq support to request-based DM

could only push ~20% of the raw iops through dm before this conversion, so this latest tree is looking really solid from a performance standpoint. Tested-by: Keith Busch keith.bu...@intel.com Signed-off-by: Mike Snitzer snit...@redhat.com --- drivers/md/dm-mpath.c | 4 +- drivers/md/dm

[PATCH 1/7] blk-mq: fix use of incorrect goto label in blk_mq_init_queue error path

If percpu_ref_init() fails the 'err_hctxs' label should be used instead of 'err_map'. Signed-off-by: Mike Snitzer snit...@redhat.com --- block/blk-mq.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/block/blk-mq.c b/block/blk-mq.c index 4f4bea2..459840c 100644 --- a/block

[PATCH 2/7] blk-mq: add blk_mq_init_allocated_queue and export blk_mq_register_disk

with alloc_dev() but the decision about what type of request_queue will be ultimately created is deferred until all component devices referenced in the DM table are processed to determine the table type (request-based, blk-mq request-based, or bio-based). Signed-off-by: Mike Snitzer snit...@redhat.com

[PATCH 7/7] dm: add 'use_blk_mq' module param and expose in per-device ro sysfs attr

and reduced work (e.g. the md-io_pool and md-rq_pool isn't created if using blk-mq). Signed-off-by: Mike Snitzer snit...@redhat.com --- drivers/md/Kconfig| 11 +++ drivers/md/dm-sysfs.c | 9 + drivers/md/dm-table.c | 6 +++--- drivers/md/dm.c | 51

Re: dm + blk-mq soft lockup complaint

On Thu, Jan 15 2015 at 3:11am -0500, Bart Van Assche bart.vanass...@sandisk.com wrote: On 01/14/15 20:00, Mike Snitzer wrote: IB aside, I haven't been following along close enough on scsi-mq developments, but does a regular iscsi initiator have support for scsi-mq? I'd like to validate

Re: dm + blk-mq soft lockup complaint

On Wed, Jan 14 2015 at 4:16am -0500, Bart Van Assche bart.vanass...@sandisk.com wrote: On 01/13/15 17:21, Mike Snitzer wrote: OK, I assume you specified the mpath device for the test that failed. Yes, of course ... This test works fine on my 100MB scsi_debug device with 4 paths

Re: dm + blk-mq soft lockup complaint

On Tue, Jan 13 2015 at 9:28am -0500, Bart Van Assche bart.vanass...@sandisk.com wrote: On 01/13/15 15:18, Mike Snitzer wrote: On Tue, Jan 13 2015 at 7:29am -0500, Bart Van Assche bart.vanass...@sandisk.com wrote: However, I hit another issue while running I/O on top of a multipath

Re: [PATCH 0/15] SCSI XCOPY support for the kernel and device mapper

On Tue, Jul 15 2014 at 3:34pm -0400, Mikulas Patocka mpato...@redhat.com wrote: This patch series makes it possible to use SCSI XCOPY offload for the block layer and device mapper. It is based on Martin Petersen's work

Re: Kernel bug triggered in multipath

it cause he was trying to get cciss working with dm-multipath. Acked-by: Mike Snitzer snit...@redhat.com -- To unsubscribe from this list: send the line unsubscribe linux-scsi in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: Kernel bug triggered in multipath

On Fri, Mar 14 2014 at 12:21pm -0400, Mike Snitzer snit...@redhat.com wrote: I have no problem with this patch, added safety-net and all, but bottomline: if scsi_dh interfaces were being called against a DM multipath request_queue that is a bug. Sorry, s/DM multipath request_queue/non-SCSI

[LSF/MM ATTEND] interest in blk-mq, scsi-mq, dm-cache, dm-thinp, dm-*

I would like to attend to participate in discussions related to topics listed in the subject. As a maintainer of DM I'd be interested to learn/discuss areas that should become a development focus in the months following LSF. Thanks, Mike -- To unsubscribe from this list: send the line

Re: SCSI's heuristics for enabling WRITE SAME still need work [was: dm mpath: disable WRITE SAME if it fails]

On Tue, Sep 24 2013 at 9:49am -0400, Martin K. Petersen martin.peter...@oracle.com wrote: Mike == Mike Snitzer snit...@redhat.com writes: Mike So are there drives like this?: Mike 1) don't support RSOC Mike 2) do support WRITE SAME Mike 3) do populate VPD page with either WRITE SAME w

Re: SCSI's heuristics for enabling WRITE SAME still need work [was: dm mpath: disable WRITE SAME if it fails]

On Tue, Sep 24 2013 at 4:44pm -0400, Martin K. Petersen martin.peter...@oracle.com wrote: The other headache is what happens if a stacking driver gets -EIO or -EREMOTEIO on a WRITE SAME request. That's what Mike was trying to fix with his patch. Maybe it would make sense for us to use -EINVAL

SCSI's heuristics for enabling WRITE SAME still need work [was: dm mpath: disable WRITE SAME if it fails]

On Thu, Sep 19 2013 at 12:13pm -0400, Mike Snitzer snit...@redhat.com wrote: Workaround the SCSI layer's problematic WRITE SAME heuristics by disabling WRITE SAME in the DM multipath device's queue_limits if an underlying device disabled it. ... This fix doesn't help configurations

[PATCH] dm mpath: disable WRITE SAME if it fails

wasn't enabled in DM multipath until v3.10. Signed-off-by: Mike Snitzer snit...@redhat.com Cc: Martin K. Petersen martin.peter...@oracle.com Cc: Hannes Reinecke h...@suse.de Cc: sta...@vger.kernel.org # 3.10+ --- drivers/md/dm-mpath.c | 11 ++- drivers/md/dm.c | 11

Re: [PATCH 2/2] dm mpath: attach scsi_dh during table resume

On Fri, Apr 26 2013 at 2:05am -0400, Hannes Reinecke h...@suse.de wrote: On 04/25/2013 05:31 PM, Mike Snitzer wrote: On Thu, Apr 25 2013 at 10:50am -0400, Mikulas Patocka mpato...@redhat.com wrote: On Thu, 25 Apr 2013, Mike Snitzer wrote: On Thu, Apr 25 2013 at 9:48am -0400

Re: [PATCH 2/2] dm mpath: attach scsi_dh during table resume

On Thu, Apr 25 2013 at 9:48am -0400, Mikulas Patocka mpato...@redhat.com wrote: On Mon, 22 Apr 2013, Mike Snitzer wrote: I spoke with Hannes at LSF, to address the potential crashes in the endio path (e.g. stpg_endio) we'd have to bump the scsi_dh_data kref where appropriate (e.g

Re: [PATCH 2/2] dm mpath: attach scsi_dh during table resume

On Thu, Apr 25 2013 at 10:50am -0400, Mikulas Patocka mpato...@redhat.com wrote: On Thu, 25 Apr 2013, Mike Snitzer wrote: On Thu, Apr 25 2013 at 9:48am -0400, Mikulas Patocka mpato...@redhat.com wrote: On Mon, 22 Apr 2013, Mike Snitzer wrote: I spoke with Hannes

Re: [PATCH 2/2] dm mpath: attach scsi_dh during table resume

On Thu, Apr 25 2013 at 11:27am -0400, Bryn M. Reeves b...@redhat.com wrote: On 04/25/2013 03:50 PM, Mikulas Patocka wrote: On Thu, 25 Apr 2013, Mike Snitzer wrote: The handler that is automatically attached _should_ be the correct handler. We now have the .match() hook for scsi_dh and it has

Re: sd: workaround invalid OPTIMAL TRANSFER LENGTH

On Wed, Apr 24 2013 at 9:19pm -0400, Martin K. Petersen martin.peter...@oracle.com wrote: Mike == Mike Snitzer snit...@redhat.com writes: Mike, Following up on our discussion at LSF/MM last week... Mike Workaround disk firmware that improperly sets OPTIMAL TRANSFER Mike LENGTH

Re: [PATCH 2/2] dm mpath: attach scsi_dh during table resume

On Mon, Apr 08 2013 at 5:50pm -0400, Mike Snitzer snit...@redhat.com wrote: Preallocate scsi_dh_data using scsi_dh_alloc_data() during table load but attach the scsi_dh for each path during table resume. This avoids a kernel crash that can happen when changing the scsi_dh during table load

[PATCH 1/2] [SCSI] scsi_dh: add scsi_dh_alloc_data

. This change will allow DM multipath to preallocate the scsi_dh_data during the multipath table load but then defer the scsi_dh_attach until the multipath table resume (memory allocation is not allowed during DM table resume). Signed-off-by: Mike Snitzer snit...@redhat.com --- drivers/md/dm-mpath.c

[PATCH 2/2] dm mpath: attach scsi_dh during table resume

is that the endio routine for the path-management request is working with structures that were freed when the handler was detached. References: http://bugzilla.redhat.com/912245 http://bugzilla.redhat.com/902595 Signed-off-by: Mike Snitzer snit...@redhat.com --- drivers/md/dm-mpath.c | 160

Re: dm: do not replace bioset for request-based dm

...@ce.jp.nec.com Cc: Alasdair G Kergon a...@redhat.com Cc: Mikulas Patocka mpato...@redhat.com Cc: Mike Snitzer snit...@redhat.com Cc: sta...@vger.kernel.org Acked-by: Mike Snitzer snit...@redhat.com -- To unsubscribe from this list: send the line unsubscribe linux-scsi in the body of a message

Re: [PATCH 2/2] dm: Avoid use-after-free of a mapped device

with v3.7+ guidance. Acked-by: Mike Snitzer snit...@redhat.com -- To unsubscribe from this list: send the line unsubscribe linux-scsi in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: [PATCH 2/2] dm: Avoid use-after-free of a mapped device

On Fri, Feb 22 2013 at 6:22am -0500, Bart Van Assche bvanass...@acm.org wrote: On 02/22/13 12:08, Mike Snitzer wrote: On Fri, Feb 22 2013 at 5:47am -0500, Bart Van Assche bvanass...@acm.org wrote: As the comment above rq_completed() explains, md members must not be touched after

Re: [PATCH 00/10] scsi: avoid linebreaks in syslog output

On Fri, Oct 12, 2012 at 4:33 AM, Hannes Reinecke h...@suse.de wrote: This patchset updates the SCSI midlayer to use dev_printk() instead of the simple printk(). The main objective here is to avoid line-breaks in syslog output; with the current state it's nearly impossible to match the output

Re: [PATCH v3] scsi_lib: rate-limit the error message from failing commands

On Thu, Aug 30, 2012 at 7:25 PM, Jens Axboe ax...@kernel.dk wrote: On 2012-08-30 14:06, Yi Zou wrote: [ Jens/James, This is a rather old rate limt patch but never gets picked up in upstream, so I am resending it here as v3, with some minor changes so it is directly applicable to the

Re: scsi_dh_alua: Enable STPG for unavailable ports

: [ ... ] d) SET TARGET PORT GROUPS; [ ... ]. Hence enable sending STPG to a target port group that is in the unavailable state. Signed-off-by: Bart Van Assche bvanass...@acm.org Cc: Hannes Reinecke h...@suse.de Cc: Mike Christie micha...@cs.wisc.edu Cc: Mike Snitzer snit...@redhat.com Cc: Babu

Re: [PATCH 2/8] block: Consolidate command flag and queue limit checks for merges

-limits.max_hw_sectors; Or queue_max_hw_sectors() here? + return blk_queue_get_max_sectors(q, rq-cmd_flags); +} + But I don't feel that strongly on these nits. So regardless: Acked-by: Mike Snitzer snit...@redhat.com (thanks for getting this patchset together Martin) -- To unsubscribe from

Re: [patch 1/2]block: handle merged discard request

On Fri, Aug 17 2012 at 11:47pm -0400, Martin K. Petersen martin.peter...@oracle.com wrote: Mike == Mike Snitzer snit...@redhat.com writes: Mike Could be I've wasted a few hours by rebasing these patches... Mike regardless, it would be great if you could share what your plans Mike

Re: [patch 1/2]block: handle merged discard request

On Mon, Aug 20 2012 at 9:58am -0400, Christoph Hellwig h...@infradead.org wrote: On Mon, Aug 20, 2012 at 09:57:39AM -0400, Mike Snitzer wrote: But I haven't put my finger on _why_ a discard bio has bio-bi_io_vec (but given my use of DM, bio comes from bio_alloc_bioset, and DM passes

Re: [patch 1/2]block: handle merged discard request

On Thu, Jun 21, 2012 at 11:46 PM, Martin K. Petersen martin.peter...@oracle.com wrote: Christoph == Christoph Hellwig h...@infradead.org writes: There are several additional commands in the pipeline where the 1:1 mapping between DMA size and block range is invalid. I want to get rid of the

Re: scsi_dh_rdac: Add empty set_params function to scsi_dh_rdac

On Wed, Aug 08 2012 at 12:10pm -0400, Moger, Babu babu.mo...@netapp.com wrote: This patch adds empty set_params function to scsi_dh_rdac. This patch is required for the following features to work properly. 1. add retain_attached_hw_handler feature

[RESEND PATCH] SCSI: scsi_lib: fix scsi_io_completion's SG_IO error propagation

in cdb Raw sense data (in hex): 70 00 05 00 00 00 00 0b 00 00 00 00 24 00 00 00 00 00 00 Signed-off-by: Mike Snitzer snit...@redhat.com Reported-by: Paolo Bonzini pbonz...@redhat.com Tested-by: Paolo Bonzini pbonz...@redhat.com Reviewed-by: Babu Moger babu.mo...@netapp.com Cc: sta

Re: scsi_lib: fix scsi_io_completion's SG_IO error propagation

Hi James, Seems this one slipped through the cracks. Please review/apply. Thanks, Mike On Thu, May 31 2012 at 4:33pm -0400, Moger, Babu babu.mo...@netapp.com wrote: Thanks Mike. Looks good. -Original Message- From: Mike Snitzer [mailto:snit...@redhat.com] Sent: Thursday, May

Re: [RESEND PATCH 2/3] dm mpath: add 'default_hw_handler' feature

On Wed, Jul 11 2012 at 6:27am -0400, Alasdair G Kergon a...@redhat.com wrote: On Wed, Jun 27, 2012 at 08:16:55AM +0200, Hannes Reinecke wrote: On 06/26/2012 08:32 PM, Mike Snitzer wrote: When specifying the feature 'default_hw_handler' multipath will use the currently attached hardware

Re: [PATCH] [SCSI] sd: make error handling more robust (v2)

:47 PM, Salyzyn, Mark [EMAIL PROTECTED] wrote: ACK Signed-off-by: Mark Salyzyn [EMAIL PROTECTED] Mike Snitzer tested this also in conjunction with the aacraid driver and it resolved the error propagation problem he experienced with MD on a 2.6.22.16 kernel, recommend that this also be added

AACRAID driver broken in 2.6.22.x (and beyond?) [WAS: Re: 2.6.22.16 MD raid1 doesn't mark removed disk faulty, MD thread goes UN]

On Jan 22, 2008 12:29 AM, Mike Snitzer [EMAIL PROTECTED] wrote: cc'ing Tanaka-san given his recent raid1 BUG report: http://lkml.org/lkml/2008/1/14/515 On Jan 21, 2008 6:04 PM, Mike Snitzer [EMAIL PROTECTED] wrote: Under 2.6.22.16, I physically pulled a SATA disk (/dev/sdac, connected

<    1   2