<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux-toradex.git/block, branch v3.17.8</title>
<subtitle>Linux kernel for Apalis and Colibri modules</subtitle>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/'/>
<entry>
<title>block: Fix computation of merged request priority</title>
<updated>2014-11-21T17:23:12+00:00</updated>
<author>
<name>Jan Kara</name>
<email>jack@suse.cz</email>
</author>
<published>2014-10-30T19:43:38+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=0d3390c09d73214cf9f44d5b791b10c2b682c7da'/>
<id>0d3390c09d73214cf9f44d5b791b10c2b682c7da</id>
<content type='text'>
commit ece9c72accdc45c3a9484dacb1125ce572647288 upstream.

Priority of a merged request is computed by ioprio_best(). If one of the
requests has undefined priority (IOPRIO_CLASS_NONE) and another request
has priority from IOPRIO_CLASS_BE, the function will return the
undefined priority which is wrong. Fix the function to properly return
priority of a request with the defined priority.

Fixes: d58cdfb89ce0c6bd5f81ae931a984ef298dbda20
Signed-off-by: Jan Kara &lt;jack@suse.cz&gt;
Reviewed-by: Jeff Moyer &lt;jmoyer@redhat.com&gt;
Signed-off-by: Jens Axboe &lt;axboe@fb.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@linuxfoundation.org&gt;

</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
commit ece9c72accdc45c3a9484dacb1125ce572647288 upstream.

Priority of a merged request is computed by ioprio_best(). If one of the
requests has undefined priority (IOPRIO_CLASS_NONE) and another request
has priority from IOPRIO_CLASS_BE, the function will return the
undefined priority which is wrong. Fix the function to properly return
priority of a request with the defined priority.

Fixes: d58cdfb89ce0c6bd5f81ae931a984ef298dbda20
Signed-off-by: Jan Kara &lt;jack@suse.cz&gt;
Reviewed-by: Jeff Moyer &lt;jmoyer@redhat.com&gt;
Signed-off-by: Jens Axboe &lt;axboe@fb.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@linuxfoundation.org&gt;

</pre>
</div>
</content>
</entry>
<entry>
<title>scsi: Fix error handling in SCSI_IOCTL_SEND_COMMAND</title>
<updated>2014-11-14T18:10:36+00:00</updated>
<author>
<name>Jan Kara</name>
<email>jack@suse.cz</email>
</author>
<published>2014-10-23T02:13:39+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=37ca06e5d8cfe85d810c93a58fd429265adf8935'/>
<id>37ca06e5d8cfe85d810c93a58fd429265adf8935</id>
<content type='text'>
commit 84ce0f0e94ac97217398b3b69c21c7a62ebeed05 upstream.

When sg_scsi_ioctl() fails to prepare request to submit in
blk_rq_map_kern() we jump to a label where we just end up copying
(luckily zeroed-out) kernel buffer to userspace instead of reporting
error. Fix the problem by jumping to the right label.

CC: Jens Axboe &lt;axboe@kernel.dk&gt;
CC: linux-scsi@vger.kernel.org
Coverity-id: 1226871
Signed-off-by: Jan Kara &lt;jack@suse.cz&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@linuxfoundation.org&gt;

Fixed up the, now unused, out label.

Signed-off-by: Jens Axboe &lt;axboe@fb.com&gt;

</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
commit 84ce0f0e94ac97217398b3b69c21c7a62ebeed05 upstream.

When sg_scsi_ioctl() fails to prepare request to submit in
blk_rq_map_kern() we jump to a label where we just end up copying
(luckily zeroed-out) kernel buffer to userspace instead of reporting
error. Fix the problem by jumping to the right label.

CC: Jens Axboe &lt;axboe@kernel.dk&gt;
CC: linux-scsi@vger.kernel.org
Coverity-id: 1226871
Signed-off-by: Jan Kara &lt;jack@suse.cz&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@linuxfoundation.org&gt;

Fixed up the, now unused, out label.

Signed-off-by: Jens Axboe &lt;axboe@fb.com&gt;

</pre>
</div>
</content>
</entry>
<entry>
<title>block: fix alignment_offset math that assumes io_min is a power-of-2</title>
<updated>2014-11-14T18:10:26+00:00</updated>
<author>
<name>Mike Snitzer</name>
<email>snitzer@redhat.com</email>
</author>
<published>2014-10-08T22:26:13+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=ab050e8e15aae8d949757a91348f8b1ebd9dfec0'/>
<id>ab050e8e15aae8d949757a91348f8b1ebd9dfec0</id>
<content type='text'>
commit b8839b8c55f3fdd60dc36abcda7e0266aff7985c upstream.

The math in both blk_stack_limits() and queue_limit_alignment_offset()
assume that a block device's io_min (aka minimum_io_size) is always a
power-of-2.  Fix the math such that it works for non-power-of-2 io_min.

This issue (of alignment_offset != 0) became apparent when testing
dm-thinp with a thinp blocksize that matches a RAID6 stripesize of
1280K.  Commit fdfb4c8c1 ("dm thin: set minimum_io_size to pool's data
block size") unlocked the potential for alignment_offset != 0 due to
the dm-thin-pool's io_min possibly being a non-power-of-2.

Signed-off-by: Mike Snitzer &lt;snitzer@redhat.com&gt;
Acked-by: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
Signed-off-by: Jens Axboe &lt;axboe@fb.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@linuxfoundation.org&gt;

</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
commit b8839b8c55f3fdd60dc36abcda7e0266aff7985c upstream.

The math in both blk_stack_limits() and queue_limit_alignment_offset()
assume that a block device's io_min (aka minimum_io_size) is always a
power-of-2.  Fix the math such that it works for non-power-of-2 io_min.

This issue (of alignment_offset != 0) became apparent when testing
dm-thinp with a thinp blocksize that matches a RAID6 stripesize of
1280K.  Commit fdfb4c8c1 ("dm thin: set minimum_io_size to pool's data
block size") unlocked the potential for alignment_offset != 0 due to
the dm-thin-pool's io_min possibly being a non-power-of-2.

Signed-off-by: Mike Snitzer &lt;snitzer@redhat.com&gt;
Acked-by: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
Signed-off-by: Jens Axboe &lt;axboe@fb.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@linuxfoundation.org&gt;

</pre>
</div>
</content>
</entry>
<entry>
<title>blk-mq: fix potential hang if rolling wakeup depth is too high</title>
<updated>2014-11-14T18:10:25+00:00</updated>
<author>
<name>Jens Axboe</name>
<email>axboe@fb.com</email>
</author>
<published>2014-10-07T14:39:20+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=7d515b57a1457c9dc4113bc471e686f2b7e79474'/>
<id>7d515b57a1457c9dc4113bc471e686f2b7e79474</id>
<content type='text'>
commit abab13b5c4fd1fec4f9a61622548012d93dc2831 upstream.

We currently divide the queue depth by 4 as our batch wakeup
count, but we split the wakeups over BT_WAIT_QUEUES number of
wait queues. This defaults to 8. If the product of the resulting
batch wake count and BT_WAIT_QUEUES is higher than the device
queue depth, we can get into a situation where a task goes to
sleep waiting for a request, but never gets woken up.

Reported-by: Bart Van Assche &lt;bvanassche@acm.org&gt;
Fixes: 4bb659b156996
Signed-off-by: Jens Axboe &lt;axboe@fb.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@linuxfoundation.org&gt;

</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
commit abab13b5c4fd1fec4f9a61622548012d93dc2831 upstream.

We currently divide the queue depth by 4 as our batch wakeup
count, but we split the wakeups over BT_WAIT_QUEUES number of
wait queues. This defaults to 8. If the product of the resulting
batch wake count and BT_WAIT_QUEUES is higher than the device
queue depth, we can get into a situation where a task goes to
sleep waiting for a request, but never gets woken up.

Reported-by: Bart Van Assche &lt;bvanassche@acm.org&gt;
Fixes: 4bb659b156996
Signed-off-by: Jens Axboe &lt;axboe@fb.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@linuxfoundation.org&gt;

</pre>
</div>
</content>
</entry>
<entry>
<title>blk-mq, percpu_ref: implement a kludge for SCSI blk-mq stall during probe</title>
<updated>2014-09-24T14:29:36+00:00</updated>
<author>
<name>Tejun Heo</name>
<email>tj@kernel.org</email>
</author>
<published>2014-09-23T19:24:32+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=0a30288da1aec914e158c2d7a3482a85f632750f'/>
<id>0a30288da1aec914e158c2d7a3482a85f632750f</id>
<content type='text'>
blk-mq uses percpu_ref for its usage counter which tracks the number
of in-flight commands and used to synchronously drain the queue on
freeze.  percpu_ref shutdown takes measureable wallclock time as it
involves a sched RCU grace period.  This means that draining a blk-mq
takes measureable wallclock time.  One would think that this shouldn't
matter as queue shutdown should be a rare event which takes place
asynchronously w.r.t. userland.

Unfortunately, SCSI probing involves synchronously setting up and then
tearing down a lot of request_queues back-to-back for non-existent
LUNs.  This means that SCSI probing may take more than ten seconds
when scsi-mq is used.

This will be properly fixed by implementing a mechanism to keep
q-&gt;mq_usage_counter in atomic mode till genhd registration; however,
that involves rather big updates to percpu_ref which is difficult to
apply late in the devel cycle (v3.17-rc6 at the moment).  As a
stop-gap measure till the proper fix can be implemented in the next
cycle, this patch introduces __percpu_ref_kill_expedited() and makes
blk_mq_freeze_queue() use it.  This is heavy-handed but should work
for testing the experimental SCSI blk-mq implementation.

Signed-off-by: Tejun Heo &lt;tj@kernel.org&gt;
Reported-by: Christoph Hellwig &lt;hch@infradead.org&gt;
Link: http://lkml.kernel.org/g/20140919113815.GA10791@lst.de
Fixes: add703fda981 ("blk-mq: use percpu_ref for mq usage count")
Cc: Kent Overstreet &lt;kmo@daterainc.com&gt;
Cc: Jens Axboe &lt;axboe@kernel.dk&gt;
Tested-by: Christoph Hellwig &lt;hch@lst.de&gt;
Signed-off-by: Jens Axboe &lt;axboe@fb.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
blk-mq uses percpu_ref for its usage counter which tracks the number
of in-flight commands and used to synchronously drain the queue on
freeze.  percpu_ref shutdown takes measureable wallclock time as it
involves a sched RCU grace period.  This means that draining a blk-mq
takes measureable wallclock time.  One would think that this shouldn't
matter as queue shutdown should be a rare event which takes place
asynchronously w.r.t. userland.

Unfortunately, SCSI probing involves synchronously setting up and then
tearing down a lot of request_queues back-to-back for non-existent
LUNs.  This means that SCSI probing may take more than ten seconds
when scsi-mq is used.

This will be properly fixed by implementing a mechanism to keep
q-&gt;mq_usage_counter in atomic mode till genhd registration; however,
that involves rather big updates to percpu_ref which is difficult to
apply late in the devel cycle (v3.17-rc6 at the moment).  As a
stop-gap measure till the proper fix can be implemented in the next
cycle, this patch introduces __percpu_ref_kill_expedited() and makes
blk_mq_freeze_queue() use it.  This is heavy-handed but should work
for testing the experimental SCSI blk-mq implementation.

Signed-off-by: Tejun Heo &lt;tj@kernel.org&gt;
Reported-by: Christoph Hellwig &lt;hch@infradead.org&gt;
Link: http://lkml.kernel.org/g/20140919113815.GA10791@lst.de
Fixes: add703fda981 ("blk-mq: use percpu_ref for mq usage count")
Cc: Kent Overstreet &lt;kmo@daterainc.com&gt;
Cc: Jens Axboe &lt;axboe@kernel.dk&gt;
Tested-by: Christoph Hellwig &lt;hch@lst.de&gt;
Signed-off-by: Jens Axboe &lt;axboe@fb.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>genhd: fix leftover might_sleep() in blk_free_devt()</title>
<updated>2014-09-22T20:45:45+00:00</updated>
<author>
<name>Jens Axboe</name>
<email>axboe@fb.com</email>
</author>
<published>2014-09-16T19:38:51+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=46f341ffcfb5d8530f7d1e60f3be06cce6661b62'/>
<id>46f341ffcfb5d8530f7d1e60f3be06cce6661b62</id>
<content type='text'>
Commit 2da78092 changed the locking from a mutex to a spinlock,
so we now longer sleep in this context. But there was a leftover
might_sleep() in there, which now triggers since we do the final
free from an RCU callback. Get rid of it.

Reported-by: Pontus Fuchs &lt;pontus.fuchs@gmail.com&gt;
Signed-off-by: Jens Axboe &lt;axboe@fb.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Commit 2da78092 changed the locking from a mutex to a spinlock,
so we now longer sleep in this context. But there was a leftover
might_sleep() in there, which now triggers since we do the final
free from an RCU callback. Get rid of it.

Reported-by: Pontus Fuchs &lt;pontus.fuchs@gmail.com&gt;
Signed-off-by: Jens Axboe &lt;axboe@fb.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>blk-mq: use blk_mq_start_hw_queues() when running requeue work</title>
<updated>2014-09-22T17:55:56+00:00</updated>
<author>
<name>Jens Axboe</name>
<email>axboe@fb.com</email>
</author>
<published>2014-09-19T19:10:29+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=8b95741569eabc5eb17da71d1d3668cdb0bef86c'/>
<id>8b95741569eabc5eb17da71d1d3668cdb0bef86c</id>
<content type='text'>
When requests are retried due to hw or sw resource shortages,
we often stop the associated hardware queue. So ensure that we
restart the queues when running the requeue work, otherwise the
queue run will be a no-op.

Signed-off-by: Jens Axboe &lt;axboe@fb.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
When requests are retried due to hw or sw resource shortages,
we often stop the associated hardware queue. So ensure that we
restart the queues when running the requeue work, otherwise the
queue run will be a no-op.

Signed-off-by: Jens Axboe &lt;axboe@fb.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>blk-mq: fix potential oops on out-of-memory in __blk_mq_alloc_rq_maps()</title>
<updated>2014-09-22T17:55:23+00:00</updated>
<author>
<name>Jens Axboe</name>
<email>axboe@fb.com</email>
</author>
<published>2014-09-19T14:04:53+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=6b55e1f2d0a5e462e52678278ab749468f1db81c'/>
<id>6b55e1f2d0a5e462e52678278ab749468f1db81c</id>
<content type='text'>
__blk_mq_alloc_rq_maps() can be invoked multiple times, if we scale
back the queue depth if we are low on memory. So don't clear
set-&gt;tags when we fail, this is handled directly in
the parent function, blk_mq_alloc_tag_set().

Reported-by: Robert Elliott  &lt;Elliott@hp.com&gt;
Signed-off-by: Jens Axboe &lt;axboe@fb.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
__blk_mq_alloc_rq_maps() can be invoked multiple times, if we scale
back the queue depth if we are low on memory. So don't clear
set-&gt;tags when we fail, this is handled directly in
the parent function, blk_mq_alloc_tag_set().

Reported-by: Robert Elliott  &lt;Elliott@hp.com&gt;
Signed-off-by: Jens Axboe &lt;axboe@fb.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>blk-mq: avoid infinite recursion with the FUA flag</title>
<updated>2014-09-22T17:55:19+00:00</updated>
<author>
<name>Christoph Hellwig</name>
<email>hch@lst.de</email>
</author>
<published>2014-09-16T21:44:07+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=a57a178a490345c7236b0077b3de005754389ed6'/>
<id>a57a178a490345c7236b0077b3de005754389ed6</id>
<content type='text'>
We should not insert requests into the flush state machine from
blk_mq_insert_request.  All incoming flush requests come through
blk_{m,s}q_make_request and are handled there, while blk_execute_rq_nowait
should only be called for BLOCK_PC requests.  All other callers
deal with requests that already went through the flush statemchine
and shouldn't be reinserted into it.

Reported-by: Robert Elliott  &lt;Elliott@hp.com&gt;
Debugged-by: Ming Lei &lt;ming.lei@canonical.com&gt;
Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;
Signed-off-by: Jens Axboe &lt;axboe@fb.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
We should not insert requests into the flush state machine from
blk_mq_insert_request.  All incoming flush requests come through
blk_{m,s}q_make_request and are handled there, while blk_execute_rq_nowait
should only be called for BLOCK_PC requests.  All other callers
deal with requests that already went through the flush statemchine
and shouldn't be reinserted into it.

Reported-by: Robert Elliott  &lt;Elliott@hp.com&gt;
Debugged-by: Ming Lei &lt;ming.lei@canonical.com&gt;
Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;
Signed-off-by: Jens Axboe &lt;axboe@fb.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>blk-mq: Avoid race condition with uninitialized requests</title>
<updated>2014-09-22T17:55:14+00:00</updated>
<author>
<name>David Hildenbrand</name>
<email>dahi@linux.vnet.ibm.com</email>
</author>
<published>2014-09-18T09:04:31+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=683d0e126232d898a481daa3a4ca032c2b1a9660'/>
<id>683d0e126232d898a481daa3a4ca032c2b1a9660</id>
<content type='text'>
This patch should fix the bug reported in
https://lkml.org/lkml/2014/9/11/249.

We have to initialize at least the atomic_flags and the cmd_flags when
allocating storage for the requests.

Otherwise blk_mq_timeout_check() might dereference uninitialized
pointers when racing with the creation of a request.

Also move the reset of cmd_flags for the initializing code to the point
where a request is freed. So we will never end up with pending flush
request indicators that might trigger dereferences of invalid pointers
in blk_mq_timeout_check().

Cc: stable@vger.kernel.org
Signed-off-by: David Hildenbrand &lt;dahi@linux.vnet.ibm.com&gt;
Reported-by: Paulo De Rezende Pinatti &lt;ppinatti@linux.vnet.ibm.com&gt;
Tested-by: Paulo De Rezende Pinatti &lt;ppinatti@linux.vnet.ibm.com&gt;
Acked-by: Christian Borntraeger &lt;borntraeger@de.ibm.com&gt;
Signed-off-by: Jens Axboe &lt;axboe@fb.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This patch should fix the bug reported in
https://lkml.org/lkml/2014/9/11/249.

We have to initialize at least the atomic_flags and the cmd_flags when
allocating storage for the requests.

Otherwise blk_mq_timeout_check() might dereference uninitialized
pointers when racing with the creation of a request.

Also move the reset of cmd_flags for the initializing code to the point
where a request is freed. So we will never end up with pending flush
request indicators that might trigger dereferences of invalid pointers
in blk_mq_timeout_check().

Cc: stable@vger.kernel.org
Signed-off-by: David Hildenbrand &lt;dahi@linux.vnet.ibm.com&gt;
Reported-by: Paulo De Rezende Pinatti &lt;ppinatti@linux.vnet.ibm.com&gt;
Tested-by: Paulo De Rezende Pinatti &lt;ppinatti@linux.vnet.ibm.com&gt;
Acked-by: Christian Borntraeger &lt;borntraeger@de.ibm.com&gt;
Signed-off-by: Jens Axboe &lt;axboe@fb.com&gt;
</pre>
</div>
</content>
</entry>
</feed>
