Skip to content

L
linux
Switch branch/tag
  1. 08 Mar, 2017 11 commits
    • NeilBrown's avatar
      blk: improve order of bio handling in generic_make_request() · 79bd9959
      NeilBrown authored 
      To avoid recursion on the kernel stack when stacked block devices
are in use, generic_make_request() will, when called recursively,
queue new requests for later handling.  They will be handled when the
make_request_fn for the current bio completes.

If any bios are submitted by a make_request_fn, these will ultimately
be handled seqeuntially.  If the handling of one of those generates
further requests, they will be added to the end of the queue.

This strict first-in-first-out behaviour can lead to deadlocks in
various ways, normally because a request might need to wait for a
previous request to the same device to complete.  This can happen when
they share a mempool, and can happen due to interdependencies
particular to the device.  Both md and dm have examples where this happens.

These deadlocks can be erradicated by more selective ordering of bios.
Specifically by handling them in depth-first order.  That is: when the
handling of one bio generates one or more further bios, they are
handled immediately after the parent, before any siblings of the
parent.  That way, when generic_make_request() calls make_request_fn
for some particular device, we can be certain that all previously
submited requests for that device have been completely handled and are
not waiting for anything in the queue of requests maintained in
generic_make_request().

An easy way to achieve this would be to use a last-in-first-out stack
instead of a queue.  However this will change the order of consecutive
bios submitted by a make_request_fn, which could have unexpected consequences.
Instead we take a slightly more complex approach.
A fresh queue is created for each call to a make_request_fn.  After it completes,
any bios for a different device are placed on the front of the main queue, followed
by any bios for the same device, followed by all bios that were already on
the queue before the make_request_fn was called.
This provides the depth-first approach without reordering bios on the same level.

This, by itself, it not enough to remove all deadlocks.  It just makes
it possible for drivers to take the extra step required themselves.

To avoid deadlocks, drivers must never risk waiting for a request
after submitting one to generic_make_request.  This includes never
allocing from a mempool twice in the one call to a make_request_fn.

A common pattern in drivers is to call bio_split() in a loop, handling
the first part and then looping around to possibly split the next part.
Instead, a driver that finds it needs to split a bio should queue
(with generic_make_request) the second part, handle the first part,
and then return.  The new code in generic_make_request will ensure the
requests to underlying bios are processed first, then the second bio
that was split off.  If it splits again, the same process happens.  In
each case one bio will be completely handled before the next one is attempted.

With this is place, it should be possible to disable the
punt_bios_to_recover() recovery thread for many block devices, and
eventually it may be possible to remove it completely.

Ref: http://www.spinics.net/lists/raid/msg54680.htmlTested-by: default avatarJinpu Wang <jinpu.wang@profitbricks.com>
Inspired-by: default avatarLars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: default avatarNeilBrown <neilb@suse.com>
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      79bd9959
    • Jan Kara's avatar
      Revert "scsi, block: fix duplicate bdi name registration crashes" · c01228db
      Jan Kara authored 
      This reverts commit 0dba1314. It causes
leaking of device numbers for SCSI when SCSI registers multiple gendisks
for one request_queue in succession. It can be easily reproduced using
Omar's script [1] on kernel with CONFIG_DEBUG_TEST_DRIVER_REMOVE.
Furthermore the protection provided by this commit is not needed anymore
as the problem it was fixing got also fixed by commit 165a5e22
"block: Move bdi_unregister() to del_gendisk()".

[1]: http://marc.info/?l=linux-block&m=148554717109098&w=2Signed-off-by: default avatarJan Kara <jack@suse.cz>
Acked-by: default avatarDan Williams <dan.j.williams@intel.com>
Tested-by: default avatarOmar Sandoval <osandov@fb.com>
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      c01228db
    • Jan Kara's avatar
      block: Make del_gendisk() safer for disks without queues · 90f16fdd
      Jan Kara authored 
      Commit 165a5e22 "block: Move bdi_unregister() to del_gendisk()"
added disk->queue dereference to del_gendisk(). Although del_gendisk()
is not supposed to be called without disk->queue valid and
blk_unregister_queue() warns in that case, this change will make it oops
instead. Return to the old more robust behavior of just warning when
del_gendisk() gets called for gendisk with disk->queue being NULL.
Reported-by: default avatarDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: default avatarJan Kara <jack@suse.cz>
Tested-by: default avatarOmar Sandoval <osandov@fb.com>
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      90f16fdd
    • Jan Kara's avatar
      bdi: Fix use-after-free in wb_congested_put() · df23de55
      Jan Kara authored 
      bdi_writeback_congested structures get created for each blkcg and bdi
regardless whether bdi is registered or not. When they are created in
unregistered bdi and the request queue (and thus bdi) is then destroyed
while blkg still holds reference to bdi_writeback_congested structure,
this structure will be referencing freed bdi and last wb_congested_put()
will try to remove the structure from already freed bdi.

With commit 165a5e22 "block: Move bdi_unregister() to
del_gendisk()", SCSI started to destroy bdis without calling
bdi_unregister() first (previously it was calling bdi_unregister() even
for unregistered bdis) and thus the code detaching
bdi_writeback_congested in cgwb_bdi_destroy() was not triggered and we
started hitting this use-after-free bug. It is enough to boot a KVM
instance with virtio-scsi device to trigger this behavior.

Fix the problem by detaching bdi_writeback_congested structures in
bdi_exit() instead of bdi_unregister(). This is also more logical as
they can get attached to bdi regardless whether it ever got registered
or not.

Fixes: 165a5e22Signed-off-by: default avatarJan Kara <jack@suse.cz>
Tested-by: default avatarOmar Sandoval <osandov@fb.com>
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      df23de55
    • Jan Kara's avatar
      block: Allow bdi re-registration · b6f8fec4
      Jan Kara authored 
      SCSI can call device_add_disk() several times for one request queue when
a device in unbound and bound, creating new gendisk each time. This will
lead to bdi being repeatedly registered and unregistered. This was not a
big problem until commit 165a5e22 "block: Move bdi_unregister() to
del_gendisk()" since bdi was only registered repeatedly (bdi_register()
handles repeated calls fine, only we ended up leaking reference to
gendisk due to overwriting bdi->owner) but unregistered only in
blk_cleanup_queue() which didn't get called repeatedly. After
165a5e22 we were doing correct bdi_register() - bdi_unregister()
cycles however bdi_unregister() is not prepared for it. So make sure
bdi_unregister() cleans up bdi in such a way that it is prepared for
a possible following bdi_register() call.

An easy way to provoke this behavior is to enable
CONFIG_DEBUG_TEST_DRIVER_REMOVE and use scsi_debug driver to create a
scsi disk which immediately hangs without this fix.

Fixes: 165a5e22Signed-off-by: default avatarJan Kara <jack@suse.cz>
Tested-by: default avatarOmar Sandoval <osandov@fb.com>
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      b6f8fec4
    • Jon Derrick's avatar
      block/sed: Fix opal user range check and unused variables · b0bfdfc2
      Jon Derrick authored 
      Fixes check that the opal user is within the range, and cleans up unused
method variables.
Signed-off-by: default avatarJon Derrick <jonathan.derrick@intel.com>
Reviewed-by: default avatarScott Bauer <scott.bauer@intel.com>
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      b0bfdfc2
    • Johannes Thumshirn's avatar
      zram: set physical queue limits to avoid array out of bounds accesses · 0bc31538
      Johannes Thumshirn authored 
      zram can handle at most SECTORS_PER_PAGE sectors in a bio's bvec. When using
the NVMe over Fabrics loopback target which potentially sends a huge bulk of
pages attached to the bio's bvec this results in a kernel panic because of
array out of bounds accesses in zram_decompress_page().
Signed-off-by: default avatarJohannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: default avatarHannes Reinecke <hare@suse.com>
Reviewed-by: default avatarSergey Senozhatsky <sergey.senozhatsky@gmail.com>
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      0bc31538
    • Ming Lei's avatar
      blk-mq: free hctx->cpumask in release handler of hctx's kobject · 01388df3
      Ming Lei authored 
      It is obviously that hctx->cpumask is per hctx, and both
share same lifetime, so this patch moves freeing of hctx->cpumask
into release handler of hctx's kobject.
Signed-off-by: default avatarMing Lei <tom.leiming@gmail.com>
Tested-by: default avatarPeter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      01388df3
    • Ming Lei's avatar
      blk-mq: make lifetime consistent between hctx and its kobject · 6c8b232e
      Ming Lei authored 
      This patch removes kobject_put() over hctx in __blk_mq_unregister_dev(),
and trys to keep lifetime consistent between hctx and hctx's kobject.

Now blk_mq_sysfs_register() and blk_mq_sysfs_unregister() become
totally symmetrical, and kobject's refcounter drops to zero just
when the hctx is freed.
Signed-off-by: default avatarMing Lei <tom.leiming@gmail.com>
Tested-by: default avatarPeter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      6c8b232e
    • Ming Lei's avatar
      blk-mq: make lifetime consitent between q/ctx and its kobject · 7ea5fe31
      Ming Lei authored 
      Currently from kobject view, both q->mq_kobj and ctx->kobj can
be released during one cycle of blk_mq_register_dev() and
blk_mq_unregister_dev(). Actually, sw queue's lifetime is
same with its request queue's, which is covered by request_queue->kobj.

So we don't need to call kobject_put() for the two kinds of
kobject in __blk_mq_unregister_dev(), instead we do that
in release handler of request queue.
Signed-off-by: default avatarMing Lei <tom.leiming@gmail.com>
Tested-by: default avatarPeter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      7ea5fe31
    • Ming Lei's avatar
      blk-mq: initialize mq kobjects in blk_mq_init_allocated_queue() · 737f98cf
      Ming Lei authored 
      Both q->mq_kobj and sw queues' kobjects should have been initialized
once, instead of doing that each add_disk context.

Also this patch removes clearing of ctx in blk_mq_init_cpu_queues()
because percpu allocator fills zero to allocated variable.

This patch fixes one issue[1] reported from Omar.

[1] kernel wearning when doing unbind/bind on one scsi-mq device

[   19.347924] kobject (ffff8800791ea0b8): tried to init an initialized object, something is seriously wrong.
[   19.349781] CPU: 1 PID: 84 Comm: kworker/u8:1 Not tainted 4.10.0-rc7-00210-g53f39eeaa263 #34
[   19.350686] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.1-20161122_114906-anatol 04/01/2014
[   19.350920] Workqueue: events_unbound async_run_entry_fn
[   19.350920] Call Trace:
[   19.350920]  dump_stack+0x63/0x83
[   19.350920]  kobject_init+0x77/0x90
[   19.350920]  blk_mq_register_dev+0x40/0x130
[   19.350920]  blk_register_queue+0xb6/0x190
[   19.350920]  device_add_disk+0x1ec/0x4b0
[   19.350920]  sd_probe_async+0x10d/0x1c0 [sd_mod]
[   19.350920]  async_run_entry_fn+0x48/0x150
[   19.350920]  process_one_work+0x1d0/0x480
[   19.350920]  worker_thread+0x48/0x4e0
[   19.350920]  kthread+0x101/0x140
[   19.350920]  ? process_one_work+0x480/0x480
[   19.350920]  ? kthread_create_on_node+0x60/0x60
[   19.350920]  ret_from_fork+0x2c/0x40

Cc: Omar Sandoval <osandov@osandov.com>
Signed-off-by: default avatarMing Lei <tom.leiming@gmail.com>
Tested-by: default avatarPeter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      737f98cf
  3. 07 Mar, 2017 29 commits
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7