[ Upstream commit 4c7ef92f6d ]
In __blk_mq_update_nr_hw_queues() the return value of
blk_mq_sysfs_register_hctxs() is not checked. If sysfs creation for hctx
fails, later changing the number of hw_queues or removing disk will
trigger the following warning:
kernfs: can not remove 'nr_tags', no directory
WARNING: CPU: 2 PID: 637 at fs/kernfs/dir.c:1707 kernfs_remove_by_name_ns+0x13f/0x160
Call Trace:
remove_files.isra.1+0x38/0xb0
sysfs_remove_group+0x4d/0x100
sysfs_remove_groups+0x31/0x60
__kobject_del+0x23/0xf0
kobject_del+0x17/0x40
blk_mq_unregister_hctx+0x5d/0x80
blk_mq_sysfs_unregister_hctxs+0x94/0xd0
blk_mq_update_nr_hw_queues+0x124/0x760
nullb_update_nr_hw_queues+0x71/0xf0 [null_blk]
nullb_device_submit_queues_store+0x92/0x120 [null_blk]
kobjct_del() was called unconditionally even if sysfs creation failed.
Fix it by checkig the kobject creation statusbefore deleting it.
Fixes: 477e19dedc ("blk-mq: adjust debugfs and sysfs register when updating nr_hw_queues")
Signed-off-by: Li Nan <linan122@huawei.com>
Reviewed-by: Yu Kuai <yukuai3@huawei.com>
Link: https://lore.kernel.org/r/20250826084854.1030545-1-linan666@huaweicloud.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 80e648042e upstream.
Fix several issues in partition probing:
- The bailout for a bad partoffset must use put_dev_sector(), since the
preceding read_part_sector() succeeded.
- If the partition table claims a silly sector size like 0xfff bytes
(which results in partition table entries straddling sector boundaries),
bail out instead of accessing out-of-bounds memory.
- We must not assume that the partition table contains proper NUL
termination - use strnlen() and strncmp() instead of strlen() and
strcmp().
Cc: stable@vger.kernel.org
Signed-off-by: Jann Horn <jannh@google.com>
Link: https://lore.kernel.org/r/20250214-partition-mac-v1-1-c1c626dffbd5@google.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 57e420c84f ]
After a recent change to clamp() and its variants [1] that increases the
coverage of the check that high is greater than low because it can be
done through inlining, certain build configurations (such as s390
defconfig) fail to build with clang with:
block/blk-iocost.c:1101:11: error: call to '__compiletime_assert_557' declared with 'error' attribute: clamp() low limit 1 greater than high limit active
1101 | inuse = clamp_t(u32, inuse, 1, active);
| ^
include/linux/minmax.h:218:36: note: expanded from macro 'clamp_t'
218 | #define clamp_t(type, val, lo, hi) __careful_clamp(type, val, lo, hi)
| ^
include/linux/minmax.h:195:2: note: expanded from macro '__careful_clamp'
195 | __clamp_once(type, val, lo, hi, __UNIQUE_ID(v_), __UNIQUE_ID(l_), __UNIQUE_ID(h_))
| ^
include/linux/minmax.h:188:2: note: expanded from macro '__clamp_once'
188 | BUILD_BUG_ON_MSG(statically_true(ulo > uhi), \
| ^
__propagate_weights() is called with an active value of zero in
ioc_check_iocgs(), which results in the high value being less than the
low value, which is undefined because the value returned depends on the
order of the comparisons.
The purpose of this expression is to ensure inuse is not more than
active and at least 1. This could be written more simply with a ternary
expression that uses min(inuse, active) as the condition so that the
value of that condition can be used if it is not zero and one if it is.
Do this conversion to resolve the error and add a comment to deter
people from turning this back into clamp().
Fixes: 7caa47151a ("blkcg: implement blk-iocost")
Link: https://lore.kernel.org/r/34d53778977747f19cce2abb287bb3e6@AcuMS.aculab.com/ [1]
Suggested-by: David Laight <david.laight@aculab.com>
Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
Closes: https://lore.kernel.org/llvm/CA+G9fYsD7mw13wredcZn0L-KBA3yeoVSTuxnss-AEWMN3ha0cA@mail.gmail.com/
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202412120322.3GfVe3vF-lkp@intel.com/
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit e9f4eee9a0 ]
When the weight of an active iocg is updated, weight_updated() is called
which in turn calls __propagate_weights() to update the active and inuse
weights so that the effective hierarchical weights are update accordingly.
The current implementation is incorrect for inner active nodes. For an
active leaf iocg, inuse can be any value between 1 and active and the
difference represents how much the iocg is donating. When weight is updated,
as long as inuse is clamped between 1 and the new weight, we're alright and
this is what __propagate_weights() currently implements.
However, that's not how an active inner node's inuse is set. An inner node's
inuse is solely determined by the ratio between the sums of inuse's and
active's of its children - ie. they're results of propagating the leaves'
active and inuse weights upwards. __propagate_weights() incorrectly applies
the same clamping as for a leaf when an active inner node's weight is
updated. Consider a hierarchy which looks like the following with saturating
workloads in AA and BB.
R
/ \
A B
| |
AA BB
1. For both A and B, active=100, inuse=100, hwa=0.5, hwi=0.5.
2. echo 200 > A/io.weight
3. __propagate_weights() update A's active to 200 and leave inuse at 100 as
it's already between 1 and the new active, making A:active=200,
A:inuse=100. As R's active_sum is updated along with A's active,
A:hwa=2/3, B:hwa=1/3. However, because the inuses didn't change, the
hwi's remain unchanged at 0.5.
4. The weight of A is now twice that of B but AA and BB still have the same
hwi of 0.5 and thus are doing the same amount of IOs.
Fix it by making __propgate_weights() always calculate the inuse of an
active inner iocg based on the ratio of child_inuse_sum to child_active_sum.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Dan Schatzberg <dschatzberg@fb.com>
Fixes: 7caa47151a ("blkcg: implement blk-iocost")
Cc: stable@vger.kernel.org # v5.4+
Link: https://lore.kernel.org/r/YJsxnLZV1MnBcqjj@slm.duckdns.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Stable-dep-of: 57e420c84f ("blk-iocost: Avoid using clamp() on inuse in __propagate_weights()")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit db84a72af6 ]
__propagate_weights() currently expects the callers to clamp inuse within
[1, active], which is needlessly fragile. The inuse adjustment logic is
going to be revamped, in preparation, let's make __propagate_weights() clamp
inuse on entry.
Also, make it avoid weight updates altogether if neither active or inuse is
changed.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Stable-dep-of: 57e420c84f ("blk-iocost: Avoid using clamp() on inuse in __propagate_weights()")
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 96a9fe64bf upstream.
Supposing first scenario with a virtio_blk driver.
CPU0 CPU1
blk_mq_try_issue_directly()
__blk_mq_issue_directly()
q->mq_ops->queue_rq()
virtio_queue_rq()
blk_mq_stop_hw_queue()
virtblk_done()
blk_mq_request_bypass_insert() 1) store
blk_mq_start_stopped_hw_queue()
clear_bit(BLK_MQ_S_STOPPED) 3) store
blk_mq_run_hw_queue()
if (!blk_mq_hctx_has_pending()) 4) load
return
blk_mq_sched_dispatch_requests()
blk_mq_run_hw_queue()
if (!blk_mq_hctx_has_pending())
return
blk_mq_sched_dispatch_requests()
if (blk_mq_hctx_stopped()) 2) load
return
__blk_mq_sched_dispatch_requests()
Supposing another scenario.
CPU0 CPU1
blk_mq_requeue_work()
blk_mq_insert_request() 1) store
virtblk_done()
blk_mq_start_stopped_hw_queue()
blk_mq_run_hw_queues() clear_bit(BLK_MQ_S_STOPPED) 3) store
blk_mq_run_hw_queue()
if (!blk_mq_hctx_has_pending()) 4) load
return
blk_mq_sched_dispatch_requests()
if (blk_mq_hctx_stopped()) 2) load
continue
blk_mq_run_hw_queue()
Both scenarios are similar, the full memory barrier should be inserted
between 1) and 2), as well as between 3) and 4) to make sure that either
CPU0 sees BLK_MQ_S_STOPPED is cleared or CPU1 sees dispatch list.
Otherwise, either CPU will not rerun the hardware queue causing
starvation of the request.
The easy way to fix it is to add the essential full memory barrier into
helper of blk_mq_hctx_stopped(). In order to not affect the fast path
(hardware queue is not stopped most of the time), we only insert the
barrier into the slow path. Actually, only slow path needs to care about
missing of dispatching the request to the low-level device driver.
Fixes: 320ae51fee ("blk-mq: new multi-queue block IO queueing mechanism")
Cc: stable@vger.kernel.org
Cc: Muchun Song <muchun.song@linux.dev>
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20241014092934.53630-4-songmuchun@bytedance.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit e972b08b91 upstream.
We're seeing crashes from rq_qos_wake_function that look like this:
BUG: unable to handle page fault for address: ffffafe180a40084
#PF: supervisor write access in kernel mode
#PF: error_code(0x0002) - not-present page
PGD 100000067 P4D 100000067 PUD 10027c067 PMD 10115d067 PTE 0
Oops: Oops: 0002 [#1] PREEMPT SMP PTI
CPU: 17 UID: 0 PID: 0 Comm: swapper/17 Not tainted 6.12.0-rc3-00013-geca631b8fe80 #11
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014
RIP: 0010:_raw_spin_lock_irqsave+0x1d/0x40
Code: 90 90 90 90 90 90 90 90 90 90 90 90 90 f3 0f 1e fa 0f 1f 44 00 00 41 54 9c 41 5c fa 65 ff 05 62 97 30 4c 31 c0 ba 01 00 00 00 <f0> 0f b1 17 75 0a 4c 89 e0 41 5c c3 cc cc cc cc 89 c6 e8 2c 0b 00
RSP: 0018:ffffafe180580ca0 EFLAGS: 00010046
RAX: 0000000000000000 RBX: ffffafe180a3f7a8 RCX: 0000000000000011
RDX: 0000000000000001 RSI: 0000000000000003 RDI: ffffafe180a40084
RBP: 0000000000000000 R08: 00000000001e7240 R09: 0000000000000011
R10: 0000000000000028 R11: 0000000000000888 R12: 0000000000000002
R13: ffffafe180a40084 R14: 0000000000000000 R15: 0000000000000003
FS: 0000000000000000(0000) GS:ffff9aaf1f280000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: ffffafe180a40084 CR3: 000000010e428002 CR4: 0000000000770ef0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
PKRU: 55555554
Call Trace:
<IRQ>
try_to_wake_up+0x5a/0x6a0
rq_qos_wake_function+0x71/0x80
__wake_up_common+0x75/0xa0
__wake_up+0x36/0x60
scale_up.part.0+0x50/0x110
wb_timer_fn+0x227/0x450
...
So rq_qos_wake_function() calls wake_up_process(data->task), which calls
try_to_wake_up(), which faults in raw_spin_lock_irqsave(&p->pi_lock).
p comes from data->task, and data comes from the waitqueue entry, which
is stored on the waiter's stack in rq_qos_wait(). Analyzing the core
dump with drgn, I found that the waiter had already woken up and moved
on to a completely unrelated code path, clobbering what was previously
data->task. Meanwhile, the waker was passing the clobbered garbage in
data->task to wake_up_process(), leading to the crash.
What's happening is that in between rq_qos_wake_function() deleting the
waitqueue entry and calling wake_up_process(), rq_qos_wait() is finding
that it already got a token and returning. The race looks like this:
rq_qos_wait() rq_qos_wake_function()
==============================================================
prepare_to_wait_exclusive()
data->got_token = true;
list_del_init(&curr->entry);
if (data.got_token)
break;
finish_wait(&rqw->wait, &data.wq);
^- returns immediately because
list_empty_careful(&wq_entry->entry)
is true
... return, go do something else ...
wake_up_process(data->task)
(NO LONGER VALID!)-^
Normally, finish_wait() is supposed to synchronize against the waker.
But, as noted above, it is returning immediately because the waitqueue
entry has already been removed from the waitqueue.
The bug is that rq_qos_wake_function() is accessing the waitqueue entry
AFTER deleting it. Note that autoremove_wake_function() wakes the waiter
and THEN deletes the waitqueue entry, which is the proper order.
Fix it by swapping the order. We also need to use
list_del_init_careful() to match the list_empty_careful() in
finish_wait().
Fixes: 38cfb5a45e ("blk-wbt: improve waking of tasks")
Cc: stable@vger.kernel.org
Signed-off-by: Omar Sandoval <osandov@fb.com>
Acked-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/d3bee2463a67b1ee597211823bf7ad3721c26e41.1729014591.git.osandov@fb.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 42c306ed72 ]
Consider the following scenario:
Process 1 Process 2 Process 3 Process 4
(BIC1) (BIC2) (BIC3) (BIC4)
Λ | | |
\-------------\ \-------------\ \--------------\|
V V V
bfqq1--------->bfqq2---------->bfqq3----------->bfqq4
ref 0 1 2 4
If Process 1 issue a new IO and bfqq2 is found, and then bfq_init_rq()
decide to spilt bfqq2 by bfq_split_bfqq(). Howerver, procress reference
of bfqq2 is 1 and bfq_split_bfqq() just clear the coop flag, which will
break the merge chain.
Expected result: caller will allocate a new bfqq for BIC1
Process 1 Process 2 Process 3 Process 4
(BIC1) (BIC2) (BIC3) (BIC4)
| | |
\-------------\ \--------------\|
V V
bfqq1--------->bfqq2---------->bfqq3----------->bfqq4
ref 0 0 1 3
Since the condition is only used for the last bfqq4 when the previous
bfqq2 and bfqq3 are already splited. Fix the problem by checking if
bfqq is the last one in the merge chain as well.
Fixes: 36eca89483 ("block, bfq: add Early Queue Merge (EQM)")
Signed-off-by: Yu Kuai <yukuai3@huawei.com>
Link: https://lore.kernel.org/r/20240902130329.3787024-4-yukuai1@huaweicloud.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 0e456dba86 ]
Consider the following merge chain:
Process 1 Process 2 Process 3 Process 4
(BIC1) (BIC2) (BIC3) (BIC4)
Λ | | |
\--------------\ \-------------\ \-------------\|
V V V
bfqq1--------->bfqq2---------->bfqq3----------->bfqq4
IO from Process 1 will get bfqf2 from BIC1 first, then
bfq_setup_cooperator() will found bfqq2 already merged to bfqq3 and then
handle this IO from bfqq3. However, the merge chain can be much deeper
and bfqq3 can be merged to other bfqq as well.
Fix this problem by iterating to the last bfqq in
bfq_setup_cooperator().
Fixes: 36eca89483 ("block, bfq: add Early Queue Merge (EQM)")
Signed-off-by: Yu Kuai <yukuai3@huawei.com>
Link: https://lore.kernel.org/r/20240902130329.3787024-3-yukuai1@huaweicloud.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 5266caaf56 ]
In blk_mq_mark_tag_wait(), __add_wait_queue() may be re-ordered
with the following blk_mq_get_driver_tag() in case of getting driver
tag failure.
Then in __sbitmap_queue_wake_up(), waitqueue_active() may not observe
the added waiter in blk_mq_mark_tag_wait() and wake up nothing, meantime
blk_mq_mark_tag_wait() can't get driver tag successfully.
This issue can be reproduced by running the following test in loop, and
fio hang can be observed in < 30min when running it on my test VM
in laptop.
modprobe -r scsi_debug
modprobe scsi_debug delay=0 dev_size_mb=4096 max_queue=1 host_max_queue=1 submit_queues=4
dev=`ls -d /sys/bus/pseudo/drivers/scsi_debug/adapter*/host*/target*/*/block/* | head -1 | xargs basename`
fio --filename=/dev/"$dev" --direct=1 --rw=randrw --bs=4k --iodepth=1 \
--runtime=100 --numjobs=40 --time_based --name=test \
--ioengine=libaio
Fix the issue by adding one explicit barrier in blk_mq_mark_tag_wait(), which
is just fine in case of running out of tag.
Cc: Jan Kara <jack@suse.cz>
Cc: Kemeng Shi <shikemeng@huaweicloud.com>
Reported-by: Changhui Zhong <czhong@redhat.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20240112122626.4181044-1-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 1b151e2435 upstream.
The special casing was originally added in pre-git history; reproducing
the commit log here:
> commit a318a92567d77
> Author: Andrew Morton <akpm@osdl.org>
> Date: Sun Sep 21 01:42:22 2003 -0700
>
> [PATCH] Speed up direct-io hugetlbpage handling
>
> This patch short-circuits all the direct-io page dirtying logic for
> higher-order pages. Without this, we pointlessly bounce BIOs up to
> keventd all the time.
In the last twenty years, compound pages have become used for more than
just hugetlb. Rewrite these functions to operate on folios instead
of pages and remove the special case for hugetlbfs; I don't think
it's needed any more (and if it is, we can put it back in as a call
to folio_test_hugetlb()).
This was found by inspection; as far as I can tell, this bug can lead
to pages used as the destination of a direct I/O read not being marked
as dirty. If those pages are then reclaimed by the MM without being
dirtied for some other reason, they won't be written out. Then when
they're faulted back in, they will not contain the data they should.
It'll take a pretty unusual setup to produce this problem with several
races all going the wrong way.
This problem predates the folio work; it could for example have been
triggered by mmaping a THP in tmpfs and using that as the target of an
O_DIRECT read.
Fixes: 800d8c63b2 ("shmem: add huge pages support")
Cc: <stable@vger.kernel.org>
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit b6f3f28f60 upstream.
The Amiga partition parser module uses signed int for partition sector
address and count, which will overflow for disks larger than 1 TB.
Use u64 as type for sector address and size to allow using disks up to
2 TB without LBD support, and disks larger than 2 TB with LBD. The RBD
format allows to specify disk sizes up to 2^128 bytes (though native
OS limitations reduce this somewhat, to max 2^68 bytes), so check for
u64 overflow carefully to protect against overflowing sector_t.
Bail out if sector addresses overflow 32 bits on kernels without LBD
support.
This bug was reported originally in 2012, and the fix was created by
the RDB author, Joanne Dow <jdow@earthlink.net>. A patch had been
discussed and reviewed on linux-m68k at that time but never officially
submitted (now resubmitted as patch 1 in this series).
This patch adds additional error checking and warning messages.
Reported-by: Martin Steigerwald <Martin@lichtvoll.de>
Closes: https://bugzilla.kernel.org/show_bug.cgi?id=43511
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Message-ID: <201206192146.09327.Martin@lichtvoll.de>
Cc: <stable@vger.kernel.org> # 5.2
Signed-off-by: Michael Schmitz <schmitzmic@gmail.com>
Reviewed-by: Geert Uytterhoeven <geert@linux-m68k.org>
Reviewed-by: Christoph Hellwig <hch@infradead.org>
Link: https://lore.kernel.org/r/20230620201725.7020-4-schmitzmic@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit ff1cc97b1f upstream.
Since gcc13, each member of an enum has the same type as the enum [1]. And
that is inherited from its members. Provided:
VTIME_PER_SEC_SHIFT = 37,
VTIME_PER_SEC = 1LLU << VTIME_PER_SEC_SHIFT,
...
AUTOP_CYCLE_NSEC = 10LLU * NSEC_PER_SEC,
the named type is unsigned long.
This generates warnings with gcc-13:
block/blk-iocost.c: In function 'ioc_weight_prfill':
block/blk-iocost.c:3037:37: error: format '%u' expects argument of type 'unsigned int', but argument 4 has type 'long unsigned int'
block/blk-iocost.c: In function 'ioc_weight_show':
block/blk-iocost.c:3047:34: error: format '%u' expects argument of type 'unsigned int', but argument 3 has type 'long unsigned int'
So split the anonymous enum with large values to a separate enum, so
that they don't affect other members.
[1] https://gcc.gnu.org/bugzilla/show_bug.cgi?id=36113
Cc: Martin Liska <mliska@suse.cz>
Cc: Tejun Heo <tj@kernel.org>
Cc: Josef Bacik <josef@toxicpanda.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: cgroups@vger.kernel.org
Cc: linux-block@vger.kernel.org
Signed-off-by: Jiri Slaby (SUSE) <jirislaby@kernel.org>
Link: https://lore.kernel.org/r/20221213120826.17446-1-jirislaby@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 5f2779dfa7 upstream.
The behavior of 'enum' types has changed in gcc-13, so now the
UNBUSY_THR_PCT constant is interpreted as a 64-bit number because
it is defined as part of the same enum definition as some other
constants that do not fit within a 32-bit integer. This in turn
leads to some inefficient code on 32-bit architectures as well
as a link error:
arm-linux-gnueabi/bin/arm-linux-gnueabi-ld: block/blk-iocost.o: in function `ioc_timer_fn':
blk-iocost.c:(.text+0x68e8): undefined reference to `__aeabi_uldivmod'
arm-linux-gnueabi-ld: blk-iocost.c:(.text+0x6908): undefined reference to `__aeabi_uldivmod'
Split the enum definition to keep the 64-bit timing constants in
a separate enum type from those constants that can clearly fit
within a smaller type.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Tejun Heo <tj@kernel.org>
Link: https://lore.kernel.org/r/20230118080706.3303186-1-arnd@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 984af1e66b ]
echo max of u64 to cost.model can cause divide by 0 error.
# echo 8:0 rbps=18446744073709551615 > /sys/fs/cgroup/io.cost.model
divide error: 0000 [#1] PREEMPT SMP
RIP: 0010:calc_lcoefs+0x4c/0xc0
Call Trace:
<TASK>
ioc_refresh_params+0x2b3/0x4f0
ioc_cost_model_write+0x3cb/0x4c0
? _copy_from_iter+0x6d/0x6c0
? kernfs_fop_write_iter+0xfc/0x270
cgroup_file_write+0xa0/0x200
kernfs_fop_write_iter+0x17d/0x270
vfs_write+0x414/0x620
ksys_write+0x73/0x160
__x64_sys_write+0x1e/0x30
do_syscall_64+0x35/0x80
entry_SYSCALL_64_after_hwframe+0x63/0xcd
calc_lcoefs() uses the input value of cost.model in DIV_ROUND_UP_ULL,
overflow would happen if bps plus IOC_PAGE_SIZE is greater than
ULLONG_MAX, it can cause divide by 0 error.
Fix the problem by setting basecost
Signed-off-by: Li Nan <linan122@huawei.com>
Signed-off-by: Yu Kuai <yukuai3@huawei.com>
Acked-by: Tejun Heo <tj@kernel.org>
Link: https://lore.kernel.org/r/20230117070806.3857142-5-yukuai1@huaweicloud.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 01542f651a ]
Commit 88022d7201 ("blk-mq: don't handle failure in .get_budget")
remove BLK_STS_RESOURCE return value and we only check if we can get
the budget from .get_budget() now.
Correct stale comment that ".get_budget() returns BLK_STS_NO_RESOURCE"
to ".get_budget() fails to get the budget".
Fixes: 88022d7201 ("blk-mq: don't handle failure in .get_budget")
Signed-off-by: Kemeng Shi <shikemeng@huaweicloud.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 98b99e9412 ]
For shared queues case, we will only wait on bitmap_tags if we fail to get
driver tag. However, rq could be from breserved_tags, then two problems
will occur:
1. io hung if no tag is currently allocated from bitmap_tags.
2. unnecessary wakeup when tag is freed to bitmap_tags while no tag is
freed to breserved_tags.
Wait on the bitmap which rq from to fix this.
Fixes: f906a6a0f4 ("blk-mq: improve tag waiting setup for non-shared tags")
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Kemeng Shi <shikemeng@huaweicloud.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit c31e76bcc3 ]
Commit 97889f9ac2 ("blk-mq: remove synchronize_rcu() from
blk_mq_del_queue_tag_set()") remove handle of TAG_SHARED in restart,
then shared_hctx_restart counted for how many hardware queues are marked
for restart is removed too.
Remove the stale comment that we still count hardware queues need restart.
Fixes: 97889f9ac2 ("blk-mq: remove synchronize_rcu() from blk_mq_del_queue_tag_set()")
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Kemeng Shi <shikemeng@huaweicloud.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 28d65729b0 ]
Flushes bypass the I/O scheduler and get added to hctx->dispatch
in blk_mq_sched_bypass_insert. This can happen while a kworker is running
hctx->run_work work item and is past the point in
blk_mq_sched_dispatch_requests where hctx->dispatch is checked.
The blk_mq_do_dispatch_sched call is not guaranteed to end in bounded time,
because the I/O scheduler can feed an arbitrary number of commands.
Since we have only one hctx->run_work, the commands waiting in
hctx->dispatch will wait an arbitrary length of time for run_work to be
rerun.
A similar phenomenon exists with dispatches from the software queue.
The solution is to poll hctx->dispatch in blk_mq_do_dispatch_sched and
blk_mq_do_dispatch_ctx and return from the run_work handler and let it
rerun.
Signed-off-by: Salman Qazi <sqazi@google.com>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Stable-dep-of: c31e76bcc3 ("blk-mq: remove stale comment for blk_mq_sched_mark_restart_hctx")
Signed-off-by: Sasha Levin <sashal@kernel.org>
v5.11 changes the blkdev lookup mechanism completely since commit
22ae8ce8b8 ("block: simplify bdev/disk lookup in blkdev_get"),
and small part of the change is to unhash part bdev inode when
deleting partition. Turns out this kind of change does fix one
nasty issue in case of BLOCK_EXT_MAJOR:
1) when one partition is deleted & closed, disk_put_part() is always
called before bdput(bdev), see blkdev_put(); so the part's devt can
be freed & re-used before the inode is dropped
2) then new partition with same devt can be created just before the
inode in 1) is dropped, then the old inode/bdev structurein 1) is
re-used for this new partition, this way causes use-after-free and
kernel panic.
It isn't possible to backport the whole big patchset of "merge struct
block_device and struct hd_struct v4" for addressing this issue.
https://lore.kernel.org/linux-block/20201128161510.347752-1-hch@lst.de/
So fixes it by unhashing part bdev in delete_partition(), and this way
is actually aligned with v5.11+'s behavior.
Backported from the following 5.10.y commit:
5f2f775605 ("block: unhash blkdev part inode when the part is deleted")
Reported-by: Shiwei Cui <cuishw@inspur.com>
Tested-by: Shiwei Cui <cuishw@inspur.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Jan Kara <jack@suse.cz>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit f829230dd5 ]
In accordance with [1] the DMA-able memory buffers must be
cacheline-aligned otherwise the cache writing-back and invalidation
performed during the mapping may cause the adjacent data being lost. It's
specifically required for the DMA-noncoherent platforms [2]. Seeing the
opal_dev.{cmd,resp} buffers are implicitly used for DMAs in the NVME and
SCSI/SD drivers in framework of the nvme_sec_submit() and sd_sec_submit()
methods respectively they must be cacheline-aligned to prevent the denoted
problem. One of the option to guarantee that is to kmalloc the buffers
[2]. Let's explicitly allocate them then instead of embedding into the
opal_dev structure instance.
Note this fix was inspired by the commit c94b7f9bab ("nvme-hwmon:
kmalloc the NVME SMART log buffer").
[1] Documentation/core-api/dma-api.rst
[2] Documentation/core-api/dma-api-howto.rst
Fixes: 455a7b238c ("block: Add Sed-opal library")
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20221107203944.31686-1-Sergey.Semin@baikalelectronics.ru
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 22b106e535 upstream.
Commit d92c370a16 ("block: really clone the block cgroup in
bio_clone_blkg_association") changed bio_clone_blkg_association() to
just clone bio->bi_blkg reference from source to destination bio. This
is however wrong if the source and destination bios are against
different block devices because struct blkcg_gq is different for each
bdev-blkcg pair. This will result in IOs being accounted (and throttled
as a result) multiple times against the same device (src bdev) while
throttling of the other device (dst bdev) is ignored. In case of BFQ the
inconsistency can even result in crashes in bfq_bic_update_cgroup().
Fix the problem by looking up correct blkcg_gq for the cloned bio.
Reported-by: Logan Gunthorpe <logang@deltatee.com>
Reported-and-tested-by: Donald Buczek <buczek@molgen.mpg.de>
Fixes: d92c370a16 ("block: really clone the block cgroup in bio_clone_blkg_association")
CC: stable@vger.kernel.org
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20220602081242.7731-1-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 075a53b78b upstream.
Bios queued into BFQ IO scheduler can be associated with a cgroup that
was already offlined. This may then cause insertion of this bfq_group
into a service tree. But this bfq_group will get freed as soon as last
bio associated with it is completed leading to use after free issues for
service tree users. Fix the problem by making sure we always operate on
online bfq_group. If the bfq_group associated with the bio is not
online, we pick the first online parent.
CC: stable@vger.kernel.org
Fixes: e21b7a0b98 ("block, bfq: add full hierarchical scheduling and cgroups support")
Tested-by: "yukuai (C)" <yukuai3@huawei.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20220401102752.8599-9-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 4e54a2493e upstream.
BFQ usage of __bio_blkcg() is a relict from the past. Furthermore if bio
would not be associated with any blkcg, the usage of __bio_blkcg() in
BFQ is prone to races with the task being migrated between cgroups as
__bio_blkcg() calls at different places could return different blkcgs.
Convert BFQ to the new situation where bio->bi_blkg is initialized in
bio_set_dev() and thus practically always valid. This allows us to save
blkcg_gq lookup and noticeably simplify the code.
CC: stable@vger.kernel.org
Fixes: 0fe061b9f0 ("blkcg: fix ref count issue with bio_blkcg() using task_css")
Tested-by: "yukuai (C)" <yukuai3@huawei.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20220401102752.8599-8-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 5f550ede5e upstream.
We call bfq_init_rq() from request merging functions where requests we
get should have already gone through bfq_init_rq() during insert and
anyway we want to do anything only if the request is already tracked by
BFQ. So replace calls to bfq_init_rq() with RQ_BFQQ() instead to simply
skip requests untracked by BFQ. We move bfq_init_rq() call in
bfq_insert_request() a bit earlier to cover request merging and thus
can transfer FIFO position in case of a merge.
CC: stable@vger.kernel.org
Tested-by: "yukuai (C)" <yukuai3@huawei.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20220401102752.8599-6-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit c1cee4ab36 upstream.
It can happen that the parent of a bfqq changes between the moment we
decide two queues are worth to merge (and set bic->stable_merge_bfqq)
and the moment bfq_setup_merge() is called. This can happen e.g. because
the process submitted IO for a different cgroup and thus bfqq got
reparented. It can even happen that the bfqq we are merging with has
parent cgroup that is already offline and going to be destroyed in which
case the merge can lead to use-after-free issues such as:
BUG: KASAN: use-after-free in __bfq_deactivate_entity+0x9cb/0xa50
Read of size 8 at addr ffff88800693c0c0 by task runc:[2:INIT]/10544
CPU: 0 PID: 10544 Comm: runc:[2:INIT] Tainted: G E 5.15.2-0.g5fb85fd-default #1 openSUSE Tumbleweed (unreleased) f1f3b891c72369aebecd2e43e4641a6358867c70
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.14.0-0-g155821a-rebuilt.opensuse.org 04/01/2014
Call Trace:
<IRQ>
dump_stack_lvl+0x46/0x5a
print_address_description.constprop.0+0x1f/0x140
? __bfq_deactivate_entity+0x9cb/0xa50
kasan_report.cold+0x7f/0x11b
? __bfq_deactivate_entity+0x9cb/0xa50
__bfq_deactivate_entity+0x9cb/0xa50
? update_curr+0x32f/0x5d0
bfq_deactivate_entity+0xa0/0x1d0
bfq_del_bfqq_busy+0x28a/0x420
? resched_curr+0x116/0x1d0
? bfq_requeue_bfqq+0x70/0x70
? check_preempt_wakeup+0x52b/0xbc0
__bfq_bfqq_expire+0x1a2/0x270
bfq_bfqq_expire+0xd16/0x2160
? try_to_wake_up+0x4ee/0x1260
? bfq_end_wr_async_queues+0xe0/0xe0
? _raw_write_unlock_bh+0x60/0x60
? _raw_spin_lock_irq+0x81/0xe0
bfq_idle_slice_timer+0x109/0x280
? bfq_dispatch_request+0x4870/0x4870
__hrtimer_run_queues+0x37d/0x700
? enqueue_hrtimer+0x1b0/0x1b0
? kvm_clock_get_cycles+0xd/0x10
? ktime_get_update_offsets_now+0x6f/0x280
hrtimer_interrupt+0x2c8/0x740
Fix the problem by checking that the parent of the two bfqqs we are
merging in bfq_setup_merge() is the same.
Link: https://lore.kernel.org/linux-block/20211125172809.GC19572@quack2.suse.cz/
CC: stable@vger.kernel.org
Fixes: 430a67f9d6 ("block, bfq: merge bursts of newly-created queues")
Tested-by: "yukuai (C)" <yukuai3@huawei.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20220401102752.8599-2-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 8a177a36da upstream.
iolatency needs to track the number of inflight IOs per cgroup. As this
tracking can be expensive, it is disabled when no cgroup has iolatency
configured for the device. To ensure that the inflight counters stay
balanced, iolatency_set_limit() freezes the request_queue while manipulating
the enabled counter, which ensures that no IO is in flight and thus all
counters are zero.
Unfortunately, iolatency_set_limit() isn't the only place where the enabled
counter is manipulated. iolatency_pd_offline() can also dec the counter and
trigger disabling. As this disabling happens without freezing the q, this
can easily happen while some IOs are in flight and thus leak the counts.
This can be easily demonstrated by turning on iolatency on an one empty
cgroup while IOs are in flight in other cgroups and then removing the
cgroup. Note that iolatency shouldn't have been enabled elsewhere in the
system to ensure that removing the cgroup disables iolatency for the whole
device.
The following keeps flipping on and off iolatency on sda:
echo +io > /sys/fs/cgroup/cgroup.subtree_control
while true; do
mkdir -p /sys/fs/cgroup/test
echo '8:0 target=100000' > /sys/fs/cgroup/test/io.latency
sleep 1
rmdir /sys/fs/cgroup/test
sleep 1
done
and there's concurrent fio generating direct rand reads:
fio --name test --filename=/dev/sda --direct=1 --rw=randread \
--runtime=600 --time_based --iodepth=256 --numjobs=4 --bs=4k
while monitoring with the following drgn script:
while True:
for css in css_for_each_descendant_pre(prog['blkcg_root'].css.address_of_()):
for pos in hlist_for_each(container_of(css, 'struct blkcg', 'css').blkg_list):
blkg = container_of(pos, 'struct blkcg_gq', 'blkcg_node')
pd = blkg.pd[prog['blkcg_policy_iolatency'].plid]
if pd.value_() == 0:
continue
iolat = container_of(pd, 'struct iolatency_grp', 'pd')
inflight = iolat.rq_wait.inflight.counter.value_()
if inflight:
print(f'inflight={inflight} {disk_name(blkg.q.disk).decode("utf-8")} '
f'{cgroup_path(css.cgroup).decode("utf-8")}')
time.sleep(1)
The monitoring output looks like the following:
inflight=1 sda /user.slice
inflight=1 sda /user.slice
...
inflight=14 sda /user.slice
inflight=13 sda /user.slice
inflight=17 sda /user.slice
inflight=15 sda /user.slice
inflight=18 sda /user.slice
inflight=17 sda /user.slice
inflight=20 sda /user.slice
inflight=19 sda /user.slice <- fio stopped, inflight stuck at 19
inflight=19 sda /user.slice
inflight=19 sda /user.slice
If a cgroup with stuck inflight ends up getting throttled, the throttled IOs
will never get issued as there's no completion event to wake it up leading
to an indefinite hang.
This patch fixes the bug by unifying enable handling into a work item which
is automatically kicked off from iolatency_set_min_lat_nsec() which is
called from both iolatency_set_limit() and iolatency_pd_offline() paths.
Punting to a work item is necessary as iolatency_pd_offline() is called
under spinlocks while freezing a request_queue requires a sleepable context.
This also simplifies the code reducing LOC sans the comments and avoids the
unnecessary freezes which were happening whenever a cgroup's latency target
is newly set or cleared.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Josef Bacik <josef@toxicpanda.com>
Cc: Liu Bo <bo.liu@linux.alibaba.com>
Fixes: 8c772a9bfc ("blk-iolatency: fix IO hang due to negative inflight counter")
Cc: stable@vger.kernel.org # v5.0+
Link: https://lore.kernel.org/r/Yn9ScX6Nx2qIiQQi@slm.duckdns.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 09f8718680 upstream.
Track whether bfq_group is still online. We cannot rely on
blkcg_gq->online because that gets cleared only after all policies are
offlined and we need something that gets updated already under
bfqd->lock when we are cleaning up our bfq_group to be able to guarantee
that when we see online bfq_group, it will stay online while we are
holding bfqd->lock lock.
CC: stable@vger.kernel.org
Tested-by: "yukuai (C)" <yukuai3@huawei.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20220401102752.8599-7-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit ea591cd4eb upstream.
When the process is migrated to a different cgroup (or in case of
writeback just starts submitting bios associated with a different
cgroup) bfq_merge_bio() can operate with stale cgroup information in
bic. Thus the bio can be merged to a request from a different cgroup or
it can result in merging of bfqqs for different cgroups or bfqqs of
already dead cgroups and causing possible use-after-free issues. Fix the
problem by updating cgroup information in bfq_merge_bio().
CC: stable@vger.kernel.org
Fixes: e21b7a0b98 ("block, bfq: add full hierarchical scheduling and cgroups support")
Tested-by: "yukuai (C)" <yukuai3@huawei.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20220401102752.8599-4-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 3bc5e683c6 upstream.
When bfqq is shared by multiple processes it can happen that one of the
processes gets moved to a different cgroup (or just starts submitting IO
for different cgroup). In case that happens we need to split the merged
bfqq as otherwise we will have IO for multiple cgroups in one bfqq and
we will just account IO time to wrong entities etc.
Similarly if the bfqq is scheduled to merge with another bfqq but the
merge didn't happen yet, cancel the merge as it need not be valid
anymore.
CC: stable@vger.kernel.org
Fixes: e21b7a0b98 ("block, bfq: add full hierarchical scheduling and cgroups support")
Tested-by: "yukuai (C)" <yukuai3@huawei.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20220401102752.8599-3-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
