Instead just call the CDROM layer functionality directly, and turn the
hot mess in isofs_get_last_session into remotely readable code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Instead just call the CDROM layer functionality directly.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Factor out a version of the CDROMMULTISESSION ioctl handler that can
be called directly from kernel space.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Factor out a version of the CDROMREADTOCENTRY ioctl handler that can
be called directly from kernel space.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Give the cdrom_read_tocentry function and ide_ prefix to not conflict
with the soon to be added generic function.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add a pointer to the CDROM information structure to struct gendisk.
This will allow various removable media file systems to call directly
into the CDROM layer instead of abusing ioctls with kernel pointers.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Wrapping numbers in strings is used by some to work around bit-width issues in
some enviroments. The problem isn't innate to json and the workaround seems to
cause more integration problems than help. Let's drop the string wrapping.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This is to help external tools to decide whether iocost_monitor has all its
requirements met or not based on the exit status of an -i0 run.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
On each IO completion, iocost decides whether the IO met or missed its latency
target. Currently, the targets are fixed numbers per IO type. While this can be
good enough for loose latency targets way higher than typical completion
latencies, the effect of IO size makes it difficult to tighten the latency
target - a target adequate for 4k IOs might be too tight for 512k IOs and
vice-versa.
iocost already has all the necessary information to account for different IO
sizes when testing whether the latency target is met as iocost can calculate the
size vtime cost of a given IO. This patch updates the completion path to
calculate the size vtime cost of the IO, deduct the nsec equivalent from the
observed latency and use the adjusted value to decide whether the target is met.
This makes latency targets independent from IO size and enables determining
adequate latency targets with fixed size fio runs.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Andy Newell <newella@fb.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The use_delay mechanism was introduced by blk-iolatency to hold memory
allocators accountable for the reclaim and other shared IOs they cause. The
duration of the delay is dynamically balanced between iolatency increasing the
value on each target miss and it auto-decaying as time passes and threads get
delayed on it.
While this works well for iolatency, iocost's control model isn't compatible
with it. There is no repeated "violation" events which can be balanced against
auto-decaying. iocost instead knows how much a given cgroup is over budget and
wants to prevent that cgroup from issuing IOs while over budget. Until now,
iocost has been adding the cost of force-issued IOs. However, this doesn't
reflect the amount which is already over budget and is simply not enough to
counter the auto-decaying allowing anon-memory leaking low priority cgroup to
go over its alloted share of IOs.
As auto-decaying doesn't make much sense for iocost, this patch introduces a
different mode of operation for use_delay - when blkcg_set_delay() are used
insted of blkcg_add/use_delay(), the delay duration is not auto-decayed until it
is explicitly cleared with blkcg_clear_delay(). iocost is updated to keep the
delay duration synchronized to the budget overage amount.
With this change, iocost can effectively police cgroups which generate
significant amount of force-issued IOs.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When replacing the bd_super check with a bd_openers I followed a logical
conclusion, which turns out to be utterly wrong. When a block device has
bd_super sets it has a mount file system on it (although not every
mounted file system sets bd_super), but that also implies it doesn't even
have partitions to start with.
So instead of trying to come up with a logical check for all openers,
just remove the check entirely.
Fixes: d3ef553627 ("block: fix busy device checking in blk_drop_partitions")
Fixes: cb6b771b05 ("block: fix busy device checking in blk_drop_partitions again")
Reported-by: Michal Koutný <mkoutny@suse.com>
Reported-by: Yang Xu <xuyang2018.jy@cn.fujitsu.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add a little helper that passes the right nowait flag to blk_queue_enter
based on the bio flag, and terminates the bio with the right error code
if entering the queue fails.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
BIO_QUEUE_ENTERED is only used for cgroup accounting now, so rename
the flag and move setting it into the cgroup code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Instead of a convoluted chain just check for REQ_OP_READ directly,
and keep all the memory stall code together in a single unlikely
branch.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The current documentation is a little weird, as it doesn't clearly
explain which function to use, and also has the guts of the information
on generic_make_request, which is the internal interface for stacking
drivers.
Fix this up by properly documenting submit_bio, and only documenting
the differences and the use case for generic_make_request.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Fix sparse warnings:
block/blk-mq-sched.c:209:5: warning: symbol '__blk_mq_sched_dispatch_requests' was not declared. Should it be static?
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Zheng Bin <zhengbin13@huawei.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When jumping to the out_put_disk label, we will call put_disk(), which will
trigger a call to disk_release(), which calls blk_put_queue().
Later in the cleanup code, we do blk_cleanup_queue(), which will also call
blk_put_queue().
Putting the queue twice is incorrect, and will generate a KASAN splat.
Set the disk->queue pointer to NULL, before calling put_disk(), so that the
first call to blk_put_queue() will not free the queue.
The second call to blk_put_queue() uses another pointer to the same queue,
so this call will still free the queue.
Fixes: 85136c0102 ("lightnvm: simplify geometry enumeration")
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Call blk_mq_make_request when no ->make_request_fn is set. This is
safe now that blk_alloc_queue always sets up the pointer for make_request
based drivers. This avoids an indirect call in the blk-mq driver I/O
fast path, which is rather expensive due to spectre mitigations.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The make_request_fn pointer should only be assigned by blk_alloc_queue.
Fix a left over manual initialization.
Fixes: ff27668ce8 ("bcache: pass the make_request methods to blk_queue_make_request")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
create_io_context just has a single caller, which also happens to not
even use the return value. Just open code it there.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
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>
Move all zoned mode related code from null_blk_main.c to
null_blk_zoned.c, avoiding an ugly #ifdef in the process.
Rename null_zone_init() into null_init_zoned_dev(), null_zone_exit()
into null_free_zoned_dev() and add the new function
null_register_zoned_dev() to finalize the zoned dev setup before
add_disk().
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
For write operations issued to a null_blk device with zoned mode
enabled, the state and write pointer position of the zone targeted by
the command should be checked before badblocks and memory backing
are handled as the write may be first failed due to, for instance, a
sector position not aligned with the zone write pointer. This order of
checking for errors reflects more accuratly the behavior of physical
zoned devices.
Furthermore, the write pointer position of the target zone should be
incremented only and only if no errors are reported by badblocks and
memory backing handling.
To fix this, introduce the small helper function null_process_cmd()
which execute null_handle_badblocks() and null_handle_memory_backed()
and use this function in null_zone_write() to correctly handle write
requests to zoned null devices depending on the type and state of the
write target zone. Also call this function in null_handle_zoned() to
process read requests to zoned null devices.
null_process_cmd() is called directly from null_handle_cmd() for
regular null devices, resulting in no functional change for these type
of devices. To have symmetric names, the function null_handle_zoned()
is renamed to null_process_zoned_cmd().
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
There are only two callers of blk_rq_map_sg/__blk_rq_map_sg that set
the dma_pad value in the queue. Move the handling into those callers
instead of burdening the common code, and move the ->extra_len field
from struct request to struct scsi_cmnd.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Don't burden the common block code with with specifics of the libata DMA
draining mechanism. Instead move most of the code to the scsi midlayer.
That also means the nr_phys_segments adjustments in the blk-mq fast path
can go away entirely, given that SCSI never looks at nr_phys_segments
after mapping the request to a scatterlist.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
scsi_init_io is the only caller of scsi_init_sgtable. Merge the two
function to make upcoming changes a little easier.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
To be able to move some of the special purpose hacks in blk_rq_map_sg
into the callers we need a variant that returns the last mapped
S/G list element to the caller. Add that variant as __blk_rq_map_sg
and make blk_rq_map_sg a trivial inline wrapper around it.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The RQF_COPY_USER is set for bio where the passthrough request mapping
helpers decided that bounce buffering is required. It is then used to
pad scatterlist for drivers that required it. But given that
non-passthrough requests are per definition aligned, and directly mapped
pass-through request must be aligned it is not actually required at all.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Dax related code already removed from this file.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Systemtap 4.2 is unable to correctly interpret the "u32 (*missed_ppm)[2]"
argument of the iocost_ioc_vrate_adj trace entry defined in
include/trace/events/iocost.h leading to the following error:
/tmp/stapAcz0G0/stap_c89c58b83cea1724e26395efa9ed4939_6321_aux_6.c:78:8:
error: expected ‘;’, ‘,’ or ‘)’ before ‘*’ token
, u32[]* __tracepoint_arg_missed_ppm
That argument type is indeed rather complex and hard to read. Looking
at block/blk-iocost.c. It is just a 2-entry u32 array. By simplifying
the argument to a simple "u32 *missed_ppm" and adjusting the trace
entry accordingly, the compilation error was gone.
Fixes: 7caa47151a ("blkcg: implement blk-iocost")
Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
invalidate_partition and bdev_unhash_inode are always paired, and
invalidate_partition already does an icache lookup for the block device
inode. Piggy back on that to remove the inode from the hash.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
invalidate_partition is only used in genhd.c, so mark it static. Also
drop the return value given that is is always ignored.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We just checked a little above that the block device for the partition
im busy. That implies no file system is mounted, and thus the only
thing in fsync_bdev that actually is used is sync_blockdev. Just call
sync_blockdev directly.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Given that the device must not be busy, most of the calls from
invalidate_partition that are related to file system metadata are
guranteed to not happen. Just open code the calls to sync_blockdev
and invalidate_bdev instead.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Use the blk_drop_partitions function instead of messing around with
ioctls that get kernel pointers. For this blk_drop_partitions needs
to be exported, which it normally shouldn't - make an exception for
s390 only.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The gendisk can be trivially deducted from the block_device.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The function has a single caller, so just open code it.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Move hd_ref_init out of line as there it isn't anywhere near a fast path,
and rename the rcu ref freeing callbacks to be more descriptive.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
All callers have the hd_struct at hand, so pass it instead of performing
another lookup.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Split each sub-command out into a separate helper, and move those helpers
to block/partitions/core.c instead of having a lot of partition
manipulation logic open coded in block/ioctl.c.
Signed-off-by: Christoph Hellwig <hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This reverts commit 7e70aa789d.
Now that we have the patches ("blk-mq: In blk_mq_dispatch_rq_list()
"no budget" is a reason to kick") and ("blk-mq: Rerun dispatching in
the case of budget contention") we should no longer need the fix in
the SCSI code. Revert it, resolving conflicts with other patches that
have touched this code.
With this revert (and the two new patches) I can run the script that
was in commit 7e70aa789d ("scsi: core: run queue if SCSI device
queue isn't ready and queue is idle") in a loop with no failure. If I
do this revert without the two new patches I can easily get a failure.
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Acked-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
If ever a thread running blk-mq code tries to get budget and fails it
immediately stops doing work and assumes that whenever budget is freed
up that queues will be kicked and whatever work the thread was trying
to do will be tried again.
One path where budget is freed and queues are kicked in the normal
case can be seen in scsi_finish_command(). Specifically:
- scsi_finish_command()
- scsi_device_unbusy()
- # Decrement "device_busy", AKA release budget
- scsi_io_completion()
- scsi_end_request()
- blk_mq_run_hw_queues()
The above is all well and good. The problem comes up when a thread
claims the budget but then releases it without actually dispatching
any work. Since we didn't schedule any work we'll never run the path
of finishing work / kicking the queues.
This isn't often actually a problem which is why this issue has
existed for a while and nobody noticed. Specifically we only get into
this situation when we unexpectedly found that we weren't going to do
any work. Code that later receives new work kicks the queues. All
good, right?
The problem shows up, however, if timing is just wrong and we hit a
race. To see this race let's think about the case where we only have
a budget of 1 (only one thread can hold budget). Now imagine that a
thread got budget and then decided not to dispatch work. It's about
to call put_budget() but then the thread gets context switched out for
a long, long time. While in this state, any and all kicks of the
queue (like the when we received new work) will be no-ops because
nobody can get budget. Finally the thread holding budget gets to run
again and returns. All the normal kicks will have been no-ops and we
have an I/O stall.
As you can see from the above, you need just the right timing to see
the race. To start with, the only case it happens if we thought we
had work, actually managed to get the budget, but then actually didn't
have work. That's pretty rare to start with. Even then, there's
usually a very small amount of time between realizing that there's no
work and putting the budget. During this small amount of time new
work has to come in and the queue kick has to make it all the way to
trying to get the budget and fail. It's pretty unlikely.
One case where this could have failed is illustrated by an example of
threads running blk_mq_do_dispatch_sched():
* Threads A and B both run has_work() at the same time with the same
"hctx". Imagine has_work() is exact. There's no lock, so it's OK
if Thread A and B both get back true.
* Thread B gets interrupted for a long time right after it decides
that there is work. Maybe its CPU gets an interrupt and the
interrupt handler is slow.
* Thread A runs, get budget, dispatches work.
* Thread A's work finishes and budget is released.
* Thread B finally runs again and gets budget.
* Since Thread A already took care of the work and no new work has
come in, Thread B will get NULL from dispatch_request(). I believe
this is specifically why dispatch_request() is allowed to return
NULL in the first place if has_work() must be exact.
* Thread B will now be holding the budget and is about to call
put_budget(), but hasn't called it yet.
* Thread B gets interrupted for a long time (again). Dang interrupts.
* Now Thread C (maybe with a different "hctx" but the same queue)
comes along and runs blk_mq_do_dispatch_sched().
* Thread C won't do anything because it can't get budget.
* Finally Thread B will run again and put the budget without kicking
any queues.
Even though the example above is with blk_mq_do_dispatch_sched() I
believe the race is possible any time someone is holding budget but
doesn't do work.
Unfortunately, the unlikely has become more likely if you happen to be
using the BFQ I/O scheduler. BFQ, by design, sometimes returns "true"
for has_work() but then NULL for dispatch_request() and stays in this
state for a while (currently up to 9 ms). Suddenly you only need one
race to hit, not two races in a row. With my current setup this is
easy to reproduce in reboot tests and traces have actually shown that
we hit a race similar to the one described above.
Note that we only need to fix blk_mq_do_dispatch_sched() and
blk_mq_do_dispatch_ctx() and not the other places that put budget. In
other cases we know that we have work to do on at least one "hctx" and
code already exists to kick that "hctx"'s queue. When that work
finally finishes all the queues will be kicked using the normal flow.
One last note is that (at least in the SCSI case) budget is shared by
all "hctx"s that have the same queue. Thus we need to make sure to
kick the whole queue, not just re-run dispatching on a single "hctx".
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We have:
* blk_mq_run_hw_queue()
* blk_mq_delay_run_hw_queue()
* blk_mq_run_hw_queues()
...but not blk_mq_delay_run_hw_queues(), presumably because nobody
needed it before now. Since we need it for a later patch in this
series, add it.
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
In blk_mq_dispatch_rq_list(), if blk_mq_sched_needs_restart() returns
true and the driver returns BLK_STS_RESOURCE then we'll kick the
queue. However, there's another case where we might need to kick it.
If we were unable to get budget we can be in much the same state as
when the driver returns BLK_STS_RESOURCE, so we should treat it the
same.
It should be noted that even if we add a whole bunch of extra kicking
to the queue in other patches this patch is still important.
Specifically any kicking that happened before we re-spliced leftover
requests into 'hctx->dispatch' wouldn't have found any work, so we
really need to make sure we kick ourselves after we've done the
splicing.
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
While trying to "dd" to the block device for a USB stick, I
encountered a hung task warning (blocked for > 120 seconds). I
managed to come up with an easy way to reproduce this on my system
(where /dev/sdb is the block device for my USB stick) with:
while true; do dd if=/dev/zero of=/dev/sdb bs=4M; done
With my reproduction here are the relevant bits from the hung task
detector:
INFO: task udevd:294 blocked for more than 122 seconds.
...
udevd D 0 294 1 0x00400008
Call trace:
...
mutex_lock_nested+0x40/0x50
__blkdev_get+0x7c/0x3d4
blkdev_get+0x118/0x138
blkdev_open+0x94/0xa8
do_dentry_open+0x268/0x3a0
vfs_open+0x34/0x40
path_openat+0x39c/0xdf4
do_filp_open+0x90/0x10c
do_sys_open+0x150/0x3c8
...
...
Showing all locks held in the system:
...
1 lock held by dd/2798:
#0: ffffff814ac1a3b8 (&bdev->bd_mutex){+.+.}, at: __blkdev_put+0x50/0x204
...
dd D 0 2798 2764 0x00400208
Call trace:
...
schedule+0x8c/0xbc
io_schedule+0x1c/0x40
wait_on_page_bit_common+0x238/0x338
__lock_page+0x5c/0x68
write_cache_pages+0x194/0x500
generic_writepages+0x64/0xa4
blkdev_writepages+0x24/0x30
do_writepages+0x48/0xa8
__filemap_fdatawrite_range+0xac/0xd8
filemap_write_and_wait+0x30/0x84
__blkdev_put+0x88/0x204
blkdev_put+0xc4/0xe4
blkdev_close+0x28/0x38
__fput+0xe0/0x238
____fput+0x1c/0x28
task_work_run+0xb0/0xe4
do_notify_resume+0xfc0/0x14bc
work_pending+0x8/0x14
The problem appears related to the fact that my USB disk is terribly
slow and that I have a lot of RAM in my system to cache things.
Specifically my writes seem to be happening at ~15 MB/s and I've got
~4 GB of RAM in my system that can be used for buffering. To write 4
GB of buffer to disk thus takes ~4000 MB / ~15 MB/s = ~267 seconds.
The 267 second number is a problem because in __blkdev_put() we call
sync_blockdev() while holding the bd_mutex. Any other callers who
want the bd_mutex will be blocked for the whole time.
The problem is made worse because I believe blkdev_put() specifically
tells other tasks (namely udev) to go try to access the device at right
around the same time we're going to hold the mutex for a long time.
Putting some traces around this (after disabling the hung task detector),
I could confirm:
dd: 437.608600: __blkdev_put() right before sync_blockdev() for sdb
udevd: 437.623901: blkdev_open() right before blkdev_get() for sdb
dd: 661.468451: __blkdev_put() right after sync_blockdev() for sdb
udevd: 663.820426: blkdev_open() right after blkdev_get() for sdb
A simple fix for this is to realize that sync_blockdev() works fine if
you're not holding the mutex. Also, it's not the end of the world if
you sync a little early (though it can have performance impacts).
Thus we can make a guess that we're going to need to do the sync and
then do it without holding the mutex. We still do one last sync with
the mutex but it should be much, much faster.
With this, my hung task warnings for my test case are gone.
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Guenter Roeck <groeck@chromium.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When remapping a mapping where a portion of a VMA is remapped
into another portion of the VMA it can cause the VMA to become
split. During the copy_vma operation the VMA can actually
be remerged if it's an anonymous VMA whose pages have not yet
been faulted. This isn't normally a problem because at the end
of the remap the original portion is unmapped causing it to
become split again.
However, MREMAP_DONTUNMAP leaves that original portion in place which
means that the VMA which was split and then remerged is not actually
split at the end of the mremap. This patch fixes a bug where
we don't detect that the VMAs got remerged and we end up
putting back VM_ACCOUNT on the next mapping which is completely
unreleated. When that next mapping is unmapped it results in
incorrectly unaccounting for the memory which was never accounted,
and eventually we will underflow on the memory comittment.
There is also another issue which is similar, we're currently
accouting for the number of pages in the new_vma but that's wrong.
We need to account for the length of the remap operation as that's
all that is being added. If there was a mapping already at that
location its comittment would have been adjusted as part of
the munmap at the start of the mremap.
A really simple repro can be seen in:
https://gist.github.com/bgaff/e101ce99da7d9a8c60acc641d07f312c
Fixes: e346b38130 ("mm/mremap: add MREMAP_DONTUNMAP to mremap()")
Reported-by: syzbot <syzkaller@googlegroups.com>
Signed-off-by: Brian Geffon <bgeffon@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>