Our current code is supposed to serialise the commits by waiting for all
the drm_crtc_commits associated to the previous HVS state.
However, assuming we have two CRTCs running and being configured and we
configure each one alternately, we end up in a situation where we're
not waiting at all.
Indeed, starting with a state (state 0) where both CRTCs are running,
and doing a commit (state 1) on the first CRTC (CRTC 0), we'll associate
its commit to its assigned FIFO in vc4_hvs_state.
If we get a new commit (state 2), this time affecting the second CRTC
(CRTC 1), the DRM core will allow both commits to execute in parallel
(assuming they don't have any share resources).
Our code in vc4_atomic_commit_tail is supposed to make sure we only get
one commit at a time and serialised by order of submission. It does so
by using for_each_old_crtc_in_state, making sure that the CRTC has a
FIFO assigned, is used, and has a commit pending. If it does, then we'll
wait for the commit before going forward.
During the transition from state 0 to state 1, as our old CRTC state we
get the CRTC 0 state 0, its commit, we wait for it, everything works fine.
During the transition from state 1 to state 2 though, the use of
for_each_old_crtc_in_state is wrong. Indeed, while the code assumes it's
returning the state of the CRTC in the old state (so CRTC 0 state 1), it
actually returns the old state of the CRTC affected by the current
commit, so CRTC 0 state 0 since it wasn't part of state 1.
Due to this, if we alternate between the configuration of CRTC 0 and
CRTC 1, we never actually wait for anything since we should be waiting
on the other every time, but it never is affected by the previous
commit.
Change the logic to, at every commit, look at every FIFO in the previous
HVS state, and if it's in use and has a commit associated to it, wait
for that commit.
Fixes: 9ec03d7f1e ("drm/vc4: kms: Wait on previous FIFO users before a commit")
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Reviewed-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Tested-by: Jian-Hong Pan <jhp@endlessos.org>
Link: https://lore.kernel.org/r/20211117094527.146275-7-maxime@cerno.tech
Our HVS global state, when duplicated, will also copy the pointer to the
drm_crtc_commit (and increase the reference count) for each FIFO if the
pointer is not NULL.
However, our atomic_setup function will overwrite that pointer without
putting the reference back leading to a memory leak.
Since the commit is only relevant during the atomic commit process, it
doesn't make sense to duplicate the reference to the commit anyway.
Let's remove it.
Fixes: 9ec03d7f1e ("drm/vc4: kms: Wait on previous FIFO users before a commit")
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Reviewed-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Tested-by: Jian-Hong Pan <jhp@endlessos.org>
Link: https://lore.kernel.org/r/20211117094527.146275-6-maxime@cerno.tech
Commit 9ec03d7f1e ("drm/vc4: kms: Wait on previous FIFO users before a
commit") introduced a wait on the previous commit done on a given HVS
FIFO.
However, we never cleared that pointer once done. Since
drm_crtc_commit_put can free the drm_crtc_commit structure directly if
we were the last user, this means that it can lead to a use-after free
if we were to duplicate the state, and that stale pointer would even be
copied to the new state.
Set the pointer to NULL once we're done with the wait so that we don't
carry over a pointer to a free'd structure.
Fixes: 9ec03d7f1e ("drm/vc4: kms: Wait on previous FIFO users before a commit")
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Reviewed-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Tested-by: Jian-Hong Pan <jhp@endlessos.org>
Link: https://lore.kernel.org/r/20211117094527.146275-5-maxime@cerno.tech
Commit 9ec03d7f1e ("drm/vc4: kms: Wait on previous FIFO users before a
commit") introduced a global state for the HVS, with each FIFO storing
the current CRTC commit so that we can properly synchronize commits.
However, the refcounting was off and we thus ended up leaking the
drm_crtc_commit structure every commit. Add a drm_crtc_commit_put to
prevent the leakage.
Fixes: 9ec03d7f1e ("drm/vc4: kms: Wait on previous FIFO users before a commit")
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Reviewed-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Tested-by: Jian-Hong Pan <jhp@endlessos.org>
Link: https://lore.kernel.org/r/20211117094527.146275-4-maxime@cerno.tech
The HVS global state functions return an error pointer, but in most
cases we check if it's NULL, possibly resulting in an invalid pointer
dereference.
Fixes: 9ec03d7f1e ("drm/vc4: kms: Wait on previous FIFO users before a commit")
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Reviewed-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Tested-by: Jian-Hong Pan <jhp@endlessos.org>
Link: https://lore.kernel.org/r/20211117094527.146275-3-maxime@cerno.tech
Several DRM/KMS atomic commits can run in parallel if they affect
different CRTC. These commits share the global HVS state, so we have
some code to make sure we run commits in sequence. This synchronization
code is one of the first thing that runs in vc4_atomic_commit_tail().
Another constraints we have is that we need to make sure the HVS clock
gets a boost during the commit. That code relies on clk_set_min_rate and
will remove the old minimum and set a new one. We also need another,
temporary, minimum for the duration of the commit.
The algorithm is thus to set a temporary minimum, drop the previous
one, do the commit, and finally set the minimum for the current mode.
However, the part that sets the temporary minimum and drops the older
one runs before the commit synchronization code.
Thus, under the proper conditions, we can end up mixing up the minimums
and ending up with the wrong one for our current step.
To avoid it, let's move the clock setup in the protected section.
Fixes: d7d96c00e5 ("drm/vc4: hvs: Boost the core clock during modeset")
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Reviewed-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Tested-by: Jian-Hong Pan <jhp@endlessos.org>
Link: https://lore.kernel.org/r/20211117094527.146275-2-maxime@cerno.tech
kernel test robot reported that RCU stall via printk() flooding is
possible [1] when stress testing.
Link: https://lkml.kernel.org/r/20211129073709.GA18483@xsang-OptiPlex-9020 [1]
Reported-by: kernel test robot <oliver.sang@intel.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
syzbot is reporting circular locking problem at __loop_clr_fd() [1], for
commit 87579e9b7d ("loop: use worker per cgroup instead of kworker")
is calling destroy_workqueue() with lo->lo_mutex held.
Since all functions where lo->lo_state matters are already checking
lo->lo_state with lo->lo_mutex held (in order to avoid racing with e.g.
ioctl(LOOP_CTL_REMOVE)), and __loop_clr_fd() can be called from either
ioctl(LOOP_CLR_FD) xor close(), lo->lo_state == Lo_rundown is considered
as an exclusive lock for __loop_clr_fd(). Therefore, hold lo->lo_mutex
inside __loop_clr_fd() only when asserting/updating lo->lo_state.
Since ioctl(LOOP_CLR_FD) depends on lo->lo_state == Lo_bound, a valid
lo->lo_backing_file must have been assigned by ioctl(LOOP_SET_FD) or
ioctl(LOOP_CONFIGURE). Thus, we can remove lo->lo_backing_file test,
and convert __loop_clr_fd() into a void function.
Link: https://syzkaller.appspot.com/bug?extid=63614029dfb79abd4383 [1]
Reported-by: syzbot <syzbot+63614029dfb79abd4383@syzkaller.appspotmail.com>
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/8ebe3b2e-8975-7f26-0620-7144a3b8b8cd@i-love.sakura.ne.jp
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Now that blk_execute_rq does not take a gendisk argument there is no need
to pass it through the scsi_ioctl callchain either.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20211126121802.2090656-6-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Remove the gendisk aregument to blk_execute_rq and blk_execute_rq_nowait
given that it is unused now. Also convert the boolean at_head parameter
to actually use the bool type while touching the prototype.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20211126121802.2090656-5-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Just use the disk attached to the request_queue instead.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20211126121802.2090656-4-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
There is a 1:1 relationship between request_queues and gendisks now, so
no need for these extra checks.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20211126121802.2090656-3-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The block layer already performs this check, no need to duplicate it in
the driver.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Miquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20211126121802.2090656-2-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The variable ret is being initialized with a value that is never
read, it is being updated later on. The assignment is redundant and
can be removed.
Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Link: https://lore.kernel.org/r/20211126230652.1175636-1-colin.i.king@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Grab a reference to the newly allocated or existing io_context in
create_task_io_context and return it. This simplifies the callers and
removes the need for double lookups.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211126115817.2087431-13-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
In __copy_io we know that the newly allocate task_struct does not have
an I/O context yet and is not exiting. So just allocate the I/O context
struct and install it directly. There is no need to lock the task
either as it is just being created.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211126115817.2087431-12-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
After the prepare side has been moved to the only I/O scheduler that
cares, do the same for the cleanup and the NULL initialization.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211126115817.2087431-9-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Move blk_mq_sched_assign_ioc so that many interfaces from the file can
be marked static. Rename the function to ioc_find_get_icq as well and
return the icq to simplify the interface.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211126115817.2087431-8-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This reverts commit 4896c4e64ba5d5d5acdbcf68c5910dd4f6d8fa62.
The helper is not needed any more.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211126115817.2087431-6-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Move the copying of the I/O context to the block layer as that is where
we can use the proper low-level interfaces.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211126115817.2087431-3-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
bio->bi_opf isn't finalized before checking the bio, so use it after
submit_bio_checks() returns.
Fixes: 5b13bc8a3f ("blk-mq: cleanup request allocation")
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Commit 7cc4ffc555 ("block, bfq: put reqs of waker and woken in
dispatch list") added a condition to bfq_insert_request() which added
waker's requests directly to dispatch list. The rationale was that
completing waker's IO is needed to get more IO for the current queue.
Although this rationale is valid, there is a hole in it. The waker does
not necessarily serve the IO only for the current queue and maybe it's
current IO is not needed for current queue to make progress. Furthermore
injecting IO like this completely bypasses any service accounting within
bfq and thus we do not properly track how much service is waker's queue
getting or that the waker is actually doing any IO. Depending on the
conditions this can result in the waker getting too much or too few
service.
Consider for example the following job file:
[global]
directory=/mnt/repro/
rw=write
size=8g
time_based
runtime=30
ramp_time=10
blocksize=1m
direct=0
ioengine=sync
[slowwriter]
numjobs=1
prioclass=2
prio=7
fsync=200
[fastwriter]
numjobs=1
prioclass=2
prio=0
fsync=200
Despite processes have very different IO priorities, they get the same
about of service. The reason is that bfq identifies these processes as
having waker-wakee relationship and once that happens, IO from
fastwriter gets injected during slowwriter's time slice. As a result bfq
is not aware that fastwriter has any IO to do and constantly schedules
only slowwriter's queue. Thus fastwriter is forced to compete with
slowwriter's IO all the time instead of getting its share of time based
on IO priority.
Drop the special injection condition from bfq_insert_request(). As a
result, requests will be tracked and queued in a normal way and on next
dispatch bfq_select_queue() can decide whether the waker's inserted
requests should be injected during the current queue's timeslice or not.
Fixes: 7cc4ffc555 ("block, bfq: put reqs of waker and woken in dispatch list")
Acked-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20211125133645.27483-8-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Waker - wakee relationships are important in deciding whether one queue
can preempt the other one. Print information about detected waker-wakee
relationships so that scheduling decisions can be better understood from
block traces.
Acked-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20211125133645.27483-7-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Instead of having helper formating bfqq pid, provide a helper to
generate full bfqq name as used in the traces. It saves some code
duplication and will save more in the coming tracepoints.
Acked-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20211125133645.27483-6-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Currently, when process A starts issuing requests shortly after process
B has completed some IO three times in a row, we decide that B is a
"waker" of A meaning that completing IO of B is needed for A to make
progress and generally stop separating A's and B's IO much. This logic
is useful to avoid unnecessary idling and thus throughput loss for cases
where workload needs to switch e.g. between the process and the
journaling thread doing IO. However the detection heuristic tends to
frequently give false positives when A and B are fighting IO bandwidth
and other processes aren't doing much IO as we are basically deemed to
eventually accumulate three occurences of a situation where one process
starts issuing requests after the other has completed some IO. To reduce
these false positives, cancel the waker detection also if we didn't
accumulate three detected wakeups within given timeout. The rationale is
that if wakeups are really rare, the pointless idling doesn't hurt
throughput that much anyway.
This significantly reduces false waker detection for workload like:
[global]
directory=/mnt/repro/
rw=write
size=8g
time_based
runtime=30
ramp_time=10
blocksize=1m
direct=0
ioengine=sync
[slowwriter]
numjobs=1
fsync=200
[fastwriter]
numjobs=1
fsync=200
Acked-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20211125133645.27483-5-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When cgroup IO scheduling is used with BFQ it does not really provide
service differentiation if the cgroup drives a big IO depth. That for
example happens with writeback which asynchronously submits lots of IO
but it can happen with AIO as well. The problem is that if we have two
cgroups that submit IO with different weights, the cgroup with higher
weight properly gets more IO time and is able to dispatch more IO.
However this causes lower weight cgroup to accumulate more requests
inside BFQ and eventually lower weight cgroup consumes most of IO
scheduler tags. At that point higher weight cgroup stops getting better
service as it is mostly blocked waiting for a scheduler tag while its
queues inside BFQ are empty and thus lower weight cgroup gets served.
Check how many requests submitting cgroup has allocated in
bfq_limit_depth() and if it consumes more requests than what would
correspond to its weight limit available depth to 1 so that the cgroup
cannot consume many more requests. With this limitation the higher
weight cgroup gets proper service even with writeback.
Reviewed-by: Michal Koutný <mkoutny@suse.com>
Acked-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20211125133645.27483-4-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Store bitmap depth shift inside bfq_data so that we can use it in
bfq_limit_depth() for proportioning when limiting number of available
request tags for a cgroup.
Acked-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20211125133645.27483-3-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When we want to limit number of requests used by each bfqq and also
cgroup, we need to track also number of requests used by each cgroup.
So track number of allocated requests for each bfq_entity.
Acked-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20211125133645.27483-2-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Currently we lookup ICQ only after the request is allocated. However BFQ
will want to decide how many scheduler tags it allows a given bfq queue
(effectively a process) to consume based on cgroup weight. So provide a
function blk_mq_sched_get_icq() so that BFQ can lookup ICQ earlier.
Acked-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20211125133645.27483-1-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The completion callback for the sdhci-pci device is invoked from a
kworker.
I couldn't identify in which context is mmc_blk_mq_req_done() invoke but
the remaining caller are from invoked from preemptible context. Here it
would make sense to complete the request directly instead scheduling
ksoftirqd for its completion.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Acked-by: Adrian Hunter <adrian.hunter@intel.com>
Link: https://lore.kernel.org/r/20211025070658.1565848-3-bigeasy@linutronix.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add blk_mq_complete_request_direct() which completes the block request
directly instead deferring it to softirq for single queue devices.
This is useful for devices which complete the requests in preemptible
context and raising softirq from means scheduling ksoftirqd.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211025070658.1565848-2-bigeasy@linutronix.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This function is trivial and is only used in one place. Having this
function is misleading because it implies that blk_crypto_register()
needs to be paired with blk_crypto_unregister(), which is not the case.
Just set disk->queue->crypto_profile to NULL directly.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211124013733.347612-1-ebiggers@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Refactor the request alloction so that blk_mq_get_cached_request tries
to find a cached request first, and the entirely separate and now
self contained blk_mq_get_new_requests allocates one or more requests
if that is not possible.
There is a small change in behavior as submit_bio_checks is called
twice now if a cached request is present but can't be used, but that
is a small price to pay for unwinding this code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211124062856.1444266-1-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Open code elevator_exit in it's only caller, and rename __elevator_exit to
elevator_exit.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211123185312.1432157-3-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>