We currently require that the back_lock is held when calling the functions
that manipulate the iscsi_task refcount. The only reason for this is to
handle races where we are handling SCSI-ml EH callbacks and the cmd is
completing at the same time the normal completion path is running, and we
can't return from the EH callback until the driver has stopped accessing
the cmd. Holding the back_lock while also accessing the task->state made it
simple to check that a cmd is completing and also get/put a refcount at the
same time, and at the time we were not as concerned about performance.
The problem is that we don't want to take the back_lock from the xmit path
for normal I/O since it causes contention with the completion path if the
user has chosen to try and split those paths on different CPUs (in this
case abusing the CPUs and ignoring caching improves perf for some uses).
Begins to remove the back_lock requirement for iscsi_get/put_task by
removing the requirement for the get path. Instead of always holding the
back_lock we detect if something has done the last put and is about to call
iscsi_free_task(). A subsequent commit will then allow iSCSI code to do the
last put on a task and only grab the back_lock if the refcount is now zero
and it's going to call iscsi_free_task().
Link: https://lore.kernel.org/r/20220616224557.115234-8-michael.christie@oracle.com
Reviewed-by: Lee Duncan <lduncan@suse.com>
Signed-off-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Commit 5923d64b7a ("scsi: libiscsi: Drop taskqueuelock") added an extra
task->state because for commit 6f8830f5bb ("scsi: libiscsi: add lock
around task lists to fix list corruption regression") we didn't know why we
ended up with cmds on the list and thought it might have been a bad target
sending a response while we were still sending the cmd. We were never able
to get a target to send us a response early, because it turns out the bug
was just a race in libiscsi/libiscsi_tcp where:
1. iscsi_tcp_r2t_rsp() queues a r2t to tcp_task->r2tqueue.
2. iscsi_tcp_task_xmit() runs iscsi_tcp_get_curr_r2t() and sees we have a
r2t. It dequeues it and iscsi_tcp_task_xmit() starts to process it.
3. iscsi_tcp_r2t_rsp() runs iscsi_requeue_task() and puts the task on the
requeue list.
4. iscsi_tcp_task_xmit() sends the data for r2t. This is the final chunk
of data, so the cmd is done.
5. target sends the response.
6. On a different CPU from #3, iscsi_complete_task() processes the
response. Since there was no common lock for the list, the lists/tasks
pointers are not fully in sync, so could end up with list corruption.
Since it was just a race on our side, remove the extra check and fix up the
comments.
Link: https://lore.kernel.org/r/20220616224557.115234-7-michael.christie@oracle.com
Reviewed-by: Wu Bo <wubo40@huawei.com>
Reviewed-by: Lee Duncan <lduncan@suse.com>
Signed-off-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
For software iSCSI, we do a session per host so there is no need to set the
target's can_queue since it's the same as the host one. Setting it just
results in extra atomic checks in the main I/O path.
Link: https://lore.kernel.org/r/20220616224557.115234-6-michael.christie@oracle.com
Reviewed-by: Lee Duncan <lduncan@suse.com>
Reviewed-by: Wu Bo <wubo40@huawei.com>
Signed-off-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
If we have more data, set the MSG_SENDPAGE_NOTLAST in case we go down the
sendpage path.
Link: https://lore.kernel.org/r/20220616224557.115234-5-michael.christie@oracle.com
Reviewed-by: Lee Duncan <lduncan@suse.com>
Signed-off-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
We don't always want to run the recv path from the network softirq because
when we have to have multiple sessions sharing the same CPUs, some sessions
can eat up the NAPI softirq budget and affect other sessions or users.
Allow us to queue the recv handling to the iscsi workqueue so we can have
the scheduler/wq code try to balance the work and CPU use across all
sessions' worker threads.
Note: It wasn't the original intent of the change but a nice side effect is
that for some workloads/configs we get a nice performance boost. For a
simple read heavy test:
fio --direct=1 --filename=/dev/dm-0 --rw=randread --bs=256K
--ioengine=libaio --iodepth=128 --numjobs=4
where the iscsi threads, fio jobs, and rps_cpus share CPUs we see a 32%
throughput boost. We also see increases for small I/O IOPs tests but it's
not as high.
Link: https://lore.kernel.org/r/20220616224557.115234-4-michael.christie@oracle.com
Reviewed-by: Lee Duncan <lduncan@suse.com>
Signed-off-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Add helpers to allow the drivers to run their recv paths from libiscsi's
workqueue.
Link: https://lore.kernel.org/r/20220616224557.115234-3-michael.christie@oracle.com
Reviewed-by: Lee Duncan <lduncan@suse.com>
Signed-off-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Rename iscsi_conn_queue_work() to iscsi_conn_queue_xmit() to reflect that
it handles queueing of xmits only.
Link: https://lore.kernel.org/r/20220616224557.115234-2-michael.christie@oracle.com
Reviewed-by: Lee Duncan <lduncan@suse.com>
Reviewed-by: Wu Bo <wubo40@huawei.com>
Signed-off-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
When the system is shutting down, iscsid is not running so we will not get
a response to the ISCSI_ERR_INVALID_HOST error event. The system shutdown
will then hang waiting on userspace to remove the session.
This has libiscsi force the destruction of the session from the kernel when
iscsi_host_remove() is called from a driver's shutdown callout.
This fixes a regression added in qedi boot with commit d1f2ce7763 ("scsi:
qedi: Fix host removal with running sessions") which made qedi use the
common session removal function that waits on userspace instead of rolling
its own kernel based removal.
Link: https://lore.kernel.org/r/20220616222738.5722-7-michael.christie@oracle.com
Fixes: d1f2ce7763 ("scsi: qedi: Fix host removal with running sessions")
Tested-by: Nilesh Javali <njavali@marvell.com>
Reviewed-by: Lee Duncan <lduncan@suse.com>
Reviewed-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
When handling errors that lead to host removal use QEDI_MODE_NORMAL. There
is currently no difference in behavior between NORMAL and SHUTDOWN, but in
a subsequent commit we will want to know when we are called from the
pci_driver shutdown callout vs remove/err_handler so we know when userspace
is up.
Link: https://lore.kernel.org/r/20220616222738.5722-6-michael.christie@oracle.com
Tested-by: Nilesh Javali <njavali@marvell.com>
Reviewed-by: Lee Duncan <lduncan@suse.com>
Reviewed-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
During qedi shutdown we need to stop the iSCSI layer from sending new nops
as pings and from responding to target ones and make sure there is no
running connection cleanups. Commit d1f2ce7763 ("scsi: qedi: Fix host
removal with running sessions") converted the driver to use the libicsi
helper to drive session removal, so the above issues could be handled. The
problem is that during system shutdown iscsid will not be running so when
we try to remove the root session we will hang waiting for userspace to
reply.
Add a helper that will drive the destruction of sessions like these during
system shutdown.
Link: https://lore.kernel.org/r/20220616222738.5722-5-michael.christie@oracle.com
Tested-by: Nilesh Javali <njavali@marvell.com>
Reviewed-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
In the next patch we allow drivers to drive session removal during
shutdown. In this case iscsid will not be running, so we need to detect
bound endpoints and disconnect them. This moves the bound ep check so we
now always check.
Link: https://lore.kernel.org/r/20220616222738.5722-4-michael.christie@oracle.com
Tested-by: Nilesh Javali <njavali@marvell.com>
Reviewed-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
iscsi_if_stop_conn() is only called from the userspace interface but in a
subsequent commit we will want to call it from the kernel interface to
allow drivers like qedi to remove sessions from inside the kernel during
shutdown. This removes the iscsi_uevent code from iscsi_if_stop_conn() so we
can call it in a new helper.
Link: https://lore.kernel.org/r/20220616222738.5722-3-michael.christie@oracle.com
Tested-by: Nilesh Javali <njavali@marvell.com>
Reviewed-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
If qla4xxx doesn't remove the connection before the session, the iSCSI
class tries to remove the connection for it. We were doing a
iscsi_put_conn() in the iter function which is not needed and will result
in a use after free because iscsi_remove_conn() will free the connection.
Link: https://lore.kernel.org/r/20220616222738.5722-2-michael.christie@oracle.com
Tested-by: Nilesh Javali <njavali@marvell.com>
Reviewed-by: Lee Duncan <lduncan@suse.com>
Reviewed-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
If CONFIG_PM_SLEEP is not set.
make ARCH=arm64 CROSS_COMPILE=aarch64-linux-gnu-, will fail:
drivers/ufs/host/ufs-mediatek.c: In function ‘ufs_mtk_vreg_fix_vcc’:
drivers/ufs/host/ufs-mediatek.c:688:46: warning: format ‘%u’ expects argument of type ‘unsigned int’, but argument 4 has type ‘long unsigned int’ [-Wformat=]
snprintf(vcc_name, MAX_VCC_NAME, "vcc-opt%u", res.a1);
~^ ~~~~~~
%lu
drivers/ufs/host/ufs-mediatek.c: In function ‘ufs_mtk_system_suspend’:
drivers/ufs/host/ufs-mediatek.c:1371:8: error: implicit declaration of function ‘ufshcd_system_suspend’; did you mean ‘ufs_mtk_system_suspend’? [-Werror=implicit-function-declaration]
ret = ufshcd_system_suspend(dev);
^~~~~~~~~~~~~~~~~~~~~
ufs_mtk_system_suspend
drivers/ufs/host/ufs-mediatek.c: In function ‘ufs_mtk_system_resume’:
drivers/ufs/host/ufs-mediatek.c:1386:9: error: implicit declaration of function ‘ufshcd_system_resume’; did you mean ‘ufs_mtk_system_resume’? [-Werror=implicit-function-declaration]
return ufshcd_system_resume(dev);
^~~~~~~~~~~~~~~~~~~~
ufs_mtk_system_resume
cc1: some warnings being treated as errors
The declaration of func "ufshcd_system_suspend()" depends on
CONFIG_PM_SLEEP, so the function wrapper ufs_mtk_system_suspend() should
wrapped by CONFIG_PM_SLEEP too.
Link: https://lore.kernel.org/r/20220619115432.205504-1-renzhijie2@huawei.com
Fixes: 3fd23b8dfb ("scsi: ufs: ufs-mediatek: Fix the timing of configuring device regulators")
Reported-by: Hulk Robot <hulkci@huawei.com>
Reviewed-by: Stanley Chu <stanley.chu@mediatek.com>
Signed-off-by: Ren Zhijie <renzhijie2@huawei.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
We have re-polling for partial IO, so a request can be polled twice. If
it used two poll entries the first time then on the second
io_arm_poll_handler() it will find the old apoll entry and NULL
kmalloc()'ed second entry, i.e. apoll->double_poll, so leaking it.
Fixes: 10c873334f ("io_uring: allow re-poll if we made progress")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/fee2452494222ecc7f1f88c8fb659baef971414a.1655852245.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The context tracking namespace is going to expand and some new functions
will require even longer names. Start shrinking the context_tracking
prefix to "ct" as is already the case for some existing macros, this
will make the introduction of new functions easier.
Acked-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Neeraj Upadhyay <quic_neeraju@quicinc.com>
Cc: Uladzislau Rezki <uladzislau.rezki@sony.com>
Cc: Joel Fernandes <joel@joelfernandes.org>
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: Nicolas Saenz Julienne <nsaenz@kernel.org>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Xiongfeng Wang <wangxiongfeng2@huawei.com>
Cc: Yu Liao <liaoyu15@huawei.com>
Cc: Phil Auld <pauld@redhat.com>
Cc: Paul Gortmaker<paul.gortmaker@windriver.com>
Cc: Alex Belits <abelits@marvell.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Reviewed-by: Nicolas Saenz Julienne <nsaenzju@redhat.com>
Tested-by: Nicolas Saenz Julienne <nsaenzju@redhat.com>
In kernels built with CONFIG_PREEMPT_RT=y, spinlocks are replaced by
rt_mutex, which can sleep. This means that acquiring a non-raw spinlock
in a critical section where preemption is disabled can trigger the
following BUG:
BUG: scheduling while atomic: ref_scale_reade/76/0x00000002
Preemption disabled at:
ref_lock_section+0x16/0x80
Call Trace:
<TASK>
dump_stack_lvl+0x5b/0x82
dump_stack+0x10/0x12
__schedule_bug.cold+0x9c/0xad
__schedule+0x839/0xc00
schedule_rtlock+0x22/0x40
rtlock_slowlock_locked+0x460/0x1350
rt_spin_lock+0x61/0xe0
ref_lock_section+0x29/0x80
rcu_scale_one_reader+0x52/0x60
ref_scale_reader+0x28d/0x490
kthread+0x128/0x150
ret_from_fork+0x22/0x30
</TASK>
This commit therefore converts spinlock to raw_spinlock.
Signed-off-by: Zqiang <qiang1.zhang@intel.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit adds warnings for allocation failure during the mem_dump_obj()
tests. It also terminates these tests upon such failure.
Signed-off-by: Li Qiong <liqiong@nfschina.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The RCU priority boosting can fail in two situations:
1) If (nr_cpus= > maxcpus=), which means if the total number of CPUs
is higher than those brought online at boot, then torture_onoff() may
later bring up CPUs that weren't online on boot. Now since rcutorture
initialization only boosts the ksoftirqds of the CPUs that have been
set online on boot, the CPUs later set online by torture_onoff won't
benefit from the boost, making RCU priority boosting fail.
2) The ksoftirqd kthreads are boosted after the creation of
rcu_torture_boost() kthreads, which opens a window large enough for these
rcu_torture_boost() kthreads to wait (despite running at FIFO priority)
for ksoftirqds that are still running at SCHED_NORMAL priority.
The issues can trigger for example with:
./kvm.sh --configs TREE01 --kconfig "CONFIG_RCU_BOOST=y"
[ 34.968561] rcu-torture: !!!
[ 34.968627] ------------[ cut here ]------------
[ 35.014054] WARNING: CPU: 4 PID: 114 at kernel/rcu/rcutorture.c:1979 rcu_torture_stats_print+0x5ad/0x610
[ 35.052043] Modules linked in:
[ 35.069138] CPU: 4 PID: 114 Comm: rcu_torture_sta Not tainted 5.18.0-rc1 #1
[ 35.096424] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.14.0-0-g155821a-rebuilt.opensuse.org 04/01/2014
[ 35.154570] RIP: 0010:rcu_torture_stats_print+0x5ad/0x610
[ 35.198527] Code: 63 1b 02 00 74 02 0f 0b 48 83 3d 35 63 1b 02 00 74 02 0f 0b 48 83 3d 21 63 1b 02 00 74 02 0f 0b 48 83 3d 0d 63 1b 02 00 74 02 <0f> 0b 83 eb 01 0f 8e ba fc ff ff 0f 0b e9 b3 fc ff f82
[ 37.251049] RSP: 0000:ffffa92a0050bdf8 EFLAGS: 00010202
[ 37.277320] rcu: De-offloading 8
[ 37.290367] RAX: 0000000000000000 RBX: 0000000000000001 RCX: 0000000000000001
[ 37.290387] RDX: 0000000000000000 RSI: 00000000ffffbfff RDI: 00000000ffffffff
[ 37.290398] RBP: 000000000000007b R08: 0000000000000000 R09: c0000000ffffbfff
[ 37.290407] R10: 000000000000002a R11: ffffa92a0050bc18 R12: ffffa92a0050be20
[ 37.290417] R13: ffffa92a0050be78 R14: 0000000000000000 R15: 000000000001bea0
[ 37.290427] FS: 0000000000000000(0000) GS:ffff96045eb00000(0000) knlGS:0000000000000000
[ 37.290448] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 37.290460] CR2: 0000000000000000 CR3: 000000001dc0c000 CR4: 00000000000006e0
[ 37.290470] Call Trace:
[ 37.295049] <TASK>
[ 37.295065] ? preempt_count_add+0x63/0x90
[ 37.295095] ? _raw_spin_lock_irqsave+0x12/0x40
[ 37.295125] ? rcu_torture_stats_print+0x610/0x610
[ 37.295143] rcu_torture_stats+0x29/0x70
[ 37.295160] kthread+0xe3/0x110
[ 37.295176] ? kthread_complete_and_exit+0x20/0x20
[ 37.295193] ret_from_fork+0x22/0x30
[ 37.295218] </TASK>
Fix this with boosting the ksoftirqds kthreads from the boosting
hotplug callback itself and before the boosting kthreads are created.
Fixes: ea6d962e80 ("rcutorture: Judge RCU priority boosting on grace periods, not callbacks")
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Currently, kvm-check-branches.sh causes each kvm.sh invocation create a
separate date-stamped directory, then after that invocation completes,
moves it into the *-group/NNNN directory. This works, but makes it more
difficult to monitor an ongoing run. This commit therefore uses the
kvm.sh --datestamp argument to make kvm.sh put the output in the right
place to start with, and also dispenses with the additional level of
datestamping. (Those wanting datestamps can find them in the log files.)
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Systems built with CONFIG_DEBUG_PREEMPT=y can trigger the following
BUG while running the rcuscale performance test:
BUG: using smp_processor_id() in preemptible [00000000] code: rcu_scale_write/69
CPU: 0 PID: 66 Comm: rcu_scale_write Not tainted 5.18.0-rc7-next-20220517-yoctodev-standard+
caller is debug_smp_processor_id+0x17/0x20
Call Trace:
<TASK>
dump_stack_lvl+0x49/0x5e
dump_stack+0x10/0x12
check_preemption_disabled+0xdf/0xf0
debug_smp_processor_id+0x17/0x20
rcu_scale_writer+0x2b5/0x580
kthread+0x177/0x1b0
ret_from_fork+0x22/0x30
</TASK>
Reproduction method:
runqemu kvm slirp nographic qemuparams="-m 4096 -smp 8" bootparams="isolcpus=2,3
nohz_full=2,3 rcu_nocbs=2,3 rcutree.dump_tree=1 rcuscale.shutdown=false
rcuscale.gp_async=true" -d
The problem is that the rcu_scale_writer() kthreads fail to set the
PF_NO_SETAFFINITY flags, which causes is_percpu_thread() to assume
that the kthread's affinity might change at any time, thus the BUG
noted above.
This commit therefore causes rcu_scale_writer() to set PF_NO_SETAFFINITY
in its kthread's ->flags field, thus preventing this BUG.
Signed-off-by: Zqiang <qiang1.zhang@intel.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The loop scanning the pipesummary[] array currently skips the last
element, which means that the diagnostics ignore those rarest of
situations, namely where some readers persist across more than ten
grace periods, but all other readers avoid spanning a full grace period.
This commit therefore adjusts the scan to include the last element of
this array.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
A recent change to the DEBUG_INFO Kconfig option means that simply adding
CONFIG_DEBUG_INFO=y to the .config file and running "make oldconfig" no
longer works. It is instead necessary to add CONFIG_DEBUG_INFO_NONE=n
and (for example) CONFIG_DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT=y.
This combination will then result in CONFIG_DEBUG_INFO being selected.
This commit therefore updates the Kconfig options produced in response
to the kvm.sh --gdb, --kasan, and --kcsan Kconfig options.
Fixes: f9b3cd2457 ("Kconfig.debug: make DEBUG_INFO selectable from a choice")
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The kernel memory leak detector located the following:
unreferenced object 0xffff95d941135b50 (size 16):
comm "swapper/0", pid 1, jiffies 4294667610 (age 1367.451s)
hex dump (first 16 bytes):
f0 c6 c2 bd d9 95 ff ff 00 00 00 00 00 00 00 00 ................
backtrace:
[<00000000bc81d9b1>] kmem_cache_alloc_trace+0x2f6/0x500
[<00000000d28be229>] rcu_torture_init+0x1235/0x1354
[<0000000032c3acd9>] do_one_initcall+0x51/0x210
[<000000003c117727>] kernel_init_freeable+0x205/0x259
[<000000003961f965>] kernel_init+0x1a/0x120
[<000000001998f890>] ret_from_fork+0x22/0x30
This is caused by the rcu_test_debug_objects() function allocating an
rcu_head structure, then failing to free it. This commit therefore adds
the needed kfree() after the last use of this structure.
Signed-off-by: Zqiang <qiang1.zhang@intel.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The existing loop has an implicit manual loop that obscures the flow
and requires an extra control variable. This commit makes this implicit
loop explicit, thus saving several lines of code.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Dumping a big ftrace buffer could lead to a RCU stall. So there is the
ftrace buffer and the stall information which needs to be printed. When
there is additionally a WARN_ON() which describes the reason for the ftrace
buffer dump and the WARN_ON() is executed _after_ ftrace buffer dump, the
information get lost in the middle of the RCU stall information.
Therefore print WARN_ON() message before dumping the ftrace buffer in
rcu_torture_writer().
[ paulmck: Add tracing_off() to avoid cruft from WARN(). ]
Signed-off-by: Anna-Maria Behnsen <anna-maria@linutronix.de>
Reviewed-by: Benedikt Spranger <b.spranger@linutronix.de>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Commit 2585014188d5 ("rcu-tasks: Be more patient for RCU Tasks
boot-time testing") fixes false positive rcu_tasks verification check
failure by repeating the test once every second until timeout using
schedule_timeout_uninterruptible().
Since rcu_tasks_verify_selft_tests() is called from do_initcalls()
as a late_initcall, this has the undesirable side effect of delaying
other late_initcall's queued after it by a second or more. Fix this by
instead using delayed_work to repeat the verification check.
Fixes: 2585014188d5 ("rcu-tasks: Be more patient for RCU Tasks boot-time testing")
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The RCU-Tasks family of grace-period primitives can take some time to
complete, and the amount of time can depend on the exact hardware and
software configuration. Some configurations boot up fast enough that the
RCU-Tasks verification process gets false-positive failures. This commit
therefore allows up to 30 seconds for the grace periods to complete, with
this value adjustable downwards using the rcupdate.rcu_task_stall_timeout
kernel boot parameter.
Reported-by: Matthew Wilcox <willy@infradead.org>
Reported-by: Zhouyi Zhou <zhouzhouyi@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Tested-by: Zhouyi Zhou <zhouzhouyi@gmail.com>
Tested-by: Mark Rutland <mark.rutland@arm.com>
This commit updates comments to reflect the changes in the series
of commits that eliminated the full task-list scan.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Cc: Neeraj Upadhyay <quic_neeraju@quicinc.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Andrii Nakryiko <andrii@kernel.org>
Cc: Martin KaFai Lau <kafai@fb.com>
Cc: KP Singh <kpsingh@kernel.org>
The rcu_tasks_trace_pregp_step() function invokes cpus_read_lock() to
disable CPU hotplug, and a later call to the rcu_tasks_trace_postscan()
function invokes cpus_read_unlock() to re-enable it. This was absolutely
necessary in the past in order to protect the intervening scan of the full
tasks list, but there is no longer such a scan. This commit therefore
improves readability by moving the cpus_read_unlock() call to the end
of the rcu_tasks_trace_pregp_step() function. This commit is a pure
code-motion commit without any (intended) change in functionality.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Cc: Neeraj Upadhyay <quic_neeraju@quicinc.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Andrii Nakryiko <andrii@kernel.org>
Cc: Martin KaFai Lau <kafai@fb.com>
Cc: KP Singh <kpsingh@kernel.org>
Currently, the RCU Tasks Trace grace-period kthread IPIs each online CPU
using smp_call_function_single() in order to track any tasks currently in
RCU Tasks Trace read-side critical sections during which the corresponding
task has neither blocked nor been preempted. These IPIs are annoying
and are also not strictly necessary because any task that blocks or is
preempted within its current RCU Tasks Trace read-side critical section
will be tracked on one of the per-CPU rcu_tasks_percpu structure's
->rtp_blkd_tasks list. So the only time that this is a problem is if
one of the CPUs runs through a long-duration RCU Tasks Trace read-side
critical section without a context switch.
Note that the task_call_func() function cannot help here because there is
no safe way to identify the target task. Of course, the task_call_func()
function will be very useful later, when processing the list of tasks,
but it needs to know the task.
This commit therefore creates a cpu_curr_snapshot() function that returns
a pointer the task_struct structure of some task that happened to be
running on the specified CPU more or less during the time that the
cpu_curr_snapshot() function was executing. If there was no context
switch during this time, this function will return a pointer to the
task_struct structure of the task that was running throughout. If there
was a context switch, then the outgoing task will be taken care of by
RCU's context-switch hook, and the incoming task was either already taken
care during some previous context switch, or it is not currently within an
RCU Tasks Trace read-side critical section. And in this latter case, the
grace period already started, so there is no need to wait on this task.
This new cpu_curr_snapshot() function is invoked on each CPU early in
the RCU Tasks Trace grace-period processing, and the resulting tasks
are queued for later quiescent-state inspection.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Neeraj Upadhyay <quic_neeraju@quicinc.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Andrii Nakryiko <andrii@kernel.org>
Cc: Martin KaFai Lau <kafai@fb.com>
Cc: KP Singh <kpsingh@kernel.org>
This commit maintains a new n_trc_holdouts counter that tracks the number
of tasks blocking the RCU Tasks grace period. This counter is useful
for debugging, and its value has been added to a diagostic message.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Cc: Neeraj Upadhyay <quic_neeraju@quicinc.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Andrii Nakryiko <andrii@kernel.org>
Cc: Martin KaFai Lau <kafai@fb.com>
Cc: KP Singh <kpsingh@kernel.org>
This commit takes off the training wheels and relies only on scanning
currently running tasks and tasks that have blocked or been preempted
within their current RCU Tasks Trace read-side critical section.
Before this commit, the time complexity of an RCU Tasks Trace grace
period is O(T), where T is the number of tasks. After this commit,
this time complexity is O(C+B), where C is the number of CPUs and B
is the number of tasks that have blocked (or been preempted) at least
once during their current RCU Tasks Trace read-side critical sections.
Of course, if all tasks have blocked (or been preempted) at least once
during their current RCU Tasks Trace read-side critical sections, this is
still O(T), but current expectations are that RCU Tasks Trace read-side
critical section will be short and that there will normally not be large
numbers of tasks blocked within such a critical section.
Dave Marchevsky kindly measured the effects of this commit on the RCU
Tasks Trace grace-period latency and the rcu_tasks_trace_kthread task's
CPU consumption per RCU Tasks Trace grace period over the course of a
fixed test, all in milliseconds:
Before After
GP latency 22.3 ms stddev > 0.1 17.0 ms stddev < 0.1
GP CPU 2.3 ms stddev 0.3 1.1 ms stddev 0.2
This was on a system with 15,000 tasks, so it is reasonable to expect
much larger savings on the systems on which this issue was first noted,
given that they sport well in excess of 100,000 tasks. CPU consumption
was measured using profiling techniques.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Cc: Neeraj Upadhyay <quic_neeraju@quicinc.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Andrii Nakryiko <andrii@kernel.org>
Cc: Martin KaFai Lau <kafai@fb.com>
Cc: KP Singh <kpsingh@kernel.org>
Tested-by: Dave Marchevsky <davemarchevsky@fb.com>
[Why&How]
GCC 12 catches the following incorrect comparison in the if arm
drivers/gpu/drm/amd/amdgpu/../dal-dev/dc/dml/dcn32/display_mode_vba_32.c: In function ‘dml32_ModeSupportAndSystemConfigurationFull’:
drivers/gpu/drm/amd/amdgpu/../dal-dev/dc/dml/dcn32/display_mode_vba_32.c:3740:33: error: the comparison will always evaluate as ‘true’ for the address of ‘USRRetrainingSupport’ will never be NULL [-Werror=address]
3740 | || &mode_lib->vba.USRRetrainingSupport[i][j])) {
| ^~
In file included from ./drivers/gpu/drm/amd/amdgpu/../dal-dev/dc/dml/display_mode_lib.h:32,
from ./drivers/gpu/drm/amd/amdgpu/../dal-dev/dc/dc.h:45,
from drivers/gpu/drm/amd/amdgpu/../dal-dev/dc/dml/dcn32/display_mode_vba_32.c:30:
./drivers/gpu/drm/amd/amdgpu/../dal-dev/dc/dml/display_mode_vba.h:1175:14: note: ‘USRRetrainingSupport’ declared here
1175 | bool USRRetrainingSupport[DC__VOLTAGE_STATES][2];
|
Fix this by remove preceding & so that value is compared instead of
address
Fixes: dda4fb85e4 ("drm/amd/display: DML changes for DCN32/321")
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why&How]
When the a 3d array is used by indexing with only one dimension in an if
condition, the addresses get compared instead of the intended value stored in the
array. GCC 12.1 caught this error:
drivers/gpu/drm/amd/amdgpu/../dal-dev/dc/dml/dcn32/display_mode_vba_32.c: In function ‘DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation’:
drivers/gpu/drm/amd/amdgpu/../dal-dev/dc/dml/dcn32/display_mode_vba_32.c:1007:45: error: the comparison will always evaluate as ‘true’ for the address of ‘use_one_row_for_frame_flip’ will never be NULL [-Werror=address]
1007 | if (v->use_one_row_for_frame_flip[k]) {
| ^
In file included from ./drivers/gpu/drm/amd/amdgpu/../dal-dev/dc/dml/display_mode_lib.h:32,
from ./drivers/gpu/drm/amd/amdgpu/../dal-dev/dc/dc.h:45,
from drivers/gpu/drm/amd/amdgpu/../dal-dev/dc/dml/dcn32/display_mode_vba_32.c:30:
./drivers/gpu/drm/amd/amdgpu/../dal-dev/dc/dml/display_mode_vba.h:605:14: note: ‘use_one_row_for_frame_flip’ declared here
605 | bool use_one_row_for_frame_flip[DC__VOLTAGE_STATES][2][DC__NUM_DPP__MAX];
|
Fix this by explicitly specifying the last two indices.
Fixes: dda4fb85e4 ("drm/amd/display: DML changes for DCN32/321")
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
there is an unexpected word "the" in the comments that need to be dropped
file: drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
line: 57
* the kernel tells the the ring what VMID to use for that command
changed to
* the kernel tells the ring what VMID to use for that command
Signed-off-by: Jiang Jian <jiangjian@cdjrlc.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Use the correct adev variable for the drm_fb_helper in
amdgpu_device_gpu_recover(). Noticed by inspection.
Fixes: 087451f372 ("drm/amdgpu: use generic fb helpers instead of setting up AMD own's.")
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The DRM_RADEON Kconfig code contains:
select BACKLIGHT_CLASS_DEVICE
So the condition these ifdefs test for is always true, drop them.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The DRM_AMDGPU Kconfig code contains:
select BACKLIGHT_CLASS_DEVICE
So the condition these ifdefs test for is always true, drop them.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
For DCN20 and above, the code that actually hooks up the provided
input_color_space got lost at some point.
Fixes COLOR_ENCODING and COLOR_RANGE doing nothing on DCN20+.
Tested using Steam Remote Play Together + gamescope.
Update other DCNs the same wasy DCN1.x was updates in
commit a1e07ba89d ("drm/amd/display: Use plane->color_space for dpp if specified")
Fixes: a1e07ba89d ("drm/amd/display: Use plane->color_space for dpp if specified")
Signed-off-by: Joshua Ashton <joshua@froggi.es>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This DC patchset brings improvements in multiple areas. In summary, we
highlight:
- Remove unnecessary code;
- Small fixes (compilation warnings, typos, etc);
- Improvements in the DPMS code;
- Fix eDP issues
- Improvements in the MST code
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Aric Cyr <aric.cyr@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
We already have DALSMC_MSG_TransferTableDram2Smu in the file dalsmc.h;
for this reason, we don't need this definition in the smu msg file.
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
In DCN32 clk hook functions, we are using the wrong reference for
get_dp_ref_clk_frequency and missing the get_dtb_ref_clk_frequency
reference. This commit adds those references.
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
For DCN32 we do not have a pme workaround function defined that sends a
BacoAudio message. Default code had uses the DCN30 function for pme
workaround. PMFW headers are inconsistent with their message ID
definitions which cause ID's to clash leading to inconsistent system
behaviour. There is a clash with FCLK message due to inconsitent PMFW
headers.
[How]
Implement a new BacoAudio function to workaround the problem of
inconsistent PMFW headers in order to avoid BacoAudio message clasing
with FCLK Enable message.
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Chaitanya Dhere <chaitanya.dhere@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Add support to get VCO frequency from registers.
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Reviewed-by: Aurabindo Jayamohanan Pillai <Aurabindo.Pillai@amd.com>
Signed-off-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>