linux.git/kernel, branch v6.10 Linux kernel source tree Merge tag 'sched_urgent_for_v6.10' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2024-07-14T17:18:25+00:00 Linus Torvalds torvalds@linux-foundation.org 2024-07-14T17:18:25+00:00 365346980ee28e6792db08bc2faa80830c2878c4 Pull scheduler fixes from Borislav Petkov: - Fix a performance regression when measuring the CPU time of a thread (clock_gettime(CLOCK_THREAD_CPUTIME_ID,...)) due to the addition of PSI IRQ time accounting in the hotpath - Fix a task_struct leak due to missing to decrement the refcount when the task is enqueued before the timer which is supposed to do that, expires - Revert an attempt to expedite detaching of movable tasks, as finding those could become very costly. Turns out the original issue wasn't even hit by anyone * tag 'sched_urgent_for_v6.10' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: sched: Move psi_account_irqtime() out of update_rq_clock_task() hotpath sched/deadline: Fix task_struct reference leak Revert "sched/fair: Make sure to try to detach at least one movable task" 
Pull scheduler fixes from Borislav Petkov:

 - Fix a performance regression when measuring the CPU time of a thread
   (clock_gettime(CLOCK_THREAD_CPUTIME_ID,...)) due to the addition of
   PSI IRQ time accounting in the hotpath

 - Fix a task_struct leak due to missing to decrement the refcount when
   the task is enqueued before the timer which is supposed to do that,
   expires

 - Revert an attempt to expedite detaching of movable tasks, as finding
   those could become very costly. Turns out the original issue wasn't
   even hit by anyone

* tag 'sched_urgent_for_v6.10' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  sched: Move psi_account_irqtime() out of update_rq_clock_task() hotpath
  sched/deadline: Fix task_struct reference leak
  Revert "sched/fair: Make sure to try to detach at least one movable task"
bpf: Defer work in bpf_timer_cancel_and_free 2024-07-10T22:59:44+00:00 Kumar Kartikeya Dwivedi memxor@gmail.com 2024-07-09T18:54:39+00:00 a6fcd19d7eac1335eb76bc16b6a66b7f574d1d69 Currently, the same case as previous patch (two timer callbacks trying to cancel each other) can be invoked through bpf_map_update_elem as well, or more precisely, freeing map elements containing timers. Since this relies on hrtimer_cancel as well, it is prone to the same deadlock situation as the previous patch. It would be sufficient to use hrtimer_try_to_cancel to fix this problem, as the timer cannot be enqueued after async_cancel_and_free. Once async_cancel_and_free has been done, the timer must be reinitialized before it can be armed again. The callback running in parallel trying to arm the timer will fail, and freeing bpf_hrtimer without waiting is sufficient (given kfree_rcu), and bpf_timer_cb will return HRTIMER_NORESTART, preventing the timer from being rearmed again. However, there exists a UAF scenario where the callback arms the timer before entering this function, such that if cancellation fails (due to timer callback invoking this routine, or the target timer callback running concurrently). In such a case, if the timer expiration is significantly far in the future, the RCU grace period expiration happening before it will free the bpf_hrtimer state and along with it the struct hrtimer, that is enqueued. Hence, it is clear cancellation needs to occur after async_cancel_and_free, and yet it cannot be done inline due to deadlock issues. We thus modify bpf_timer_cancel_and_free to defer work to the global workqueue, adding a work_struct alongside rcu_head (both used at _different_ points of time, so can share space). Update existing code comments to reflect the new state of affairs. Fixes: b00628b1c7d5 ("bpf: Introduce bpf timers.") Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com> Link: https://lore.kernel.org/r/20240709185440.1104957-3-memxor@gmail.com Signed-off-by: Alexei Starovoitov <ast@kernel.org> 
Currently, the same case as previous patch (two timer callbacks trying
to cancel each other) can be invoked through bpf_map_update_elem as
well, or more precisely, freeing map elements containing timers. Since
this relies on hrtimer_cancel as well, it is prone to the same deadlock
situation as the previous patch.

It would be sufficient to use hrtimer_try_to_cancel to fix this problem,
as the timer cannot be enqueued after async_cancel_and_free. Once
async_cancel_and_free has been done, the timer must be reinitialized
before it can be armed again. The callback running in parallel trying to
arm the timer will fail, and freeing bpf_hrtimer without waiting is
sufficient (given kfree_rcu), and bpf_timer_cb will return
HRTIMER_NORESTART, preventing the timer from being rearmed again.

However, there exists a UAF scenario where the callback arms the timer
before entering this function, such that if cancellation fails (due to
timer callback invoking this routine, or the target timer callback
running concurrently). In such a case, if the timer expiration is
significantly far in the future, the RCU grace period expiration
happening before it will free the bpf_hrtimer state and along with it
the struct hrtimer, that is enqueued.

Hence, it is clear cancellation needs to occur after
async_cancel_and_free, and yet it cannot be done inline due to deadlock
issues. We thus modify bpf_timer_cancel_and_free to defer work to the
global workqueue, adding a work_struct alongside rcu_head (both used at
_different_ points of time, so can share space).

Update existing code comments to reflect the new state of affairs.

Fixes: b00628b1c7d5 ("bpf: Introduce bpf timers.")
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20240709185440.1104957-3-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
bpf: Fail bpf_timer_cancel when callback is being cancelled 2024-07-10T22:59:44+00:00 Kumar Kartikeya Dwivedi memxor@gmail.com 2024-07-09T18:54:38+00:00 d4523831f07a267a943f0dde844bf8ead7495f13 Given a schedule: timer1 cb timer2 cb bpf_timer_cancel(timer2); bpf_timer_cancel(timer1); Both bpf_timer_cancel calls would wait for the other callback to finish executing, introducing a lockup. Add an atomic_t count named 'cancelling' in bpf_hrtimer. This keeps track of all in-flight cancellation requests for a given BPF timer. Whenever cancelling a BPF timer, we must check if we have outstanding cancellation requests, and if so, we must fail the operation with an error (-EDEADLK) since cancellation is synchronous and waits for the callback to finish executing. This implies that we can enter a deadlock situation involving two or more timer callbacks executing in parallel and attempting to cancel one another. Note that we avoid incrementing the cancelling counter for the target timer (the one being cancelled) if bpf_timer_cancel is not invoked from a callback, to avoid spurious errors. The whole point of detecting cur->cancelling and returning -EDEADLK is to not enter a busy wait loop (which may or may not lead to a lockup). This does not apply in case the caller is in a non-callback context, the other side can continue to cancel as it sees fit without running into errors. Background on prior attempts: Earlier versions of this patch used a bool 'cancelling' bit and used the following pattern under timer->lock to publish cancellation status. lock(t->lock); t->cancelling = true; mb(); if (cur->cancelling) return -EDEADLK; unlock(t->lock); hrtimer_cancel(t->timer); t->cancelling = false; The store outside the critical section could overwrite a parallel requests t->cancelling assignment to true, to ensure the parallely executing callback observes its cancellation status. It would be necessary to clear this cancelling bit once hrtimer_cancel is done, but lack of serialization introduced races. Another option was explored where bpf_timer_start would clear the bit when (re)starting the timer under timer->lock. This would ensure serialized access to the cancelling bit, but may allow it to be cleared before in-flight hrtimer_cancel has finished executing, such that lockups can occur again. Thus, we choose an atomic counter to keep track of all outstanding cancellation requests and use it to prevent lockups in case callbacks attempt to cancel each other while executing in parallel. Reported-by: Dohyun Kim <dohyunkim@google.com> Reported-by: Neel Natu <neelnatu@google.com> Fixes: b00628b1c7d5 ("bpf: Introduce bpf timers.") Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com> Link: https://lore.kernel.org/r/20240709185440.1104957-2-memxor@gmail.com Signed-off-by: Alexei Starovoitov <ast@kernel.org> 
Given a schedule:

timer1 cb			timer2 cb

bpf_timer_cancel(timer2);	bpf_timer_cancel(timer1);

Both bpf_timer_cancel calls would wait for the other callback to finish
executing, introducing a lockup.

Add an atomic_t count named 'cancelling' in bpf_hrtimer. This keeps
track of all in-flight cancellation requests for a given BPF timer.
Whenever cancelling a BPF timer, we must check if we have outstanding
cancellation requests, and if so, we must fail the operation with an
error (-EDEADLK) since cancellation is synchronous and waits for the
callback to finish executing. This implies that we can enter a deadlock
situation involving two or more timer callbacks executing in parallel
and attempting to cancel one another.

Note that we avoid incrementing the cancelling counter for the target
timer (the one being cancelled) if bpf_timer_cancel is not invoked from
a callback, to avoid spurious errors. The whole point of detecting
cur->cancelling and returning -EDEADLK is to not enter a busy wait loop
(which may or may not lead to a lockup). This does not apply in case the
caller is in a non-callback context, the other side can continue to
cancel as it sees fit without running into errors.

Background on prior attempts:

Earlier versions of this patch used a bool 'cancelling' bit and used the
following pattern under timer->lock to publish cancellation status.

lock(t->lock);
t->cancelling = true;
mb();
if (cur->cancelling)
	return -EDEADLK;
unlock(t->lock);
hrtimer_cancel(t->timer);
t->cancelling = false;

The store outside the critical section could overwrite a parallel
requests t->cancelling assignment to true, to ensure the parallely
executing callback observes its cancellation status.

It would be necessary to clear this cancelling bit once hrtimer_cancel
is done, but lack of serialization introduced races. Another option was
explored where bpf_timer_start would clear the bit when (re)starting the
timer under timer->lock. This would ensure serialized access to the
cancelling bit, but may allow it to be cleared before in-flight
hrtimer_cancel has finished executing, such that lockups can occur
again.

Thus, we choose an atomic counter to keep track of all outstanding
cancellation requests and use it to prevent lockups in case callbacks
attempt to cancel each other while executing in parallel.

Reported-by: Dohyun Kim <dohyunkim@google.com>
Reported-by: Neel Natu <neelnatu@google.com>
Fixes: b00628b1c7d5 ("bpf: Introduce bpf timers.")
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20240709185440.1104957-2-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
bpf: fix order of args in call to bpf_map_kvcalloc 2024-07-10T22:31:19+00:00 Mohammad Shehar Yaar Tausif sheharyaar48@gmail.com 2024-07-10T10:05:22+00:00 af253aef183a31ce62d2e39fc520b0ebfb562bb9 The original function call passed size of smap->bucket before the number of buckets which raises the error 'calloc-transposed-args' on compilation. Vlastimil Babka added: The order of parameters can be traced back all the way to 6ac99e8f23d4 ("bpf: Introduce bpf sk local storage") accross several refactorings, and that's why the commit is used as a Fixes: tag. In v6.10-rc1, a different commit 2c321f3f70bc ("mm: change inlined allocation helpers to account at the call site") however exposed the order of args in a way that gcc-14 has enough visibility to start warning about it, because (in !CONFIG_MEMCG case) bpf_map_kvcalloc is then a macro alias for kvcalloc instead of a static inline wrapper. To sum up the warning happens when the following conditions are all met: - gcc-14 is used (didn't see it with gcc-13) - commit 2c321f3f70bc is present - CONFIG_MEMCG is not enabled in .config - CONFIG_WERROR turns this from a compiler warning to error Fixes: 6ac99e8f23d4 ("bpf: Introduce bpf sk local storage") Reviewed-by: Andrii Nakryiko <andrii@kernel.org> Tested-by: Christian Kujau <lists@nerdbynature.de> Signed-off-by: Mohammad Shehar Yaar Tausif <sheharyaar48@gmail.com> Signed-off-by: Vlastimil Babka <vbabka@suse.cz> Link: https://lore.kernel.org/r/20240710100521.15061-2-vbabka@suse.cz Signed-off-by: Alexei Starovoitov <ast@kernel.org> 
The original function call passed size of smap->bucket before the number of
buckets which raises the error 'calloc-transposed-args' on compilation.

Vlastimil Babka added:

The order of parameters can be traced back all the way to 6ac99e8f23d4
("bpf: Introduce bpf sk local storage") accross several refactorings,
and that's why the commit is used as a Fixes: tag.

In v6.10-rc1, a different commit 2c321f3f70bc ("mm: change inlined
allocation helpers to account at the call site") however exposed the
order of args in a way that gcc-14 has enough visibility to start
warning about it, because (in !CONFIG_MEMCG case) bpf_map_kvcalloc is
then a macro alias for kvcalloc instead of a static inline wrapper.

To sum up the warning happens when the following conditions are all met:

- gcc-14 is used (didn't see it with gcc-13)
- commit 2c321f3f70bc is present
- CONFIG_MEMCG is not enabled in .config
- CONFIG_WERROR turns this from a compiler warning to error

Fixes: 6ac99e8f23d4 ("bpf: Introduce bpf sk local storage")
Reviewed-by: Andrii Nakryiko <andrii@kernel.org>
Tested-by: Christian Kujau <lists@nerdbynature.de>
Signed-off-by: Mohammad Shehar Yaar Tausif <sheharyaar48@gmail.com>
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Link: https://lore.kernel.org/r/20240710100521.15061-2-vbabka@suse.cz
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
mm: optimize the redundant loop of mm_update_owner_next() 2024-07-03T19:29:24+00:00 Jinliang Zheng alexjlzheng@tencent.com 2024-06-20T12:21:24+00:00 cf3f9a593dab87a032d2b6a6fb205e7f3de4f0a1 When mm_update_owner_next() is racing with swapoff (try_to_unuse()) or /proc or ptrace or page migration (get_task_mm()), it is impossible to find an appropriate task_struct in the loop whose mm_struct is the same as the target mm_struct. If the above race condition is combined with the stress-ng-zombie and stress-ng-dup tests, such a long loop can easily cause a Hard Lockup in write_lock_irq() for tasklist_lock. Recognize this situation in advance and exit early. Link: https://lkml.kernel.org/r/20240620122123.3877432-1-alexjlzheng@tencent.com Signed-off-by: Jinliang Zheng <alexjlzheng@tencent.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Christian Brauner <brauner@kernel.org> Cc: Jens Axboe <axboe@kernel.dk> Cc: Mateusz Guzik <mjguzik@gmail.com> Cc: Matthew Wilcox (Oracle) <willy@infradead.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Tycho Andersen <tandersen@netflix.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> 
When mm_update_owner_next() is racing with swapoff (try_to_unuse()) or
/proc or ptrace or page migration (get_task_mm()), it is impossible to
find an appropriate task_struct in the loop whose mm_struct is the same as
the target mm_struct.

If the above race condition is combined with the stress-ng-zombie and
stress-ng-dup tests, such a long loop can easily cause a Hard Lockup in
write_lock_irq() for tasklist_lock.

Recognize this situation in advance and exit early.

Link: https://lkml.kernel.org/r/20240620122123.3877432-1-alexjlzheng@tencent.com
Signed-off-by: Jinliang Zheng <alexjlzheng@tencent.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Mateusz Guzik <mjguzik@gmail.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Tycho Andersen <tandersen@netflix.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
sched: Move psi_account_irqtime() out of update_rq_clock_task() hotpath 2024-07-01T11:01:44+00:00 John Stultz jstultz@google.com 2024-06-18T21:58:55+00:00 ddae0ca2a8fe12d0e24ab10ba759c3fbd755ada8 It was reported that in moving to 6.1, a larger then 10% regression was seen in the performance of clock_gettime(CLOCK_THREAD_CPUTIME_ID,...). Using a simple reproducer, I found: 5.10: 100000000 calls in 24345994193 ns => 243.460 ns per call 100000000 calls in 24288172050 ns => 242.882 ns per call 100000000 calls in 24289135225 ns => 242.891 ns per call 6.1: 100000000 calls in 28248646742 ns => 282.486 ns per call 100000000 calls in 28227055067 ns => 282.271 ns per call 100000000 calls in 28177471287 ns => 281.775 ns per call The cause of this was finally narrowed down to the addition of psi_account_irqtime() in update_rq_clock_task(), in commit 52b1364ba0b1 ("sched/psi: Add PSI_IRQ to track IRQ/SOFTIRQ pressure"). In my initial attempt to resolve this, I leaned towards moving all accounting work out of the clock_gettime() call path, but it wasn't very pretty, so it will have to wait for a later deeper rework. Instead, Peter shared this approach: Rework psi_account_irqtime() to use its own psi_irq_time base for accounting, and move it out of the hotpath, calling it instead from sched_tick() and __schedule(). In testing this, we found the importance of ensuring psi_account_irqtime() is run under the rq_lock, which Johannes Weiner helpfully explained, so also add some lockdep annotations to make that requirement clear. With this change the performance is back in-line with 5.10: 6.1+fix: 100000000 calls in 24297324597 ns => 242.973 ns per call 100000000 calls in 24318869234 ns => 243.189 ns per call 100000000 calls in 24291564588 ns => 242.916 ns per call Reported-by: Jimmy Shiu <jimmyshiu@google.com> Originally-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: John Stultz <jstultz@google.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Chengming Zhou <chengming.zhou@linux.dev> Reviewed-by: Qais Yousef <qyousef@layalina.io> Link: https://lore.kernel.org/r/20240618215909.4099720-1-jstultz@google.com 
It was reported that in moving to 6.1, a larger then 10%
regression was seen in the performance of
clock_gettime(CLOCK_THREAD_CPUTIME_ID,...).

Using a simple reproducer, I found:
5.10:
100000000 calls in 24345994193 ns => 243.460 ns per call
100000000 calls in 24288172050 ns => 242.882 ns per call
100000000 calls in 24289135225 ns => 242.891 ns per call

6.1:
100000000 calls in 28248646742 ns => 282.486 ns per call
100000000 calls in 28227055067 ns => 282.271 ns per call
100000000 calls in 28177471287 ns => 281.775 ns per call

The cause of this was finally narrowed down to the addition of
psi_account_irqtime() in update_rq_clock_task(), in commit
52b1364ba0b1 ("sched/psi: Add PSI_IRQ to track IRQ/SOFTIRQ
pressure").

In my initial attempt to resolve this, I leaned towards moving
all accounting work out of the clock_gettime() call path, but it
wasn't very pretty, so it will have to wait for a later deeper
rework. Instead, Peter shared this approach:

Rework psi_account_irqtime() to use its own psi_irq_time base
for accounting, and move it out of the hotpath, calling it
instead from sched_tick() and __schedule().

In testing this, we found the importance of ensuring
psi_account_irqtime() is run under the rq_lock, which Johannes
Weiner helpfully explained, so also add some lockdep annotations
to make that requirement clear.

With this change the performance is back in-line with 5.10:
6.1+fix:
100000000 calls in 24297324597 ns => 242.973 ns per call
100000000 calls in 24318869234 ns => 243.189 ns per call
100000000 calls in 24291564588 ns => 242.916 ns per call

Reported-by: Jimmy Shiu <jimmyshiu@google.com>
Originally-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Chengming Zhou <chengming.zhou@linux.dev>
Reviewed-by: Qais Yousef <qyousef@layalina.io>
Link: https://lore.kernel.org/r/20240618215909.4099720-1-jstultz@google.com
sched/deadline: Fix task_struct reference leak 2024-07-01T11:01:44+00:00 Wander Lairson Costa wander@redhat.com 2024-06-20T12:56:17+00:00 b58652db66c910c2245f5bee7deca41c12d707b9 During the execution of the following stress test with linux-rt: stress-ng --cyclic 30 --timeout 30 --minimize --quiet kmemleak frequently reported a memory leak concerning the task_struct: unreferenced object 0xffff8881305b8000 (size 16136): comm "stress-ng", pid 614, jiffies 4294883961 (age 286.412s) object hex dump (first 32 bytes): 02 40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 .@.............. 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ debug hex dump (first 16 bytes): 53 09 00 00 00 00 00 00 00 00 00 00 00 00 00 00 S............... backtrace: [<00000000046b6790>] dup_task_struct+0x30/0x540 [<00000000c5ca0f0b>] copy_process+0x3d9/0x50e0 [<00000000ced59777>] kernel_clone+0xb0/0x770 [<00000000a50befdc>] __do_sys_clone+0xb6/0xf0 [<000000001dbf2008>] do_syscall_64+0x5d/0xf0 [<00000000552900ff>] entry_SYSCALL_64_after_hwframe+0x6e/0x76 The issue occurs in start_dl_timer(), which increments the task_struct reference count and sets a timer. The timer callback, dl_task_timer, is supposed to decrement the reference count upon expiration. However, if enqueue_task_dl() is called before the timer expires and cancels it, the reference count is not decremented, leading to the leak. This patch fixes the reference leak by ensuring the task_struct reference count is properly decremented when the timer is canceled. Fixes: feff2e65efd8 ("sched/deadline: Unthrottle PI boosted threads while enqueuing") Signed-off-by: Wander Lairson Costa <wander@redhat.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Acked-by: Juri Lelli <juri.lelli@redhat.com> Link: https://lore.kernel.org/r/20240620125618.11419-1-wander@redhat.com 
During the execution of the following stress test with linux-rt:

stress-ng --cyclic 30 --timeout 30 --minimize --quiet

kmemleak frequently reported a memory leak concerning the task_struct:

unreferenced object 0xffff8881305b8000 (size 16136):
  comm "stress-ng", pid 614, jiffies 4294883961 (age 286.412s)
  object hex dump (first 32 bytes):
    02 40 00 00 00 00 00 00 00 00 00 00 00 00 00 00  .@..............
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
  debug hex dump (first 16 bytes):
    53 09 00 00 00 00 00 00 00 00 00 00 00 00 00 00  S...............
  backtrace:
    [<00000000046b6790>] dup_task_struct+0x30/0x540
    [<00000000c5ca0f0b>] copy_process+0x3d9/0x50e0
    [<00000000ced59777>] kernel_clone+0xb0/0x770
    [<00000000a50befdc>] __do_sys_clone+0xb6/0xf0
    [<000000001dbf2008>] do_syscall_64+0x5d/0xf0
    [<00000000552900ff>] entry_SYSCALL_64_after_hwframe+0x6e/0x76

The issue occurs in start_dl_timer(), which increments the task_struct
reference count and sets a timer. The timer callback, dl_task_timer,
is supposed to decrement the reference count upon expiration. However,
if enqueue_task_dl() is called before the timer expires and cancels it,
the reference count is not decremented, leading to the leak.

This patch fixes the reference leak by ensuring the task_struct
reference count is properly decremented when the timer is canceled.

Fixes: feff2e65efd8 ("sched/deadline: Unthrottle PI boosted threads while enqueuing")
Signed-off-by: Wander Lairson Costa <wander@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Juri Lelli <juri.lelli@redhat.com>
Link: https://lore.kernel.org/r/20240620125618.11419-1-wander@redhat.com
Revert "sched/fair: Make sure to try to detach at least one movable task" 2024-07-01T11:01:43+00:00 Josh Don joshdon@google.com 2024-06-20T21:44:50+00:00 2feab2492deb2f14f9675dd6388e9e2bf669c27a This reverts commit b0defa7ae03ecf91b8bfd10ede430cff12fcbd06. b0defa7ae03ec changed the load balancing logic to ignore env.max_loop if all tasks examined to that point were pinned. The goal of the patch was to make it more likely to be able to detach a task buried in a long list of pinned tasks. However, this has the unfortunate side effect of creating an O(n) iteration in detach_tasks(), as we now must fully iterate every task on a cpu if all or most are pinned. Since this load balance code is done with rq lock held, and often in softirq context, it is very easy to trigger hard lockups. We observed such hard lockups with a user who affined O(10k) threads to a single cpu. When I discussed this with Vincent he initially suggested that we keep the limit on the number of tasks to detach, but increase the number of tasks we can search. However, after some back and forth on the mailing list, he recommended we instead revert the original patch, as it seems likely no one was actually getting hit by the original issue. Fixes: b0defa7ae03e ("sched/fair: Make sure to try to detach at least one movable task") Signed-off-by: Josh Don <joshdon@google.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org> Link: https://lore.kernel.org/r/20240620214450.316280-1-joshdon@google.com 
This reverts commit b0defa7ae03ecf91b8bfd10ede430cff12fcbd06.

b0defa7ae03ec changed the load balancing logic to ignore env.max_loop if
all tasks examined to that point were pinned. The goal of the patch was
to make it more likely to be able to detach a task buried in a long list
of pinned tasks. However, this has the unfortunate side effect of
creating an O(n) iteration in detach_tasks(), as we now must fully
iterate every task on a cpu if all or most are pinned. Since this load
balance code is done with rq lock held, and often in softirq context, it
is very easy to trigger hard lockups. We observed such hard lockups with
a user who affined O(10k) threads to a single cpu.

When I discussed this with Vincent he initially suggested that we keep
the limit on the number of tasks to detach, but increase the number of
tasks we can search. However, after some back and forth on the mailing
list, he recommended we instead revert the original patch, as it seems
likely no one was actually getting hit by the original issue.

Fixes: b0defa7ae03e ("sched/fair: Make sure to try to detach at least one movable task")
Signed-off-by: Josh Don <joshdon@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20240620214450.316280-1-joshdon@google.com
Merge tag 'tty-6.10-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty 2024-06-30T15:57:43+00:00 Linus Torvalds torvalds@linux-foundation.org 2024-06-30T15:57:43+00:00 3e334486ec5cc6e79e7b0c4f58757fe8e05fbe5a Pull tty / serial / console fixes from Greg KH: "Here are a bunch of fixes/reverts for 6.10-rc6. Include in here are: - revert the bunch of tty/serial/console changes that landed in -rc1 that didn't quite work properly yet. Everyone agreed to just revert them for now and will work on making them better for a future release instead of trying to quick fix the existing changes this late in the release cycle - 8250 driver port count bugfix - Other tiny serial port bugfixes for reported issues All of these have been in linux-next this week with no reported issues" * tag 'tty-6.10-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty: Revert "printk: Save console options for add_preferred_console_match()" Revert "printk: Don't try to parse DEVNAME:0.0 console options" Revert "printk: Flag register_console() if console is set on command line" Revert "serial: core: Add support for DEVNAME:0.0 style naming for kernel console" Revert "serial: core: Handle serial console options" Revert "serial: 8250: Add preferred console in serial8250_isa_init_ports()" Revert "Documentation: kernel-parameters: Add DEVNAME:0.0 format for serial ports" Revert "serial: 8250: Fix add preferred console for serial8250_isa_init_ports()" Revert "serial: core: Fix ifdef for serial base console functions" serial: bcm63xx-uart: fix tx after conversion to uart_port_tx_limited() serial: core: introduce uart_port_tx_limited_flags() Revert "serial: core: only stop transmit when HW fifo is empty" serial: imx: set receiver level before starting uart tty: mcf: MCF54418 has 10 UARTS serial: 8250_omap: Implementation of Errata i2310 tty: serial: 8250: Fix port count mismatch with the device 
Pull tty / serial / console fixes from Greg KH:
 "Here are a bunch of fixes/reverts for 6.10-rc6.  Include in here are:

   - revert the bunch of tty/serial/console changes that landed in -rc1
     that didn't quite work properly yet.

     Everyone agreed to just revert them for now and will work on making
     them better for a future release instead of trying to quick fix the
     existing changes this late in the release cycle

   - 8250 driver port count bugfix

   - Other tiny serial port bugfixes for reported issues

  All of these have been in linux-next this week with no reported
  issues"

* tag 'tty-6.10-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:
  Revert "printk: Save console options for add_preferred_console_match()"
  Revert "printk: Don't try to parse DEVNAME:0.0 console options"
  Revert "printk: Flag register_console() if console is set on command line"
  Revert "serial: core: Add support for DEVNAME:0.0 style naming for kernel console"
  Revert "serial: core: Handle serial console options"
  Revert "serial: 8250: Add preferred console in serial8250_isa_init_ports()"
  Revert "Documentation: kernel-parameters: Add DEVNAME:0.0 format for serial ports"
  Revert "serial: 8250: Fix add preferred console for serial8250_isa_init_ports()"
  Revert "serial: core: Fix ifdef for serial base console functions"
  serial: bcm63xx-uart: fix tx after conversion to uart_port_tx_limited()
  serial: core: introduce uart_port_tx_limited_flags()
  Revert "serial: core: only stop transmit when HW fifo is empty"
  serial: imx: set receiver level before starting uart
  tty: mcf: MCF54418 has 10 UARTS
  serial: 8250_omap: Implementation of Errata i2310
  tty: serial: 8250: Fix port count mismatch with the device
Merge tag 'smp_urgent_for_v6.10_rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2024-06-30T15:41:42+00:00 Linus Torvalds torvalds@linux-foundation.org 2024-06-30T15:41:42+00:00 3ffea9a7a6f71b5fb6f997d7b2ad95634dae57ab Pull smp fixes from Borislav Petkov: - Fix "nosmp" and "maxcpus=0" after the parallel CPU bringup work went in and broke them - Make sure CPU hotplug dynamic prepare states are actually executed * tag 'smp_urgent_for_v6.10_rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: cpu: Fix broken cmdline "nosmp" and "maxcpus=0" cpu/hotplug: Fix dynstate assignment in __cpuhp_setup_state_cpuslocked() 
Pull smp fixes from Borislav Petkov:

 - Fix "nosmp" and "maxcpus=0" after the parallel CPU bringup work went
   in and broke them

 - Make sure CPU hotplug dynamic prepare states are actually executed

* tag 'smp_urgent_for_v6.10_rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  cpu: Fix broken cmdline "nosmp" and "maxcpus=0"
  cpu/hotplug: Fix dynstate assignment in __cpuhp_setup_state_cpuslocked()