The commit cited below should not have been merged. It attemted to fix an
existing problem ansd thereby introduced new problems by keeping the
pi_state in state Q_REQUEUE_PI_IN_PROGRESS and leaking it.
Based on the commit description the intention was to handle the case
when task_blocks_on_rt_mutex() returns -EDEADLK and the following
remove_waiter() dereferences the NULL pointer in waiter->task.
That is already handled by Davidlohr in commit 40a25d59e8
("locking/rtmutex: Skip remove_waiter() when waiter is not enqueued") and
requires no further acting.
Revert the commit breaking the "waiter == owner" case again.
Fixes: 74e144274a ("futex/requeue: Prevent NULL pointer dereference in remove_waiter() on self-deadlock")
Reported-by: Michael Bommarito <michael.bommarito@gmail.com>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Cc: stable@vger.kernel.org
Link: https://patch.msgid.link/20260701131150.0Ijhq4Dw@linutronix.de
Closes: https://lore.kernel.org/all/20260629020049.2082397-1-michael.bommarito@gmail.com
Pull locking updates from Ingo Molnar:
"Futex updates:
- Optimize futex hash bucket access patterns (Peter Zijlstra)
- Large series to address the robust futex unlock race for real, by
Thomas Gleixner:
"The robust futex unlock mechanism is racy in respect to the
clearing of the robust_list_head::list_op_pending pointer because
unlock and clearing the pointer are not atomic.
The race window is between the unlock and clearing the pending op
pointer. If the task is forced to exit in this window, exit will
access a potentially invalid pending op pointer when cleaning up
the robust list.
That happens if another task manages to unmap the object
containing the lock before the cleanup, which results in an UAF.
In the worst case this UAF can lead to memory corruption when
unrelated content has been mapped to the same address by the time
the access happens.
User space can't solve this problem without help from the kernel.
This series provides the kernel side infrastructure to help it
along:
1) Combined unlock, pointer clearing, wake-up for the
contended case
2) VDSO based unlock and pointer clearing helpers with a
fix-up function in the kernel when user space was interrupted
within the critical section.
... with help by André Almeida:
- Add a note about robust list race condition (André Almeida)
- Add self-tests for robust release operations (André Almeida)
Context analysis updates:
- Implement context analysis for 'struct rt_mutex'. (Bart Van Assche)
- Bump required Clang version to 23 (Marco Elver)
Guard infrastructure updates:
- Series to remove NULL check from unconditional guards (Dmitry
Ilvokhin)
Lockdep updates:
- Restore self-test migrate_disable() and sched_rt_mutex state on
PREEMPT_RT (Karl Mehltretter)
Membarriers updates:
- Use per-CPU mutexes for targeted commands (Aniket Gattani)
- Modernize membarrier_global_expedited with cleanup guards (Aniket
Gattani)
- Add rseq stress test for CFS throttle interactions (Aniket Gattani)
percpu-rwsems updates:
- Extract __percpu_up_read() to optimize inlining overhead (Dmitry
Ilvokhin)
Seqlocks updates:
- Allow UBSAN_ALIGNMENT to fail optimizing (Heiko Carstens)
Lock tracing:
- Add contended_release tracepoint to sleepable locks such as
mutexes, percpu-rwsems, rtmutexes, rwsems and semaphores (Dmitry
Ilvokhin)
MAINTAINERS updates:
- MAINTAINERS: Add RUST [SYNC] entry (Boqun Feng)
Misc updates and fixes by Randy Dunlap, YE WEI-HONG, Fabricio Parra,
Dmitry Ilvokhin and Peter Zijlstra"
* tag 'locking-core-2026-06-14' of gitolite.kernel.org:pub/scm/linux/kernel/git/tip/tip: (36 commits)
locking: Add contended_release tracepoint to sleepable locks
locking/percpu-rwsem: Extract __percpu_up_read()
tracing/lock: Remove unnecessary linux/sched.h include
futex: Optimize futex hash bucket access patterns
rust: sync: completion: Mark inline complete_all and wait_for_completion
MAINTAINERS: Add RUST [SYNC] entry
cleanup: Specify nonnull argument index
selftests: futex: Add tests for robust release operations
Documentation: futex: Add a note about robust list race condition
x86/vdso: Implement __vdso_futex_robust_try_unlock()
x86/vdso: Prepare for robust futex unlock support
futex: Provide infrastructure to plug the non contended robust futex unlock race
futex: Add robust futex unlock IP range
futex: Add support for unlocking robust futexes
futex: Cleanup UAPI defines
x86: Select ARCH_MEMORY_ORDER_TSO
uaccess: Provide unsafe_atomic_store_release_user()
futex: Provide UABI defines for robust list entry modifiers
futex: Move futex related mm_struct data into a struct
futex: Make futex_mm_init() void
...
Breno reported significant c2c HITM in a futex hash heavy workload.
It turns out that the hash bucket to private hash table reverse pointer
(futex_hash_bucket::priv) was to blame. Notably when the hash buckets are
heavily contended, the: 'fph = bh->priv;' load in futex_hash() will typically
miss and consequently become quite expensive.
Since this load in particular is quite superfluous, removing it is fairly
straight forward. However, removing it does not in fact achieve anything much.
The pain moves to the next user, notably: futex_hash_put().
Therefore rework the whole private hash refcounting to avoid needing this back
pointer (and removing it). Instead of passing around 'struct futex_hash_bucket
*hb', pass around a new structure that contains it and the related 'struct
futex_private_hash *fph' pointer in tandem.
Funnily this turns out to remove more code than it adds and significantly
improves futex hash performance (as measured by 'perf bench futex hash'):
SKL dual socket 112 threads:
Baseline Patched
shared (16k) 1571857 1641435 + 4.4%
autosize (512) 646390 903371 +39.7%
-b 256 464395 587014 +26.4%
-b 512 715687 995943 +39.2%
-b 1024 995085 1396328 +40.3%
-b 2048 1293114 1668395 +29.0%
-b 4096 2124438 2240228 + 5.5%
Zen3 dual socket 256 threads:
Baseline Patched
shared (16k) 1275840 1381279 + 8.2%
autosize (512) 1252745 1482179 +18.3%
-b 256 856274 955455 +11.5%
-b 512 1267490 1544010 +21.8%
-b 1024 1424013 1625424 +14.1%
-b 2048 1505181 1669342 +10.9%
-b 4096 1465993 1688932 +15.2%
AMD EPYC 9D64 (Zen4, single socket) 176 threads:
Baseline Patched Delta
shared (16k) 1,230,599 1,368,655 +11.2%
autosize (1024) 1,285,440 1,556,946 +21.1%
-b 256 1,341,471 1,520,303 +13.3%
-b 512 1,438,330 1,599,319 +11.2%
-b 1024 1,443,772 1,622,493 +12.4%
-b 2048 1,472,108 1,643,975 +11.7%
-b 4096 1,333,098 1,570,897 +17.8%
Reported-by: Breno Leitao <leitao@debian.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Breno Leitao <leitao@debian.org>
Tested-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260610135510.GB1430057@noisy.programming.kicks-ass.net
When the FUTEX_ROBUST_UNLOCK mechanism is used for unlocking (PI-)futexes,
then the unlock sequence in user space looks like this:
1) robust_list_set_op_pending(mutex);
2) robust_list_remove(mutex);
lval = gettid();
3) if (atomic_try_cmpxchg(&mutex->lock, lval, 0))
4) robust_list_clear_op_pending();
else
5) sys_futex(OP | FUTEX_ROBUST_UNLOCK, ....);
That still leaves a minimal race window between #3 and #4 where the mutex
could be acquired by some other task, which observes that it is the last
user and:
1) unmaps the mutex memory
2) maps a different file, which ends up covering the same address
When then the original task exits before reaching #5 then the kernel robust
list handling observes the pending op entry and tries to fix up user space.
In case that the newly mapped data contains the TID of the exiting thread
at the address of the mutex/futex the kernel will set the owner died bit in
that memory and therefore corrupt unrelated data.
On X86 this boils down to this simplified assembly sequence:
mov %esi,%eax // Load TID into EAX
xor %ecx,%ecx // Set ECX to 0
#3 lock cmpxchg %ecx,(%rdi) // Try the TID -> 0 transition
.Lstart:
jnz .Lend
#4 movq %rcx,(%rdx) // Clear list_op_pending
.Lend:
If the cmpxchg() succeeds and the task is interrupted before it can clear
list_op_pending in the robust list head (#4) and the task crashes in a
signal handler or gets killed then it ends up in do_exit() and subsequently
in the robust list handling, which then might run into the unmap/map issue
described above.
This is only relevant when user space was interrupted and a signal is
pending. The fix-up has to be done before signal delivery is attempted
because:
1) The signal might be fatal so get_signal() ends up in do_exit()
2) The signal handler might crash or the task is killed before returning
from the handler. At that point the instruction pointer in pt_regs is
not longer the instruction pointer of the initially interrupted unlock
sequence.
The right place to handle this is in __exit_to_user_mode_loop() before
invoking arch_do_signal_or_restart() as this covers obviously both
scenarios.
As this is only relevant when the task was interrupted in user space, this
is tied to RSEQ and the generic entry code as RSEQ keeps track of user
space interrupts unconditionally even if the task does not have a RSEQ
region installed. That makes the decision very lightweight:
if (current->rseq.user_irq && within(regs, csr->unlock_ip_range))
futex_fixup_robust_unlock(regs, csr);
futex_fixup_robust_unlock() then invokes a architecture specific function
to return the pending op pointer or NULL. The function evaluates the
register content to decide whether the pending ops pointer in the robust
list head needs to be cleared.
Assuming the above unlock sequence, then on x86 this decision is the
trivial evaluation of the zero flag:
return regs->eflags & X86_EFLAGS_ZF ? regs->dx : NULL;
Other architectures might need to do more complex evaluations due to LLSC,
but the approach is valid in general. The size of the pointer is determined
from the matching range struct, which covers both 32-bit and 64-bit builds
including COMPAT.
The unlock sequence is going to be placed in the VDSO so that the kernel
can keep everything synchronized, especially the register usage. The
resulting code sequence for user space is:
if (__vdso_futex_robust_list$SZ_try_unlock(lock, tid, &pending_op) != tid)
err = sys_futex($OP | FUTEX_ROBUST_UNLOCK,....);
Both the VDSO unlock and the kernel side unlock ensure that the pending_op
pointer is always cleared when the lock becomes unlocked.
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: André Almeida <andrealmeid@igalia.com>
Link: https://patch.msgid.link/20260602090535.773669210@kernel.org
There will be a VDSO function to unlock robust futexes in user space. The
unlock sequence is racy vs. clearing the list_pending_op pointer in the
tasks robust list head. To plug this race the kernel needs to know the
instruction window. As the VDSO is per MM the addresses are stored in
mm_struct::futex.
Architectures which implement support for this have to update these
addresses when the VDSO is (re)mapped and indicate the pending op pointer
size which is matching the IP.
Arguably this could be resolved by chasing mm->context->vdso->image, but
that's architecture specific and requires to touch quite some cache
lines. Having it in mm::futex reduces the cache line impact and avoids
having yet another set of architecture specific functionality.
To support multi size robust list applications (gaming) this provides two
ranges when COMPAT is enabled.
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: André Almeida <andrealmeid@igalia.com>
Link: https://patch.msgid.link/20260602090535.718926819@kernel.org
Unlocking robust non-PI futexes happens in user space with the following
sequence:
1) robust_list_set_op_pending(mutex);
2) robust_list_remove(mutex);
lval = 0;
3) lval = atomic_xchg(lock, lval);
4) if (lval & WAITERS)
5) sys_futex(WAKE,....);
6) robust_list_clear_op_pending();
That opens a window between #3 and #6 where the mutex could be acquired by
some other task which observes that it is the last user and:
A) unmaps the mutex memory
B) maps a different file, which ends up covering the same address
When the original task exits before reaching #6 then the kernel robust list
handling observes the pending op entry and tries to fix up user space.
In case that the newly mapped data contains the TID of the exiting thread
at the address of the mutex/futex the kernel will set the owner died bit in
that memory and therefore corrupting unrelated data.
PI futexes have a similar problem both for the non-contented user space
unlock and the in kernel unlock:
1) robust_list_set_op_pending(mutex);
2) robust_list_remove(mutex);
lval = gettid();
3) if (!atomic_try_cmpxchg(lock, lval, 0))
4) sys_futex(UNLOCK_PI,....);
5) robust_list_clear_op_pending();
Address the first part of the problem where the futexes have waiters and
need to enter the kernel anyway. Add a new FUTEX_ROBUST_UNLOCK flag, which
is valid for the sys_futex() FUTEX_UNLOCK_PI, FUTEX_WAKE, FUTEX_WAKE_BITSET
operations.
This deliberately omits FUTEX_WAKE_OP from this treatment as it's unclear
whether this is needed and there is no usage of it in glibc either to
investigate.
For the futex2 syscall family this needs to be implemented with a new
syscall.
The sys_futex() case [ab]uses the @uaddr2 argument to hand the pointer to
robust_list_head::list_pending_op into the kernel. This argument is only
evaluated when the FUTEX_ROBUST_UNLOCK bit is set and is therefore backward
compatible.
This is an explicit argument to avoid the lookup of the robust list pointer
and retrieving the pending op pointer from there. User space has the
pointer already available so it can just put it into the @uaddr2
argument. Aside of that this allows the usage of multiple robust lists in
the future without any changes to the internal functions as they just operate
on the provided pointer.
This requires a second flag FUTEX_ROBUST_LIST32 which indicates that the
robust list pointer points to an u32 and not to an u64. This is required
for two reasons:
1) sys_futex() has no compat variant
2) The gaming emulators use both both 64-bit and compat 32-bit robust
lists in the same 64-bit application
As a consequence 32-bit applications have to set this flag unconditionally
so they can run on a 64-bit kernel in compat mode unmodified. 32-bit
kernels return an error code when the flag is not set. 64-bit kernels will
happily clear the full 64 bits if user space fails to set it.
In case of FUTEX_UNLOCK_PI this clears the robust list pending op when the
unlock succeeded. In case of errors, the user space value is still locked
by the caller and therefore the above cannot happen.
In case of FUTEX_WAKE* this does the unlock of the futex in the kernel and
clears the robust list pending op when the unlock was successful. If not,
the user space value is still locked and user space has to deal with the
returned error. That means that the unlocking of non-PI robust futexes has
to use the same try_cmpxchg() unlock scheme as PI futexes.
If the clearing of the pending list op fails (fault) then the kernel clears
the registered robust list pointer if it matches to prevent that exit()
will try to handle invalid data. That's a valid paranoid decision because
the robust list head sits usually in the TLS and if the TLS is not longer
accessible then the chance for fixing up the resulting mess is very close
to zero.
The problem of non-contended unlocks still exists and will be addressed
separately.
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: André Almeida <andrealmeid@igalia.com>
Link: https://patch.msgid.link/20260602090535.670514505@kernel.org
The marker for PI futexes in the robust list is a hardcoded 0x1 which lacks
any sensible form of documentation.
Provide proper defines for the bit and the mask and fix up the usage
sites. Thereby convert the boolean pi argument into a modifier argument,
which allows new modifier bits to be trivially added and conveyed.
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Reviewed-by: André Almeida <andrealmeid@igalia.com>
Link: https://patch.msgid.link/20260602090535.458758556@kernel.org
Having all these members in mm_struct along with the required #ifdeffery is
annoying, does not allow efficient initializing of the data with
memset() and makes extending it tedious.
Move it into a data structure and fix up all usage sites.
The extra struct for the private hash is intentional to make integration of
other conditional mechanisms easier in terms of initialization and separation.
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260602090535.407756793@kernel.org
When FUTEX_CMP_REQUEUE_PI requeues a non-top waiter that already owns the
target PI futex, task_blocks_on_rt_mutex() returns -EDEADLK before setting
waiter->task.
The subsequent remove_waiter() in rt_mutex_start_proxy_lock() dereferences
the NULL waiter->task, causing a kernel crash.
Add a self-deadlock check for non-top waiters before calling
rt_mutex_start_proxy_lock(), analogous to the top-waiter check in
futex_lock_pi_atomic().
Fixes: 3bfdc63936 ("rtmutex: Use waiter::task instead of current in remove_waiter()")
Signed-off-by: Ji'an Zhou <eilaimemedsnaimel@gmail.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Cc: stable@vger.kernel.org
During wait-requeue-pi (task A) and requeue-PI (task B) the following
race can happen:
Task A Task B
futex_wait_requeue_pi()
futex_setup_timer()
futex_do_wait()
futex_requeue()
CLASS(hb, hb1)(&key1);
CLASS(hb, hb2)(&key2);
*timeout*
futex_requeue_pi_wakeup_sync()
requeue_state = Q_REQUEUE_PI_IGNORE
*blocks on hb->lock*
futex_proxy_trylock_atomic()
futex_requeue_pi_prepare()
Q_REQUEUE_PI_IGNORE => -EAGAIN
double_unlock_hb(hb1, hb2)
*retry*
Task B acquires both hb locks and attempts to acquire the PI-lock of the
top most waiter (task B). Task A is leaving early due to a signal/
timeout and started removing itself from the queue. It updates its
requeue_state but can not remove it from the list because this requires
the hb lock which is owned by task B.
Usually task A is able to swoop the lock after task B unlocked it.
However if task B is of higher priority then task A may not be able to
wake up in time and acquire the lock before task B gets it again.
Especially on a UP system where A is never scheduled.
As a result task A blocks on the lock and task B busy loops, trying to
make progress but live locks the system instead. Tragic.
This can be fixed by removing the top most waiter from the list in this
case. This allows task B to grab the next top waiter (if any) in the
next iteration and make progress.
Remove the top most waiter if futex_requeue_pi_prepare() fails.
Let the waiter conditionally remove itself from the list in
handle_early_requeue_pi_wakeup().
Fixes: 07d91ef510 ("futex: Prevent requeue_pi() lock nesting issue on RT")
Reported-by: Moritz Klammler <Moritz.Klammler@ferchau.com>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260428103425.dywXyPd3@linutronix.de
Closes: https://lore.kernel.org/all/VE1PR06MB6894BE61C173D802365BE19DFF4CA@VE1PR06MB6894.eurprd06.prod.outlook.com
Pull locking updates from Ingo Molnar:
"Mutexes:
- Add killable flavor to guard definitions (Davidlohr Bueso)
- Remove the list_head from struct mutex (Matthew Wilcox)
- Rename mutex_init_lockep() (Davidlohr Bueso)
rwsems:
- Remove the list_head from struct rw_semaphore and
replace it with a single pointer (Matthew Wilcox)
- Fix logic error in rwsem_del_waiter() (Andrei Vagin)
Semaphores:
- Remove the list_head from struct semaphore (Matthew Wilcox)
Jump labels:
- Use ATOMIC_INIT() for initialization of .enabled (Thomas Weißschuh)
- Remove workaround for old compilers in initializations
(Thomas Weißschuh)
Lock context analysis changes and improvements:
- Add context analysis for rwsems (Peter Zijlstra)
- Fix rwlock and spinlock lock context annotations (Bart Van Assche)
- Fix rwlock support in <linux/spinlock_up.h> (Bart Van Assche)
- Add lock context annotations in the spinlock implementation
(Bart Van Assche)
- signal: Fix the lock_task_sighand() annotation (Bart Van Assche)
- ww-mutex: Fix the ww_acquire_ctx function annotations
(Bart Van Assche)
- Add lock context support in do_raw_{read,write}_trylock()
(Bart Van Assche)
- arm64, compiler-context-analysis: Permit alias analysis through
__READ_ONCE() with CONFIG_LTO=y (Marco Elver)
- Add __cond_releases() (Peter Zijlstra)
- Add context analysis for mutexes (Peter Zijlstra)
- Add context analysis for rtmutexes (Peter Zijlstra)
- Convert futexes to compiler context analysis (Peter Zijlstra)
Rust integration updates:
- Add atomic fetch_sub() implementation (Andreas Hindborg)
- Refactor various rust_helper_ methods for expansion (Boqun Feng)
- Add Atomic<*{mut,const} T> support (Boqun Feng)
- Add atomic operation helpers over raw pointers (Boqun Feng)
- Add performance-optimal Flag type for atomic booleans, to avoid
slow byte-sized RMWs on architectures that don't support them.
(FUJITA Tomonori)
- Misc cleanups and fixes (Andreas Hindborg, Boqun Feng, FUJITA
Tomonori)
LTO support updates:
- arm64: Optimize __READ_ONCE() with CONFIG_LTO=y (Marco Elver)
- compiler: Simplify generic RELOC_HIDE() (Marco Elver)
Miscellaneous fixes and cleanups by Peter Zijlstra, Randy Dunlap,
Thomas Weißschuh, Davidlohr Bueso and Mikhail Gavrilov"
* tag 'locking-core-2026-04-13' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (39 commits)
compiler: Simplify generic RELOC_HIDE()
locking: Add lock context annotations in the spinlock implementation
locking: Add lock context support in do_raw_{read,write}_trylock()
locking: Fix rwlock support in <linux/spinlock_up.h>
lockdep: Raise default stack trace limits when KASAN is enabled
cleanup: Optimize guards
jump_label: remove workaround for old compilers in initializations
jump_label: use ATOMIC_INIT() for initialization of .enabled
futex: Convert to compiler context analysis
locking/rwsem: Fix logic error in rwsem_del_waiter()
locking/rwsem: Add context analysis
locking/rtmutex: Add context analysis
locking/mutex: Add context analysis
compiler-context-analysys: Add __cond_releases()
locking/mutex: Remove the list_head from struct mutex
locking/semaphore: Remove the list_head from struct semaphore
locking/rwsem: Remove the list_head from struct rw_semaphore
rust: atomic: Update a safety comment in impl of `fetch_add()`
rust: sync: atomic: Update documentation for `fetch_add()`
rust: sync: atomic: Add fetch_sub()
...
Fuzzying/stressing futexes triggered:
WARNING: kernel/futex/core.c:825 at wait_for_owner_exiting+0x7a/0x80, CPU#11: futex_lock_pi_s/524
When futex_lock_pi_atomic() sees the owner is exiting, it returns -EBUSY
and stores a refcounted task pointer in 'exiting'.
After wait_for_owner_exiting() consumes that reference, the local pointer
is never reset to nil. Upon a retry, if futex_lock_pi_atomic() returns a
different error, the bogus pointer is passed to wait_for_owner_exiting().
CPU0 CPU1 CPU2
futex_lock_pi(uaddr)
// acquires the PI futex
exit()
futex_cleanup_begin()
futex_state = EXITING;
futex_lock_pi(uaddr)
futex_lock_pi_atomic()
attach_to_pi_owner()
// observes EXITING
*exiting = owner; // takes ref
return -EBUSY
wait_for_owner_exiting(-EBUSY, owner)
put_task_struct(); // drops ref
// exiting still points to owner
goto retry;
futex_lock_pi_atomic()
lock_pi_update_atomic()
cmpxchg(uaddr)
*uaddr ^= WAITERS // whatever
// value changed
return -EAGAIN;
wait_for_owner_exiting(-EAGAIN, exiting) // stale
WARN_ON_ONCE(exiting)
Fix this by resetting upon retry, essentially aligning it with requeue_pi.
Fixes: 3ef240eaff ("futex: Prevent exit livelock")
Signed-off-by: Davidlohr Bueso <dave@stgolabs.net>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Cc: stable@vger.kernel.org
Link: https://patch.msgid.link/20260326001759.4129680-1-dave@stgolabs.net
During futex_key_to_node_opt() execution, vma->vm_policy is read under
speculative mmap lock and RCU. Concurrently, mbind() may call
vma_replace_policy() which frees the old mempolicy immediately via
kmem_cache_free().
This creates a race where __futex_key_to_node() dereferences a freed
mempolicy pointer, causing a use-after-free read of mpol->mode.
[ 151.412631] BUG: KASAN: slab-use-after-free in __futex_key_to_node (kernel/futex/core.c:349)
[ 151.414046] Read of size 2 at addr ffff888001c49634 by task e/87
[ 151.415969] Call Trace:
[ 151.416732] __asan_load2 (mm/kasan/generic.c:271)
[ 151.416777] __futex_key_to_node (kernel/futex/core.c:349)
[ 151.416822] get_futex_key (kernel/futex/core.c:374 kernel/futex/core.c:386 kernel/futex/core.c:593)
Fix by adding rcu to __mpol_put().
Fixes: c042c50521 ("futex: Implement FUTEX2_MPOL")
Reported-by: Hao-Yu Yang <naup96721@gmail.com>
Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Hao-Yu Yang <naup96721@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Acked-by: David Hildenbrand (Arm) <david@kernel.org>
Link: https://patch.msgid.link/20260324174418.GB1850007@noisy.programming.kicks-ass.net
Nicholas reported that his LLM found it was possible to create a UaF
when sys_futex_requeue() is used with different flags. The initial
motivation for allowing different flags was the variable sized futex,
but since that hasn't been merged (yet), simply mandate the flags are
identical, as is the case for the old style sys_futex() requeue
operations.
Fixes: 0f4b5f9722 ("futex: Add sys_futex_requeue()")
Reported-by: Nicholas Carlini <npc@anthropic.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
This was done entirely with mindless brute force, using
git grep -l '\<k[vmz]*alloc_objs*(.*, GFP_KERNEL)' |
xargs sed -i 's/\(alloc_objs*(.*\), GFP_KERNEL)/\1)/'
to convert the new alloc_obj() users that had a simple GFP_KERNEL
argument to just drop that argument.
Note that due to the extreme simplicity of the scripting, any slightly
more complex cases spread over multiple lines would not be triggered:
they definitely exist, but this covers the vast bulk of the cases, and
the resulting diff is also then easier to check automatically.
For the same reason the 'flex' versions will be done as a separate
conversion.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This is the result of running the Coccinelle script from
scripts/coccinelle/api/kmalloc_objs.cocci. The script is designed to
avoid scalar types (which need careful case-by-case checking), and
instead replace kmalloc-family calls that allocate struct or union
object instances:
Single allocations: kmalloc(sizeof(TYPE), ...)
are replaced with: kmalloc_obj(TYPE, ...)
Array allocations: kmalloc_array(COUNT, sizeof(TYPE), ...)
are replaced with: kmalloc_objs(TYPE, COUNT, ...)
Flex array allocations: kmalloc(struct_size(PTR, FAM, COUNT), ...)
are replaced with: kmalloc_flex(*PTR, FAM, COUNT, ...)
(where TYPE may also be *VAR)
The resulting allocations no longer return "void *", instead returning
"TYPE *".
Signed-off-by: Kees Cook <kees@kernel.org>
Pull futex updates from Ingo Molnar:
- Standardize on ktime_t in restart_block::time as well (Thomas
Weißschuh)
- Futex selftests:
- Add robust list testcases (André Almeida)
- Formatting fixes/cleanups (Carlos Llamas)
* tag 'locking-futex-2025-12-10' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
futex: Store time as ktime_t in restart block
selftests/futex: Create test for robust list
selftests/futex: Skip tests if shmget unsupported
selftests/futex: Add newline to ksft_exit_fail_msg()
selftests/futex: Remove unused test_futex_mpol()
Pull scoped user access updates from Thomas Gleixner:
"Scoped user mode access and related changes:
- Implement the missing u64 user access function on ARM when
CONFIG_CPU_SPECTRE=n.
This makes it possible to access a 64bit value in generic code with
[unsafe_]get_user(). All other architectures and ARM variants
provide the relevant accessors already.
- Ensure that ASM GOTO jump label usage in the user mode access
helpers always goes through a local C scope label indirection
inside the helpers.
This is required because compilers are not supporting that a ASM
GOTO target leaves a auto cleanup scope. GCC silently fails to emit
the cleanup invocation and CLANG fails the build.
[ Editor's note: gcc-16 will have fixed the code generation issue
in commit f68fe3ddda4 ("eh: Invoke cleanups/destructors in asm
goto jumps [PR122835]"). But we obviously have to deal with clang
and older versions of gcc, so.. - Linus ]
This provides generic wrapper macros and the conversion of affected
architecture code to use them.
- Scoped user mode access with auto cleanup
Access to user mode memory can be required in hot code paths, but
if it has to be done with user controlled pointers, the access is
shielded with a speculation barrier, so that the CPU cannot
speculate around the address range check. Those speculation
barriers impact performance quite significantly.
This cost can be avoided by "masking" the provided pointer so it is
guaranteed to be in the valid user memory access range and
otherwise to point to a guaranteed unpopulated address space. This
has to be done without branches so it creates an address dependency
for the access, which the CPU cannot speculate ahead.
This results in repeating and error prone programming patterns:
if (can_do_masked_user_access())
from = masked_user_read_access_begin((from));
else if (!user_read_access_begin(from, sizeof(*from)))
return -EFAULT;
unsafe_get_user(val, from, Efault);
user_read_access_end();
return 0;
Efault:
user_read_access_end();
return -EFAULT;
which can be replaced with scopes and automatic cleanup:
scoped_user_read_access(from, Efault)
unsafe_get_user(val, from, Efault);
return 0;
Efault:
return -EFAULT;
- Convert code which implements the above pattern over to
scope_user.*.access(). This also corrects a couple of imbalanced
masked_*_begin() instances which are harmless on most
architectures, but prevent PowerPC from implementing the masking
optimization.
- Add a missing speculation barrier in copy_from_user_iter()"
* tag 'core-uaccess-2025-11-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
lib/strn*,uaccess: Use masked_user_{read/write}_access_begin when required
scm: Convert put_cmsg() to scoped user access
iov_iter: Add missing speculation barrier to copy_from_user_iter()
iov_iter: Convert copy_from_user_iter() to masked user access
select: Convert to scoped user access
x86/futex: Convert to scoped user access
futex: Convert to get/put_user_inline()
uaccess: Provide put/get_user_inline()
uaccess: Provide scoped user access regions
arm64: uaccess: Use unsafe wrappers for ASM GOTO
s390/uaccess: Use unsafe wrappers for ASM GOTO
riscv/uaccess: Use unsafe wrappers for ASM GOTO
powerpc/uaccess: Use unsafe wrappers for ASM GOTO
x86/uaccess: Use unsafe wrappers for ASM GOTO
uaccess: Provide ASM GOTO safe wrappers for unsafe_*_user()
ARM: uaccess: Implement missing __get_user_asm_dword()
Shrikanth noted that the per-cpu reference counter was still some 10%
slower than the old immutable option (which removes the reference
counting entirely).
Further optimize the per-cpu reference counter by:
- switching from RCU to preempt;
- using __this_cpu_*() since we now have preempt disabled;
- switching from smp_load_acquire() to READ_ONCE().
This is all safe because disabling preemption inhibits the RCU grace
period exactly like rcu_read_lock().
Having preemption disabled allows using __this_cpu_*() provided the
only access to the variable is in task context -- which is the case
here.
Furthermore, since we know changing fph->state to FR_ATOMIC demands a
full RCU grace period we can rely on the implied smp_mb() from that to
replace the acquire barrier().
This is very similar to the percpu_down_read_internal() fast-path.
The reason this is significant for PowerPC is that it uses the generic
this_cpu_*() implementation which relies on local_irq_disable() (the
x86 implementation relies on it being a single memop instruction to be
IRQ-safe). Switching to preempt_disable() and __this_cpu*() avoids
this IRQ state swizzling. Also, PowerPC needs LWSYNC for the ACQUIRE
barrier, not having to use explicit barriers safes a bunch.
Combined this reduces the performance gap by half, down to some 5%.
Fixes: 760e6f7bef ("futex: Remove support for IMMUTABLE")
Reported-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Tested-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://patch.msgid.link/20251106092929.GR4067720@noisy.programming.kicks-ass.net
Pull futex updates from Thomas Gleixner:
"A set of updates for futexes and related selftests:
- Plug the ptrace_may_access() race against a concurrent exec() which
allows to pass the check before the target's process transition in
exec() by taking a read lock on signal->ext_update_lock.
- A large set of cleanups and enhancement to the futex selftests. The
bulk of changes is the conversion to the kselftest harness"
* tag 'locking-futex-2025-09-29' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (25 commits)
selftest/futex: Fix spelling mistake "boundarie" -> "boundary"
selftests/futex: Remove logging.h file
selftests/futex: Drop logging.h include from futex_numa
selftests/futex: Refactor futex_numa_mpol with kselftest_harness.h
selftests/futex: Refactor futex_priv_hash with kselftest_harness.h
selftests/futex: Refactor futex_waitv with kselftest_harness.h
selftests/futex: Refactor futex_requeue with kselftest_harness.h
selftests/futex: Refactor futex_wait with kselftest_harness.h
selftests/futex: Refactor futex_wait_private_mapped_file with kselftest_harness.h
selftests/futex: Refactor futex_wait_unitialized_heap with kselftest_harness.h
selftests/futex: Refactor futex_wait_wouldblock with kselftest_harness.h
selftests/futex: Refactor futex_wait_timeout with kselftest_harness.h
selftests/futex: Refactor futex_requeue_pi_signal_restart with kselftest_harness.h
selftests/futex: Refactor futex_requeue_pi_mismatched_ops with kselftest_harness.h
selftests/futex: Refactor futex_requeue_pi with kselftest_harness.h
selftests: kselftest: Create ksft_print_dbg_msg()
futex: Don't leak robust_list pointer on exec race
selftest/futex: Compile also with libnuma < 2.0.16
selftest/futex: Reintroduce "Memory out of range" numa_mpol's subtest
selftest/futex: Make the error check more precise for futex_numa_mpol
...
sys_get_robust_list() and compat_get_robust_list() use ptrace_may_access()
to check if the calling task is allowed to access another task's
robust_list pointer. This check is racy against a concurrent exec() in the
target process.
During exec(), a task may transition from a non-privileged binary to a
privileged one (e.g., setuid binary) and its credentials/memory mappings
may change. If get_robust_list() performs ptrace_may_access() before
this transition, it may erroneously allow access to sensitive information
after the target becomes privileged.
A racy access allows an attacker to exploit a window during which
ptrace_may_access() passes before a target process transitions to a
privileged state via exec().
For example, consider a non-privileged task T that is about to execute a
setuid-root binary. An attacker task A calls get_robust_list(T) while T
is still unprivileged. Since ptrace_may_access() checks permissions
based on current credentials, it succeeds. However, if T begins exec
immediately afterwards, it becomes privileged and may change its memory
mappings. Because get_robust_list() proceeds to access T->robust_list
without synchronizing with exec() it may read user-space pointers from a
now-privileged process.
This violates the intended post-exec access restrictions and could
expose sensitive memory addresses or be used as a primitive in a larger
exploit chain. Consequently, the race can lead to unauthorized
disclosure of information across privilege boundaries and poses a
potential security risk.
Take a read lock on signal->exec_update_lock prior to invoking
ptrace_may_access() and accessing the robust_list/compat_robust_list.
This ensures that the target task's exec state remains stable during the
check, allowing for consistent and synchronized validation of
credentials.
Suggested-by: Jann Horn <jann@thejh.net>
Signed-off-by: Pranav Tyagi <pranav.tyagi03@gmail.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/linux-fsdevel/1477863998-3298-5-git-send-email-jann@thejh.net/
Link: https://github.com/KSPP/linux/issues/119
syzbot managed to trigger the following race:
T1 T2
futex_wait_requeue_pi()
futex_do_wait()
schedule()
futex_requeue()
futex_proxy_trylock_atomic()
futex_requeue_pi_prepare()
requeue_pi_wake_futex()
futex_requeue_pi_complete()
/* preempt */
* timeout/ signal wakes T1 *
futex_requeue_pi_wakeup_sync() // Q_REQUEUE_PI_LOCKED
futex_hash_put()
// back to userland, on stack futex_q is garbage
/* back */
wake_up_state(q->task, TASK_NORMAL);
In this scenario futex_wait_requeue_pi() is able to leave without using
futex_q::lock_ptr for synchronization.
This can be prevented by reading futex_q::task before updating the
futex_q::requeue_state. A reference on the task_struct is not needed
because requeue_pi_wake_futex() is invoked with a spinlock_t held which
implies a RCU read section.
Even if T1 terminates immediately after, the task_struct will remain valid
during T2's wake_up_state(). A READ_ONCE on futex_q::task before
futex_requeue_pi_complete() is enough because it ensures that the variable
is read before the state is updated.
Read futex_q::task before updating the requeue state, use it for the
following wakeup.
Fixes: 07d91ef510 ("futex: Prevent requeue_pi() lock nesting issue on RT")
Reported-by: syzbot+034246a838a10d181e78@syzkaller.appspotmail.com
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Closes: https://lore.kernel.org/all/68b75989.050a0220.3db4df.01dd.GAE@google.com/
To avoid a memory leak via mm_alloc() + mmdrop() the futex cleanup code
has been moved to __mmdrop(). This resulted in a warnings if the futex
hash table has been allocated via vmalloc() the mmdrop() was invoked
from atomic context.
The free path must stay in __mmput() to ensure it is invoked from
preemptible context.
In order to avoid the memory leak, delay the allocation of
mm_struct::mm->futex_ref to futex_hash_allocate(). This works because
neither the per-CPU counter nor the private hash has been allocated and
therefore
- futex_private_hash() callers (such as exit_pi_state_list()) don't
acquire reference if there is no private hash yet. There is also no
reference put.
- Regular callers (futex_hash()) fallback to global hash. No reference
counting here.
The futex_ref member can be allocated in futex_hash_allocate() before
the private hash itself is allocated. This happens either while the
first thread is created or on request. In both cases the process has
just a single thread so there can be either futex operation in progress
or the request to create a private hash.
Move futex_hash_free() back to __mmput();
Move the allocation of mm_struct::futex_ref to futex_hash_allocate().
[ bp: Fold a follow-up fix to prevent a use-after-free:
https://lore.kernel.org/r/20250830213806.sEKuuGSm@linutronix.de ]
Fixes: e703b7e247 ("futex: Move futex cleanup to __mmdrop()")
Closes: https://lore.kernel.org/all/20250821102721.6deae493@kernel.org/
Reported-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lkml.kernel.org/r/20250822141238.PfnkTjFb@linutronix.de
Commit cec199c5e3 ("futex: Implement FUTEX2_NUMA") introduced the
futex_put_value() helper to write a value to the given user
address.
However, it uses user_read_access_begin() before the write. For
architectures that differentiate between read and write accesses, like
PowerPC, futex_put_value() fails with -EFAULT.
Fix that by using the user_write_access_begin/user_write_access_end() pair
instead.
Fixes: cec199c5e3 ("futex: Implement FUTEX2_NUMA")
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/all/20250811141147.322261-1-longman@redhat.com
The FH_FLAG_IMMUTABLE flag was meant to avoid the reference counting on
the private hash and so to avoid the performance regression on big
machines.
With the switch to per-CPU counter this is no longer needed. That flag
was never useable on any released kernel.
Remove any support for IMMUTABLE while preserve the flags argument and
enforce it to be zero.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20250710110011.384614-5-bigeasy@linutronix.de
The use of rcuref_t for reference counting introduces a performance bottleneck
when accessed concurrently by multiple threads during futex operations.
Replace rcuref_t with special crafted per-CPU reference counters. The
lifetime logic remains the same.
The newly allocate private hash starts in FR_PERCPU state. In this state, each
futex operation that requires the private hash uses a per-CPU counter (an
unsigned int) for incrementing or decrementing the reference count.
When the private hash is about to be replaced, the per-CPU counters are
migrated to a atomic_t counter mm_struct::futex_atomic.
The migration process:
- Waiting for one RCU grace period to ensure all users observe the
current private hash. This can be skipped if a grace period elapsed
since the private hash was assigned.
- futex_private_hash::state is set to FR_ATOMIC, forcing all users to
use mm_struct::futex_atomic for reference counting.
- After a RCU grace period, all users are guaranteed to be using the
atomic counter. The per-CPU counters can now be summed up and added to
the atomic_t counter. If the resulting count is zero, the hash can be
safely replaced. Otherwise, active users still hold a valid reference.
- Once the atomic reference count drops to zero, the next futex
operation will switch to the new private hash.
call_rcu_hurry() is used to speed up transition which otherwise might be
delay with RCU_LAZY. There is nothing wrong with using call_rcu(). The
side effects would be that on auto scaling the new hash is used later
and the SET_SLOTS prctl() will block longer.
[bigeasy: commit description + mm get/ put_async]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20250710110011.384614-3-bigeasy@linutronix.de
Once the global hash is requested there is no way back to switch back to
the per-task private hash. This is checked at the begin of the function.
It is possible that two threads simultaneously request the global hash
and both pass the initial check and block later on the
mm::futex_hash_lock. In this case the first thread performs the switch
to the global hash. The second thread will also attempt to switch to the
global hash and while doing so, accessing the nonexisting slot 1 of the
struct futex_private_hash.
The same applies if the hash is made immutable: There is no reference
counting and the hash must not be replaced.
Verify under mm_struct::futex_phash that neither the global hash nor an
immutable hash in use.
Tested-by: "Lai, Yi" <yi1.lai@linux.intel.com>
Reported-by: "Lai, Yi" <yi1.lai@linux.intel.com>
Closes: https://lore.kernel.org/all/aDwDw9Aygqo6oAx+@ly-workstation/
Fixes: bd54df5ea7 ("futex: Allow to resize the private local hash")
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/all/20250610104400.1077266-5-bigeasy@linutronix.de/
Pull more MM updates from Andrew Morton:
- "zram: support algorithm-specific parameters" from Sergey Senozhatsky
adds infrastructure for passing algorithm-specific parameters into
zram. A single parameter `winbits' is implemented at this time.
- "memcg: nmi-safe kmem charging" from Shakeel Butt makes memcg
charging nmi-safe, which is required by BFP, which can operate in NMI
context.
- "Some random fixes and cleanup to shmem" from Kemeng Shi implements
small fixes and cleanups in the shmem code.
- "Skip mm selftests instead when kernel features are not present" from
Zi Yan fixes some issues in the MM selftest code.
- "mm/damon: build-enable essential DAMON components by default" from
SeongJae Park reworks DAMON Kconfig to make it easier to enable
CONFIG_DAMON.
- "sched/numa: add statistics of numa balance task migration" from Libo
Chen adds more info into sysfs and procfs files to improve visibility
into the NUMA balancer's task migration activity.
- "selftests/mm: cow and gup_longterm cleanups" from Mark Brown
provides various updates to some of the MM selftests to make them
play better with the overall containing framework.
* tag 'mm-stable-2025-06-01-14-06' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm: (43 commits)
mm/khugepaged: clean up refcount check using folio_expected_ref_count()
selftests/mm: fix test result reporting in gup_longterm
selftests/mm: report unique test names for each cow test
selftests/mm: add helper for logging test start and results
selftests/mm: use standard ksft_finished() in cow and gup_longterm
selftests/damon/_damon_sysfs: skip testcases if CONFIG_DAMON_SYSFS is disabled
sched/numa: add statistics of numa balance task
sched/numa: fix task swap by skipping kernel threads
tools/testing: check correct variable in open_procmap()
tools/testing/vma: add missing function stub
mm/gup: update comment explaining why gup_fast() disables IRQs
selftests/mm: two fixes for the pfnmap test
mm/khugepaged: fix race with folio split/free using temporary reference
mm: add CONFIG_PAGE_BLOCK_ORDER to select page block order
mmu_notifiers: remove leftover stub macros
selftests/mm: deduplicate test names in madv_populate
kcov: rust: add flags for KCOV with Rust
mm: rust: make CONFIG_MMU ifdefs more narrow
mmu_gather: move tlb flush for VM_PFNMAP/VM_MIXEDMAP vmas into free_pgtables()
mm/damon/Kconfig: enable CONFIG_DAMON by default
...
Extend the futex2 interface to be aware of mempolicy.
When FUTEX2_MPOL is specified and there is a MPOL_PREFERRED or
home_node specified covering the futex address, use that hash-map.
Notably, in this case the futex will go to the global node hashtable,
even if it is a PRIVATE futex.
When FUTEX2_NUMA|FUTEX2_MPOL is specified and the user specified node
value is FUTEX_NO_NODE, the MPOL lookup (as described above) will be
tried first before reverting to setting node to the local node.
[bigeasy: add CONFIG_FUTEX_MPOL, add MPOL to FUTEX2_VALID_MASK, write
the node only to user if FUTEX_NO_NODE was supplied]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20250416162921.513656-18-bigeasy@linutronix.de
Extend the futex2 interface to be numa aware.
When FUTEX2_NUMA is specified for a futex, the user value is extended
to two words (of the same size). The first is the user value we all
know, the second one will be the node to place this futex on.
struct futex_numa_32 {
u32 val;
u32 node;
};
When node is set to ~0, WAIT will set it to the current node_id such
that WAKE knows where to find it. If userspace corrupts the node value
between WAIT and WAKE, the futex will not be found and no wakeup will
happen.
When FUTEX2_NUMA is not set, the node is simply an extension of the
hash, such that traditional futexes are still interleaved over the
nodes.
This is done to avoid having to have a separate !numa hash-table.
[bigeasy: ensure to have at least hashsize of 4 in futex_init(), add
pr_info() for size and allocation information. Cast the naddr math to
void*]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20250416162921.513656-17-bigeasy@linutronix.de
My initial testing showed that:
perf bench futex hash
reported less operations/sec with private hash. After using the same
amount of buckets in the private hash as used by the global hash then
the operations/sec were about the same.
This changed once the private hash became resizable. This feature added
an RCU section and reference counting via atomic inc+dec operation into
the hot path.
The reference counting can be avoided if the private hash is made
immutable.
Extend PR_FUTEX_HASH_SET_SLOTS by a fourth argument which denotes if the
private should be made immutable. Once set (to true) the a further
resize is not allowed (same if set to global hash).
Add PR_FUTEX_HASH_GET_IMMUTABLE which returns true if the hash can not
be changed.
Update "perf bench" suite.
For comparison, results of "perf bench futex hash -s":
- Xeon CPU E5-2650, 2 NUMA nodes, total 32 CPUs:
- Before the introducing task local hash
shared Averaged 1.487.148 operations/sec (+- 0,53%), total secs = 10
private Averaged 2.192.405 operations/sec (+- 0,07%), total secs = 10
- With the series
shared Averaged 1.326.342 operations/sec (+- 0,41%), total secs = 10
-b128 Averaged 141.394 operations/sec (+- 1,15%), total secs = 10
-Ib128 Averaged 851.490 operations/sec (+- 0,67%), total secs = 10
-b8192 Averaged 131.321 operations/sec (+- 2,13%), total secs = 10
-Ib8192 Averaged 1.923.077 operations/sec (+- 0,61%), total secs = 10
128 is the default allocation of hash buckets.
8192 was the previous amount of allocated hash buckets.
- Xeon(R) CPU E7-8890 v3, 4 NUMA nodes, total 144 CPUs:
- Before the introducing task local hash
shared Averaged 1.810.936 operations/sec (+- 0,26%), total secs = 20
private Averaged 2.505.801 operations/sec (+- 0,05%), total secs = 20
- With the series
shared Averaged 1.589.002 operations/sec (+- 0,25%), total secs = 20
-b1024 Averaged 42.410 operations/sec (+- 0,20%), total secs = 20
-Ib1024 Averaged 740.638 operations/sec (+- 1,51%), total secs = 20
-b65536 Averaged 48.811 operations/sec (+- 1,35%), total secs = 20
-Ib65536 Averaged 1.963.165 operations/sec (+- 0,18%), total secs = 20
1024 is the default allocation of hash buckets.
65536 was the previous amount of allocated hash buckets.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Link: https://lore.kernel.org/r/20250416162921.513656-16-bigeasy@linutronix.de
The mm_struct::futex_hash_lock guards the futex_hash_bucket assignment/
replacement. The futex_hash_allocate()/ PR_FUTEX_HASH_SET_SLOTS
operation can now be invoked at runtime and resize an already existing
internal private futex_hash_bucket to another size.
The reallocation is based on an idea by Thomas Gleixner: The initial
allocation of struct futex_private_hash sets the reference count
to one. Every user acquires a reference on the local hash before using
it and drops it after it enqueued itself on the hash bucket. There is no
reference held while the task is scheduled out while waiting for the
wake up.
The resize process allocates a new struct futex_private_hash and drops
the initial reference. Synchronized with mm_struct::futex_hash_lock it
is checked if the reference counter for the currently used
mm_struct::futex_phash is marked as DEAD. If so, then all users enqueued
on the current private hash are requeued on the new private hash and the
new private hash is set to mm_struct::futex_phash. Otherwise the newly
allocated private hash is saved as mm_struct::futex_phash_new and the
rehashing and reassigning is delayed to the futex_hash() caller once the
reference counter is marked DEAD.
The replacement is not performed at rcuref_put() time because certain
callers, such as futex_wait_queue(), drop their reference after changing
the task state. This change will be destroyed once the futex_hash_lock
is acquired.
The user can change the number slots with PR_FUTEX_HASH_SET_SLOTS
multiple times. An increase and decrease is allowed and request blocks
until the assignment is done.
The private hash allocated at thread creation is changed from 16 to
16 <= 4 * number_of_threads <= global_hash_size
where number_of_threads can not exceed the number of online CPUs. Should
the user PR_FUTEX_HASH_SET_SLOTS then the auto scaling is disabled.
[peterz: reorganize the code to avoid state tracking and simplify new
object handling, block the user until changes are in effect, allow
increase and decrease of the hash].
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20250416162921.513656-15-bigeasy@linutronix.de
The futex hash is system wide and shared by all tasks. Each slot
is hashed based on futex address and the VMA of the thread. Due to
randomized VMAs (and memory allocations) the same logical lock (pointer)
can end up in a different hash bucket on each invocation of the
application. This in turn means that different applications may share a
hash bucket on the first invocation but not on the second and it is not
always clear which applications will be involved. This can result in
high latency's to acquire the futex_hash_bucket::lock especially if the
lock owner is limited to a CPU and can not be effectively PI boosted.
Introduce basic infrastructure for process local hash which is shared by
all threads of process. This hash will only be used for a
PROCESS_PRIVATE FUTEX operation.
The hashmap can be allocated via:
prctl(PR_FUTEX_HASH, PR_FUTEX_HASH_SET_SLOTS, num);
A `num' of 0 means that the global hash is used instead of a private
hash.
Other values for `num' specify the number of slots for the hash and the
number must be power of two, starting with two.
The prctl() returns zero on success. This function can only be used
before a thread is created.
The current status for the private hash can be queried via:
num = prctl(PR_FUTEX_HASH, PR_FUTEX_HASH_GET_SLOTS);
which return the current number of slots. The value 0 means that the
global hash is used. Values greater than 0 indicate the number of slots
that are used. A negative number indicates an error.
For optimisation, for the private hash jhash2() uses only two arguments
the address and the offset. This omits the VMA which is always the same.
[peterz: Use 0 for global hash. A bit shuffling and renaming. ]
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20250416162921.513656-13-bigeasy@linutronix.de
futex_lock_pi() and __fixup_pi_state_owner() acquire the
futex_q::lock_ptr without holding a reference assuming the previously
obtained hash bucket and the assigned lock_ptr are still valid. This
isn't the case once the private hash can be resized and becomes invalid
after the reference drop.
Introduce futex_q_lockptr_lock() to lock the hash bucket recorded in
futex_q::lock_ptr. The lock pointer is read in a RCU section to ensure
that it does not go away if the hash bucket has been replaced and the
old pointer has been observed. After locking the pointer needs to be
compared to check if it changed. If so then the hash bucket has been
replaced and the user has been moved to the new one and lock_ptr has
been updated. The lock operation needs to be redone in this case.
The locked hash bucket is not returned.
A special case is an early return in futex_lock_pi() (due to signal or
timeout) and a successful futex_wait_requeue_pi(). In both cases a valid
futex_q::lock_ptr is expected (and its matching hash bucket) but since
the waiter has been removed from the hash this can no longer be
guaranteed. Therefore before the waiter is removed and a reference is
acquired which is later dropped by the waiter to avoid a resize.
Add futex_q_lockptr_lock() and use it.
Acquire an additional reference in requeue_pi_wake_futex() and
futex_unlock_pi() while the futex_q is removed, denote this extra
reference in futex_q::drop_hb_ref and let the waiter drop the reference
in this case.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20250416162921.513656-11-bigeasy@linutronix.de
To support runtime resizing of the process private hash, it's required
to not use the obtained hash bucket once the reference count has been
dropped. The reference will be dropped after the unlock of the hash
bucket.
The amount of waiters is decremented after the unlock operation. There
is no requirement that this needs to happen after the unlock. The
increment happens before acquiring the lock to signal early that there
will be a waiter. The waiter can avoid blocking on the lock if it is
known that there will be no waiter.
There is no difference in terms of ordering if the decrement happens
before or after the unlock.
Decrease the waiter count before the unlock operation.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20250416162921.513656-10-bigeasy@linutronix.de
futex_wait_multiple_setup() changes task_struct::__state to
!TASK_RUNNING and then enqueues on multiple futexes. Every
futex_q_lock() acquires a reference on the global hash which is
dropped later.
If a rehash is in progress then the loop will block on
mm_struct::futex_hash_bucket for the rehash to complete and this will
lose the previously set task_struct::__state.
Acquire a reference on the local hash to avoiding blocking on
mm_struct::futex_hash_bucket.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20250416162921.513656-9-bigeasy@linutronix.de
