| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Fix those:
./kernel/futex/futex.h:208: warning: Function parameter or struct member 'drop_hb_ref' not described in 'futex_q'
./kernel/futex/waitwake.c:343: warning: expecting prototype for futex_wait_queue(). Prototype was for futex_do_wait() instead
./kernel/futex/waitwake.c:594: warning: Function parameter or struct member 'task' not described in 'futex_wait_setup'
Fixes: 93f1b6d79a73 ("futex: Move futex_queue() into futex_wait_setup()")
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20250512185641.0450a99b@canb.auug.org.au # report
Link: https://lore.kernel.org/r/20250515171641.24073-1-bp@kernel.org # submission
|
|
The following commit added an rcu_assign_pointer() assignment to
futex_mm_init() in <linux/futex.h>:
bd54df5ea7ca ("futex: Allow to resize the private local hash")
Which breaks the build on older compilers (gcc-9, x86-64 defconfig):
CC io_uring/futex.o
In file included from ./arch/x86/include/generated/asm/rwonce.h:1,
from ./include/linux/compiler.h:390,
from ./include/linux/array_size.h:5,
from ./include/linux/kernel.h:16,
from io_uring/futex.c:2:
./include/linux/futex.h: In function 'futex_mm_init':
./include/linux/rcupdate.h:555:36: error: dereferencing pointer to incomplete type 'struct futex_private_hash'
The problem is that this variant of rcu_assign_pointer() wants to
know the full type of 'struct futex_private_hash', which type
is local to futex.c:
kernel/futex/core.c:struct futex_private_hash {
There are a couple of mechanical solutions for this bug:
- we can uninline futex_mm_init() and move it into futex/core.c
- or we can share the structure definition with kernel/fork.c.
But both of these solutions have disadvantages: the first one adds
runtime overhead, while the second one dis-encapsulates private
futex types.
A third solution, implemented by this patch, is to just initialize
mm->futex_phash with NULL like the patch below, it's not like this
new MM's ->futex_phash can be observed externally until the task
is inserted into the task list, which guarantees full store ordering.
The relaxation of this initialization might also give a tiny speedup
on certain platforms.
Fixes: bd54df5ea7ca ("futex: Allow to resize the private local hash")
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: André Almeida <andrealmeid@igalia.com>
Cc: Darren Hart <dvhart@infradead.org>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Valentin Schneider <vschneid@redhat.com>
Cc: Waiman Long <longman@redhat.com>
Link: https://lore.kernel.org/r/aB8SI00EHBri23lB@gmail.com
|
|
Since commit 2070887fdeac ("futex: fix restart in wait_requeue_pi"),
futex_wait_requeue_pi() no longer uses restart_block. Update the comment
accordingly.
Signed-off-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/20250428193445.4571-1-namcao@linutronix.de
|
|
Implement a simple NUMA aware spinlock for testing and howto purposes.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
|
|
Test the basic functionality for the NUMA and MPOL flags:
- FUTEX2_NUMA should take the NUMA node which is after the uaddr
and use it.
- Only update the node if FUTEX_NO_NODE was set by the user
- FUTEX2_MPOL should use the memory based on the policy. I attempted to
set the node with mbind() and then use this with MPOL but this fails
and futex falls back to the default node for the current CPU.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20250416162921.513656-22-bigeasy@linutronix.de
|
|
Test the basic functionality of the private hash:
- Upon start, with no threads there is no private hash.
- The first thread initializes the private hash.
- More than four threads will increase the size of the private hash if
the system has more than 16 CPUs online.
- Once the user sets the size of private hash, auto scaling is disabled.
- The user is only allowed to use numbers to the power of two.
- The user may request the global or make the hash immutable.
- Once the global hash has been set or the hash has been made immutable,
further changes are not allowed.
- Futex operations should work the whole time. It must be possible to
hold a lock, such a PI initialised mutex, during the resize 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/20250416162921.513656-21-bigeasy@linutronix.de
|
|
Make it build without relying on recent headers.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
|
|
Add the -b/ --buckets argument to specify the number of hash buckets for
the private futex hash. This is directly passed to
prctl(PR_FUTEX_HASH, PR_FUTEX_HASH_SET_SLOTS, buckets, immutable)
and must return without an error if specified. The `immutable' is 0 by
default and can be set to 1 via the -I/ --immutable argument.
The size of the private hash is verified with PR_FUTEX_HASH_GET_SLOTS.
If PR_FUTEX_HASH_GET_SLOTS failed then it is assumed that an older
kernel was used without the support and that the global hash is used.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20250416162921.513656-20-bigeasy@linutronix.de
|
|
Synchronize prctl.h with current uapi version after adding
PR_FUTEX_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/20250416162921.513656-19-bigeasy@linutronix.de
|
|
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
|
|
Allocate a private futex hash with 16 slots if a task forks its first
thread.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20250416162921.513656-14-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
|
|
Factor out the futex_hash_bucket initialisation into a helpr function.
The helper function will be used in a follow up patch implementing
process private hash buckets.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20250416162921.513656-12-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
|
|
This gets us:
fph = futex_private_hash(key) /* gets fph and inc users */
futex_private_hash_get(fph) /* inc users */
futex_private_hash_put(fph) /* dec users */
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/20250416162921.513656-8-bigeasy@linutronix.de
|
|
This gets us:
hb = futex_hash(key) /* gets hb and inc users */
futex_hash_get(hb) /* inc users */
futex_hash_put(hb) /* dec users */
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/20250416162921.513656-7-bigeasy@linutronix.de
|
|
Create explicit scopes for hb variables; almost pure re-indent.
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/20250416162921.513656-6-bigeasy@linutronix.de
|
|
Getting the hash bucket and queuing it are two distinct actions. In
light of wanting to add a put hash bucket function later, untangle
them.
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/20250416162921.513656-5-bigeasy@linutronix.de
|
|
futex_wait_setup() has a weird calling convention in order to return
hb to use as an argument to futex_queue().
Mostly such that requeue can have an extra test in between.
Reorder code a little to get rid of this and keep the hb usage inside
futex_wait_setup().
[bigeasy: fixes]
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/20250416162921.513656-4-bigeasy@linutronix.de
|
|
To enable node specific hash-tables using huge pages if possible.
[bigeasy: use __vmalloc_node_range_noprof(), add nommu bits, inline
vmalloc_huge]
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>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20250416162921.513656-3-bigeasy@linutronix.de
|
|
rcuref_read() returns the number of references that are currently held.
If 0 is returned then it is not safe to assume that the object ca be
scheduled for deconstruction because it is marked DEAD. This happens if
the return value of rcuref_put() is ignored and assumptions are made.
If 0 is returned then the counter transitioned from 0 to RCUREF_NOREF.
If rcuref_put() did not return to the caller then the counter did not
yet transition from RCUREF_NOREF to RCUREF_DEAD. This means that there
is still a chance that the counter will transition from RCUREF_NOREF to
0 meaning it is still valid and must not be deconstructed. In this brief
window rcuref_read() will return 0.
Provide rcuref_is_dead() to determine if the counter is marked as
RCUREF_DEAD.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20250416162921.513656-2-bigeasy@linutronix.de
|
|
|
|
Ondrej reports that certain SELinux tests are failing after commit
fc2a169c56de ("sunrpc: clean cache_detail immediately when flush is
written frequently"), merged during the v6.15 merge window.
Reported-by: Ondrej Mosnacek <omosnace@redhat.com>
Fixes: fc2a169c56de ("sunrpc: clean cache_detail immediately when flush is written frequently")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
|
|
There is a code path in dequeue_entities() that can set the slice of a
sched_entity to U64_MAX, which sometimes results in a crash.
The offending case is when dequeue_entities() is called to dequeue a
delayed group entity, and then the entity's parent's dequeue is delayed.
In that case:
1. In the if (entity_is_task(se)) else block at the beginning of
dequeue_entities(), slice is set to
cfs_rq_min_slice(group_cfs_rq(se)). If the entity was delayed, then
it has no queued tasks, so cfs_rq_min_slice() returns U64_MAX.
2. The first for_each_sched_entity() loop dequeues the entity.
3. If the entity was its parent's only child, then the next iteration
tries to dequeue the parent.
4. If the parent's dequeue needs to be delayed, then it breaks from the
first for_each_sched_entity() loop _without updating slice_.
5. The second for_each_sched_entity() loop sets the parent's ->slice to
the saved slice, which is still U64_MAX.
This throws off subsequent calculations with potentially catastrophic
results. A manifestation we saw in production was:
6. In update_entity_lag(), se->slice is used to calculate limit, which
ends up as a huge negative number.
7. limit is used in se->vlag = clamp(vlag, -limit, limit). Because limit
is negative, vlag > limit, so se->vlag is set to the same huge
negative number.
8. In place_entity(), se->vlag is scaled, which overflows and results in
another huge (positive or negative) number.
9. The adjusted lag is subtracted from se->vruntime, which increases or
decreases se->vruntime by a huge number.
10. pick_eevdf() calls entity_eligible()/vruntime_eligible(), which
incorrectly returns false because the vruntime is so far from the
other vruntimes on the queue, causing the
(vruntime - cfs_rq->min_vruntime) * load calulation to overflow.
11. Nothing appears to be eligible, so pick_eevdf() returns NULL.
12. pick_next_entity() tries to dereference the return value of
pick_eevdf() and crashes.
Dumping the cfs_rq states from the core dumps with drgn showed tell-tale
huge vruntime ranges and bogus vlag values, and I also traced se->slice
being set to U64_MAX on live systems (which was usually "benign" since
the rest of the runqueue needed to be in a particular state to crash).
Fix it in dequeue_entities() by always setting slice from the first
non-empty cfs_rq.
Fixes: aef6987d8954 ("sched/eevdf: Propagate min_slice up the cgroup hierarchy")
Signed-off-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lkml.kernel.org/r/f0c2d1072be229e1bdddc73c0703919a8b00c652.1745570998.git.osandov@fb.com
|
|
With ACPI in place, gicv2m_get_fwnode() is registered with the pci
subsystem as pci_msi_get_fwnode_cb(), which may get invoked at runtime
during a PCI host bridge probe. But, the call back is wrongly marked as
__init, causing it to be freed, while being registered with the PCI
subsystem and could trigger:
Unable to handle kernel paging request at virtual address ffff8000816c0400
gicv2m_get_fwnode+0x0/0x58 (P)
pci_set_bus_msi_domain+0x74/0x88
pci_register_host_bridge+0x194/0x548
This is easily reproducible on a Juno board with ACPI boot.
Retain the function for later use.
Fixes: 0644b3daca28 ("irqchip/gic-v2m: acpi: Introducing GICv2m ACPI support")
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Marc Zyngier <maz@kernel.org>
Cc: stable@vger.kernel.org
|
|
In function kvm_pre_enter_guest(), it prepares to enter guest and check
whether there are pending signals or events. And it will not enter guest
if there are, PMU pass-through preparation for guest should be cancelled
and host should own PMU hardware.
Cc: stable@vger.kernel.org
Fixes: f4e40ea9f78f ("LoongArch: KVM: Add PMU support for guest")
Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
|
|
Some registers such as LOONGARCH_CSR_ESTAT and LOONGARCH_CSR_GINTC are
partly cleared with function _kvm_setcsr(). This comes from the hardware
specification, some bits are read only in VM mode, and however they can
be written in host mode. So they are partly cleared in VM mode, and can
be fully cleared in host mode.
These read only bits show pending interrupt or exception status. When VM
reset, the read-only bits should be cleared, otherwise vCPU will receive
unknown interrupts in boot stage.
Here registers LOONGARCH_CSR_ESTAT/LOONGARCH_CSR_GINTC are fully cleared
in ioctl KVM_REG_LOONGARCH_VCPU_RESET vCPU reset path.
Cc: stable@vger.kernel.org
Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
|
|
Fix multiple typos inside arch/loongarch/kvm.
Cc: stable@vger.kernel.org
Reviewed-by: Yuli Wang <wangyuli@uniontech.com>
Reviewed-by: Bibo Mao <maobibo@loongson.cn>
Signed-off-by: Yulong Han <wheatfox17@icloud.com>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
|
|
LoongArch's huge_pte_offset() currently returns a pointer to a PMD slot
even if the underlying entry points to invalid_pte_table (indicating no
mapping). Callers like smaps_hugetlb_range() fetch this invalid entry
value (the address of invalid_pte_table) via this pointer.
The generic is_swap_pte() check then incorrectly identifies this address
as a swap entry on LoongArch, because it satisfies the "!pte_present()
&& !pte_none()" conditions. This misinterpretation, combined with a
coincidental match by is_migration_entry() on the address bits, leads to
kernel crashes in pfn_swap_entry_to_page().
Fix this at the architecture level by modifying huge_pte_offset() to
check the PMD entry's content using pmd_none() before returning. If the
entry is invalid (i.e., it points to invalid_pte_table), return NULL
instead of the pointer to the slot.
Cc: stable@vger.kernel.org
Acked-by: Peter Xu <peterx@redhat.com>
Co-developed-by: Hongchen Zhang <zhanghongchen@loongson.cn>
Signed-off-by: Hongchen Zhang <zhanghongchen@loongson.cn>
Signed-off-by: Ming Wang <wangming01@loongson.cn>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
|
|
Remove dead code. LoongArch does not have a DMA memory zone (24bit DMA).
The architecture does not even define MAX_DMA_PFN.
Cc: stable@vger.kernel.org
Reviewed-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Signed-off-by: Petr Tesarik <ptesarik@suse.com>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
|
|
Like the other relevant symbols, export some fp, lsx, lasx and lbt
assembly symbols and put the function declarations in header files
rather than source files.
While at it, use "asmlinkage" for the other existing C prototypes
of assembly functions and also do not use the "extern" keyword with
function declarations according to the document coding-style.rst.
Cc: stable@vger.kernel.org # 6.6+
Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
|
|
Currently, interrupts need to be disabled before single-step mode is
set, it requires that CSR_PRMD_PIE be cleared in save_local_irqflag()
which is called by setup_singlestep(), this is reasonable.
But in the first kprobe breakpoint exception, if the irq is enabled at
the beginning of do_bp(), it will not be disabled at the end of do_bp()
due to the CSR_PRMD_PIE has been cleared in save_local_irqflag(). So for
this case, it may corrupt exception context when restoring the exception
after do_bp() in handle_bp(), this is not reasonable.
In order to restore exception safely in handle_bp(), it needs to ensure
the irq is disabled at the end of do_bp(), so just add a local variable
to record the original interrupt status in the parent context, then use
it as the check condition to enable and disable irq in do_bp().
While at it, do the similar thing for other do_xyz() exception handlers
to make them more robust.
Fixes: 6d4cc40fb5f5 ("LoongArch: Add kprobes support")
Suggested-by: Jinyang He <hejinyang@loongson.cn>
Suggested-by: Huacai Chen <chenhuacai@loongson.cn>
Co-developed-by: Tianyang Zhang <zhangtianyang@loongson.cn>
Signed-off-by: Tianyang Zhang <zhangtianyang@loongson.cn>
Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
|
|
In the current code, the definition of regs_irqs_disabled() is actually
"!(regs->csr_prmd & CSR_CRMD_IE)" because arch_irqs_disabled_flags() is
defined as "!(flags & CSR_CRMD_IE)", it looks a little strange.
Define regs_irqs_disabled() as !(regs->csr_prmd & CSR_PRMD_PIE) directly
to make it more clear, no functional change.
While at it, the return value of regs_irqs_disabled() is true or false,
so change its type to reflect that and also make it always inline.
Fixes: 803b0fc5c3f2 ("LoongArch: Add process management")
Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
|
|
As of commit dce44566192e ("mm/memtest: add ARCH_USE_MEMTEST"),
architectures must select ARCH_USE_MEMTESET to enable CONFIG_MEMTEST.
Commit 628c3bb40e9a ("LoongArch: Add boot and setup routines") added
support for early_memtest but did not select ARCH_USE_MEMTESET.
Fixes: 628c3bb40e9a ("LoongArch: Add boot and setup routines")
Tested-by: Erpeng Xu <xuerpeng@uniontech.com>
Tested-by: Yuli Wang <wangyuli@uniontech.com>
Signed-off-by: Yuli Wang <wangyuli@uniontech.com>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
|
|
Make sure that CAN_USE_BPF_ST test (compute_live_registers/store) is
enabled when __clang_major__ >= 18.
Fixes: 2ea8f6a1cda7 ("selftests/bpf: test cases for compute_live_registers()")
Signed-off-by: Peilin Ye <yepeilin@google.com>
Link: https://lore.kernel.org/r/20250425213712.1542077-1-yepeilin@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
When building the latest samples/bpf on LoongArch Fedora
make M=samples/bpf
There are compilation errors as follows:
In file included from ./linux/samples/bpf/sockex2_kern.c:2:
In file included from ./include/uapi/linux/in.h:25:
In file included from ./include/linux/socket.h:8:
In file included from ./include/linux/uio.h:9:
In file included from ./include/linux/thread_info.h:60:
In file included from ./arch/loongarch/include/asm/thread_info.h:15:
In file included from ./arch/loongarch/include/asm/processor.h:13:
In file included from ./arch/loongarch/include/asm/cpu-info.h:11:
./arch/loongarch/include/asm/loongarch.h:13:10: fatal error: 'larchintrin.h' file not found
^~~~~~~~~~~~~~~
1 error generated.
larchintrin.h is included in /usr/lib64/clang/14.0.6/include,
and the header file location is specified at compile time.
Test on LoongArch Fedora:
https://github.com/fedora-remix-loongarch/releases-info
Signed-off-by: Haoran Jiang <jianghaoran@kylinos.cn>
Signed-off-by: zhangxi <zhangxi@kylinos.cn>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20250425095042.838824-1-jianghaoran@kylinos.cn
|
|
Add namespace to BPF internal symbols used by light skeleton
to prevent abuse and document with the code their allowed usage.
Fixes: b1d18a7574d0 ("bpf: Extend sys_bpf commands for bpf_syscall programs.")
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/bpf/20250425014542.62385-1-alexei.starovoitov@gmail.com
|
|
Add test that modifies the map while it's being iterated in such a way that
hangs the kernel thread unless the _safe fix is applied to
bpf_for_each_hash_elem.
Signed-off-by: Brandon Kammerdiener <brandon.kammerdiener@intel.com>
Link: https://lore.kernel.org/r/20250424153246.141677-3-brandon.kammerdiener@intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Hou Tao <houtao1@huawei.com>
|
|
The _safe variant used here gets the next element before running the callback,
avoiding the endless loop condition.
Signed-off-by: Brandon Kammerdiener <brandon.kammerdiener@intel.com>
Link: https://lore.kernel.org/r/20250424153246.141677-2-brandon.kammerdiener@intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Hou Tao <houtao1@huawei.com>
|
|
Especially the port manager (tcpm.c) is so major driver that
it should have somebody watching over it who really
understands it, and the port controller interface in
general. Assigning Badhri as the designated reviewer and
restoring the status to Maintained from Orphan.
Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Cc: Badhri Jagan Sridharan <badhri@google.com>
Acked-by: Badhri Jagan Sridharan <badhri@google.com>
Link: https://lore.kernel.org/r/20250407133306.387576-1-heikki.krogerus@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Currently, setxattrat(2) and getxattrat(2) are wrongly handling the
calls of the from setxattrat(AF_FDCWD, NULL, AT_EMPTY_PATH, ...) and
fail with -EBADF error instead of operating on CWD. Fix it.
Fixes: 6140be90ec70 ("fs/xattr: add *at family syscalls")
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/20250424132246.16822-2-jack@suse.cz
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
I used to maintain Allwinner SoC cpufreq and thermal drivers and
have some work experience in the F2FS file system.
I volunteered to maintain the code together with Slava and Adrian.
Signed-off-by: Yangtao Li <frank.li@vivo.com>
Link: https://lore.kernel.org/20250423123423.2062619-1-frank.li@vivo.com
Acked-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
pipe_clear_nowait has two noinline macros, but we only need one.
I checked the whole tree, and this is the only occurrence:
$ grep -r "noinline .* noinline"
fs/splice.c:static noinline void noinline pipe_clear_nowait(struct file *file)
$
Fixes: 0f99fc513ddd ("splice: clear FMODE_NOWAIT on file if splice/vmsplice is used")
Signed-off-by: "T.J. Mercier" <tjmercier@google.com>
Link: https://lore.kernel.org/20250423180025.2627670-1-tjmercier@google.com
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
The recent move of the bdev_statx call to the low-level vfs_getattr_nosec
helper caused it being used by devtmpfs, which leads to deadlocks in
md teardown due to the block device lookup and put interfering with the
unusual lifetime rules in md.
But as handle_remove only works on inodes created and owned by devtmpfs
itself there is no need to use vfs_getattr_nosec vs simply reading the
mode from the inode directly. Switch to that to avoid the bdev lookup
or any other unintentional side effect.
Reported-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Reported-by: Xiao Ni <xni@redhat.com>
Fixes: 777d0961ff95 ("fs: move the bdex_statx call to vfs_getattr_nosec")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/20250423045941.1667425-1-hch@lst.de
Tested-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Tested-by: Xiao Ni <xni@redhat.com>
Tested-by: Ayush Jain <Ayush.jain3@amd.com>
Tested-by: Heiko Carstens <hca@linux.ibm.com>
Reviewed-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
ublk_cancel_cmd() calls io_uring_cmd_done() to complete uring_cmd, but
we may have scheduled task work via io_uring_cmd_complete_in_task() for
dispatching request, then kernel crash can be triggered.
Fix it by not trying to canceling the command if ublk block request is
started.
Fixes: 216c8f5ef0f2 ("ublk: replace monitor with cancelable uring_cmd")
Reported-by: Jared Holzman <jholzman@nvidia.com>
Tested-by: Jared Holzman <jholzman@nvidia.com>
Closes: https://lore.kernel.org/linux-block/d2179120-171b-47ba-b664-23242981ef19@nvidia.com/
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20250425013742.1079549-3-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
Borislav Petkov (AMD)
Ingo Molnar
Nam Cao
Peter Zijlstra
Sebastian Andrzej Siewior
Linus Torvalds
Chuck Lever
Omar Sandoval
Suzuki K Poulose
Bibo Mao
Yulong Han
Ming Wang
Petr Tesarik
Tiezhu Yang
Yuli Wang
Peilin Ye
Haoran Jiang
Alexei Starovoitov
Brandon Kammerdiener
Heikki Krogerus
Jan Kara
Yangtao Li
T.J. Mercier
Christoph Hellwig
Ming Lei