Pull scheduler fixes from Ingo Molnar:
"Fix two dl_server regressions: a race that can end up leaving the
dl_server stuck, and a dl_server throttling bug causing lag to fair
tasks"

* tag 'sched-urgent-2025-09-26' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/deadline: Fix dl_server behaviour
sched/deadline: Fix dl_server getting stuck

Pull locking fixes from Ingo Molnar:
"Fix a PI-futexes race, and fix a copy_process() futex cleanup bug"

* tag 'locking-urgent-2025-09-26' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
futex: Use correct exit on failure from futex_hash_allocate_default()
futex: Prevent use-after-free during requeue-PI

John reported undesirable behaviour with the dl_server since commit:
cccb45d7c4295 ("sched/deadline: Less agressive dl_server handling").

When starving fair tasks on purpose (starting spinning FIFO tasks),
his fair workload, which often goes (briefly) idle, would delay fair
invocations for a second, running one invocation per second was both
unexpected and terribly slow.

The reason this happens is that when dl_se->server_pick_task() returns
NULL, indicating no runnable tasks, it would yield, pushing any later
jobs out a whole period (1 second).

Instead simply stop the server. This should restore behaviour in that
a later wakeup (which restarts the server) will be able to continue
running (subject to the CBS wakeup rules).

Notably, this does not re-introduce the behaviour cccb45d7c4295 set
out to solve, any start/stop cycle is naturally throttled by the timer
period (no active cancel).

Fixes: cccb45d7c4295 ("sched/deadline: Less agressive dl_server handling")
Reported-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: John Stultz <jstultz@google.com>

Pull core fix from Ingo Molnar:
"Fix a CONFIG_CC_HAS_ASM_GOTO_OUTPUT=y bug on older Clang versions"

* tag 'core-urgent-2025-09-26' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
kbuild: Disable CC_HAS_ASM_GOTO_OUTPUT on clang < 17

copy_process() uses the wrong error exit path from futex_hash_allocate_default().
After exiting from futex_hash_allocate_default(), neither tasklist_lock
nor siglock has been acquired. The exit label bad_fork_core_free unlocks
both of these locks which is wrong.

The next exit label, bad_fork_cancel_cgroup, is the correct exit.
sched_cgroup_fork() did not allocate any resources that need to freed.

Use bad_fork_cancel_cgroup on error exit from futex_hash_allocate_default().

Fixes: 7c4f75a21f636 ("futex: Allow automatic allocation of process wide futex hash")
Reported-by: syzbot+80cb3cc5c14fad191a10@syzkaller.appspotmail.com
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Closes: https://lore.kernel.org/all/68cb1cbd.050a0220.2ff435.0599.GAE@google.com

John found it was easy to hit lockup warnings when running locktorture
on a 2 CPU VM, which he bisected down to: commit cccb45d7c429
("sched/deadline: Less agressive dl_server handling").

While debugging it seems there is a chance where we end up with the
dl_server dequeued, with dl_se->dl_server_active. This causes
dl_server_start() to return without enqueueing the dl_server, thus it
fails to run when RT tasks starve the cpu.

When this happens, dl_server_timer() catches the
'!dl_se->server_has_tasks(dl_se)' case, which then calls
replenish_dl_entity() and dl_server_stopped() and finally return
HRTIMER_NO_RESTART.

This ends in no new timer and also no enqueue, leaving the dl_server
'dead', allowing starvation.

What should have happened is for the bandwidth timer to start the
zero-laxity timer, which in turn would enqueue the dl_server and cause
dl_se->server_pick_task() to be called -- which will stop the
dl_server if no fair tasks are observed for a whole period.

IOW, it is totally irrelevant if there are fair tasks at the moment of
bandwidth refresh.

This removes all dl_se->server_has_tasks() users, so remove the whole
thing.

Fixes: cccb45d7c4295 ("sched/deadline: Less agressive dl_server handling")
Reported-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: John Stultz <jstultz@google.com>

Pull smb client fixes from Steve French:

- Fix unlink bug

- Fix potential out of bounds access in processing compound requests

* tag 'v6.17rc7-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6:
smb: client: fix wrong index reference in smb2_compound_op()
smb: client: handle unlink(2) of files open by different clients

clang < 17 fails to use scope local labels with CONFIG_CC_HAS_ASM_GOTO_OUTPUT=y:

{
__label__ local_lbl;
...
unsafe_get_user(uval, uaddr, local_lbl);
...
return 0;
local_lbl:
return -EFAULT;
}

when two such scopes exist in the same function:

error: cannot jump from this asm goto statement to one of its possible targets

There are other failure scenarios. Shuffling code around slightly makes it
worse and fail even with one instance.

That issue prevents using local labels for a cleanup based user access
mechanism.

After failed attempts to provide a simple enough test case for the 'depends
on' test in Kconfig, the initial cure was to mark ASM goto broken on clang
versions < 17 to get this road block out of the way.

But Nathan pointed out that this is a known clang issue and indeed affects
clang < version 17 in combination with cleanup(). It's not even required to
use local labels for that.

The clang issue tracker has a small enough test case, which can be used as
a test in the 'depends on' section of CC_HAS_ASM_GOTO_OUTPUT:

void bar(void **);
void* baz(void);

int foo (void) {
{
asm goto("jmp %l0"::::l0);
return 0;
l0:
return 1;
}
void *x __attribute__((cleanup(bar))) = baz();
{
asm goto("jmp %l0"::::l1);
return 42;
l1:
return 0xff;
}
}

Add another dependency to config CC_HAS_ASM_GOTO_OUTPUT for it and use the
clang issue tracker test case for detection by condensing it to obfuscated
C-code contest format. This reliably catches the problem on clang < 17 and
did not show any issues on the non broken GCC versions.

That test might be sufficient to catch all issues and therefore could
replace the existing test, but keeping that around does no harm either.

Thanks to Nathan for pointing to the relevant clang issue!

Suggested-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Nathan Chancellor <nathan@kernel.org>
Link: https://github.com/ClangBuiltLinux/linux/issues/1886
Link: https://github.com/llvm/llvm-project/commit/f023f5cdb2e6c19026f04a15b5a935c041835d14

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: 07d91ef510fb1 ("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/

Pull vfs fixes from Christian Brauner:

- Prevent double unlock in netfs

- Fix a NULL pointer dereference in afs_put_server()

- Fix a reference leak in netfs

* tag 'vfs-6.17-rc8.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:
netfs: fix reference leak
afs: Fix potential null pointer dereference in afs_put_server
netfs: Prevent duplicate unlocking

In smb2_compound_op(), the loop that processes each command's response
uses wrong indices when accessing response bufferes.

This incorrect indexing leads to improper handling of command results.
Also, if incorrectly computed index is greather than or equal to
MAX_COMPOUND, it can cause out-of-bounds accesses.

Fixes: 3681c74d342d ("smb: client: handle lack of EA support in smb2_query_path_info()") # 6.14
Reviewed-by: Paulo Alcantara (Red Hat) <pc@manguebit.org>
Signed-off-by: Sang-Heon Jeon <ekffu200098@gmail.com>
Signed-off-by: Steve French <stfrench@microsoft.com>

Pull generic phy driver fixes from Vinod Koul:

- Qualcomm repeater override properties, qmp pcie bindings fix for
clocks and initialization sequence for firmware power down case

- Marvell comphy bindings clock and child node constraints

- Tegra xusb device reference leaks fix

- TI omap usb device ref leak on unbind and RGMII IS settings fix

* tag 'phy-fix-6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/phy/linux-phy:
phy: qcom: qmp-pcie: Fix PHY initialization when powered down by firmware
phy: ti: gmii-sel: Always write the RGMII ID setting
dt-bindings: phy: qcom,sc8280xp-qmp-pcie-phy: Update pcie phy bindings
phy: ti-pipe3: fix device leak at unbind
phy: ti: omap-usb2: fix device leak at unbind
phy: tegra: xusb: fix device and OF node leak at probe
dt-bindings: phy: marvell,comphy-cp110: Fix clock and child node constraints
phy: qualcomm: phy-qcom-eusb2-repeater: fix override properties

Pull pmdomain fix from Ulf Hansson:

- mediatek: Make sure MT8195 AUDIO power domain isn't left powered-on

* tag 'pmdomain-v6.17-rc2-2' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/linux-pm:
pmdomain: mediatek: set default off flag for MT8195 AUDIO power domain

Commit 20d72b00ca81 ("netfs: Fix the request's work item to not
require a ref") modified netfs_alloc_request() to initialize the
reference counter to 2 instead of 1. The rationale was that the
requet's "work" would release the second reference after completion
(via netfs_{read,write}_collection_worker()). That works most of the
time if all goes well.

However, it leaks this additional reference if the request is released
before the I/O operation has been submitted: the error code path only
decrements the reference counter once and the work item will never be
queued because there will never be a completion.

This has caused outages of our whole server cluster today because
tasks were blocked in netfs_wait_for_outstanding_io(), leading to
deadlocks in Ceph (another bug that I will address soon in another
patch). This was caused by a netfs_pgpriv2_begin_copy_to_cache() call
which failed in fscache_begin_write_operation(). The leaked
netfs_io_request was never completed, leaving `netfs_inode.io_count`
with a positive value forever.

All of this is super-fragile code. Finding out which code paths will
lead to an eventual completion and which do not is hard to see:

- Some functions like netfs_create_write_req() allocate a request, but
will never submit any I/O.

- netfs_unbuffered_read_iter_locked() calls netfs_unbuffered_read()
and then netfs_put_request(); however, netfs_unbuffered_read() can
also fail early before submitting the I/O request, therefore another
netfs_put_request() call must be added there.

A rule of thumb is that functions that return a `netfs_io_request` do
not submit I/O, and all of their callers must be checked.

For my taste, the whole netfs code needs an overhaul to make reference
counting easier to understand and less fragile & obscure. But to fix
this bug here and now and produce a patch that is adequate for a
stable backport, I tried a minimal approach that quickly frees the
request object upon early failure.

I decided against adding a second netfs_put_request() each time
because that would cause code duplication which obscures the code
further. Instead, I added the function netfs_put_failed_request()
which frees such a failed request synchronously under the assumption
that the reference count is exactly 2 (as initially set by
netfs_alloc_request() and never touched), verified by a
WARN_ON_ONCE(). It then deinitializes the request object (without
going through the "cleanup_work" indirection) and frees the allocation
(with RCU protection to protect against concurrent access by
netfs_requests_seq_start()).

All code paths that fail early have been changed to call
netfs_put_failed_request() instead of netfs_put_request().
Additionally, I have added a netfs_put_request() call to
netfs_unbuffered_read() as explained above because the
netfs_put_failed_request() approach does not work there.

Fixes: 20d72b00ca81 ("netfs: Fix the request's work item to not require a ref")
Signed-off-by: Max Kellermann <max.kellermann@ionos.com>
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Paulo Alcantara <pc@manguebit.org>
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
cc: stable@vger.kernel.org
Signed-off-by: Christian Brauner <brauner@kernel.org>