Remove unnecessary kfree(main_task) call in ref_scale_cleanup() as
torture_stop_kthread() already handles the memory cleanup for the
task structure internally.
The additional kfree(main_task) call after torture_stop_kthread()
is redundant and confusing since torture_stop_kthread() sets the
pointer to NULL, making this a no-op.
This pattern is consistent with other torture test modules where
torture_stop_kthread() is called without explicit kfree() of the
task pointer, as the torture framework manages the task lifecycle
internally.
Signed-off-by: Kaushlendra Kumar <kaushlendra.kumar@intel.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Currently, the srcu_gp_start_if_needed() is always be invoked in
preempt disable's critical section, this commit therefore remove
redundant preempt_disable/enable() in srcu_gp_start_if_needed()
and adds a call to lockdep_assert_preemption_disabled() in order
to enable lockdep to diagnose mistaken invocations of this function
from preempts-enabled code.
Fixes: 65b4a59557 ("srcu: Make Tiny SRCU explicitly disable preemption")
Signed-off-by: Zqiang <qiang.zhang@linux.dev>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The smp_mb() memory barrier at the end of srcu_flip() has a comment,
but that comment does not make it clear that this memory barrier is an
optimization, as opposed to being needed for correctness. This commit
therefore adds this information and points out that it is omitted
for SRCU-fast, where a much heavier weight synchronize_srcu() would
be required.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: <bpf@vger.kernel.org>
When doing load balance and the target cfs_rq is in throttled hierarchy,
whether to allow balancing there is a question.
The good side to allow balancing is: if the target CPU is idle or less
loaded and the being balanced task is holding some kernel resources,
then it seems a good idea to balance the task there and let the task get
the CPU earlier and release kernel resources sooner. The bad part is, if
the task is not holding any kernel resources, then the balance seems not
that useful.
While theoretically it's debatable, a performance test[0] which involves
200 cgroups and each cgroup runs hackbench(20 sender, 20 receiver) in
pipe mode showed a performance degradation on AMD Genoa when allowing
load balance to throttled cfs_rq. Analysis[1] showed hackbench doesn't
like task migration across LLC boundary. For this reason, add a check in
can_migrate_task() to forbid balancing to a cfs_rq that is in throttled
hierarchy. This reduced task migration a lot and performance restored.
[0]: https://lore.kernel.org/lkml/20250822110701.GB289@bytedance/
[1]: https://lore.kernel.org/lkml/20250903101102.GB42@bytedance/
Signed-off-by: Aaron Lu <ziqianlu@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: K Prateek Nayak <kprateek.nayak@amd.com>
With the introduction of task based throttle model, task in a throttled
hierarchy is allowed to continue to run till it gets throttled on its
ret2user path.
For this reason, remove those throttled_hierarchy() checks in the
following functions so that those tasks can get their turn as normal
tasks: dequeue_entities(), check_preempt_wakeup_fair() and
yield_to_task_fair().
The benefit of doing it this way is: if those tasks gets the chance to
run earlier and if they hold any kernel resources, they can release
those resources earlier. The downside is, if they don't hold any kernel
resouces, all they can do is to throttle themselves on their way back to
user space so the favor to let them run seems not that useful and for
check_preempt_wakeup_fair(), that favor may be bad for curr.
K Prateek Nayak pointed out prio_changed_fair() can send a throttled
task to check_preempt_wakeup_fair(), further tests showed the affinity
change path from move_queued_task() can also send a throttled task to
check_preempt_wakeup_fair(), that's why the check of task_is_throttled()
in that function.
Signed-off-by: Aaron Lu <ziqianlu@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
With task based throttle model, tasks in a throttled hierarchy are
allowed to continue to run if they are running in kernel mode. For this
reason, PELT clock is not stopped for these cfs_rqs in throttled
hierarchy when they still have tasks running or queued.
Since PELT clock is not stopped, whether to allow update_cfs_group()
doing its job for cfs_rqs which are in throttled hierarchy but still
have tasks running/queued is a question.
The good side is, continue to run update_cfs_group() can get these
cfs_rq entities with an up2date weight and that up2date weight can be
useful to derive an accurate load for the CPU as well as ensure fairness
if multiple tasks of different cgroups are running on the same CPU.
OTOH, as Benjamin Segall pointed: when unthrottle comes around the most
likely correct distribution is the distribution we had at the time of
throttle.
In reality, either way may not matter that much if tasks in throttled
hierarchy don't run in kernel mode for too long. But in case that
happens, let these cfs_rq entities have an up2date weight seems a good
thing to do.
Signed-off-by: Aaron Lu <ziqianlu@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Before task based throttle model, propagating load will stop at a
throttled cfs_rq and that propagate will happen on unthrottle time by
update_load_avg().
Now that there is no update_load_avg() on unthrottle for throttled
cfs_rq and all load tracking is done by task related operations, let the
propagate happen immediately.
While at it, add a comment to explain why cfs_rqs that are not affected
by throttle have to be added to leaf cfs_rq list in
propagate_entity_cfs_rq() per my understanding of commit 0258bdfaff
("sched/fair: Fix unfairness caused by missing load decay").
Signed-off-by: Aaron Lu <ziqianlu@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Chengming Zhou <chengming.zhou@linux.dev>
If create_monitor_dir() fails, the function returns directly without
releasing rv_interface_lock. This leaves the mutex locked and causes
subsequent monitor registration attempts to deadlock.
Fix it by making the error path jump to out_unlock, ensuring that the
mutex is always released before returning.
Fixes: 24cbfe18d5 ("rv: Merge struct rv_monitor_def into struct rv_monitor")
Signed-off-by: Zhen Ni <zhen.ni@easystack.cn>
Reviewed-by: Gabriele Monaco <gmonaco@redhat.com>
Reviewed-by: Nam Cao <namcao@linutronix.de>
Link: https://lore.kernel.org/r/20250903065112.1878330-1-zhen.ni@easystack.cn
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
Argument 'p' of enabled_monitors_next() is not a pointer to struct
rv_monitor, it is actually a pointer to the list_head inside struct
rv_monitor. Therefore it is wrong to cast 'p' to struct rv_monitor *.
This wrong type cast has been there since the beginning. But it still
worked because the list_head was the first field in struct rv_monitor_def.
This is no longer true since commit 24cbfe18d5 ("rv: Merge struct
rv_monitor_def into struct rv_monitor") moved the list_head, and this wrong
type cast became a functional problem.
Properly use container_of() instead.
Fixes: 24cbfe18d5 ("rv: Merge struct rv_monitor_def into struct rv_monitor")
Signed-off-by: Nam Cao <namcao@linutronix.de>
Reviewed-by: Gabriele Monaco <gmonaco@redhat.com>
Link: https://lore.kernel.org/r/20250806120911.989365-1-namcao@linutronix.de
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
Pull timer fix from Ingo Molnar:
"Fix a lost-timeout CPU hotplug bug in the hrtimer code, which can
trigger with certain hardware configs and regular HZ"
* tag 'timers-urgent-2025-09-14' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
hrtimers: Unconditionally update target CPU base after offline timer migration
Patch series "efi: Fix EFI boot with kexec handover (KHO)", v3.
This patch series fixes a kernel panic that occurs when booting with both
EFI and KHO (Kexec HandOver) enabled.
The issue arises because EFI's `reserve_regions()` clears all memory
regions with `memblock_remove(0, PHYS_ADDR_MAX)` before rebuilding them
from EFI data. This destroys KHO scratch regions that were set up early
during device tree scanning, causing a panic as the kernel has no valid
memory regions for early allocations.
The first patch introduces `is_kho_boot()` to allow early boot components
to reliably detect if the kernel was booted via KHO-enabled kexec. The
existing `kho_is_enabled()` only checks the command line and doesn't
verify if an actual KHO FDT was passed.
The second patch modifies EFI's `reserve_regions()` to selectively remove
only non-KHO memory regions when KHO is active, preserving the critical
scratch regions while still allowing EFI to rebuild its memory map.
This patch (of 3):
During early initialisation, after a kexec, other components, like EFI
need to know if a KHO enabled kexec is performed. The `kho_is_enabled`
function is not enough as in the early stages, it only reflects whether
the cmdline has KHO enabled, not if an actual KHO FDT exists.
Extend the KHO API with `is_kho_boot()` to provide a way for components to
check if a KHO enabled kexec is performed.
Link: https://lkml.kernel.org/r/cover.1755721529.git.epetron@amazon.de
Link: https://lkml.kernel.org/r/7dc6674a76bf6e68cca0222ccff32427699cc02e.1755721529.git.epetron@amazon.de
Signed-off-by: Evangelos Petrongonas <epetron@amazon.de>
Reviewed-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Reviewed-by: Pratyush Yadav <pratyush@kernel.org>
Cc: Alexander Graf <graf@amazon.com>
Cc: Ard Biesheuvel <ardb@kernel.org>
Cc: Baoquan He <bhe@redhat.com>
Cc: Changyuan Lyu <changyuanl@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
crash_exclude_mem_range seems to be a simple function but there have been
multiple attempts to fix it,
- commit a2e9a95d21 ("kexec: Improve & fix crash_exclude_mem_range()
to handle overlapping ranges")
- commit 6dff315972 ("crash_core: fix and simplify the logic of
crash_exclude_mem_range()")
So add a set of unit tests to verify the correctness of current
implementation. Shall we change the function in the future, the unit
tests can also help prevent any regression. For example, we may make the
function smarter by allocating extra crash_mem range on demand thus there
is no need for the caller to foresee any memory range split or address
-ENOMEM failure.
The testing strategy is to verify the correctness of base case. The
base case is there is one to-be-excluded range A and one existing range
B. Then we can exhaust all possibilities of the position of A regarding
B. For example, here are two combinations,
Case: A is completely inside B (causes split)
Original: [----B----]
Exclude: {--A--}
Result: [B1] .. [B2]
Case: A overlaps B's left part
Original: [----B----]
Exclude: {---A---}
Result: [..B..]
In theory we can prove the correctness by induction,
- Base case: crash_exclude_mem_range is correct in the case where n=1
(n is the number of existing ranges).
- Inductive step: If crash_exclude_mem_range is correct for n=k
existing ranges, then the it's also correct for n=k+1 ranges.
But for the sake of simplicity, simply use unit tests to cover the base
case together with two regression tests.
Note most of the exclude_single_range_test() code is generated by Google
Gemini with some small tweaks. The function specification, function body
and the exhausting test strategy are presented as prompts.
[akpm@linux-foundation.org: export crash_exclude_mem_range() to modules, for kernel/crash_core_test.c]
Link: https://lkml.kernel.org/r/20250904093855.1180154-2-coxu@redhat.com
Signed-off-by: Coiby Xu <coxu@redhat.com>
Assisted-by: Google Gemini
Cc: Baoquan He <bhe@redhat.com>
Cc: Borislav Betkov <bp@alien8.de>
Cc: Dave Young <dyoung@redhat.com>
Cc: fuqiang wang <fuqiang.wang@easystack.cn>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Thomas Gleinxer <tglx@linutronix.de>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Unlike sys_clone(), these helpers have only in kernel users which should
pass the correct "flags" argument. lower_32_bits(flags) just adds the
unnecessary confusion and doesn't allow to use the CLONE_ flags which
don't fit into 32 bits.
create_io_thread() looks especially confusing because:
- "flags" is a compile-time constant, so lower_32_bits() simply
has no effect
- .exit_signal = (lower_32_bits(flags) & CSIGNAL) is harmless but
doesn't look right, copy_process(CLONE_THREAD) will ignore this
argument anyway.
None of these helpers actually need CLONE_UNTRACED or "& ~CSIGNAL", but
their presence does not add any confusion and improves code clarity.
Link: https://lkml.kernel.org/r/20250820163946.GA18549@redhat.com
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Kees Cook <kees@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Since we use 16-bit precision, the raw data will undergo integer division,
which may sometimes result in data loss. This can lead to slightly
inaccurate CPU utilization calculations. Under normal circumstances, this
isn't an issue. However, when CPU utilization reaches 100%, the
calculated result might exceed 100%. For example, with raw data like the
following:
sample_period 400000134 new_stat 83648414036 old_stat 83247417494
sample_period=400000134/2^24=23
new_stat=83648414036/2^24=4985
old_stat=83247417494/2^24=4961
util=105%
Below log will output：
CPU#3 Utilization every 0s during lockup:
#1: 0% system, 0% softirq, 105% hardirq, 0% idle
#2: 0% system, 0% softirq, 105% hardirq, 0% idle
#3: 0% system, 0% softirq, 100% hardirq, 0% idle
#4: 0% system, 0% softirq, 105% hardirq, 0% idle
#5: 0% system, 0% softirq, 105% hardirq, 0% idle
To avoid confusion, we enforce a 100% display cap when calculations exceed
this threshold.
We also round to the nearest multiple of 16.8 milliseconds to improve the
accuracy.
[yaozhenguo1@gmail.com: make get_16bit_precision() more accurate, fix comment layout]
Link: https://lkml.kernel.org/r/20250818081438.40540-1-yaozhenguo@jd.com
Link: https://lkml.kernel.org/r/20250812082510.32291-1-yaozhenguo@jd.com
Signed-off-by: ZhenguoYao <yaozhenguo1@gmail.com>
Cc: Bitao Hu <yaoma@linux.alibaba.com>
Cc: Li Huafei <lihuafei1@huawei.com>
Cc: Max Kellermann <max.kellermann@ionos.com>
Cc: Thomas Gleinxer <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
KCOV Remote uses two separate memory buffers, one private to the kernel
space (kcov_remote_areas) and the second one shared between user and
kernel space (kcov->area). After every pair of kcov_remote_start() and
kcov_remote_stop(), the coverage data collected in the kcov_remote_areas
is copied to kcov->area so the user can read the collected coverage data.
This memcpy() is located in kcov_move_area().
The load/store pattern on the kernel-side [1] is:
```
/* dst_area === kcov->area, dst_area[0] is where the count is stored */
dst_len = READ_ONCE(*(unsigned long *)dst_area);
...
memcpy(dst_entries, src_entries, ...);
...
WRITE_ONCE(*(unsigned long *)dst_area, dst_len + entries_moved);
```
And for the user [2]:
```
/* cover is equivalent to kcov->area */
n = __atomic_load_n(&cover[0], __ATOMIC_RELAXED);
```
Without a write-memory barrier, the atomic load for the user can
potentially read fresh values of the count stored at cover[0], but
continue to read stale coverage data from the buffer itself. Hence, we
recommend adding a write-memory barrier between the memcpy() and the
WRITE_ONCE() in kcov_move_area().
Link: https://lkml.kernel.org/r/20250728184318.1839137-1-soham.bagchi@utah.edu
Link: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/tree/kernel/kcov.c?h=master#n978 [1]
Link: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/tree/Documentation/dev-tools/kcov.rst#n364 [2]
Signed-off-by: Soham Bagchi <soham.bagchi@utah.edu>
Reviewed-by: Marco Elver <elver@google.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Andrey Konovalov <andreyknvl@gmail.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Dmitriy Vyukov <dvyukov@google.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Thomas Gleinxer <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Currently, the kexec_file_load syscall on x86 does not support passing a
device tree blob to the new kernel. Some embedded x86 systems use device
trees. On these systems, failing to pass a device tree to the new kernel
causes a boot failure.
To add support for this, we copy the behavior of ARM64 and PowerPC and
copy the current boot's device tree blob for use in the new kernel. We do
this on x86 by passing the device tree blob as a setup_data entry in
accordance with the x86 boot protocol.
This behavior is gated behind the KEXEC_FILE_FORCE_DTB flag.
Link: https://lkml.kernel.org/r/20250805211527.122367-3-makb@juniper.net
Signed-off-by: Brian Mak <makb@juniper.net>
Cc: Alexander Graf <graf@amazon.com>
Cc: Baoquan He <bhe@redhat.com>
Cc: Borislav Betkov <bp@alien8.de>
Cc: Dave Young <dyoung@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Rob Herring <robh@kernel.org>
Cc: Saravana Kannan <saravanak@google.com>
Cc: Thomas Gleinxer <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
When specifying fixed sized scratch areas, the parser only parses the
three scratch sizes and ignores the rest of the argument. This means the
argument can have any bogus trailing characters.
For example, "kho_scratch=256M,512M,512Mfoobar" results in successful
parsing:
[ 0.000000] KHO: scratch areas: lowmem: 256MiB global: 512MiB pernode: 512MiB
It is generally a good idea to parse arguments as strictly as possible.
In addition, if bogus trailing characters are allowed in the kho_scratch
argument, it is possible that some people might end up using them and
later extensions to the argument format will cause unexpected breakages.
Make sure the argument is fully consumed after all three scratch sizes are
parsed. With this change, the bogus argument
"kho_scratch=256M,512M,512Mfoobar" results in:
[ 0.000000] Malformed early option 'kho_scratch'
Link: https://lkml.kernel.org/r/20250826123817.64681-1-pratyush@kernel.org
Signed-off-by: Pratyush Yadav <pratyush@kernel.org>
Reviewed-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Cc: Alexander Graf <graf@amazon.com>
Cc: Baoquan He <bhe@redhat.com>
Cc: Changyuan Lyu <changyuanl@google.com>
Cc: Pratyush Yadav <pratyush@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Patch series "prctl: extend PR_SET_THP_DISABLE to only provide THPs when
advised", v5.
This will allow individual processes to opt-out of THP = "always" into THP
= "madvise", without affecting other workloads on the system. This has
been extensively discussed on the mailing list and has been summarized
very well by David in the first patch which also includes the links to
alternatives, please refer to the first patch commit message for the
motivation for this series.
Patch 1 adds the PR_THP_DISABLE_EXCEPT_ADVISED flag to implement this,
along with the MMF changes.
Patch 2 is a cleanup patch for tva_flags that will allow the forced
collapse case to be transmitted to vma_thp_disabled (which is done in
patch 3).
Patch 4 adds documentation for PR_SET_THP_DISABLE/PR_GET_THP_DISABLE.
Patches 6-7 implement the selftests for PR_SET_THP_DISABLE for completely
disabling THPs (old behaviour) and only enabling it at advise
(PR_THP_DISABLE_EXCEPT_ADVISED).
This patch (of 7):
People want to make use of more THPs, for example, moving from the "never"
system policy to "madvise", or from "madvise" to "always".
While this is great news for every THP desperately waiting to get
allocated out there, apparently there are some workloads that require a
bit of care during that transition: individual processes may need to
opt-out from this behavior for various reasons, and this should be
permitted without needing to make all other workloads on the system
similarly opt-out.
The following scenarios are imaginable:
(1) Switch from "none" system policy to "madvise"/"always", but keep THPs
disabled for selected workloads.
(2) Stay at "none" system policy, but enable THPs for selected
workloads, making only these workloads use the "madvise" or "always"
policy.
(3) Switch from "madvise" system policy to "always", but keep the
"madvise" policy for selected workloads: allocate THPs only when
advised.
(4) Stay at "madvise" system policy, but enable THPs even when not advised
for selected workloads -- "always" policy.
Once can emulate (2) through (1), by setting the system policy to
"madvise"/"always" while disabling THPs for all processes that don't want
THPs. It requires configuring all workloads, but that is a user-space
problem to sort out.
(4) can be emulated through (3) in a similar way.
Back when (1) was relevant in the past, as people started enabling THPs,
we added PR_SET_THP_DISABLE, so relevant workloads that were not ready yet
(i.e., used by Redis) were able to just disable THPs completely. Redis
still implements the option to use this interface to disable THPs
completely.
With PR_SET_THP_DISABLE, we added a way to force-disable THPs for a
workload -- a process, including fork+exec'ed process hierarchy. That
essentially made us support (1): simply disable THPs for all workloads
that are not ready for THPs yet, while still enabling THPs system-wide.
The quest for handling (3) and (4) started, but current approaches
(completely new prctl, options to set other policies per process,
alternatives to prctl -- mctrl, cgroup handling) don't look particularly
promising. Likely, the future will use bpf or something similar to
implement better policies, in particular to also make better decisions
about THP sizes to use, but this will certainly take a while as that work
just started.
Long story short: a simple enable/disable is not really suitable for the
future, so we're not willing to add completely new toggles.
While we could emulate (3)+(4) through (1)+(2) by simply disabling THPs
completely for these processes, this is a step backwards, because these
processes can no longer allocate THPs in regions where THPs were
explicitly advised: regions flagged as VM_HUGEPAGE. Apparently, that
imposes a problem for relevant workloads, because "not THPs" is certainly
worse than "THPs only when advised".
Could we simply relax PR_SET_THP_DISABLE, to "disable THPs unless not
explicitly advised by the app through MAD_HUGEPAGE"? *maybe*, but this
would change the documented semantics quite a bit, and the versatility to
use it for debugging purposes, so I am not 100% sure that is what we want
-- although it would certainly be much easier.
So instead, as an easy way forward for (3) and (4), add an option to
make PR_SET_THP_DISABLE disable *less* THPs for a process.
In essence, this patch:
(A) Adds PR_THP_DISABLE_EXCEPT_ADVISED, to be used as a flag in arg3
of prctl(PR_SET_THP_DISABLE) when disabling THPs (arg2 != 0).
prctl(PR_SET_THP_DISABLE, 1, PR_THP_DISABLE_EXCEPT_ADVISED).
(B) Makes prctl(PR_GET_THP_DISABLE) return 3 if
PR_THP_DISABLE_EXCEPT_ADVISED was set while disabling.
Previously, it would return 1 if THPs were disabled completely. Now
it returns the set flags as well: 3 if PR_THP_DISABLE_EXCEPT_ADVISED
was set.
(C) Renames MMF_DISABLE_THP to MMF_DISABLE_THP_COMPLETELY, to express
the semantics clearly.
Fortunately, there are only two instances outside of prctl() code.
(D) Adds MMF_DISABLE_THP_EXCEPT_ADVISED to express "no THP except for VMAs
with VM_HUGEPAGE" -- essentially "thp=madvise" behavior
Fortunately, we only have to extend vma_thp_disabled().
(E) Indicates "THP_enabled: 0" in /proc/pid/status only if THPs are
disabled completely
Only indicating that THPs are disabled when they are really disabled
completely, not only partially.
For now, we don't add another interface to obtained whether THPs
are disabled partially (PR_THP_DISABLE_EXCEPT_ADVISED was set). If
ever required, we could add a new entry.
The documented semantics in the man page for PR_SET_THP_DISABLE "is
inherited by a child created via fork(2) and is preserved across
execve(2)" is maintained. This behavior, for example, allows for
disabling THPs for a workload through the launching process (e.g., systemd
where we fork() a helper process to then exec()).
For now, MADV_COLLAPSE will *fail* in regions without VM_HUGEPAGE and
VM_NOHUGEPAGE. As MADV_COLLAPSE is a clear advise that user space thinks
a THP is a good idea, we'll enable that separately next (requiring a bit
of cleanup first).
There is currently not way to prevent that a process will not issue
PR_SET_THP_DISABLE itself to re-enable THP. There are not really known
users for re-enabling it, and it's against the purpose of the original
interface. So if ever required, we could investigate just forbidding to
re-enable them, or make this somehow configurable.
Link: https://lkml.kernel.org/r/20250815135549.130506-1-usamaarif642@gmail.com
Link: https://lkml.kernel.org/r/20250815135549.130506-2-usamaarif642@gmail.com
Acked-by: Zi Yan <ziy@nvidia.com>
Acked-by: Usama Arif <usamaarif642@gmail.com>
Tested-by: Usama Arif <usamaarif642@gmail.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Signed-off-by: Usama Arif <usamaarif642@gmail.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: Barry Song <baohua@kernel.org>
Cc: Dev Jain <dev.jain@arm.com>
Cc: Jann Horn <jannh@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Mariano Pache <npache@redhat.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Rik van Riel <riel@surriel.com>
Cc: Ryan Roberts <ryan.roberts@arm.com>
Cc: SeongJae Park <sj@kernel.org>
Cc: Shakeel Butt <shakeel.butt@linux.dev>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Yafang <laoar.shao@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Goto-san reported confusing pgpromote statistics where the
pgpromote_success count significantly exceeded pgpromote_candidate.
On a system with three nodes (nodes 0-1: DRAM 4GB, node 2: NVDIMM 4GB):
# Enable demotion only
echo 1 > /sys/kernel/mm/numa/demotion_enabled
numactl -m 0-1 memhog -r200 3500M >/dev/null &
pid=$!
sleep 2
numactl memhog -r100 2500M >/dev/null &
sleep 10
kill -9 $pid # terminate the 1st memhog
# Enable promotion
echo 2 > /proc/sys/kernel/numa_balancing
After a few seconds, we observeed `pgpromote_candidate < pgpromote_success`
$ grep -e pgpromote /proc/vmstat
pgpromote_success 2579
pgpromote_candidate 0
In this scenario, after terminating the first memhog, the conditions for
pgdat_free_space_enough() are quickly met, and triggers promotion.
However, these migrated pages are only counted for in PGPROMOTE_SUCCESS,
not in PGPROMOTE_CANDIDATE.
To solve these confusing statistics, introduce PGPROMOTE_CANDIDATE_NRL to
count the missed promotion pages. And also, not counting these pages into
PGPROMOTE_CANDIDATE is to avoid changing the existing algorithm or
performance of the promotion rate limit.
Link: https://lkml.kernel.org/r/20250901090122.124262-1-ruansy.fnst@fujitsu.com
Link: https://lkml.kernel.org/r/20250729035101.1601407-1-ruansy.fnst@fujitsu.com
Fixes: c6833e1000 ("memory tiering: rate limit NUMA migration throughput")
Co-developed-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Ruan Shiyang <ruansy.fnst@fujitsu.com>
Reported-by: Yasunori Gotou (Fujitsu) <y-goto@fujitsu.com>
Suggested-by: Huang Ying <ying.huang@linux.alibaba.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Reviewed-by: Huang Ying <ying.huang@linux.alibaba.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Ben Segall <bsegall@google.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Valentin Schneider <vschneid@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
rseq_need_restart() reads and clears task::rseq_event_mask with preemption
disabled to guard against the scheduler.
But membarrier() uses an IPI and sets the PREEMPT bit in the event mask
from the IPI, which leaves that RMW operation unprotected.
Use guard(irq) if CONFIG_MEMBARRIER is enabled to fix that.
Fixes: 2a36ab717e ("rseq/membarrier: Add MEMBARRIER_CMD_PRIVATE_EXPEDITED_RSEQ")
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Boqun Feng <boqun.feng@gmail.com>
Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: stable@vger.kernel.org
Currently if a user enqueue a work item using schedule_delayed_work() the
used wq is "system_wq" (per-cpu wq) while queue_delayed_work() use
WORK_CPU_UNBOUND (used when a cpu is not specified). The same applies to
schedule_work() that is using system_wq and queue_work(), that makes use
again of WORK_CPU_UNBOUND.
This lack of consistentcy cannot be addressed without refactoring the API.
alloc_workqueue() treats all queues as per-CPU by default, while unbound
workqueues must opt-in via WQ_UNBOUND.
This default is suboptimal: most workloads benefit from unbound queues,
allowing the scheduler to place worker threads where they’re needed and
reducing noise when CPUs are isolated.
This default is suboptimal: most workloads benefit from unbound queues,
allowing the scheduler to place worker threads where they’re needed and
reducing noise when CPUs are isolated.
This patch adds a new WQ_PERCPU flag to explicitly request the use of
the per-CPU behavior. Both flags coexist for one release cycle to allow
callers to transition their calls.
Once migration is complete, WQ_UNBOUND can be removed and unbound will
become the implicit default.
With the introduction of the WQ_PERCPU flag (equivalent to !WQ_UNBOUND),
any alloc_workqueue() caller that doesn’t explicitly specify WQ_UNBOUND
must now use WQ_PERCPU.
All existing users have been updated accordingly.
Suggested-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Currently if a user enqueue a work item using schedule_delayed_work() the
used wq is "system_wq" (per-cpu wq) while queue_delayed_work() use
WORK_CPU_UNBOUND (used when a cpu is not specified). The same applies to
schedule_work() that is using system_wq and queue_work(), that makes use
again of WORK_CPU_UNBOUND.
This lack of consistentcy cannot be addressed without refactoring the API.
system_unbound_wq should be the default workqueue so as not to enforce
locality constraints for random work whenever it's not required.
Adding system_dfl_wq to encourage its use when unbound work should be used.
queue_work() / queue_delayed_work() / mod_delayed_work() will now use the
new unbound wq: whether the user still use the old wq a warn will be
printed along with a wq redirect to the new one.
The old system_unbound_wq will be kept for a few release cycles.
Suggested-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Cross-merge networking fixes after downstream PR (net-6.17-rc6).
Conflicts:
net/netfilter/nft_set_pipapo.c
net/netfilter/nft_set_pipapo_avx2.c
c4eaca2e10 ("netfilter: nft_set_pipapo: don't check genbit from packetpath lookups")
84c1da7b38 ("netfilter: nft_set_pipapo: use avx2 algorithm for insertions too")
Only trivial adjacent changes (in a doc and a Makefile).
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
