| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Pull misc vfs updates from Christian Brauner:
"Features:
- Add CONFIG_DEBUG_VFS infrastucture:
- Catch invalid modes in open
- Use the new debug macros in inode_set_cached_link()
- Use debug-only asserts around fd allocation and install
- Place f_ref to 3rd cache line in struct file to resolve false
sharing
Cleanups:
- Start using anon_inode_getfile_fmode() helper in various places
- Don't take f_lock during SEEK_CUR if exclusion is guaranteed by
f_pos_lock
- Add unlikely() to kcmp()
- Remove legacy ->remount_fs method from ecryptfs after port to the
new mount api
- Remove invalidate_inodes() in favour of evict_inodes()
- Simplify ep_busy_loopER by removing unused argument
- Avoid mmap sem relocks when coredumping with many missing pages
- Inline getname()
- Inline new_inode_pseudo() and de-staticize alloc_inode()
- Dodge an atomic in putname if ref == 1
- Consistently deref the files table with rcu_dereference_raw()
- Dedup handling of struct filename init and refcounts bumps
- Use wq_has_sleeper() in end_dir_add()
- Drop the lock trip around I_NEW wake up in evict()
- Load the ->i_sb pointer once in inode_sb_list_{add,del}
- Predict not reaching the limit in alloc_empty_file()
- Tidy up do_sys_openat2() with likely/unlikely
- Call inode_sb_list_add() outside of inode hash lock
- Sort out fd allocation vs dup2 race commentary
- Turn page_offset() into a wrapper around folio_pos()
- Remove locking in exportfs around ->get_parent() call
- try_lookup_one_len() does not need any locks in autofs
- Fix return type of several functions from long to int in open
- Fix return type of several functions from long to int in ioctls
Fixes:
- Fix watch queue accounting mismatch"
* tag 'vfs-6.15-rc1.misc' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs: (30 commits)
fs: sort out fd allocation vs dup2 race commentary, take 2
fs: call inode_sb_list_add() outside of inode hash lock
fs: tidy up do_sys_openat2() with likely/unlikely
fs: predict not reaching the limit in alloc_empty_file()
fs: load the ->i_sb pointer once in inode_sb_list_{add,del}
fs: drop the lock trip around I_NEW wake up in evict()
fs: use wq_has_sleeper() in end_dir_add()
VFS/autofs: try_lookup_one_len() does not need any locks
fs: dedup handling of struct filename init and refcounts bumps
fs: consistently deref the files table with rcu_dereference_raw()
exportfs: remove locking around ->get_parent() call.
fs: use debug-only asserts around fd allocation and install
fs: dodge an atomic in putname if ref == 1
vfs: Remove invalidate_inodes()
ecryptfs: remove NULL remount_fs from super_operations
watch_queue: fix pipe accounting mismatch
fs: place f_ref to 3rd cache line in struct file to resolve false sharing
epoll: simplify ep_busy_loop by removing always 0 argument
fs: Turn page_offset() into a wrapper around folio_pos()
kcmp: improve performance adding an unlikely hint to task comparisons
...
|
|
Merge updates related to system sleep for 6.15-rc1 including fixes,
cleanups and a rework of the "smart suspend" driver flag handling to
avoid issues that may occur when drivers using it depend on some other
drivers:
- Rework the handling of the "smart suspend" driver flag in the PM core
to avoid issues hat may occur when drivers using it depend on some
other drivers and clean up the related PM core code (Rafael Wysocki,
Colin Ian King).
- Fix the handling of devices with the power.direct_complete flag set
if device_suspend() returns an error for at least one device to avoid
situations in which some of them may not be resumed (Rafael Wysocki).
- Use mutex_trylock() in hibernate_compressor_param_set() to avoid a
possible deadlock that may occur if the "compressor" hibernation
module parameter is accessed during the registration of a new
ieee80211 device (Lizhi Xu).
- Suppress sleeping parent warning in device_pm_add() in the case when
new children are added under a device with the power.direct_complete
set after it has been processed by device_resume() (Xu Yang).
- Remove needless return in three void functions related to system
wakeup (Zijun Hu).
- Replace deprecated kmap_atomic() with kmap_local_page() in the
hibernation core code (David Reaver).
- Remove unused helper functions related to system sleep (David Alan
Gilbert).
- Clean up s2idle_enter() so it does not lock and unlock CPU offline
in vain and update comments in it (Ulf Hansson).
- Clean up broken white space in dpm_wait_for_children() (Geert
Uytterhoeven).
* pm-sleep:
PM: sleep: Fix bit masking operation
PM: sleep: Fix handling devices with direct_complete set on errors
PM: sleep: core: Fix indentation in dpm_wait_for_children()
PM: s2idle: Extend comment in s2idle_enter()
PM: s2idle: Drop redundant locks when entering s2idle
PM: sleep: Remove unused pm_generic_ wrappers
PM: sleep: Rearrange dpm_async_fn() and async state clearing
PM: sleep: Rename power.async_in_progress to power.work_in_progress
PM: core: Tweak pm_runtime_block_if_disabled() return value
PM: runtime: Convert pm_runtime_blocked() to static inline
PM: sleep: Update power.smart_suspend under PM spinlock
PM: sleep: Adjust check before setting power.must_resume
PM: wakeup: Remove needless return in three void APIs
PM: sleep: Suppress sleeping parent warning in special case
PM: hibernate: Avoid deadlock in hibernate_compressor_param_set()
PM: sleep: Avoid unnecessary checks in device_prepare_smart_suspend()
PM: sleep: Use DPM_FLAG_SMART_SUSPEND conditionally
PM: runtime: Introduce pm_runtime_blocked()
PM: Block enabling of runtime PM during system suspend
PM: hibernate: Replace deprecated kmap_atomic() with kmap_local_page()
|
|
Merge Energy Model handling code updates and updates of the runtime PM
core code for 6.15-rc1:
- Clean up the Energy Model handling code somewhat (Rafael Wysocki).
- Use kfree_rcu() to simplify the handling of runtime Energy Model
updates (Li RongQing).
- Add an entry for the Energy Model framework to MAINTAINERS as
properly maintained (Lukasz Luba).
- Address RCU-related sparse warnings in the Energy Model code (Rafael
Wysocki).
- Remove ENERGY_MODEL dependency on SMP and allow it to be selected
when DEVFREQ is set without CPUFREQ so it can be used on a wider
range of systems (Jeson Gao).
- Unify error handling during runtime suspend and runtime resume in the
core to help drivers to implement more consistent runtime PM error
handling (Rafael Wysocki).
- Drop a redundant check from pm_runtime_force_resume() and rearrange
documentation related to __pm_runtime_disable() (Rafael Wysocki).
* pm-em:
PM: EM: Rework the depends on for CONFIG_ENERGY_MODEL
PM: EM: Address RCU-related sparse warnings
PM: EM: Consify two parameters of em_dev_register_perf_domain()
MAINTAINERS: Add Energy Model framework as properly maintained
PM: EM: use kfree_rcu() to simplify the code
PM: EM: Slightly reduce em_check_capacity_update() overhead
PM: EM: Drop unused parameter from em_adjust_new_capacity()
* pm-runtime:
PM: runtime: Unify error handling during suspend and resume
PM: runtime: Drop status check from pm_runtime_force_resume()
PM: Rearrange documentation related to __pm_runtime_disable()
|
|
Convert the event_hash array in trace_output.c to use the generic
hashtable implementation from hashtable.h instead of the manually
implemented hash table.
This simplifies the code and makes it more maintainable by using the
standard hashtable API defined in hashtable.h.
Rename EVENT_HASHSIZE to EVENT_HASH_BITS to properly reflect its new
meaning as the number of bits for the hashtable size.
Link: https://lore.kernel.org/20250323132800.3010783-1-sashal@kernel.org
Link: https://lore.kernel.org/20250319190545.3058319-1-sashal@kernel.org
Signed-off-by: Sasha Levin <sashal@kernel.org>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
Currently, using synth_event_delete() will fail if the event is being
used (tracing in progress), but that is normally done in the module exit
function. At that stage, failing is problematic as returning a non-zero
status means the module will become locked (impossible to unload or
reload again).
Instead, ensure the module exit function does not get called in the
first place by increasing the module refcnt when the event is enabled.
Cc: stable@vger.kernel.org
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Fixes: 35ca5207c2d11 ("tracing: Add synthetic event command generation functions")
Link: https://lore.kernel.org/20250318180906.226841-1-douglas.raillard@arm.com
Signed-off-by: Douglas Raillard <douglas.raillard@arm.com>
Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
When __ftrace_event_enable_disable invokes the class callback to
unregister the event, the return value is not reported up to the
caller, hence leading to event unregister failures being silently
ignored.
This patch assigns the ret variable to the invocation of the
event unregister callback, so that its return value is stored
and reported to the caller, and it raises a warning in case
of error.
Link: https://lore.kernel.org/20250321170821.101403-1-gpaoloni@redhat.com
Signed-off-by: Gabriele Paoloni <gpaoloni@redhat.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
Lockdep reports this deadlock log:
osnoise: could not start sampling thread
============================================
WARNING: possible recursive locking detected
--------------------------------------------
CPU0
----
lock(cpu_hotplug_lock);
lock(cpu_hotplug_lock);
Call Trace:
<TASK>
print_deadlock_bug+0x282/0x3c0
__lock_acquire+0x1610/0x29a0
lock_acquire+0xcb/0x2d0
cpus_read_lock+0x49/0x120
stop_per_cpu_kthreads+0x7/0x60
start_kthread+0x103/0x120
osnoise_hotplug_workfn+0x5e/0x90
process_one_work+0x44f/0xb30
worker_thread+0x33e/0x5e0
kthread+0x206/0x3b0
ret_from_fork+0x31/0x50
ret_from_fork_asm+0x11/0x20
</TASK>
This is the deadlock scenario:
osnoise_hotplug_workfn()
guard(cpus_read_lock)(); // first lock call
start_kthread(cpu)
if (IS_ERR(kthread)) {
stop_per_cpu_kthreads(); {
cpus_read_lock(); // second lock call. Cause the AA deadlock
}
}
It is not necessary to call stop_per_cpu_kthreads() which stops osnoise
kthread for every other CPUs in the system if a failure occurs during
hotplug of a certain CPU.
For start_per_cpu_kthreads(), if the start_kthread() call fails,
this function calls stop_per_cpu_kthreads() to handle the error.
Therefore, similarly, there is no need to call stop_per_cpu_kthreads()
again within start_kthread().
So just remove stop_per_cpu_kthreads() from start_kthread to solve this issue.
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/20250321095249.2739397-1-ranxiaokai627@163.com
Fixes: c8895e271f79 ("trace/osnoise: Support hotplug operations")
Signed-off-by: Ran Xiaokai <ran.xiaokai@zte.com.cn>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
The vast majority of ftrace event print fmt consist of a space-separated
field=value pair. Synthetic event currently use a comma-separated
field=value pair, which sticks out from events created via more
classical means.
Align the format of synth events so they look just like any other event,
for better consistency and less headache when doing crude text-based
data processing.
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Link: https://lore.kernel.org/20250319215028.1680278-1-douglas.raillard@arm.com
Signed-off-by: Douglas Raillard <douglas.raillard@arm.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
syzbot reported the following splat [0].
In check_atomic_load/store(), register validity is not checked before
atomic_ptr_type_ok(). This causes the out-of-bounds read in is_ctx_reg()
called from atomic_ptr_type_ok() when the register number is MAX_BPF_REG
or greater.
Call check_load_mem()/check_store_reg() before atomic_ptr_type_ok()
to avoid the OOB read.
However, some tests introduced by commit ff3afe5da998 ("selftests/bpf: Add
selftests for load-acquire and store-release instructions") assume
calling atomic_ptr_type_ok() before checking register validity.
Therefore the swapping of order unintentionally changes verifier messages
of these tests.
For example in the test load_acquire_from_pkt_pointer(), expected message
is 'BPF_ATOMIC loads from R2 pkt is not allowed' although actual messages
are different.
validate_msgs:FAIL:754 expect_msg
VERIFIER LOG:
=============
Global function load_acquire_from_pkt_pointer() doesn't return scalar. Only those are supported.
0: R1=ctx() R10=fp0
; asm volatile ( @ verifier_load_acquire.c:140
0: (61) r2 = *(u32 *)(r1 +0) ; R1=ctx() R2_w=pkt(r=0)
1: (d3) r0 = load_acquire((u8 *)(r2 +0))
invalid access to packet, off=0 size=1, R2(id=0,off=0,r=0)
R2 offset is outside of the packet
processed 2 insns (limit 1000000) max_states_per_insn 0 total_states 0 peak_states 0 mark_read 0
=============
EXPECTED SUBSTR: 'BPF_ATOMIC loads from R2 pkt is not allowed'
#505/19 verifier_load_acquire/load-acquire from pkt pointer:FAIL
This is because instructions in the test don't pass check_load_mem() and
therefore don't enter the atomic_ptr_type_ok() path.
In this case, we have to modify instructions so that they pass the
check_load_mem() and trigger atomic_ptr_type_ok().
Similarly for store-release tests, we need to modify instructions so that
they pass check_store_reg().
Like load_acquire_from_pkt_pointer(), modify instructions in:
load_acquire_from_sock_pointer()
store_release_to_ctx_pointer()
store_release_to_pkt_pointer()
Also in store_release_to_sock_pointer(), check_store_reg() returns error
early and atomic_ptr_type_ok() is not triggered, since write to sock
pointer is not possible in general.
We might be able to remove the test, but for now let's leave it and just
change the expected message.
[0]
BUG: KASAN: slab-out-of-bounds in is_ctx_reg kernel/bpf/verifier.c:6185 [inline]
BUG: KASAN: slab-out-of-bounds in atomic_ptr_type_ok+0x3d7/0x550 kernel/bpf/verifier.c:6223
Read of size 4 at addr ffff888141b0d690 by task syz-executor143/5842
CPU: 1 UID: 0 PID: 5842 Comm: syz-executor143 Not tainted 6.14.0-rc3-syzkaller-gf28214603dc6 #0
Call Trace:
<TASK>
__dump_stack lib/dump_stack.c:94 [inline]
dump_stack_lvl+0x241/0x360 lib/dump_stack.c:120
print_address_description mm/kasan/report.c:408 [inline]
print_report+0x16e/0x5b0 mm/kasan/report.c:521
kasan_report+0x143/0x180 mm/kasan/report.c:634
is_ctx_reg kernel/bpf/verifier.c:6185 [inline]
atomic_ptr_type_ok+0x3d7/0x550 kernel/bpf/verifier.c:6223
check_atomic_store kernel/bpf/verifier.c:7804 [inline]
check_atomic kernel/bpf/verifier.c:7841 [inline]
do_check+0x89dd/0xedd0 kernel/bpf/verifier.c:19334
do_check_common+0x1678/0x2080 kernel/bpf/verifier.c:22600
do_check_main kernel/bpf/verifier.c:22691 [inline]
bpf_check+0x165c8/0x1cca0 kernel/bpf/verifier.c:23821
Reported-by: syzbot+a5964227adc0f904549c@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=a5964227adc0f904549c
Tested-by: syzbot+a5964227adc0f904549c@syzkaller.appspotmail.com
Fixes: e24bbad29a8d ("bpf: Introduce load-acquire and store-release instructions")
Fixes: ff3afe5da998 ("selftests/bpf: Add selftests for load-acquire and store-release instructions")
Signed-off-by: Kohei Enju <enjuk@amazon.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20250322045340.18010-5-enjuk@amazon.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Kairui reported a UAF issue in print_graph_function_flags() during
ftrace stress testing [1]. This issue can be reproduced if puting a
'mdelay(10)' after 'mutex_unlock(&trace_types_lock)' in s_start(),
and executing the following script:
$ echo function_graph > current_tracer
$ cat trace > /dev/null &
$ sleep 5 # Ensure the 'cat' reaches the 'mdelay(10)' point
$ echo timerlat > current_tracer
The root cause lies in the two calls to print_graph_function_flags
within print_trace_line during each s_show():
* One through 'iter->trace->print_line()';
* Another through 'event->funcs->trace()', which is hidden in
print_trace_fmt() before print_trace_line returns.
Tracer switching only updates the former, while the latter continues
to use the print_line function of the old tracer, which in the script
above is print_graph_function_flags.
Moreover, when switching from the 'function_graph' tracer to the
'timerlat' tracer, s_start only calls graph_trace_close of the
'function_graph' tracer to free 'iter->private', but does not set
it to NULL. This provides an opportunity for 'event->funcs->trace()'
to use an invalid 'iter->private'.
To fix this issue, set 'iter->private' to NULL immediately after
freeing it in graph_trace_close(), ensuring that an invalid pointer
is not passed to other tracers. Additionally, clean up the unnecessary
'iter->private = NULL' during each 'cat trace' when using wakeup and
irqsoff tracers.
[1] https://lore.kernel.org/all/20231112150030.84609-1-ryncsn@gmail.com/
Cc: stable@vger.kernel.org
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Zheng Yejian <zhengyejian1@huawei.com>
Link: https://lore.kernel.org/20250320122137.23635-1-wutengda@huaweicloud.com
Fixes: eecb91b9f98d ("tracing: Fix memleak due to race between current_tracer and trace")
Closes: https://lore.kernel.org/all/CAMgjq7BW79KDSCyp+tZHjShSzHsScSiJxn5ffskp-QzVM06fxw@mail.gmail.com/
Reported-by: Kairui Song <kasong@tencent.com>
Signed-off-by: Tengda Wu <wutengda@huaweicloud.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
CONFIG_TRACE_BRANCH_PROFILING inserts a call to ftrace_likely_update()
for each use of likely() or unlikely(). That breaks noinstr rules if
the affected function is annotated as noinstr.
Disable branch profiling for files with noinstr functions. In addition
to some individual files, this also includes the entire arch/x86
subtree, as well as the kernel/entry, drivers/cpuidle, and drivers/idle
directories, all of which are noinstr-heavy.
Due to the nature of how sched binaries are built by combining multiple
.c files into one, branch profiling is disabled more broadly across the
sched code than would otherwise be needed.
This fixes many warnings like the following:
vmlinux.o: warning: objtool: do_syscall_64+0x40: call to ftrace_likely_update() leaves .noinstr.text section
vmlinux.o: warning: objtool: __rdgsbase_inactive+0x33: call to ftrace_likely_update() leaves .noinstr.text section
vmlinux.o: warning: objtool: handle_bug.isra.0+0x198: call to ftrace_likely_update() leaves .noinstr.text section
...
Reported-by: Ingo Molnar <mingo@kernel.org>
Suggested-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Josh Poimboeuf <jpoimboe@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/r/fb94fc9303d48a5ed370498f54500cc4c338eb6d.1742586676.git.jpoimboe@kernel.org
|
|
Improve readability and maintainability by replacing a hard coded string
allocation and formatting by using the kasprintf() helper.
It also eliminates the GCC compiler warning (with CONFIG_WERROR=y, which
is default, it becomes an error:
kernel/relay.c:357:42: error: `snprintf' output may be truncated before the last format character [-Werror=format-truncation=]
Link: https://lkml.kernel.org/r/20250317212948.1811176-1-andriy.shevchenko@linux.intel.com
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Replace open coded variant of DEFINE_RES(). No functional changes intended.
Link: https://lkml.kernel.org/r/20250317181412.1560630-5-andriy.shevchenko@linux.intel.com
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Replace open coded variants of DEFINE_RES_*_NAMED().
Link: https://lkml.kernel.org/r/20250317181412.1560630-4-andriy.shevchenko@linux.intel.com
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Replace open coded variant of DEFINE_RES_NAMED_DESC().
Link: https://lkml.kernel.org/r/20250317181412.1560630-3-andriy.shevchenko@linux.intel.com
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Patch series "hung_task: Dump the blocking task stacktrace", v4.
The hung_task detector is very useful for detecting the lockup. However,
since it only dumps the blocked (uninterruptible sleep) processes, it is
not enough to identify the root cause of that lockup.
For example, if a process holds a mutex and sleep an event in
interruptible state long time, the other processes will wait on the mutex
in uninterruptible state. In this case, the waiter processes are dumped,
but the blocker process is not shown because it is sleep in interruptible
state.
This adds a feature to dump the blocker task which holds a mutex
when detecting a hung task. e.g.
INFO: task cat:115 blocked for more than 122 seconds.
Not tainted 6.14.0-rc3-00003-ga8946be3de00 #156
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
task:cat state:D stack:13432 pid:115 tgid:115 ppid:106 task_flags:0x400100 flags:0x00000002
Call Trace:
<TASK>
__schedule+0x731/0x960
? schedule_preempt_disabled+0x54/0xa0
schedule+0xb7/0x140
? __mutex_lock+0x51b/0xa60
? __mutex_lock+0x51b/0xa60
schedule_preempt_disabled+0x54/0xa0
__mutex_lock+0x51b/0xa60
read_dummy+0x23/0x70
full_proxy_read+0x6a/0xc0
vfs_read+0xc2/0x340
? __pfx_direct_file_splice_eof+0x10/0x10
? do_sendfile+0x1bd/0x2e0
ksys_read+0x76/0xe0
do_syscall_64+0xe3/0x1c0
? exc_page_fault+0xa9/0x1d0
entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x4840cd
RSP: 002b:00007ffe99071828 EFLAGS: 00000246 ORIG_RAX: 0000000000000000
RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00000000004840cd
RDX: 0000000000001000 RSI: 00007ffe99071870 RDI: 0000000000000003
RBP: 00007ffe99071870 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000001000000 R11: 0000000000000246 R12: 0000000000001000
R13: 00000000132fd3a0 R14: 0000000000000001 R15: ffffffffffffffff
</TASK>
INFO: task cat:115 is blocked on a mutex likely owned by task cat:114.
task:cat state:S stack:13432 pid:114 tgid:114 ppid:106 task_flags:0x400100 flags:0x00000002
Call Trace:
<TASK>
__schedule+0x731/0x960
? schedule_timeout+0xa8/0x120
schedule+0xb7/0x140
schedule_timeout+0xa8/0x120
? __pfx_process_timeout+0x10/0x10
msleep_interruptible+0x3e/0x60
read_dummy+0x2d/0x70
full_proxy_read+0x6a/0xc0
vfs_read+0xc2/0x340
? __pfx_direct_file_splice_eof+0x10/0x10
? do_sendfile+0x1bd/0x2e0
ksys_read+0x76/0xe0
do_syscall_64+0xe3/0x1c0
? exc_page_fault+0xa9/0x1d0
entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x4840cd
RSP: 002b:00007ffe3e0147b8 EFLAGS: 00000246 ORIG_RAX: 0000000000000000
RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00000000004840cd
RDX: 0000000000001000 RSI: 00007ffe3e014800 RDI: 0000000000000003
RBP: 00007ffe3e014800 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000001000000 R11: 0000000000000246 R12: 0000000000001000
R13: 000000001a0a93a0 R14: 0000000000000001 R15: ffffffffffffffff
</TASK>
TBD: We can extend this feature to cover other locks like rwsem and
rt_mutex, but rwsem requires to dump all the tasks which acquire and wait
that rwsem. We can follow the waiter link but the output will be a bit
different compared with mutex case.
This patch (of 2):
The "hung_task" shows a long-time uninterruptible slept task, but most
often, it's blocked on a mutex acquired by another task. Without dumping
such a task, investigating the root cause of the hung task problem is very
difficult.
This introduce task_struct::blocker_mutex to point the mutex lock which
this task is waiting for. Since the mutex has "owner" information, we can
find the owner task and dump it with hung tasks.
Note: the owner can be changed while dumping the owner task, so
this is "likely" the owner of the mutex.
With this change, the hung task shows blocker task's info like below;
INFO: task cat:115 blocked for more than 122 seconds.
Not tainted 6.14.0-rc3-00003-ga8946be3de00 #156
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
task:cat state:D stack:13432 pid:115 tgid:115 ppid:106 task_flags:0x400100 flags:0x00000002
Call Trace:
<TASK>
__schedule+0x731/0x960
? schedule_preempt_disabled+0x54/0xa0
schedule+0xb7/0x140
? __mutex_lock+0x51b/0xa60
? __mutex_lock+0x51b/0xa60
schedule_preempt_disabled+0x54/0xa0
__mutex_lock+0x51b/0xa60
read_dummy+0x23/0x70
full_proxy_read+0x6a/0xc0
vfs_read+0xc2/0x340
? __pfx_direct_file_splice_eof+0x10/0x10
? do_sendfile+0x1bd/0x2e0
ksys_read+0x76/0xe0
do_syscall_64+0xe3/0x1c0
? exc_page_fault+0xa9/0x1d0
entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x4840cd
RSP: 002b:00007ffe99071828 EFLAGS: 00000246 ORIG_RAX: 0000000000000000
RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00000000004840cd
RDX: 0000000000001000 RSI: 00007ffe99071870 RDI: 0000000000000003
RBP: 00007ffe99071870 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000001000000 R11: 0000000000000246 R12: 0000000000001000
R13: 00000000132fd3a0 R14: 0000000000000001 R15: ffffffffffffffff
</TASK>
INFO: task cat:115 is blocked on a mutex likely owned by task cat:114.
task:cat state:S stack:13432 pid:114 tgid:114 ppid:106 task_flags:0x400100 flags:0x00000002
Call Trace:
<TASK>
__schedule+0x731/0x960
? schedule_timeout+0xa8/0x120
schedule+0xb7/0x140
schedule_timeout+0xa8/0x120
? __pfx_process_timeout+0x10/0x10
msleep_interruptible+0x3e/0x60
read_dummy+0x2d/0x70
full_proxy_read+0x6a/0xc0
vfs_read+0xc2/0x340
? __pfx_direct_file_splice_eof+0x10/0x10
? do_sendfile+0x1bd/0x2e0
ksys_read+0x76/0xe0
do_syscall_64+0xe3/0x1c0
? exc_page_fault+0xa9/0x1d0
entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x4840cd
RSP: 002b:00007ffe3e0147b8 EFLAGS: 00000246 ORIG_RAX: 0000000000000000
RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00000000004840cd
RDX: 0000000000001000 RSI: 00007ffe3e014800 RDI: 0000000000000003
RBP: 00007ffe3e014800 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000001000000 R11: 0000000000000246 R12: 0000000000001000
R13: 000000001a0a93a0 R14: 0000000000000001 R15: ffffffffffffffff
</TASK>
[akpm@linux-foundation.org: implement debug_show_blocker() in C rather than in CPP]
Link: https://lkml.kernel.org/r/174046694331.2194069.15472952050240807469.stgit@mhiramat.tok.corp.google.com
Link: https://lkml.kernel.org/r/174046695384.2194069.16796289525958195643.stgit@mhiramat.tok.corp.google.com
Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Reviewed-by: Waiman Long <longman@redhat.com>
Reviewed-by: Lance Yang <ioworker0@gmail.com>
Reviewed-by: Sergey Senozhatsky <senozhatsky@chromium.org>
Cc: Anna Schumaker <anna.schumaker@oracle.com>
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Joel Granados <joel.granados@kernel.org>
Cc: Kent Overstreet <kent.overstreet@linux.dev>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tomasz Figa <tfiga@chromium.org>
Cc: Will Deacon <will@kernel.org>
Cc: Yongliang Gao <leonylgao@tencent.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Replace __vmalloc_node_range() by __vmalloc_node(). The last variant
requires less parameters and it uses exactly the same arguments which are
partly now hidden inside __vmalloc_node().
This change does not change any functionality. It makes the code a bit
simpler.
Link: https://lkml.kernel.org/r/20250317163614.166502-1-urezki@gmail.com
Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
When tracepoint_debug is set, we may get the output in kernel log:
[ 380.013843] Probe 0 : 00000000f0d68cda
It is not readable, so change to print the function symbol.
After this patch, the output may becomes:
[ 55.225555] Probe 0 : perf_trace_sched_wakeup_template+0x0/0x20
Link: https://lore.kernel.org/20250307033858.4134-1-shijie@os.amperecomputing.com
Signed-off-by: Huang Shijie <shijie@os.amperecomputing.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
Lei Chen raised an issue with CLOCK_MONOTONIC_COARSE seeing time
inconsistencies.
Lei tracked down that this was being caused by the adjustment
tk->tkr_mono.xtime_nsec -= offset;
which is made to compensate for the unaccumulated cycles in offset when the
multiplicator is adjusted forward, so that the non-_COARSE clockids don't
see inconsistencies.
However, the _COARSE clockid getter functions use the adjusted xtime_nsec
value directly and do not compensate the negative offset via the
clocksource delta multiplied with the new multiplicator. In that case the
caller can observe time going backwards in consecutive calls.
By design, this negative adjustment should be fine, because the logic run
from timekeeping_adjust() is done after it accumulated approximately
multiplicator * interval_cycles
into xtime_nsec. The accumulated value is always larger then the
mult_adj * offset
value, which is subtracted from xtime_nsec. Both operations are done
together under the tk_core.lock, so the net change to xtime_nsec is always
always be positive.
However, do_adjtimex() calls into timekeeping_advance() as well, to to
apply the NTP frequency adjustment immediately. In this case,
timekeeping_advance() does not return early when the offset is smaller then
interval_cycles. In that case there is no time accumulated into
xtime_nsec. But the subsequent call into timekeeping_adjust(), which
modifies the multiplicator, subtracts from xtime_nsec to correct
for the new multiplicator.
Here because there was no accumulation, xtime_nsec becomes smaller than
before, which opens a window up to the next accumulation, where the _COARSE
clockid getters, which don't compensate for the offset, can observe the
inconsistency.
To fix this, rework the timekeeping_advance() logic so that when invoked
from do_adjtimex(), the time is immediately forwarded to accumulate also
the sub-interval portion into xtime. That means the remaining offset
becomes zero and the subsequent multiplier adjustment therefore does not
modify xtime_nsec.
There is another related inconsistency. If xtime is forwarded due to the
instantaneous multiplier adjustment, the NTP error, which was accumulated
with the previous setting, becomes meaningless.
Therefore clear the NTP error as well, after forwarding the clock for the
instantaneous multiplier update.
Fixes: da15cfdae033 ("time: Introduce CLOCK_REALTIME_COARSE")
Reported-by: Lei Chen <lei.chen@smartx.com>
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/20250320200306.1712599-1-jstultz@google.com
Closes: https://lore.kernel.org/lkml/20250310030004.3705801-1-lei.chen@smartx.com/
|
|
Pull scheduler fix from Ingo Molnar:
"Revert a scheduler performance optimization that regressed other
workloads"
* tag 'sched-urgent-2025-03-21' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
Revert "sched/core: Reduce cost of sched_move_task when config autogroup"
|
|
Replace the legacy crypto compression interface with the new acomp
interface.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
|
|
Setting pci_msi_ignore_mask inhibits the toggling of the mask bit for both
MSI and MSI-X entries globally, regardless of the IRQ chip they are using.
Only Xen sets the pci_msi_ignore_mask when routing physical interrupts over
event channels, to prevent PCI code from attempting to toggle the maskbit,
as it's Xen that controls the bit.
However, the pci_msi_ignore_mask being global will affect devices that use
MSI interrupts but are not routing those interrupts over event channels
(not using the Xen pIRQ chip). One example is devices behind a VMD PCI
bridge. In that scenario the VMD bridge configures MSI(-X) using the
normal IRQ chip (the pIRQ one in the Xen case), and devices behind the
bridge configure the MSI entries using indexes into the VMD bridge MSI
table. The VMD bridge then demultiplexes such interrupts and delivers to
the destination device(s). Having pci_msi_ignore_mask set in that scenario
prevents (un)masking of MSI entries for devices behind the VMD bridge.
Move the signaling of no entry masking into the MSI domain flags, as that
allows setting it on a per-domain basis. Set it for the Xen MSI domain
that uses the pIRQ chip, while leaving it unset for the rest of the
cases.
Remove pci_msi_ignore_mask at once, since it was only used by Xen code, and
with Xen dropping usage the variable is unneeded.
This fixes using devices behind a VMD bridge on Xen PV hardware domains.
Albeit Devices behind a VMD bridge are not known to Xen, that doesn't mean
Linux cannot use them. By inhibiting the usage of
VMD_FEAT_CAN_BYPASS_MSI_REMAP and the removal of the pci_msi_ignore_mask
bodge devices behind a VMD bridge do work fine when use from a Linux Xen
hardware domain. That's the whole point of the series.
Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Juergen Gross <jgross@suse.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
Message-ID: <20250219092059.90850-4-roger.pau@citrix.com>
Signed-off-by: Juergen Gross <jgross@suse.com>
|
|
Pull dma-mapping fix from Marek Szyprowski:
- fix missing clear bdr in check_ram_in_range_map() (Baochen Qiang)
* tag 'dma-mapping-6.14-2025-03-21' of git://git.kernel.org/pub/scm/linux/kernel/git/mszyprowski/linux:
dma-mapping: fix missing clear bdr in check_ram_in_range_map()
|
|
This patch adds struct_ops context information to struct bpf_prog_aux.
This context information will be used in the kfunc filter.
Currently the added context information includes struct_ops member
offset and a pointer to struct bpf_struct_ops.
Signed-off-by: Juntong Deng <juntong.deng@outlook.com>
Signed-off-by: Amery Hung <ameryhung@gmail.com>
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Link: https://patch.msgid.link/20250319215358.2287371-2-ameryhung@gmail.com
|
|
Cross-merge networking fixes after downstream PR (net-6.14-rc8).
Conflict:
tools/testing/selftests/net/Makefile
03544faad761 ("selftest: net: add proc_net_pktgen")
3ed61b8938c6 ("selftests: net: test for lwtunnel dst ref loops")
tools/testing/selftests/net/config:
85cb3711acb8 ("selftests: net: Add test cases for link and peer netns")
3ed61b8938c6 ("selftests: net: test for lwtunnel dst ref loops")
Adjacent commits:
tools/testing/selftests/net/Makefile
c935af429ec2 ("selftests: net: add support for testing SO_RCVMARK and SO_RCVPRIORITY")
355d940f4d5a ("Revert "selftests: Add IPv6 link-local address generation tests for GRE devices."")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Now that the rstat lock is being re-acquired on every CPU iteration in
cgroup_rstat_flush_locked(), having the initially acquire the lock is
unnecessary and unclear.
Inline cgroup_rstat_flush_locked() into cgroup_rstat_flush() and move
the lock/unlock calls to the beginning and ending of the loop body to
make the critical section obvious.
cgroup_rstat_flush_hold/release() do not make much sense with the lock
being dropped and reacquired internally. Since it has no external
callers, remove it and explicitly acquire the lock in
cgroup_base_stat_cputime_show() instead.
This leaves the code with a single flushing function,
cgroup_rstat_flush().
Signed-off-by: Yosry Ahmed <yosry.ahmed@linux.dev>
Signed-off-by: Tejun Heo <tj@kernel.org>
|
|
Commit 779dbc2e78d7 ("printk: Avoid non-panic CPUs writing to ringbuffer")
aimed to isolate panic-related messages. However, when panic() itself
malfunctions, messages from non-panic CPUs become crucial for debugging.
While commit bcc954c6caba ("printk/panic: Allow cpu backtraces to
be written into ringbuffer during panic") enables non-panic CPU
backtraces, it may not provide sufficient diagnostic information.
Introduce the "debug_non_panic_cpus" command-line option, enabling
non-panic CPU messages to be stored in the ring buffer during a panic.
This also prevents discarding non-finalized messages from non-panic CPUs
during console flushing, providing a more comprehensive view of system
state during critical failures.
Link: https://lore.kernel.org/all/Z8cLEkqLL2IOyNIj@pathway/
Signed-off-by: Donghyeok Choe <d7271.choe@samsung.com>
Reviewed-by: Petr Mladek <pmladek@suse.com>
Link: https://lore.kernel.org/r/20250318022320.2428155-1-d7271.choe@samsung.com
[pmladek@suse.com: Added documentation, added module_parameter, removed printk_ prefix.]
Tested-by: Petr Mladek <pmladek@suse.com>
Signed-off-by: Petr Mladek <pmladek@suse.com>
|
|
This is another attempt trying to make pidfd polling for multi-threaded
exec and premature thread-group leader exit consistent.
A quick recap of these two cases:
(1) During a multi-threaded exec by a subthread, i.e., non-thread-group
leader thread, all other threads in the thread-group including the
thread-group leader are killed and the struct pid of the
thread-group leader will be taken over by the subthread that called
exec. IOW, two tasks change their TIDs.
(2) A premature thread-group leader exit means that the thread-group
leader exited before all of the other subthreads in the thread-group
have exited.
Both cases lead to inconsistencies for pidfd polling with PIDFD_THREAD.
Any caller that holds a PIDFD_THREAD pidfd to the current thread-group
leader may or may not see an exit notification on the file descriptor
depending on when poll is performed. If the poll is performed before the
exec of the subthread has concluded an exit notification is generated
for the old thread-group leader. If the poll is performed after the exec
of the subthread has concluded no exit notification is generated for the
old thread-group leader.
The correct behavior would be to simply not generate an exit
notification on the struct pid of a subhthread exec because the struct
pid is taken over by the subthread and thus remains alive.
But this is difficult to handle because a thread-group may exit
prematurely as mentioned in (2). In that case an exit notification is
reliably generated but the subthreads may continue to run for an
indeterminate amount of time and thus also may exec at some point.
So far there was no way to distinguish between (1) and (2) internally.
This tiny series tries to address this problem by discarding
PIDFD_THREAD notification on premature thread-group leader exit.
If that works correctly then no exit notifications are generated for a
PIDFD_THREAD pidfd for a thread-group leader until all subthreads have
been reaped. If a subthread should exec aftewards no exit notification
will be generated until that task exits or it creates subthreads and
repeates the cycle.
Co-Developed-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Link: https://lore.kernel.org/r/20250320-work-pidfs-thread_group-v4-1-da678ce805bf@kernel.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
'event_trigger_ops mwifiex_if_ops' are not modified in these drivers.
Constifying these structures moves some data to a read-only section, so
increase overall security, especially when the structure holds some
function pointers.
On a x86_64, with allmodconfig, as an example:
Before:
======
text data bss dec hex filename
31368 9024 6200 46592 b600 kernel/trace/trace_events_trigger.o
After:
=====
text data bss dec hex filename
31752 8608 6200 46560 b5e0 kernel/trace/trace_events_trigger.o
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Link: https://lore.kernel.org/66e8f990e649678e4be37d4d1a19158ca0dea2f4.1741521295.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
|
|
All the big Linux distros enable CONFIG_SCHED_DEBUG, because
the various features it provides help not just with kernel
development, but with system administration and user-space
software development as well.
Reflect this reality and enable this functionality
unconditionally.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Shrikanth Hegde <sshegde@linux.ibm.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>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/r/20250317104257.3496611-4-mingo@kernel.org
|
|
With CONFIG_SCHED_DEBUG becoming unconditional, remove the
extra 'const_debug' indirection towards __read_mostly.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Shrikanth Hegde <sshegde@linux.ibm.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>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/r/20250317104257.3496611-3-mingo@kernel.org
|
|
The scheduler has this special SCHED_WARN() facility that
depends on CONFIG_SCHED_DEBUG.
Since CONFIG_SCHED_DEBUG is getting removed, convert
SCHED_WARN() to WARN_ON_ONCE().
Note that the warning output isn't 100% equivalent:
#define SCHED_WARN_ON(x) WARN_ONCE(x, #x)
Because SCHED_WARN_ON() would output the 'x' condition
as well, while WARN_ONCE() will only show a backtrace.
Hopefully these are rare enough to not really matter.
If it does, we should probably introduce a new WARN_ON()
variant that outputs the condition in stringified form,
or improve WARN_ON() itself.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Shrikanth Hegde <sshegde@linux.ibm.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>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/r/20250317104257.3496611-2-mingo@kernel.org
|
|
cgroup_rstat_flush_locked() grabs the irq safe cgroup_rstat_lock while
iterating all possible cpus. It only drops the lock if there is
scheduler or spin lock contention. If neither, then interrupts can be
disabled for a long time. On large machines this can disable interrupts
for a long enough time to drop network packets. On 400+ CPU machines
I've seen interrupt disabled for over 40 msec.
Prevent rstat from disabling interrupts while processing all possible
cpus. Instead drop and reacquire cgroup_rstat_lock for each cpu. This
approach was previously discussed in
https://lore.kernel.org/lkml/ZBz%2FV5a7%2F6PZeM7S@slm.duckdns.org/,
though this was in the context of an non-irq rstat spin lock.
Benchmark this change with:
1) a single stat_reader process with 400 threads, each reading a test
memcg's memory.stat repeatedly for 10 seconds.
2) 400 memory hog processes running in the test memcg and repeatedly
charging memory until oom killed. Then they repeat charging and oom
killing.
v6.14-rc6 with CONFIG_IRQSOFF_TRACER with stat_reader and hogs, finds
interrupts are disabled by rstat for 45341 usec:
# => started at: _raw_spin_lock_irq
# => ended at: cgroup_rstat_flush
#
#
# _------=> CPU#
# / _-----=> irqs-off/BH-disabled
# | / _----=> need-resched
# || / _---=> hardirq/softirq
# ||| / _--=> preempt-depth
# |||| / _-=> migrate-disable
# ||||| / delay
# cmd pid |||||| time | caller
# \ / |||||| \ | /
stat_rea-96532 52d.... 0us*: _raw_spin_lock_irq
stat_rea-96532 52d.... 45342us : cgroup_rstat_flush
stat_rea-96532 52d.... 45342us : tracer_hardirqs_on <-cgroup_rstat_flush
stat_rea-96532 52d.... 45343us : <stack trace>
=> memcg1_stat_format
=> memory_stat_format
=> memory_stat_show
=> seq_read_iter
=> vfs_read
=> ksys_read
=> do_syscall_64
=> entry_SYSCALL_64_after_hwframe
With this patch the CONFIG_IRQSOFF_TRACER doesn't find rstat to be the
longest holder. The longest irqs-off holder has irqs disabled for
4142 usec, a huge reduction from previous 45341 usec rstat finding.
Running stat_reader memory.stat reader for 10 seconds:
- without memory hogs: 9.84M accesses => 12.7M accesses
- with memory hogs: 9.46M accesses => 11.1M accesses
The throughput of memory.stat access improves.
The mode of memory.stat access latency after grouping by of 2 buckets:
- without memory hogs: 64 usec => 16 usec
- with memory hogs: 64 usec => 8 usec
The memory.stat latency improves.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Greg Thelen <gthelen@google.com>
Tested-by: Greg Thelen <gthelen@google.com>
Acked-by: Michal Koutný <mkoutny@suse.com>
Reviewed-by: Yosry Ahmed <yosry.ahmed@linux.dev>
Signed-off-by: Tejun Heo <tj@kernel.org>
|
|
Since out-of-order unlocks are unsupported for rqspinlock, and irqsave
variants enforce strict FIFO ordering anyway, make the same change for
normal non-irqsave variants, such that FIFO ordering is enforced.
Two new verifier state fields (active_lock_id, active_lock_ptr) are used
to denote the top of the stack, and prev_id and prev_ptr are ascertained
whenever popping the topmost entry through an unlock.
Take special care to make these fields part of the state comparison in
refsafe.
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20250316040541.108729-25-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Introduce verifier-side support for rqspinlock kfuncs. The first step is
allowing bpf_res_spin_lock type to be defined in map values and
allocated objects, so BTF-side is updated with a new BPF_RES_SPIN_LOCK
field to recognize and validate.
Any object cannot have both bpf_spin_lock and bpf_res_spin_lock, only
one of them (and at most one of them per-object, like before) must be
present. The bpf_res_spin_lock can also be used to protect objects that
require lock protection for their kfuncs, like BPF rbtree and linked
list.
The verifier plumbing to simulate success and failure cases when calling
the kfuncs is done by pushing a new verifier state to the verifier state
stack which will verify the failure case upon calling the kfunc. The
path where success is indicated creates all lock reference state and IRQ
state (if necessary for irqsave variants). In the case of failure, the
state clears the registers r0-r5, sets the return value, and skips kfunc
processing, proceeding to the next instruction.
When marking the return value for success case, the value is marked as
0, and for the failure case as [-MAX_ERRNO, -1]. Then, in the program,
whenever user checks the return value as 'if (ret)' or 'if (ret < 0)'
the verifier never traverses such branches for success cases, and would
be aware that the lock is not held in such cases.
We push the kfunc state in check_kfunc_call whenever rqspinlock kfuncs
are invoked. We introduce a kfunc_class state to avoid mixing lock
irqrestore kfuncs with IRQ state created by bpf_local_irq_save.
With all this infrastructure, these kfuncs become usable in programs
while satisfying all safety properties required by the kernel.
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20250316040541.108729-24-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Introduce four new kfuncs, bpf_res_spin_lock, and bpf_res_spin_unlock,
and their irqsave/irqrestore variants, which wrap the rqspinlock APIs.
bpf_res_spin_lock returns a conditional result, depending on whether the
lock was acquired (NULL is returned when lock acquisition succeeds,
non-NULL upon failure). The memory pointed to by the returned pointer
upon failure can be dereferenced after the NULL check to obtain the
error code.
Instead of using the old bpf_spin_lock type, introduce a new type with
the same layout, and the same alignment, but a different name to avoid
type confusion.
Preemption is disabled upon successful lock acquisition, however IRQs
are not. Special kfuncs can be introduced later to allow disabling IRQs
when taking a spin lock. Resilient locks are safe against AA deadlocks,
hence not disabling IRQs currently does not allow violation of kernel
safety.
__irq_flag annotation is used to accept IRQ flags for the IRQ-variants,
with the same semantics as existing bpf_local_irq_{save, restore}.
These kfuncs will require additional verifier-side support in subsequent
commits, to allow programs to hold multiple locks at the same time.
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20250316040541.108729-23-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Convert all LPM trie usage of raw_spinlock to rqspinlock.
Note that rcu_dereference_protected in trie_delete_elem is switched over
to plain rcu_dereference, the RCU read lock should be held from BPF
program side or eBPF syscall path, and the trie->lock is just acquired
before the dereference. It is not clear the reason the protected variant
was used from the commit history, but the above reasoning makes sense so
switch over.
Closes: https://lore.kernel.org/lkml/000000000000adb08b061413919e@google.com
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20250316040541.108729-22-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Convert the percpu_freelist.c code to use rqspinlock, and remove the
extralist fallback and trylock-based acquisitions to avoid deadlocks.
Key thing to note is the retained while (true) loop to search through
other CPUs when failing to push a node due to locking errors. This
retains the behavior of the old code, where it would keep trying until
it would be able to successfully push the node back into the freelist of
a CPU.
Technically, we should start iteration for this loop from
raw_smp_processor_id() + 1, but to avoid hitting the edge of nr_cpus,
we skip execution in the loop body instead.
Closes: https://lore.kernel.org/bpf/CAPPBnEa1_pZ6W24+WwtcNFvTUHTHO7KUmzEbOcMqxp+m2o15qQ@mail.gmail.com
Closes: https://lore.kernel.org/bpf/CAPPBnEYm+9zduStsZaDnq93q1jPLqO-PiKX9jy0MuL8LCXmCrQ@mail.gmail.com
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20250316040541.108729-21-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Convert hashtab.c from raw_spinlock to rqspinlock, and drop the hashed
per-cpu counter crud from the code base which is no longer necessary.
Closes: https://lore.kernel.org/bpf/675302fd.050a0220.2477f.0004.GAE@google.com
Closes: https://lore.kernel.org/bpf/000000000000b3e63e061eed3f6b@google.com
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20250316040541.108729-20-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Introduce locktorture support for rqspinlock using the newly added
macros as the first in-kernel user and consumer. Guard the code with
CONFIG_BPF_SYSCALL ifdef since rqspinlock is not available otherwise.
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20250316040541.108729-19-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Ensure that the rqspinlock code is only built when the BPF subsystem is
compiled in. Depending on queued spinlock support, we may or may not end
up building the queued spinlock slowpath, and instead fallback to the
test-and-set implementation. Also add entries to MAINTAINERS file.
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20250316040541.108729-18-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
We ripped out PV and virtualization related bits from rqspinlock in an
earlier commit, however, a fair lock performs poorly within a virtual
machine when the lock holder is preempted. As such, retain the
virt_spin_lock fallback to test and set lock, but with timeout and
deadlock detection. We can do this by simply depending on the
resilient_tas_spin_lock implementation from the previous patch.
We don't integrate support for CONFIG_PARAVIRT_SPINLOCKS yet, as that
requires more involved algorithmic changes and introduces more
complexity. It can be done when the need arises in the future.
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20250316040541.108729-15-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Include a test-and-set fallback when queued spinlock support is not
available. Introduce a rqspinlock type to act as a fallback when
qspinlock support is absent.
Include ifdef guards to ensure the slow path in this file is only
compiled when CONFIG_QUEUED_SPINLOCKS=y. Subsequent patches will add
further logic to ensure fallback to the test-and-set implementation
when queued spinlock support is unavailable on an architecture.
Unlike other waiting loops in rqspinlock code, the one for test-and-set
has no theoretical upper bound under contention, therefore we need a
longer timeout than usual. Bump it up to a second in this case.
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20250316040541.108729-14-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
While the timeout logic provides guarantees for the waiter's forward
progress, the time until a stalling waiter unblocks can still be long.
The default timeout of 1/4 sec can be excessively long for some use
cases. Additionally, custom timeouts may exacerbate recovery time.
Introduce logic to detect common cases of deadlocks and perform quicker
recovery. This is done by dividing the time from entry into the locking
slow path until the timeout into intervals of 1 ms. Then, after each
interval elapses, deadlock detection is performed, while also polling
the lock word to ensure we can quickly break out of the detection logic
and proceed with lock acquisition.
A 'held_locks' table is maintained per-CPU where the entry at the bottom
denotes a lock being waited for or already taken. Entries coming before
it denote locks that are already held. The current CPU's table can thus
be looked at to detect AA deadlocks. The tables from other CPUs can be
looked at to discover ABBA situations. Finally, when a matching entry
for the lock being taken on the current CPU is found on some other CPU,
a deadlock situation is detected. This function can take a long time,
therefore the lock word is constantly polled in each loop iteration to
ensure we can preempt detection and proceed with lock acquisition, using
the is_lock_released check.
We set 'spin' member of rqspinlock_timeout struct to 0 to trigger
deadlock checks immediately to perform faster recovery.
Note: Extending lock word size by 4 bytes to record owner CPU can allow
faster detection for ABBA. It is typically the owner which participates
in a ABBA situation. However, to keep compatibility with existing lock
words in the kernel (struct qspinlock), and given deadlocks are a rare
event triggered by bugs, we choose to favor compatibility over faster
detection.
The release_held_lock_entry function requires an smp_wmb, while the
release store on unlock will provide the necessary ordering for us. Add
comments to document the subtleties of why this is correct. It is
possible for stores to be reordered still, but in the context of the
deadlock detection algorithm, a release barrier is sufficient and
needn't be stronger for unlock's case.
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20250316040541.108729-13-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
When we run out of maximum rqnodes, the original queued spin lock slow
path falls back to a try lock. In such a case, we are again susceptible
to stalls in case the lock owner fails to make progress. We use the
timeout as a fallback to break out of this loop and return to the
caller. This is a fallback for an extreme edge case, when on the same
CPU we run out of all 4 qnodes. When could this happen? We are in slow
path in task context, we get interrupted by an IRQ, which while in the
slow path gets interrupted by an NMI, whcih in the slow path gets
another nested NMI, which enters the slow path. All of the interruptions
happen after node->count++.
We use RES_DEF_TIMEOUT as our spinning duration, but in the case of this
fallback, no fairness is guaranteed, so the duration may be too small
for contended cases, as the waiting time is not bounded. Since this is
an extreme corner case, let's just prefer timing out instead of
attempting to spin for longer.
Reviewed-by: Barret Rhoden <brho@google.com>
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20250316040541.108729-12-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Implement the wait queue cleanup algorithm for rqspinlock. There are
three forms of waiters in the original queued spin lock algorithm. The
first is the waiter which acquires the pending bit and spins on the lock
word without forming a wait queue. The second is the head waiter that is
the first waiter heading the wait queue. The third form is of all the
non-head waiters queued behind the head, waiting to be signalled through
their MCS node to overtake the responsibility of the head.
In this commit, we are concerned with the second and third kind. First,
we augment the waiting loop of the head of the wait queue with a
timeout. When this timeout happens, all waiters part of the wait queue
will abort their lock acquisition attempts. This happens in three steps.
First, the head breaks out of its loop waiting for pending and locked
bits to turn to 0, and non-head waiters break out of their MCS node spin
(more on that later). Next, every waiter (head or non-head) attempts to
check whether they are also the tail waiter, in such a case they attempt
to zero out the tail word and allow a new queue to be built up for this
lock. If they succeed, they have no one to signal next in the queue to
stop spinning. Otherwise, they signal the MCS node of the next waiter to
break out of its spin and try resetting the tail word back to 0. This
goes on until the tail waiter is found. In case of races, the new tail
will be responsible for performing the same task, as the old tail will
then fail to reset the tail word and wait for its next pointer to be
updated before it signals the new tail to do the same.
We terminate the whole wait queue because of two main reasons. Firstly,
we eschew per-waiter timeouts with one applied at the head of the wait
queue. This allows everyone to break out faster once we've seen the
owner / pending waiter not responding for the timeout duration from the
head. Secondly, it avoids complicated synchronization, because when not
leaving in FIFO order, prev's next pointer needs to be fixed up etc.
Lastly, all of these waiters release the rqnode and return to the
caller. This patch underscores the point that rqspinlock's timeout does
not apply to each waiter individually, and cannot be relied upon as an
upper bound. It is possible for the rqspinlock waiters to return early
from a failed lock acquisition attempt as soon as stalls are detected.
The head waiter cannot directly WRITE_ONCE the tail to zero, as it may
race with a concurrent xchg and a non-head waiter linking its MCS node
to the head's MCS node through 'prev->next' assignment.
One notable thing is that we must use RES_DEF_TIMEOUT * 2 as our maximum
duration for the waiting loop (for the wait queue head), since we may
have both the owner and pending bit waiter ahead of us, and in the worst
case, need to span their maximum permitted critical section lengths.
Reviewed-by: Barret Rhoden <brho@google.com>
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20250316040541.108729-11-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
The pending bit is used to avoid queueing in case the lock is
uncontended, and has demonstrated benefits for the 2 contender scenario,
esp. on x86. In case the pending bit is acquired and we wait for the
locked bit to disappear, we may get stuck due to the lock owner not
making progress. Hence, this waiting loop must be protected with a
timeout check.
To perform a graceful recovery once we decide to abort our lock
acquisition attempt in this case, we must unset the pending bit since we
own it. All waiters undoing their changes and exiting gracefully allows
the lock word to be restored to the unlocked state once all participants
(owner, waiters) have been recovered, and the lock remains usable.
Hence, set the pending bit back to zero before returning to the caller.
Introduce a lockevent (rqspinlock_lock_timeout) to capture timeout
event statistics.
Reviewed-by: Barret Rhoden <brho@google.com>
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20250316040541.108729-10-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Currently, for rqspinlock usage, the implementation of
smp_cond_load_acquire (and thus, atomic_cond_read_acquire) are
susceptible to stalls on arm64, because they do not guarantee that the
conditional expression will be repeatedly invoked if the address being
loaded from is not written to by other CPUs. When support for
event-streams is absent (which unblocks stuck WFE-based loops every
~100us), we may end up being stuck forever.
This causes a problem for us, as we need to repeatedly invoke the
RES_CHECK_TIMEOUT in the spin loop to break out when the timeout
expires.
Let us import the smp_cond_load_acquire_timewait implementation Ankur is
proposing in [0], and then fallback to it once it is merged.
While we rely on the implementation to amortize the cost of sampling
check_timeout for us, it will not happen when event stream support is
unavailable. This is not the common case, and it would be difficult to
fit our logic in the time_expr_ns >= time_limit_ns comparison, hence
just let it be.
[0]: https://lore.kernel.org/lkml/20250203214911.898276-1-ankur.a.arora@oracle.com
Cc: Ankur Arora <ankur.a.arora@oracle.com>
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20250316040541.108729-9-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Introduce policy macro RES_CHECK_TIMEOUT which can be used to detect
when the timeout has expired for the slow path to return an error. It
depends on being passed two variables initialized to 0: ts, ret. The
'ts' parameter is of type rqspinlock_timeout.
This macro resolves to the (ret) expression so that it can be used in
statements like smp_cond_load_acquire to break the waiting loop
condition.
The 'spin' member is used to amortize the cost of checking time by
dispatching to the implementation every 64k iterations. The
'timeout_end' member is used to keep track of the timestamp that denotes
the end of the waiting period. The 'ret' parameter denotes the status of
the timeout, and can be checked in the slow path to detect timeouts
after waiting loops.
The 'duration' member is used to store the timeout duration for each
waiting loop. The default timeout value defined in the header
(RES_DEF_TIMEOUT) is 0.25 seconds.
This macro will be used as a condition for waiting loops in the slow
path. Since each waiting loop applies a fresh timeout using the same
rqspinlock_timeout, we add a new RES_RESET_TIMEOUT as well to ensure the
values can be easily reinitialized to the default state.
Reviewed-by: Barret Rhoden <brho@google.com>
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20250316040541.108729-8-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
Linus Torvalds
Rafael J. Wysocki
Sasha Levin
Douglas Raillard
Gabriele Paoloni
Ran Xiaokai
Kohei Enju
Tengda Wu
Josh Poimboeuf
Andy Shevchenko
Masami Hiramatsu (Google)
Uladzislau Rezki (Sony)
Huang Shijie
John Stultz
Herbert Xu
Roger Pau Monne
Juntong Deng
Paolo Abeni
Yosry Ahmed
Donghyeok Choe
Christian Brauner
Christophe JAILLET
Joerg Roedel
Ingo Molnar
Eric Dumazet
Kumar Kartikeya Dwivedi