diff options
| author | Tim Chen <tim.c.chen@linux.intel.com> | 2026-09-21 17:37:22 -0700 |
|---|---|---|
| committer | Ingo Molnar <mingo@kernel.org> | 2026-09-22 10:49:53 +0200 |
| commit | 0d6526f82c3cdefcca47f73f5fc08dc6f335eac6 (patch) | |
| tree | 176dc2d685debcfb082c151ee94f2425ac96b364 /Documentation/driver-api/basics.rst | |
| download | linux-stable-0d6526f82c3cdefcca47f73f5fc08dc6f335eac6.tar.gz linux-stable-0d6526f82c3cdefcca47f73f5fc08dc6f335eac6.zip | |
sched/cache: Keep nr_pref_llc_running in the runnable domain, to fix LLC mis-scheduling buggrafted
alb_break_llc() decides whether to break LLC preference during active
load balance. It does so by testing that every runnable fair task on the
source rq prefers its LLC:
env->src_rq->nr_pref_llc_running == env->src_rq->cfs.h_nr_runnable
But the two counters cover different sets. nr_pref_llc_running is updated
in account_llc_enqueue()/account_llc_dequeue(), next to cfs_rq->nr_queued,
so it follows queued tasks. h_nr_runnable is updated in set_delayed()/
clear_delayed() and drops delay-dequeued tasks.
So under DELAY_DEQUEUE, a preferring task that goes to sleep stays counted
in nr_pref_llc_running while h_nr_runnable falls. The equality then breaks,
alb_break_llc() returns false, and active balance is free to pull a task
off its preferred LLC. Active balance only moves runnable tasks, and this
is the only LLC check it consults: once the stopper runs, LBF_ACTIVE_LB
skips the per-task test in can_migrate_task(). The runnable set is the one
we want.
Fix it on the counter side. A task should be counted in
nr_pref_llc_running exactly while it is both queued on its preferred LLC
(pref_llc_queued) and runnable (!sched_delayed). Define that membership
once in task_pref_llc_runnable(), and adjust the counter only through
pref_llc_running_inc()/pref_llc_running_dec() from the four sites that
change either input: account_llc_enqueue(), account_llc_dequeue(),
set_delayed() and clear_delayed(). Gating every update on the same
predicate keeps the delay, wake and dequeue paths from double-counting
or underflowing; see the comments at those sites for the ordering.
nr_llc_running and sd->llc_counts are not touched and stay on queued
semantics.
Fixes: 714059f79ff0 ("sched/cache: Handle moving single tasks to/from their preferred LLC")
Closes: https://lore.kernel.org/lkml/20260827135000.735138-1-zhanxusheng@xiaomi.com/
Reported-by: Zhan Xusheng <zhanxusheng@xiaomi.com>
Suggested-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Kayra Cizmeci <kayracizmeci@gmail.com>
Cc: <stable@kernel.org> # v7.2.x
Link: https://patch.msgid.link/06af61afedac32e6477f57feb4d658f6c411c3af.1790035273.git.tim.c.chen@linux.intel.com
Diffstat (limited to 'Documentation/driver-api/basics.rst')
| -rw-r--r-- | Documentation/driver-api/basics.rst | 151 |
1 files changed, 151 insertions, 0 deletions
diff --git a/Documentation/driver-api/basics.rst b/Documentation/driver-api/basics.rst new file mode 100644 index 000000000..3b182cfdf --- /dev/null +++ b/Documentation/driver-api/basics.rst @@ -0,0 +1,151 @@ +Driver Basics +============= + +Driver Entry and Exit points +---------------------------- + +.. kernel-doc:: include/linux/module.h + :internal: + +Driver device table +------------------- + +.. kernel-doc:: include/linux/mod_devicetable.h + :internal: + :no-identifiers: pci_device_id + + +Delaying and scheduling routines +-------------------------------- + +.. kernel-doc:: include/linux/sched.h + :internal: + +.. kernel-doc:: kernel/sched/core.c + :export: + +.. kernel-doc:: kernel/sched/cpupri.c + :internal: + +.. kernel-doc:: kernel/sched/fair.c + :internal: + +.. kernel-doc:: include/linux/completion.h + :internal: + +Time and timer routines +----------------------- + +.. kernel-doc:: include/linux/jiffies.h + :internal: + +.. kernel-doc:: kernel/time/time.c + :export: + +.. kernel-doc:: kernel/time/timer.c + :export: + +High-resolution timers +---------------------- + +.. kernel-doc:: include/linux/ktime.h + :internal: + +.. kernel-doc:: include/linux/hrtimer.h + :internal: + +.. kernel-doc:: kernel/time/hrtimer.c + :export: + +Wait queues and Wake events +--------------------------- + +.. kernel-doc:: include/linux/wait.h + :internal: + +.. kernel-doc:: kernel/sched/wait.c + :export: + +Internal Functions +------------------ + +.. kernel-doc:: kernel/exit.c + :internal: + +.. kernel-doc:: kernel/signal.c + :internal: + +.. kernel-doc:: include/linux/kthread.h + :internal: + +.. kernel-doc:: kernel/kthread.c + :export: + +Reference counting +------------------ + +.. kernel-doc:: include/linux/refcount.h + :internal: + +.. kernel-doc:: lib/refcount.c + :export: + +.. kernel-doc:: include/linux/percpu-refcount.h +.. kernel-doc:: lib/percpu-refcount.c + +Atomics +------- + +.. kernel-doc:: include/linux/atomic/atomic-instrumented.h + :internal: + +.. kernel-doc:: include/linux/atomic/atomic-arch-fallback.h + :internal: + +.. kernel-doc:: include/linux/atomic/atomic-long.h + :internal: + +Kernel objects manipulation +--------------------------- + +.. kernel-doc:: lib/kobject.c + :export: + +.. kernel-doc:: lib/kobject_uevent.c + :export: + +Kernel utility functions +------------------------ + +.. kernel-doc:: include/linux/array_size.h + :internal: + +.. kernel-doc:: include/linux/container_of.h + :internal: + +.. kernel-doc:: include/linux/kstrtox.h + :internal: + :no-identifiers: kstrtol kstrtoul + +.. kernel-doc:: include/linux/stddef.h + :internal: + +.. kernel-doc:: include/linux/util_macros.h + :internal: + +.. kernel-doc:: include/linux/wordpart.h + :internal: + +.. kernel-doc:: kernel/printk/printk.c + :export: + :no-identifiers: printk + +.. kernel-doc:: kernel/panic.c + :export: + +Device Resource Management +-------------------------- + +.. kernel-doc:: drivers/base/devres.c + :export: + |
