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 /sound/Kconfig | |
| 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 'sound/Kconfig')
| -rw-r--r-- | sound/Kconfig | 112 |
1 files changed, 112 insertions, 0 deletions
diff --git a/sound/Kconfig b/sound/Kconfig new file mode 100644 index 000000000..8b4020539 --- /dev/null +++ b/sound/Kconfig @@ -0,0 +1,112 @@ +# SPDX-License-Identifier: GPL-2.0-only +menuconfig SOUND + tristate "Sound card support" + depends on HAS_IOMEM || INDIRECT_IOMEM + help + If you have a sound card in your computer, i.e. if it can say more + than an occasional beep, say Y. + +if SOUND + +config SOUND_OSS_CORE + bool + default n + +config SOUND_OSS_CORE_PRECLAIM + bool "Preclaim OSS device numbers" + depends on SOUND_OSS_CORE + default y + help + With this option enabled, the kernel will claim all OSS device + numbers if any OSS support (native or emulation) is enabled + whether the respective module is loaded or not and try to load the + appropriate module using sound-slot/service-* and char-major-* + module aliases when one of the device numbers is opened. With + this option disabled, kernel will only claim actually in-use + device numbers and opening a missing device will generate only the + standard char-major-* aliases. + + The only visible difference is use of additional module aliases + and whether OSS sound devices appear multiple times in + /proc/devices. sound-slot/service-* module aliases are scheduled + to be removed (ie. PRECLAIM won't be available) and this option is + to make the transition easier. This option can be overridden + during boot using the kernel parameter soundcore.preclaim_oss. + + Disabling this allows alternative OSS implementations. + + If unsure, say Y. + +source "sound/oss/dmasound/Kconfig" + +menuconfig SND + tristate "Advanced Linux Sound Architecture" + help + Say 'Y' or 'M' to enable ALSA (Advanced Linux Sound Architecture), + the new base sound system. + + For more information, see <http://www.alsa-project.org/> + +if SND + +source "sound/core/Kconfig" + +source "sound/drivers/Kconfig" + +source "sound/isa/Kconfig" + +source "sound/pci/Kconfig" + +source "sound/hda/Kconfig" + +source "sound/ppc/Kconfig" + +source "sound/ac97/Kconfig" + +source "sound/aoa/Kconfig" + +source "sound/arm/Kconfig" + +source "sound/atmel/Kconfig" + +source "sound/spi/Kconfig" + +source "sound/mips/Kconfig" + +source "sound/sh/Kconfig" + +# the following will depend on the order of config. +# here assuming USB is defined before ALSA +source "sound/usb/Kconfig" + +source "sound/firewire/Kconfig" + +# the following will depend on the order of config. +# here assuming PCMCIA is defined before ALSA +source "sound/pcmcia/Kconfig" + +source "sound/sparc/Kconfig" + +source "sound/parisc/Kconfig" + +source "sound/soc/Kconfig" + +source "sound/x86/Kconfig" + +source "sound/synth/Kconfig" + +source "sound/xen/Kconfig" + +source "sound/virtio/Kconfig" + +endif # SND + +endif # SOUND + +config AC97_BUS + tristate + help + This is used to avoid config and link hard dependencies between the + sound subsystem and other function drivers completely unrelated to + sound although they're sharing the AC97 bus. Concerned drivers + should "select" this. |
