From 0d6526f82c3cdefcca47f73f5fc08dc6f335eac6 Mon Sep 17 00:00:00 2001 From: Tim Chen Date: Mon, 21 Sep 2026 17:37:22 -0700 Subject: sched/cache: Keep nr_pref_llc_running in the runnable domain, to fix LLC mis-scheduling bug 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 Suggested-by: Chen Yu Signed-off-by: Tim Chen Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Reviewed-by: Kayra Cizmeci Cc: # v7.2.x Link: https://patch.msgid.link/06af61afedac32e6477f57feb4d658f6c411c3af.1790035273.git.tim.c.chen@linux.intel.com --- arch/alpha/lib/Makefile | 46 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 46 insertions(+) create mode 100644 arch/alpha/lib/Makefile (limited to 'arch/alpha/lib/Makefile') diff --git a/arch/alpha/lib/Makefile b/arch/alpha/lib/Makefile new file mode 100644 index 000000000..84046e730 --- /dev/null +++ b/arch/alpha/lib/Makefile @@ -0,0 +1,46 @@ +# SPDX-License-Identifier: GPL-2.0 +# +# Makefile for alpha-specific library files.. +# + +asflags-y := $(KBUILD_CFLAGS) + +# Many of these routines have implementations tuned for ev6. +# Choose them iff we're targeting ev6 specifically. +ev6-$(CONFIG_ALPHA_EV6) := ev6- + +# Several make use of the cttz instruction introduced in ev67. +ev67-$(CONFIG_ALPHA_EV67) := ev67- + +lib-y = __divqu.o __remqu.o __divlu.o __remlu.o \ + udiv-qrnnd.o \ + udelay.o \ + $(ev6-y)memset.o \ + $(ev6-y)memcpy.o \ + memmove.o \ + checksum.o \ + csum_partial_copy.o \ + $(ev67-y)strlen.o \ + stycpy.o \ + styncpy.o \ + $(ev67-y)strchr.o \ + $(ev67-y)strrchr.o \ + $(ev6-y)memchr.o \ + $(ev6-y)copy_user.o \ + $(ev6-y)clear_user.o \ + $(ev6-y)csum_ipv6_magic.o \ + $(ev6-y)clear_page.o \ + $(ev6-y)copy_page.o \ + fpreg.o \ + callback_srm.o srm_puts.o srm_printk.o \ + fls.o + +# The division routines are built from single source, with different defines. +AFLAGS___divqu.o = -DDIV +AFLAGS___remqu.o = -DREM +AFLAGS___divlu.o = -DDIV -DINTSIZE +AFLAGS___remlu.o = -DREM -DINTSIZE + +$(addprefix $(obj)/,__divqu.o __remqu.o __divlu.o __remlu.o): \ + $(src)/$(ev6-y)divide.S FORCE + $(call if_changed_rule,as_o_S) -- cgit v1.3.1