[PATCH v7 0/8] sched/fair: Get rid of sched_domains_curr_level hack for tl->cpumask()

[Shrikanth Hegde]

On 8/29/25 1:23 PM, Valentin Schneider wrote:
> On 26/08/25 12:13, Peter Zijlstra wrote:
>> Subject: sched/fair: Get rid of sched_domains_curr_level hack for tl->cpumask()
>> From: Peter Zijlstra <peterz at infradead.org>
>> Date: Mon, 25 Aug 2025 12:02:44 +0000
>>
>> Leon [1] and Vinicius [2] noted a topology_span_sane() warning during
>> their testing starting from v6.16-rc1. Debug that followed pointed to
>> the tl->mask() for the NODE domain being incorrectly resolved to that of
>> the highest NUMA domain.
>>
>> tl->mask() for NODE is set to the sd_numa_mask() which depends on the
>> global "sched_domains_curr_level" hack. "sched_domains_curr_level" is
>> set to the "tl->numa_level" during tl traversal in build_sched_domains()
>> calling sd_init() but was not reset before topology_span_sane().
>>
>> Since "tl->numa_level" still reflected the old value from
>> build_sched_domains(), topology_span_sane() for the NODE domain trips
>> when the span of the last NUMA domain overlaps.
>>
>> Instead of replicating the "sched_domains_curr_level" hack, get rid of
>> it entirely and instead, pass the entire "sched_domain_topology_level"
>> object to tl->cpumask() function to prevent such mishap in the future.
>>
>> sd_numa_mask() now directly references "tl->numa_level" instead of
>> relying on the global "sched_domains_curr_level" hack to index into
>> sched_domains_numa_masks[].
>>
> 
> Eh, of course I see this *after* looking at the v6 patch.
> 
> I tested this again for good measure, but given I only test this under
> x86 and the changes with v6 are in s390/ppc, I didn't expect to see much
> change :-)
> 
> Reviewed-by: Valentin Schneider <vschneid at redhat.com>
> Tested-by: Valentin Schneider <vschneid at redhat.com>
> 

I was looking at: https://git.kernel.org/pub/scm/linux/kernel/git/peterz/queue.git/log/?h=sched/core

Current code doesn't allow one to enable/disable SCHED_MC on ppc since it is set always in kconfig.
Used the below patch:

I think since the config is there, it would be good to provide a option to disable. no?

---

diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index fc0d1c19f5a1..da5b2f8d3686 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -170,9 +170,8 @@ config PPC
  	select ARCH_STACKWALK
  	select ARCH_SUPPORTS_ATOMIC_RMW
  	select ARCH_SUPPORTS_DEBUG_PAGEALLOC	if PPC_BOOK3S || PPC_8xx
-	select ARCH_SUPPORTS_SCHED_SMT		if PPC64 && SMP
  	select ARCH_SUPPORTS_SCHED_MC		if PPC64 && SMP
-	select SCHED_MC				if ARCH_SUPPORTS_SCHED_MC
+	select ARCH_SUPPORTS_SCHED_SMT		if PPC64 && SMP
  	select ARCH_USE_BUILTIN_BSWAP
  	select ARCH_USE_CMPXCHG_LOCKREF		if PPC64
  	select ARCH_USE_MEMTEST
diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c
index 68edb66c2964..458ec5bd859e 100644
--- a/arch/powerpc/kernel/smp.c
+++ b/arch/powerpc/kernel/smp.c
@@ -1706,10 +1706,12 @@ static void __init build_sched_topology(void)
  			SDTL_INIT(tl_cache_mask, powerpc_shared_cache_flags, CACHE);
  	}
  
+#ifdef CONFIG_SCHED_MC
  	if (has_coregroup_support()) {
  		powerpc_topology[i++] =
  			SDTL_INIT(tl_mc_mask, powerpc_shared_proc_flags, MC);
  	}
+#endif
  
  	powerpc_topology[i++] = SDTL_INIT(tl_pkg_mask, powerpc_shared_proc_flags, PKG);