x86, mce: Ensure offline CPU's don't participate in mce rendezvous process.
PATCH 1/1 x86, mce: Ensure offline CPU's don't participate in mce rendezvous process.
Linux has logical cpu offline capability. That can be triggered by:
# echo 0 > /sys/devices/system/cpu/cpuX/online
In Intel Architecture, MCE's are broadcasted to all CPUs in the system.
This includes the CPUs marked offline by Linux. Unless the CPU's were removed via an ACPI notification, in which case the cpu's are removed from the cpu_present_map.
This patch ensures offline CPU's don't participate in MCE rendezvous, but simply perform clearing some status bits to ensure a second MCE wont cause automatic shutdown.
Without the patch, mce_start will increment mce_callin, but mce_start would only wait for all online_cpus. So offline cpu's should avoid participating in the rendezvous process.
--- arch/x86/kernel/cpu/mcheck/mce.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-)
arch/x86/kernel/cpu/mcheck/mce.c +14 -1
@@ -998,7 +998,20 @@ void do_machine_check(struct pt_regs *regs, long error_code)u64 recover_paddr = ~0ull ;int flags =MF_ACTION_REQUIRED ;int lmce =0 ; +unsigned int cpu =smp_processor_id (); +/* +* if this cpu is offline, just bail out. +*/ +if (cpu_is_offline (cpu)) { +u64 mcgstatus; + + mcgstatus =mce_rdmsrl (MSR_IA32_MCG_STATUS ); +if (mcgstatus &MCG_STATUS_RIPV ) { +mce_wrmsrl (MSR_IA32_MCG_STATUS ,0 ); +return ; + } + }ist_enter (regs);this_cpu_inc (mce_exception_count);
@@ -1142,8 +1155,8 @@ void do_machine_check(struct pt_regs *regs, long error_code)if (worst >0 )mce_report_event (regs); -mce_wrmsrl (MSR_IA32_MCG_STATUS ,0 ); out: +mce_wrmsrl (MSR_IA32_MCG_STATUS ,0 );sync_core ();if (recover_paddr == ~0ull )
Signature · 1 line
2.4.3
Quoted 52 lines · Ashok Raj
Linux has logical cpu offline capability. That can be triggered by:
# echo 0 > /sys/devices/system/cpu/cpuX/onlineIn Intel Architecture, MCE's are broadcasted to all CPUs in the system.
This includes the CPUs marked offline by Linux. Unless the CPU's were removed via an ACPI notification, in which case the cpu's are removed from the
cpu_present_map.This patch ensures offline CPU's don't participate in MCE rendezvous, but simply perform clearing some status bits to ensure a second MCE wont cause automatic shutdown.
Without the patch,
mce_startwill incrementmce_callin, butmce_startwould only wait for allonline_cpus. So offline cpu's should avoid participating in the rendezvous process.Reviewed-by Tony Luck <[email protected]>Signed-off-by Ashok Raj <[email protected]>--- arch/x86/kernel/cpu/mcheck/mce.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-)arch/x86/kernel/cpu/mcheck/mce.c +13
-0@@ -998,7 +998,20 @@ void do_machine_check(struct pt_regs *regs, long error_code)u64 recover_paddr = ~0ull ;int flags =MF_ACTION_REQUIRED ;int lmce =0 ; +unsigned int cpu =smp_processor_id (); +/* +* if this cpu is offline, just bail out. +*/ +if (cpu_is_offline (cpu)) { +u64 mcgstatus; + + mcgstatus =mce_rdmsrl (MSR_IA32_MCG_STATUS ); +if (mcgstatus &MCG_STATUS_RIPV ) { +mce_wrmsrl (MSR_IA32_MCG_STATUS ,0 ); +return ; + } + }ist_enter (regs);this_cpu_inc (mce_exception_count);
arch/x86/kernel/cpu/mcheck/mce.c void do_machine_check(struct pt_regs *regs, long error_code) · line 1162
Quoted 9 lines
arch/x86/kernel/cpu/mcheck/mce.c +1
-1@@ -1142,8 +1155,8 @@ void do_machine_check(struct pt_regs *regs, long error_code)if (worst >0 )mce_report_event (regs); -mce_wrmsrl (MSR_IA32_MCG_STATUS ,0 ); out: +mce_wrmsrl (MSR_IA32_MCG_STATUS ,0 );sync_core ();if (recover_paddr == ~0ull )
With that hunk here you want to clear MSR_IA32_MCG_STATUS in the !cfg->banks case, right?
If so, this should be a separate patch. Also, CC: stable with an explanation why.
Thanks.
Signature · 5 lines
Regards/Gruss,
Boris.
SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg)
-- Quoted 2 lines
With that hunk here you want to clear
MSR_IA32_MCG_STATUSin the !cfg->banks case, right?
I can't imagine how we'd get into do_machine_check without any banks.
Would indeed be a separate patch ... but value seems limited.
-Tony
���{.n�+�������+%�����ݶ��w��{.n�+����{��G�����{ay�ʇڙ�,j��f���h��������z_�(�階�ݢj"���m������G������?����&���~��iO��z��v�^�m����������?�I�
Quoted 6 lines · Tony
Quoted 2 lines
With that hunk here you want to clear
MSR_IA32_MCG_STATUSin the !cfg->banks case, right?I can't imagine how we'd get into
do_machine_checkwithout any banks.Would indeed be a separate patch ... but value seems limited.
Then we should kill that hunk. I'm all for not covering improbable cases.
Signature · 5 lines
Regards/Gruss,
Boris.
SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg)
--