x86/mce: Detect local MCEs properly

Check the MCG_STATUS_LMCES bit on Intel to verify that current MCE is
local. It is always local on AMD.

Signed-off-by: Yazen Ghannam <[email protected]>
[ Massaged it a bit. Reflowed comments. Shut up -Wmaybe-uninitialized. ]
Signed-off-by: Borislav Petkov <[email protected]>
Cc: Andy Lutomirski <[email protected]>
Cc: Borislav Petkov <[email protected]>
Cc: Brian Gerst <[email protected]>
Cc: Denys Vlasenko <[email protected]>
Cc: H. Peter Anvin <[email protected]>
Cc: Linus Torvalds <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Tony Luck <[email protected]>
Cc: linux-edac <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Ingo Molnar <[email protected]>
This commit is contained in:
Yazen Ghannam 2016-04-30 14:33:57 +02:00 • committed by Ingo Molnar
commit fead35c689

View file

@ -1038,11 +1038,12 @@ void do_machine_check(struct pt_regs *regs, long error_code)
int i;
int worst = 0;
int severity;
/*
* Establish sequential order between the CPUs entering the machine
* check handler.
*/
int order;
int order = -1;
/*
* If no_way_out gets set, there is no safe way to recover from this
* MCE. If mca_cfg.tolerant is cranked up, we'll try anyway.
@ -1056,7 +1057,12 @@ void do_machine_check(struct pt_regs *regs, long error_code)
DECLARE_BITMAP(toclear, MAX_NR_BANKS);
DECLARE_BITMAP(valid_banks, MAX_NR_BANKS);
char *msg = "Unknown";
int lmce = 0;
/*
* MCEs are always local on AMD. Same is determined by MCG_STATUS_LMCES
* on Intel.
*/
int lmce = 1;
/* If this CPU is offline, just bail out. */
if (cpu_is_offline(smp_processor_id())) {
@ -1095,19 +1101,20 @@ void do_machine_check(struct pt_regs *regs, long error_code)
kill_it = 1;
/*
* Check if this MCE is signaled to only this logical processor
* Check if this MCE is signaled to only this logical processor,
* on Intel only.
*/
if (m.mcgstatus & MCG_STATUS_LMCES)
lmce = 1;
else {
/*
* Go through all the banks in exclusion of the other CPUs.
* This way we don't report duplicated events on shared banks
* because the first one to see it will clear it.
* If this is a Local MCE, then no need to perform rendezvous.
*/
if (m.cpuvendor == X86_VENDOR_INTEL)
lmce = m.mcgstatus & MCG_STATUS_LMCES;
/*
* Go through all banks in exclusion of the other CPUs. This way we
* don't report duplicated events on shared banks because the first one
* to see it will clear it. If this is a Local MCE, then no need to
* perform rendezvous.
*/
if (!lmce)
order = mce_start(&no_way_out);
}
for (i = 0; i < cfg->banks; i++) {
__clear_bit(i, toclear);