[PATCH v3 1/3] powernv/mce: reduce mce console logs to lesser lines.
Mahesh Salgaonkar
mahesh at linux.vnet.ibm.com
Tue Apr 30 04:15:48 AEST 2019
Also add cpu number while displaying mce log. This will help cleaner logs
when mce hits on multiple cpus simultaneously.
before the changes the mce o/p was:
[ 127.223515] Severe Machine check interrupt [Recovered]
[ 127.223530] NIP [d00000000ba80280]: insert_slb_entry.constprop.0+0x278/0x2c0 [mcetest_slb]
[ 127.223539] Initiator: CPU
[ 127.223544] Error type: SLB [Multihit]
[ 127.223550] Effective address: d00000000ba80280
After this patch series changes the mce o/p will be:
[ 471.959843] MCE: CPU80: machine check (Warning) Host SLB Multihit [Recovered]
[ 471.959870] MCE: CPU80: NIP: [d00000000b550280] insert_slb_entry.constprop.0+0x278/0x2c0 [mcetest_slb]
[ 471.959892] MCE: CPU80: Probable software error (some chance of hardware cause)
UE in host application:
[ 1001.831517] MCE: CPU48: machine check (Severe) Host UE Load/Store DAR: 00007fffc6079a80 paddr: 0000000f8e260000 [Not recovered]
[ 1001.831518] MCE: CPU48: PID: 4584 Comm: find NIP: [0000000010023368]
[ 1001.831519] MCE: CPU48: Hardware error
and for MCE in Guest:
[ 1289.447571] MCE: CPU80: machine check (Warning) Guest SLB Multihit DAR: 000001001b6e0320 [Recovered]
[ 1289.447615] MCE: CPU80: PID: 24765 Comm: qemu-system-ppc Guest NIP: [00007fffa309dc60]
[ 1289.447634] MCE: CPU80: Probable software error (some chance of hardware cause)
Signed-off-by: Mahesh Salgaonkar <mahesh at linux.vnet.ibm.com>
---
CHange in v3:
- Print physical address if available.
Change in v2:
- Address comments from Michael.
---
arch/powerpc/include/asm/mce.h | 2 -
arch/powerpc/kernel/mce.c | 89 ++++++++++++++++++++++------------------
2 files changed, 49 insertions(+), 42 deletions(-)
diff --git a/arch/powerpc/include/asm/mce.h b/arch/powerpc/include/asm/mce.h
index 17996bc9382b..8d0b1c24c636 100644
--- a/arch/powerpc/include/asm/mce.h
+++ b/arch/powerpc/include/asm/mce.h
@@ -116,7 +116,7 @@ struct machine_check_event {
enum MCE_Initiator initiator:8; /* 0x03 */
enum MCE_ErrorType error_type:8; /* 0x04 */
enum MCE_Disposition disposition:8; /* 0x05 */
- uint8_t reserved_1[2]; /* 0x06 */
+ uint16_t cpu; /* 0x06 */
uint64_t gpr3; /* 0x08 */
uint64_t srr0; /* 0x10 */
uint64_t srr1; /* 0x18 */
diff --git a/arch/powerpc/kernel/mce.c b/arch/powerpc/kernel/mce.c
index b5fec1f9751a..25a8b20cbbdc 100644
--- a/arch/powerpc/kernel/mce.c
+++ b/arch/powerpc/kernel/mce.c
@@ -112,6 +112,7 @@ void save_mce_event(struct pt_regs *regs, long handled,
mce->srr1 = regs->msr;
mce->gpr3 = regs->gpr[3];
mce->in_use = 1;
+ mce->cpu = get_paca()->paca_index;
/* Mark it recovered if we have handled it and MSR(RI=1). */
if (handled && (regs->msr & MSR_RI))
@@ -310,7 +311,11 @@ static void machine_check_process_queued_event(struct irq_work *work)
void machine_check_print_event_info(struct machine_check_event *evt,
bool user_mode, bool in_guest)
{
- const char *level, *sevstr, *subtype;
+ const char *level, *sevstr, *subtype, *err_type;
+ uint64_t ea = 0, pa = 0;
+ int n = 0;
+ char dar_str[50];
+ char pa_str[50];
static const char *mc_ue_types[] = {
"Indeterminate",
"Instruction fetch",
@@ -384,101 +389,103 @@ void machine_check_print_event_info(struct machine_check_event *evt,
break;
}
- printk("%s%s Machine check interrupt [%s]\n", level, sevstr,
- evt->disposition == MCE_DISPOSITION_RECOVERED ?
- "Recovered" : "Not recovered");
-
- if (in_guest) {
- printk("%s Guest NIP: %016llx\n", level, evt->srr0);
- } else if (user_mode) {
- printk("%s NIP: [%016llx] PID: %d Comm: %s\n", level,
- evt->srr0, current->pid, current->comm);
- } else {
- printk("%s NIP [%016llx]: %pS\n", level, evt->srr0,
- (void *)evt->srr0);
- }
-
- printk("%s Initiator: %s\n", level,
- evt->initiator == MCE_INITIATOR_CPU ? "CPU" : "Unknown");
switch (evt->error_type) {
case MCE_ERROR_TYPE_UE:
+ err_type = "UE";
subtype = evt->u.ue_error.ue_error_type <
ARRAY_SIZE(mc_ue_types) ?
mc_ue_types[evt->u.ue_error.ue_error_type]
: "Unknown";
- printk("%s Error type: UE [%s]\n", level, subtype);
if (evt->u.ue_error.effective_address_provided)
- printk("%s Effective address: %016llx\n",
- level, evt->u.ue_error.effective_address);
+ ea = evt->u.ue_error.effective_address;
if (evt->u.ue_error.physical_address_provided)
- printk("%s Physical address: %016llx\n",
- level, evt->u.ue_error.physical_address);
+ pa = evt->u.ue_error.physical_address;
break;
case MCE_ERROR_TYPE_SLB:
+ err_type = "SLB";
subtype = evt->u.slb_error.slb_error_type <
ARRAY_SIZE(mc_slb_types) ?
mc_slb_types[evt->u.slb_error.slb_error_type]
: "Unknown";
- printk("%s Error type: SLB [%s]\n", level, subtype);
if (evt->u.slb_error.effective_address_provided)
- printk("%s Effective address: %016llx\n",
- level, evt->u.slb_error.effective_address);
+ ea = evt->u.slb_error.effective_address;
break;
case MCE_ERROR_TYPE_ERAT:
+ err_type = "ERAT";
subtype = evt->u.erat_error.erat_error_type <
ARRAY_SIZE(mc_erat_types) ?
mc_erat_types[evt->u.erat_error.erat_error_type]
: "Unknown";
- printk("%s Error type: ERAT [%s]\n", level, subtype);
if (evt->u.erat_error.effective_address_provided)
- printk("%s Effective address: %016llx\n",
- level, evt->u.erat_error.effective_address);
+ ea = evt->u.erat_error.effective_address;
break;
case MCE_ERROR_TYPE_TLB:
+ err_type = "TLB";
subtype = evt->u.tlb_error.tlb_error_type <
ARRAY_SIZE(mc_tlb_types) ?
mc_tlb_types[evt->u.tlb_error.tlb_error_type]
: "Unknown";
- printk("%s Error type: TLB [%s]\n", level, subtype);
if (evt->u.tlb_error.effective_address_provided)
- printk("%s Effective address: %016llx\n",
- level, evt->u.tlb_error.effective_address);
+ ea = evt->u.tlb_error.effective_address;
break;
case MCE_ERROR_TYPE_USER:
+ err_type = "User";
subtype = evt->u.user_error.user_error_type <
ARRAY_SIZE(mc_user_types) ?
mc_user_types[evt->u.user_error.user_error_type]
: "Unknown";
- printk("%s Error type: User [%s]\n", level, subtype);
if (evt->u.user_error.effective_address_provided)
- printk("%s Effective address: %016llx\n",
- level, evt->u.user_error.effective_address);
+ ea = evt->u.user_error.effective_address;
break;
case MCE_ERROR_TYPE_RA:
+ err_type = "Real address";
subtype = evt->u.ra_error.ra_error_type <
ARRAY_SIZE(mc_ra_types) ?
mc_ra_types[evt->u.ra_error.ra_error_type]
: "Unknown";
- printk("%s Error type: Real address [%s]\n", level, subtype);
if (evt->u.ra_error.effective_address_provided)
- printk("%s Effective address: %016llx\n",
- level, evt->u.ra_error.effective_address);
+ ea = evt->u.ra_error.effective_address;
break;
case MCE_ERROR_TYPE_LINK:
+ err_type = "Link";
subtype = evt->u.link_error.link_error_type <
ARRAY_SIZE(mc_link_types) ?
mc_link_types[evt->u.link_error.link_error_type]
: "Unknown";
- printk("%s Error type: Link [%s]\n", level, subtype);
if (evt->u.link_error.effective_address_provided)
- printk("%s Effective address: %016llx\n",
- level, evt->u.link_error.effective_address);
+ ea = evt->u.link_error.effective_address;
break;
default:
case MCE_ERROR_TYPE_UNKNOWN:
- printk("%s Error type: Unknown\n", level);
+ err_type = "Unknown";
+ subtype = "";
break;
}
+
+ dar_str[0] = pa_str[0] = '\0';
+ if (ea && evt->srr0 != ea) {
+ /* Load/Store address */
+ n = sprintf(dar_str, "DAR: %016llx ", ea);
+ if (pa)
+ sprintf(dar_str + n, "paddr: %016llx ", pa);
+ } else if (pa) {
+ sprintf(pa_str, " paddr: %016llx", pa);
+ }
+
+ printk("%sMCE: CPU%d: machine check (%s) %s %s %s %s[%s]\n",
+ level, evt->cpu, sevstr, in_guest ? "Guest" : "Host",
+ err_type, subtype, dar_str,
+ evt->disposition == MCE_DISPOSITION_RECOVERED ?
+ "Recovered" : "Not recovered");
+
+ if (in_guest || user_mode) {
+ printk("%sMCE: CPU%d: PID: %d Comm: %s %sNIP: [%016llx]%s\n",
+ level, evt->cpu, current->pid, current->comm,
+ in_guest ? "Guest " : "", evt->srr0, pa_str);
+ } else {
+ printk("%sMCE: CPU%d: NIP: [%016llx] %pS%s\n",
+ level, evt->cpu, evt->srr0, (void *)evt->srr0, pa_str);
+ }
}
EXPORT_SYMBOL_GPL(machine_check_print_event_info);
More information about the Linuxppc-dev
mailing list