This warning can be triggered if NUMA is disabled and the system
boots with fewer CPUs than the number of CPUs in die 0.
WARNING: CPU: 9 PID: 7257 at uncore.c:1146 uncore_pci_pmu_register+0x136/0x160 [intel_uncore]
Currently, the discovery table continues to be parsed even if all CPUs
in the associated die are offline. This can lead to an array overflow
at "pmu->boxes[die] = box" in uncore_pci_pmu_register(), which may
trigger the warning above or cause other issues.
However, the Intel uncore driver assumes all die IDs are consecutive
and does not support gaps in the die ID space.
Reported-by: Steve Wahl <steve.wahl@hpe.com>
Tested-by: Steve Wahl <steve.wahl@hpe.com>
Fixes: edae1f06c2cd ("perf/x86/intel/uncore: Parse uncore discovery tables")
Reviewed-by: Dapeng Mi <dapeng1.mi@linux.intel.com>
Signed-off-by: Zide Chen <zide.chen@intel.com>
---
V2:
- Add the Tested-by tag
- Rebase onto perf/core (base commit: a491c02c2770)
V3:
- Remove the overly cautious WARN_ON()
V4:
- Add Reviewed-by tag.
v6:
- Change WARNING line from 1157 to 1146 to reflect the new rebased code.
- Add changelog to clarify that it doesn't support die ID gaps.
---
arch/x86/events/intel/uncore_discovery.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/x86/events/intel/uncore_discovery.c b/arch/x86/events/intel/uncore_discovery.c
index 427939be2fb4..583cbd06b9b8 100644
--- a/arch/x86/events/intel/uncore_discovery.c
+++ b/arch/x86/events/intel/uncore_discovery.c
@@ -371,7 +371,7 @@ static bool uncore_discovery_pci(struct uncore_discovery_domain *domain)
(val & UNCORE_DISCOVERY_DVSEC2_BIR_MASK) * UNCORE_DISCOVERY_BIR_STEP;
die = get_device_die_id(dev);
- if (die < 0)
+ if ((die < 0) || (die >= uncore_max_dies()))
continue;
parse_discovery_table(domain, dev, die, bar_offset, &parsed);
--
2.53.0