perf/x86: Fix undefined shift on 32-bit kernels
authorAndrey Ryabinin <aryabinin@virtuozzo.com>
Wed, 11 May 2016 13:51:51 +0000 (16:51 +0300)
committerIngo Molnar <mingo@kernel.org>
Thu, 12 May 2016 08:14:31 +0000 (10:14 +0200)
Jim reported:

UBSAN: Undefined behaviour in arch/x86/events/intel/core.c:3708:12
shift exponent 35 is too large for 32-bit type 'long unsigned int'

The use of 'unsigned long' type obviously is not correct here, make it
'unsigned long long' instead.

Reported-by: Jim Cromie <jim.cromie@gmail.com>
Signed-off-by: Andrey Ryabinin <aryabinin@virtuozzo.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: <stable@vger.kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Imre Palik <imrep@amazon.de>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Fixes: 2c33645d366d ("perf/x86: Honor the architectural performance monitoring version")
Link: http://lkml.kernel.org/r/1462974711-10037-1-git-send-email-aryabinin@virtuozzo.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
arch/x86/events/intel/core.c

index a6fd4db..5210eaa 100644 (file)
@@ -3708,7 +3708,7 @@ __init int intel_pmu_init(void)
                                c->idxmsk64 |= (1ULL << x86_pmu.num_counters) - 1;
                        }
                        c->idxmsk64 &=
-                               ~(~0UL << (INTEL_PMC_IDX_FIXED + x86_pmu.num_counters_fixed));
+                               ~(~0ULL << (INTEL_PMC_IDX_FIXED + x86_pmu.num_counters_fixed));
                        c->weight = hweight64(c->idxmsk64);
                }
        }