summaryrefslogtreecommitdiff
path: root/arch/x86/events/intel
diff options
context:
space:
mode:
authorPeter Zijlstra <peterz@infradead.org>2020-10-30 15:50:32 +0100
committerPeter Zijlstra <peterz@infradead.org>2020-11-09 18:12:33 +0100
commit267fb27352b6fc9fdbad753127a239f75618ecbc (patch)
tree2f99dd2dedf21e53d329e16cdb615e2c234202bb /arch/x86/events/intel
parent7bdb157cdebbf95a1cd94ed2e01b338714075d00 (diff)
perf: Reduce stack usage of perf_output_begin()
__perf_output_begin() has an on-stack struct perf_sample_data in the unlikely case it needs to generate a LOST record. However, every call to perf_output_begin() must already have a perf_sample_data on-stack. Reported-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lkml.kernel.org/r/20201030151954.985416146@infradead.org
Diffstat (limited to 'arch/x86/events/intel')
-rw-r--r--arch/x86/events/intel/ds.c4
1 files changed, 2 insertions, 2 deletions
diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c
index 404315df1e16..cd2ae14a0a98 100644
--- a/arch/x86/events/intel/ds.c
+++ b/arch/x86/events/intel/ds.c
@@ -642,8 +642,8 @@ int intel_pmu_drain_bts_buffer(void)
rcu_read_lock();
perf_prepare_sample(&header, &data, event, &regs);
- if (perf_output_begin(&handle, event, header.size *
- (top - base - skip)))
+ if (perf_output_begin(&handle, &data, event,
+ header.size * (top - base - skip)))
goto unlock;
for (at = base; at < top; at++) {