From: Daniel Borkmann Date: Sat, 4 Jun 2016 18:50:59 +0000 (+0200) Subject: bpf, trace: use READ_ONCE for retrieving file ptr X-Git-Tag: v4.6.3~67 X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=a6878f9fa2040bd2c2407a678736512f4685292f;p=thirdparty%2Fkernel%2Fstable.git bpf, trace: use READ_ONCE for retrieving file ptr [ Upstream commit 5b6c1b4d46b0dae4edea636a776d09f2064f4cd7 ] In bpf_perf_event_read() and bpf_perf_event_output(), we must use READ_ONCE() for fetching the struct file pointer, which could get updated concurrently, so we must prevent the compiler from potential refetching. We already do this with tail calls for fetching the related bpf_prog, but not so on stored perf events. Semantics for both are the same with regards to updates. Fixes: a43eec304259 ("bpf: introduce bpf_perf_event_output() helper") Fixes: 35578d798400 ("bpf: Implement function bpf_perf_event_read() that get the selected hardware PMU conuter") Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 3e4ffb3ace5fa..d0289414cdeb1 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -194,7 +194,7 @@ static u64 bpf_perf_event_read(u64 r1, u64 index, u64 r3, u64 r4, u64 r5) if (unlikely(index >= array->map.max_entries)) return -E2BIG; - file = (struct file *)array->ptrs[index]; + file = READ_ONCE(array->ptrs[index]); if (unlikely(!file)) return -ENOENT; @@ -238,7 +238,7 @@ static u64 bpf_perf_event_output(u64 r1, u64 r2, u64 index, u64 r4, u64 size) if (unlikely(index >= array->map.max_entries)) return -E2BIG; - file = (struct file *)array->ptrs[index]; + file = READ_ONCE(array->ptrs[index]); if (unlikely(!file)) return -ENOENT;