Re: Oops (NULL pointer) with 'perf record' of selftest 'null_syscall'
flat view
From: Christophe Leroy <hidden>
Date: 2021-06-16 06:28:44
Le 16/06/2021 à 05:40, Athira Rajeev a écrit :
quoted
On 16-Jun-2021, at 8:53 AM, Madhavan Srinivasan [off-list ref] wrote: On 6/15/21 8:35 PM, Christophe Leroy wrote:quoted
For your information, I'm getting the following Oops. Detected with 5.13-rc6, it also oopses on 5.12 and 5.11. Runs ok on 5.10. I'm starting bisecting now.Thanks for reporting, got the issue. What has happened in this case is that, pmu device is not registered and trying to access the instruction point which will land in perf_instruction_pointer(). And recently I have added a workaround patch for power10 DD1 which has caused this breakage. My bad. We are working on a fix patch for the same and will post it out. Sorry again.Hi Christophe, Can you please try with below patch in your environment and test if it works for you. From 55d3afc9369dfbe28a7152c8e9f856c11c7fe43d Mon Sep 17 00:00:00 2001 From: Athira Rajeev <redacted> Date: Tue, 15 Jun 2021 22:28:11 -0400 Subject: [PATCH] powerpc/perf: Fix crash with 'perf_instruction_pointer' when pmu is not set On systems without any specific PMU driver support registered, running perf record causes oops: [ 38.841073] NIP [c00000000013af54] perf_instruction_pointer+0x24/0x100 [ 38.841079] LR [c0000000003c7358] perf_prepare_sample+0x4e8/0x820 [ 38.841085] --- interrupt: 300 [ 38.841088] [c00000001cf03440] [c0000000003c6ef8] perf_prepare_sample+0x88/0x820 (unreliable) [ 38.841096] [c00000001cf034a0] [c0000000003c76d0] perf_event_output_forward+0x40/0xc0 [ 38.841104] [c00000001cf03520] [c0000000003b45e8] __perf_event_overflow+0x88/0x1b0 [ 38.841112] [c00000001cf03570] [c0000000003b480c] perf_swevent_hrtimer+0xfc/0x1a0 [ 38.841119] [c00000001cf03740] [c0000000002399cc] __hrtimer_run_queues+0x17c/0x380 [ 38.841127] [c00000001cf037c0] [c00000000023a5f8] hrtimer_interrupt+0x128/0x2f0 [ 38.841135] [c00000001cf03870] [c00000000002962c] timer_interrupt+0x13c/0x370 [ 38.841143i] [c00000001cf038d0] [c000000000009ba4] decrementer_common_virt+0x1a4/0x1b0 [ 38.841151] --- interrupt: 900 at copypage_power7+0xd4/0x1c0 During perf record session, perf_instruction_pointer() is called to capture the sample ip. This function in core-book3s accesses ppmu->flags. If a platform specific PMU driver is not registered, ppmu is set to NULL and accessing its members results in a crash. Fix this crash by checking if ppmu is set. Signed-off-by: Athira Rajeev <redacted> Reported-by: Christophe Leroy <redacted>
Fixes: 2ca13a4cc56c ("powerpc/perf: Use regs->nip when SIAR is zero")
Cc: stable@vger.kernel.org
Tested-by: Christophe Leroy <redacted>
quoted hunk ↗ jump to hunk
--- arch/powerpc/perf/core-book3s.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)diff --git a/arch/powerpc/perf/core-book3s.c b/arch/powerpc/perf/core-book3s.c index 16d4d1b6a1ff..816756588cb7 100644 --- a/arch/powerpc/perf/core-book3s.c +++ b/arch/powerpc/perf/core-book3s.c@@ -2254,7 +2254,7 @@ unsigned long perf_instruction_pointer(struct pt_regs *regs) bool use_siar = regs_use_siar(regs); unsigned long siar = mfspr(SPRN_SIAR); - if (ppmu->flags & PPMU_P10_DD1) { + if (ppmu && ppmu->flags & PPMU_P10_DD1) { if (siar) return siar; else