Re: perf: fuzzer PEBS record without PEBS event! crash

2015-09-08 Thread Vince Weaver
On Tue, 8 Sep 2015, Stephane Eranian wrote: > Has to do with inlining. Make the function not inlined. OK I'll do that. This current kernel isn't crashing, but I'm getting this a lot [ 7877.413225] PEBS record without PEBS event! status=1 pebs_enabled=8 active_mask=0 vince@haswell:~$ dmesg

Re: perf: fuzzer PEBS record without PEBS event! crash

2015-09-08 Thread Stephane Eranian
On Tue, Sep 8, 2015 at 5:32 PM, Vince Weaver wrote: > On Mon, 7 Sep 2015, Stephane Eranian wrote: > >> On Mon, Sep 7, 2015 at 2:44 AM, Vince Weaver >> wrote: >> > On Sun, 6 Sep 2015, Stephane Eranian wrote: >> > >> >> > [ 1800.285152] [ cut here ] >> >> > [ 1800.290757]

Re: perf: fuzzer PEBS record without PEBS event! crash

2015-09-08 Thread Vince Weaver
On Mon, 7 Sep 2015, Stephane Eranian wrote: > On Mon, Sep 7, 2015 at 2:44 AM, Vince Weaver wrote: > > On Sun, 6 Sep 2015, Stephane Eranian wrote: > > > >> > [ 1800.285152] [ cut here ] > >> > [ 1800.290757] WARNING: CPU: 0 PID: 0 at > >> >

Re: perf: fuzzer PEBS record without PEBS event! crash

2015-09-08 Thread Stephane Eranian
On Tue, Sep 8, 2015 at 5:32 PM, Vince Weaver wrote: > On Mon, 7 Sep 2015, Stephane Eranian wrote: > >> On Mon, Sep 7, 2015 at 2:44 AM, Vince Weaver >> wrote: >> > On Sun, 6 Sep 2015, Stephane Eranian wrote: >> > >> >> > [ 1800.285152]

Re: perf: fuzzer PEBS record without PEBS event! crash

2015-09-08 Thread Vince Weaver
On Tue, 8 Sep 2015, Stephane Eranian wrote: > Has to do with inlining. Make the function not inlined. OK I'll do that. This current kernel isn't crashing, but I'm getting this a lot [ 7877.413225] PEBS record without PEBS event! status=1 pebs_enabled=8 active_mask=0 vince@haswell:~$ dmesg

Re: perf: fuzzer PEBS record without PEBS event! crash

2015-09-08 Thread Vince Weaver
On Mon, 7 Sep 2015, Stephane Eranian wrote: > On Mon, Sep 7, 2015 at 2:44 AM, Vince Weaver wrote: > > On Sun, 6 Sep 2015, Stephane Eranian wrote: > > > >> > [ 1800.285152] [ cut here ] > >> > [ 1800.290757] WARNING: CPU: 0 PID: 0 at > >> >

Re: perf: fuzzer PEBS record without PEBS event! crash

2015-09-07 Thread Stephane Eranian
On Mon, Sep 7, 2015 at 2:44 AM, Vince Weaver wrote: > On Sun, 6 Sep 2015, Stephane Eranian wrote: > >> > [ 1800.285152] [ cut here ] >> > [ 1800.290757] WARNING: CPU: 0 PID: 0 at >> > arch/x86/kernel/cpu/perf_event_intel_ds.c:1210 >> >

Re: perf: fuzzer PEBS record without PEBS event! crash

2015-09-07 Thread Stephane Eranian
On Mon, Sep 7, 2015 at 2:44 AM, Vince Weaver wrote: > On Sun, 6 Sep 2015, Stephane Eranian wrote: > >> > [ 1800.285152] [ cut here ] >> > [ 1800.290757] WARNING: CPU: 0 PID: 0 at >> > arch/x86/kernel/cpu/perf_event_intel_ds.c:1210 >> >

Re: perf: fuzzer PEBS record without PEBS event! crash

2015-09-06 Thread Vince Weaver
On Sun, 6 Sep 2015, Stephane Eranian wrote: > > [ 1800.285152] [ cut here ] > > [ 1800.290757] WARNING: CPU: 0 PID: 0 at > > arch/x86/kernel/cpu/perf_event_intel_ds.c:1210 > > intel_pmu_drain_pebs_nhm+0x305/0x320() > > [ 1800.303301] PEBS record without PEBS event!

Re: perf: fuzzer PEBS record without PEBS event! crash

2015-09-06 Thread Stephane Eranian
Hi Vince, On Thu, Sep 3, 2015 at 11:04 PM, Vince Weaver wrote: > > Hello > > running a fresh copy of current git and I can still crash my Haswell test > machine using the perf_fuzzer. > > It looks like the pebs issue still, it triggered the new warning, and > was stuck looping with these

Re: perf: fuzzer PEBS record without PEBS event! crash

2015-09-06 Thread Vince Weaver
On Sun, 6 Sep 2015, Stephane Eranian wrote: > > [ 1800.285152] [ cut here ] > > [ 1800.290757] WARNING: CPU: 0 PID: 0 at > > arch/x86/kernel/cpu/perf_event_intel_ds.c:1210 > > intel_pmu_drain_pebs_nhm+0x305/0x320() > > [ 1800.303301] PEBS record without PEBS event!

Re: perf: fuzzer PEBS record without PEBS event! crash

2015-09-06 Thread Stephane Eranian
Hi Vince, On Thu, Sep 3, 2015 at 11:04 PM, Vince Weaver wrote: > > Hello > > running a fresh copy of current git and I can still crash my Haswell test > machine using the perf_fuzzer. > > It looks like the pebs issue still, it triggered the new warning, and > was stuck

perf: fuzzer PEBS record without PEBS event! crash

2015-09-03 Thread Vince Weaver
Hello running a fresh copy of current git and I can still crash my Haswell test machine using the perf_fuzzer. It looks like the pebs issue still, it triggered the new warning, and was stuck looping with these messages until I rebooted the system. [ 1800.285152] [ cut here

perf: fuzzer PEBS record without PEBS event! crash

2015-09-03 Thread Vince Weaver
Hello running a fresh copy of current git and I can still crash my Haswell test machine using the perf_fuzzer. It looks like the pebs issue still, it triggered the new warning, and was stuck looping with these messages until I rebooted the system. [ 1800.285152] [ cut here