On Mon, Jun 06, 2016 at 09:37:41AM +0200, Christian Borntraeger wrote:
> Peter, Hendrik,
>
> commit 26657848502b ("perf/core: Verify we have a single perf_hw_context
> PMU") seems to
> trigger the newly created warning on a z196.
>
>
> [ 2.202363] ------------[ cut here ]------------
> [ 2.202372] WARNING: CPU: 5 PID: 1 at kernel/events/core.c:8485
> perf_pmu_register+0x420/0x428
> [ 2.202373] Modules linked in:
> [ 2.202377] CPU: 5 PID: 1 Comm: swapper/0 Not tainted 4.7.0-rc1+ #2
> [ 2.202379] task: 00000009c5240000 ti: 00000009c5234000 task.ti:
> 00000009c5234000
> [ 2.202381] Krnl PSW : 0704c00180000000 0000000000220c50
> (perf_pmu_register+0x420/0x428)
> [ 2.202385] R:0 T:1 IO:1 EX:1 Key:0 M:1 W:0 P:0 AS:3 CC:0 PM:0
> RI:0 EA:3
> Krnl GPRS: ffffffffffffffff 0000000000b15ac6 0000000000000000 00000009cb440000
> [ 2.202388] 000000000022087a 0000000000000000 0000000000b78fa0
> 0000000000000000
> [ 2.202390] 0000000000a9aa90 0000000000000084 0000000000000005
> 000000000088a97a
> [ 2.202405] 0000000000000004 0000000000749dd0 000000000022087a
> 00000009c5237cc0
> [ 2.202415] Krnl Code: 0000000000220c44: a7f4ff54 brc
> 15,220aec
> 0000000000220c48: 92011000 mvi 0(%r1),1
> #0000000000220c4c: a7f40001 brc 15,220c4e
> >0000000000220c50: a7f4ff12 brc 15,220a74
> 0000000000220c54: 0707 bcr 0,%r7
> 0000000000220c56: 0707 bcr 0,%r7
> 0000000000220c58: ebdff0800024 stmg %r13,%r15,128(%r15)
> 0000000000220c5e: a7f13fe0 tmll %r15,16352
> [ 2.202431] Call Trace:
> [ 2.202433] ([<000000000022087a>] perf_pmu_register+0x4a/0x428)
> [ 2.202438] ([<0000000000b2c25c>] init_cpum_sampling_pmu+0x14c/0x1f8)
> [ 2.202441] ([<0000000000100248>] do_one_initcall+0x48/0x140)
> [ 2.202444] ([<0000000000b25d26>] kernel_init_freeable+0x1e6/0x2a0)
> [ 2.202449] ([<000000000072bda4>] kernel_init+0x24/0x138)
> [ 2.202453] ([<000000000073495e>] kernel_thread_starter+0x6/0xc)
> [ 2.202455] ([<0000000000734958>] kernel_thread_starter+0x0/0xc)
> [ 2.202456] Last Breaking-Event-Address:
> [ 2.202458] [<0000000000220c4c>] perf_pmu_register+0x41c/0x428
> [ 2.202460] ---[ end trace 0c6ef9f5b771ad97 ]---
>
> Looks like perf_pmu_register does not like to be called twice (once for the
> counter
> and once for the sampling facility).
Twice isn't the problem per se, its trying to register two PMUs for
perf_hw_context that is the problem.
The perf core does not expect or deal well with that.
The perf core expects a single HW PMU in that when it schedules
hw_context events, and encounters an failure to pmu::add() (because the
hw pmu is 'full') it stops trying to add more events.
So if you mix two PMUs in, you get horrible PMU utilization issues,
because as soon as one is full, it will not try and add more events and
can leave the other empty.
