Re: [RFC PATCH 2/3 v3] perf: Implement Nehalem uncore pmu

[Stephane Eranian]

On Mon, Jan 17, 2011 at 2:29 AM, Lin Ming <ming.m.lin@xxxxxxxxx> wrote:
> On Fri, 2011-01-14 at 01:14 +0800, Stephane Eranian wrote:
>> Lin,
>
> Hi, Stephane,
>
> Sorry for late response, I'm just back from vacation.
>
>>
>> On Thu, Dec 2, 2010 at 6:20 AM, Lin Ming <ming.m.lin@xxxxxxxxx> wrote:
>> > +static void uncore_pmu_enable_all(int nmi_core)
>> > +{
>> > + Â Â Â u64 ctrl;
>> > +
>> > + Â Â Â ctrl = ((1 << UNCORE_NUM_GENERAL_COUNTERS) - 1) | MSR_UNCORE_PERF_GLOBAL_CTRL_EN_FC0;
>> > +
>> > + Â Â Â /* Route all interrupts to the first core that accesses uncore */
>> > + Â Â Â ctrl |= 1ULL << (48 + nmi_core);
>> > +
>> > + Â Â Â wrmsrl(MSR_UNCORE_PERF_GLOBAL_CTRL, ctrl);
>> > +}
>>
>> Are you sure nmi_core is always between 0-3 on a 4-core system and 0-5
>> on a 6-core system?
>> In other words, is that what topology_core_id(raw_smp_processor_id()) returns?
>
> I just have a look at a 6-core system, the core id is not 0-5
>
> $ cat /proc/cpuinfo |grep "core id"
> core id     : 0
> core id     : 1
> core id     : 2
> core id     : 8
> core id     : 9
> core id     : 10
>
> So we'd better route all the interrupts to the first core of the socket.
>
I recently realized the issue with 0,1,2,8,9,10. At the time I wrote the perfmon
support for uncore, those systems did not exist. Sparse APIC id is a major pain
for uncore PMU interrupt routine given the way UNC_GLOBAL_CTRL works.

Unfortunately, routing to core 0 (core_cpu_id=0) won't be enough in the
presence of HOTPLUG CPU. Imagine I disable the first three 3 cores.
Now you the cpu you have to play with are 8,9,10. You need to remap
to a number between 0-5.


> Thanks for the catch.
> Lin Ming
>
>>
>> Note that, unfortunately, I have not seen documentation that says on
>> 6-core system
>> UNC_GLOBAL_CTRL has 6 interrupt target bits, but it would make sense.
>>
>>
>> Otherwise, you will get a kernel panic when you wrmsr UNC_GLOBAL_CTRL.
>>
>> > +
>> > + Â Â Â if (uncore->n_events == 1) {
>> > + Â Â Â Â Â Â Â nmi_core = topology_core_id(raw_smp_processor_id());
>> > + Â Â Â Â Â Â Â uncore->nmi_core = nmi_core;
>> > + Â Â Â Â Â Â Â uncore_pmu_enable_all(nmi_core);
>> > + Â Â Â }
>
>
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/