CPU underload

[Adrian Chadd]

Yeah, I'll poke at it tomorrow.

I should just be able to fire up this in gdb and figure it out.

How's it all looking so far? Is it still stable and still performing OK?



-a



On 1 November 2015 at 12:31, Eugene Grosbein <eugen at grosbein.net> wrote:
> 02.11.2015 2:36, Adrian Chadd пишет:
>>
>> Hi,
>>
>> Whilst I'm thinking about that, how about we take a look at the
>> unaligned accesses?
>>
>> I bet it's pmcstat, so can you do this:
>>
>> gdb pmcstat
>> list *0x(address of the 'pc' value returned by the kernel messages)
>>
>> I'd like to figure out where in pmcstat/libpmc it's doing stupid stuff
>> and undo it.
>>
>> Thanks!
>
>
> I'd like to, but there is no space for gdb on real device.
> Is it possible to use qemu for that?
>
>