Hi Stephane and Paul,
After comparing sources, I didn't find any remarkable differences
between the two (bare 2.6.28-rc6 and Stephane's perfmon2-enhanced
2.6.28-rc6)
It didn't immediately occur to me that I could hit SysRq when the system
hung, so I decided to give that a try, and I did get a
Corey,
I looked at the pfm_handle_work() code. Normally
this functions gets called only if the TIF_PERFMON_WORK
flag is set. On Power, it seems the logic is coded differently.
You call systematically and in pfm_handle_work() you check
the flag, and if not set you return.
During boot, I don't expe
Hi,
I have fixed the problem now. It was due to the fact that we now tag
samples with pid/tid when they are added to hash table. But in aggregation,
you want to ignore pid/tid. So the patch makes pid/tid anonymous. Also
found a serialization problem.
Pull pfmon from CVS and try again. Let me know
Hi Stephane,
We are working through the example programs in the libpfm distribution
and came across multiplex.c and multiplex2.c
After fixing a superficial problem with get_cpu_speed(), we find that on
ppc970mp we are getting an error in pfm_write_pmds, where the smpl_pmds
field is set to some
Hello,
Today, I have fixed some important issues in pfmon. They were related
to --follow-*, dlopen hooks, trigger breakpoints support.
Sometimes, you would get:
'[xxx] interrupted @0x. for unknown reason'
And pfmon would hang.
There was a problem with how breakpoints were inherited ac
Hey, sounds interesting. Can you describe this for us geeks?
Phil
On Dec 18, 2008, at 10:54 AM, stephane eranian wrote:
> There was a problem with how breakpoints were inherited across fork
> and pthread_create. Fixing it was not easy as there is a race
> condition
> in the way a new thread/pr
Phil,
On Thu, Dec 18, 2008 at 5:21 PM, Philip Mucci wrote:
> Hey, sounds interesting. Can you describe this for us geeks?
>
> Phil
>
> On Dec 18, 2008, at 10:54 AM, stephane eranian wrote:
>
>> There was a problem with how breakpoints were inherited across fork
>> and pthread_create. Fixing it w
On Thu, Dec 18, 2008 at 11:00 PM, Corey J Ashford wrote:
> Thanks for your reply, Stephane,
>
> "stephane eranian" wrote on 12/18/2008 01:56:16
> AM:
>
>> Corey,
>>
>> I looked at the pfm_handle_work() code. Normally
>> this functions gets called only if the TIF_PERFMON_WORK
>> flag is set. On Po
Thanks for your reply, Stephane,
"stephane eranian" wrote on 12/18/2008 01:56:16
AM:
> Corey,
>
> I looked at the pfm_handle_work() code. Normally
> this functions gets called only if the TIF_PERFMON_WORK
> flag is set. On Power, it seems the logic is coded differently.
> You call systematical
Hi Stephane,
stephane eranian wrote:
> On Thu, Dec 18, 2008 at 11:00 PM, Corey J Ashford wrote:
>> Thanks for your reply, Stephane,
>>
>> "stephane eranian" wrote on 12/18/2008 01:56:16
>> AM:
>>
>>> Corey,
>>>
>>> I looked at the pfm_handle_work() code. Normally
>>> this functions gets called o
Corey,
On Fri, Dec 19, 2008 at 1:26 AM, Corey Ashford
wrote:
>
> I think you were pretty close to putting your finger on the problem. After
> stepping through the code, I see that it is looping, and I believe I know
> what the problem is. When CONFIG_PERFMON is defined, we call
> pfm_handle_work
11 matches
Mail list logo
