[ltt-dev] Kernel and User Space tracers have different timestamp reference

[ltt-dev] Kernel and User Space tracers have different timestamp reference

[Alexandre Montplaisir]

Hi folks,

I've received this bug report about kernel/UST trace synchronization. 
Anyone knows what's the status on this?

Thanks,
Alexandre


On 10-11-26 03:24 AM, Florent Boudet wrote:
> Hello,
>
> I am sorry to contact you directly instead of filling a bug, but it
> seems the option is disabled on the launchpad project...
>
> I have recorded userspace traces at the same time as kernel traces, and
> opened both with LTTv.
> Timestamp for userspace traces and timestamp for kernel traces have
> different origins, which make it difficult to analyze.
> I am not sure about the cause, this might be related to a GMT/localtime
> issue...
>
> Regards,
> Florent Boudet

[ltt-dev] Kernel and User Space tracers have different timestamp reference

[Mathieu Desnoyers]

Hi!

Yesterday evening, Julien and I were actively finishing the vDSO work
that is needed to have synchronized timestamps between the kernel LTTng
and UST on x86. It should be ready very soon. Keep posted for a new
release. :)

By the way, in the initial implementation, only 64-bit processes on
64-bit x86 systems with synchronized TSCs will benefit from the vDSO
speed. All other x86 setups will fallback on a system call, which will
still work, albeit being a bit slower.

Thanks,

Mathieu

* Alexandre Montplaisir (alexandre.montplaisir at polymtl.ca) wrote:
> Hi folks,
>
> I've received this bug report about kernel/UST trace synchronization.  
> Anyone knows what's the status on this?
>
> Thanks,
> Alexandre
>
>
> On 10-11-26 03:24 AM, Florent Boudet wrote:
>> Hello,
>>
>> I am sorry to contact you directly instead of filling a bug, but it
>> seems the option is disabled on the launchpad project...
>>
>> I have recorded userspace traces at the same time as kernel traces, and
>> opened both with LTTv.
>> Timestamp for userspace traces and timestamp for kernel traces have
>> different origins, which make it difficult to analyze.
>> I am not sure about the cause, this might be related to a GMT/localtime
>> issue...
>>
>> Regards,
>> Florent Boudet
>
>
> _______________________________________________
> ltt-dev mailing list
> ltt-dev at lists.casi.polymtl.ca
> http://lists.casi.polymtl.ca/cgi-bin/mailman/listinfo/ltt-dev
>

-- 
Mathieu Desnoyers
Operating System Efficiency R&D Consultant
EfficiOS Inc.
http://www.efficios.com

[ltt-dev] Kernel and User Space tracers have different timestamp reference

[David Goulet]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Yes,

And also, the UST patch is done and ready for merge :)

Very soon guys!

David

On 10-11-26 08:32 AM, Mathieu Desnoyers wrote:
> Hi!
> 
> Yesterday evening, Julien and I were actively finishing the vDSO work
> that is needed to have synchronized timestamps between the kernel LTTng
> and UST on x86. It should be ready very soon. Keep posted for a new
> release. :)
> 
> By the way, in the initial implementation, only 64-bit processes on
> 64-bit x86 systems with synchronized TSCs will benefit from the vDSO
> speed. All other x86 setups will fallback on a system call, which will
> still work, albeit being a bit slower.
> 
> Thanks,
> 
> Mathieu
> 
> * Alexandre Montplaisir (alexandre.montplaisir at polymtl.ca) wrote:
>> Hi folks,
>>
>> I've received this bug report about kernel/UST trace synchronization.  
>> Anyone knows what's the status on this?
>>
>> Thanks,
>> Alexandre
>>
>>
>> On 10-11-26 03:24 AM, Florent Boudet wrote:
>>> Hello,
>>>
>>> I am sorry to contact you directly instead of filling a bug, but it
>>> seems the option is disabled on the launchpad project...
>>>
>>> I have recorded userspace traces at the same time as kernel traces, and
>>> opened both with LTTv.
>>> Timestamp for userspace traces and timestamp for kernel traces have
>>> different origins, which make it difficult to analyze.
>>> I am not sure about the cause, this might be related to a GMT/localtime
>>> issue...
>>>
>>> Regards,
>>> Florent Boudet
>>
>>
>> _______________________________________________
>> ltt-dev mailing list
>> ltt-dev at lists.casi.polymtl.ca
>> http://lists.casi.polymtl.ca/cgi-bin/mailman/listinfo/ltt-dev
>>
> 

- -- 
David Goulet
LTTng project, DORSAL Lab.

1024D/16BD8563
BE3C 672B 9331 9796 291A  14C6 4AF7 C14B 16BD 8563
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)

iEYEARECAAYFAkzvunoACgkQSvfBSxa9hWP1gQCfTTq08bZsAX5kZd9aL98gTUPJ
1bsAnRJ2LzczcavguNjeADa5WrtN1MoW
=jWhQ
-----END PGP SIGNATURE-----

[ltt-dev] Kernel and User Space tracers have different timestamp reference

[florent boudet]

Hi!

Thank you for your answers, I will keep posted for new releases.
By the way, LTTng and UST are awesome tools, thank you for the hard work!

Florent

On Fri, Nov 26, 2010 at 2:47 PM, David Goulet <david.goulet at polymtl.ca> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Yes,
>
> And also, the UST patch is done and ready for merge :)
>
> Very soon guys!
>
> David
>
> On 10-11-26 08:32 AM, Mathieu Desnoyers wrote:
>> Hi!
>>
>> Yesterday evening, Julien and I were actively finishing the vDSO work
>> that is needed to have synchronized timestamps between the kernel LTTng
>> and UST on x86. It should be ready very soon. Keep posted for a new
>> release. :)
>>
>> By the way, in the initial implementation, only 64-bit processes on
>> 64-bit x86 systems with synchronized TSCs will benefit from the vDSO
>> speed. All other x86 setups will fallback on a system call, which will
>> still work, albeit being a bit slower.
>>
>> Thanks,
>>
>> Mathieu
>>
>> * Alexandre Montplaisir (alexandre.montplaisir at polymtl.ca) wrote:
>>> Hi folks,
>>>
>>> I've received this bug report about kernel/UST trace synchronization.
>>> Anyone knows what's the status on this?
>>>
>>> Thanks,
>>> Alexandre
>>>
>>>
>>> On 10-11-26 03:24 AM, Florent Boudet wrote:
>>>> Hello,
>>>>
>>>> I am sorry to contact you directly instead of filling a bug, but it
>>>> seems the option is disabled on the launchpad project...
>>>>
>>>> I have recorded userspace traces at the same time as kernel traces, and
>>>> opened both with LTTv.
>>>> Timestamp for userspace traces and timestamp for kernel traces have
>>>> different origins, which make it difficult to analyze.
>>>> I am not sure about the cause, this might be related to a GMT/localtime
>>>> issue...
>>>>
>>>> Regards,
>>>> Florent Boudet
>>>
>>>
>>> _______________________________________________
>>> ltt-dev mailing list
>>> ltt-dev at lists.casi.polymtl.ca
>>> http://lists.casi.polymtl.ca/cgi-bin/mailman/listinfo/ltt-dev
>>>
>>
>
> - --
> David Goulet
> LTTng project, DORSAL Lab.
>
> 1024D/16BD8563
> BE3C 672B 9331 9796 291A ?14C6 4AF7 C14B 16BD 8563
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.10 (GNU/Linux)
>
> iEYEARECAAYFAkzvunoACgkQSvfBSxa9hWP1gQCfTTq08bZsAX5kZd9aL98gTUPJ
> 1bsAnRJ2LzczcavguNjeADa5WrtN1MoW
> =jWhQ
> -----END PGP SIGNATURE-----
>