[xdebug-general] Re: Profiling a Joomla page -- huge mismatch between xdebug profiling information and actual performance

Hi,

Wincachegrind's reported timings must be multiplied by 10. xDebug 2.0.0 RC4
changed the way timings are reported (µ seconds instead of tenth of µ
seconds), and unfortunately Wincachegrind hasn't been updated for years.

So if you see 1.5 seconds in Wincachegrind, it really is 15 seconds :)

-----Message d'origine-----
De : xdebug-general-bounce[@]lists.xdebug.org
[mailto:xdebug-general-bounce[@]lists.xdebug.org] De la part de Emiliano Heyns
Envoyé : lundi 30 mars 2009 18:03
À : A Paella
Cc : xdebug-general[@]lists.xdebug.org
Objet : [xdebug-general] Re: Profiling a Joomla page -- huge mismatch
between xdebug profiling information and actual performance

I'm using wincachegrind, currently, but I have a linux system at home
I could use for other clients. How so? Would different data be
reported by different viewers?

Met vriendelijke groet,
Emiliano Heyns

On 30 mrt 2009, at 17:43, A Paella <apaella[@]gmail.com> wrote:

> Wich client are you using to visualize the cachegrind file?
>
> On Mon, Mar 30, 2009 at 3:07 PM, Emiliano Heyns
> <Emiliano.Heyns[@]iris-advies.nl> wrote:
>> Hi,
>>
>> We're trying to find out why our pages are loading so slow (12-13
>> seconds),
>> so we've ran xdebug against it, which is showing that the page
>> executes in
>> 1.5 seconds. It's not a network delay, as we've loaded this page
>> from the
>> console of the webserver itself, using lynx (so no javascript, css
>> or images
>> causing the slowdown, either). The slowdown appears to be at the
>> "tail" of
>> the request, since the timestamp for the request in the apache log
>> and the
>> timestamp of the xdebug output file are the same. Does anyone have
>> any idea
>> where else we might look?
>>
>> Thanks,
>> Emile
>>
>>
Received on Mon Mar 30 2009 - 20:04:12 BST