[prev in list] [next in list] [prev in thread] [next in thread] 

List:       freedesktop-dbus
Subject:    Re: Optimization problem when sending image data
From:       Sander Jansen <s.jansen () gmail ! com>
Date:       2009-08-20 17:31:21
Message-ID: 2d51e2b10908201031n75893cfbw975a8f216e30b99f () mail ! gmail ! com
[Download RAW message or body]

Actually, fixing 1) does make it behave the same on all optimization levels.
Thanks for the help,

Sander

On Thu, Aug 20, 2009 at 12:26 PM, Sander Jansen<s.jansen@gmail.com> wrote:
> It seems you are correct. I actually had two mistakes:
>
> 1) the bool being passed wasn't a 32 bit int. Fixing this one  didn't
> change the fact that in 02 and 03 optimized behaved different than the
> 01 optimized.
>
> 2) passing true with has_alpha  seems to make it work at any
> optimization level. Ofcourse now I'm not sure what has_alpha means,
> since we already pass the number of channels as well. The spec isn't
> very clear about it...
>
> Since just fixing 1) doesn't make it work the same across all
> optimizations levels, I'm still wondering what else could be wrong
>
> Cheers,
>
> Sander
>
> On Thu, Aug 20, 2009 at 12:00 PM, iain<iain@openedhand.com> wrote:
>> On Thu, 2009-08-20 at 11:33 -0500, Sander Jansen wrote:
>>> I'm trying to send a image to the notification daemon, but somehow the
>>> image get messed up (see attachment) whenever I turn on optimization
>>> to -O2 or -O3. The weird thing is that it used to work, so I'm kind of
>>> baffled why this is not working anymore.
>>
>> It looks like either you're passing alpha and the alpha isn't being
>> rendered, or you're not passing alpha but the notification daemon is
>> expecting alpha.
>>
>> iain
>>
>>
>
>
>
> --
> "And any fool knows a dog needs a home
> A shelter from pigs on the wing"
>



-- 
"And any fool knows a dog needs a home
A shelter from pigs on the wing"
_______________________________________________
dbus mailing list
dbus@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dbus

[prev in list] [next in list] [prev in thread] [next in thread]