wined3d: ActivateContext: try to fix wrong parameter?

[Rico Schüller]

Hi,

as suggested I send this patch for a discussion. This is a workaround 
for this bug http://bugs.winehq.org/show_bug.cgi?id=11584 .

I don't know if this patch is the right way to fix the problem.

Could someone with more knowledge have a look at this?

Cheers
Rico

[Roderick Colenbrander]

> Hi,
>
> as suggested I send this patch for a discussion. This is a workaround 
> for this bug http://bugs.winehq.org/show_bug.cgi?id=11584 .
>
> I don't know if this patch is the right way to fix the problem.
>
> Could someone with more knowledge have a look at this?
>
> Cheers
> Rico

It is not a workaround but a patch which restores the original behavior of the LoadLocation function. During the merge of the two codepaths I missed this part.

Roderick

[Stefan Dösinger]

Am Donnerstag, 21. Februar 2008 20:12:24 schrieb Roderick Colenbrander:
> It is not a workaround but a patch which restores the original behavior of
> the LoadLocation function. During the merge of the two codepaths I missed
> this part.

What happens if e.g. the current render target is an offscreen target and the 
application tries to read back the front or back buffer? In this case you'll 
read the offscreen target back.

So I am not sure this patch is correct

[Rico Schüller]

Stefan Dösinger wrote:
> What happens if e.g. the current render target is an offscreen target and the 
> application tries to read back the front or back buffer? In this case you'll 
> read the offscreen target back.
>
> So I am not sure this patch is correct

I have no idea I just tried a bit around and got cod4 
(http://bugs.winehq.org/show_bug.cgi?id=11584) working again with this 
patch.

I didn't step deep into this problem. So if I could help with some logs 
or other things, please let me know. Properly, the real bug is somewhere 
else?

Cheers
Rico

[Stefan Dösinger]

Am Freitag, 22. Februar 2008 22:45:23 schrieb Rico Schüller:
> I didn't step deep into this problem. So if I could help with some logs
> or other things, please let me know. Properly, the real bug is somewhere
> else?

Did you intend to attach a file to the mail, or did you refer to the last 
attachment to the bug?

[Rico Schüller]

I refer to the last attachment to the bug (and the patch which was sent 
in the initial mail 
http://www.winehq.org/pipermail/wine-devel/2008-February/062835.html - 
it's the same).

I just want to say if my patch is wrong and the other way is wrong (see 
bug http://bugs.winehq.org/show_bug.cgi?id=11584 ) then the real issue 
must be somewhere else!
To find that my knowledge isn't good enough yet. So I just want to have 
some help/hints to find the real source of the bug ;-)

Cheers
Rico

Stefan Dösinger wrote:
> Am Freitag, 22. Februar 2008 22:45:23 schrieb Rico Schüller:
> > I didn't step deep into this problem. So if I could help with some logs
> > or other things, please let me know. Properly, the real bug is somewhere
> > else?
>
> Did you intend to attach a file to the mail, or did you refer to the last 
> attachment to the bug?