wine 1.1.38/1.1.39 is totaly broken testet on fglrx 10.1/10.2

[Qaridarium]

wine 1.1.38/1.1.39 is totaly broken testet on fglrx 10.1/10.2

http://www.codeweavers.com/support/tickets/browse/?ticket_id=784736;layout=user_tickets;sort[status]=ASC

what is doing wrong???

[RealNC]

Your link says:

Access Denied!

You are not allowed to access this page. This is usually because the page is restricted to certain user accounts.
Contact us at website@codeweavers.com if you feel you have received this message in error.

[rbmorse]

Have you tried sending your "bug report" to WineHQ? A little more information about your system might be helpful, too.

[monraaf]

This link works:

http://www.codeweavers.com/support/t...cket_id=784736

You should really read this how to get Wine regressions fixed:

http://wiki.winehq.org/RegressionTesting

However the most likely explanation is that a valid commit in Wine may have triggered a bug in fglrx. If that's the case, well then you have to live with the bug . The fglrx team only cares about workstation customers, so they won't fix it and I don't think the Wine devs feel like providing a workaround for yet another fglrx bug.

[Qaridarium]

This link works:

http://www.codeweavers.com/support/t...cket_id=784736

You should really read this how to get Wine regressions fixed:

http://wiki.winehq.org/RegressionTesting

However the most likely explanation is that a valid commit in Wine may have triggered a bug in fglrx. If that's the case, well then you have to live with the bug . The fglrx team only cares about workstation customers, so they won't fix it and I don't think the Wine devs feel like providing a workaround for yet another fglrx bug.

ok thank you for you help...
may it can be a fglrx bug yes...

mor info abaut my system: its a hd4670 1gb vram and a dualsocket F opteron system with nvidia nforce3600 chipset.

i run kubuntu 9.10 and kernel 2.6.32

"The fglrx team only cares about workstation customers"

not really in the past they fix wine bugs to!

"I don't think the Wine devs feel like providing a workaround for yet another fglrx bug."

codewavers devs say to me they wana fix the fglrx only 'do not request on 16bit compatibility check' bug..

but codewavers do nor care abaut wine 1.1.38/39.. only crossover 9...

[Kano]

I used the git bisect way often enough. Together with ccache it takes some time, but it is ok. Why can't you do that on your one once?

[monraaf]

"The fglrx team only cares about workstation customers"

not really in the past they fix wine bugs to!

That must have been a fluke then. Probably fixed by an intern, who didn't get the message.

[bridgman]

Nope, quite the opposite in fact. The guy who interacts most with the Wine devs is the OpenGL driver architect.

If someone points out a problem with the OpenGL driver (eg for this operation behaviour A is expected but behavior B is observed) it's going to be looked at no matter what the parent app is.

On the other hand if the bug report is just "this app doesn't run right" then workstation apps and usage scenarios are often going to take precedence over consumer ones.

[monraaf]

Nope, quite the opposite in fact. The guy who interacts most with the Wine devs is the OpenGL driver architect.

If someone points out a problem with the OpenGL driver (eg for this operation behaviour A is expected but behavior B is observed) it's going to be looked at no matter what the parent app is.

Maybe from an ATI point of view. But somehow I get the feeling that the Wine devs are quite frustrated with the (lack of) interaction from ATI's part and ATI's (lack of) bug fixing. Allow me to quote parts of emails from some wine devs to public mailing lists.

It's not that we hate ATI. It's that there are plenty of bugs in
applications on Wine that ATI hasn't fixed, but work fine in
Nvidia/Intel. Stefand put a lot of work into making testcases, and
filed bugs with ATI about them, but they are reluctant to fix them
unless they affect 'a large number of popular applications'.

Mesa (that includes e.g. the intel and radeon drivers) is generally
responsive to bug reports, as is NVIDIA. It's really just AMD and
Apple that are problematic wrt. getting bugs fixed.

Just for the record, I'm not trying to just make AMD look bad here, I
think it's great that they released the GPU documentation, and that
they're paying people to work on the Mesa driver. Also, going by what
I hear from other people, fglrx improved a lot in the more recent
versions (certainly compared to the kind of support my Radeon 9800 Pro
had...), so it's not all bad. They just *really* need to fix their
developer relations stuff.

[bridgman]

How recent are those quotes ? They sound more like something I would expect to have heard a year ago than today.