[PATCH] complete (I hope!) i830 EXA support

[Jesse Barnes]

On Saturday, September 10, 2005 1:31 pm, Bernhard Kaindl wrote:
> I see that it woudl be called is in I830DRIInitBuffers which is
> also the last debug message which I pasted before the call, I
> see it directly after the this debug message:
>
>    if (I810_DEBUG & DEBUG_VERBOSE_DRI)
>       ErrorF("I830DRIInitBuffers\n");
>
>    I830SetupForSolidFill(pScrn, 0, GXcopy, -1);

Ah, DRI, I missed that.  I'll have to fix that up somehow.

> > So you're using the patch but using "AccelMethod" "XAA"?
>
> No, I use Option "AccelMethod" "exa" and with -verbose 8, I get these
> messages from the driver (excerpts only):

Ok, sounds like you've got the right one, it just sounded like a bug I 
fixed awhile back.  Either it came back or this is a different bug 
(probably the latter).

> In the patch which I see, I see both EXA and XAA enabled by default:

Right, they're both compiled in unless you change the Imakefile.

> I think it makes sense to compile both by default as long as there
> is still some valid reason why to use XAA. I don't know but if XAA
> could become obsolete by EXA and unmaintained, then the ifdefs could
> be good to isolate the code which could be deleted at a later date.

I agree, that was my feeling too.

> I'd assume that i810_dri.c would get changed accordingly, but
> I don't see a change to i810_dri.c in the patch which I have.

You're right, I just wasn't exercising that codepath so I didn't see the 
problem.  It definitely needs to get fixed.

Thanks a lot for your thorough testing and bug reports, they're very 
helpful.

Jesse