Bug 17889

Summary:	s3v dri module calls undefined function s3vInitSpanFuncs
Product:	Mesa	Reporter:	Alan Coopersmith <alan.coopersmith>
Component:	Drivers/DRI/S3V	Assignee:	Default DRI bug account <dri-devel>
Status:	RESOLVED FIXED	QA Contact:
Severity:	normal
Priority:	medium	CC:	MostAwesomedude
Version:	7.1
Hardware:	Other
OS:	All
Whiteboard:
i915 platform:		i915 features:

Description Alan Coopersmith 2008-10-03 10:45:33 UTC

[Actually seen in Mesa 7.2, but that's not in the bugzilla version list yet, 
 and this bug appears to have been around for a while.]

Noticed while building Mesa 7.2 with all DRI modules using ld -z defs to fail
on undefined symbols - src/mesa/drivers/dri/s3v/s3v_context.c contains a call
to s3vInitSpanFuncs(), but that function was removed in this commit:

commit 687918b794d12a0ff5d395b8cf87880dbe8012bf
Author: Brian Paul <brian.paul@tungstengraphics.com>
Date:   Sat Sep 3 16:43:02 2005 +0000

    SetBuffer, renderbuffer changes

(I don't have any s3v hardware to see if this causes actual failure or not.)

Comment 1 maximlevitsky 2008-10-03 12:11:24 UTC

I actually have one in the closet, but s3virge has no drm module, but only mesa module (drm module was developed, but it is very very very outdated, and isn't included anywhere but in 0.01 branch in drm tree)

So mesa support is just dead code.

Comment 2 Alan Coopersmith 2008-10-04 15:45:53 UTC

Ah, I didn't even check if it was usable, it was just in the default
set of modules built by the autoconf configuration in 7.2.

Comment 3 maximlevitsky 2009-10-21 14:26:19 UTC

how about removing s3virge support from mesa?
And closing this bug...

Comment 4 Corbin Simpson 2009-10-21 19:12:18 UTC

You probably won't ever get the code dropped from Mesa. I'd recommend finding somebody adventurous with the proper HW that wouldn't mind updating the driver and actually attempting to keep it.

I'm not at home so I can't really check to see if I have one.

Comment 5 Ian Romanick 2009-10-21 23:56:25 UTC

(In reply to comment #4)
> You probably won't ever get the code dropped from Mesa. I'd recommend finding
> somebody adventurous with the proper HW that wouldn't mind updating the driver
> and actually attempting to keep it.
> 
> I'm not at home so I can't really check to see if I have one.

Since this driver was never finish and has never worked, I vote to 'git rm -r' it.  It's just an annoyance at this point.  The last actually development commit to this driver (not just compile fixes) was in 2004.  I won't hold my breath while we wait for someone to come along and care about this driver.

If you want to finish it, anholt and I can give you a PILE of Virge cards next time you're in Portland. :)

Comment 6 Corbin Simpson 2009-10-22 12:32:08 UTC

Ah. If it doesn't work, then mercy killing is probably the way to go. Can always revive it from history if anybody decides to start caring about it.

Closing as FIXED with commit 8e4657a.

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.