Trivial one-liner.
>From d1d5480209e83891dc929dea95aded0c0581fb2b Mon Sep 17 00:00:00 2001
From: Nathan Kidd
Date: Wed, 1 Aug 2012 22:10:20 -0400
Subject: [PATCH 4/8] Clarify glx pixmap unit test description
---
server/fakerut.cpp | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
We talked about this in the past, but I don't recall why virtualgl.org
stayed with opt-in-stencils rather than opt-out. I've been carrying
this patch for a few years with no complaints.
-Nathan
>From 9830736048ff418941d1e93ab73324af83c5960f Mon Sep 17 00:00:00 2001
From: Nathan Kidd
Date: Tue,
Sweeping through my tree I found a few things that I think are generally
useful.
For this patch I don't remember the application that triggered this, but
I did see the problem in practice in some obscure test application, IIRC.
-Nathan
>From 204399556e5a62964a3ba442f2abb9648313c6fa Mon Sep 17 00:
This seems tangentially related to the recent MATLAB post on users@.
Obviously this is only a band-aid for the potential general issue
(assuming extensions exist in the 3D X server without actually
checking), but it is maybe the right band-aid, given I've never seen
issues related to this before.
It's sort of pedantic. The assumption is that, in the first three
tests, the Pixmap is 3D, because it isn't meaningful to test XCopyArea()
otherwise (2D->2D would pass through to X11 without VirtualGL's
interference.) I don't have any real problem changing this, but I think
the first two (Win
I apparently even agreed to it:
http://sourceforge.net/p/virtualgl/mailman/message/28581272/
Not sure why it slipped through the cracks, but I just checked it in.
On 3/11/15 8:52 PM, Nathan Kidd wrote:
> We talked about this in the past, but I don't recall why virtualgl.org
> stayed with opt-in
