- 25 Jul, 2012 2 commits
-
-
Henri Verbeet authored
-
Henri Verbeet authored
-
- 01 Jun, 2012 1 commit
-
-
Francois Gouget authored
-
- 11 Jan, 2012 1 commit
-
-
Henri Verbeet authored
-
- 10 Jan, 2012 1 commit
-
-
Henri Verbeet authored
-
- 20 Dec, 2011 1 commit
-
-
Henri Verbeet authored
-
- 14 Oct, 2011 1 commit
-
-
Andrew Talbot authored
-
- 25 Aug, 2011 1 commit
-
-
Matteo Bruni authored
-
- 25 Jul, 2011 1 commit
-
-
Henri Verbeet authored
-
- 14 Jul, 2011 1 commit
-
-
Henri Verbeet authored
Instead of an entire stateblock. This is mainly useful is we ever want to call state handlers on state not stored in a stateblock.
-
- 06 Jul, 2011 1 commit
-
-
Henri Verbeet authored
-
- 26 May, 2011 1 commit
-
-
Stefan Dösinger authored
-
- 17 May, 2011 1 commit
-
-
Henri Verbeet authored
-
- 31 Jan, 2011 1 commit
-
-
Henri Verbeet authored
-
- 02 Dec, 2010 3 commits
-
-
Henri Verbeet authored
-
Henri Verbeet authored
-
Henri Verbeet authored
-
- 28 Sep, 2010 2 commits
-
-
Henri Verbeet authored
-
Henri Verbeet authored
-
- 24 Sep, 2010 1 commit
-
-
Henri Verbeet authored
-
- 16 Sep, 2010 1 commit
-
-
Henri Verbeet authored
-
- 15 Sep, 2010 1 commit
-
-
Henri Verbeet authored
We'd like to not pass an entire stateblock to things like state handlers and the shader backend, because those then use the stateblock to get to the device and through there to all the rest of wined3d. This would also be required for serialization of wined3d draw and state change commands into a single GL context. Resource updates would be explicitly excluded from serialization.
-
- 14 Sep, 2010 1 commit
-
-
Henri Verbeet authored
-
- 03 Sep, 2010 1 commit
-
-
Henri Verbeet authored
-
- 25 May, 2010 2 commits
-
-
Henri Verbeet authored
-
Henri Verbeet authored
-
- 03 May, 2010 1 commit
-
-
Henri Verbeet authored
-
- 01 Apr, 2010 2 commits
-
-
Henri Verbeet authored
We should never call the handler for states that have a rep.
-
Henri Verbeet authored
-
- 04 Feb, 2010 1 commit
-
-
Henri Verbeet authored
On the subject of actually supporting software devices, we should probably implement those by creating a software or reference adapter with appropriate gl_info and pipe implementations and creating a device based on that.
-
- 10 Dec, 2009 1 commit
-
-
Henri Verbeet authored
-
- 29 Oct, 2009 1 commit
-
-
Henri Verbeet authored
-
- 22 Oct, 2009 2 commits
-
-
Henri Verbeet authored
-
Henri Verbeet authored
-
- 07 Aug, 2009 1 commit
-
-
Henri Verbeet authored
-
- 03 Aug, 2009 1 commit
-
-
Henri Verbeet authored
-
- 17 Jul, 2009 1 commit
-
-
Henri Verbeet authored
-
- 07 Jul, 2009 1 commit
-
-
Henri Verbeet authored
The compiler should be smart enough to fix most of these, but it looks sloppy.
-
- 26 Jun, 2009 1 commit
-
-
Henri Verbeet authored
-
- 10 Jun, 2009 1 commit
-
-
Stefan Dösinger authored
Thanks to Roland Scheidegger from Tungsten Graphics for the suggestion to replace the 2 movs + dp2add with two MADs, where one can conveniently be coissued with the other dp2add.
-