Age | Commit message (Collapse) | Author | |
---|---|---|---|
2010-04-14 | r300g: simplify get_shader_constant to be rc-constant-state specific | Marek Olšák | |
2010-04-14 | r300g: remove the CS compenstation for non-atomized states | Marek Olšák | |
Now there aren't any. If CS overflow occurs, the fix should go directly before the problematic BEGIN_CS. | |||
2010-04-14 | r300g: remove always_dirty flag and correct leftovers | Marek Olšák | |
2010-04-14 | r300g: atomize beginning occlusion query | Marek Olšák | |
2010-04-14 | r300g: atomize VS constant buffer | Marek Olšák | |
2010-04-14 | r300g: emit VS immediates along with VS code | Marek Olšák | |
2010-04-14 | r300g: make setting up fragment depth output less hackish | Marek Olšák | |
2010-04-14 | r300g: kill off r300_fragment_shader::shadow_samplers and friends | Marek Olšák | |
2010-04-14 | trace: Ignore transfer_flush_region too. | José Fonseca | |
2010-04-13 | i915g: Whitespace & formating | Jakob Bornecrantz | |
2010-04-13 | i915g: Mark shared layouts more clearly | Jakob Bornecrantz | |
2010-04-13 | Merge branch '7.8' | Jesse Barnes | |
2010-04-13 | llvmpipe: disable half float formats | Alan Hourihane | |
2010-04-13 | nvfx: use dynamically sized rotating BO pool for fragment programs | Luca Barbieri | |
Currently we used a single buffer for each fragment programs, leading to rendering synchronization. This patch uses a doubly linked list of BOs, which is dynamically resized if all the BOs are busy. Note that inline image transfers could be an alternative option: this will be explored later. This removes one of the big performance limitations of the current driver. We also stop using pipe_resource internally in favor of using nouveau_bo directly. | |||
2010-04-13 | r300g: Silence uninitialized variable warning. | Vinson Lee | |
2010-04-13 | nvfx: fix for 64-bit systems | Luca Barbieri | |
Thanks to Patrice Mandin for debugging this. | |||
2010-04-13 | r300g: add generating texture coordinates for point sprites | Marek Olšák | |
[airlied - Convert sprite coord index to a per-coord enable bit set the rasteriser block up correctly for point sprites. The inputs to the RS hw block change for sprite coords, so fix them up properly - this fixes piglit point-sprite test. ] Signed-off-by: Dave Airlie <airlied@redhat.com> | |||
2010-04-12 | nvfx: add LRP in vertprog | Luca Barbieri | |
glsl generates these. | |||
2010-04-12 | nvfx: add SIN and COS in vertprog | Luca Barbieri | |
2010-04-12 | nvfx: add missing vertprog setcond instructions | Luca Barbieri | |
Trivially adds SEQ, SGT, SLE, SNE, SFL, STR and SSG which were missing. | |||
2010-04-12 | nvfx: allocate a bigger block for queries | Luca Barbieri | |
This patch allocates a bigger chunk of memory to store queries in, increasing the (hidden) outstanding query limit. | |||
2010-04-12 | nvfx: support an unlimited number of occlusion queries | Luca Barbieri | |
Currently on nv30/nv40 an assert will be triggered once 32 queries are outstanding. This violates the OpenGL/Gallium interface, which requires support for an unlimited number of fences. This patch fixes the problem by putting queries in a linked list and waiting on the oldest one if allocation fails. nVidia seems to use a similar strategy, but with 1024 instead of 32 fences. The next patch will improve this. | |||
2010-04-12 | r300g: Move declaration before code. | Vinson Lee | |
Fixes SCons build. | |||
2010-04-12 | nvfx: fix TXL opcode value | Luca Barbieri | |
Was broken during unification | |||
2010-04-12 | nvfx: Add support for two sided color | Luca Barbieri | |
This patch adds support for two-sided vertex color to nv30/nv40. When set, the COLOR0/1 fs inputs on back faces will be wired to vs outputs BCOLOR0/1. This makes OpenGL two sided lighting work, which can be tested with progs/demos/projtex. | |||
2010-04-12 | nvfx: fix compile error | Luca Barbieri | |
2010-04-12 | nvfx: don't use hw index buffers on nv3x and nv44 | Luca Barbieri | |
They only apparently work on nv40 grclass cards, and this was the previous behavior of the driver. This really needs to be investigated more. | |||
2010-04-12 | nvfx: expose PIPE_FORMAT_B8G8R8X8_UNORM support | Luca Barbieri | |
This is implemented in nvfx_state_fb and fragtex but was missing in nvfx_screen. This allows to avoid glCopyTexSubImage CPU fallbacks and makes Doom 3 much faster as a result. | |||
2010-04-12 | nvfx: add support for ARB_half_float_vertex | Luca Barbieri | |
2010-04-12 | nvfx: allocate buffers in GART if !NOUVEAU_SWTNL | Luca Barbieri | |
Should improve performance and fix serious regressions on AGP cards. | |||
2010-04-12 | nvfx: move check for NOUVEAU_SWTNL | Luca Barbieri | |
2010-04-12 | nvfx: remove #include "nouveau_stateobj.h" | Luca Barbieri | |
No longer used, finally! | |||
2010-04-12 | nvfx: rework validation logic to use code and avoid stateobjs | Luca Barbieri | |
This makes the code faster due to the lack of indirect calls and also makes it much easier to understand what is actually going on. | |||
2010-04-12 | nvfx: add NVFX_NEW_{FRAG,VERT}CONST | Luca Barbieri | |
2010-04-12 | nvfx: so->RING_3D: vertprog | Luca Barbieri | |
We could use an sb here instead, but there is no obvious performance advantage, and perhaps there could be a disadvantage. | |||
2010-04-12 | nvfx: so->RING_3D: sr | Luca Barbieri | |
2010-04-12 | nvfx: so->sb: blend | Luca Barbieri | |
2010-04-12 | nvfx: so->sb: zsa | Luca Barbieri | |
2010-04-12 | nvfx: so->sb: rasterizer | Luca Barbieri | |
2010-04-12 | nvfx: so->RING_3D: screen | Luca Barbieri | |
2010-04-12 | nvfx: so->RING_3D: viewport | Luca Barbieri | |
2010-04-12 | nvfx: so->RING_3D: stipple | Luca Barbieri | |
2010-04-12 | nvfx: so->RING_3D: scissor | Luca Barbieri | |
2010-04-12 | nvfx: so->RING_3D: blend | Luca Barbieri | |
2010-04-12 | nvfx: so->RING_3D: fb | Luca Barbieri | |
2010-04-12 | nvfx: so->RING_3D: fragprog | Luca Barbieri | |
2010-04-12 | nvfx: so->RING_3D: fragtex | Luca Barbieri | |
2010-04-12 | nvfx: so->RING_3D: vbo | Luca Barbieri | |
2010-04-12 | nvfx: stop using flush_notify | Luca Barbieri | |
Rather than emitting relocations on flush notifications, emit them in nvfx_state_start. | |||
2010-04-12 | nouveau: add state buffers, lightweight replacement for state objects | Luca Barbieri | |
Just a dumb buffer, allowed by the RING_3D/fixed subchannel binding and no support for relocations. This is *much* faster than state objects. |