Commit Graph

39536 Commits

Author SHA1 Message Date
Tom Stellard b27a809266 r300/compiler: Don't merge instructions that write output regs and ALU result
https://bugs.freedesktop.org/show_bug.cgi?id=30415

NOTE: This is a candidate for the 7.9 branch.
2010-09-28 23:52:41 -07:00
Tom Stellard 1b76dde0cd r300/compiler: Don't use rc_error() unless the error is unrecoverable
https://bugs.freedesktop.org/show_bug.cgi?id=30416

NOTE: This is a candidate for the 7.9 branch.
2010-09-28 23:52:41 -07:00
Tom Stellard d40ff5510c r300/compiler: Fix segfault in error path
https://bugs.freedesktop.org/show_bug.cgi?id=30415

NOTE: This is a candidate for the 7.9 branch.
2010-09-28 23:52:41 -07:00
Zhenyu Wang 73dab75b41 i965: fallback lineloop on sandybridge for now
Until we fixed GS hang issue.
2010-09-29 14:35:19 +08:00
Zhenyu Wang 0a1910c267 i965: Always set tiling for depth buffer on sandybridge
Sandybridge only support tiling depth buffer, always set tiling bit.

Fix 'fbo_firecube' demo.
2010-09-29 14:02:37 +08:00
Dave Airlie 28b57c56e2 r600g: remove old assert from new codepath
this fixes draw-elements-base-vertex
2010-09-29 14:52:39 +10:00
Dave Airlie 914b669b08 r600g: add initial vertex translate support. 2010-09-29 14:41:16 +10:00
Kenneth Graunke 565ff67688 glsl: "Copyright", not "Constantright"
Clearly this started out as ir_copy_propagation.cpp, but the search and
replace was a bit overzealous.
2010-09-28 21:17:33 -07:00
Eric Anholt 1747aa6755 i965: Add support for builtin uniforms to the new FS backend.
Fixes 8 piglit tests.
2010-09-28 16:31:10 -07:00
Eric Anholt daacaac3c8 mesa: Move the list of builtin uniform info from ir_to_mesa to shared code.
I'm still not pleased with how builtin uniforms are handled, but as
long as we're relying on the prog_statevar stuff this seems about as
good as it'll get.
2010-09-28 16:26:58 -07:00
Eric Anholt 9ac910cfcd i965: Clean up obsolete FINISHME comment. 2010-09-28 16:26:58 -07:00
Eric Anholt ff0eb45f47 i965: Fix array indexing of arrays of matrices.
The deleted code was meant to be handling indexing of a matrix, which
would have been a noop if it had been correct.
2010-09-28 16:26:49 -07:00
Dave Airlie 301ab49605 r600g: move radeon.h members around to add back map flushing. 2010-09-29 09:19:22 +10:00
Dave Airlie 53b3933ce6 r600g: add evergreen texture border support to new path 2010-09-29 09:19:22 +10:00
Dave Airlie 23be883c9b r600g: add back evergreen name. 2010-09-29 09:19:22 +10:00
Eric Anholt 17f3b8097d i965: Don't try to emit interpolation for unused varying slots.
Fixes:
glsl-fs-varying-array
glsl-texcoord-array
glsl-texcoord-array-2
glsl-vs-varying-array
2010-09-28 14:53:36 -07:00
Eric Anholt 5272c6a7a2 i965: Do interpolation for varying matrices and arrays in the FS backend.
Fixes:
glsl-array-varying-01
glsl-vs-mat-add-1
glsl-vs-mat-div-1
glsl-vs-mat-div-2
glsl-vs-mat-mul-2
glsl-vs-mat-mul-3
2010-09-28 14:50:59 -07:00
Eric Anholt 586b4b500f glsl: Also update implicit sizes of varyings at link time.
Otherwise, we'll often end up with gl_TexCoord being 0 length, for
example.  With ir_to_mesa, things ended up working out anyway, as long
as multiple implicitly-sized arrays weren't involved.
2010-09-28 14:37:26 -07:00
Eric Anholt b9a59f0358 i965: Add support for ARB_fragment_coord_conventions to the new FS backend.
Fixes:
glsl-arb-frag-coord-conventions
glsl-fs-fragcoord
2010-09-28 13:42:52 -07:00
Eric Anholt 701c5f11c9 i965: Add support for ir_loop counters to the new FS backend.
Fixes:
glsl1-discard statement in for loop
glsl-fs-loop-two-counter-02
glsl-fs-loop-two-counter-04
2010-09-28 13:31:01 -07:00
Tilman Sauerbeck 35f94b1942 r600g: Cleaned up index buffer reference handling in the draw module.
This fixes a buffer leak.

Signed-off-by: Tilman Sauerbeck <tilman@code-monkey.de>
2010-09-28 22:12:23 +02:00
Eric Anholt 89f6783d17 i965: Add support for MRT to the new FS backend.
Fixes these tests using gl_FragData or just gl_FragDepth:
glsl1-Preprocessor test (extension test 1)
glsl1-Preprocessor test (extension test 2)
glsl-bug-22603
2010-09-28 12:37:21 -07:00
Eric Anholt 86fd11262c i965: Add support for non-color render target write data to new FS backend.
This is the first time these payload bits have made sense to me,
outside of brw_wm_pass* structure.

Fixes: glsl1-gl_FragDepth writing
2010-09-28 12:37:21 -07:00
Vinson Lee f46a61554f scons: Add program/sampler.cpp to SCons build.
This is a follow-up to commit a32893221c.

Fixes MinGW SCons build.
2010-09-28 12:03:45 -07:00
Eric Anholt 2999a44968 i965: Set up sampler numbers in the FS backend.
+10 piglits
2010-09-28 11:37:08 -07:00
Eric Anholt a32893221c mesa: Pull ir_to_mesa's sampler number fetcher out to shared code. 2010-09-28 11:37:08 -07:00
Jerome Glisse 723a655ed3 r600g: avoid rebuilding the vertex shader if no change to input format
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-28 14:34:25 -04:00
Jerome Glisse fe790a3c34 r600g: suspend/resume occlusion query around clear/copy
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-28 14:24:18 -04:00
Marek Olšák 11eb422a16 configure.ac: do not build xorg-r300g by default
NOTE: This is a candidate for the 7.9 branch.
2010-09-28 19:38:40 +02:00
Marek Olšák a1aec2e2be configure.ac: look for libdrm_radeon before building gallium/r300,r600
NOTE: This is a candidate for the 7.9 branch.
2010-09-28 19:38:39 +02:00
Eric Anholt 9e96c737f8 i965: Subtract instead of adding when computing y delta in new FS backend.
Fixes 7 piglit cases.
2010-09-28 10:19:54 -07:00
Eric Anholt 5f7bd68149 i965: Add support for gl_FrontFacing to the new FS backend.
Fixes:
glsl1-gl_FrontFacing var (1)
glsl1-gl_FrontFacing var (2)
2010-09-28 10:10:44 -07:00
Eric Anholt ef8e002c75 i965: Fix up part of my Sandybridge attributes support patch.
I confused the array sizing for number of files for the number of regs
in a file.
2010-09-28 10:10:42 -07:00
Eric Anholt f1dba03056 i965: Fix all non-snb regression in the snb attribute interpolation commit.
This apparently had never been tested elsewhere before being merged to
master.
2010-09-28 10:10:42 -07:00
Eric Anholt 6bf12c8b73 i965: Add support for struct, array, and matrix uniforms to FS backend.
Fixes 16 piglit cases.
2010-09-28 09:33:31 -07:00
Eric Anholt ba481f2046 i965: Add support for dereferencing structs to the new FS backend.
Fixes: glsl1-struct(2)
2010-09-28 09:33:31 -07:00
Eric Anholt 07fc8eed8f i965: Set the variable type when dereferencing an array.
We don't set the type on the array virtual reg as a whole, so here's
the right place.

Fixes:
glsl1-GLSL 1.20 arrays
glsl1-temp array with constant indexing, fragment shader
glsl1-temp array with swizzled variable indexing
2010-09-28 09:33:31 -07:00
Eric Anholt 719f84d9ab i965: Fix up the FS backend for the variable array indexing pass.
We need to re-run channel expressions afterwards as it generates new
vector expressions, and we need to successfully support conditional
assignment (brw_CMP takes 2 operands, not 1).
2010-09-28 09:33:30 -07:00
Eric Anholt 57edd7c5c1 i965: Fix valgrind complaint about base_ir for new FS debugging. 2010-09-28 09:33:30 -07:00
Eric Anholt 1723fdb3f0 i965: Apply the same set of lowering passes to new FS as to Mesa IR.
While much of this we will want to support natively, this should make
the task of reaching the Mesa IR backend's quality easier.

Fixes:
glsl-fs-main-return.
2010-09-28 09:33:30 -07:00
Eric Anholt e10508812a i965: Actually track the "if" depth in loop in the new FS backend.
Fixes:
glsl-fs-if-nested-loop.
2010-09-28 09:33:30 -07:00
Eric Anholt fceb78e3cc i965: Fix negation in the new FS backend.
Fixes:
glsl1-Negation
glsl1-Negation2
2010-09-28 09:33:30 -07:00
Jerome Glisse 7ee8fa0421 r600g: switch to new design
New design seems to be on parity according to piglit,
make it default to get more exposure and see if there
is any show stopper in the coming days.

Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-28 11:37:30 -04:00
Jerome Glisse b534eb16a2 r600g: fix remaining piglit issue in new design
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-28 11:12:03 -04:00
Jerome Glisse 5a38cec7c8 r600g: use ptr for blit depth uncompress function
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2010-09-28 09:51:08 -04:00
Christoph Bumiller e0b93c5beb nv50: fix GP state bind and validate 2010-09-28 11:22:59 +02:00
Dave Airlie 175261a1f1 r600g: on evergreen the centroid isn't set in this register. 2010-09-28 19:02:46 +10:00
Zhenyu Wang 45b37c4b12 i965: fallback bitmap operation on sandybridge
Need to bring back correct fb write with header to set pixel
write mask. Fallback for now.
2010-09-28 15:58:21 +08:00
Zhenyu Wang 3074b61f64 i965: fix occlusion query on sandybridge
Fix pipe control command for depth stall and PS_DEPTH_COUNT write.
2010-09-28 15:58:21 +08:00
Zhenyu Wang ec99833e92 i965: fix point sprite on sandybridge
Need to set point sprite function in fixed SF state now on sandybridge.
2010-09-28 15:58:21 +08:00