summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* i965/gen9: Handle the GL_TEXTURE_{1D, 1D_ARRAY} targets inside switchAnuj Phogat2015-10-191-4/+4
| | | | | Signed-off-by: Anuj Phogat <[email protected]> Reviewed-by: Topi Pohjolainen <[email protected]>
* i965/fs: Localize variables' scopes.Matt Turner2015-10-191-8/+5
|
* i965/fs: Consider type mismatches in saturate propagation.Matt Turner2015-10-191-3/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | NIR considers bcsel to produce and consume unsigned types, leading to SEL instructions operating on unsigned types when the data is really floating-point. Previous to this patch, saturate propagation would happily transform (+f0) sel g20:UD, g30:UD, g40:UD mov.sat g50:F, g20:F into (+f0) sel.sat g20:UD, g30:UD, g40:UD mov g50:F, g20:F But since the meaning of .sat is dependent on the type of the destination register, this is not valid. Instead, allow saturate propagation to change the types of dest/source on instructions that are simply copying data in order to propagate the saturate modifier. Fixes bad code gen in 158 programs. Reviewed-by: Jason Ekstrand <[email protected]>
* i965: Extract can_change_source_types() functions.Matt Turner2015-10-196-27/+30
| | | | | | | | | Make them members of fs_inst/vec4_instruction for use elsewhere. Also fix the fs version to check that dst.type == src[1].type and for !saturate. Reviewed-by: Jason Ekstrand <[email protected]>
* i965/vs: Move URB entry_size and read_length calculations to compile_vsJason Ekstrand2015-10-192-34/+34
| | | | Reviewed-By: Eduardo Lima Mitev <[email protected]>
* i965: Move the entire compiler API into a single fileJason Ekstrand2015-10-1910-619/+665
| | | | | | | | | | | At this point, the compiler API has been substantially simplified. In the spirit of Kristian's making a compiler library, this commit makes a single header file that contains, more-or-less, the entire compiler API. There's still a bit of cleanup to do particularly in the area of geometry shaders. However, this gets us much closer to having a separate compiler. Reviewed-by: Topi Pohjolainen <[email protected]>
* i965: Rename brw_foo_emit to brw_compile_fooJason Ekstrand2015-10-1911-66/+67
| | | | Reviewed-by: Topi Pohjolainen <[email protected]>
* i965/fs: Move some of the prog_data setup into brw_wm_emitJason Ekstrand2015-10-192-98/+100
| | | | | | | | This commit moves the common/modern stuff. Some legacy stuff such as setting use_alt_mode was left because it needs to know whether or not we're an ARB program. Reviewed-by: Topi Pohjolainen <[email protected]>
* i965/cs: Rework cs_emit to take a nir_shader and a brw_compilerJason Ekstrand2015-10-193-29/+42
| | | | | | | This commit removes all dependence on GL state by getting rid of the brw_context parameter and the GL data structures. Reviewed-by: Topi Pohjolainen <[email protected]>
* i965/gs: Rework gs_emit to take a nir_shader and a brw_compilerJason Ekstrand2015-10-193-40/+29
| | | | | | | | | This commit removes all dependence on GL state by getting rid of the brw_context parameter and the GL data structures. Unfortunately, we still have to pass in the gl_shader_program for gen6 because it's needed for transform feedback. Reviewed-by: Topi Pohjolainen <[email protected]>
* i965/vs: Rework vs_emit to take a nir_shader and a brw_compilerJason Ekstrand2015-10-193-49/+49
| | | | | | | | | | | This commit removes all dependence on GL state by getting rid of the brw_context parameter and the GL data structures. v2 (Jason Ekstrand): - Patch use_legacy_snorm_formula through as a function argument rather than trying to go through the shader key. Reviewed-by: Topi Pohjolainen <[email protected]>
* i965/fs: Rework wm_fs_emit to take a nir_shader and a brw_compilerJason Ekstrand2015-10-193-39/+47
| | | | | | | This commit removes all dependence on GL state by getting rid of the brw_context parameter and the GL data structures. Reviewed-by: Topi Pohjolainen <[email protected]>
* i965: Use a const nir_shader in backend_shaderJason Ekstrand2015-10-1911-12/+12
| | | | Reviewed-by: Topi Pohjolainen <[email protected]>
* i965/vec4: Remove gl_program and gl_shader_program from the generatorJason Ekstrand2015-10-194-29/+20
| | | | Reviewed-by: Topi Pohjolainen <[email protected]>
* i965/fs: Remove the gl_program from the generatorJason Ekstrand2015-10-195-9/+5
| | | | Reviewed-by: Topi Pohjolainen <[email protected]>
* nir/info: Add a few bits of info for fragment shadersJason Ekstrand2015-10-193-0/+29
| | | | Reviewed-by: Topi Pohjolainen <[email protected]>
* nir/info: Add compute shader local size to nir_shader_infoJason Ekstrand2015-10-192-0/+12
| | | | Reviewed-by: Topi Pohjolainen <[email protected]>
* nir/info: Move the GS info into a stage-specific info unionJason Ekstrand2015-10-192-8/+18
| | | | | | | | This way we can have other stage-specific info without consuming too much extra space. While we're at it, we make sure that the geometry info is only set if we're actually a goemetry shader. Reviewed-by: Topi Pohjolainen <[email protected]>
* mesa: Move gl_frag_depth_layout from mtypes.h to shader_enums.hJason Ekstrand2015-10-192-18/+17
| | | | Reviewed-by: Topi Pohjolainen <[email protected]>
* nir: Add a label to nir_shader_infoJason Ekstrand2015-10-193-0/+7
| | | | Reviewed-by: Topi Pohjolainen <[email protected]>
* i965/asm: Explicitly use a nir_instr for IR annotationsJason Ekstrand2015-10-194-19/+5
| | | | | | | Now that everything goes through NIR, we don't need this to be a void pointer anymore. Reviewed-by: Topi Pohjolainen <[email protected]>
* scons: Build nir/glsl_types.cpp once.Jose Fonseca2015-10-198-29/+8
| | | | | | | | | | | | Undoes early hacks, and ensures nir/glsl_types.cpp is built once, and only once. The root problem is that SCons doesn't know about NIR nor any source file in the NIR_FILES source list. Tested with libgl-gdi and libgl-xlib scons targets. Reviewed-by: Brian Paul <[email protected]>
* svga: fix incorrect round-down arithmeticBrian Paul2015-10-191-1/+1
| | | | | | | | | Spotted by Roland. Luckily, this code should never really be hit since the const buffer size and offset should already be multiples of 16. I could probably add more assertions to that effect, but let's just fix the arithmetic for now. Reviewed-by: Roland Scheidegger <[email protected]>
* glsl: fix segfault when indirect indexing a buffer variable which is an arraySamuel Iglesias Gonsalvez2015-10-191-1/+2
| | | | | | | Fixes a regression added by bb5aeb854915ba67abc56257f830d002c956439e. Signed-off-by: Samuel Iglesias Gonsalvez <[email protected]> Reviewed-by: Timothy Arceri <[email protected]>
* st/va: Added support for NV12 to IYUV conversion in vlVaGetImageIndrajit Das2015-10-191-3/+5
| | | | Reviewed-by: Christian König <[email protected]>
* st/va: Used correct parameter to derive the value of the "h" variable in ↵Indrajit Das2015-10-191-1/+1
| | | | | | | | vlVaCreateImage Cc: "11.0" <[email protected]> Reviewed-by: Christian König <[email protected]> Reviewed-by: Emil Velikov <[email protected]>
* glsl_to_tgsi: Use {Num}UniformBlocks instead of {Num}BufferInterfaceBlocksIago Toral Quiroga2015-10-191-2/+2
| | | | | | | The latter holds both UBOs and SSBOs, but here we only want UBOs. Reviewed-by: Kristian Høgsberg <[email protected]> Reviewed-by: Marek Olšák <[email protected]>
* st/mesa: Use {Num}UniformBlocks instead of {Num}BufferInterfaceBlocksIago Toral Quiroga2015-10-191-2/+2
| | | | | | | The latter holds both UBOs and SSBOs, but here we only want UBOs. Reviewed-by: Kristian Høgsberg <[email protected]> Reviewed-by: Marek Olšák <[email protected]>
* i965: Do not use NumBufferInterfaceBlocksIago Toral Quiroga2015-10-191-1/+1
| | | | | | | | | | This is the only place in the driver where we use this. Since we now work with separate index spaces, always use NumUniformBlocks and NumShaderStorageBlocks instead of NumBufferInterfaceBlocks to be more consistent with the rest of the code. Reviewed-by: Kristian Høgsberg <[email protected]> Reviewed-by: Marek Olšák <[email protected]>
* main: GL_ACTIVE_UNIFORM_BLOCK_MAX_NAME_LENGTH is about UBOS, not SSBOsIago Toral Quiroga2015-10-191-2/+2
| | | | | Reviewed-by: Kristian Høgsberg <[email protected]> Reviewed-by: Marek Olšák <[email protected]>
* main: Use NumUniformBlocks to count UBOsIago Toral Quiroga2015-10-191-5/+1
| | | | | | | | | Now that we have separate index spaces for UBOs and SSBOs we do not need to iterate through BufferInterfaceBlocks any more, we can just take the UBO count directly from NumUniformBlocks. Reviewed-by: Kristian Høgsberg <[email protected]> Reviewed-by: Marek Olšák <[email protected]>
* ilo: set VME for 3DSTATE_PSChia-I Wu2015-10-181-1/+6
| | | | | When the bit is not set, we can see sampling artifacts on triangle edges when the mip filter is not GEN6_MIPFILTER_NONE.
* ilo: ignore prefer_linear_threshold when zeroChia-I Wu2015-10-182-3/+3
| | | | This was the intended behavior but it did not work as intended until now.
* ilo: remove some unused kernel paramsChia-I Wu2015-10-182-22/+0
|
* ilo: remove unused ilo_shader_get_type()Chia-I Wu2015-10-182-12/+0
|
* ilo: remove u_debug.h inclusion from ilo_core.hChia-I Wu2015-10-182-1/+2
| | | | Move it to ilo_debug.h.
* ilo: remove u_memory.h inclusion from ilo_core.hChia-I Wu2015-10-183-1/+3
| | | | We do not make allocations generally in the core.
* nvc0: do not bind input params at compute state init on FermiSamuel Pitoiset2015-10-181-8/+0
| | | | | | | | | | | | | | | | | | It looks like binding a constant buffer on compute overwrites the 3D state. To avoid that, we already re-bind all the 3D constant buffers after launching a compute grid but this is not enough. Binding the constant buffer of input parameters for the compute state at initialization corrupts the 3D constant buffers, and it's just useless to bind it because this is not needed until we really launch a grid. This fixes some piglit regressions related to interpolation tests introduced in "nvc0: enable compute support by default on Fermi". Fixes: 00d6186 (nvc0: enable compute support by default on Fermi) Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Ilia Mirkin <[email protected]>
* i965/vs: Drop hack that created NIR for fixed function vertex programs.Kenneth Graunke2015-10-171-12/+0
| | | | | | | | Marek made core Mesa call ProgramStringNotify(), which solves this properly. The hack is no longer needed. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Matt Turner <[email protected]>
* i965/nir: Switch on shader stage in nir_lower_outputs().Kenneth Graunke2015-10-171-5/+21
| | | | | | | | | | | VS, GS, and FS continue doing the same thing they did before. We can simplify the FS code a bit because it is always scalar. Compute shaders now assert that there are no outputs instead of doing a loop over 0 outputs. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Jason Ekstrand <[email protected]>
* radeonsi: don't use the AMDGPU intrinsic for CMPMarek Olšák2015-10-171-9/+22
| | | | | | | No difference according to shader-db. Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* radeonsi: use LRP from gallivmMarek Olšák2015-10-171-2/+0
| | | | | | | | | | | | | | | | | | Totals: SGPRS: 344552 -> 344368 (-0.05 %) VGPRS: 197132 -> 197552 (0.21 %) Code Size: 7375376 -> 7366304 (-0.12 %) bytes LDS: 91 -> 91 (0.00 %) blocks Scratch: 1679360 -> 1615872 (-3.78 %) bytes per wave Totals from affected shaders: SGPRS: 47736 -> 47552 (-0.39 %) VGPRS: 27952 -> 28372 (1.50 %) Code Size: 1392724 -> 1383652 (-0.65 %) bytes LDS: 39 -> 39 (0.00 %) blocks Scratch: 513024 -> 449536 (-12.38 %) bytes per wave Reviewed-by: Michel Dänzer <[email protected]>
* radeonsi: don't emit AMDGPU intrinsics for integer abs, min, maxMarek Olšák2015-10-171-10/+50
| | | | | | | No difference according to shader-db. (with the new S_ABS_I32 pattern) Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* radeonsi: don't emit AMDGPU intrinsics for EX2, ROUND, TRUNCMarek Olšák2015-10-171-3/+3
| | | | | | | No difference according to shader-db. Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* radeonsi: initialize output, temp, and address registers to "undef"Marek Olšák2015-10-171-4/+15
| | | | | | | | | | | | | | | | | | | | | This removes "v_mov v0, 0" which typically occurs before exports. Totals: SGPRS: 345216 -> 344552 (-0.19 %) VGPRS: 197684 -> 197132 (-0.28 %) Code Size: 7390408 -> 7375376 (-0.20 %) bytes LDS: 91 -> 91 (0.00 %) blocks Scratch: 1842176 -> 1679360 (-8.84 %) bytes per wave Totals from affected shaders: SGPRS: 101336 -> 100672 (-0.66 %) VGPRS: 53920 -> 53368 (-1.02 %) Code Size: 2170176 -> 2155144 (-0.69 %) bytes LDS: 2 -> 2 (0.00 %) blocks Scratch: 1015808 -> 852992 (-16.03 %) bytes per wave Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* gallivm: implement the correct version of LRPMarek Olšák2015-10-171-6/+13
| | | | | | | | | The previous version has precision issues. This can be a problem with tessellation. Sadly, I can't find the article where I read it anymore. I'm not sure if the unsafe-fp-math flag would be enough to revert this. v2: added the comment
* gallivm: set correct opcode info from unary/binary/ternary emitsMarek Olšák2015-10-171-3/+6
| | | | | | | | and clear the emit_data structure. The new radeonsi min/max opcode implementation requires this. (it looks good according to Roland S.)
* radeonsi: implement vertex color clampingMarek Olšák2015-10-175-4/+52
| | | | | | This is only supported in the compatibility profile (without GS and tess). Reviewed-by: Michel Dänzer <[email protected]>
* radeonsi: implement fragment color clampingMarek Olšák2015-10-176-2/+18
| | | | | | using the shader key for now. Reviewed-by: Michel Dänzer <[email protected]>
* radeonsi: clean up other scratch buffer functionsMarek Olšák2015-10-171-15/+8
| | | | Reviewed-by: Michel Dänzer <[email protected]>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-27 01:36:53 +0000