summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* nir/lower_double_pack: fixup for new nir_foreach_block()Jason Ekstrand2016-05-051-26/+21
| | | | | | Signed-off-by: Jason Ekstrand <[email protected]> Reviewed-by: Connor Abbott <[email protected]> Reviewed-by: Ian Romanick <[email protected]>
* nir/gather_info: fixup for new foreach_block()Jason Ekstrand2016-05-051-5/+5
| | | | | | Signed-off-by: Jason Ekstrand <[email protected]> Reviewed-by: Connor Abbott <[email protected]> Reviewed-by: Ian Romanick <[email protected]>
* nir/lower_two_sided_color: fixup for new foreach_block()Connor Abbott2016-05-051-3/+5
| | | | | Reviewed-by: Jason Ekstrand <[email protected]> Reviewed-by: Ian Romanick <[email protected]>
* nir/lower_tex: fixup for new foreach_block()Connor Abbott2016-05-051-25/+22
| | | | | Reviewed-by: Jason Ekstrand <[email protected]> Reviewed-by: Ian Romanick <[email protected]>
* nir/lower_outputs_to_temporaries: fixup for new foreach_block()Connor Abbott2016-05-051-16/+12
| | | | | Reviewed-by: Jason Ekstrand <[email protected]> Reviewed-by: Ian Romanick <[email protected]>
* i965: Switch to scalar TCS by default.Kenneth Graunke2016-05-051-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Normally, we expect SIMD8 shaders to be more instructions than SIMD4x2 shaders, as it takes four instructions to operate on a vec4, rather than a single instruction. However, the benefit is that it can process 8 objects per shader thread instead of 2. Surprisingly, the shader-db statistics show an improvement in both instruction and cycle counts: Synmark: -31.25% instructions, -29.27% cycles, 0 hurt. Tessmark: -36.92% instructions, -37.81% cycles, 0 hurt. Unigine Heaven: -3.42% instructions, -17.95% cycles, 0 hurt. Shadow of Mordor: +13.24% instructions (26 with fewer instructions, 45 with more), -5.23% cycles (44 with fewer cycles, 27 with more cycles). Presumably, this is because the SIMD8 URB messages are a much more natural fit than the SIMD4x2 URB messages - there's a ton less header setup. I benchmarked Shadow of Mordor and Unigine Heaven on my Skylake GT3e, and the performance seems to be the same or increase ever so slightly (< 1 FPS difference). So I believe it's strictly superior. There's also a lot more optimization potential we can do in scalar mode. This will also help us finish fp64 support, as scalar support is going to land much sooner than vec4-mode support. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Matt Turner <[email protected]>
* nir: Optimize out stores of undefs.Kenneth Graunke2016-05-051-0/+30
| | | | | | | | | | | | | | | | | | | | | | | | | | There are a couple of cycle count changes in shader-db, but it's basically a wash. However, with the Broadwell scalar TCS backend enabled, many Shadow of Mordor shaders benefit from this patch. Because we don't batch up output writes for TCS, vec4 outputs might not have all components defined. Many output writes have a value of undef, which is useless. With scalar TCS, stats for tessellation shaders on Broadwell: total instructions in shared programs: 1283000 -> 1280444 (-0.20%) instructions in affected programs: 34302 -> 31746 (-7.45%) helped: 71 HURT: 0 total cycles in shared programs: 10798768 -> 10780682 (-0.17%) cycles in affected programs: 158004 -> 139918 (-11.45%) helped: 71 HURT: 0 Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Matt Turner <[email protected]>
* nir: Replace vecN(undef, undef, ...) with a single undef.Kenneth Graunke2016-05-051-0/+33
| | | | | | | | | | | | | | | | | | | | | | | shader-db statistics on Broadwell: total instructions in shared programs: 8963409 -> 8962455 (-0.01%) instructions in affected programs: 60858 -> 59904 (-1.57%) helped: 318 HURT: 0 total cycles in shared programs: 71408022 -> 71406276 (-0.00%) cycles in affected programs: 398416 -> 396670 (-0.44%) helped: 199 HURT: 51 GAINED: 1 The only shaders affected were in Dota 2 Reborn. It also sets up for the next optimization. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Matt Turner <[email protected]>
* nir: Rename opt_undef_alu to opt_undef_csel; update comments.Kenneth Graunke2016-05-051-12/+13
| | | | | | | | | | | | This better reflects what it does. I plan to add other ALU optimizations as well, so the old name would be confusing. In preparation for that, also move the file comments about csels above the opt_undef_csel function, and delete the ones about there not being other optimizations. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Matt Turner <[email protected]>
* i965: Rework passthrough TCS checks.Kenneth Graunke2016-05-054-2/+5
| | | | | | | | | | According to Timothy, using program_string_id == 0 to identify the passthrough TCS is going to be problematic for his shader cache work. So, change it to strcmp() the name at visitor creation time. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Matt Turner <[email protected]>
* swr: [rasterizer core] Faster modulo operator in ProcessVertsTim Rowley2016-05-051-1/+4
| | | | | | Avoid % operator, since we know that curVertex is always incrementing. Reviewed-by: Bruce Cherniak <[email protected]>
* swr: [rasterizer] Small warning cleanupTim Rowley2016-05-052-8/+4
| | | | Reviewed-by: Bruce Cherniak <[email protected]>
* swr: [rasterizer] Add SWR_ASSUME / SWR_ASSUME_ASSERT macrosTim Rowley2016-05-052-14/+52
| | | | | | Fix static code analysis errors found by coverity on Linux Reviewed-by: Bruce Cherniak <[email protected]>
* swr: [rasterizer] Miscellaneous backend changesTim Rowley2016-05-053-22/+31
| | | | Reviewed-by: Bruce Cherniak <[email protected]>
* swr: [rasterizer] Add support for X24_TYPELESS_G8_UINT formatTim Rowley2016-05-053-7/+41
| | | | Reviewed-by: Bruce Cherniak <[email protected]>
* swr: [rasterizer jitter] Fix printing bugs for tracing.Tim Rowley2016-05-051-81/+24
| | | | Reviewed-by: Bruce Cherniak <[email protected]>
* swr: [rasterizer memory] Add missing store tiles functionTim Rowley2016-05-051-1/+4
| | | | | | Storing color hot tile to 8bit w-major stencil format. Reviewed-by: Bruce Cherniak <[email protected]>
* swr: [rasterizer jitter] Add asserts for supported formats in fetch shaderTim Rowley2016-05-051-0/+2
| | | | Reviewed-by: Bruce Cherniak <[email protected]>
* swr: [rasterizer core] Fix thread allocationTim Rowley2016-05-051-17/+47
| | | | | | | | Fix windows in 32-bit mode when hyperthreading is disabled on Xeons. Some support for asymmetric processor topologies. Reviewed-by: Bruce Cherniak <[email protected]>
* swr: [rasterizer core] Fix threadviz support in bucketsTim Rowley2016-05-053-12/+14
| | | | | | | Need to do lazy eval of the threadviz knob since order of globals is undefined. Reviewed-by: Bruce Cherniak <[email protected]>
* swr: [rasterizer] Whitespace cleanup and misc changesTim Rowley2016-05-055-5/+2
| | | | Reviewed-by: Bruce Cherniak <[email protected]>
* radeonsi: mark descriptor loads as using dynamically uniform indicesNicolai Hähnle2016-05-051-5/+17
| | | | | | | | This tells LLVM to always use SMEM loads for descriptors. It fixes a regression in piglit's arb_shader_storage_buffer_object/execution/indirect.shader_test that was caused by LLVM r268259 (but the proper fix is really here in Mesa). Reviewed-by: Marek Olšák <[email protected]>
* i965/fs: Don't follow pow with an instruction with two dest regs.Matt Turner2016-05-051-0/+18
| | | | | | | | | | | | | | | | | Beginning with commit 7b208a73, Unigine Valley began hanging the GPU on Gen >= 8 platforms. Evidently that commit allowed the scheduler to make different choices that somehow finally ran afoul of a hardware bug in which POW and FDIV instructions may not be followed by an instruction with two destination registers (including compressed instructions). I presume the conditions are more complex than that, but the internal hardware bug report (BDWGFX bug_de 1696294) does not contain much more information. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94924 Reviewed-by: Topi Pohjolainen <[email protected]> [v1] Tested-by: Mark Janes <[email protected]> [v1] Reviewed-by: Francisco Jerez <[email protected]>
* swr: Remove stall waiting for core query counters.Bruce Cherniak2016-05-054-124/+81
| | | | | | | | When gathering query results, swr_gather_stats was unnecessarily stalling the entire pipeline. Results are now collected asynchronously, with a fence marking completion. Reviewed-By: George Kyriazis <[email protected]>
* mesa/ubo: add missing compute cases for ubo/atomic buffersDave Airlie2016-05-051-0/+6
| | | | | | | This fixes: GL43-CTS.compute_shader.resource-ubo Reviewed-by: Anuj Phogat <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* mesa/compute: drop pointless casts.Dave Airlie2016-05-051-3/+3
| | | | | | | | We already are a GLintptr, casting won't help. Reviewed-by: Alejandro Piñeiro <[email protected]> Reviewed-by: Anuj Phogat <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* mesa: remove null check before freeThomas Hindoe Paaboel Andersen2016-05-052-4/+2
| | | | Reviewed-by: Eduardo Lima Mitev <[email protected]>
* freedreno: remove null check before freeThomas Hindoe Paaboel Andersen2016-05-051-2/+1
| | | | Reviewed-by: Eduardo Lima Mitev <[email protected]>
* nir: fix assert for wildcard pairsThomas Hindoe Paaboel Andersen2016-05-051-1/+1
| | | | | | | | | The assert was null checking dest_arr_parent twice. The intention seems to be to check both dest_ and src_. Added in d3636da9 Reviewed-by: Eduardo Lima Mitev <[email protected]>
* glapi: fix parameter type for GetSamplerParameterIuivEXT() in es_EXT.xmlBrian Paul2016-05-041-2/+2
| | | | | | | | | | The function returns GLuint, not GLfloat values. v2: also fix the OES function Cc: "11.2" <[email protected]> Reviewed-by: Charmaine Lee <[email protected]> Reviewed-by: Ilia Mirkin <[email protected]>
* mesa: include texture format in glGenerateMipmap error messageBrian Paul2016-05-041-1/+2
| | | | Reviewed-by: Charmaine Lee <[email protected]>
* main: uses casts to silence some _mesa_debug() format warningsBrian Paul2016-05-041-4/+6
| | | | | | | Silences warnings with 32-bit Linux gcc builds and MinGW which doesn't recognize the ‘t’ conversion character. Reviewed-by: Sinclair Yeh <[email protected]>
* docs: Mark GL_ARB_query_buffer_object as done for i965/hsw+Jordan Justen2016-05-042-1/+2
| | | | Signed-off-by: Jordan Justen <[email protected]>
* i965: Implement ARB_query_buffer_object for HSW+Jordan Justen2016-05-048-3/+501
| | | | | | | | | | | | | | | v2: * Declare loop index variable at loop site (idr) * Make arrays of MI_MATH instructions 'static const' (idr) * Remove commented debug code (idr) * Updated comment in set_query_availability (Ken) * Replace switch with if/else in hsw_result_to_gpr0 (Ken) * Only divide GL_FRAGMENT_SHADER_INVOCATIONS_ARB by 4 on hsw and gen8 (Ken) Signed-off-by: Jordan Justen <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]> Reviewed-by: Ian Romanick <[email protected]>
* i965/gen6+: Add load register immediate helper functionsJordan Justen2016-05-042-0/+36
| | | | | Signed-off-by: Jordan Justen <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* i965/hsw+: Add support for copying a registerJordan Justen2016-05-043-0/+18
| | | | | Signed-off-by: Jordan Justen <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* i965/gen6+: Add support for storing immediate data into a bufferJordan Justen2016-05-043-0/+50
| | | | | Signed-off-by: Jordan Justen <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* i965: Add MI_MATH reg defs for HSW+Jordan Justen2016-05-041-0/+38
| | | | | Signed-off-by: Jordan Justen <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* i965: Add brw_store_register_mem32Jordan Justen2016-05-042-0/+28
| | | | | Signed-off-by: Jordan Justen <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* i965: Use offset instead of index in brw_store_register_mem64Jordan Justen2016-05-045-54/+52
| | | | | | | | | | This matches the byte based offset of brw_load_register_mem*. The function is also moved into intel_batchbuffer.c like brw_load_register_mem*. Signed-off-by: Jordan Justen <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* r600,compute: create vtx buffer for text + rodataJan Vesely2016-05-041-2/+10
| | | | | | | Reserve buffer id 2 Signed-off-by: Jan Vesely <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* freedreno: allow ctx->draw_vbo to failRob Clark2016-05-045-30/+37
| | | | | | | Pretty much only happens if shader variant compile fails. But in this case, if we haven't emitted cmdstream, we don't want to set needs_flush. Signed-off-by: Rob Clark <[email protected]>
* freedreno: move shader-stage dirty bits to global dirty flagRob Clark2016-05-048-59/+41
| | | | | | | | | | | This was always a bit overly complicated, and had some issues (like ctx->prog.dirty not getting reset at the end of the batch). It also required some special hacks to avoid resetting dirty state on binning pass. So just move it all into ctx->dirty (leaving some free bits for future shader stages), and make FD_DIRTY_PROG just be the union of all FD_SHADER_DIRTY_*. Signed-off-by: Rob Clark <[email protected]>
* freedreno/a4xx: fix bogus offset for f32x24s8 stencil restoreRob Clark2016-05-041-4/+5
| | | | | | fixes: $piglit/bin/fbo-clear-formats GL_ARB_depth_buffer_float Signed-off-by: Rob Clark <[email protected]>
* freedreno: add some debug_asserts() to catch insane offsetsRob Clark2016-05-041-0/+2
| | | | | | | Ofc won't catch *all* faults, but at least helpful for catching offsets which are completely bogus. Signed-off-by: Rob Clark <[email protected]>
* freedreno/a4xx: deal with VS which do not write positionRob Clark2016-05-041-0/+7
| | | | | | | | Fixes $piglit/bin/glsl-1.40-tf-no-position a3xx may need similar? Signed-off-by: Rob Clark <[email protected]>
* freedreno/ir3: remove a couple redundant is_flow()sRob Clark2016-05-042-2/+2
| | | | | | | Now that the opc's encode the instruction category (making them unique) we no longer need to check the category in addition to the opc. Signed-off-by: Rob Clark <[email protected]>
* freedreno/ir3: cp small negative integers tooRob Clark2016-05-041-1/+2
| | | | Signed-off-by: Rob Clark <[email protected]>
* freedreno/ir3: fix # of registersRob Clark2016-05-041-1/+1
| | | | | | | The instruction encoding allows for more registers, but at least on a3xx/a4xx they don't actually exist. Signed-off-by: Rob Clark <[email protected]>
* freedreno/ir3: lower immeds to constRob Clark2016-05-043-4/+80
| | | | | | | | | | | | | | | | | Helps reduce register pressure and instruction counts for immediates that would otherwise require a mov into gpr. total instructions in shared programs: 4455332 -> 4369297 (-1.93%) total dwords in shared programs: 8807872 -> 8614432 (-2.20%) total full registers used in shared programs: 263062 -> 250846 (-4.64%) total half registers used in shader programs: 9845 -> 9845 (0.00%) total const registers used in shared programs: 1029735 -> 1466993 (42.46%) half full const instr dwords helped 0 10415 0 17861 5912 hurt 0 1157 21458 947 33 Signed-off-by: Rob Clark <[email protected]>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-26 23:46:21 +0000