summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* draw/so: maintain an exact number of written verticesZack Rusin2013-04-033-7/+33
| | | | | | | | | It's quite helpful during the rendering when we know exactly the count of the vertices available in the buffer. Signed-off-by: Zack Rusin <[email protected]> Reviewed-by: José Fonseca <[email protected]>
* draw: Implement support for primitive idZack Rusin2013-04-038-8/+33
| | | | | | | We were largely ignoring primitive id. Signed-off-by: Zack Rusin <[email protected]> Reviewed-by: José Fonseca <[email protected]>
* draw/so: Fix bogus assertZack Rusin2013-04-031-1/+0
| | | | | | | We do support so with multiple primitives. Signed-off-by: Zack Rusin <[email protected]> Reviewed-by: José Fonseca <[email protected]>
* draw/gs: Fix memory corruption with multiple primitivesZack Rusin2013-04-031-10/+15
| | | | | | | | | | We were flushing with incorrect number of primitives. TGSI exec can only work with a single primitive at a time. Plus the fetching with multiple primitives on llvm paths wasn't copying the last element. Signed-off-by: Zack Rusin <[email protected]> Reviewed-by: José Fonseca <[email protected]>
* gallivm: cleanup the gs interfaceZack Rusin2013-04-033-50/+85
| | | | | | | | Instead of void pointers use a base interface. Signed-off-by: Zack Rusin <[email protected]> Reviewed-by: Brian Paul <[email protected]> Reviewed-by: José Fonseca <[email protected]>
* svga: add new memory-used HUD queryBrian Paul2013-04-038-1/+33
| | | | | | | To track the amount of memory used by all pipe_resources (textures and buffers). Reviewed-by: Jose Fonseca <[email protected]>
* util: add new util_resource_size() function in u_resource.[ch]Brian Paul2013-04-032-1/+98
| | | | Reviewed-by: Jose Fonseca <[email protected]>
* util: move functions from u_resource.c to u_transfer.cBrian Paul2013-04-032-75/+74
| | | | | | | | | The functions are prototyped in u_transfer.h and are related to the other functions in u_transfer.c. The next patch will re-use the u_resource.c file for new code. Reviewed-by: Jose Fonseca <[email protected]>
* r600g/llvm: Do not override llvm provided stack_sizeVincent Lejeune2013-04-031-1/+2
|
* r600g/llvm: Do not change cf_alu inst when adding alusVincent Lejeune2013-04-031-7/+2
|
* radeonsi: add more cases for copying unsupported formats to resource_copy_regionMarek Olšák2013-04-031-0/+12
| | | | | | | | | | | Ported from r600g commit: 8891b2f9c91b2f6c8625184c23a10b8e55875dc0 Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Christian König <[email protected]> NOTE: This is a candidate for the 9.1 branch.
* svga: add HUD queries for number of draw calls, number of fallbacksBrian Paul2013-04-034-0/+61
| | | | | | | The fallbacks count is the number of drawing calls that use a "draw" module fallback, such as polygon stipple. Reviewed-by: Jose Fonseca <[email protected]>
* svga: refactor occlusion query codeBrian Paul2013-04-031-94/+124
| | | | | | This is in preparation for adding new query types for the HUD. Reviewed-by: Jose Fonseca <[email protected]>
* gallium/hud: try L8 texture for font if I8 format isn't supportedBrian Paul2013-04-031-4/+13
|
* svga: add case for PIPE_CAP_QUERY_PIPELINE_STATISTICSBrian Paul2013-04-031-0/+1
|
* st/mesa: rewrite comment in st_manager.cBrian Paul2013-04-031-3/+2
|
* nv50,nvc0: remove MS resolve formats hackChristoph Bumiller2013-04-032-15/+0
| | | | Mesa now allows BlitFramebuffer resolve between RGBA and BGRA.
* nvc0: fix 128 bit compressed storage type selectionChristoph Bumiller2013-04-031-1/+1
|
* nvc0: place staging textures in GART and map them directlyChristoph Bumiller2013-04-037-11/+76
|
* nv50: account for pesky prefetch in size calculation of linear texturesChristoph Bumiller2013-04-031-1/+6
|
* nvc0: honour scaled coordiantes setting for linear texturesChristoph Bumiller2013-04-031-6/+5
|
* nvc0: fix for 2d engine R source formats writing RRR1 and not R001Christoph Bumiller2013-04-033-52/+148
|
* nv50,nvc0: disable DEPTH_RANGE_NEAR/FAR clipping during blitChristoph Bumiller2013-04-033-1/+5
| | | | | We send position.z == 0, DEPTH_RANGE may be some arbitrary range not including 0 (for exmaple in piglit's hiz tests).
* st/mesa: fix bitmap,drawpix,drawtex for PIPE_CAP_TGSI_TEXCOORDChristoph Bumiller2013-04-033-2/+8
| | | | | | | NOTE: Changed the semantic index for the drawtex coordinate to be the texture unit index instead of always 0. Not sure if this is correct but since the value seems to depend on the unit it would make sense to use different varying slots.
* nouveau: accelerate buffer copies in resource_copy_regionChristoph Bumiller2013-04-035-11/+47
|
* nvc0: demagic some of the NVE4_COMPUTE_UPLOAD methodsChristoph Bumiller2013-04-033-49/+129
| | | | It's actually the same as P2MF.
* nvc0: read PM counters for each warp scheduler separatelyChristoph Bumiller2013-04-032-61/+138
|
* nvc0: add some metrics to driver specific queriesChristoph Bumiller2013-04-032-58/+160
|
* nvc0: add some driver statistics queriesChristoph Bumiller2013-04-0314-25/+279
|
* nvc0: disable compressed storage type 0xdb for nowChristoph Bumiller2013-04-031-1/+3
| | | | Single-sample color compression doesn't seem that useful anyway.
* nvc0: use correct hw query for PRIMITIVES_GENERATEDChristoph Bumiller2013-04-031-4/+7
| | | | It was the same as SO_STATISTICS[1] before.
* nvc0: use fence to check state of queries that don't write sequenceChristoph Bumiller2013-04-031-1/+5
| | | | | | This still isn't optimal, since the fence will signal a bit late, but better than checking on the bo, which may never be ready if it is shared (which is likely).
* gallium/hud: add support for PIPE_QUERY_PIPELINE_STATISTICSChristoph Bumiller2013-04-034-9/+52
| | | | | | | Also, renamed "pixels-rendered" to "samples-passed" because the occlusion counter increments even if colour and depth writes are disabled, or (on some implementations) for killed fragments that passed the depth test when PS early_fragment_tests is set.
* gallium/docs: fix definition of PIPE_QUERY_SO_STATISTICSChristoph Bumiller2013-04-031-3/+5
| | | | Reviewed-by: Marek Olšák <[email protected]>
* gallium: add PIPE_CAP_QUERY_PIPELINE_STATISTICSChristoph Bumiller2013-04-0312-1/+17
| | | | Reviewed-by: Marek Olšák <[email protected]>
* i965: Reduce code duplication in handling of depth, stencil, and HiZ.Paul Berry2013-04-025-151/+178
| | | | | | | | | | | | | | | | | | | | | | | | | This patch consolidates duplicate code in the brw_depthbuffer and gen7_depthbuffer state atoms. Previously, these state atoms contained 5 chunks of code for emitting the _3DSTATE_DEPTH_BUFFER packet (3 for Gen4-6 and 2 for Gen7). Also a lot of logic for determining the appropriate buffer setup was duplicated between the Gen4-6 and Gen7 functions. This refactor splits the code into three separate functions: brw_emit_depthbuffer(), which determines the appropriate buffer setup in a mostly generation-independent way, brw_emit_depth_stencil_hiz(), which emits the appropriate state packets for Gen4-6, and gen7_emit_depth_stencil_hiz(), which emits the appropriate state packets for Gen7. Tested using Piglit on Gen5-7 (no regressions). v2: Re-word some comments. Fix an assertion that incorrectly prohibited packed depth/stencil formats on Gen6 (these are allowed provided that HiZ is disabled). Reviewed-by: Chad Versace <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* Revert "glsl: Replace constant-index vector array accesses with swizzles"Paul Berry2013-04-022-83/+83
| | | | | | | This reverts commit dbf94d105a48b7aafb2c8cf64d8b4392d87efea1, which was working around a bug in the handling of array indexing when constant folding built-in functions. Now that the constant folding bug has been fixed, the workaround is no longer needed.
* glsl: Fix array indexing when constant folding built-in functions.Paul Berry2013-04-021-1/+1
| | | | | | | | | | | | | | | | | | | Mesa constant-folds built-in functions by using a miniature GLSL interpreter (see ir_function_signature::constant_expression_evaluate_expression_list()). This interpreter had a bug in its handling of array indexing, which caused expressions like "m[i][j]" (where m is a matrix) to be handled incorrectly. Specifically, it incorrectly treated j as indexing into the whole matrix (rather than indexing just into the vector m[i]); as a result the offset computed for m[i] was lost and m[i][j] was treated as m[j][0]. Fixes piglit tests inverse-mat[234].{vert,frag}. NOTE: This is a candidate for the 9.1 and 9.0 branches. Reviewed-by: Ian Romanick <[email protected]> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=57436
* gallivm: bring back optimized but incorrect float to smallfloat optimizationsRoland Scheidegger2013-04-021-38/+78
| | | | | | | | | | | | | Conceptually the same as previously done in float_to_half. Should cut down number of instructions from 14 to 10 or so, but will promote some NaNs to Infs, so it's disabled. It gets a bit tricky though handling all the cases correctly... Passes basic tests either way (though there are no tests testing special cases, but some manual tests injecting them seemed promising). v2: style and comment fixes suggested by Jose Reviewed-by: Jose Fonseca <[email protected]>
* gallivm: consolidate code for float-to-half and float-to-packed conversion.Roland Scheidegger2013-04-023-108/+102
| | | | | | | | | | | | | | | | | | This replaces the existing float-to-half implementation. There are definitely a couple of differences - the old implementation had unspecified(?) rounding behavior, and could at least in theory construct Inf values out of NaNs. NaNs and Infs should now always be properly propagated, and rounding behavior is now towards zero (note this means too large but non-Infinity values get propagated to max representable value, not Infinity). The implementation will definitely not match util code, however (which does nearest rounding, which also means too large values will get propagated to Infinity). Also fix a bogus round mask probably leading to rounding bugs... v2: fix a logic bug in handling infs/nans. Reviewed-by: Jose Fonseca <[email protected]>
* r600g: don't reserve more stack space than required v5Vadim Girlin2013-04-023-56/+142
| | | | | | | | | | | Reduced stack size allows to run more threads in some cases, improving performance for the shaders that use stack (that is, for the shaders with control flow instructions). E.g. with unigine-based apps. v4: implement exact computation taking into account wavefront size v5: add cases for RV620, RS880 Signed-off-by: Vadim Girlin <[email protected]>
* r600g: fix range handling for tgsi input declarations v2Vadim Girlin2013-04-021-3/+7
| | | | Signed-off-by: Vadim Girlin <[email protected]>
* gallium/hud: do .xxxx swizzling for the font texture in the fragment shaderMarek Olšák2013-04-021-6/+30
| | | | | | This allows using L8 and R8 for the font if I8 isn't supported. Tested-by: Brian Paul <[email protected]>
* hud: flush/unmap the vertex buffer before drawingBrian Paul2013-04-021-0/+3
| | | | | | | The VMware svga driver is picky about making sure the VBO is unmapped before drawing. Reviewed-by: Marek Olšák <[email protected]>
* draw: use pipe_transfer_unmap() to match pipe_transfer_map()Brian Paul2013-04-021-1/+1
|
* gallivm: fix signed small float to float conversionRoland Scheidegger2013-04-021-1/+1
| | | | | | Introduced by 5f41e08cf39d585d600aa506cdcd2f5380c60ddd, just a silly typo. Fixes https://bugs.freedesktop.org/show_bug.cgi?id=62921.
* radeonsi: add instance divisor support v3Christian König2013-04-024-59/+94
| | | | | | | | v2: reduce key size, don't copy key around to much. v3: remove key size reduction Signed-off-by: Christian König <[email protected]> Reviewed-by: Michel Dänzer <[email protected]>
* radeonsi: add start instance supportChristian König2013-04-023-10/+21
| | | | | | | | This works different than on R600, we need to add the start instance manually. Signed-off-by: Christian König <[email protected]> Reviewed-by: Michel Dänzer <[email protected]> Tested-by: Michel Dänzer <[email protected]>
* radeonsi: add instanceid supportChristian König2013-04-024-7/+43
| | | | | | Signed-off-by: Christian König <[email protected]> Reviewed-by: Michel Dänzer <[email protected]> Tested-by: Michel Dänzer <[email protected]>
* radeon/llvm: move system value fetching to common codeChristian König2013-04-022-12/+12
| | | | | | | | This should be used by both SI and R600. Signed-off-by: Christian König <[email protected]> Reviewed-by: Michel Dänzer <[email protected]> Tested-by: Michel Dänzer <[email protected]>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-17 01:48:39 +0000