summaryrefslogtreecommitdiffstats
path: root/src/gallium
Commit message (Collapse)AuthorAgeFilesLines
* radeonsi: Fix handling of geometry shader output vertex IDMichel Dänzer2014-01-291-7/+19
| | | | | | | | | It needs to increment at shader runtime, not at shader compile time, as the geometry shader can emit vertices in loops. LLVM automagically converts the ID back to an immediate value if its value can be determined at compile time. Reviewed-by: Marek Olšák <[email protected]>
* radeonsi: Initial geometry shader supportMichel Dänzer2014-01-2911-191/+1084
| | | | | | | Partly based on the corresponding r600g work by Vadim Girlin and Dave Airlie. Reviewed-by: Marek Olšák <[email protected]>
* radeonsi: Refactor shader input / output handling codeMichel Dänzer2014-01-291-218/+268
| | | | | | In preparation for adding geometry shader support. Reviewed-by: Marek Olšák <[email protected]>
* r600g,radeonsi: Don't set resource_create in r600_common_screen_init()Tom Stellard2014-01-282-1/+1
| | | | | | | | r600g and radeonsi have different implementations of resource_create. https://bugs.freedesktop.org/show_bug.cgi?id=74139 Reviewed-by: Marek Olšák <[email protected]>
* gallivm: Workaround http://llvm.org/PR18600José Fonseca2014-01-281-2/+4
| | | | | | | | | | | | | | | | | | We have code generation paths that carry out swizzles of AoS vectors via bitwise shifts, as these tend to generate more efficient code than straightforward byte shuffles. But when the input is a constant the additional bitwise arithmetic operations somehow don't really get constant propagated properly, evenutally causing assertion failure in InstCombine pass. Therefore avoid the bug by using the trivial shuffles for constant inputs. Although the sample LLVM IR can cause a crash with any LLVM version, this was only seen in practice with LLVM 3.2. Reviewed-by: Matthew McClure <[email protected]> Reviewed-by: Roland Scheidegger <[email protected]>
* r600g: s/r600_llvm_gpu_string/r600_get_llvm_processor_name/Michel Dänzer2014-01-281-1/+1
| | | | | | Fixes build failure introduced by commit 65dc588bfd3b8145131340ffe77f216be58378ac ('r600g,radeonsi: consolidate get_compute_param'), which consolidated the former into the latter.
* radeonsi: cleanup includes, add missing licenseMarek Olšák2014-01-2812-86/+49
| | | | | Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* radeonsi: remove open-coded PS_PARTIAL_FLUSH eventMarek Olšák2014-01-281-5/+3
| | | | | Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* radeonsi: move some inline functions from si_pipe.h to si_state.cMarek Olšák2014-01-282-39/+30
| | | | | | | | | And si_tex_aniso_filter is unused. v2: remove INLINE occurences Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* radeonsi: remove si_resource.hMarek Olšák2014-01-284-50/+13
| | | | | Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* radeonsi: remove si.hMarek Olšák2014-01-283-45/+5
| | | | | Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* radeonsi: move si_upload_const_buffer to a better placeMarek Olšák2014-01-285-70/+29
| | | | | | | This gets rid of another file. Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* radeonsi: inline si_translate_index_bufferMarek Olšák2014-01-284-60/+19
| | | | | Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* radeonsi: inline si_upload_index_bufferMarek Olšák2014-01-283-13/+2
| | | | | Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* r600g,radeonsi: consolidate remaining obviously duplicated pipe_screen codeMarek Olšák2014-01-284-57/+30
| | | | | Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* r600g,radeonsi: consolidate get_compute_paramMarek Olšák2014-01-287-267/+162
| | | | | | | v2: added fprintf to r600_get_llvm_processor_name Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* r600g,radeonsi: consolidate get_paramf and get_video_paramMarek Olšák2014-01-284-118/+70
| | | | | | | | | | radeonsi now reports PIPE_VIDEO_CAP_SUPPORTS_PROGRESSIVE = true if UVD support isn't available. It's what all the other drivers do. Also, some #include directives were missing in radeon_uvd.h. Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* r600g,radeonsi: consolidate variables for CS tracingMarek Olšák2014-01-2810-46/+42
| | | | | Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* r600g,radeonsi: consolidate get_timestamp, get_driver_query_infoMarek Olšák2014-01-284-41/+34
| | | | | | | This enables more queries for the Gallium HUD with radeonsi. Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* r600g,radeonsi: consolidate get_name and get_vendor queriesMarek Olšák2014-01-283-74/+50
| | | | | Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* radeon: place context-related functions first in r600_pipe_common.cMarek Olšák2014-01-281-79/+87
| | | | | | | To follow the unwritten convention of r600g and radeonsi. Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* r600g,radeonsi: consolidate the contents of r600_resource.cMarek Olšák2014-01-2813-163/+45
| | | | | Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* radeonsi: advertise the pipeline statistics queryMarek Olšák2014-01-281-1/+1
| | | | | | | | | | Implemented by the common code. You can now visualize the statistics with the HUD, see GALLIUM_HUD=help for all available queries. For example: GALLIUM_HUD=clipper-primitives-generated Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* radeonsi: use queries from r600gMarek Olšák2014-01-289-757/+40
| | | | | Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* r600g: remove a no-op while loopMarek Olšák2014-01-281-1/+1
| | | | | | | | | | | for (;;) { } while (); I was surprised to see such a statement. Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* r600g: convert query emission code to radeon_emitMarek Olšák2014-01-281-50/+50
| | | | | Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* r600g: only emit NOP relocations for queries if VM is disabledMarek Olšák2014-01-283-23/+19
| | | | | Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* r600g: move queries to drivers/radeonMarek Olšák2014-01-2812-239/+282
| | | | | Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* mesa: Change many Type P MESA_FORMATs to meet naming specMark Mueller2014-01-271-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Conversion of Type P formats as follows (w/related comment fixes): s/MESA_FORMAT_RGB565\b/MESA_FORMAT_B5G6R5_UNORM/g s/MESA_FORMAT_RGB565_REV\b/MESA_FORMAT_R5G6B5_UNORM/g s/MESA_FORMAT_ARGB4444\b/MESA_FORMAT_B4G4R4A4_UNORM/g s/MESA_FORMAT_ARGB4444_REV\b/MESA_FORMAT_A4R4G4B4_UNORM/g s/MESA_FORMAT_RGBA5551\b/MESA_FORMAT_A1B5G5R5_UNORM/g s/MESA_FORMAT_XBGR8888_SNORM\b/MESA_FORMAT_R8G8B8X8_SNORM/g s/MESA_FORMAT_XBGR8888_SRGB\b/MESA_FORMAT_R8G8B8X8_SRGB/g s/MESA_FORMAT_ARGB1555\b/MESA_FORMAT_B5G5R5A1_UNORM/g s/MESA_FORMAT_ARGB1555_REV\b/MESA_FORMAT_A1R5G5B5_UNORM/g s/MESA_FORMAT_AL44\b/MESA_FORMAT_L4A4_UNORM/g s/MESA_FORMAT_RGB332\b/MESA_FORMAT_B2G3R3_UNORM/g s/MESA_FORMAT_ARGB2101010\b/MESA_FORMAT_B10G10R10A2_UNORM/g s/MESA_FORMAT_Z24_S8\b/MESA_FORMAT_S8_UINT_Z24_UNORM/g s/MESA_FORMAT_S8_Z24\b/MESA_FORMAT_Z24_UNORM_S8_UINT/g s/MESA_FORMAT_X8_Z24\b/MESA_FORMAT_Z24_UNORM_X8_UINT/g s/MESA_FORMAT_Z24_X8\b/MESA_FORMAT_X8Z24_UNORM/g s/MESA_FORMAT_RGB9_E5_FLOAT\b/MESA_FORMAT_R9G9B9E5_FLOAT/g s/MESA_FORMAT_R11_G11_B10_FLOAT\b/MESA_FORMAT_R11G11B10_FLOAT/g s/MESA_FORMAT_Z32_FLOAT_X24S8\b/MESA_FORMAT_Z32_FLOAT_S8X24_UINT/g s/MESA_FORMAT_ABGR2101010_UINT\b/MESA_FORMAT_R10G10B10A2_UINT/g s/MESA_FORMAT_XRGB4444_UNORM\b/MESA_FORMAT_B4G4R4X4_UNORM/g s/MESA_FORMAT_XRGB1555_UNORM\b/MESA_FORMAT_B5G5R5X1_UNORM/g s/MESA_FORMAT_XRGB2101010_UNORM\b/MESA_FORMAT_B10G10R10X2_UNORM/g s/MESA_FORMAT_AL88\b/MESA_FORMAT_L8A8_UNORM/g s/MESA_FORMAT_AL88_REV\b/MESA_FORMAT_A8L8_UNORM/g s/MESA_FORMAT_AL1616\b/MESA_FORMAT_L16A16_UNORM/g s/MESA_FORMAT_AL1616_REV\b/MESA_FORMAT_A16L16_UNORM/g s/MESA_FORMAT_RG88\b/MESA_FORMAT_G8R8_UNORM/g s/MESA_FORMAT_GR88\b/MESA_FORMAT_R8G8_UNORM/g s/MESA_FORMAT_GR1616\b/MESA_FORMAT_R16G16_UNORM/g s/MESA_FORMAT_RG1616\b/MESA_FORMAT_G16R16_UNORM/g s/MESA_FORMAT_SRGBA8\b/MESA_FORMAT_A8B8G8R8_SRGB/g s/MESA_FORMAT_SARGB8\b/MESA_FORMAT_B8G8R8A8_SRGB/g s/MESA_FORMAT_SLA8\b/MESA_FORMAT_L8A8_SRGB/g Conflicts: src/mesa/drivers/dri/i965/brw_surface_formats.c src/mesa/main/format_pack.c src/mesa/main/format_unpack.c src/mesa/main/formats.c src/mesa/main/texformat.c src/mesa/main/texstore.c
* mesa: Rename 4 color component unsigned byte MESA_FORMATsMark Mueller2014-01-271-2/+2
| | | | | | | | | | | | | Change all 4 color component unsigned byte formats to meet spec for P Type formats: s/MESA_FORMAT_RGBA8888\b/MESA_FORMAT_A8B8G8R8_UNORM/g s/MESA_FORMAT_RGBA8888_REV\b/MESA_FORMAT_R8G8B8A8_UNORM/g s/MESA_FORMAT_ARGB8888\b/MESA_FORMAT_B8G8R8A8_UNORM/g s/MESA_FORMAT_ARGB8888_REV\b/MESA_FORMAT_A8R8G8B8_UNORM/g s/MESA_FORMAT_RGBX8888\b/MESA_FORMAT_X8B8G8R8_UNORM/g s/MESA_FORMAT_RGBX8888_REV\b/MESA_FORMAT_R8G8B8X8_UNORM/g s/MESA_FORMAT_XRGB8888\b/MESA_FORMAT_B8G8R8X8_UNORM/g s/MESA_FORMAT_XRGB8888_REV\b/MESA_FORMAT_X8R8G8B8_UNORM/g
* mesa: change gl_format to mesa_formatMark Mueller2014-01-271-2/+2
| | | | s/\bgl_format\b/mesa_format/g. Use better name for Mesa Formats enum
* r600g/compute: Emit DEALLOC_STATE on cayman after dispatching a compute shader.Tom Stellard2014-01-274-14/+10
| | | | | | | | | | | | This is necessary to prevent the next SURFACE_SYNC packet from hanging the GPU. https://bugs.freedesktop.org/show_bug.cgi?id=73418 Reviewed-by: Marek Olšák <[email protected]> Reviewed-by: Alex Deucher <[email protected]> CC: "9.2" "10.0" <[email protected]>
* nv50, nvc0: update reported glsl version to 330Ilia Mirkin2014-01-272-2/+2
| | | | Signed-off-by: Ilia Mirkin <[email protected]>
* nv50: add more RGB10A2 formatsChristoph Bumiller2014-01-271-0/+7
|
* nv50: enable seamless cube maps on all hwIlia Mirkin2014-01-271-1/+1
| | | | | | | | | | Some of the hardware support is missing. The NVIDIA-provided driver, which claims seamless cube map support fails the relevant tests as well. As this is the last extension before we can have OpenGL 3.2, doing this allows us to expose geometry shaders without doing the additional work involved in supporting ARB_geometry_shader4. Signed-off-by: Ilia Mirkin <[email protected]>
* nv50: report glsl 1.50 now that gp tests passIlia Mirkin2014-01-271-1/+1
| | | | Signed-off-by: Ilia Mirkin <[email protected]>
* nv50: add support for texelFetch'ing MS textures, ARB_texture_multisampleIlia Mirkin2014-01-277-7/+164
| | | | | | | | | | | | | | Creates two areas in the AUX constbuf: - Sample offsets for MS textures - Per-texture MS settings When executing a texelFetch with a MS sampler, looks up that texture's settings and adjusts the parameters given to the texfetch instruction. With this change, all the ARB_texture_multisample piglits pass, so turn on PIPE_CAP_TEXTURE_MULTISAMPLE. Signed-off-by: Ilia Mirkin <[email protected]>
* nv50: copy nvc0's get_sample_position implementationIlia Mirkin2014-01-271-0/+46
| | | | Signed-off-by: Ilia Mirkin <[email protected]>
* nv50: add comments about CB_AUX contentsIlia Mirkin2014-01-273-5/+15
| | | | | | | Updates a few inconsistencies as well, like the size of the buffer, location of the runout, etc. Signed-off-by: Ilia Mirkin <[email protected]>
* nvc0: don't forget to also clear additional layersIlia Mirkin2014-01-271-5/+27
| | | | Signed-off-by: Ilia Mirkin <[email protected]>
* nv50: don't forget to also clear additional layersIlia Mirkin2014-01-271-5/+27
| | | | | | Fixes most of the tests/spec/gl-3.2/layered-rendering/* piglits. Signed-off-by: Ilia Mirkin <[email protected]>
* nv50: allocate an extra code bo to avoid dmesg spamIlia Mirkin2014-01-271-1/+5
| | | | | | | | | Each code BO is a heap that allocates at the end first, and so GPs are allocated at the very end of the allocated space. When executing, we see PAGE_NOT_PRESENT errors for the next page. Just over-allocate to make sure that there's something there. Signed-off-by: Ilia Mirkin <[email protected]>
* nv50: GP_REG_ALLOC_RESULT must be positiveIlia Mirkin2014-01-271-0/+2
| | | | | | Set max_out to 1 when there are no outputs. Signed-off-by: Ilia Mirkin <[email protected]>
* nv50: VP_RESULT_MAP_SIZE has to be positiveIlia Mirkin2014-01-271-0/+4
| | | | | | | | Make sure that we never try to use a 0-sized map. This can happen when using a gp, so add a dummy mapping when computing vp_gp_mapping in that case. Signed-off-by: Ilia Mirkin <[email protected]>
* nv50: enable primitive id generation when it is an FP input without GPIlia Mirkin2014-01-272-9/+8
| | | | Signed-off-by: Ilia Mirkin <[email protected]>
* nv50: handle gl_Layer writes in GPIlia Mirkin2014-01-274-1/+19
| | | | | | | | Marks gl_Layer as only having one component, and makes sure to keep track of where it is and emit it in the output map, since it is not an input to the FP. Signed-off-by: Ilia Mirkin <[email protected]>
* nv50: properly set the PRIMITIVE_ID enable flag when it is a gp input.Ilia Mirkin2014-01-271-0/+3
| | | | Signed-off-by: Ilia Mirkin <[email protected]>
* nv50/ir: add support for gl_PrimitiveIDInIlia Mirkin2014-01-273-5/+11
| | | | | | | | | | | | | Note that the primitive id is stored in a[0x18], while usually the geometry instructions are of the form a[$a1 + 0x4] which gets mapped to p[] space. We need to avoid the change from a[] to p[] here, so it's keyed on whether the access is indirect or not. Note that there's also a use-case for accessing e.g. a[$r1], however that's not supported for now. (Could be added by checking the register file of the indirect parameter.) Signed-off-by: Ilia Mirkin <[email protected]>
* nv50/ir: fix support for shader input + immediate in gpIlia Mirkin2014-01-272-1/+7
| | | | | | This only works for up to $a3, hopefully we won't go that high. Signed-off-by: Ilia Mirkin <[email protected]>
* nv50/ir: disallow shader input + cbuf in same instruction in gpIlia Mirkin2014-01-271-1/+6
| | | | Signed-off-by: Ilia Mirkin <[email protected]>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-21 10:39:06 +0000