summaryrefslogtreecommitdiffstats
path: root/src/gallium/drivers
Commit message (Collapse)AuthorAgeFilesLines
* nv50/ir: optimize SUB(a, b) to MOV(a - b)Samuel Pitoiset2016-09-181-0/+10
| | | | | | | | | | | | | | | | | | | This helps shaders in UE4 demos, especially with Elemental (+1% perf). This optimization reduces spilling usage in one shader which explains the little gain. GF100/GK104: total instructions in shared programs :2838551 -> 2838045 (-0.02%) total gprs used in shared programs :396706 -> 396684 (-0.01%) total local used in shared programs :34432 -> 34416 (-0.05%) local gpr inst bytes helped 1 19 112 112 hurt 0 0 0 0 Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Ilia Mirkin <[email protected]>
* gk110/ir: fix wrong emission of OP_NOTSamuel Pitoiset2016-09-181-1/+1
| | | | | | | | | This should emit src0 instead of src1. Found by inspection. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Ilia Mirkin <[email protected]> Cc: [email protected]
* r600g/sb: fix struct/class declaration conflictsMartina Kollarova2016-09-181-5/+1
| | | | | | | | | A couple of forward-declarations were causing warnings in clang: 'value' defined as a class here but previously declared as a struct [-Wmismatched-tags] Signed-off-by: Martina Kollarova <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* svga: relax restriction of compressed formats for texture uploadCharmaine Lee2016-09-171-3/+22
| | | | | | | | | | | | | | | This patch relaxes the restriction of compressed formats for texture upload buffer. For now, 3D texture with compressed format is still not supported in the texture upload buffer path. As Brian noted, ETQW does many texture updates with glCompressedTexSubImage. This patch greatly improves the performance of the ETQW trace. Tested with ETQW, MTT piglit, glretrace, conform, viewperf v2: Per Brian's suggestion, removed the subregion boundary check. Reviewed-by: Brian Paul <[email protected]>
* svga: skip query flush if we already have the query resultBrian Paul2016-09-171-5/+5
| | | | | | | | | This reduces the number of times we flush in some situations (the arbocclude demo is one trivial example). Tested with Piglit, ETQW, Sauerbraten, arbocclude. Reviewed-by: Charmaine Lee <[email protected]>
* svga: remove unneeded svga_context_flush() in svga_end_query()Brian Paul2016-09-171-5/+0
| | | | | | | | | Since commit 99d8fe20abe1f we don't have to flush the command buffer when we end a query. Tested with Piglit, Sauerbraten, arbocclude, ETQW (noticably faster now). Reviewed-by: José Fonseca <[email protected]>
* svga: use upload buffer for upload texture.Charmaine Lee2016-09-176-26/+279
| | | | | | | | | | | | | With this patch, when running with vgpu10, instead of mapping directly to the guest backed memory for texture update, we'll use the texture upload buffer and use the transfer from buffer command to update the host side texture memory. This optimization yields about 20% performance improvement with Lightsmark2008 and about 40% with Tropics. Tested with Lightsmark2008, Tropics, Heaven, MTT piglit, glretrace, conform. Reviewed-by: Brian Paul <[email protected]>
* svga: refactor svga_texture_transfer_map/unmap functionsCharmaine Lee2016-09-171-217/+272
| | | | | | | | | Split the functions into separate functions for dma and direct map to make the code more readable. Tested with MTT piglit, glretrace, viewperf, conform, various OpenGL apps Reviewed-by: Brian Paul <[email protected]>
* svga: add SVGA3d_vgpu10_TransferFromBuffer()Charmaine Lee2016-09-173-0/+51
| | | | | | Also add the corresponding dump function to dump the TransferFromBuffer command. Reviewed-by: Brian Paul <[email protected]>
* svga: single sample surface can be created as non-multisamples surfaceCharmaine Lee2016-09-172-2/+6
| | | | | | | | | With this patch, single sample surface will be created as non-multisamples surface. Tested with piglit, glretrace. Reviewed-by: Brian Paul <[email protected]>
* svga: fix memory leak with sampler stateCharmaine Lee2016-09-171-3/+0
| | | | | | | | | | | This patch fixes a memory leak with sampler state when piglit is run with HW version 11. Sampler state clean up was incorrectly skipped in svga_cleanup_sampler_state() for vgpu9. Tested with piglit. Reviewed-by: Neha Bhende <[email protected]> Reviewed-by: Brian Paul <[email protected]>
* svga: fix prim type check/assignment in translate_indices()Brian Paul2016-09-171-2/+2
| | | | | | | | Left over test code spotted by Sinclair. Tested with piglit, Google Earth, Lightsmark, Heaven4, glretraces, etc. Reviewed-by: Sinclair Yeh <[email protected]>
* svga: use SVGA3D_QUERYTYPE_MAX for svga query type checkCharmaine Lee2016-09-171-3/+3
| | | | | | | | | Use SVGA3D_QUERYTYPE_MAX instead of SVGA_QUERY_MAX for svga query type check. Tested with various OpenGL apps with GALLIUM_HUD set. Reviewed-by: Brian Paul <[email protected]>
* svga: split the num-resources-mapped hud to textures & buffersCharmaine Lee2016-09-175-12/+26
| | | | | | | | Replace the num-resources-mapped hud with num-textures-mapped and num-buffers-mapped, so we can differentiate the map counts for these two different resources. Reviewed-by: Brian Paul <[email protected]>
* svga: change svga hud defines to enumsCharmaine Lee2016-09-171-22/+25
| | | | | | This will make it easier to add new hud types. Reviewed-by: Brian Paul <[email protected]>
* svga: implement an index buffer translation cacheBrian Paul2016-09-175-6/+84
| | | | | | | | | | | | | | | Some OpenGL apps, like Cinebench R15, have many glDrawElements(GL_QUADS) calls. Since we don't directly support quads we have to convert these calls into GL_TRIANGLES which involves generating a new index buffer. This patch saves the new/translated index buffer in the hope that it can be reused for a later draw call. Cinebench R15 increases by about 20% with this change. The NobelClinician Viewer app also hits this code. Tested with full piglit run. Reviewed-by: Charmaine Lee <[email protected]>
* svga: try to emit fewer buffer rebind commandsBrian Paul2016-09-175-5/+60
| | | | | | | | | | | | | | | | | If a consecutive sequence of drawing commands references the same vertex/index buffers, there should be no need to rebind the surfaces for the second and subsequent drawing commands. Apps that use multiple display lists benefit from this since the vertex data for several display lists is often stored in one buffer. In the case of the legacy E&S Glaze demo, this reduces the size of our command buffers from 91KB to 44KB. One WSI Fusion trace shows a 33% reduction in command buffer sizes. Tested with full piglit run. Reviewed-by: Charmaine Lee <[email protected]>
* svga: reduce unmapping/remapping of the default constant bufferBrian Paul2016-09-173-6/+40
| | | | | | | | | | | | | | | Previously, every time we put shader constants into the default constant buffer we called u_upload_alloc(), which mapped the buffer, and u_upload_unmap(). We had to unmap the buffer before calling svga_buffer_handle() to get the winsys handle for the buffer. But we really only need to do that the first time we reference the const buffer. Now we try to keep the upload manager's buffer mapped until we fill it or flush the command buffer. v2: add additional comment on the buffer unmapping code in svga_context_flush(), per Charmaine. Reviewed-by: Charmaine Lee <[email protected]>
* svga: optimize memcpy() in svga_buffer_update_hw()Brian Paul2016-09-171-1/+8
| | | | | | | When we migrate a buffer from sw/malloc storage to a hardware buffer, don't memcpy the whole buffer, just copy the part we've written to. Reviewed-by: Charmaine Lee <[email protected]>
* svga: Use comparison between svga texture types to use PredCopyRegion commandNeha Bhende2016-09-171-6/+9
| | | | | | | | | | | | | | PredCopyRegion support copy between same type of textures. Instead of comparing src and dst pipe texture type, compare svga texture type which can avoid some software fallback. for example, it avoids a software blit with the Redway3D Aston demo. Tested piglit tests on VGPU9 and VGPU10 on GL/DX11Renderer, Redway3D Aston demo v2: some nit pick suggested by Charmaine. Reviewed-by: Charmaine Lee <[email protected]> Reviewed-by: Brian Paul <[email protected]>
* svga: Add function svga_resource_type()Neha Bhende2016-09-172-19/+21
| | | | | | | This function returns svga texture type for corresponding pipe texture. Reviewed-by: Charmaine Lee <[email protected]> Reviewed-by: Brian Paul <[email protected]>
* nvc0/ir: fix subops for IMADSamuel Pitoiset2016-09-171-4/+6
| | | | | | | | | Offset was wrong, it's at bit 8, not 4. Also, uses subr instead of sub when src2 has neg. Similar to GK110 now. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Ilia Mirkin <[email protected]> Cc: [email protected]
* nvc0/ir: fix comments about instructions infoSamuel Pitoiset2016-09-171-2/+3
| | | | | | | | | The comment for the commutative flags was wrong because OP_MUL is before OP_MAD. While we are at it add missing opcodes, and fix the comment about the short forms. Signed-off-by: Samuel Pitoiset <[email protected]> Acked-by: Ilia Mirkin <[email protected]>
* radeonsi/compute: Use the HSA abi for non-TGSI compute shaders v3Tom Stellard2016-09-162-18/+227
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch switches non-TGSI compute shaders over to using the HSA ABI described here: https://github.com/RadeonOpenCompute/ROCm-Docs/blob/master/AMDGPU-ABI.md The HSA ABI provides a much cleaner interface for compute shaders and allows us to share more code in the compiler with the HSA stack. The main changes in this patch are: - We now pass the scratch buffer resource into the shader via user sgprs rather than using relocations. - Grid/Block sizes are now passed to the shader via the dispatch packet rather than at the beginning of the kernel arguments. Typically for HSA, the CP firmware will create the dispatch packet and set up the user sgprs automatically. However, in Mesa we let the driver do this work. The main reason for this is that I haven't researched how to get the CP to do all these things, and I'm not sure if it is supported for all GPUs. v2: - Add comments explaining why we are setting certain bits of the scratch resource descriptor. v3: - Use amdgcn-mesa-mesa3d triple instead of amdgcn--mesa3d. Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi/compute: Add some more debug printfsTom Stellard2016-09-161-0/+3
|
* nir: Add a flag to lower_io to force "sample" interpolationJason Ekstrand2016-09-151-1/+1
| | | | | | Signed-off-by: Jason Ekstrand <[email protected]> Reviewed-by: Anuj Phogat <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* nir: Report progress from nir_lower_phis_to_scalar.Kenneth Graunke2016-09-142-3/+2
| | | | | Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Eric Anholt <[email protected]>
* nir: Report progress from nir_lower_alu_to_scalar.Kenneth Graunke2016-09-142-2/+2
| | | | | Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Eric Anholt <[email protected]>
* radeonsi: reload PS inputs with direct indexing at each use (v2)Marek Olšák2016-09-143-22/+41
| | | | | | | | | | | | | | | | | | | | | The LLVM compiler can CSE interp intrinsics thanks to LLVMReadNoneAttribute. 26011 shaders in 14651 tests Totals: SGPRS: 1146340 -> 1132676 (-1.19 %) VGPRS: 727371 -> 711730 (-2.15 %) Spilled SGPRs: 2218 -> 2078 (-6.31 %) Spilled VGPRs: 369 -> 369 (0.00 %) Scratch VGPRs: 1344 -> 1344 (0.00 %) dwords per thread Code Size: 35841268 -> 36009732 (0.47 %) bytes LDS: 767 -> 767 (0.00 %) blocks Max Waves: 222559 -> 224779 (1.00 %) Wait states: 0 -> 0 (0.00 %) v2: don't call load_input for fragment shaders in emit_declaration Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: get rid of constant buffer preloadingMarek Olšák2016-09-141-24/+14
| | | | | | | | | | | | | | | | | 26011 shaders in 14651 tests Totals: SGPRS: 1152636 -> 1146340 (-0.55 %) VGPRS: 728198 -> 727371 (-0.11 %) Spilled SGPRs: 3776 -> 2218 (-41.26 %) Spilled VGPRs: 369 -> 369 (0.00 %) Scratch VGPRs: 1344 -> 1344 (0.00 %) dwords per thread Code Size: 35835152 -> 35841268 (0.02 %) bytes LDS: 767 -> 767 (0.00 %) blocks Max Waves: 222372 -> 222559 (0.08 %) Wait states: 0 -> 0 (0.00 %) Reviewed-by: Nicolai Hähnle <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]>
* radeonsi: get rid of img/buf/sampler descriptor preloading (v2)Marek Olšák2016-09-141-132/+47
| | | | | | | | | | | | | | | | | | 26011 shaders in 14651 tests Totals: SGPRS: 1251920 -> 1152636 (-7.93 %) VGPRS: 728421 -> 728198 (-0.03 %) Spilled SGPRs: 16644 -> 3776 (-77.31 %) Spilled VGPRs: 369 -> 369 (0.00 %) Scratch VGPRs: 1344 -> 1344 (0.00 %) dwords per thread Code Size: 36001064 -> 35835152 (-0.46 %) bytes LDS: 767 -> 767 (0.00 %) blocks Max Waves: 222221 -> 222372 (0.07 %) Wait states: 0 -> 0 (0.00 %) v2: merge codepaths where possible Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: rename get_sampler_desc -> load_sampler_descMarek Olšák2016-09-141-11/+11
| | | | | Reviewed-by: Nicolai Hähnle <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]>
* radeonsi: cosmetic changes in si_shader.cMarek Olšák2016-09-141-3/+5
| | | | | Reviewed-by: Nicolai Hähnle <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]>
* radeonsi: load streamout buffer descriptors before use (v2)Marek Olšák2016-09-141-33/+14
| | | | | | v2: inline the code and remove the conditional that's a no-op now Reviewed-by: Nicolai Hähnle <[email protected]>
* vc4: Implement job shufflingEric Anholt2016-09-148-194/+333
| | | | | | | | | | | | | | | Track rendering to each FBO independently and flush rendering only when necessary. This lets us avoid the overhead of storing and loading the frame when an application momentarily switches to rendering to some other texture in order to continue rendering the main scene. Improves glmark -b desktop:effect=shadow:windows=4 by 27% Improves glmark -b desktop:blur-radius=5:effect=blur:passes=1:separable=true:windows=4 by 17% While I haven't tested other apps, this should help X rendering a lot, and I've heard GLBenchmark needed it too.
* vc4: Handle resolve skipping at job submit time.Eric Anholt2016-09-143-31/+37
| | | | | | This is done in vc4_flush currently, but I'm going to make the job always track the surfaces it might be rendering to instead of putting in the destinations at flush time.
* vc4: Move the render job state into a separate structure.Eric Anholt2016-09-1412-255/+287
| | | | | This is a preparation step for having multiple jobs being queued up at the same time.
* vc4: Always unref the current job surfaces at job reset time.Eric Anholt2016-09-143-36/+21
| | | | | Drops some tricky logic in vc4_flush() trying to update the pointers, and fixes a broken lack of unref for MSAA surfaces at context destroy time.
* vc4: Move job-submit skip cases to vc4_job_submit().Eric Anholt2016-09-142-12/+12
| | | | For calling job_submit() directly, I need the skipping here.
* vc4: Move bin CL trailer to job_submit() time.Eric Anholt2016-09-142-11/+14
| | | | | To implement job shuffling, I want to be able to call submit() on specific jobs, turning vc4_flush() into the context's flush-all-jobs hook.
* vc4: Simplify the DISCARD_RANGE handlingEric Anholt2016-09-141-12/+15
| | | | | | | It's really just an upgrade to attempting WHOLE_RESOURCE. Pulling the logic out caught two bugs in it: We would try to do so on cubemaps (even though we're only mapping 1 of the 6 slices), and we would break persistent coherent mappings by trying to reallocate when we shouldn't.
* vc4: Fix incorrect clearing of Z/stencil when cleared separately.Eric Anholt2016-09-143-15/+38
| | | | | | | | | | | | | | | | | The clear of Z or stencil will end up clearing the other as well, instead of masking. There's no way around this that I know of, so if we are clearing just one then we need to draw a quad. Fixes a regression in the job-shuffling code, where the clear values move to the job and don't just have the last clear's value laying around when you do glClear(DEPTH) and then glClear(STENCIL) separately (ext_framebuffer_multisample-clear 4 depth)). This causes regressions in ext_framebuffer_multisample/multisample-blit depth and ext_framebuffer_multisample/no-color depth, but these were formerly false positives due to the reference image also being black. Now the reference and test images are both being drawn, and it looks like there's an incorrect resolve of depth during blitting to an MSAA FBO.
* radeonsi: fix FP64 UBO loads with indirect uniform block indexingMarek Olšák2016-09-131-2/+1
| | | | | | | No known tests. Cc: [email protected] Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: clean up CP DMA emit codeMarek Olšák2016-09-131-84/+60
| | | | | | | | Unify the clear and copy paths, clean up the definitions. It looks more like a rework. It's a preparation for GDS support, which might or might not come. Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: print the IB and buffer list in VM fault reportsMarek Olšák2016-09-131-1/+2
| | | | | | This is a fallout from reworking the debug flags. Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: add sampler view BOs to the BO list lastMarek Olšák2016-09-131-7/+10
| | | | | | | | | If si_sampler_view_add_buffer ends up flushing, then the code in begin_new_cs would previously have added the buffer(s) for whatever was previously bound to that slot. Now it would add only the new buffer. Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: export SampleMask from pixel shaders at full rateMarek Olšák2016-09-133-16/+56
| | | | | | | Heaven and Valley write gl_SampleMask and not Z. Use 16_ABGR instead of 32_ABGR if Z isn't written. Reviewed-by: Nicolai Hähnle <[email protected]>
* gallium/radeon: set new r600_resource fields correctly in other places tooMarek Olšák2016-09-131-0/+11
| | | | | | | | | | | | | This was missed in: commit 0d2e43fcb1198a6e67c85feadb1ca8c360ddc284 Author: Marek Olšák <[email protected]> Date: Thu Aug 18 16:30:00 2016 +0200 gallium/radeon: derive buffer placement and flags only at initialization Tested-by: Michel Dänzer <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]>
* ddebug: dump shader buffers and imagesMarek Olšák2016-09-131-3/+27
| | | | | | this was unimplemented Reviewed-by: Nicolai Hähnle <[email protected]>
* ddebug: fix a crash in resource_get_handleMarek Olšák2016-09-131-1/+1
| | | | | | broken recently Reviewed-by: Nicolai Hähnle <[email protected]>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-26 22:15:05 +0000