summaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* freedreno/a6xx: improve setup_slices() debug msgsRob Clark2018-12-221-6/+5
| | | | Signed-off-by: Rob Clark <[email protected]>
* freedreno/a6xx: simplify special case for 3d layoutRob Clark2018-12-221-9/+10
| | | | | | | | This logic can be re-written as the two cases for 3d (ie. before/after the miplevel sizes start reducing) vs everything else. I think it is easier to read this way. Signed-off-by: Rob Clark <[email protected]>
* freedreno: combine fd_resource_layer_offset()/fd_resource_offset()Rob Clark2018-12-221-13/+2
| | | | | | We really only need this logic in one place. Signed-off-by: Rob Clark <[email protected]>
* freedreno/ir3: don't treat all inputs/outputs as vec4Rob Clark2018-12-222-14/+38
| | | | | | | | | This was a hold-over from the early TGSI days, and mostly not needed with NIR. This avoids burning an entire 4 consecutive scalar regs for vec3 outputs, for example. Which fixes a few places that we were doing worse that we should on register usage. Signed-off-by: Rob Clark <[email protected]>
* freedreno/ir3: fix fallout of extra assertRob Clark2018-12-211-1/+1
| | | | | | | | | | | | | | | Fixes the following crash that happened after d6110d4d The problem happens if we first compile a "vanilla" shader with nothing lowered in NIR, which perform the final lowering passes on so->shader-> nir (including nir_lower_locals_to_regs()), and then later we have compile a shader with some lowering. The second time through we would have already done nir_lower_locals_to_regs(). Arguably this was already a bug, just one we hadn't noticed yet. Fixes: d6110d4d547 intel/compiler: move nir_lower_bool_to_int32 before nir_lower_locals_to_regs Signed-off-by: Rob Clark <[email protected]>
* st/nir: Drop unused gl_program parameter in VS input handling helper.Kenneth Graunke2018-12-211-2/+2
| | | | | | | Nobody uses this, so let's drop it. This makes the helper callable from places without a gl_program. Reviewed-by: Marek Olšák <[email protected]>
* st/nir: Gather info after applying lowering FS variant featuresKenneth Graunke2018-12-211-0/+4
| | | | | | | | DrawPixels lowering, for example, adds new varyings that need to be accounted for in inputs_read. The earlier info gathering at link time cannot account for this. Reviewed-by: Marek Olšák <[email protected]>
* st/mesa: Combine the DrawPixels and Bitmap passthrough VS programs.Kenneth Graunke2018-12-214-33/+22
| | | | | | They're now identical, so we can just compile it once. Reviewed-by: Marek Olšák <[email protected]>
* st/mesa: Don't open code the drawpixels vertex shader.Kenneth Graunke2018-12-211-27/+11
| | | | | | Now that we always copy color, we can just use the util function. Reviewed-by: Marek Olšák <[email protected]>
* st/mesa: Drop !passColor optimization in drawpixels shaders.Kenneth Graunke2018-12-212-28/+20
| | | | | | | | | | | | | | | | | | | | The glDrawPixels passthrough vertex shader copies position and texcoord vertex attributes to varying outputs. It also optionally copies a third gl_Color attribute, which sometimes is unnecessary. Until now, we've compiled separate variants of the shader, one of which does this extra copy, and the other of which doesn't. We have done this since 2007. But, the vertex shader runs for a whopping four vertices, and so the cost of a copying a single input to output is likely inconsequential. In theory, we could bind one fewer vertex element - but we always bind all three regardless. So, we don't even get that savings. This patch unifies the two, so we always copy the optional color, and save having to compile the variant. It also makes the VS input interface match up with the vertex element state without any dead (unused) input attributes. Reviewed-by: Marek Olšák <[email protected]>
* st/mesa: Drop dead 'passthrough_fs' field.Kenneth Graunke2018-12-211-2/+0
| | | | | | | Dead since 2015 (commit 5142564734bd68f165b02e29e384ebbcf91cce38). Reviewed-by: Ilia Mirkin <[email protected]> Reviewed-by: Marek Olšák <[email protected]>
* radv: Fix wrongly positioned paren.Bas Nieuwenhuizen2018-12-211-1/+1
| | | | | | Trivial. Fixes: 9f0bfbed11f "radv: Work around non-renderable 128bpp compressed 3d textures on GFX9."
* gallivm: abort when trying to use non-existing intrinsicRoland Scheidegger2018-12-211-0/+10
| | | | | | | | | | | Whenever llvm removes an intrinsic (we're using), we're hitting segfaults due to llvm doing calls to address 0 in the jitted code instead. However, Jose figured out we can actually detect this with LLVMGetIntrinsicID(), so use this to abort, so we don't have to wonder what got broken. (Of course, someone still needs to fix the code to no longer use this intrinsic.) Reviewed-by: Jose Fonseca <[email protected]>
* gallivm: don't use pavg.b intrinsic on llvm >= 6.0Roland Scheidegger2018-12-212-51/+95
| | | | | | | | | | | | | | | | | | | | | This intrinsic disppeared with llvm 6.0, using it ends up in segfaults (due to llvm issuing call to NULL address in the jited shaders). Add code doing the same thing as the autoupgrade code in llvm so it can be matched and replaced back with a pavgb. While here, also improve lp_test_format, so it tests both with and without cache (as it was, it tested the cache versions only, whereas cache is actually disabled in llvmpipe, and in any case even with it enabled vertex and geometry shaders wouldn't use it). (Although at least for the unorm8 uncached fetch, the code is still quite different to what llvmpipe is using, since that would use unorm8x16 type, whereas the test code is using unorm8x4 type, hence disabling some intrinsic paths.) Fixes: 6f4083143bb8 ("gallivm: use llvm jit code for decoding s3tc") Reviewed-by: Jose Fonseca <[email protected]> Tested-by: Michel Dänzer <[email protected]>
* glx/test: meson: assorted include fixesEmil Velikov2018-12-121-2/+2
| | | | | | | | | Swap '..' with the symbolic inc_glx and add glproto as dependency. That will pull the correct include, effectively fixing the tests on macOS. Fixes: a47c525f328 ("meson: build glx") Signed-off-by: Emil Velikov <[email protected]> Reviewed-by: Dylan Baker <[email protected]>
* glx: meson: wire up the dispatch-index-check testEmil Velikov2018-12-121-0/+5
| | | | | | | | Accidentally dropped with earlier commit.! Fixes: 4ccb9816737 ("meson: Use consistent style for tests") Signed-off-by: Emil Velikov <[email protected]> Reviewed-by: Dylan Baker <[email protected]>
* glx: meson: drop includes from a link-only libraryEmil Velikov2018-12-121-1/+0
| | | | | | | | | When producing the final libGL.so/libGLX_mesa.so we only link the local static helper lib (libglx). Thus there's no reason for the includes. Fixes: a47c525f328 ("meson: build glx") Signed-off-by: Emil Velikov <[email protected]> Reviewed-by: Dylan Baker <[email protected]>
* TODO: glx: meson: build dri based glx tests, only with -Dglx=driEmil Velikov2018-12-122-19/+16
| | | | | | | | | | | | | | | | The library itself (libGL) is only built when -Dglx=dri, yet it's accompanying tests are build even with -Dglx=xlib. Adjust the guards, so we don't build the tests when they are not applicable v2: - Reword commit message (Dylan) - Drop build_by_default hunk (Dylan) Fixes: a47c525f328 ("meson: build glx") Signed-off-by: Emil Velikov <[email protected]> Reviewed-by: Dylan Baker <[email protected]>
* pipe-loader: meson: reference correct libraryEmil Velikov2018-12-131-1/+1
| | | | | | | | The library is called libgalliumvl_stub - note singular. Fixes: 42ea0631f10 ("meson: build clover") Signed-off-by: Emil Velikov <[email protected]> Reviewed-by: Dylan Baker <[email protected]>
* vc4: Hook up perf_debug() output to GL_ARB_debug_output as well.Eric Anholt2018-12-202-0/+3
| | | | | This is the right channel to report these things, so that end-users don't need to know each driver's custom debug options.
* vc4: Wire up core pipe_debug_callbackRhys Kidd2018-12-202-0/+14
| | | | | | | This lets the driver use pipe_debug_message() for GL_ARB_debug_output. Signed-off-by: Rhys Kidd <[email protected]> Reviewed-by: Eric Anholt <[email protected]>
* v3d: Hook up perf_debug() output to GL_ARB_debug output as well.Eric Anholt2018-12-202-0/+3
| | | | | This is the right channel to report these things, so that end-users don't need to know each driver's custom debug options.
* v3d: Wire up core pipe_debug_callbackRhys Kidd2018-12-202-0/+14
| | | | | | | This lets the driver use pipe_debug_message() for GL_ARB_debug_output. Signed-off-by: Rhys Kidd <[email protected]> Reviewed-by: Eric Anholt <[email protected]>
* v3d: Drop shadow comparison state from shader variant key.Eric Anholt2018-12-202-4/+0
| | | | The shadow state is now in the sampler.
* v3d: Fix simulator mode on i915 render nodes.Eric Anholt2018-12-201-28/+73
| | | | | | i915 render nodes refuse the dumb ioctls, so the simulator would crash on the original non-apitrace shader-db. Replace them with direct i915 calls if we detect that we're on one of their gem fds.
* radv: enable shaderStorageImageMultisample feature on GFX8+Samuel Pitoiset2018-12-203-4/+4
| | | | | | | Untested on older chips. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: add support for FMASK expandSamuel Pitoiset2018-12-207-0/+335
| | | | | | | Original patch by Dave Airlie. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: initialize FMASK for images in fully expanded modeSamuel Pitoiset2018-12-204-0/+39
| | | | | | | The value depends on the number of samples. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* ac/nir: restrict fmask lookup to image load intrinsicsSamuel Pitoiset2018-12-201-1/+1
| | | | | | | | | | | We don't ever want to do the fmask lookup on a atomic or store, the fmask should have been decompressed if the surface has been moved to IMAGE_LAYOUT. Original patch by Dave Airlie. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* spirv: add support for SpvCapabilityStorageImageMultisampleSamuel Pitoiset2018-12-202-1/+5
| | | | | Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: compute optimal VM alignment for imported buffersSamuel Pitoiset2018-12-201-1/+30
| | | | | | | | | | This fixes GPU hangs on GFX9 with dEQP-VK.memory.external_memory_host.bind_image_memory_and_render.with_zero_offset.* Copied from RadeonSI. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: Work around non-renderable 128bpp compressed 3d textures on GFX9.Bas Nieuwenhuizen2018-12-205-8/+41
| | | | | | | | | | | Exactly what title says, the new addrlib does not allow the above with certain dimensions that the CTS seems to hit. Work around it by not allowing the app to render to it via compat with other 128bpp formats and do not render to it ourselves during copies. Fixes: 776b9113656 "amd/addrlib: update Mesa's copy of addrlib" Reviewed-by: Nicolai Hähnle <[email protected]> Reviewed-by: Samuel Pitoiset <[email protected]>
* radv: fix subpass image transitions with multiviewsSamuel Pitoiset2018-12-201-0/+11
| | | | | | | | The driver needs to decompress all image layers if a fast depth/color clear has been performed. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: drop the amdgpu-skip-threshold=1 workaround for LLVM 8Samuel Pitoiset2018-12-201-3/+9
| | | | | | | | | This workaround has been introduced by 135e4d434f6 for fixing DXVK GPU hangs with many games. It is no longer needed since LLVM r345718. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* ac/nir: remove the bitfield_extract workaround for LLVM 8Samuel Pitoiset2018-12-201-9/+15
| | | | | | | | This workaround has been introduced by 3d41757788a and it is no longer needed since LLVM r346422. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Marek Olšák <[email protected]>
* intel/compiler: move nir_lower_bool_to_int32 before nir_lower_locals_to_regsIago Toral Quiroga2018-12-202-2/+4
| | | | | | | | | | | | The former expects to see SSA-only things, but the latter injects registers. The assertions in the lowering where not seeing this because they asserted on the bit_size values only, not on the is_ssa field, so add that assertion too. Fixes: 11dc1307794e "nir: Add a bool to int32 lowering pass" CC: [email protected] Reviewed-by: Jason Ekstrand <[email protected]>
* st/mesa: remove sampler associated with buffer texture in pbo logicIlia Mirkin2018-12-201-5/+0
| | | | | | | | | | | | | | A long time ago, when this was first implemented, not having a sampler bound would cause problems on Fermi. I didn't work out the reasons, but the solution was simple -- just put the samplers back in. Since then, regular texturing paths appear to have lost their associated samplers which required a fuller investigation and fix in nouveau. Now that this is done, this code should no longer need a sampler state for fetching texels from a buffer texture. Signed-off-by: Ilia Mirkin <[email protected]> Reviewed-by: Marek Olšák <[email protected]>
* gallivm: use llvm jit code for decoding s3tcRoland Scheidegger2018-12-207-383/+2239
| | | | | | | | | | | | This is (much) faster than using the util fallback. (Note that there's two methods here, one would use a cache, similar to the existing code (although the cache was disabled), except the block decode is done with jit code, the other directly decodes the required pixels. For now don't use the cache (being direct-mapped is suboptimal, but it's difficult to come up with something better which doesn't have too much overhead.) Reviewed-by: Jose Fonseca <[email protected]>
* radv/query: Use 1-bit booleans in query shadersJason Ekstrand2018-12-191-21/+21
| | | | | | | | Fixes: 44227453ec03f "nir: Switch to using 1-bit Booleans for almost..." Reviewed-by: Rhys Perry <[email protected]> Tested-by: Rhys Perry <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]> Tested-by: Bas Nieuwenhuizen <[email protected]>
* radv/query: Add a nir_test_flag helperJason Ekstrand2018-12-191-15/+16
| | | | | | | | | | This is little more than an iadd_imm right now but it will help in the next commit where we refactor things further. Reviewed-by: Rhys Perry <[email protected]> Tested-by: Rhys Perry <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]> Tested-by: Bas Nieuwenhuizen <[email protected]>
* freedreno/ir3: Handle GL_NONE in get_num_components_for_glformat()Eduardo Lima Mitev2018-12-191-3/+8
| | | | | | | | | | An earlier patch that introduced the function failed to handle the case where an image format layout qualifier is not specified, which is allowed on desktop GL profiles. In these cases, nir_variable's image format is GL_NONE, and we don't need to print a debug message for those. Reviewed-by: Ilia Mirkin <[email protected]> Reviewed-by: Rob Clark <[email protected]>
* v3d: Load and store aligned utiles all at once.Eric Anholt2018-12-191-8/+114
| | | | | | This calls the expensive uif offset function once per utile, but it still gets us a 212.218% +/- 2.41216% (n=10) win on 1024x1024 glTexImage over calling it on each pixel.
* v3d: Add a fallthrough path for utile load/store of 32 byte lines.Eric Anholt2018-12-191-12/+16
| | | | | | Now that V3D has 8 byte per pixel formats exposed, we've got stride==32 utiles to load and store. Just handle them through the non-NEON paths for now.
* vc4: Move the utile load/store functions to a header for reuse by v3d.Eric Anholt2018-12-194-202/+234
| | | | | These implementations of whole-utile load/stores would be the same for v3d, though the layouts of blocks of utiles has changed.
* v3d: Implement texture_subdata to reduce teximage upload copies.Eric Anholt2018-12-191-29/+85
| | | | | | | This lets us store the non-PBO glTexImage data directly into the tiled image without making an extra untiled memcpy for the gallium transfer. Improves 1024x1024 TexImage perf by ~19%, mostly from not thrashing around in the kernel mapping and unmapping the transfer's temporary area.
* v3d: Remove dead prototypes for load/store utile functions.Eric Anholt2018-12-191-2/+0
|
* v3d: Don't try to create shadow tiled temporaries for 1D textures.Eric Anholt2018-12-191-1/+2
| | | | | | | They're raster order anyway, so we'd assertion fail along with wasting bandwidth. Fixes: 6ad9e8690d14 ("v3d: Add support for texturing from linear.")
* v3d: Fix check for TFU job completion in the simulator.Eric Anholt2018-12-191-1/+1
| | | | | | | | | | We're waiting for the jobs-completed count to increment (with wrapping), not to reach its starting state. This mostly ended up working out because the next v3d_hw_tick() for a submit CL would end up doing the TFU operation first, but it did fail when a blit was used for glReadPixels() at the end of a test. Fixes: ee0549ff9ab3 ("v3d: Add the V3D TFU submit interface to the simulator.")
* v3d: Put the dst bo first in the list of BOs for TFU calls.Eric Anholt2018-12-191-2/+2
| | | | | | | | | | | | In the UAPI, the first BO is the destination, and the one the kernel should do an exclusive reservation on. Currently we only do exclusive reservations, anyway. However, in the simulator path I was only copying back the "destination" BO (actually src in this case), and this caused regressions once I fixed the simulator to actually complete TFU before returning (since otherwise, the TFU op would happen at the start of the next CL submit and the draw would get the right contents). Fixes: 976ea90bdca2 ("v3d: Add support for using the TFU to do some blits.")
* nir: properly find the entry to keep in copy_prop_varsCaio Marcelo de Oliveira Filho2018-12-191-3/+16
| | | | | | | | | | | | | | | | | | | When copy propagation handles a store/copy, it iterates the current copy entries to remove aliases, but keeps the "equal" entry (if exists) to be updated. The removal step may swap the entries around (to ensure there are no holes), invalidating previous iteration pointers. The bug was saving such pointer to use later. Change the code to first perform the removals and then find the remaining right entry. This was causing updates to be lost since they were being made to an entry that was not part of the current copies. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=108624 Fixes: b3c61469255 "nir: Copy propagation between blocks" Cc: [email protected] Reviewed-by: Jason Ekstrand <[email protected]>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-22 18:05:40 +0000