summaryrefslogtreecommitdiffstats
path: root/src/gallium/drivers/radeonsi
Commit message (Collapse)AuthorAgeFilesLines
* radeonsi: remove r600_pipe_common::blit_decompress_depthMarek Olšák2018-04-052-7/+12
| | | | Acked-by: Timothy Arceri <[email protected]>
* radeonsi: remove r600_pipe_common::decompress_dccMarek Olšák2018-04-054-6/+5
| | | | Acked-by: Timothy Arceri <[email protected]>
* radeonsi: remove r600_pipe_common::invalidate_bufferMarek Olšák2018-04-051-20/+0
| | | | Acked-by: Timothy Arceri <[email protected]>
* radeonsi: remove r600_pipe_common::rebind_bufferMarek Olšák2018-04-052-4/+7
| | | | Acked-by: Timothy Arceri <[email protected]>
* radeonsi: remove r600_common_context::set_occlusion_query_stateMarek Olšák2018-04-052-4/+4
| | | | | | and remove unused old_enable parameter. Acked-by: Timothy Arceri <[email protected]>
* radeonsi: remove r600_pipe_common::save_qbo_stateMarek Olšák2018-04-052-2/+2
| | | | Acked-by: Timothy Arceri <[email protected]>
* radeonsi: remove unused query codeMarek Olšák2018-04-051-32/+0
| | | | | | The get_size perf counter callback is also inlined and removed. Acked-by: Timothy Arceri <[email protected]>
* radeonsi: use num_cs_dw_queries_suspendMarek Olšák2018-04-051-2/+6
| | | | Acked-by: Timothy Arceri <[email protected]>
* radeonsi: remove r600_pipe_common::need_gfx_cs_spaceMarek Olšák2018-04-051-7/+0
| | | | Acked-by: Timothy Arceri <[email protected]>
* radeonsi: remove r600_pipe_common::set_atom_dirtyMarek Olšák2018-04-051-1/+0
| | | | Acked-by: Timothy Arceri <[email protected]>
* radeonsi: remove r600_pipe_common::check_vm_faultsMarek Olšák2018-04-051-1/+0
| | | | Acked-by: Timothy Arceri <[email protected]>
* radeonsi: call CS flush functions directly whenever possibleMarek Olšák2018-04-054-11/+11
| | | | Acked-by: Timothy Arceri <[email protected]>
* radeonsi: skip DCC render feedback checking if color writes are disabledMarek Olšák2018-04-053-5/+23
|
* radeonsi/gfx9: fix bad LLVM params in monolithic LS+HSMarek Olšák2018-04-031-1/+5
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: Fix include for LLVMAddPromoteMemoryToRegisterPassMike Lothian2018-04-021-0/+3
| | | | | | | | | Include llvm-c/Transforms/Utils.h with the newest LLVM 7 Signed-of-by: Mike Lothian <[email protected]> Reviewed-by: Samuel Pitoiset <[email protected]> Tested-by: Dieter Nützel <[email protected]> Signed-off-by: Marek Olšák <[email protected]>
* radeonsi: implement GL_KHR_blend_equation_advancedMarek Olšák2018-04-0212-17/+203
| | | | | | | MSAA is supported using sample shading. Layered rendering and all texture targets are also supported. Tested-by: Dieter Nützel <[email protected]>
* radeonsi: rename unpack_param -> si_unpack_paramMarek Olšák2018-04-022-27/+31
| | | | Tested-by: Dieter Nützel <[email protected]>
* radeonsi: move FMASK shader logic to shared codeMarek Olšák2018-04-021-72/+2
| | | | | | We'll need it for FBFETCH in both TGSI and NIR paths. Tested-by: Dieter Nützel <[email protected]>
* radeonsi: add R600_DEBUG=nofmask to disable MSAA compressionMarek Olšák2018-04-023-8/+9
| | | | | | For testing. Tested-by: Dieter Nützel <[email protected]>
* radeonsi/nir: fix explicit component packing for geom/tess doublesTimothy Arceri2018-04-021-8/+11
| | | | Reviewed-by: Marek Olšák <[email protected]>
* radeonsi/nir: gather buffers declared more accurately and use const fast pathTimothy Arceri2018-04-022-6/+90
| | | | | | | | For now we skip SI && HAVE_LLVM < 0x0600 for simplicity. We also skip setting the more accurate masks for builtin uniforms for now as it causes some piglit regressions. Reviewed-by: Marek Olšák <[email protected]>
* radeonsi: create load_const_buffer_desc_fast_path() helperTimothy Arceri2018-04-021-39/+49
| | | | | | | | This will be shared by the TGSI and NIR backends. For simplicity we leave the SI LLVM 5.0 and lower work around only in the TGSI backend. Reviewed-by: Marek Olšák <[email protected]>
* radeonsi/nir: set TGSI_PROPERTY_NEXT_SHADERTimothy Arceri2018-04-021-0/+3
| | | | Reviewed-by: Marek Olšák <[email protected]>
* util: Move util_is_power_of_two to bitscan.h and rename to ↵Ian Romanick2018-03-291-1/+1
| | | | | | | | | | | util_is_power_of_two_or_zero The new name make the zero-input behavior more obvious. The next patch adds a new function with different zero-input behavior. Signed-off-by: Ian Romanick <[email protected]> Suggested-by: Matt Turner <[email protected]> Reviewed-by: Alejandro Piñeiro <[email protected]>
* radeonsi: don't use the SPI barrier management bug workaroundMarek Olšák2018-03-281-0/+5
| | | | Reviewed-by: Samuel Pitoiset <[email protected]>
* radeonsi: use maximum OFFCHIP_BUFFERING on Vega12Marek Olšák2018-03-281-1/+8
| | | | Reviewed-by: Samuel Pitoiset <[email protected]>
* radeonsi: add support for Vega12Marek Olšák2018-03-284-1/+7
| | | | Reviewed-by: Alex Deucher <[email protected]>
* ac/radeonsi: pass bindless bool to load_sampler_desc()Timothy Arceri2018-03-281-1/+1
| | | | | | | | We also fix the base_index for bindless by using the driver location. Reviewed-by: Samuel Pitoiset <[email protected]> Reviewed-by: Marek Olšák <[email protected]>
* radeonsi/nir: set uses_bindless_samplers for samplersTimothy Arceri2018-03-281-0/+3
| | | | | Reviewed-by: Samuel Pitoiset <[email protected]> Reviewed-by: Marek Olšák <[email protected]>
* radeonsi/nir: fix input processing for packed varyingsTimothy Arceri2018-03-281-3/+2
| | | | | | | | | The location was only being incremented the first time we processed a location. This meant we would incorrectly skip some elements of an array if the first element was packed and proccessed previously but other elements were not. Reviewed-by: Marek Olšák <[email protected]>
* radeonsi/nir: fix scanning of multi-slot output varyingsTimothy Arceri2018-03-281-109/+127
| | | | | | | | | | This fixes tcs/tes varying arrays where we dont lower indirects and therefore don't split arrays. Here we also fix useagemask for dual slot doubles. Fixes a number of arb_tessellation_shader piglit tests. Reviewed-by: Marek Olšák <[email protected]>
* nir: Rename image intrinsics to image_varJason Ekstrand2018-03-231-9/+9
| | | | | | | | | | | Generated with git grep -l nir_intrinsic_image | xargs \ sed -i 's/nir_intrinsic_image/nir_intrinsic_image_var/g' and some manual fixing in nir_intrinsics.h Reviewed-by: Timothy Arceri <[email protected]>
* radeonsi: fix a snprintf warning on gcc 7.3.0Marek Olšák2018-03-211-1/+1
|
* radeonsi/gfx9: print the swizzle mode for testdmaMarek Olšák2018-03-211-2/+16
| | | | Tested-by: Dieter Nützel <[email protected]>
* st/radeonsi: enable uniform packing in NIR backendTimothy Arceri2018-03-201-1/+5
| | | | Reviewed-by: Marek Olšák <[email protected]>
* gallium: add packed uniform CAPTimothy Arceri2018-03-201-0/+1
| | | | Reviewed-by: Marek Olšák <[email protected]>
* st/nir/radeonsi: move nir_lower_uniforms_to_ubo() to the state trackerTimothy Arceri2018-03-201-10/+0
| | | | | | | | | | | | | This will only ever be used by gallium drivers so it probably doesn't belong in the nir toolkit. Also we want to pass it some non NIR things in the following patch. To avoid regressions we wrap the lowering calls that have been moved to st_glsl_to_nir with a quick hack so that they are only called for radeonsi, we will replace the hack with a check for uniform packing in a following patch. Reviewed-by: Marek Olšák <[email protected]>
* ac/nir: pass the nir variable through tcs loading.Dave Airlie2018-03-141-5/+4
| | | | | | | | | | | | I was going to have to add another parameter to this monster, so we should just pass the nir_variable in, I can't find any reason this would be a bad idea. This needed for the next fix. Fixes: 94f9591995 (radv/ac: add support for TCS/TES inputs/outputs.) Reviewed-by: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* ac/nir: Use lower_vote_eq_to_ballot instead of ac_nir_lower_subgroupsJason Ekstrand2018-03-131-0/+1
| | | | Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radeonsi: add a workaround for GFX9 hang with init_config alignmentMarek Olšák2018-03-091-1/+2
| | | | | Fixes: 75c5d25f0f34cd702 "radeonsi: align command buffer starting address to fix some Raven hangs" Cc: 17.3 18.0 <[email protected]>
* radeonsi: remove chip_class parameter from si_lower_nirMarek Olšák2018-03-084-10/+6
| | | | | | | We can get it from si_screen. Reviewed-by: Timothy Arceri <[email protected]> Acked-by: Alex Deucher <[email protected]>
* radeonsi: expand constbuf 0 address correctly to fix Vega10 hangsMarek Olšák2018-03-081-4/+17
| | | | | | | | | | This is only required with the latest libdrm. This fixes 32-bit support with high addresses. (and possibly 64-bit support too because the high bits need to be masked out) Acked-by: Christian König <[email protected]> Acked-by: Alex Deucher <[email protected]>
* radeonsi: align command buffer starting address to fix some Raven hangsMarek Olšák2018-03-081-2/+3
| | | | | | Cc: 17.3 18.0 <[email protected]> Reviewed-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]>
* ac/radeonsi: add emit_kill to the abiTimothy Arceri2018-03-081-0/+1
| | | | | | | | This should fix a regression with Rocket League grass rendering on the NIR backend. Reviewed-by: Marek Olšák <[email protected]> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104717
* radeonsi: add si_llvm_emit_kill() helperTimothy Arceri2018-03-082-12/+21
| | | | Reviewed-by: Marek Olšák <[email protected]>
* radeonsi: make use of if/loop build helpers in acTimothy Arceri2018-03-082-160/+11
| | | | Reviewed-by: Marek Olšák <[email protected]>
* radeonsi: remove si_llvm_add_attributeMarek Olšák2018-03-073-25/+16
|
* radeonsi: fix passing address32_hi to LLVM for high valuesMarek Olšák2018-03-071-2/+5
| | | | The old function treats high values as negative, which LLVM interprets as 0.
* radeonsi: add/update assertions for 32-bit address spaceMarek Olšák2018-03-071-1/+5
|
* radeonsi: prevent a negative buffer offset in si_upload_descriptorsMarek Olšák2018-03-071-4/+3
|