summaryrefslogtreecommitdiffstats
path: root/src/amd
Commit message (Collapse)AuthorAgeFilesLines
* radeonsi: document a CP DMA bug that doesn't need a workaround yetMarek Olšák2016-12-011-1/+5
| | | | | | This one is easy to miss, because it's not documented in any internal doc. Reviewed-by: Nicolai Hähnle <[email protected]>
* ac/nir: Fix out of bounds array access.Bas Nieuwenhuizen2016-11-301-1/+1
| | | | | | | With nir_intrinsic_ssbo_atomic_comp_swap we run out of params. Signed-off-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* radv: set spi_baryc_cntl.pos_float_location to 0Dave Airlie2016-11-291-1/+1
| | | | | | | | | | | This fixes: dEQP-VK.pipeline.multisample_interpolation.offset_interpolate_at_sample_position.* This should probably be 2 when sample shading is enabled, but I'm not sure. Reviewed-by: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: force persample shading when required.Dave Airlie2016-11-294-6/+23
| | | | | | | | | | | | | | | | | | | We need to force persample shading when a) shader uses sample_id b) shader uses sample_position c) shader uses sample qualifier. Also since ps_iter_samples can now change independently of the rasterizer samples we need to move setting the regs more often. This fixes: dEQP-VK.pipeline.multisample_interpolation.centroid_interpolate_at_consistency.* dEQP-VK.pipeline.multisample_interpolation.centroid_qualifier_inside_primitive.137_191_1.* dEQP-VK.pipeline.multisample_interpolation.sample_interpolate_at_distinct_values.* dEQP-VK.pipeline.multisample_interpolation.sample_qualifier_distinct_values.128_128_1.* Reviewed-by: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* ac/nir: Fix accessing an unitialized value.Bas Nieuwenhuizen2016-11-291-1/+2
| | | | | Signed-off-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* radv: Initialize the shader_stats_dump flag.Bas Nieuwenhuizen2016-11-291-0/+1
| | | | | | | | Meta was using it before it was set. I suspect we typically don't want to dump meta shaders, so just set it to false in the beginning. Signed-off-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* radv: Use different intrinsic for ubo loads.Bas Nieuwenhuizen2016-11-291-1/+29
| | | | | | | | | | | Not sure about the deprecation path, but this intrinsic can be lowered to SMEM loads. This results in a significant Talos performance improvement. v2: Fix for LLVM attribute changes. Signed-off-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Dave Airlie <[email protected]>
* anv: set maxFragmentDualSrcAttachments to 1Dave Airlie2016-11-291-1/+1
| | | | | | | | Reviewed-by: Kenneth Graunke <[email protected]> Reviewed-by: Jason Ekstrand <[email protected]> Reported-by: Ilia Mirkin <[email protected]> Cc: "13.0" <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: don't return VK_SUCCESS if radv_device_get_cache_uuid() failsEmil Velikov2016-11-281-0/+2
| | | | | | | | | If radv_device_get_cache_uuid() fails result will be VK_SUCCESS as set by the radv_init_wsi() call above. Fixes: d943839 (radv: Use library mtime for cache UUID.) Signed-off-by: Emil Velikov <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: don't leak the fd if radv_physical_device_init() succeedsEmil Velikov2016-11-281-0/+1
| | | | | | | | | | radv_amdgpu_winsys_create() does not take ownership of the fd, thus we end up leaking it as we return with VK_SUCCESS. Cc: Dave Airlie <[email protected]> Cc: "13.0" <[email protected]> Signed-off-by: Emil Velikov <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: Make radv_finishme only warn once per call-siteEmil Velikov2016-11-281-1/+7
| | | | | Signed-off-by: Emil Velikov <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* radv: fix 3D clears with baseMiplevelDave Airlie2016-11-281-1/+1
| | | | | | | | | | | | This fixes: dEQP-VK.api.image_clearing.clear_color_image.3d* These were hitting an assert as the code wasn't taking the baseMipLevel into account when minify the image depth. Reviewed-by: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Dave Airlie <[email protected]> Cc: "13.0" <[email protected]>
* radv: brown-paper bag for a forgotten else.Dave Airlie2016-11-281-1/+1
| | | | | | | | This fixes the fix: radv/ac/llvm: fix regression with shadow samplers fix Signed-off-by: Dave Airlie <[email protected]> Cc: "13.0" <[email protected]>
* radv/ac/llvm: fix regression with shadow samplers fixDave Airlie2016-11-281-3/+3
| | | | | | | | | | | This fixes b56b54cbf1d8e70c87a434da5350d11533e5fed8: radv/ac/llvm: shadow samplers only return one value It makes sure we only do that for shadow sampling, as opposed to sizing requests. Signed-off-by: Dave Airlie <[email protected]> Cc: "13.0" <[email protected]>
* radv/ac/llvm: shadow samplers only return one value.Dave Airlie2016-11-271-1/+3
| | | | | | | | The intrinsic engine asserts in llvm due to this. Reported-by: Christoph Haag <[email protected]> Cc: "13.0" <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv/si: fix optimal micro tile selectionDave Airlie2016-11-271-7/+7
| | | | | | | | The same fix was posted for radeonsi, so port it here. Reviewed-by: Edward O'Callaghan <[email protected]> Cc: "13.0" <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: honour the number of properties availableEmil Velikov2016-11-271-11/+7
| | | | | | | | | | | Cap up-to the number of properties available while copying the data. Otherwise we might crash and/or leak data. Cc: Dave Airlie <[email protected]> Cc: "13.0" <[email protected]> Signed-off-by: Emil Velikov <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: drop the return type for radv_queue_init()Mun Gwan-gyeong2016-11-271-3/+1
| | | | | | | | radv_queue_init() always returns VK_SUCCESS, so caller does not need to check return value of radv_queue_init(). Signed-off-by: Mun Gwan-gyeong <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: Don't generate radv_timestamp.hBas Nieuwenhuizen2016-11-242-8/+1
| | | | | | | | Not needed anymore. Signed-off-by: Bas Nieuwenhuizen <[email protected]> Reviewed-by: Dave Airlie <[email protected]> Reviewed-by: Emil Velikov <[email protected]>
* radv: fix texel fetch offset with 2d arrays.Dave Airlie2016-11-241-3/+4
| | | | | | | | | | | | The code didn't limit the offsets to the number supplied, so if we expected 3 but only got 2 we were accessing undefined memory. This fixes random failures in: dEQP-VK.glsl.texture_functions.texelfetchoffset.sampler2darray_* Reviewed-by: Bas Nieuwenhuizen <[email protected]> Cc: "13.0" <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: fix flipped blitsDave Airlie2016-11-231-57/+111
| | | | | | | | This fixes: dEQP-VK.api.copy_and_blit.blit_image.simple_tests.mirror* Reviewed-by: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv/meta: just local vars for src/dst subresources.Dave Airlie2016-11-231-8/+10
| | | | | | | | This is just a cleanup before I rework this code to fix mirrored blits. Reviewed-by: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: add support for VK_AMD_draw_indirect_countFredrik Höglund2016-11-242-14/+84
| | | | Reviewed-by: Dave Airlie <[email protected]>
* radv: add support for VK_AMD_negative_viewport_heightFredrik Höglund2016-11-241-0/+4
| | | | | | The driver already supports this extension in practice. Reviewed-by: Dave Airlie <[email protected]>
* radv: add support for VK_KHR_sampler_mirror_clamp_to_edgeFredrik Höglund2016-11-241-0/+4
| | | | | | | radv_tex_wrap() already supports VK_SAMPLER_ADDRESS_MODE_MIRROR_CLAMP_TO_EDGE, so all that's needed is to advertise support for the extension. Reviewed-by: Dave Airlie <[email protected]>
* radv: add support for anisotropic filtering on SI-CIFredrik Höglund2016-11-242-9/+33
| | | | | | | | | Ported from radeonsi. Note that si_make_texture_descriptor() already sets img7 to the mask value referred to in the comment. Reviewed-by: Dave Airlie <[email protected]>
* radv: move pipeline barrier image transitions after src flushingDave Airlie2016-11-231-8/+12
| | | | | | | | | This seems like it would conform better with the spec. noticed while digging into fast clears. Reviewed-by: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: add support for shader stats dumpDave Airlie2016-11-223-0/+87
| | | | | | | | | | | I've started working on a shader-db alike for Vulkan, it's based on vktrace and it records pipelines, this adds support to dump the shader stats exactly like radeonsi does, so I can reuse the shader-db scripts it uses. Reviewed-by: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: fix sample id loadingDave Airlie2016-11-221-1/+18
| | | | | | | | The sample id is packed into bits 8-12, so adjust things properly. Reviewed-by: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv/ac: add implementation of load_sample_pos intrinsic.Dave Airlie2016-11-221-0/+12
| | | | | | | This fixes a bunch of crashes in CTS tests looking for this. Reviewed-by: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv/ac: cleanup ddxy emissionDave Airlie2016-11-221-93/+43
| | | | | | | | | | This cleans up the ddxy emission along the same lines as radeonsi. It also means we don't use LDS on VI chips we use the dspermute interface, it also removes some duplicated code. Reviewed-by: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv/meta: cleanup resolve vertex state emissionDave Airlie2016-11-221-47/+2
| | | | | | | | For the hw resolve there is no need to emit any sort of texture coordinates, so drop them all in the meta path. Reviewed-by: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: Incorporate GPU family into cache UUID.Bas Nieuwenhuizen2016-11-221-3/+5
| | | | | | Invalidates the cache when someone switches cards. Signed-off-by: Bas Nieuwenhuizen <[email protected]>
* radv: Use library mtime for cache UUID.Bas Nieuwenhuizen2016-11-221-4/+32
| | | | | | | | | | | | We want to also invalidate the cache when LLVM gets changed. As the specific LLVM revision is not fixed at build time, we will need to check at runtime. Computing a checksum for LLVM is going to be very expensive, so just use the mtime. Tested on my computer that the returned DSO for the LLVM symbol is actually the LLVM DSO. Signed-off-by: Bas Nieuwenhuizen <[email protected]>
* radv: Store UUID in physical device.Bas Nieuwenhuizen2016-11-223-14/+16
| | | | | | | No sense in repeatedly determining it. Also, it might be dependent on the device as shaders get compiled differently for SI/CIK/VI etc. Signed-off-by: Bas Nieuwenhuizen <[email protected]>
* radv: spir-v allows texture size query with and without lod.Dave Airlie2016-11-211-1/+4
| | | | | | | | | | The translation to llvm was failing here due to required lod. This fixes some new SteamVR shaders. Cc: "13.0" <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: fix image view creation for depth and stencil onlyDave Airlie2016-11-212-13/+9
| | | | | | | | | | | | | | | This fixes the image view for sampling just the depth. It removes some pointless swizzle code, and adds a missing case for the x8_d24 format. Fixes: dEQP-VK.renderpass.formats.d32_sfloat_s8_uint.input.* dEQP-VK.renderpass.formats.d24_unorm_s8_uint.input.* dEQP-VK.renderpass.formats.x8_d24_unorm_pack32.input.* Cc: "13.0" <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: make sure to flush input attachments correctly.Dave Airlie2016-11-211-0/+1
| | | | | | | | | | This fixes 9 of the dEQP-VK.renderpass.attachment_allocation.input_output.* tests. Cc: "13.0" <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* ac/nir/llvm: fix channel in texture gather lowering code.Dave Airlie2016-11-161-1/+1
| | | | | | | | This fixes a number of CTS tests like: dEQP-VK.glsl.texture_gather.basic.2d.rgba8ui.size_npot.clamp_to_edge_repeat Cc: "13.0" <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: don't crash on null swapchain destroy.Dave Airlie2016-11-161-0/+3
| | | | | | | | | Just return if the passed in swapchain is NULL. Fixes: dEQP-VK.wsi.xlib.swapchain.destroy.null_handle Cc: "13.0" <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* amd: automake: android: rename sources lists to foo_FILESEmil Velikov2016-11-153-5/+5
| | | | | | | | Autotools goes smart on us warning that foo_SOURCES variable is present yet a target with name foo is missing. Rename things (like we do throughout the build) to silence the warnings. Signed-off-by: Emil Velikov <[email protected]>
* amd: flatten amd/common makefile structureMauro Rossi2016-11-153-38/+14
| | | | | | | | | | | This pulls amd/common build rules into upper level makefile, along with amd/addlib which is already there. v2: [Emil Velikov] - Move NEED_RADEON_LLVM conditional, drop amd/common from SUBDIRS - Drop AM_ from common_libamd_common_la* Signed-off-by: Emil Velikov <[email protected]>
* radv/pipeline: Don't dereference NULL dynamic state pointersDarren Salt2016-11-121-20/+47
| | | | | | | | | | | | | | | | | This is a port of commit a4a59172482d50318a5ae7f99021bcf0125e0f53: Add guards to prevent dereferencing NULL dynamic pipeline state. Asserts of pCreateInfo members are moved to the earliest points at which they should not be NULL. This fixes a segfault, related to pColorBlendState, seen in Talos Principle which I've observed after startup is completed and when exiting the menus, depending on when Vulkan rendering is selected. v2: moved the NULL check in radv_pipeline_init_blend_state to after the declarations. Acked-by: Edward O'Callaghan <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* ac/nir/llvm: Fix setting function attributes for intrinsicsDaniel Scharrer2016-11-111-1/+5
| | | | | | | | This fixes a NULL pointer dereference for intrinsics with more than one function attribute introduced in commit 2fdaf38. The fix is ported from the lp_build_intrinsic changes in commit 8bdd52c. Reviewed-by: Bas Nieuwenhuizen <[email protected]>
* vulkan: import latest public vulkan headers + and fix drivers.Dave Airlie2016-11-111-1/+1
| | | | | | | | | I just noticed the new vulkan headers changed a prototype, so I've decided to import them and fix the drivers to use the new API. Acked-by: Jason Ekstrand <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: fix texturesamples to handle single sample caseDave Airlie2016-11-111-2/+10
| | | | | | | | | | | | We can only read the valid samples if this is an MSAA texture, which means the type field must be 0x14 or 0x15. This fixes: dEQP-VK.glsl.texture_functions.query.texturesamples.* Cc: "13.0" <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* radv: automake: list correct file in the EXTRA_DISTEmil Velikov2016-11-101-1/+1
| | | | | | | | | Earlier commit renamed the file radeon_icd.json{,.in} but missed one reference of the file - in EXTRA_DIST. Cc: "13.0" <[email protected]> Fixes: 0f434a68a ("radv: Suffix the radeon_icd file with the host CPU") Signed-off-by: Emil Velikov <[email protected]>
* radv: fixup botched llvm API changes.Dave Airlie2016-11-101-4/+3
| | | | | Reported-by: Jan Vesely <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* ac/nir/llvm: adopt to new LLVM attribute API.Dave Airlie2016-11-101-36/+108
| | | | | | | | Ported from corresponding changes to gallivm. tested build against 3.9 and master. Signed-off-by: Dave Airlie <[email protected]>
* radv: Suffix the radeon_icd file with the host CPUEmil Velikov2016-11-093-3/+9
| | | | | | | | | | | Port of the anv commit d96345de989 ("anv: Suffix the intel_icd file with the host CPU"). v2: s/intel_icd/radeon_icd/ in commit summary (Gražvydas) Cc: "13.0" <[email protected]> Signed-off-by: Emil Velikov <[email protected]> Reviewed-by: Dave Airlie <[email protected]> (IRC)
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-21 08:44:37 +0000