encounter
/
dawn-cmake
mirror of https://github.com/encounter/dawn-cmake.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
dawn-cmake/test/vk-gl-cts/graphicsfuzz/spv-stable-quicksort-dontin.../0-opt.spvasm.expected.spvasm

542 lines
22 KiB
Plaintext
Raw Normal View History

Add tests derived from VK-GL-CTS This adds SPIR-V assembly and WGSL tests derived from VK-GL-CTS commit 571256871c2e2f03995373e1e4a02958d8cd8cf5. The following procedure was followed: - Those .amber files in VK-GL-CTS wholly owned by Google were identified - All GLSL and SPIR-V shaders were extracted from the Amber files and converted into SPIR-V binaries - The compact-ids pass of spirv-opt was applied to each binary - Duplicate binaries were removed - spirv-opt -O was used to obtain an optimized version of each remaining binary, with duplicates discarded - Binaries that failed validation using spirv-val with target environment SPIR-V 1.3 were discarded - Those binaries that tint could not successfully convert into WGSL were put aside for further investigation - SPIR-V assembly versions of the remaining binaries are included in this CL - test-runner with -generate-expected and -generate-skip was used to generate expected .spvasm, .msl, .hlsl and .wgsl outputs for these SPIR-V assembly tests - Each successfully-generated .expected.wgsl is included in this CL again, as a WGLSL test - test-runner with -generate-expected and -generate-skip was used again, to generate expected outputs for these WGSL tests Change-Id: Ibe9baf2729cf97e0b633db9a426f53362a5de540 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/58842 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: Ben Clayton <bclayton@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-07-23 13:10:12 +00:00
; SPIR-V
; Version: 1.3
; Generator: Google Tint Compiler; 0
; Bound: 374
; Schema: 0
OpCapability Shader
%343 = OpExtInstImport "GLSL.std.450"
OpMemoryModel Logical GLSL450
spirv: Use generic transform to process shader IO The refactored CanonicalizeEntryPointIO transform makes it much easier to handle SPIR-V style IO as well, and doing this removes a lot of duplicated code. Remove all of the SPIR-V transform code for shader IO and vertex point size. Bug: tint:920 Change-Id: Id1b97517619b4d2fd09b45d5aee848259f3dfa77 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/60840 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: James Price <jrprice@google.com> Auto-Submit: James Price <jrprice@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-08-05 17:34:19 +00:00
OpEntryPoint Fragment %main "main" %gl_FragCoord_param_1 %x_GLF_color_1_1
Add tests derived from VK-GL-CTS This adds SPIR-V assembly and WGSL tests derived from VK-GL-CTS commit 571256871c2e2f03995373e1e4a02958d8cd8cf5. The following procedure was followed: - Those .amber files in VK-GL-CTS wholly owned by Google were identified - All GLSL and SPIR-V shaders were extracted from the Amber files and converted into SPIR-V binaries - The compact-ids pass of spirv-opt was applied to each binary - Duplicate binaries were removed - spirv-opt -O was used to obtain an optimized version of each remaining binary, with duplicates discarded - Binaries that failed validation using spirv-val with target environment SPIR-V 1.3 were discarded - Those binaries that tint could not successfully convert into WGSL were put aside for further investigation - SPIR-V assembly versions of the remaining binaries are included in this CL - test-runner with -generate-expected and -generate-skip was used to generate expected .spvasm, .msl, .hlsl and .wgsl outputs for these SPIR-V assembly tests - Each successfully-generated .expected.wgsl is included in this CL again, as a WGLSL test - test-runner with -generate-expected and -generate-skip was used again, to generate expected outputs for these WGSL tests Change-Id: Ibe9baf2729cf97e0b633db9a426f53362a5de540 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/58842 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: Ben Clayton <bclayton@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-07-23 13:10:12 +00:00
OpExecutionMode %main OriginUpperLeft
spirv: Use generic transform to process shader IO The refactored CanonicalizeEntryPointIO transform makes it much easier to handle SPIR-V style IO as well, and doing this removes a lot of duplicated code. Remove all of the SPIR-V transform code for shader IO and vertex point size. Bug: tint:920 Change-Id: Id1b97517619b4d2fd09b45d5aee848259f3dfa77 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/60840 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: James Price <jrprice@google.com> Auto-Submit: James Price <jrprice@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-08-05 17:34:19 +00:00
OpName %gl_FragCoord_param_1 "gl_FragCoord_param_1"
OpName %x_GLF_color_1_1 "x_GLF_color_1_1"
Add tests derived from VK-GL-CTS This adds SPIR-V assembly and WGSL tests derived from VK-GL-CTS commit 571256871c2e2f03995373e1e4a02958d8cd8cf5. The following procedure was followed: - Those .amber files in VK-GL-CTS wholly owned by Google were identified - All GLSL and SPIR-V shaders were extracted from the Amber files and converted into SPIR-V binaries - The compact-ids pass of spirv-opt was applied to each binary - Duplicate binaries were removed - spirv-opt -O was used to obtain an optimized version of each remaining binary, with duplicates discarded - Binaries that failed validation using spirv-val with target environment SPIR-V 1.3 were discarded - Those binaries that tint could not successfully convert into WGSL were put aside for further investigation - SPIR-V assembly versions of the remaining binaries are included in this CL - test-runner with -generate-expected and -generate-skip was used to generate expected .spvasm, .msl, .hlsl and .wgsl outputs for these SPIR-V assembly tests - Each successfully-generated .expected.wgsl is included in this CL again, as a WGLSL test - test-runner with -generate-expected and -generate-skip was used again, to generate expected outputs for these WGSL tests Change-Id: Ibe9baf2729cf97e0b633db9a426f53362a5de540 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/58842 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: Ben Clayton <bclayton@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-07-23 13:10:12 +00:00
OpName %QuicksortObject "QuicksortObject"
OpMemberName %QuicksortObject 0 "numbers"
OpName %obj "obj"
OpName %gl_FragCoord "gl_FragCoord"
OpName %buf0 "buf0"
OpMemberName %buf0 0 "resolution"
OpName %x_32 "x_32"
OpName %x_GLF_color "x_GLF_color"
OpName %swap_i1_i1_ "swap_i1_i1_"
OpName %i "i"
OpName %j "j"
OpName %temp "temp"
OpName %performPartition_i1_i1_ "performPartition_i1_i1_"
OpName %l "l"
OpName %h "h"
OpName %pivot "pivot"
OpName %i_1 "i_1"
OpName %j_1 "j_1"
OpName %param "param"
OpName %param_1 "param_1"
OpName %param_2 "param_2"
OpName %param_3 "param_3"
OpName %quicksort_ "quicksort_"
OpName %l_1 "l_1"
OpName %h_1 "h_1"
OpName %top "top"
OpName %stack "stack"
OpName %p "p"
OpName %param_4 "param_4"
OpName %param_5 "param_5"
OpName %main_1 "main_1"
OpName %i_2 "i_2"
OpName %uv "uv"
OpName %color "color"
OpName %main_out "main_out"
OpMemberName %main_out 0 "x_GLF_color_1"
spirv: Use generic transform to process shader IO The refactored CanonicalizeEntryPointIO transform makes it much easier to handle SPIR-V style IO as well, and doing this removes a lot of duplicated code. Remove all of the SPIR-V transform code for shader IO and vertex point size. Bug: tint:920 Change-Id: Id1b97517619b4d2fd09b45d5aee848259f3dfa77 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/60840 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: James Price <jrprice@google.com> Auto-Submit: James Price <jrprice@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-08-05 17:34:19 +00:00
OpName %main_inner "main_inner"
OpName %gl_FragCoord_param "gl_FragCoord_param"
Add tests derived from VK-GL-CTS This adds SPIR-V assembly and WGSL tests derived from VK-GL-CTS commit 571256871c2e2f03995373e1e4a02958d8cd8cf5. The following procedure was followed: - Those .amber files in VK-GL-CTS wholly owned by Google were identified - All GLSL and SPIR-V shaders were extracted from the Amber files and converted into SPIR-V binaries - The compact-ids pass of spirv-opt was applied to each binary - Duplicate binaries were removed - spirv-opt -O was used to obtain an optimized version of each remaining binary, with duplicates discarded - Binaries that failed validation using spirv-val with target environment SPIR-V 1.3 were discarded - Those binaries that tint could not successfully convert into WGSL were put aside for further investigation - SPIR-V assembly versions of the remaining binaries are included in this CL - test-runner with -generate-expected and -generate-skip was used to generate expected .spvasm, .msl, .hlsl and .wgsl outputs for these SPIR-V assembly tests - Each successfully-generated .expected.wgsl is included in this CL again, as a WGLSL test - test-runner with -generate-expected and -generate-skip was used again, to generate expected outputs for these WGSL tests Change-Id: Ibe9baf2729cf97e0b633db9a426f53362a5de540 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/58842 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: Ben Clayton <bclayton@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-07-23 13:10:12 +00:00
OpName %main "main"
spirv: Use generic transform to process shader IO The refactored CanonicalizeEntryPointIO transform makes it much easier to handle SPIR-V style IO as well, and doing this removes a lot of duplicated code. Remove all of the SPIR-V transform code for shader IO and vertex point size. Bug: tint:920 Change-Id: Id1b97517619b4d2fd09b45d5aee848259f3dfa77 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/60840 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: James Price <jrprice@google.com> Auto-Submit: James Price <jrprice@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-08-05 17:34:19 +00:00
OpDecorate %gl_FragCoord_param_1 BuiltIn FragCoord
OpDecorate %x_GLF_color_1_1 Location 0
Add tests derived from VK-GL-CTS This adds SPIR-V assembly and WGSL tests derived from VK-GL-CTS commit 571256871c2e2f03995373e1e4a02958d8cd8cf5. The following procedure was followed: - Those .amber files in VK-GL-CTS wholly owned by Google were identified - All GLSL and SPIR-V shaders were extracted from the Amber files and converted into SPIR-V binaries - The compact-ids pass of spirv-opt was applied to each binary - Duplicate binaries were removed - spirv-opt -O was used to obtain an optimized version of each remaining binary, with duplicates discarded - Binaries that failed validation using spirv-val with target environment SPIR-V 1.3 were discarded - Those binaries that tint could not successfully convert into WGSL were put aside for further investigation - SPIR-V assembly versions of the remaining binaries are included in this CL - test-runner with -generate-expected and -generate-skip was used to generate expected .spvasm, .msl, .hlsl and .wgsl outputs for these SPIR-V assembly tests - Each successfully-generated .expected.wgsl is included in this CL again, as a WGLSL test - test-runner with -generate-expected and -generate-skip was used again, to generate expected outputs for these WGSL tests Change-Id: Ibe9baf2729cf97e0b633db9a426f53362a5de540 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/58842 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: Ben Clayton <bclayton@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-07-23 13:10:12 +00:00
OpMemberDecorate %QuicksortObject 0 Offset 0
OpDecorate %_arr_int_uint_10 ArrayStride 4
OpDecorate %buf0 Block
OpMemberDecorate %buf0 0 Offset 0
OpDecorate %x_32 NonWritable
OpDecorate %x_32 DescriptorSet 0
OpDecorate %x_32 Binding 0
OpMemberDecorate %main_out 0 Offset 0
spirv: Use generic transform to process shader IO The refactored CanonicalizeEntryPointIO transform makes it much easier to handle SPIR-V style IO as well, and doing this removes a lot of duplicated code. Remove all of the SPIR-V transform code for shader IO and vertex point size. Bug: tint:920 Change-Id: Id1b97517619b4d2fd09b45d5aee848259f3dfa77 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/60840 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: James Price <jrprice@google.com> Auto-Submit: James Price <jrprice@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-08-05 17:34:19 +00:00
%float = OpTypeFloat 32
%v4float = OpTypeVector %float 4
%_ptr_Input_v4float = OpTypePointer Input %v4float
%gl_FragCoord_param_1 = OpVariable %_ptr_Input_v4float Input
%_ptr_Output_v4float = OpTypePointer Output %v4float
%7 = OpConstantNull %v4float
%x_GLF_color_1_1 = OpVariable %_ptr_Output_v4float Output %7
Add tests derived from VK-GL-CTS This adds SPIR-V assembly and WGSL tests derived from VK-GL-CTS commit 571256871c2e2f03995373e1e4a02958d8cd8cf5. The following procedure was followed: - Those .amber files in VK-GL-CTS wholly owned by Google were identified - All GLSL and SPIR-V shaders were extracted from the Amber files and converted into SPIR-V binaries - The compact-ids pass of spirv-opt was applied to each binary - Duplicate binaries were removed - spirv-opt -O was used to obtain an optimized version of each remaining binary, with duplicates discarded - Binaries that failed validation using spirv-val with target environment SPIR-V 1.3 were discarded - Those binaries that tint could not successfully convert into WGSL were put aside for further investigation - SPIR-V assembly versions of the remaining binaries are included in this CL - test-runner with -generate-expected and -generate-skip was used to generate expected .spvasm, .msl, .hlsl and .wgsl outputs for these SPIR-V assembly tests - Each successfully-generated .expected.wgsl is included in this CL again, as a WGLSL test - test-runner with -generate-expected and -generate-skip was used again, to generate expected outputs for these WGSL tests Change-Id: Ibe9baf2729cf97e0b633db9a426f53362a5de540 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/58842 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: Ben Clayton <bclayton@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-07-23 13:10:12 +00:00
%int = OpTypeInt 32 1
%uint = OpTypeInt 32 0
%uint_10 = OpConstant %uint 10
%_arr_int_uint_10 = OpTypeArray %int %uint_10
%QuicksortObject = OpTypeStruct %_arr_int_uint_10
%_ptr_Private_QuicksortObject = OpTypePointer Private %QuicksortObject
spirv: Use generic transform to process shader IO The refactored CanonicalizeEntryPointIO transform makes it much easier to handle SPIR-V style IO as well, and doing this removes a lot of duplicated code. Remove all of the SPIR-V transform code for shader IO and vertex point size. Bug: tint:920 Change-Id: Id1b97517619b4d2fd09b45d5aee848259f3dfa77 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/60840 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: James Price <jrprice@google.com> Auto-Submit: James Price <jrprice@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-08-05 17:34:19 +00:00
%15 = OpConstantNull %QuicksortObject
%obj = OpVariable %_ptr_Private_QuicksortObject Private %15
Add tests derived from VK-GL-CTS This adds SPIR-V assembly and WGSL tests derived from VK-GL-CTS commit 571256871c2e2f03995373e1e4a02958d8cd8cf5. The following procedure was followed: - Those .amber files in VK-GL-CTS wholly owned by Google were identified - All GLSL and SPIR-V shaders were extracted from the Amber files and converted into SPIR-V binaries - The compact-ids pass of spirv-opt was applied to each binary - Duplicate binaries were removed - spirv-opt -O was used to obtain an optimized version of each remaining binary, with duplicates discarded - Binaries that failed validation using spirv-val with target environment SPIR-V 1.3 were discarded - Those binaries that tint could not successfully convert into WGSL were put aside for further investigation - SPIR-V assembly versions of the remaining binaries are included in this CL - test-runner with -generate-expected and -generate-skip was used to generate expected .spvasm, .msl, .hlsl and .wgsl outputs for these SPIR-V assembly tests - Each successfully-generated .expected.wgsl is included in this CL again, as a WGLSL test - test-runner with -generate-expected and -generate-skip was used again, to generate expected outputs for these WGSL tests Change-Id: Ibe9baf2729cf97e0b633db9a426f53362a5de540 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/58842 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: Ben Clayton <bclayton@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-07-23 13:10:12 +00:00
%_ptr_Private_v4float = OpTypePointer Private %v4float
spirv: Use generic transform to process shader IO The refactored CanonicalizeEntryPointIO transform makes it much easier to handle SPIR-V style IO as well, and doing this removes a lot of duplicated code. Remove all of the SPIR-V transform code for shader IO and vertex point size. Bug: tint:920 Change-Id: Id1b97517619b4d2fd09b45d5aee848259f3dfa77 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/60840 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: James Price <jrprice@google.com> Auto-Submit: James Price <jrprice@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-08-05 17:34:19 +00:00
%gl_FragCoord = OpVariable %_ptr_Private_v4float Private %7
Add tests derived from VK-GL-CTS This adds SPIR-V assembly and WGSL tests derived from VK-GL-CTS commit 571256871c2e2f03995373e1e4a02958d8cd8cf5. The following procedure was followed: - Those .amber files in VK-GL-CTS wholly owned by Google were identified - All GLSL and SPIR-V shaders were extracted from the Amber files and converted into SPIR-V binaries - The compact-ids pass of spirv-opt was applied to each binary - Duplicate binaries were removed - spirv-opt -O was used to obtain an optimized version of each remaining binary, with duplicates discarded - Binaries that failed validation using spirv-val with target environment SPIR-V 1.3 were discarded - Those binaries that tint could not successfully convert into WGSL were put aside for further investigation - SPIR-V assembly versions of the remaining binaries are included in this CL - test-runner with -generate-expected and -generate-skip was used to generate expected .spvasm, .msl, .hlsl and .wgsl outputs for these SPIR-V assembly tests - Each successfully-generated .expected.wgsl is included in this CL again, as a WGLSL test - test-runner with -generate-expected and -generate-skip was used again, to generate expected outputs for these WGSL tests Change-Id: Ibe9baf2729cf97e0b633db9a426f53362a5de540 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/58842 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: Ben Clayton <bclayton@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-07-23 13:10:12 +00:00
%v2float = OpTypeVector %float 2
%buf0 = OpTypeStruct %v2float
%_ptr_Uniform_buf0 = OpTypePointer Uniform %buf0
%x_32 = OpVariable %_ptr_Uniform_buf0 Uniform
spirv: Use generic transform to process shader IO The refactored CanonicalizeEntryPointIO transform makes it much easier to handle SPIR-V style IO as well, and doing this removes a lot of duplicated code. Remove all of the SPIR-V transform code for shader IO and vertex point size. Bug: tint:920 Change-Id: Id1b97517619b4d2fd09b45d5aee848259f3dfa77 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/60840 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: James Price <jrprice@google.com> Auto-Submit: James Price <jrprice@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-08-05 17:34:19 +00:00
%x_GLF_color = OpVariable %_ptr_Private_v4float Private %7
Add tests derived from VK-GL-CTS This adds SPIR-V assembly and WGSL tests derived from VK-GL-CTS commit 571256871c2e2f03995373e1e4a02958d8cd8cf5. The following procedure was followed: - Those .amber files in VK-GL-CTS wholly owned by Google were identified - All GLSL and SPIR-V shaders were extracted from the Amber files and converted into SPIR-V binaries - The compact-ids pass of spirv-opt was applied to each binary - Duplicate binaries were removed - spirv-opt -O was used to obtain an optimized version of each remaining binary, with duplicates discarded - Binaries that failed validation using spirv-val with target environment SPIR-V 1.3 were discarded - Those binaries that tint could not successfully convert into WGSL were put aside for further investigation - SPIR-V assembly versions of the remaining binaries are included in this CL - test-runner with -generate-expected and -generate-skip was used to generate expected .spvasm, .msl, .hlsl and .wgsl outputs for these SPIR-V assembly tests - Each successfully-generated .expected.wgsl is included in this CL again, as a WGLSL test - test-runner with -generate-expected and -generate-skip was used again, to generate expected outputs for these WGSL tests Change-Id: Ibe9baf2729cf97e0b633db9a426f53362a5de540 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/58842 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: Ben Clayton <bclayton@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-07-23 13:10:12 +00:00
%void = OpTypeVoid
%_ptr_Function_int = OpTypePointer Function %int
%23 = OpTypeFunction %void %_ptr_Function_int %_ptr_Function_int
%31 = OpConstantNull %int
%uint_0 = OpConstant %uint 0
%_ptr_Private_int = OpTypePointer Private %int
%49 = OpTypeFunction %int %_ptr_Function_int %_ptr_Function_int
%int_1 = OpConstant %int 1
%bool = OpTypeBool
%109 = OpTypeFunction %void
%_ptr_Function__arr_int_uint_10 = OpTypePointer Function %_arr_int_uint_10
%117 = OpConstantNull %_arr_int_uint_10
%int_0 = OpConstant %int 0
%int_9 = OpConstant %int 9
%int_n1 = OpConstant %int -1
%_ptr_Function_v2float = OpTypePointer Function %v2float
%189 = OpConstantNull %v2float
%v3float = OpTypeVector %float 3
%_ptr_Function_v3float = OpTypePointer Function %v3float
%193 = OpConstantNull %v3float
%int_10 = OpConstant %int 10
%_ptr_Uniform_v2float = OpTypePointer Uniform %v2float
%float_1 = OpConstant %float 1
%float_2 = OpConstant %float 2
%float_3 = OpConstant %float 3
%231 = OpConstantComposite %v3float %float_1 %float_2 %float_3
%_ptr_Function_float = OpTypePointer Function %float
%float_0_25 = OpConstant %float 0.25
%float_0_5 = OpConstant %float 0.5
%int_2 = OpConstant %int 2
%uint_1 = OpConstant %uint 1
%float_0_75 = OpConstant %float 0.75
%int_3 = OpConstant %int 3
%uint_2 = OpConstant %uint 2
%int_4 = OpConstant %int 4
%int_5 = OpConstant %int 5
%int_6 = OpConstant %int 6
%int_7 = OpConstant %int 7
%int_8 = OpConstant %int 8
%main_out = OpTypeStruct %v4float
spirv: Use generic transform to process shader IO The refactored CanonicalizeEntryPointIO transform makes it much easier to handle SPIR-V style IO as well, and doing this removes a lot of duplicated code. Remove all of the SPIR-V transform code for shader IO and vertex point size. Bug: tint:920 Change-Id: Id1b97517619b4d2fd09b45d5aee848259f3dfa77 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/60840 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: James Price <jrprice@google.com> Auto-Submit: James Price <jrprice@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-08-05 17:34:19 +00:00
%361 = OpTypeFunction %main_out %v4float
Add tests derived from VK-GL-CTS This adds SPIR-V assembly and WGSL tests derived from VK-GL-CTS commit 571256871c2e2f03995373e1e4a02958d8cd8cf5. The following procedure was followed: - Those .amber files in VK-GL-CTS wholly owned by Google were identified - All GLSL and SPIR-V shaders were extracted from the Amber files and converted into SPIR-V binaries - The compact-ids pass of spirv-opt was applied to each binary - Duplicate binaries were removed - spirv-opt -O was used to obtain an optimized version of each remaining binary, with duplicates discarded - Binaries that failed validation using spirv-val with target environment SPIR-V 1.3 were discarded - Those binaries that tint could not successfully convert into WGSL were put aside for further investigation - SPIR-V assembly versions of the remaining binaries are included in this CL - test-runner with -generate-expected and -generate-skip was used to generate expected .spvasm, .msl, .hlsl and .wgsl outputs for these SPIR-V assembly tests - Each successfully-generated .expected.wgsl is included in this CL again, as a WGLSL test - test-runner with -generate-expected and -generate-skip was used again, to generate expected outputs for these WGSL tests Change-Id: Ibe9baf2729cf97e0b633db9a426f53362a5de540 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/58842 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: Ben Clayton <bclayton@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-07-23 13:10:12 +00:00
%swap_i1_i1_ = OpFunction %void None %23
%i = OpFunctionParameter %_ptr_Function_int
%j = OpFunctionParameter %_ptr_Function_int
%29 = OpLabel
%temp = OpVariable %_ptr_Function_int Function %31
%33 = OpLoad %int %i
%36 = OpAccessChain %_ptr_Private_int %obj %uint_0 %33
%37 = OpLoad %int %36
OpStore %temp %37
%39 = OpLoad %int %i
%41 = OpLoad %int %j
%42 = OpAccessChain %_ptr_Private_int %obj %uint_0 %41
%43 = OpLoad %int %42
%44 = OpAccessChain %_ptr_Private_int %obj %uint_0 %39
OpStore %44 %43
%46 = OpLoad %int %j
%47 = OpLoad %int %temp
%48 = OpAccessChain %_ptr_Private_int %obj %uint_0 %46
OpStore %48 %47
OpReturn
OpFunctionEnd
%performPartition_i1_i1_ = OpFunction %int None %49
%l = OpFunctionParameter %_ptr_Function_int
%h = OpFunctionParameter %_ptr_Function_int
%53 = OpLabel
%pivot = OpVariable %_ptr_Function_int Function %31
%i_1 = OpVariable %_ptr_Function_int Function %31
%j_1 = OpVariable %_ptr_Function_int Function %31
%param = OpVariable %_ptr_Function_int Function %31
%param_1 = OpVariable %_ptr_Function_int Function %31
%param_2 = OpVariable %_ptr_Function_int Function %31
%param_3 = OpVariable %_ptr_Function_int Function %31
%62 = OpLoad %int %h
%63 = OpAccessChain %_ptr_Private_int %obj %uint_0 %62
%64 = OpLoad %int %63
OpStore %pivot %64
%66 = OpLoad %int %l
%68 = OpISub %int %66 %int_1
OpStore %i_1 %68
%70 = OpLoad %int %l
OpStore %j_1 %70
OpBranch %71
%71 = OpLabel
OpLoopMerge %72 %73 None
OpBranch %74
%74 = OpLabel
%75 = OpLoad %int %j_1
%77 = OpLoad %int %h
%78 = OpISub %int %77 %int_1
%79 = OpSLessThanEqual %bool %75 %78
OpSelectionMerge %81 None
OpBranchConditional %79 %82 %83
%82 = OpLabel
OpBranch %81
%83 = OpLabel
OpBranch %72
%81 = OpLabel
%84 = OpLoad %int %j_1
%85 = OpAccessChain %_ptr_Private_int %obj %uint_0 %84
%86 = OpLoad %int %85
%87 = OpLoad %int %pivot
%88 = OpSLessThanEqual %bool %86 %87
OpSelectionMerge %89 None
OpBranchConditional %88 %90 %89
%90 = OpLabel
%91 = OpLoad %int %i_1
%92 = OpIAdd %int %91 %int_1
OpStore %i_1 %92
%93 = OpLoad %int %i_1
OpStore %param %93
%94 = OpLoad %int %j_1
OpStore %param_1 %94
%95 = OpFunctionCall %void %swap_i1_i1_ %param %param_1
OpBranch %89
%89 = OpLabel
OpBranch %73
%73 = OpLabel
%98 = OpLoad %int %j_1
%99 = OpIAdd %int %98 %int_1
OpStore %j_1 %99
OpBranch %71
%72 = OpLabel
%100 = OpLoad %int %i_1
%101 = OpIAdd %int %100 %int_1
OpStore %i_1 %101
%102 = OpLoad %int %i_1
OpStore %param_2 %102
%104 = OpLoad %int %h
OpStore %param_3 %104
%105 = OpFunctionCall %void %swap_i1_i1_ %param_2 %param_3
%108 = OpLoad %int %i_1
OpReturnValue %108
OpFunctionEnd
%quicksort_ = OpFunction %void None %109
%111 = OpLabel
%l_1 = OpVariable %_ptr_Function_int Function %31
%h_1 = OpVariable %_ptr_Function_int Function %31
%top = OpVariable %_ptr_Function_int Function %31
%stack = OpVariable %_ptr_Function__arr_int_uint_10 Function %117
%p = OpVariable %_ptr_Function_int Function %31
%param_4 = OpVariable %_ptr_Function_int Function %31
%param_5 = OpVariable %_ptr_Function_int Function %31
OpStore %l_1 %int_0
OpStore %h_1 %int_9
OpStore %top %int_n1
%124 = OpLoad %int %top
%125 = OpIAdd %int %124 %int_1
OpStore %top %125
%126 = OpLoad %int %l_1
%127 = OpAccessChain %_ptr_Function_int %stack %125
OpStore %127 %126
%128 = OpLoad %int %top
%129 = OpIAdd %int %128 %int_1
OpStore %top %129
%130 = OpLoad %int %h_1
%131 = OpAccessChain %_ptr_Function_int %stack %129
OpStore %131 %130
OpBranch %132
%132 = OpLabel
OpLoopMerge %133 %134 None
OpBranch %135
%135 = OpLabel
%136 = OpLoad %int %top
%137 = OpSGreaterThanEqual %bool %136 %int_0
OpSelectionMerge %138 None
OpBranchConditional %137 %139 %140
%139 = OpLabel
OpBranch %138
%140 = OpLabel
OpBranch %133
%138 = OpLabel
%141 = OpLoad %int %top
%142 = OpISub %int %141 %int_1
OpStore %top %142
%143 = OpAccessChain %_ptr_Function_int %stack %141
%144 = OpLoad %int %143
OpStore %h_1 %144
%145 = OpLoad %int %top
%146 = OpISub %int %145 %int_1
OpStore %top %146
%147 = OpAccessChain %_ptr_Function_int %stack %145
%148 = OpLoad %int %147
OpStore %l_1 %148
%149 = OpLoad %int %l_1
OpStore %param_4 %149
%150 = OpLoad %int %h_1
OpStore %param_5 %150
%151 = OpFunctionCall %int %performPartition_i1_i1_ %param_4 %param_5
OpStore %p %151
%154 = OpLoad %int %p
%155 = OpLoad %int %l_1
%156 = OpISub %int %154 %int_1
%157 = OpSGreaterThan %bool %156 %155
OpSelectionMerge %158 None
OpBranchConditional %157 %159 %158
%159 = OpLabel
%160 = OpLoad %int %top
%161 = OpIAdd %int %160 %int_1
OpStore %top %161
%162 = OpLoad %int %l_1
%163 = OpAccessChain %_ptr_Function_int %stack %161
OpStore %163 %162
%164 = OpLoad %int %top
%165 = OpIAdd %int %164 %int_1
OpStore %top %165
%166 = OpLoad %int %p
%167 = OpAccessChain %_ptr_Function_int %stack %165
%168 = OpISub %int %166 %int_1
OpStore %167 %168
OpBranch %158
%158 = OpLabel
%169 = OpLoad %int %p
%170 = OpLoad %int %h_1
%171 = OpIAdd %int %169 %int_1
%172 = OpSLessThan %bool %171 %170
OpSelectionMerge %173 None
OpBranchConditional %172 %174 %173
%174 = OpLabel
%175 = OpLoad %int %top
%176 = OpIAdd %int %175 %int_1
OpStore %top %176
%177 = OpLoad %int %p
%178 = OpAccessChain %_ptr_Function_int %stack %176
%179 = OpIAdd %int %177 %int_1
OpStore %178 %179
%180 = OpLoad %int %top
%181 = OpIAdd %int %180 %int_1
OpStore %top %181
%182 = OpLoad %int %h_1
%183 = OpAccessChain %_ptr_Function_int %stack %181
OpStore %183 %182
OpBranch %173
%173 = OpLabel
OpBranch %134
%134 = OpLabel
OpBranch %132
%133 = OpLabel
OpReturn
OpFunctionEnd
%main_1 = OpFunction %void None %109
%185 = OpLabel
%i_2 = OpVariable %_ptr_Function_int Function %31
%uv = OpVariable %_ptr_Function_v2float Function %189
%color = OpVariable %_ptr_Function_v3float Function %193
OpStore %i_2 %int_0
OpBranch %194
%194 = OpLabel
OpLoopMerge %195 %196 None
OpBranch %197
%197 = OpLabel
%198 = OpLoad %int %i_2
%200 = OpSLessThan %bool %198 %int_10
OpSelectionMerge %201 None
OpBranchConditional %200 %202 %203
%202 = OpLabel
OpBranch %201
%203 = OpLabel
OpBranch %195
%201 = OpLabel
%204 = OpLoad %int %i_2
%205 = OpLoad %int %i_2
%206 = OpAccessChain %_ptr_Private_int %obj %uint_0 %204
%207 = OpISub %int %int_10 %205
OpStore %206 %207
%208 = OpLoad %int %i_2
%209 = OpLoad %int %i_2
%210 = OpAccessChain %_ptr_Private_int %obj %uint_0 %209
%211 = OpLoad %int %210
%212 = OpLoad %int %i_2
%213 = OpAccessChain %_ptr_Private_int %obj %uint_0 %212
%214 = OpLoad %int %213
%215 = OpAccessChain %_ptr_Private_int %obj %uint_0 %208
%216 = OpIMul %int %211 %214
OpStore %215 %216
OpBranch %196
%196 = OpLabel
%217 = OpLoad %int %i_2
%218 = OpIAdd %int %217 %int_1
OpStore %i_2 %218
OpBranch %194
%195 = OpLabel
%219 = OpFunctionCall %void %quicksort_
%220 = OpLoad %v4float %gl_FragCoord
%222 = OpAccessChain %_ptr_Uniform_v2float %x_32 %uint_0
%223 = OpLoad %v2float %222
%224 = OpCompositeExtract %float %220 0
%225 = OpCompositeExtract %float %220 1
%226 = OpCompositeConstruct %v2float %224 %225
%227 = OpFDiv %v2float %226 %223
OpStore %uv %227
OpStore %color %231
%232 = OpAccessChain %_ptr_Private_int %obj %uint_0 %int_0
%233 = OpLoad %int %232
%235 = OpAccessChain %_ptr_Function_float %color %uint_0
%236 = OpLoad %float %235
%237 = OpAccessChain %_ptr_Function_float %color %uint_0
%238 = OpConvertSToF %float %233
%239 = OpFAdd %float %236 %238
OpStore %237 %239
%240 = OpAccessChain %_ptr_Function_float %uv %uint_0
%241 = OpLoad %float %240
%243 = OpFOrdGreaterThan %bool %241 %float_0_25
OpSelectionMerge %244 None
OpBranchConditional %243 %245 %244
%245 = OpLabel
%246 = OpAccessChain %_ptr_Private_int %obj %uint_0 %int_1
%247 = OpLoad %int %246
%248 = OpAccessChain %_ptr_Function_float %color %uint_0
%249 = OpLoad %float %248
%250 = OpAccessChain %_ptr_Function_float %color %uint_0
%251 = OpConvertSToF %float %247
%252 = OpFAdd %float %249 %251
OpStore %250 %252
OpBranch %244
%244 = OpLabel
%253 = OpAccessChain %_ptr_Function_float %uv %uint_0
%254 = OpLoad %float %253
%256 = OpFOrdGreaterThan %bool %254 %float_0_5
OpSelectionMerge %257 None
OpBranchConditional %256 %258 %257
%258 = OpLabel
%260 = OpAccessChain %_ptr_Private_int %obj %uint_0 %int_2
%261 = OpLoad %int %260
%263 = OpAccessChain %_ptr_Function_float %color %uint_1
%264 = OpLoad %float %263
%265 = OpAccessChain %_ptr_Function_float %color %uint_1
%266 = OpConvertSToF %float %261
%267 = OpFAdd %float %264 %266
OpStore %265 %267
OpBranch %257
%257 = OpLabel
%268 = OpAccessChain %_ptr_Function_float %uv %uint_0
%269 = OpLoad %float %268
%271 = OpFOrdGreaterThan %bool %269 %float_0_75
OpSelectionMerge %272 None
OpBranchConditional %271 %273 %272
%273 = OpLabel
%275 = OpAccessChain %_ptr_Private_int %obj %uint_0 %int_3
%276 = OpLoad %int %275
%278 = OpAccessChain %_ptr_Function_float %color %uint_2
%279 = OpLoad %float %278
%280 = OpAccessChain %_ptr_Function_float %color %uint_2
%281 = OpConvertSToF %float %276
%282 = OpFAdd %float %279 %281
OpStore %280 %282
OpBranch %272
%272 = OpLabel
%284 = OpAccessChain %_ptr_Private_int %obj %uint_0 %int_4
%285 = OpLoad %int %284
%286 = OpAccessChain %_ptr_Function_float %color %uint_1
%287 = OpLoad %float %286
%288 = OpAccessChain %_ptr_Function_float %color %uint_1
%289 = OpConvertSToF %float %285
%290 = OpFAdd %float %287 %289
OpStore %288 %290
%291 = OpAccessChain %_ptr_Function_float %uv %uint_1
%292 = OpLoad %float %291
%293 = OpFOrdGreaterThan %bool %292 %float_0_25
OpSelectionMerge %294 None
OpBranchConditional %293 %295 %294
%295 = OpLabel
%297 = OpAccessChain %_ptr_Private_int %obj %uint_0 %int_5
%298 = OpLoad %int %297
%299 = OpAccessChain %_ptr_Function_float %color %uint_0
%300 = OpLoad %float %299
%301 = OpAccessChain %_ptr_Function_float %color %uint_0
%302 = OpConvertSToF %float %298
%303 = OpFAdd %float %300 %302
OpStore %301 %303
OpBranch %294
%294 = OpLabel
%304 = OpAccessChain %_ptr_Function_float %uv %uint_1
%305 = OpLoad %float %304
%306 = OpFOrdGreaterThan %bool %305 %float_0_5
OpSelectionMerge %307 None
OpBranchConditional %306 %308 %307
%308 = OpLabel
%310 = OpAccessChain %_ptr_Private_int %obj %uint_0 %int_6
%311 = OpLoad %int %310
%312 = OpAccessChain %_ptr_Function_float %color %uint_1
%313 = OpLoad %float %312
%314 = OpAccessChain %_ptr_Function_float %color %uint_1
%315 = OpConvertSToF %float %311
%316 = OpFAdd %float %313 %315
OpStore %314 %316
OpBranch %307
%307 = OpLabel
%317 = OpAccessChain %_ptr_Function_float %uv %uint_1
%318 = OpLoad %float %317
%319 = OpFOrdGreaterThan %bool %318 %float_0_75
OpSelectionMerge %320 None
OpBranchConditional %319 %321 %320
%321 = OpLabel
%323 = OpAccessChain %_ptr_Private_int %obj %uint_0 %int_7
%324 = OpLoad %int %323
%325 = OpAccessChain %_ptr_Function_float %color %uint_2
%326 = OpLoad %float %325
%327 = OpAccessChain %_ptr_Function_float %color %uint_2
%328 = OpConvertSToF %float %324
%329 = OpFAdd %float %326 %328
OpStore %327 %329
OpBranch %320
%320 = OpLabel
%331 = OpAccessChain %_ptr_Private_int %obj %uint_0 %int_8
%332 = OpLoad %int %331
%333 = OpAccessChain %_ptr_Function_float %color %uint_2
%334 = OpLoad %float %333
%335 = OpAccessChain %_ptr_Function_float %color %uint_2
%336 = OpConvertSToF %float %332
%337 = OpFAdd %float %334 %336
OpStore %335 %337
%338 = OpAccessChain %_ptr_Function_float %uv %uint_0
%339 = OpLoad %float %338
%340 = OpAccessChain %_ptr_Function_float %uv %uint_1
%341 = OpLoad %float %340
%344 = OpFSub %float %339 %341
%342 = OpExtInst %float %343 FAbs %344
%345 = OpFOrdLessThan %bool %342 %float_0_25
OpSelectionMerge %346 None
OpBranchConditional %345 %347 %346
%347 = OpLabel
%348 = OpAccessChain %_ptr_Private_int %obj %uint_0 %int_9
%349 = OpLoad %int %348
%350 = OpAccessChain %_ptr_Function_float %color %uint_0
%351 = OpLoad %float %350
%352 = OpAccessChain %_ptr_Function_float %color %uint_0
%353 = OpConvertSToF %float %349
%354 = OpFAdd %float %351 %353
OpStore %352 %354
OpBranch %346
%346 = OpLabel
%355 = OpLoad %v3float %color
%356 = OpExtInst %v3float %343 Normalize %355
%357 = OpCompositeExtract %float %356 0
%358 = OpCompositeExtract %float %356 1
%359 = OpCompositeExtract %float %356 2
%360 = OpCompositeConstruct %v4float %357 %358 %359 %float_1
OpStore %x_GLF_color %360
OpReturn
OpFunctionEnd
spirv: Use generic transform to process shader IO The refactored CanonicalizeEntryPointIO transform makes it much easier to handle SPIR-V style IO as well, and doing this removes a lot of duplicated code. Remove all of the SPIR-V transform code for shader IO and vertex point size. Bug: tint:920 Change-Id: Id1b97517619b4d2fd09b45d5aee848259f3dfa77 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/60840 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: James Price <jrprice@google.com> Auto-Submit: James Price <jrprice@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-08-05 17:34:19 +00:00
%main_inner = OpFunction %main_out None %361
%gl_FragCoord_param = OpFunctionParameter %v4float
Add tests derived from VK-GL-CTS This adds SPIR-V assembly and WGSL tests derived from VK-GL-CTS commit 571256871c2e2f03995373e1e4a02958d8cd8cf5. The following procedure was followed: - Those .amber files in VK-GL-CTS wholly owned by Google were identified - All GLSL and SPIR-V shaders were extracted from the Amber files and converted into SPIR-V binaries - The compact-ids pass of spirv-opt was applied to each binary - Duplicate binaries were removed - spirv-opt -O was used to obtain an optimized version of each remaining binary, with duplicates discarded - Binaries that failed validation using spirv-val with target environment SPIR-V 1.3 were discarded - Those binaries that tint could not successfully convert into WGSL were put aside for further investigation - SPIR-V assembly versions of the remaining binaries are included in this CL - test-runner with -generate-expected and -generate-skip was used to generate expected .spvasm, .msl, .hlsl and .wgsl outputs for these SPIR-V assembly tests - Each successfully-generated .expected.wgsl is included in this CL again, as a WGLSL test - test-runner with -generate-expected and -generate-skip was used again, to generate expected outputs for these WGSL tests Change-Id: Ibe9baf2729cf97e0b633db9a426f53362a5de540 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/58842 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: Ben Clayton <bclayton@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-07-23 13:10:12 +00:00
%365 = OpLabel
spirv: Use generic transform to process shader IO The refactored CanonicalizeEntryPointIO transform makes it much easier to handle SPIR-V style IO as well, and doing this removes a lot of duplicated code. Remove all of the SPIR-V transform code for shader IO and vertex point size. Bug: tint:920 Change-Id: Id1b97517619b4d2fd09b45d5aee848259f3dfa77 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/60840 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: James Price <jrprice@google.com> Auto-Submit: James Price <jrprice@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-08-05 17:34:19 +00:00
OpStore %gl_FragCoord %gl_FragCoord_param
%366 = OpFunctionCall %void %main_1
%367 = OpLoad %v4float %x_GLF_color
%368 = OpCompositeConstruct %main_out %367
OpReturnValue %368
Add tests derived from VK-GL-CTS This adds SPIR-V assembly and WGSL tests derived from VK-GL-CTS commit 571256871c2e2f03995373e1e4a02958d8cd8cf5. The following procedure was followed: - Those .amber files in VK-GL-CTS wholly owned by Google were identified - All GLSL and SPIR-V shaders were extracted from the Amber files and converted into SPIR-V binaries - The compact-ids pass of spirv-opt was applied to each binary - Duplicate binaries were removed - spirv-opt -O was used to obtain an optimized version of each remaining binary, with duplicates discarded - Binaries that failed validation using spirv-val with target environment SPIR-V 1.3 were discarded - Those binaries that tint could not successfully convert into WGSL were put aside for further investigation - SPIR-V assembly versions of the remaining binaries are included in this CL - test-runner with -generate-expected and -generate-skip was used to generate expected .spvasm, .msl, .hlsl and .wgsl outputs for these SPIR-V assembly tests - Each successfully-generated .expected.wgsl is included in this CL again, as a WGLSL test - test-runner with -generate-expected and -generate-skip was used again, to generate expected outputs for these WGSL tests Change-Id: Ibe9baf2729cf97e0b633db9a426f53362a5de540 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/58842 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: Ben Clayton <bclayton@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-07-23 13:10:12 +00:00
OpFunctionEnd
%main = OpFunction %void None %109
spirv: Use generic transform to process shader IO The refactored CanonicalizeEntryPointIO transform makes it much easier to handle SPIR-V style IO as well, and doing this removes a lot of duplicated code. Remove all of the SPIR-V transform code for shader IO and vertex point size. Bug: tint:920 Change-Id: Id1b97517619b4d2fd09b45d5aee848259f3dfa77 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/60840 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: James Price <jrprice@google.com> Auto-Submit: James Price <jrprice@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-08-05 17:34:19 +00:00
%370 = OpLabel
%372 = OpLoad %v4float %gl_FragCoord_param_1
%371 = OpFunctionCall %main_out %main_inner %372
%373 = OpCompositeExtract %v4float %371 0
OpStore %x_GLF_color_1_1 %373
Add tests derived from VK-GL-CTS This adds SPIR-V assembly and WGSL tests derived from VK-GL-CTS commit 571256871c2e2f03995373e1e4a02958d8cd8cf5. The following procedure was followed: - Those .amber files in VK-GL-CTS wholly owned by Google were identified - All GLSL and SPIR-V shaders were extracted from the Amber files and converted into SPIR-V binaries - The compact-ids pass of spirv-opt was applied to each binary - Duplicate binaries were removed - spirv-opt -O was used to obtain an optimized version of each remaining binary, with duplicates discarded - Binaries that failed validation using spirv-val with target environment SPIR-V 1.3 were discarded - Those binaries that tint could not successfully convert into WGSL were put aside for further investigation - SPIR-V assembly versions of the remaining binaries are included in this CL - test-runner with -generate-expected and -generate-skip was used to generate expected .spvasm, .msl, .hlsl and .wgsl outputs for these SPIR-V assembly tests - Each successfully-generated .expected.wgsl is included in this CL again, as a WGLSL test - test-runner with -generate-expected and -generate-skip was used again, to generate expected outputs for these WGSL tests Change-Id: Ibe9baf2729cf97e0b633db9a426f53362a5de540 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/58842 Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: Ben Clayton <bclayton@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
2021-07-23 13:10:12 +00:00
OpReturn
OpFunctionEnd