| Age | Commit message (Collapse) | Author |
|---|
|
When copying a D24_UNORM_S8_UINT depth/stencil buffer to or from the host, the depth and stencil data must be copied separately. This change adds extra space to the end of the Amber host buffer for the stencil data.
|
|
There's no need to restrict binding a vertex buffer to multiple locations. Now that we have OFFSET supported this has a real use case where the same buffer is used with different offsets.
|
|
This CL adds a DATA initializer to the IMAGE command. This allows
setting the data to be used in the image as part of the initializer.
|
|
Adds parameters to the VERTEX_DATA AmberScript command: FORMAT, OFFSET, STRIDE.
|
|
* Various fixes for size comparisons
This CL fixes up various size comparison issues in amber along with
running the formatter.
* Roll Deps
Roll third_party/dxc/ c607bb3c0..f1f60648d (23 commits)
https://github.com/Microsoft/DirectXShaderCompiler/compare/c607bb3c032d...f1f60648df03
$ git log c607bb3c0..f1f60648d --date=short --no-merges --format='%ad %ae %s'
2020-11-12 hekotas Fix hctbuild to enable passing of quoted values and values with parenthesis (#3256)
2020-11-12 marijns95 Linux: Implement prefix-counted BSTR allocation in SysAllocStringLen (#3250)
2020-11-11 hekotas Fix dxilconv crash when initialization fails (#3254)
2020-11-10 grroth Handle scalar args to out vector params (#3246)
2020-11-09 jaebaek [spirv] emit OpLine at the end of entry function (#3244)
2020-11-06 hekotas Support for ARM64EC and custom cmake (#3245)
2020-11-05 hekotas Minor refactoring to reduce size of dxilconv.dll (#3240)
2020-11-04 texr Fix lost debug info on DispatchMesh payload in AS (#3243)
2020-11-04 jaebaek Revert "[spirv] add location for DebugDeclare (#3230)" (#3242)
2020-11-04 grroth Initialize conversion remarks result (#3241)
2020-11-03 31109344+adam-yang Made exit values not have to dominate latch in structurize loop exit (#3220)
2020-11-03 hekotas Fix DIA SDK detection for ARM64 and VS2017 (#3239)
2020-11-02 grroth New approach to avoid skipallocas (#3219)
2020-11-02 jaebaek [spirv] add location for DebugDeclare (#3230)
2020-11-02 jaebaek [spirv] wrap instructions with OpNoLine (#3229)
2020-11-02 hekotas Fix DIA SDK detection (#3231)
2020-11-02 Junda.Liu [spirv] Fix [[vk::push_constant]] on ConstantBuffer part II (#3216)
2020-11-02 Junda.Liu [spirv] globallycoherent should generate SPIR-V decoration Coherent (#3217)
2020-10-30 texr Fix precise propagation through various missing paths (#3224)
2020-10-30 texr Fix regression in lib vector arg case when skipping copy-out (#3232)
2020-10-30 31109344+adam-yang Bitpiece crash when value/addr are null (#3227)
2020-10-29 hekotas Add caps check for optional 16bit ops support to several execution tests. (#3222)
2020-10-28 31109344+adam-yang Make sure bitpiece not generated that covers the entire variable (#3223)
Created with:
roll-dep third_party/dxc
Roll third_party/clspv/ 3a11614ee..e0406e705 (22 commits)
https://github.com/google/clspv/compare/3a11614ee409...e0406e7053d1
$ git log 3a11614ee..e0406e705 --date=short --no-merges --format='%ad %ae %s'
2020-11-10 alanbaker Implement fdim (#675)
2020-11-10 marco.antognini Lower OpenCL vloadN and vstoreN (#670)
2020-11-09 alanbaker Implement isnormal (#674)
2020-11-07 alanbaker Implement islessgreater (#672)
2020-11-06 alanbaker Implement isordered and isunordered (#671)
2020-11-05 alanbaker Implement ctz (#669)
2020-11-05 alanbaker Implement sub_sat (#668)
2020-11-03 alanbaker Implement mad_sat (#667)
2020-11-03 marco.antognini Lower function call and initial support for LLVM intrinsics (#666)
2020-11-02 kpet Fix hadd/rhadd for signed types (#665)
2020-11-02 alanbaker Fix non-32-bit clz (#664)
2020-10-30 marco.antognini Lower UnaryOperator and handle fast math flags (#663)
2020-10-29 alanbaker Implement rotate in terms of llvm.fshl (#662)
2020-10-29 alanbaker Implement hadd and rhadd (#660)
2020-10-29 alanbaker Add support for OpenCL 2.0 atomic functions (#657)
2020-10-29 alanbaker Update deps (#661)
2020-10-29 marco.antognini Recusiverly lower instructions (#655)
2020-10-28 alanbaker Add convergent attribute in more cases (#658)
2020-10-26 alanbaker Implement add_sat (#656)
2020-10-22 alanbaker Prevent duplicate types and decorations (#654)
2020-10-19 kpet Add support for -cl-std=CL3.0 (#653)
2020-10-19 marco.antognini Lower function parameters and return type (#646)
Created with:
roll-dep third_party/clspv
Roll third_party/shaderc/ 813ef3dc0..88f9156d7 (3 commits)
https://github.com/google/shaderc/compare/813ef3dc0d5e...88f9156d7f6a
$ git log 813ef3dc0..88f9156d7 --date=short --no-merges --format='%ad %ae %s'
2020-11-09 dneto Replace C-style cast with static_cast
2020-11-12 dneto Roll back googletest to earlier version
2020-11-02 rharrison Roll 6 dependencies (#1144)
Created with:
roll-dep third_party/shaderc
Roll third_party/spirv-tools/ df4198e50..1f2fcddd3 (21 commits)
https://github.com/KhronosGroup/SPIRV-Tools/compare/df4198e50eaa...1f2fcddd3963
$ git log df4198e50..1f2fcddd3 --date=short --no-merges --format='%ad %ae %s'
2020-11-13 afdx spirv-opt: Set parent when adding basic block (#4021)
2020-11-13 jaebaek spirv-opt: properly preserve DebugValue indexes operand (#4022)
2020-11-11 dneto Use less stack space when validating Vulkan builtins (#4019)
2020-11-05 46493288+sfricke-samsung spirv-val: Fix SPV_KHR_fragment_shading_rate VUID label (#4014)
2020-11-05 46493288+sfricke-samsung spirv-val: Label Layer and ViewportIndex VUIDs (#4013)
2020-11-05 alanbaker Add dead function elimination to -O (#4015)
2020-11-04 jaebaek Add DebugValue for invisible store in single_store_elim (#4002)
2020-11-04 dnovillo Fix SSA re-writing in the presence of variable pointers. (#4010)
2020-11-04 afdx spirv-fuzz: Fixes to pass management (#4011)
2020-11-03 afdx spirv-fuzz: Add support for reining in rogue fuzzer passes (#3987)
2020-11-03 vasniktel spirv-fuzz: Fix assertion failure in FuzzerPassAddCompositeExtract (#3995)
2020-11-03 vasniktel spirv-fuzz: Fix invalid equation facts (#4009)
2020-11-03 vasniktel spirv-fuzz: Fix bugs in TransformationFlattenConditionalBranch (#4006)
2020-11-03 andreperezmaselco.developer spirv-fuzz: Fix bug related to transformation applicability (#3990)
2020-10-30 jaebaek Temporarily add EmptyPass to prevent glslang from failing (#4004)
2020-10-30 Junda.Liu spirv-opt: Add support to prevent functions from being inlined if they have DontInline flag (#3858)
2020-10-29 jaebaek Propagate OpLine to all applied instructions in spirv-opt (#3951)
2020-10-29 bclayton CMake: Add SPIRV_TOOLS_BUILD_STATIC flag (#3910)
2020-10-29 dneto Avoid copying a ref in a loop (#4000)
2020-10-28 justsid spirv-val: Allow the ViewportIndex and Layer built-ins on SPIR-V 1.5 (#3986)
2020-10-28 dnovillo Simplify logic to decide whether CCP modified the IR (#3997)
Created with:
roll-dep third_party/spirv-tools
Roll third_party/glslang/ 740ae9f60..beec2e4a7 (17 commits)
https://github.com/KhronosGroup/glslang/compare/740ae9f60b00...beec2e4a7c4d
$ git log 740ae9f60..beec2e4a7 --date=short --no-merges --format='%ad %ae %s'
2020-11-12 mbechard tweak local_size comparison a bit (#2456)
2020-11-12 dneto Avoid spuriously adding Geometry capability for vert, tesc, tese (#2462)
2020-11-12 greg New nonuniform analysis (#2457)
2020-11-09 jhall1024 Implement GL_EXT_terminate_invocation (#2454)
2020-11-06 rdb Fix token-pasting macros not working in preprocessor directives. (#2453)
2020-11-06 laddoc Fix warning in iomapper. (#2449)
2020-11-04 TobyHector Add GL_EXT_shader_image_int64 support (#2409)
2020-11-04 laddoc 8. io mapping refine & qualifier member check & resolver expand (#2396)
2020-11-02 courtneygo Fix build error with Chromium & ANGLE (#2446)
2020-11-02 dev Add new SpirvToolsDisassemble API interface + Improve Doc on existing API interface (#2442)
2020-11-02 justsid Support for CapabilityShaderViewportIndex and CapabilityShaderLayer (#2432)
2020-11-02 jaebaek Do not use PropagateLineInfoPass and RedundantLineInfoElimPass (#2440)
2020-09-09 greg Add texture sample to nonuniform test
2020-09-08 greg Add buffer store to nonuniform tests
2020-09-03 bas SPV: Add NonUniform decoration for constructors.
2020-09-03 bas SPV: Add NonUniform decoration for OpImages created during lowering.
2020-08-24 bas SPV: Add NonUniform decorations for stores.
Created with:
roll-dep third_party/glslang
Roll third_party/spirv-headers/ 7845730ca..5ab5c9619 (2 commits)
https://github.com/KhronosGroup/SPIRV-Headers/compare/7845730cab6e...5ab5c96198f3
$ git log 7845730ca..5ab5c9619 --date=short --no-merges --format='%ad %ae %s'
2020-11-04 michael.kinsner Reserve additional loop control bit for Intel extension (NoFusionINTEL) (#175)
2020-11-02 4464295+XAMPPRocky Add EmbarkStudios/rust-gpu to vendor list. (#174)
Created with:
roll-dep third_party/spirv-headers
Roll third_party/vulkan-loader/ 2b0892e15..be6ccb9ec (7 commits)
https://github.com/KhronosGroup/Vulkan-Loader/compare/2b0892e15bd8...be6ccb9ecaf7
$ git log 2b0892e15..be6ccb9ec --date=short --no-merges --format='%ad %ae %s'
2020-11-02 shabbyx Reland "Generate libvulkan.so.1 on Linux instead of libvulkan.so"
2020-11-10 shannon build: Update known-good files for 1.2.160 header
2020-11-02 shannon build: Update known-good files for 1.2.159 header
2020-10-30 shabbyx Revert "Generate libvulkan.so.1 on Linux instead of libvulkan.so"
2020-10-22 xantares09 Add an option to disable MASM
2020-10-20 xantares09 Configure vulkan.pc when pkg-config is available
2020-10-22 shabbyx Generate libvulkan.so.1 on Linux instead of libvulkan.so
Created with:
roll-dep third_party/vulkan-loader
Roll third_party/clspv-llvm/ b91a236ee..a6ac2b32f (2881 commits)
https://github.com/llvm/llvm-project/compare/b91a236ee1c3...a6ac2b32fbab
$ git log b91a236ee..a6ac2b32f --date=short --no-merges --format='%ad %ae %s'
2020-11-16 schuett Convert TypeSpecifiersPipe from Specifiers.h to a scoped enum; NFC
2020-11-16 v.g.vassilev Reland "Move the test compiler setup in a common place. NFCI"
2020-11-16 dmitry.preobrazhensky [AMDGPU][MC] Corrected error position for some operands and modifiers
2020-11-16 mkazantsev [Test] More tests on range checks
2020-11-11 caroline.concatto [AArch64] Add check for widening instruction for SVE.
2020-11-12 matthew.malcomson Hwasan reporting check for dladdr failing
2020-11-16 vitalybuka Revert "Move the test compiler setup in a common place. NFCI"
2020-11-16 vitalybuka [NFC][tsan] Simplify call_pthread_cancel_with_cleanup
2020-11-16 mkazantsev [Test] More tests on range check elimination
2020-11-13 mgorny [lldb] [Process/FreeBSDRemote] Optimize regset pointer logic
2020-11-11 mgorny [lldb] [Process/FreeBSDRemote] Modernize and simplify YMM logic
2020-11-11 mgorny [lldb] [Process/FreeBSDRemote] Access debug registers via offsets
2020-11-11 mgorny [lldb] [Process/FreeBSDRemote] Access FPR via RegisterInfo offsets
2020-11-16 dmitry.preobrazhensky [AMDGPU][MC] Corrected error position for swizzle()
2020-11-16 dmitry.preobrazhensky [AMDGPU][MC] Corrected error position for hwreg() and sendmsg()
2020-11-10 mgorny [lldb] [Process/FreeBSDRemote] Access GPR via reginfo offsets
2020-11-16 nicolas.vasilache [mlir][Linalg] Add support for tileAndDistribute on tensors.
2020-11-13 marukawa [VE] Add pfchv intrinsic instructions
2020-11-16 vitalybuka [sanitizer] Fix StackDepotPrint testing
2020-11-16 benny.kra [LoopFlatten] Fold variable into assert. NFC.
2020-11-16 flo [IRGen] Add !annotation metadata for auto-init stores.
2020-11-09 sjoerd.meijer [LoopFlatten] Widen the IV
2020-11-16 david.penry [ARM] Cortex-M7 schedule
2020-11-16 goncharov.mikhail [lld] Use %t file in test
2020-11-02 protze [OpenMP][Tool] Add Archer option to disable data race analysis for sequential part
2020-11-13 fraser [RISCV] Use a macro to simplify getTargetNodeName
2020-11-16 marukawa [VE] Change variable capitalization
2020-11-16 simon.moll [VP][NFC] Rename to HANDLE_VP_TO_OPC
2020-11-13 git [clang][cli] Add support for options with two flags for controlling the same field.
2020-11-16 lhames [MCJIT] Profile the code generated by MCJIT engine using Intel VTune profiler
2020-11-16 simon.moll [VE] LVLGen sets VL before vector insts
2020-11-16 mkazantsev Recommit "[NFC] Move code between functions as a preparation step for further improvement"
2020-11-15 kazu [JumpThreading] Call eraseBlock when folding a conditional branch
2020-11-15 kazu [IR] Use llvm::is_contained in BasicBlock::removePredecessor (NFC)
2020-11-15 kazu [Loop Fusion] Use pred_empty and succ_empty (NFC)
2020-11-15 kazu [ADCE] Use succ_empty (NFC)
2020-11-16 qshanz [NFC][Test] Add test coverage for IEEE Long Double on Power8
2020-11-15 kazu [BranchProbabilityInfo] Use predecessors(BB) and successors(BB) (NFC)
2020-11-15 kazu [TRE] Use successors(BB) (NFC)
2020-11-15 craig.topper [X86] Fix crash with i64 bitreverse on 32-bit targets with XOP.
2020-11-15 kazu [SanitizerCoverage] Use llvm::all_of (NFC)
2020-11-10 serguei.katkov [IRCE] Use the same min runtime iteration threshold for BPI and BFI checks
2020-11-16 pengfei.wang [CodeGen][X86] Remove unused trivial check-prefixes from all CodeGen/X86 directory.
2020-11-15 richard PR48169: Fix crash generating debug info for class non-type template parameters.
2020-11-15 spatel [InstCombine] reduce code for flip of masked bit; NFC
2020-11-15 spatel [InstCombine] add multi-use demanded bits fold for add with low-bit mask
2020-11-15 spatel [InstCombine] add vector tests for multi-use demanded bits; NFC
2020-11-14 martin [OpenMP] Fix building for windows after adding omp_calloc
2020-11-15 nikita.ppv [AA] Add missing AAQI parameter
2020-11-15 flo [MemorySSA] Add pointer decrement loop clobber test case.
(...)
2020-10-19 martin [libcxx] [test] Mark tests that require specific allocation behaviours as libcpp only
2020-10-19 martin [libcxx] [test] Exclude domain socket tests on windows, like bsd/darwin
2020-10-19 martin [libcxx] [test] Add LIBCPP_ONLY() around another test for an implementation detail
2020-10-19 martin [libcxx] [test] Don't require fs::path::operator(string_type&&) to be noexcept
2020-10-19 martin [libcxx] [test] Allow fs::permissions(path, perms, perm_options, error_code) to be noexcept
2020-10-19 martin [libcxx] [test] Do error printfs to stderr in filesystems tests
2020-10-19 martin [clang] Automatically link against oldnames just as linking against libcmt
2020-10-14 dexonsmith clang/{Format,Rewrite}: Stop using SourceManager::getBuffer, NFC
2020-10-15 eugenis Add alloca size threshold for StackTagging initializer merging.
2020-10-19 aeubanks [test][NPM] Fix already-vectorized.ll under NPM
2020-10-19 craig.topper [SelectionDAG][X86] Enable SimplifySetCC CTPOP transforms for vector splats
2020-10-19 craig.topper [X86] Add i32->float and i64->double bitcast pseudo instructions to store folding table.
2020-10-17 Matthew.Arsenault Fix typo
2020-10-19 lhames [ORC] Fix a missing include.
2020-10-19 aeubanks [NPM][test] Fix some LoopVectorize tests under NPM
2020-10-19 nikita.ppv [BatchAA] Add test for incorrect phi cycle result (NFC)
2020-10-19 Tony.Tye [AMDGPU] Correct comment typo in SIMemoryLegaliizer.cpp
2020-10-19 aeubanks [NPM][opt] Run -O# after other passes in legacy PM compatibility mode
2020-10-19 clementval [flang][openacc] Switch to use TODO from D88909
2020-10-19 alexandre.ganea [LLDB][TestSuite] Improve skipIfWindowsAndNonEnglish in decorators.py
2020-10-19 mcinally [SVE] Update vector reduction intrinsics in new tests.
2020-10-13 michaelrj [libc] Add LLVM libc specific functions to llvm_libc_ext.td.
2020-10-19 jay.foad [AMDGPU] Remove MUL_LOHI_U24/MUL_LOHI_I24
2020-10-19 jhuber6 [OpenMP] Fixing OpenMP/driver.c failing on 32-bit hosts
2020-10-19 atmndp [LangRef] Define mustprogress attribute
2020-10-19 mikhail.maltsev [clang] Use SourceLocation as key in std::map, NFCI
2020-10-19 flo [SCEV] Add tests where assumes can be used to improve tripe multiple.
2020-10-16 ldionne [libc++] Add more tests for operator<< on std::complex
2020-10-15 amy.kwan1 [DAGCombiner][PowerPC] Remove isMulhCheaperThanMulShift TLI hook, Use isOperationLegalOrCustom directly instead.
2020-10-19 jonas [llvm] Make obj2yaml and yaml2obj LLVM utilities instead of tools
2020-10-19 Tony.Tye [AMDGPU] Simplify cumode handling in SIMemoryLegalizer
2020-10-17 Tony.Tye [AMDGPU] Extend hip-toolchin-features.hip test
2020-10-17 Tony.Tye [AMDGPU] Correct hsa-diag-v3.s test
2020-10-19 mtrofin [NFC][MC] Type [MC]Register uses in MachineTraceMetrics
2020-10-19 Alexander.Richardson [libc++] Skip tests using constexpr destructors with older clang versions
2020-10-06 sameeranjayant.joshi [Flang][OpenACC] Fix for branching out issue in OpenACC parallel construct.
2020-10-16 ldionne [libc++] Refactor the fuzzing tests
2020-10-19 lhames [ORC] Fix unused variable warning.
2020-10-19 llvm-dev [InstCombine] foldOrOfICmps - only fold (icmp_eq B, 0) | (icmp_ult/gt A, B) for integer types
2020-10-17 mtrofin [ML] Add final reward logging facility.
2020-10-19 nabeel.omer [Dexter][NFC] Add Missing Commands to Commands.md Contents
2019-10-01 ldionne [libc++] Define new/delete in libc++abi only by default
2020-10-19 Casey [libc++][test] Portably silence warnings
2020-10-19 Alexander.Richardson [lit] Avoid calling realpath() for every printed message
2020-10-19 llvm-dev Revert rGa704d8238c86bac: "[InstCombine] Add or((icmp ult/ule (A + C1), C3), (icmp ult/ule (A + C2), C3)) uniform vector support"
2020-10-19 n.james93 [clang-tidy] Add scoped enum constants to identifier naming check
2020-10-16 psteinfeld [flang] Add name match checks for interface procedures
2020-10-19 sam.mccall [clangd] Add test for structured-binding completion. NFC
2020-10-19 georgemitenk0v [MLIR][SPIRVToLLVM] Updated documentation for composite ops
2020-10-19 llvm-dev [InstCombine] Add (icmp ne A, 0) | (icmp ne B, 0) --> (icmp ne (A|B), 0) vector support
Created with:
roll-dep third_party/clspv-llvm
Roll third_party/swiftshader/ df17a7610..e4c1a25cc (56 commits)
https://swiftshader.googlesource.com/SwiftShader.git/+log/df17a76102df..e4c1a25cc679
$ git log df17a7610..e4c1a25cc --date=short --no-merges --format='%ad %ae %s'
2020-11-12 capn Use LLVM JIT for MemorySanitizer GN builds
2020-11-12 srisser Update vulkan headers to 1.2.160
2020-11-12 swiftshader.regress Regres: Update test lists @ 4804ac87
2020-11-11 zzyiwei Disable VK_KHR_external_memory_fd for Android build
2020-10-28 natsu Handle image properties from external device memory
2020-11-12 capn Fix conditionally enabling Reactor MSan instrumentation
2020-11-02 capn Implement MemorySanitizer instrumentation of Reactor routines
2020-08-25 nicolascapens Fix backslash interpreted as part of URL
2020-10-26 srisser Implement VK_KHR_shader_float_controls
2020-11-10 amaiorano Subzero: clang-format crosstest, pnacl-llvm, runtime, and unittest
2020-11-10 amaiorano Subzero: clang-format src
2020-11-02 capn Fix LLVM host triple for ORCv2 JIT
2020-11-11 srisser Add SPIR-V tools dependency to BUILD.gn files
2020-11-10 natsu Fix Android build
2020-11-06 amaiorano Add a unit test for bug fixed by swiftshader-cl/50008
2020-11-09 srisser Track SPIR-V version in VkConfig.hpp
2019-04-24 vhau Modify gralloc0 implementation
2020-11-05 jaebaek SpirvShaderDebugger: Correct member index for DebugValue
2020-11-05 capn Fix Android.bp build after SPIRV-Tools update
2020-11-04 amaiorano Subzero: fix invalid arg access on Windows x86 for split variables
2020-11-03 amaiorano Subzero: fix non-deterministic stack layout and code-gen
2020-11-05 srisser Squashed 'third_party/SPIRV-Tools/' changes from f7da52775..a61d07a72
2020-11-04 capn Fix Vulkan benchmarks build
2020-10-02 capn Fix unaligned access on depth values
2020-11-02 capn Fix LLVM alloca array size type
2020-11-02 capn Resolve MSan symbols dynamically
2020-11-02 capn Switch Chromium/Fuchsia to use the LLVM ORCv2 JIT
2020-11-03 nicolascapens Clarify the ICD/loader compatibility
2020-11-03 swiftshader.regress Regres: Update test lists @ b5bf8269
2020-11-03 capn Set LLVM_ENABLE_ABI_BREAKING_CHECKS for Debug builds only
2020-10-30 jari.komppa Add script to verify proper Memset template use
2020-11-02 capn Set the LLVM module's target triple
2020-10-30 capn Consistently remove double colon from llvm namespace
2020-10-30 capn Emulate gather/scatter for MSan builds
2020-10-30 capn Prevent 'most vexing parse' issue
2020-11-02 sugoi Squashed 'third_party/SPIRV-Tools/' changes from 82b378d671..f7da527757
2020-10-23 amaiorano Subzero: fix Ice::BitVector::grow not copying old to new data
2020-10-30 jari.komppa Ignore third party libraries during presubmit
2020-10-28 jari.komppa Add check for vtable in memset template
2020-10-29 sugoi Squashed 'third_party/SPIRV-Tools/' changes from 8a5500656..82b378d67
2020-10-29 sugoi Squashed 'third_party/SPIRV-Headers/' changes from 979924c8b..7845730ca
2020-10-23 capn Fix LLVM pointer element type
2020-10-29 capn Avoid false-positive MSan error for thread_local
2020-10-29 capn Squashed 'third_party/marl/' changes from f1c446ccd..3285a2dfb
2020-10-28 capn Temporarily support both ORCv1 and ORCv2
2020-10-26 amaiorano Apply google3 upstream fixes for latest LLVM
2020-10-22 amaiorano LLVMJIT: fix asserts in latest LLVM
2020-10-08 capn Support MemorySanitizer builds with CMake
2020-10-15 srisser Implement VK_KHR_shader_subgroup_extended_types
2020-10-19 srisser Update VkStringify to use vulkan.hpp to_string
2020-10-21 bclayton LLVMJIT: Fix memory leak
2020-10-10 natsu Update SwiftShader VK AHB to support generic gralloc
2020-10-21 amaiorano Fix when built against latest LLVM (11)
2020-10-20 amaiorano Fix Android build
2020-10-19 amaiorano Subzero: workaround bad codegen for Windows x86 at Om1 optimization level
2020-10-19 bclayton LLVM: Replace legacy ORC JIT with new API
Created with:
roll-dep third_party/swiftshader
Roll third_party/vulkan-headers/ 320af06cb..11c6670b4 (3 commits)
https://github.com/KhronosGroup/Vulkan-Headers/compare/320af06cbdd2...11c6670b4a4f
$ git log 320af06cb..11c6670b4 --date=short --no-merges --format='%ad %ae %s'
2020-11-15 oddhack Update for Vulkan-Docs 1.2.161
2020-11-08 oddhack Update for Vulkan-Docs 1.2.160
2020-11-01 oddhack Update for Vulkan-Docs 1.2.159
Created with:
roll-dep third_party/vulkan-headers
Roll third_party/vulkan-validationlayers/ 87b0951f8..9d3ef3258 (70 commits)
https://github.com/KhronosGroup/Vulkan-ValidationLayers/compare/87b0951f8dba...9d3ef3258715
$ git log 87b0951f8..9d3ef3258 --date=short --no-merges --format='%ad %ae %s'
2020-11-13 pankratz layers: Removed duplicate type declaration
2020-11-13 mark scripts: Suppress warning for empty difflist
2020-11-12 mark scripts: Remove format checking shell scripts
2020-11-11 mark scripts: Add python format-checking script and update travis
2020-11-05 mark tests: Add oldSwapchain-from-different-device test
2020-11-09 mark threading: Make swapchain objects traceable to parent
2020-11-10 mark tests: Added specialuse extension BP tests
2020-11-10 mark practices: Add checking of specialuse extensions
2020-11-09 mark practices: Check for specialuse extensions and warn
2020-11-09 mark practices: Added specialuse extension table
2020-11-10 karl layers: Fix typedef and spelling
2020-11-09 jeremyg syncval: Convert enum SyncStageAccessFlags to a std::bitset
2020-11-09 jeremyg syncval: Change generated lookup tables to be const.
2020-11-06 marshall tests: Test queue retirement over timeline semaphores
2020-11-06 marshall layers: Fix queue retirement via waited-on timeline semaphores
2020-11-06 rgarcia layers: Check nullDescriptor when validating vertex attributes
2020-11-02 shannon build: Update known-good files for 1.2.159 header
2020-11-05 orbea cmake: Install the header files into the vulkan directory
2020-11-04 liyl build: Allow generating layers json file for cross-compiling
2020-11-05 courtneygo Fix missing source reference
2020-10-31 s.fricke layers: Change copy2 VU from int to string
2020-11-04 orbea cmake: Silence warnings
2020-11-03 orbea cmake: Add BUILD_WERROR
2020-11-01 s.fricke layers: Add VUID 01591
2020-11-01 s.fricke tests: Add VUID 04113 and 04114
2020-11-01 s.fricke layers: Add VUID 04113 and 04114
2020-11-01 s.fricke layers: Label 04101 and 04010
2020-11-03 s.fricke layers: Add Android 24 and 25 support
2020-11-03 tony docs: Fix enabling debug printf
2020-10-31 skdkksfadkadkk layers: Fix checks for 01508 and 01509
2020-11-03 neel.bedekar layers: Remove unnecessary FDM tiling check
2020-11-02 mark tests: Add 2718 security VUID tests
2020-10-30 mark stateless: Fix clang-formatting issues
2020-10-30 mark stateless: Comment out unused legacy vuids
2020-10-30 mark stateless: Output correct VUID for validate_string_array
2020-10-30 mark stateless: Remove non-VU pNext chain cycle check
2020-10-30 mark stateless: Remove warnings from viewport validation
2020-10-30 mark stateless: Use correct VUIDs for drawCount check
2020-10-29 jzulauf syncval: Add raster order rules to load op validation
2020-10-23 jzulauf syncval: Add layout transition info w/o prev access
2020-10-22 jzulauf syncval: Remove unsafe/unused code path.
2020-10-21 jzulauf syncval: Support range gen w/ empty subres ranges
2020-10-19 jzulauf syncval: Add dst scope barriers to all ILT
2020-10-06 jzulauf tests: Add multi dependency syncval test
2020-10-06 jzulauf syncval: Multidep subpass layout transition rework
2020-10-02 jzulauf syncval: Simplify input attachment read tracking
2020-09-29 jzulauf syncval: vkCmdPipelineBarrier multidep support
2020-09-23 jzulauf syncval: Cleanup input attachment read tracking
2020-10-29 tony layers: Fix checks for 02502 and 02503
2020-10-27 tony gpu: Report OpLine info even if no source
2020-10-27 tony tests: Add test for variable count descriptors
2020-10-27 tony layers: Add validation for variable descriptor count
2020-10-27 tony layers: Fix error message
2020-10-28 jeremyg syncval: Add override keyword to ValidationObject method overrides.
2020-10-26 s.fricke layers: Relabel vkCmdDrawIndirectByteCountEXT VUs
2020-10-30 nathaniel layers: Fix test error introduced by #2276
2020-10-29 s.fricke layers: Add extra api names to errors
2020-10-28 s.fricke tests: Add SampleMask Array Limit VUID
2020-10-26 s.fricke layers: Add SampleMask Array Limit VUID
2020-10-24 s.fricke layers: Add table for buffer image copy VUs
2020-10-29 s.fricke layers: Add VK_EXT_4444_formats support
2020-10-30 mark contributing: Updated for Travis CI migration
2020-10-30 mark readme: Updated for Travis migration
2020-10-29 s.fricke layers: Add missing 1.2 overload functions
2020-10-21 s.fricke tests: Add input attachment aspectMask/layout
2020-10-20 s.fricke layers: Remove non-input aspectMask check
2020-10-28 shannon build: Reconcile final 158 VUID inconsistencies
2020-10-21 mark tests: Add invalid stride test for GetQueryPoolResults
2020-10-21 mark corechecks: Added 4519 vuid for GetQueryPoolResults stride
2020-10-28 krzysio.kurek stateless: Add a missing "not" to ValidateGreaterThan
Created with:
roll-dep third_party/vulkan-validationlayers
* Roll deps
Roll third_party/dxc/ f1f60648d..489c2e4d3 (6 commits)
https://github.com/Microsoft/DirectXShaderCompiler/compare/f1f60648df03...489c2e4d3241
$ git log f1f60648d..489c2e4d3 --date=short --no-merges --format='%ad %ae %s'
2020-11-17 rkarrenberg Move force-zero-store-lifetimes flag to help-hidden, remove accidental check from test. (#3264)
2020-11-17 marijns95 dxcisense: Allocate "TM" classes using IMalloc instead of new (#3258)
2020-11-17 rkarrenberg Enable generation of llvm.lifetime.start/.end intrinsics (#3034)
2020-11-16 ehsannas [spirv] Fix compound MulAssign with type mismatches. (#3255)
2020-11-16 ehsannas [spirv] Use isResourceType to cover SubpassInput. (#3253)
2020-11-16 ehsannas [spirv] Move descriptor scalar replacement to legalization stage. (#3261)
Created with:
roll-dep third_party/dxc
Roll third_party/glslang/ beec2e4a7..7f6559d28 (1 commit)
https://github.com/KhronosGroup/glslang/compare/beec2e4a7c4d...7f6559d2802d
$ git log beec2e4a7..7f6559d28 --date=short --no-merges --format='%ad %ae %s'
2020-11-16 ShabbyX Compile out code for GL_EXT_shader_image_int64 for ANGLE (#2463)
Created with:
roll-dep third_party/glslang
Roll third_party/clspv-llvm/ a6ac2b32f..7e30989da (134 commits)
https://github.com/llvm/llvm-project/compare/a6ac2b32fbab...7e30989dabce
$ git log a6ac2b32f..7e30989da --date=short --no-merges --format='%ad %ae %s'
2020-11-17 llvm-dev [IR] ShuffleVectorInst::isIdentityWithPadding - bail on non-fixed-type vector shuffles.
2020-11-17 llvm-dev [X86] nontemporal.ll - replace X32 check prefix with X86. NFC.
2020-11-17 llvm-dev [X86] subvector-broadcast.ll - replace X32 check prefix with X86. NFC.
2020-11-17 jonathan.l.peyton [OpenMP] NFC: remove tabs in message catalog file
2020-11-17 wingo [WebAssembly] Fix parsing of linking section for named global imports
2020-11-05 jonathan.l.peyton [OpenMP][stats] reset serial state when re-entering serial region
2020-11-17 Matthew.Arsenault Linker: Fix linking of byref types
2020-11-16 jay.foad [AMDGPU] Precommit more vccz workaround tests
2020-11-17 anton.a.afanasyev [SLPVectorizer] Fix assert
2020-11-17 powerman1st [clang][AVR] Improve avr-ld command line options
2020-11-16 a.bataev [OPENMP]Fix PR48076: mapping of data member pointer.
2020-11-17 flo [Matrix] Add inline assembly test case.
2020-10-30 anton.a.afanasyev [SLP] Make SLPVectorizer to use `llvm.masked.gather` intrinsic
2020-11-17 wingo [WebAssembly][MC] Remove useless overrides in MCWasmStreamer
2020-11-17 joe.ellis [AArch64][SVE] Support implicit lax vector conversions for SVE types
2020-11-14 paul [TableGen] Clean up Target .td include files
2020-11-17 benny.kra [MLIR] Fix up integration tests after b7382ed3fea08da27530a6d6d53f168fc704e4c4
2020-11-17 andrzej.warzynski [flang][nfc] Add missing dependency in CMake
2020-08-05 flo [ConstraintElimination] Add support for And.
2020-11-17 kadircet [clang-tidy] Make clang-format and include-order-check coherent
2020-11-16 erich.keane Update 'note-candiate' functions to skip lambda-conversion-op-overloads
2020-11-12 sander.desmalen Reland [LoopVectorizer] NFCI: Calculate register usage based on TLI.getTypeLegalizationCost.
2020-11-17 marukawa [VE] Implement JumpTable
2020-11-17 herhut [mlir][std] Canonicalize extract_element(tensor_cast).
2020-11-17 herhut [mlir][std] Fold dim(dynamic_tensor_from_elements, %cst)
2020-11-17 herhut [mlir][linalg] Allow tensor_to_memref in dependence analysis
2020-11-17 marukawa [VE] Correct getMnemonic
2020-11-15 zinenko [mlir] Allow for using interface class name in ODS interface definitions
2020-11-17 zinenko [mlir] Fix Python tests after "module_terminator" migrated to ODS
2020-11-17 n.james93 [clang][NFC] Use SmallString instead of SmallVector<char
2020-11-17 llvm-dev [X86] vector-sext.ll - replace X32 check prefix with X86. NFC.
2020-11-17 llvm-dev [X86] insert-into-constant-vector.ll - replace X32 check prefix with X86. NFC.
2020-11-17 llvm-dev [X86] avx-intel-ocl.ll - replace X32 check prefix with X86. NFC.
2020-11-17 llvm-dev [X86] 2011-10-19-widen_vselect.ll - replace X32 check prefix with X86. NFC.
2020-11-17 llvm-dev [clang][driver] Fix definition/declaration argument name mismatches. NFCI.
2020-11-17 omair.javaid [LLDB] Fix SVE reginfo for sequential offset in g packet
2020-11-04 luke.drummond Escape command line arguments in backtraces
2020-11-17 omair.javaid [LLDB] Update SVE Z reg info to remove invalidate regs
2020-11-17 flo [AsmPrinter] Use getMnemonic for instruction-mix remark.
2020-11-17 sven.vanhaastregt [OpenCL] Stop opencl-c-base.h leaking extension enabling
2020-11-17 sguelton [build] Fix dependencies of LLVMExegesis
2020-11-16 zinenko [mlir] Add basic support for attributes in ODS-generated Python bindings
2020-11-17 hokein.wu [clang-tidy] Fix an abseil-redundant-strcat-calls crash on 0-parameter StrCat().
2020-11-15 marukawa [VE] Add vbrd intrinsic instructions
2020-11-17 powerman1st [AVR] Optimize the 16-bit NEGW pseudo instruction
2020-11-17 flo [AsmWriter] Factor out mnemonic generation to accessible getMnemonic.
2020-11-13 sguelton [build] normalize components dependencies
2020-11-17 lkail [PowerPC] Add negated abs test using llvm.abs intrinsic. NFC.
2020-11-17 hokein.wu [clang-tidy] Verify the fixes in abseil-redundant-strcat-calls test, NFC
2020-11-17 riddleriver [mlir][NFC] Remove references to Module.h and Function.h
(...)
2020-11-14 marukawa [VE] Add lvm/svm intrinsic instructions
2020-11-16 jonas [crashlog] Implement parser for JSON encoded crashlogs
2020-11-16 flo [ConstraintElimination] Add support for Or.
2020-11-16 listmail [LoopVec] Introduce an api for detecting uniform memory ops
2020-11-13 ldionne [libc++] Move the GDB pretty printer tests to the DSL
2020-11-09 kostyak [scudo][standalone] Correct min/max region indices
2020-11-05 silvasean [mlir] Add DecomposeCallGraphTypes pass.
2020-11-16 thakis lld/mach-o: Infer darwinnew from filename ld64.lld.darwinnew too
2020-11-16 listmail [SCEV] Factor out part of wrap flag detection logic [NFC](try 2)
2020-11-16 aschwaighofer [coro] Async coroutines: Make sure we can handle control flow in suspend point dispatch function
2020-11-16 flo [ConstraintElimination] Add complex test cases.
2020-11-16 spatel Revert "[InstCombine] add multi-use demanded bits fold for add with low-bit mask"
2020-11-16 aaron Another speculative fix for lldb related to ConstexprSpecKind
2020-11-16 aaron Speculatively fix the lldb build
2020-11-16 schuett Convert ConstexprKind from Specifiers.h to a scoped enum; NFC
2020-11-16 jonas [dsymutil] Generalize handling of aliases external symbols
2020-11-13 amara [AArch64][GlobalISel] Look through a G_ZEXT when trying to match shift-extended register offsets.
2020-11-16 Scott.Linder [YAMLIO] Correctly diagnose empty alias/anchor
2020-11-16 craig.topper [RISCV] Add MC tests for the User Vector CSRs
2020-11-16 ldionne [libc++] Remove transitional #error message
2020-11-16 vtjnash Reland "[AsmPrinter] fix -disable-debug-info option"
2020-11-16 praveen [Flang][OpenMP] Add semantic checks for OpenMP Depend clause.
2020-11-16 craig.topper [RISCV] When matching SROIW, check all 64 bits of the OR mask
2020-11-13 aeubanks [SimplifyCFG] Respect optforfuzzing in NPM pass
2020-11-16 teemperor [lldb] Skip TestCppKeywordsAsCIdentifiers on Windows
2020-11-13 pklausler [flang] Unsplit COMPLEX operations
2020-11-16 metzman [fuzzer] Disable uncaught-exception on non-Win+undo bad fix
2020-11-16 xun [Coroutine] Allocas used by StoreInst does not always escape
2020-11-16 metzman [fuzzer] Add allocator_may_return_null to uncaught-exception.test.
2020-11-10 Matthew.Arsenault AMDGPU: Select global saddr mode from SGPR pointer
2020-11-16 Mirko.Brkusanin [AMDGPU][GlobalISel] Fix lowerShlSat
2020-11-09 Matthew.Arsenault AMDGPU: Split large offsets when selecting global saddr mode
2020-11-16 wei.huang Fix the compilation assertion due to unreachable BB pruning not deleting the associated BB from the jump tables
2020-11-16 jay.foad [AMDGPU] Add comments. NFC.
2020-11-16 csigg [mlir] Require std.alloc() ops to have canonical layout during LLVM lowering.
2020-11-16 andrzej.warzynski [flang][f18] Remove "-fdebug-resolve-names"
2020-11-15 marukawa [VE] Optimize leaf functions
2020-11-16 david.truby [MLIR][OpenMP] Add omp.wsloop operation
2020-11-16 simon.moll [VE] fastcc and vreg-to-vreg copy
2020-11-14 yhs BPF: make __builtin_btf_type_id() return 64bit int
2020-11-16 Xiangling.Liao [AIX][driver] Include crti[_64].o and -bcdtors also for C language link invocations by default
2020-11-16 teemperor [lldb] Add support for using variables with C++ keywords names in non-C++ expressions
2020-11-16 llvmgnsyncbot [gn build] Port 8dbe44cb293
2020-11-16 flo Add pass to add !annotate metadata from @llvm.global.annotations.
2020-11-16 hanchung [mlir][StandardToSPIRV] Extend support for lowering cmpi to SPIRV.
2020-11-16 francesco.petrogalli [clang][SveEmitter] Fix enum declarations. [NFCI]
2020-11-14 marukawa [VE] Add lsv/lvs intrinsic instructions
2020-11-16 aaron Fix bots that are running with assertions enabled; NFC.
2020-11-16 aaron Adding some explicit casts to appease build bots; NFC
2020-11-16 schuett Convert TypeSpecifierSign from Specifiers.h to a scoped enum; NFC
Created with:
roll-dep third_party/clspv-llvm
Roll third_party/swiftshader/ e4c1a25cc..04515da40 (3 commits)
https://swiftshader.googlesource.com/SwiftShader.git/+log/e4c1a25cc679..04515da400d5
$ git log e4c1a25cc..04515da40 --date=short --no-merges --format='%ad %ae %s'
2020-11-12 srisser Don't auto append value of stringified enums
2020-11-13 capn Work around LLVM ORCv2 issue affecting Cuttlefish
2020-11-06 capn Enable Reactor Print() in Debug/RelWithDebInfo builds
Created with:
roll-dep third_party/swiftshader
Roll third_party/vulkan-validationlayers/ 9d3ef3258..0cb8cc8cf (4 commits)
https://github.com/KhronosGroup/Vulkan-ValidationLayers/compare/9d3ef3258715...0cb8cc8cfcb2
$ git log 9d3ef3258..0cb8cc8cf --date=short --no-merges --format='%ad %ae %s'
2020-11-11 karl tests: Add test for destroyed buffer during draw validate
2020-11-11 karl layers: Remove duplicate handles from Descriptor subclasses
2020-11-16 mark repo: Update issue templates
2020-11-16 mark repo: Update issue templates
Created with:
roll-dep third_party/vulkan-validationlayers
* suppress on bot
|
|
I added tests/cases/debugger_hlsl_function_call.amber from Ben's Amber
repo. This change will make Amber successfully test debugging more
shaders.
|
|
This works as an optimization but also fixes an issue where binding a single buffer multiple times (to read-only descriptors) in a pipeline would fail because Amber would try to copy back to the Amber buffer multiple times.
|
|
|
|
This CL adds the ability to specify the data_type for an AmberScript buffer to be an array.
|
|
Demonstrates that reading a column vector from a row-major matrix
is a gather operation, and writing it is a scatter operation.
|
|
|
|
`compute` was called before fully initializing the `ssbo` buffer.
This lead to writes beyond the end of the buffer, corrupting memory in exciting ways when using the swiftshader driver.
Fixes: #725
|
|
Some buffers were added twice if vkscript format was used.
Also fix bug in `tests/run_tests.py` that prevented vkscript files from being tested.
|
|
|
|
Also refactored IMAGE parsing to allow arbitrary order of parameters like it's done for BUFFER parsing.
|
|
Also modified Vulkan engine to use separate bindings for each vertex buffer as input rate is set per binding.
Fixes #870
|
|
Fixes #866
|
|
Add support for uniform/storage texel buffers.
Fixes #861
|
|
Add instanced drawing support for AmberScript
Fixes #565
|
|
Added STENCIL support for Amber Script.
|
|
|
|
Also moved delegate from options to Amber object to be able to access it from a parser.
|
|
* Added implementation for loading buffer data from a binary file.
* Use PNG, BINARY, and TEXT to select the type of file to load buffer data from.
|
|
* Add support for kernel decls
* no-op
* Add support for pod args in push constants
* piggybacks push constant support
* pod buffer generation supports push constants
* New test
* Don't assert on unknown interface types in clspv
* update OpenCL SET example
* Refactor some code into CreatePushConstantBuffer
* Update clspv deps
|
|
If the `dap::Source` does not provide the embedded source, scan the virtual file system for the file by path.
Also fix a race condition where we would instruct the debugger to step, but not actually wait for it to notify us that the step had completed before making further checks.
Updated the `debugger_hlsl_line_stepping.amber` to:
* Use `VIRTUAL_FILE` for the file content
* Fix the scoping of `CompilationUnit` <- `Function` <- `LexicalBlock`
* Remove the unused and incorrect debug info from the fragment shader. We'll have proper testing for this once DXC produces correct debug info.
|
|
`VIRTUAL_FILE` declares a file in a new virtual file system.
These files can now be referenced by the `SHADER` commands instead of using inline source.
The DXC compiler will first look in the virtual file system for any `#include`s before falling back to the standard file system.
This allows us to write tests that exercise the compiler and debugger handling of multiple files.
Added `relative_includes_hlsl.amber` which checks that DXC correctly includes relative to the current file, not the root file.
|
|
|
|
This CL adds the `FILE` parameter to `BUFFER` commands. This allows loading file data into the buffer. Currently only PNG data is supported.
|
|
When the -work-dim or -global-offset flags are passed to Clspv, it
will generate push constants. The sizes and offsets of the push
constant values are provided in the descriptor map. Amber needs to add
these to the pipeline layout and create a buffer for them.
|
|
Polygon mode was only supported by VkScript. This adds the support for AmberScript.
|
|
Handling of 0.0 as float16 was broken. It can be easily tested by replacing 2.4 in tests/cases/float16.amber with 0.0, which makes amber assert in float16_helper.cc:48, unable to convert 0.0 to float16. This PR handles this special case and flushes denormals to zero.
|
|
This commits adds support for requesting 16-bit storage features as
specified in VkPhysicalDevice16BitStorageFeaturesKHR by using the
following feature names:
* Storage16BitFeatures.storageBuffer16BitAccess
* Storage16BitFeatures.uniformAndStorageBuffer16BitAccess
* Storage16BitFeatures.storagePushConstant16
* Storage16BitFeatures.storageInputOutput16
|
|
Add support for draw_grid, which is fairly similar to draw_rect,
except it also takes parameters to split the rectangle into
desired number of columns and rows of cells.
This functionality is meant for the testing of vertex shaders,
giving an easy way to produce a lot of vertices.
The implementation is based on draw_rect. Dawn rendering is not
implemented.
|
|
This pull request adds support for 8-bit integers to be used in shaders, both as a basic type and for input/output as part of push constants and uniform/storage blocks. It adds 4 new features understood by Amberscript which are needed to request the corresponding Vulkan features. The original structure is called "8BitStorageFeatures" but it has been renamed to "Storage8BitFeatures" due to the leading digit causing problems to the parser.
|
|
Currently it is possible for the command buffer to have begun work and
not be reset before we shut down the vulkan engine. This CL moves the
reset call into the command buffer and moves the guards into the command
buffer methods themselves.
|
|
Checks the callstack is as expected.
Note: I've updated the .amber scripts with the current SwiftShader
output. It's clear that this is somewhat wonky, and will change.
|
|
SwiftShader now ignores SPIR-V 'non-statement' opcodes for the debugger. These are no longer steppable.
`tests/cases/debugger_spirv_line_stepping.amber` has been updated to match this new behaviour.
|
|
One for SPIR-V debugging, one for HLSL debugging using `OpenCL.DebugInfo.100`.
|
|
* Parse literal sampler descriptor map entries
* add them to the pipeline sampler info
* Before generating engine pipelines, create the sampler objects for the
literal samplers
* Tests
|
|
Contributes to #747
Add support for specifying buffers as images via a new command.
* takes dimensionality and dimensions
* Generalize image support to support 1D and 3D images.
Add some examples of the new functionality.
Simplify some Vulkan image calls.
|
|
This CL adds the float16 data type into amber_script and the required plumbing to enable the feature.
Fixes #485
|
|
This CL builds swiftshader on all platforms and runs the integration tests where possible.
|
|
|
|
Fixes #714
|
|
Previously we could only extract buffers from the first pipeline in the
file. This Cl extends the -B flag to allow providing the pipeline name
as a prefix. So, my_pipeline:0:2 would extract from the pipeline named
my_pipeline, descriptor set 0, binding 2.
Fixes #732
|
|
* Add descriptor map parsing for images and samplers
* Allow samplers to be bound by arg name or ordinal
* no custom support for program scope samplers
* Add parsing of buffer width and height as an alternative to SIZE for use as an image
* add tests for new functionality
* update docs
* add new full flow tests
* update deps and allow rolls
|
|
Currently the test output is always emitted. This is confusing when the
test is suppressed. This CL only outputs the command results if the
script is not suppressed and not expected to fail.
|
|
This test currently fails due to incorrect UClamp generation in the
SPIRV-Opt pass.
(https://github.com/KhronosGroup/SPIRV-Tools/issues/3072)
|
|
This CL adds a --use-opencl flag to run_tests.py which will enable the
OpenCL test cases.
|
|
Fixes #711
|
