blob: 2f2968e2b708c28d71f7fd121ec40545aedb145f [file] [log] [blame]
Revision history for SPIRV-Tools
v2020.1-dev 2019-12-11
- Start v2020.1-dev
v2019.5 2019-12-11
- General:
- Export SPIRV-Tools targets on installation
- SPIRV-Tools support for SPIR-V 1.5 (#2865)
- Add WebGPU SPIR-V Assembler in JavaScript. (#2876)
- Add Bazel build configuration. (#2891)
- Add support for building with emscripten (#2948)
- Update SPIR-V binary header test for SPIR-V 1.5 (#2967)
- Add fuzzer for spirv-as call path (#2976)
- Improved CMake install step. (#2963)
- Add fuzzer for spirv-dis call path (#2977)
- Ensure timestamp does not vary with timezone. (#2982)
- Add a vscode extension for SPIR-V disassembly files (#2987)
- Add iOS as a supported platform (#3001)
- utils/vscode: Add SPIR-V language server support
- Respect CMAKE_INSTALL_LIBDIR in installed CMake files (#3054)
- Permit the debug instructions in WebGPU SPIR-V (#3063)
- Add support for Fuchsia. (#3062)
- Optimizer
- Add descriptor array scalar replacement (#2742)
- Add pass to wrap OpKill in a function call (#2790)
- Fold FMix during constant folding. (#2818)
- Add pass to replace AMD shader ballot extension (#2811)
- Add pass to make Float32 operation relax precision (#2808)
- Add pass to make relax precision operation Float16 (#2808)
- Add pass to replace uses of 3 AMD extensions (#2814)
- Fold Min, Max, and Clamp instructions. (#2836)
- Better handling of OpKill in continues (#2842,#2922,#2933)
- Enable OpTypeCooperativeMatrix specialization (#2927)
- Support constant-folding UConvert and SConvert (#2960)
- Update Offset to ConstOffset bitmask if operand is constant. (#3024)
- Improve RegisterSizePasses (#3059)
- Folding: perform add and sub on mismatched integer types (#3084)
- Graphics robust access: use signed clamp (#3073)
- Instrument: Fix version 2 output record write for tess eval shaders. (#2782)
- Instrument: Add support for Buffer Device Address extension (#2792)
- Fix check for changed binary in API call. (#2798)
- For WebGPU<->Vulkan optimization, set correct execution environment (#2834)
- Handle OpConstantNull in copy-prop-arrays. (#2870)
- Use OpReturn* in wrap-opkill (#2886)
- Validator
- Add generic builtin validation of target (#2843)
- Extra resource interface validation (#2864)
- Adding valilidation checks for OpEntryPoint duplicate names and execution mode (#2862)
- Relaxed bitcast with pointers (#2878)
- Validate physical storage buffer restrictions (#2930)
- Add SPV_KHR_shader_clock validation (#2879, #3013)
- Validate that selections are structured (#2962)
- Disallow use of OpCompositeExtract/OpCompositeInsert with no indices (#2980)
- Check that derivatives operate on 32-bit values (#2983)
- Validate array stride does not cause overlap (#3028)
- Validate nested constructs (#3068)
- Fix validation of constant matrices (#2794)
- Update "remquor" validation
- Only allow previously declared forward refs in structs (#2920)
- Reduce
- Remove relaxed precision decorations (#2797)
- Reduce/fuzz: improve command line args (#2932)
- Improve remove unref instr pass (#2945)
- Fuzz
- Fix add-dead-break and add-dead-continue passes to respect dominance (#2838)
- Add fuzzer pass to copy objects (#2853)
- Add fuzzer pass to replace ids with synonyms (#2857)
- Allow validation during spirv-fuzz replay (#2873)
- Employ the "swarm testing" idea in spirv-fuzz (#2890)
- reduce/fuzz: improve command line args (#2932)
- option to convert shader into a form that renders red (#2934)
- Add fuzzer pass to change selection controls (#2944)
- add transformation and pass to construct composites (#2941)
- Add fuzzer pass to change loop controls (#2949)
- Add fuzzer pass to change function controls (#2951)
- Add fuzzer pass to add NoContraction decorations (#2950)
- Add missing functionality for matrix composites (#2974)
- Fuzzer pass to adjust memory access operands (#2968)
- Transformation to extract from a composite object (#2991)
- Vector shuffle transformation (#3015)
- Improve debugging facilities (#3074)
- Function outlining fuzzer pass (#3078)
v2019.4 2019-08-08
- General:
- Memory model support for SPIR-V 1.4
- Add new spirv-fuzz tool
- Add option for base branch in
- Removed MarkV and Stats code. (#2576)
- Instrument: Add version 2 of record formats (#2630)
- Linker: Better type comparison for OpTypeArray and OpTypeForwardPointer (#2580)
- Optimizer
- Bindless Validation: Instrument descriptor-based loads and stores (#2583)
- Better folding for OpSpecConstantOp (#2585, #2614)
- Add in individual flags for Vulkan <-> WebGPU passes (#2615)
- Handle nested breaks from switches. (#2624)
- Optimizer: Handle array type with OpSpecConstantOp length (#2652)
- Perform merge return with single return in loop. (#2714)
- Add --preserve-bindings and --preserve-spec-constants (#2693)
- Remove Common Uniform Elimination Pass (#2731)
- Allow ray tracing shaders in inst bindle check pass. (#2733)
- Add pass to inject code for robust-buffer-access semantics (#2771)
- Treat access chain indexes as signed in SROA (#2776)
- Handle RelaxedPrecision in SROA (#2788)
- Add descriptor array scalar replacement (#2742)
- Handle decorations better in some optimizations (#2716)
- Change the order branches are simplified in dead branch elim (#2728)
- Fix bug in merge return (#2734)
- SSA rewriter: Don't use trivial phis (#2757)
- Record correct dominators in merge return (#2760)
- Process OpDecorateId in ADCE (#2761)
- Fix check for unreachable blocks in merge-return (#2762)
- Handle out-of-bounds scalar replacements. (#2767)
- Don't move debug or decorations when folding (#2772)
- Protect against out-of-bounds references when folding OpCompositeExtract (#2774)
- Validator
- Validate loop merge (#2579)
- Validate construct exits (#2459)
- Validate OpenCL memory and addressing model environment rules (#2589)
- Validate OpenCL environment rules for OpTypeImage (#2606)
- Allow breaks to switch merge from nested construct (#2604)
- Validate OpenCL environment rules for OpImageWrite (#2619)
- Allow arrays of out per-primitive builtins for mesh shaders (#2617)
- Validate OpenCL rules for ImageRead and OpImageSampleExplicitLod (#2643)
- Add validation for SPV_EXT_fragment_shader_interlock (#2650)
- Add builtin validation for SPV_NV_shader_sm_builtins (#2656)
- Add validation for Subgroup builtins (#2637)
- Validate variable initializer type (#2668)
- Disallow stores to UBOs (#2651)A
- Validate Volatile memory semantics bit (#2672)
- Basic validation for Component decorations (#2679)
- Validate that in OpenGL env block variables have Binding (#2685)
- Validate usage of 8- and 16-bit types with only storage capabilities (#2704)
- Add validation for SPV_EXT_demote_to_helper_invocation (#2707)
- Extra small storage validation (#2732)
- For Vulkan, disallow structures containing opaque types (#2546)
- Validate storage class OpenCL environment rules for atomics (#2750)
- Update OpControlBarriers rules for WebGPU (#2769)
- Update OpMemoryBarriers rules for WebGPU (#2775)
- Update WebGPU validation rules of OpAtomic*s (#2777)
- Disallow merge targeting block with OpLoopMerge (#2610)
- Update vloadn and vstoren validation to match the OpenCL Extended
Instruction Set Specification (#2599)
- Update memory scope rules for WebGPU (#2725)
- Allow LOD ops in compute shaders with derivative group execution modes (#2752)
- Reduce
v2019.3 2019-05-14
- General:
- Require Python 3 since Python 2 will out of service soon.
- Add a continuous test that does memory checks using the address sanitizer.
- Fix the build files so the SPIRV_USE_SANITIZER=address build works.
- Packaging top of tree build artifacts again.
- Added support for SPIR-V 1.4. (#2550)
- Optimizer
- Remove duplicates from list of interface IDs in OpEntryPoint instruction (#2449)
- Bindless Validation: Descriptor Initialization Check (#2419)
- Add option to validate after each pass (#2462)
- Add legalization pass to fix mismatched pointer (#2430, #2535)
- Add error messages when the input contains unknown instructions. (#2487)
- Add pass to convert from WebGPU Spir-V to Vulkan Spir-V and back. (#2495)
- #2412: Dead memeber elimination should not change input and output variables.
- #2405: Fix OpDot folding of half float vectors.
- #2391: Dead branch elim should not fold away back edges.
- #2441: Removing decorations when doing constant propagation.
- #2455: Maintain inst to block mapping in merge return.
- #2453: Fix merge return in the face of breaks.
- #2456: Handle dead infinite loops in DCE.
- #2458: Handle variable pointer in some optimizations.
- #2452: Fix dead branch elimination to handle unreachable blocks better.
- #2528: Fix undefined bit shift in sroa.
- #2539: Change implementation of post order CFG traversal.
- Validator
- Add validation of storage classes for WebGPU (#2446)
- Add validation for ExecutionMode in WebGPU (#2443)
- Implement WebGPU specific CFG validation (#2386)
- Allow NonWritable to target struct members. (#2420)
- Allow storage type mismatch for parameter in relaxed addressing mode.
- Allow non memory objects as parameter in relaxed addressing mode.
- Disallow nested Blocks and buffer blocks (#2410).
- Add validation for SPV_NV_cooperative_matrix (#2404)
- Add --strip-atomic-counter-memory (#2413)
- Check OpSampledImage is only passed into valid instructions (#2467)
- Handle function decls in Structured CFG analysis (#2474)
- Validate that OpUnreacahble is not statically reachable (#2473)
- Add pass to generate needed initializers for WebGPU (#2481)
- Allow images without format for OpenCL. (#2470)
- Remove unreachable block validation (#2525)
- Reduce runtime of array layout checks (#2534)
- Add validation specific to OpExecutionModeId (#2536)
- Validate sign of int types. (#2549)
- VK_KHR_uniform_buffer_standard_layout validation (#2562)
- #2439: Add missing DepthGreater case to Fragment only check.
- #2168: Disallow BufferBlock on StorageBuffer variables for Vulkan.
- #2408: Restrict and Aliased decorations cannot be applied to the same id.
- #2447: Improve function call parameter check.
- Reduce
- Add Pass to remove unreferenced blocks. (#2398)
- Allows passing options to the validator. (#2401)
- Improve reducer algorithm and other changes (#2472)
- Add Pass to remove selections (#2485)
- Add passes to simplify branches (#2507)
- #2478: fix loop to selection pass for loops with combined header/continue block
v2019.2 2019-02-20
- General:
- Support SPV_EXT_physical_storage_buffer
- A number of memory leak have been fixed.
- Removed use of deprecated Google test macro:
- Changed the to only build tests in Chromium.
- Optimizer
- Upgrade memory model improvments for modf and frexp.
- Add a new pass to move loads closer to their uses: code sinking.
- Invalidating the type manager now invalidates the constnat manager.
- Expand instrumentation pass for bindless bounds checking to runtime-sized descriptor arrays.
- Add a new pass that removes members from structs that are not used: dead member elimination.
- #2292: Remove undefined behaviour when folding bit shifts.
- #2294: Fixes for instrumentation code.
- #2293: Fix overflow when folding -INT_MIN.
- #2374: Don't merge unreachable blocks when merging blocks.
- Validator
- Support SPV_KHR_no_integer_wrap and related decorations.
- Validate Vulkan rules for OpTypeRuntimeArray.
- Validate NonWritable decoration.
- Many WebGPU specific validation rules were added.
- Validate variable pointer related function call rules.
- Better error messages.
- #2307: Check forwards references in OpTypeArray.
- #2315, #2303: Fixed the layout check for relaxed layout.
- #1628: Emit an error when an OpSwitch target is not an OpLabel.
- Reduce
- Added more documentation for spirv-reduce.
- Add ability to remove OpPhi instructions.
- Add ability to merge two basic blocks.
- Add ability to remove unused functions and unused basic blocks.
v2019.1 2019-01-07
- General:
- Created a new tool called spirv-reduce.
- Add cmake option to turn off SPIRV_TIMER_ENABLED (#2103)
- New optimization pass to update the memory model from GLSL450 to VulkanKHR.
- Recognize OpTypeAccelerationStructureNV as a type instruction and ray tracing storage classes.
- Fix GCC8 build.
- Add --target-env flag to spirv-opt.
- Add --webgpu-mode flag to run optimizations for webgpu.
- The output disassembled line number stead of byte offset in validation errors. (#2091)
- Optimizer
- Added the instrumentation passes for bindless validation.
- Added passes to help preserve OpLine information (#2027)
- Add basic support for EXT_fragment_invocation_density (#2100)
- Fix invalid OpPhi generated by merge-return. (#2172)
- Constant and type manager have been turned into analysies. (#2251)
- #2018: Don't inline functions with a return in a structured CFG contstruct.
- #2047: Fix bug in folding when volatile stores are present.
- #2053: Fix check for when folding floating pointer values is allowed.
- #2130: Don't inline recursive functions.
- #2202: Handle multiple edges between two basic blocks in SSA-rewriter.
- #2205: Don't unswitch a latch condition during loop unswitch.
- #2245: Don't fold branch in loop unswitch. Run dead branch elimination to fold them.
- #2204: Fix eliminate common uniform to place OpPhi instructions correctly.
- #2247: Fix type mismatches caused by scalar replacement.
- #2248: Fix missing OpPhi after merge return.
- #2211: After merge return, fix invalid continue target.
- #2210: Fix loop invariant code motion to not place code between merge instruction and branch.
- #2258: Handle CompositeInsert with no indices in VDCE.
- #2261: Have replace load size handle extact with no index.
- Validator
- Changed the naming convention of outputing ids with names in diagnostic messages.
- Added validation rules for UniformConstant variables in Vulkan.
- #1949: Validate uniform variable type in Vulkan
- Ensure for OpVariable that result type and storage class operand agree (#2052)
- Validator: Support VK_EXT_scalar_block_layout
- Added Vulkan memory model semantics validation
- Added validation checkes spefic to WebGPU environment.
- Add support for VK_EXT_Transform_feedback capabilities (#2088)
- Add validation for OpArrayLength. (#2117)
- Ensure that function parameter's type is not void (#2118)
- Validate pointer variables (#2111)
- Add check for QueueFamilyKHMR memory scope (#2144)
- Validate PushConstants annotation and type (#2140)
- Allow Float16/Int8 for Vulkan 1.0 (#2153)
- Check binding annotations in resource variables (#2151, #2167)
- Validate OpForwardPointer (#2156)
- Validate operation for OpSpecConstantOp (#2260)
- #2049: Allow InstanceId for NV ray tracing
- Reduce
- Initial commit wit a few passes to reduce test cases.
- Validation is run after each reduction step.
v2018.6 2018-11-07
- General:
- Added support for the Nvidia Turing and ray tracing extensions.
- Make C++11 the CXX standard in CMakeLists.txt.
- Enabled a parallel build for MSVC.
- Enable pre-compiled headers for MSVC.
- Added a code of conduct.
- EFFCEE and RE2 are now required when build the tests.
- Optimizer
- Unrolling loops marked for unrolling in the legalization passes.
- Improved the compile time of loop unrolling.
- Changee merge-return to create a dummy loop around the function.
- Small improvement to merge-blocks to allow it to merge more often.
- Enforce an upper bound for the ids, and add option to set it.
- #1966: Report error if there are unreachable block before running merge return
- #1917: Allow 0 (meaning unlimited) as a parameter to --scalar-replacement
- #1915: Improve handling of group decorations.
- #1942: Fix incorrect uses of the constant manager. Avoids type mismatches in generated code.
- #1997: Fix dead branch elimination when there is a loop in folded selection.
- #1991: Fixes legality check in if-conversion.
- #1987: Add nullptr check to array copy propagation.
- #1984: Better handling of OpUnreachable in ADCE.
- #1983: Run merge return on reachable functions only.
- #1956: Handled atomic operations in ADCE.
- #1963: Fold integer divisions by 0 to 0.
- #2019: Handle MemberDecorateStringGOOGLE in ADCE and strip reflect.
- Validator
- Added validation for OpGroupNonUniformBallotBitCount.
- Added validation for the Vulkan memory model.
- Added support for VK_KHR_shader_atddomic_int64.
- Added validation for execution modes.
- Added validation for runtime array layouts.
- Added validation for 8-bit storage.
- Added validation of OpPhi instructions with pointer result type.
- Added checks for the Vulkan memory model.
- Validate MakeTexelAvailableKHR and MakeTexelVisibleKHR
- Allow atomic function pointer for OpenCL.
- FPRounding mode checks were implemented.
- Added validation for the id bound with an option to set the max id bound.
- #1882: Improve the validation of decorations to reduce memory usage.
- #1891: Fix an potential infinite loop in dead-branch-elimination.
- #1405: Validate the storage class of boolean objects.
- #1880: Identify arrays of type void as invalid.
- #487: Validate OpImageTexelPointer.
- #1922: Validate OpPhi instructions are at the start of a block correctly.
- #1923: Validate function scope variable are at the start of the entry block.
v2018.5 2018-09-07
- General:
- Support SPV_KHR_vulkan_memory_model
- Update Dim capabilities, to match SPIR-V 1.3 Rev 4
- Automated build bots no run tests for the VS2013 case
- Support Chromium GN build
- Use Kokoro bots:
- Disable Travis-CI bots
- Disable AppVeyor VisualStudio Release builds. Keep VS 2017 Debug build
- Don't check export symbols on OSX (Darwin): some installations don't have 'objdump'
- Reorganize source files and namespaces
- Fixes for ClangTidy, and whitespace (passes 'git cl presumit --all -uf')
- Fix unused param compile warnings/errors when Effcee not present
- Avoid including time headers when timer functionality is disabled
- Avoid too-stringent warnings flags for Clang on Windows
- Internal refactoring
- Add hooks for automated fuzzing
- Add testing of command line executables
- #1688: Use binary mode on stdin; fixes "spirv-dis <foo.spv" on Windows
- Optimizer
- The optimizer validates the module before it begins
- Add API to register passes by string name
- Fold a vector shuffle feeding a vector shuffle
- Add -combine-access-chains transform
- Refactor how IRContext is handled by passes
- Improve bookkeeping for instruction result type and result id
- Fix over-duplication of decorations
- Fix handling of exits from selections in dead-branch elimination, and dead code
- Fix handling of certain kinds of flow control in merge-return
- #1721: Fix size bug when folding vector shuffles
- #1722: Fix size infinite loop when folding vector shuffles
- #1724: Fix finding a constant of a specific type
- #1727: Dead branch elim: Reorder blocks if needed to satisfy dominance rule
- #1729: Handle VariablePointers cases in various optimizations
- #1731: Fix vector shuffle with literal id indicating undef value
- #1736: Fix handling of decorations and phis in merge-return
- #1787: Fix handling of decorations related to access chains
- #1865: Avoid leaking memory for SPIR-V constant values
- Validator
- Improve error messages
- Avoid platform-dependent traversal ordering, to ensure consistent messages
- Use libspirv::Instruction where possible
- Add option to skip all block layout checks
- Validate all type IDs
- Validate uses of OpFunction
- Validate uses of OpTypeFunction
- Disallow a struct containing its own type
- #1685: Vulkan permits non-monotonic offsets for block members
- #1697: Enforce block layout rules even when using relaxed block layout option
- #1719: Fix line number for vector shuffle valiation error
- #1789: Avoid assertion failure when validating some functions
- #1800: Fix validation of OpCopyMemorySized
- #1822: Stop enforcing struct member offset montonicity
- #1831: Disallow void members in structs
v2018.4 2018-07-08
- General:
- Support SPV_KHR_8bit_storage
- Add gclient and presubmit configurations
- Enable Kokoro build bots (#1625)
- Group tests into fewer executables, reduces load on CI
- Port test script to Python 3
- Symbol export tests respect SPIRV_SKIP_TESTS
- #1596: Operand lookup succeeds if enabled by a capability
- #1624: Instruction lookup succeeds if enabled by a capability
- Refactoring namespaces:
- #1678: Change libspirv to spvtools
- Code in source/utils moved into spvtools::utils
- Code in source/comp moved into spvtools::comp
- Optimizer:
- Remove insert-extract-elim pass. Use simplification pass instead.
- Preserve instruction-to-block mapping in most passes, to reduce runtime.
- Small vector optimization for operands
- Add pass to move Private variables to Function. Increase opportunity to optimize.
- Fixes:
#1120: Check static uses of entry point interfaces
#1372: Avoid merging some structs, to preserve names for reflection
#1577: Scalar replacement uses only undecorated types.
#1578: Fix handling of forward-pointer types, and types embedding pointers
to themselves.
#1591: Inliner: Callee variable with initializer should have a store at the call site.
#1634: Fix crash: Use type id in vector type lookup
#1649: Fix assert in compact-ids pass
Fix constant folder: ensure it uses the right type
#1659: Folding rules added to IRContext. Avoids leak.
- Validator
- Add work-in-progress WebGPU environment. Disallows OpUndef
- #670, #1581: Improve error messages; disassemble instruction
- #491: Check structured switches
- #937: Check layout rules for Block and BufferBlock in Uniform, StorageBuffer, PushConstant
- #1281: Check invalid branches into structured constructs
- #1522: Disallow array-of-arrays with DescriptorSets
- #1577: Allow duplicate pointer types.
- #1581: Better messages: output ID names along with numbers in more cases.
- #1597: Check Vulkan 1.1 capabilities
- #1618: Check invalid exit from structured case construct
- #1622: Run IdPass before DataRulesPass
- #1632: Reduce test time by artificially lowering limits in limit test
- #1638: Block-decorated structs member order must respect offset order
- #1657: Improve CFG validation diagnostics
- Khronos SPIR-V #337: GLSL.std.450 Refract instruction Eta param can be any float scalar.
- #1606: PushConstant Blocks follow storage-buffer layout rules
- #1664: Check layout of StorageBuffer variables with Block decoration, using storage buffer
- #1666: Layout validation should permit {vec3; float} packing
- #1637, #1668: Layout validation uses RowMajor, ArrayStride, MatrixStride properly
- Linker
- Avoid buffer overrun when creating OpModuleProcessed
v2018.3 2018-05-25
- General:
- Support SPV_EXT_descriptor_indexing
- Support SPV_GOOGLE_decorate_string
- Support SPV_GOOGLE_hlsl_functionality1
- Support SPV_NV_shader_subgroup_partitioned
- Use "unified1" grammar from SPIRV-Headers
- Simplify support for new extensions. Assembler, disassembler, and simple validation
support is automatic if new tokens are introduced with appropriate extension
attributes in the "unified1" SPIR-V core grammar.
- Disassembler: Emit more digits on floating point, to reliably reproduce all
significand bits. (Use std::max_digits10 instead of std::digits10)
- Fix compilation for old XCode versions: Explicit construction required for std::set.
- Optimizer:
- Add --strip-reflect
- Add --time-report
- Add --loop-fission
- Add lop fusion.
- Add loop peeling pass and internal utility.
- Improve optimizer runtime.
- Merge-return now works with structured control flow.
- New (faster) SSA rewriter to convert local loads and stores to SSA IDs and phis.
Can replace load/store elimination passes.
- Fix instruction folding case: insertion that feeds and extract, when the extract
- Fold OpDot.
- Fold OpFNegate.
- Fold multply and divide of same value.
- Fold FClamp feeding a compare.
- Fold OpLoad feeding an extract, to reduce excessive copying. (#1547)
- Fold Fmix feeding an extract.
- Use simplification pass instead of insert-extract elimination.
- Constant fold OpVectorTimesScalar.
- Copy propagate arrays, in simple cases.
- Aggressive dead code elimination: Can remove more instructions, e.g. derivatives.
- Aggressive dead code elimination: Remove Workgroup variables that are written but not read.
- Better handling of OpImageTexelPointer
- Initial utilities for scalar evolution.
- Add Vector dead code elimination.
- Each pass can only run once.
- Allow code hosting in if-conversion.
- Add external interface for adding a PassToken, so external code can make their own
- Fixes:
#1404: Don't optimize away the compute compute workgroup size constant.
#1407: Remove a bad assertion
#1456: Fix bug in SSA rewriter related to variables updated in loops.
#1487: Fix long runtime in Dead insertion elimination: Don't revist select phi nodes.
#1492: Aggressive dead code elimination can remove OpDecorateStringGOOGLE.
#1527: Fix inlining of functions having OpKill and OpUnreachable.
#1559: Fix assert failure in reduce-load-size pass.
#1556: Aggressive dead code elimination: Fix handling of OpCopyMemory.
- Validator:
- Check Vulkan built-in variables
- Check Vulkan-specific atomic result type rule.
- Relax control barrier check for SPIR-V 1.3. Fixes #1427
- Check OpPhi.
- Check OpMemoryModel.
- Stop checking sizes derived from spec-constants.
- Re-enable checks for OpUConvert.
- Vulkan: Fix check for PrimitiveId: Permit as Input in fragment shader.
- Validate binary version for the given target environment.
- Add tests for OpBranch checks.
- Vulkan 1.1: Check scope for non-uniform subgroup operations.
- Fix checks for SPV_AMD_gpu_shader_int16.
- Fix logical layout check for OpDecorateId.
- Fix checks for ViewportIndex & Layer for Vulkan and SPV_EXT_shader_viewport_index_layer.
- Fixes:
#1470: Vulkan: Don't restrict WorkgroupSize to Input storage class.
#1469: Vulkan: Permit Subgroup memory scope for Vulkan 1.1.
#1472: Per-vertex variable validation fixes.
#1483: Valdiate barrier execution scopes for Vulkan 1.1.
- Fixes:
#898: Linker properly removes FuncParamAttr from imported symbols.
#924, #1174: Fix handling of decoration groups in optimizer, linker.
v2018.2 2018-03-07
- General:
- Support SPIR-V 1.3 and Vulkan 1.1.
- Default target environment is now SPIR-V 1.3. For command-line tools,
use the --target-env option to override the default. Examples:
# Generate a SPIR-V 1.0 binary instead of SPIR-V 1.3
spirv-as --target-env spv1.0 a.spvasm -o a.spv
spirv-as --target-env vulkan1.0 a.spvasm -o a.spv
# Validate as Vulkan 1.0
spirv-val --target-env vulkan1.0 a.spv
- Support SPV_GOOGLE_decorate_string and SPV_GOOGLE_hlsl_functionality1
- Fixes:
- Fix build. Compilation was failing due to missing definitions of
SpvCapabilityFloat16ImageAMD and other enumerated values.
- Optimizer: Avoid generating duplicate names when merging types.
- #1375: Validator: SPV_AMD_gpu_shaer_half_float implicitly allows declaration
of the 16-bit floating point type.
- #1376: Optimizer: Avoid folding half-precision float.
v2018.1 2018-03-02
- General:
- Support Visual Studio 2013 again. (Continue support for VS 2015 and VS 2017.)
- Support building SPIRV-Tools as a shared library.
- Improve the HLSL legalization optimization recipe. #1311
- Optimizer:
- General speedups.
- Remove generic dead code elimination functionality from transforms:
To recover the previous behaviour, a recipe using those transforms should now
also invoke the --eliminate-dead-code-aggressive transform.
- Improve folding, including coverage for floating point, OpSelect, and arithmetic
with non-trivial constant operands.
- Add loop-invariant code motion pass.
- Add loop-unrolling pass, for honouring unroll hits.
- Add loop-unswitch pass.
- Add instruction simplification pass.
- Aggressive dead code elimination: Understands capability hierarchy when finding
instructions it can eliminate (combinators). (PR #1268)
- CCP can now fold floating point arithmetic. #1311
- Validator:
- Validate barrier instructions.
- Check Vulkan-specific rules for atomics.
- Check Vulkan prohibition of Location or Component decorations on BuiltIn variables.
- Linker:
- Add --verify-ids option
- Add option to allow a resulting module to be partially linked.
- Handle OpModuleProcessed (instructions in SPIR-V layout section 7c)
- Fixes:
- #1265: Optimizer: Fix use-after free bug in if-conversion. (Fix object lifecycle bug
in type manager.)
- #1282: Fix new warnings found by GCC 8.0.1.
- #1285: Optimizer: Fix random failures during inlining. (Dangling references in DefUseManager)
- #1295: Optimizer: Fix incorrect handling of Phi nodes in CCP.
- #1300: Fix CCP: avoid bad CCP transitions and unsettled values.
- #1304: Avoid static-duration variables of class type (with constructors).
- #1323: Fix folding of an insert composite feeding a composite extract.
- #1339: Fix CCP: Handle OpConstantNull boolean values as conditions.
- #1341: DCEInst: Keep atomic instructions (and some others with side effects).
- #1354: Don't fold integer division.
- #1357: Support OpConstantNull in folding.
- #1361: CCP: Fix handling of non-constant module-scope values
v2018.0 2018-02-02
- General
- VisualStudio 2013 is no longer supported. VisualStudio 2015 is supported.
- Use "include/unified1" directory from SPIRV-Headers. Requires recent SPIRV-Headers source.
- Disassembler: spirv-dis adds --color option to force color disassembly.
- Optimizer:
- Add pass to eliminate dead insertions.
- Aggressive dead code elimination now removes OpSwitch constructs.
- Block merging occurs in more cases.
- Add driver workaround transform: replace OpUnreachable with harmless branch to merge.
- Improve instruction folding framework.
- Add loop analysis.
- Add scalar replacement of aggregates to size-optimization recipe.
- Add pass to replace instructions invalid for a shader stage, with a harmless value.
This changes the semantics of the program! Not for general use!
- Rearragne and add passes to performance-optimization recipe, to produce better results.
- Validator:
- Validate OpenCL extended instructions.
- Shaders can't perform atomics on floats.
- Validate memory semantics values in atomics.
- Validate instruction-adjacency constraints, e.g. OpPhi predecessors, merge instructions
immediately precede branches.
- Fixes:
- PR 1198: Optimizer: Fix CCP in presence of matrix constants.
- #1199: Optimizer: Fix CCP: don't propagate spec constants.
- #1203: Optimizer: Fix common uniform elim bug introduced by refactoring.
- #1210: Optimizer: Aggressive dead code elimination: Fix 'break' identification.
- #1212: Optimizer: Aggressive dead code elimination: Was skipping too many instructions.
- #1214: Optimizer: Aggressive dead code elimination: Fix infinite loop.
- #1228: Optimizer: Fix CCP: Handling of varying Phi nodes; was resulting in infinite loop.
- #1245: Optimizer: Dead branch elimination: Avoid a null pointer dereference.
- #1250: Optimizer: Dead branch elimination: Avoid spuriously reporting a change.
v2017.3 2018-01-12
- General:
- Support DebugInfo extended instruction set, targeted at OpenCL environments.
See the SPIR-V Registry.
- Generate a SPIRV-Tools.pc file for pkg-config.
- Optimizer:
- Progress for legalization of code generated from HLSL (issue #1118):
- Add --legalize-hlsl option to run transforms used to transform intermediate
code generated by HLSL to SPIR-V for Vulkan compilers. Those compilers
normally run these transforms automatically. This option is used for developing
those transforms.
- Add Private-to-Function variable conversion for modules with logical
- Add --ccp: SSA Conditional Constant Propagation (CCP)
- Add --print-all to show disassembly for each optimization pass.
- Internal: Add loop descriptors and post-order tree iterator.
- Generalized dead branch elimination
- Aggressive dead code elimination (ADCE) now removes dead functions and
module-scope variables.
- Vector extract/insert elimination now optimizes through some cases of
VectorShuffle, and GLSL.std.450 Mix extended instruction.
- Validator:
- Add validation for GLSL.std.450 extended instruction set.
- Check out of bounds composite accesses, where that's statically computable.
Fixes #1112.
- Check upper bits of literal numbers that aren't a multiple of 32-bits wide.
- More validation of primitive instructions
- Add optional "relaxed" checking logical addressing mode to permit some
cases of pointer-to-pointer. Contributes to HLSL legalization (issue #1118).
- Fixes:
#1100: Validator: Image operand Sample can be used with OpImageSparseFetch,
#1108: Remove duplicates transform was incorrectly removing non-duplicate
#1111: Optimizer's type manager could reference deleted memory.
#1112: Fix decoration equality check, e.g. it is now symmetric.
#1129: Validator now disallows Dim=SupbassData for OpImageSparseRead.
#1143: Fix CCP: Was generating incorrect code for loops.
#1153: Fix CCP crash.
#1154: Optimizer's internal instruction-to-block mappings were sometimes
#1159: Fix CCP infinite loop.
#1168: Fix dead branch elimination intermittently generating incorrect code.
#1186: Fix validation of PackDouble2x32 and UnpackDouble2x32
v2017.2 2017-12-15
- General:
- Support OpenCL 1.2, 2.0 target environments, including embedded profiles
- Add
- Fix exit status code for spirv-link
- Disassember: Enable emitting ANSI colour codes to a string
- Library avoids polluting global namespace. The libraries can export C and C++
symbols starting with "spv", or in a C++ namespace. Add a test for this.
- Linux release builds include debug information, for easier profiling
- Build bots no longer test VisualStudio 2013
- Testing dependency RE2 requires VisualStudio 2015 or later
- Build bots check code formatting
- Optimizer:
- Add --skip-validation to spirv-opt
- Add dominance tree analysis
- Add generic value propagation engine
- Add global redundancy elimination within a function
- Add scalar replacement of function-scope variables of composite type
- Aggressive dead code elimination: Remove empty loops
- Killing an instruction notifies the IRContext
- IRContext::KillInst deletes the instruction
- Move CFG analysis to IRContext
- Add constant manager
- Fix: Don't consider derivative instructions as combinators.
- Fix: Don't delete an instruction twice in local dead-code-elimination
- Fix: Don't consider derivative instructions as combinators.
- Validator:
- Finish checking of image instructions (Section 3.32.10)
- Check sparse image instructions
- Check OpTypeImage, OpTypeSampleImage
- Check composite instructions (Section 3.32.12)
- Check atomic instructions (Section 3.32.18)
- Check OpEmitStreamVertex, OpEndStreamPrimitive instructions
- Re-enable validation of OpCopyObject
- OpKill, image ImplicitLod and QueryLod instructions can only be used in Fragment
- Fixes for image instruction validation:
- Lod image operand only usable with ExplicitLod and OpImageFetch
- ExplicitLod Lod image operand must be float scalar
- OpImageFectch Lod image operand must be int scalar
- OpImageGather component operand must be 32-bits (integer scalar)
- OpImageQuerySizeLod Lod must be integer scalar
- Fixes:
#622: Remove names and decorations when inlining
#989: Aggressive dead code elim: Don't optimize away live breaks from a loop
#991: Fix validation of SPV_AMD_shader_ballot
#1004: Use after free of an instruction, in remove-duplicates transform
#1007: OpImageRead not required to return 4-component vector
#1009: OpImageRead can return scalar int/float types
#1011: OpImageWrite should allow scalar int/float texel types
#1012: Fix validat Dref type check
#1017: Load-store elimination considers variable initializations
#1034: Fix Windows debug build: operator< should be a weak ordering
#1083: Inlining: Set parent (function) for each inlined basic block.
#1075: Aggressive dead code elimination: Was leaving dangling references to
removed blocks.
v2017.1 2017-11-23
- Update README with details on the mailing list.
- General:
- Automatically deploy built artifacts to GitHub Releases
- Add a Linker (module combiner). Under development.
- Add for Android NDK builds.
- Add the 'effcee' library as an optional dependency for use in tests.
Eventually it will be a required dependency, once downstream projects have
a chance to adjust. Requires 're2' library.
- Avoid static-duration variables of class type (with constructors).
- Hack around bugs in gcc-4.8.1 template handling
- Faster opcode lookup
- Validator:
- Recognize extensions listed on SPIR-V registry,
through #25 SPV_AMD_shader_fragment_mask
- Validator issues an info message when it sees an unrecognized extension.
- Type check basic arithmetic operations
- Type check carry/extended arithmetic operations
- Type check vector arithmetic operations
- Type check Relational and Logical instructions
- Type check Bit instructions
- Check type uniqueness rules
- Check conversion instructions
- Check image instructions
- Check derivative instructions
- Check OpVectorShuffle
- Check OpBranchConditional
- OpModuleProcessed is only allowed after debug names section and before annotations
- Checks the right kind of return is called for each function (void or non-void).
- Add option to relax type check when storing structs (--relax-store-struct)
- Optimizer:
- Refactoring internal representation of the module, including:
- IRContext: owns a module and manages analyses
- Instructions are owned by intrusive lists, and have unique IDs
- BasicBlock owns its instruction list.
- DefUseManager: change representation of uses, for faster processing
on large modules.
- Add high level recipes: -O, -Os, and -Oconfig
Recipes for -O and -Os are under development.
- Add eliminate-dead-function transform
- Add strength reduction transform: For now, convert multiply by power of 2
to a bit shift.
- Add CFG cleanup transform
- Add removal of dead module-scope variables
- Add merge-return transform for modules without structured control flow
- Add redundancy elimination within a basic block (local value numbering)
- Extract-insert elimination:
- Recognize the case where the first instruction in the sequence is an
OpCompositeConstruct or OpConstantComposite
- Handle some cases of nested structs
- Dead branch elimination now can eliminate entire selection constructs
when all arms are dead.
- Compressing codec:
- Updated algorithm to 1.01, 1.02, 1.03
- Not built by default. Use -DSPIRV_BUILD_COMPRESSION=ON to build.
- Codec can be parameterized by a customized model.
- Fixes:
#728: Fix decoration of inlined functions
#798: spirv-as should fail when given unrecognized long option
#800: Inliner: Fix inlining function into header of multi-block loop
#824: Eliminate-local-multi-store: Fix a crash
#826: Elimiante-local-multi-store: Fix a crash
#827: Fix crash when compact-ids transform runs before another transform.
#834: Add Cmake option to build the compressing codec. Off by default.
#911: Fix classification of Line and NoLine instructions
v2017.0 2017-09-01
- Update README to describe that assembler, disassembler, and binary parser support
are based on grammar files from the SPIRV-Headers repository.
v2016.7 2017-09-01
- Add SPIR-V 1.2
- OpenCL 2.2 support is now based on SPIR-V 1.2
- Support AMD extensions in assembler, disassembler:
- Optimizer: Add support for:
- Inline all function calls in entry points.
- Flatten decoration groups. Fixes #602
- Id compaction (minimize Id bound). Fixes #624
- Eliminate redundant composite insert followed by extract
- Simplify access chains to local variables
- Eliminate local variables with a single store, if possible
- Eliminate local variables with a several stores, if possible
- Eliminate loads and stores in same block to local variables
- Eliminate redundant insert/extract to composite values
- Aggressive dead instruction elimination
- Eliminate dead branches
- Merge blocks when the second can only be preceded by the first
- Eliminate ommon uniform loads
- Assembler: Add option to preserve numeric ids. Fixes #625
- Add build target spirv-tools-vimsyntax to generate spvasm.vim, a SPIR-V
assembly syntax file for Vim.
- Version string: Allow overriding of wall clock timestamp with contents
of environment variable SOURCE_DATE_EPOCH.
- Validator implements relaxed rules for SPV_KHR_16bit_storage.
- CMake installation rules use GNUInstallDirs. For example, libraries
will be installed into a lib64 directory if that's the norm for the
current system.
- Fixes:
#500: Parameterize validator limit checks
#508: Support compilation under CYGWIN
#517: Fix validation when continue (or case) contstruct is also the head of a
nested control construct.
#551: If a merge block is reachable, it must be *strictly* dominated by its
#548: Validator: Error when the reserved OpImageSparseSampleProj* opcodes
are used.
#611: spvtools::Optimizer was failing to save the module to the output
binary vector when all passes succeded without changes.
#629: The inline-entry-points-all optimization could generate invalidly
structured code when the inlined function had early returns.
#697: Optimizer's Instruction::ForEachInId method was skipping semantics-id
and scope-id.
#755: Inliner: Fix inlining of callee with single Return appearing before
the end of the function.
#776: Fix dead branch elimination in presence of complex but dead control
#781: SPV_KHR_variable_pointers allows duplicate pointer types
#782: Inliner: Fix remapping of non-label forward references in callee
#787: Inliner: Fix remapping of inlined entry block when called from
single block loop.
#790: Inliner: Fix remapping of inlined entry block when callee has
multiple returns.
v2016.6 2016-12-13
- Published the C++ interface for assembling, disassembling, validation, and
- Support SPV_KHR_shader_draw_parameters in assembler, disassembler, parser.
- Validator:
- Add validator API accepting raw binary words
- Increased coverage:
- Checks "Data rules" in Universal Validation Rules, section 2.16.1
- WIP: Universal Limits.
- The minimum mandated upper bounds are checked.
- TODO: Parameterize the validator to allow larger limits accepted by
a more than minimally capable implementation.
- OpSampledImage checks
- OpConstantComposite checks
- Id bound check
- Disasssembler:
- Generates friendly GLSL-based names for more builtin variables
- Generates friendly names for numeric OpConstant values
- Vendor tool info extracted from SPIR-V XML registry file.
- Fixes issues:
#429: Validator: Allow OpTypeForwardPointer and OpTypeStruct to reference
undefined IDs
#482: Validator: OpVariable initializer can be an ID of a module-scope variable
v2016.5 2016-09-16
- Support SPV_KHR_shader_ballot in assembler, disassembler, parser.
- Disassembler: Generate friendly names for built-in variables.
- Partial fixes:
#359: Add Emacs helper for automatically diassembling/assembling a SPIR-V
binary on file load/save.
- Fixes:
#414: Validator: Allow OpUndef for composite constants
#415: Validator: Phi can use its own value in some cases.
v2016.4 2016-09-01
- Relicensed under Apache 2.0
- Add optimization passes (in API and spirv-opt command)
- Fold spec constants defined with OpSpecConstantOp and
OpSpecConstantComposite to normal constants with fixed value(s).
- Fixes issues:
#318: Relicensed under Apache 2.0
v2016.3 2016-08-24
- Add target environment enums for OpenCL 2.1, OpenCL 2.2,
OpenGL 4.0, OpenGL 4.1, OpenGL 4.2, OpenGL 4.3, OpenGL 4.5.
- Add spirv-cfg, an experimental tool to dump the control flow graph
as a GraphiViz "dot" graph
- Add optimization pass: Eliminate dead constants.
- Add filter utility
- Fixes issues:
#288: Check def-use dominance rules for OpPhi (variable,parent) operands
#339: Allow OpUndef in types-constants-global-vars section, as required
by SPIR-V 1.0 Rev7, 1.1 Rev 3.
#340: Avoid race on mkdir during build
#365: Relax PointSize, ClipDistance, CullDistance capability check in all
environments not just Vulkan 1.0.
v2016.2 2016-08-05
- Validator is incomplete
- Checks ID use block is dominated by definition block
- Add optimization passes (in API and spirv-opt command)
- Strip debug info instructions
- Freeze spec constant to their default values
- Allow INotEqual as operation for OpSpecConstantOp
- Fixes bugs:
#270: validator: crash when continue construct is unreachable
#279: validator: infinite loop when analyzing some degenerate control
flow graphs
#286: validator: don't incorrectly generate def-use error for
(variable,parent) parameters to OpPhi
#290: disassembler: never generate bare % for an identifier
#295: validator: def-use dominance check should ignore unreachable uses
#276: validator: allow unreachable continue constructs
#297: validator: allow an unreachable block to branch to a reachable
merge block
v2016.1 2016-07-19
- Fix
Turn off ClipDistance and CullDistance capability checks for Vulkan.
- The disassembler can emit friendly names based on debug info (OpName
instructions), and will infer somewhat friendly names for most types.
This is turned on by default for the spirv-dis command line tool.
- Updated to support SPIR-V 1.1 rev 2
- Input StorageClass, Sampled1D capability, and SampledBuffer capability
do not require Shader capability anymore.
v2016.0 2016-07-04
- Adds v<year>.<index> versioning, with "-dev" indicating
work in progress. The intent is to more easly report
and summarize functionality when SPIRV-Tools is incorporated
in downstream projects.
- Summary of functionality (See the for more):
- Supports SPIR-V 1.1 Rev 1
- Supports SPIR-V 1.0 Rev 5
- Supports GLSL std450 extended instructions 1.0 Rev 3
- Supports OpenCL extended instructions 1.0 Rev 2
- Assembler, disassembler are complete
- Supports floating point widths of 16, 32, 64 bits
- Supports integer widths up to 64 bits
- Validator is incomplete
- Checks capability requirements in most cases
- Checks module layout constraints
- Checks ID use-definition ordering constraints,
ignoring control flow
- Checks some control flow graph rules
- Optimizer is introduced, with few available transforms.
- Supported on Linux, OSX, Android, Windows
- Fixes bugs:
- #143: OpenCL pow and pown arguments