summaryrefslogtreecommitdiff
path: root/source/slang/slang-ir-autodiff-rev.cpp
AgeCommit message (Collapse)Author
2025-07-01extend fiddle to allow custom lua splices in more places (#7559)Ellie Hermaszewska
* Add fkYAML submodule * Generate slang-ir-inst-defs.h from slang-ir-inst-defs.yaml * generate ir-inst-defs.h * neaten things * neaten inst def parser * add rapidyaml submodule * remove fkyaml * remove fkyaml submodule * remove use of ir-inst-defs.h * format and warnings * fix wasm build * tidy * remove rapidyaml * Extend fiddle to allow custom splices in more places * Use lua to describe ir insts * fix * neaten * neaten * neaten * spelling * neaten * comment comment out assert * merge
2025-05-10Add debug information for slang inling (#6621)Mukund Keshava
2025-02-25Fix `UseGraph::replace` (#6395)Sai Praveen Bangaru
* Fix `UseGraph::isTrivial()` test. * Fix. * Fix. * Refactor `UseGraph` and `UseChain` * Update slang-ir-autodiff-primal-hoist.cpp * Update all auto-diff locations that handle pointers to treat user pointers as regular values * Update test to use direct-SPIRV only --------- Co-authored-by: Yong He <yonghe@outlook.com>
2025-01-13Don't initialize temp var for out parameters. (#6076)Sai Praveen Bangaru
2025-01-10Fix `markNonContextParamsAsSideEffectFree`. (#6054)Yong He
2025-01-09[Auto-diff] Overhaul auto-diff type tracking + Overhaul dynamic dispatch for ↵Sai Praveen Bangaru
differentiable functions (#5866) * Overhauled the auto-diff system for dynamic dispatch * More fixes * remove intermediate dumps * Update slang-ast-type.h * More fixes + add a workaround for existential no-diff * Update reverse-control-flow-3.slang * remove dumps * remove more dumps * Delete working-reverse-control-flow-3.hlsl * Cleanup comments + unused variables * More comment cleanup * Add support for lowering `DiffPairType(TypePack)` & `MakePair(MakeValuePack, MakeValuePack)` * Fix array of issues in Falcor tests. * Update slang-ir-autodiff-pairs.cpp * More fixes for Falcor image tests * Small fixups. --------- Co-authored-by: Yong He <yonghe@outlook.com>
2025-01-02Correct IR generation for no-diff pointer type (#5976)kaizhangNV
* Correct IR generation for no-diff pointer type Close #5805 There is an issue on checking whether a pointer type parameter is no_diff, we should first check whether this parameter is an Attribute type first, then check the data type. In the back-propagate pass, for the pointer type parameter, we should load this parameter to a temp variable, then pass it to the primal function call. Otherwise, the temp variable will no be initialized, which will cause the following calculation wrong.
2024-11-05Move switch statement bodies to their own lines (#5493)Ellie Hermaszewska
* Move switch statement bodies to their own lines * format --------- Co-authored-by: Yong He <yonghe@outlook.com>
2024-10-29formatEllie Hermaszewska
* format * Minor test fixes * enable checking cpp format in ci
2024-10-28Replace the word stdlib or standard-library with core-module for source code ↵Jay Kwak
(#5415) This commit changes the word "stdlib" or "standard library" to "core module" in the source code.
2024-10-28Assorted auto-diff enhancements for increased performance & more streamlined ↵Sai Praveen Bangaru
auto-diff results (#5394) * Various AD enhancements * Fix issue with pt-loop test * Update pt-loop.slang * More fixes for perf. Final minimal context test now passes. * Fix issue with loop-elimination pass not running after dce * Try fix wgpu test by removing select operator * Disable wgpu * Delete out.wgsl * Remove comments * Update slang-ir-util.cpp * Fix header relative paths for slang-embed * Disbale wgpu for a few other tests * Better way of determining which params to ignore for side-effects * Update slang-ir-dce.cpp * Fix issue with circular reference from previous AD pass being left behind for the next AD pass * Update slang-ir-dce.cpp
2024-09-19Support `IDifferentiablePtrType` (#5031)Sai Praveen Bangaru
* initial diff-ref-type interface * Initial support for `IDifferentiablePtrType` * Fix unused vars * More tests + fix switch case fallthrough. * Update slang-ir-autodiff.cpp * Update diff-ptr-type-loop.slang * Add optimization to allow more complex pair types * Update slang-ir-autodiff-primal-hoist.cpp * Update diff-ptr-type-loop.slang * Update slang-ir-autodiff-primal-hoist.cpp * More fixes to address reviews * Update slang-check-expr.cpp * Optimizations + rename `differentiableRefInterfaceType` -> `differentiablePtrInterfaceType` * Move pair logic to ir-builder, unify the type dictionaries. --------- Co-authored-by: Yong He <yonghe@outlook.com>
2024-09-18Report AD checkpoint contexts (#5058)venkataram-nv
* Transferring source locations when creating phi instructions * Tracking for simple variables * Deriving source locations for loop counters * Printing checkpoint structure breakdown * More readable output format * Special behavior for loop counters * Writing report to file * Add slangc option to enable checkpoint reports * Display types of checkpointed fields * Message in case there are no checkpointing contexts * Catch source locations for function calls * Source cleanup * Fix compilation warnings * Remove stray dump() * Provide the report through diagnostic notes * Add missing path for sourceLoc during unzip pass * Add tests for reporting intermediates * Include more transfer cases for source locations * Fix ordering in address elimination * Fill in more holes with source location transfer * Remove debugging line * Reverting changes to diagnostic sink * Simplify address elimination using source location RAII contexts * Eliminating manual source loc transfers in forward transcription * Fix local var adaptation to use RAII location setter * Simplify primal hoisting logic for source location transfer * Simplify unzipping with RAII location scopes * Simplify transpose logic * Cleaning up for rev.cpp * Reverting spacing changes * Fix mistake with source loc RAII instantiation * Fix formatting issues
2024-08-27Adds a warning for using `[PreferRecompute]` on methods that may contain ↵Sai Praveen Bangaru
side effects (#4707) * Adds a warning for using prefer-recompute on methods that contain side effects * Rename `SideEffects` -> `SideEffectBehavior` --------- Co-authored-by: Yong He <yonghe@outlook.com>
2024-02-28[SPIRV] Add NonSemanticDebugInfo for step-through debugging. (#3644)Yong He
* [SPIRV] Add NonSemanticDebugInfo for step-through debugging. * Fix. * Fix.
2023-09-28Support `constref` parameters passing. (#3249)Yong He
* Support `constref` parameters passing. * Fix. * Fix. * Add test and diagnostic on mix use of __constref and no_diff. * check for [constref] on differentiable member method. --------- Co-authored-by: Yong He <yhe@nvidia.com>
2023-09-21Revert inlining change in #3217. (#3229)Yong He
Co-authored-by: Yong He <yhe@nvidia.com>
2023-09-20Move force inlining step to before `processAutodiffCalls` (and run in loop) ↵Sai Praveen Bangaru
(#3217) * Move auto-diff force inlining step to before `processAutodiffCalls` * Fix `replaceUsesWith` to handle existing inst defined after current use. * Fix. --------- Co-authored-by: Yong He <yhe@nvidia.com>
2023-09-19Added `[AutoPyBindCUDA]` for automatic kernel binding + `[PyExport]` for ↵Sai Praveen Bangaru
exporting type information (#3209) * Initial: add a DiffTensor impl * Auto-binding and diff tensor implementations now work * Refactored diff-tensor implementation + added py-export for struct types * Cleanup * Update slang-ir-pytorch-cpp-binding.cpp * Updated test names * Update autodiff-data-flow.slang.expected * Add more versions of load/store & default generic args for DiffTensorView. * Add diagnostic for default generic arg and more tests * Add more `[AutoPyBind]` tests
2023-09-08Don't inline callees with custom derivative before autodiff. (#3196)Yong He
Co-authored-by: Yong He <yhe@nvidia.com>
2023-08-17Initial support for differentiating existential types (#3111)Sai Praveen Bangaru
* Merge * WIP: Complete auto-diff logic for existential types * Revert "Add compiler option for generating representative hash" This reverts commit 13b09ef4621e73844c96d64d9c111a8ed0d45aae. * More fixes for fwd-mode AD on existential types * Add anyValueSize inference pass * Fix checking of `Differential.Differential==Differential` * In-progress: infer any-value-size for existential types * Existentials now work in forward-mode * Overhaul handling of existential AD types. Fwd-mode works, reverse-mode requires front-end changes * Reverse-mode now works on existentials * Cleanup * Remove diff rules for create existential object for now * Revert treat-as-differentiable changes * Fixes * More fixes * Cleanup * more cleanup * signed/unsigned * Revert "Cleanup" This reverts commit e4f7d71f07bb207736f90708961eeecd09a1b652. * Cleanup (again) * Remove public/export/keep-alive on null differential after AD pass * Minor fix * Update dictionary accessors * Keep export decoration * More fixes + Support for `kIROp_PackAnyValue` * Merge upstream * Update expected-failure.txt
2023-08-16Use ankerl/unordered_dense as a hashmap implementation (#3036)Ellie Hermaszewska
* Correct namespace for getClockFrequency * missing const * Add missing assignment operator * Remove unused variables * Return correct modified variable * Use stable hash code for file system identity * terse static_assert * Structured binding for map iteration * Make (==) and getHashCode const on many structs * Add ConstIterator for LinkedList * Replace uses of ItemProxy::getValue with Dictionary::at * Extract list of loads from gradientsMap before updating it * Const correctness in type layout * Add unordered_dense hashmap submodule * Use wyhash or getHashCode in slang-hash.h * refactor slang-hash.h * Use ankerl/unordered_dense as a hashmap implementation Notable changes: - The subscript operator returns a reference directly to the value, rather than a lazy ItemProxy (pair of dict pointer and key) slang-profile time (95% over 10 runs): - Before: 6.3913906 (±0.0746) - After: 5.9276123 (±0.0964) * 64 bit hash for strings So they have the same hash as char buffers with the same contents * Narrowing warnings for gcc to match msvc * revert back to c++17 * Correct c++ version for msvc * Use path to unordered_dense which keeps tests happy * Do not assign to and read from map in same expression * Remove redundant map operations in primal-hoist * Split out stable hash functions into slang-stable-hash.h * 64 bit hash by default * regenerate vs projects * Correct return type from HashSetBase::getCount() * correct width for call to Dictionary::reserve * Use stable hash for obfuscated module ids * Signed int for reserve * clearer variable naming * Parameterize Dictionary on hash and equality functors * Allow heterogenous lookup for Dictionary * missing const * Use set over operator[] in some places * Remove unused function * s/at/getValue
2023-08-14Add auto-diff support for `IRSwizzleStore` (#3102)Sai Praveen Bangaru
* Add auto-diff support for `IRSwizzleStore` - Lower IRSwizzleStore to multiple stores during AD preprocess. - Fix typo in `transcribeNonDiffInst` * Remove unnecessary file & add more robust check for 'local' addresses * Fix. * Update slang-ir-autodiff-fwd.cpp --------- Co-authored-by: Yong He <yonghe@outlook.com>
2023-05-26Fix bug in legalizeFuncType that leads to invalid IR. (#2902)Yong He
* Fix bug in legalizeFuncType that leads to invalid IR. * Diagnose on functions that never returns when differentiate it. --------- Co-authored-by: Yong He <yhe@nvidia.com>
2023-05-05Fix checkpoint hints for higher-order autodiff (#2869)Sai Praveen Bangaru
2023-04-26Fix most of the disabled warnings on gcc/clang (#2839)Ellie Hermaszewska
2023-04-26Autodiff support for dynamically dispatched generic method. (#2846)Yong He
* Autodiff support for dynamically dispatched generic method. * Fix. * Support dynamically dispatched generic type. --------- Co-authored-by: Yong He <yhe@nvidia.com>
2023-04-25Support recomputing phi params in bwd prop func. (#2841)Yong He
2023-04-25Cleanup checkpointing policy impl. (#2837)Yong He
* Cleanup checkpointing policy impl. * More cleanup. --------- Co-authored-by: Yong He <yhe@nvidia.com>
2023-04-25Dictionary using lowerCamel (#2835)jsmall-nvidia
* #include an absolute path didn't work - because paths were taken to always be relative. * WIP lowerCamel Dictionary. * WIP more lowerCamel fixes for Dictionary. * Add/Remove/Clear * GetValue/Contains * Fix tabs in dictionary. Count -> getCount * Fix fields with caps. * Key -> key Value -> value Use m_ for members where appropriate. Use lowerCamel in linked list. * Some small fixes/improvements to Dictionary. * Kick CI.
2023-04-24Change AD checkpointing policy to recompute more. (#2836)Yong He
Co-authored-by: Yong He <yhe@nvidia.com>
2023-04-21Refactor checkpointing policy and availability pass. (#2826)Yong He
Co-authored-by: Yong He <yhe@nvidia.com>
2023-04-10Fix inlining. (#2786)Yong He
Co-authored-by: Yong He <yhe@nvidia.com>
2023-03-30Fix autodiff pass duplicates exported functions. (#2759)Yong He
Co-authored-by: Yong He <yhe@nvidia.com>
2023-03-22Type legalization and autodiff bug fixes. (#2722)Yong He
* Bug fixes. * Fix. * Only perform autodiff for functions whose derivative is actually used. * Fix loop optimize bug. * Fix high order diff. * Fix trivial diff func generation. * Fixes. * Cleanup. --------- Co-authored-by: Yong He <yhe@nvidia.com>
2023-03-21[TreatAsDifferentiable] functions. (#2720)Yong He
2023-03-16Fix Phi simplification bug. (#2710)Yong He
* Fix Phi simplification bug. * Fix up. * Fix. * Fix. * Fix. * Fix. * Fix. * Fix test. * Fix test. --------- Co-authored-by: Yong He <yhe@nvidia.com>
2023-03-15AD: Primal-Hoisting Rework + Checkpoint Policy Framework (#2702)Sai Praveen Bangaru
2023-03-13Support high order diff pattern: `bwd_diff(fwd_diff(f))`. (#2695)Yong He
* Support high order diff pattern: `bwd_diff(fwd_diff(f))`. * Fix. --------- Co-authored-by: Yong He <yhe@nvidia.com>
2023-03-08Add support for `[PrimalSubstitute]` and `[PrimalSubstituteOf]`. (#2691)Yong He
* Add support for `[PrimalSubstitute]` and `[PrimalSubstituteOf]`. * Fix * Fix. * Cleanup. --------- Co-authored-by: Yong He <yhe@nvidia.com>
2023-02-27More fixes for reverse-mode on complicated loops (#2675)Sai Praveen Bangaru
* Multiple fixes to get various loop tests to pass. * Create reverse-nested-loop.slang * Fix for variables becoming inaccessible during cfg normalization * Removed comments and moved break-branch-normalization to eliminateMultiLevelBreaks * Fix. * Override liveness tests
2023-02-22Reverse-mode AD fixes for loops with non-trivial break region (#2671)Sai Praveen Bangaru
* Fix crash when applying autodiff to functions with no arguments * Fixes for loops where the break region is non-trivial * Minor fix * Implement array legalization correctly. * Fix array legalization. --------- Co-authored-by: Yong He <yhe@nvidia.com>
2023-02-20Miscellaneous backward autodiff fixes. (#2665)Yong He
* Fix differentiable type registration * Fix use of non-differentiable return value in a differentiable func. * Fix use of primal inst that does not dominate the diff block. * Fix primal inst hoisting, and add missing type legalization logic. * Make `detach` defined on all differentiable T. --------- Co-authored-by: Yong He <yhe@nvidia.com>
2023-02-17AD: More legacy type handling cleanup + user-defined reverse-mode fix (#2662)Sai Praveen Bangaru
* WIP: Remove all legacy type checking * Fixed issue with user-defined backward derivatives not bypassing the AD process --------- Co-authored-by: Yong He <yonghe@outlook.com>
2023-02-16Remove `SharedIRBuilder`. (#2657)Yong He
Co-authored-by: Yong He <yhe@nvidia.com>
2023-02-13Various auto-diff bug fixes. (#2646)Yong He
Co-authored-by: Yong He <yhe@nvidia.com>
2023-02-09Reverse-mode Loop Support (#2635)Sai Praveen Bangaru
* Full loop support now working. MaxItersAttr in progress * Lookup table updates? * Fixed the max iters decoration * Minox fixes & remove superfluous code * fixup warnings * Revert "Lookup table updates?" This reverts commit 7d9b0793fb5239f31d1155776e846dcf1892d8d9. * Update 07-autodiff.md * Change maxiters to MaxIters * Added asserts * Update 07-autodiff.md
2023-02-07Arithmetic simplifications and more IR clean up logic. (#2632)Yong He
2023-02-04Patch transcription of `inout` non differentiable params. (#2623)Yong He
2023-02-03Overhaul `transposeParameterBlock` to support `inout` params. (#2621)Yong He
* Overhaul `transposeParameterBlock` to support `inout` params. * Small bug fixes. * Bug fix on differentiable intrinsic specialization. * Fixes. * Run autodiff tests on CPU. * Clean up. * More bug fixes., * Add test coverage on inout param. * Fix language server hinting for transcribed mutable params. --------- Co-authored-by: Yong He <yhe@nvidia.com>
generated by cgit v1.2.3 (git 2.39.1) at 2026-07-16 07:55:53 +0000