d8a40ab | Yong He | 22 March 2023, 04:29:13 UTC | [TreatAsDifferentiable] functions. (#2720) | 22 March 2023, 04:29:13 UTC |
8387673 | Yong He | 22 March 2023, 03:03:12 UTC | Fix associated type resolution bug. (#2719) * Fix associated type resolution bug. * Fix. * Fix language server hinting messed up by breadcrumb nodes. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 22 March 2023, 03:03:12 UTC |
5d2c5dd | Yong He | 21 March 2023, 22:44:43 UTC | Fix IntVal type check. (#2718) * Fix IntVal type check. * add test case result. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 21 March 2023, 22:44:43 UTC |
96caba7 | Yong He | 21 March 2023, 22:44:21 UTC | Add texture tri-linear autodiff example. (#2715) * Add quad texture example. * delete output image * remove irrelavent files * update project files * fix * Update example. * Fix. * remove out-texture --------- Co-authored-by: Yong He <yhe@nvidia.com> | 21 March 2023, 22:44:21 UTC |
7f11f88 | Yong He | 17 March 2023, 22:57:22 UTC | Add support for emitting cuda kernel and host functions. (#2712) * Add support for emitting cuda kernel and host functions. * Update test. * Fix cuda preamble emit. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 17 March 2023, 22:57:22 UTC |
4b55bf6 | Yong He | 17 March 2023, 22:14:44 UTC | Rework differentiability dataflow check. (#2711) Co-authored-by: Yong He <yhe@nvidia.com> | 17 March 2023, 22:14:44 UTC |
29abe39 | Yong He | 17 March 2023, 17:53:48 UTC | Update 07-autodiff.md | 17 March 2023, 17:53:48 UTC |
8a61b9d | jsmall-nvidia | 17 March 2023, 15:05:15 UTC | Support for producing SourceMap on emit (#2707) * #include an absolute path didn't work - because paths were taken to always be relative. * WIP source map. * Split out handling of RttiTypeFuncs to a map type. * Make RttiTypeFuncsMap hold default impls. * Slightly more sophisticated RttiTypeFuncsMap * Source map decoding. * Fix tabs. * Fix asserts due to negative values. * Use less obscure mechanisms in SourceMap. * Source map decoding. Simplifying SourceMap usage. * First attempt at ouputting a source map as part of emit. * Added support for -source-map option. SourceMap is added to the artifact. | 17 March 2023, 15:05:15 UTC |
9476d45 | Yong He | 17 March 2023, 06:46:14 UTC | Fix Phi simplification bug. (#2710) * Fix Phi simplification bug. * Fix up. * Fix. * Fix. * Fix. * Fix. * Fix. * Fix test. * Fix test. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 17 March 2023, 06:46:14 UTC |
77d3630 | Yong He | 17 March 2023, 05:17:51 UTC | Fix name mangling of `FuncCallIntVal` in func signatures. (#2709) * Fix name mangling of `FuncCallIntVal` in func signatures. * Enhance the test. * Fix. * Fix. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 17 March 2023, 05:17:51 UTC |
fc9cba5 | Yong He | 17 March 2023, 05:17:34 UTC | Add `[CudaDeviceExport]` to allow exporting CUDA device functions. (#2708) * Add `[CudaDeviceExport]` to allow exporting CUDA device functions. * Fix. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 17 March 2023, 05:17:34 UTC |
2fd1ac6 | jsmall-nvidia | 17 March 2023, 05:13:59 UTC | SourceMap encoding (#2706) * #include an absolute path didn't work - because paths were taken to always be relative. * WIP source map. * Split out handling of RttiTypeFuncs to a map type. * Make RttiTypeFuncsMap hold default impls. * Slightly more sophisticated RttiTypeFuncsMap * Source map decoding. * Fix tabs. * Fix asserts due to negative values. * Use less obscure mechanisms in SourceMap. * Source map decoding. Simplifying SourceMap usage. | 17 March 2023, 05:13:59 UTC |
4cb899f | jsmall-nvidia | 16 March 2023, 19:19:20 UTC | Preliminary SourceMap support (#2701) * #include an absolute path didn't work - because paths were taken to always be relative. * WIP source map. * Split out handling of RttiTypeFuncs to a map type. * Make RttiTypeFuncsMap hold default impls. * Slightly more sophisticated RttiTypeFuncsMap * Source map decoding. * Fix tabs. * Fix asserts due to negative values. * Use less obscure mechanisms in SourceMap. | 16 March 2023, 19:19:20 UTC |
1036d1a | Yong He | 16 March 2023, 18:06:01 UTC | Add test case for `makeStruct` transcription. (#2705) * Add test case for `makeStruct` transcription. * Update documentation on `no_diff`. * Update toc --------- Co-authored-by: Yong He <yhe@nvidia.com> | 16 March 2023, 18:06:01 UTC |
b92f280 | Samuel Kogler | 16 March 2023, 18:05:33 UTC | Support GL_EXT_fragment_shader_barycentric (#2704) * Support GL_EXT_fragment_shader_barycentric * Support pervertex with GL_EXT_fragment_shader_barycentric | 16 March 2023, 18:05:33 UTC |
e960534 | Yong He | 16 March 2023, 06:31:09 UTC | Update 07-autodiff.md | 16 March 2023, 06:31:09 UTC |
f7431f9 | Yong He | 16 March 2023, 06:26:14 UTC | Fix `transcribeConstruct` for `makeStruct`. (#2703) Co-authored-by: Yong He <yhe@nvidia.com> | 16 March 2023, 06:26:14 UTC |
3a1a6c4 | Yong He | 16 March 2023, 03:53:48 UTC | Extend DCE to handle pure function calls that returns via `inout` parameters. (#2700) | 16 March 2023, 03:53:48 UTC |
71efd94 | Sai Praveen Bangaru | 16 March 2023, 02:26:58 UTC | AD: Primal-Hoisting Rework + Checkpoint Policy Framework (#2702) | 16 March 2023, 02:26:58 UTC |
38e6219 | Yong He | 15 March 2023, 18:12:15 UTC | Update 07-autodiff.md | 15 March 2023, 18:12:15 UTC |
bf30824 | Yong He | 15 March 2023, 16:39:21 UTC | Properly implement differential witness of intermediate context type. (#2699) * Properly implement differential witness of intermediate context type. * Modify test to include a loop. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 15 March 2023, 16:39:21 UTC |
176eaa9 | Yong He | 14 March 2023, 16:33:16 UTC | Update 07-autodiff.md | 14 March 2023, 16:33:16 UTC |
e291f60 | Yong He | 14 March 2023, 16:31:51 UTC | Support `fwd_diff(bwd_diff(f))`. (#2697) * Support `fwd_diff(bwd_diff(f))`. * Fix. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 14 March 2023, 16:31:51 UTC |
a911ca6 | Yong He | 13 March 2023, 17:57:28 UTC | Support high order diff pattern: `bwd_diff(fwd_diff(f))`. (#2695) * Support high order diff pattern: `bwd_diff(fwd_diff(f))`. * Fix. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 13 March 2023, 17:57:28 UTC |
3fea56e | jsmall-nvidia | 10 March 2023, 19:19:48 UTC | Support for PDB output from DXC (#2693) * #include an absolute path didn't work - because paths were taken to always be relative. * Add versioning to CompileOptions for DownstreamCompiler so we can add new options without breaking binary interface. * Add support for debug info format to API/command line processing. * Small simplification. * Add support for adding PDB output from a compilation. * Use builtin offset of directly. * Fix typo in debug. | 10 March 2023, 19:19:48 UTC |
e39893e | jsmall-nvidia | 10 March 2023, 17:24:51 UTC | Add struct version to DownstreamCompiler::CompileOptions interface (#2692) * #include an absolute path didn't work - because paths were taken to always be relative. * Add versioning to CompileOptions for DownstreamCompiler so we can add new options without breaking binary interface. * Use builtin offset of directly. | 10 March 2023, 17:24:51 UTC |
e06cfb3 | Yong He | 09 March 2023, 23:53:51 UTC | Update restrictions sections in autodiff doc. (#2694) * Update restrictions section in autodiff doc. * Update restrictions sections in autodiff doc. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 09 March 2023, 23:53:51 UTC |
a611d4b | Yong He | 09 March 2023, 23:18:36 UTC | Update user guide on `[PrimalSubstitute]` | 09 March 2023, 23:18:36 UTC |
86fc50c | Yong He | 09 March 2023, 05:52:34 UTC | Add support for `[PrimalSubstitute]` and `[PrimalSubstituteOf]`. (#2691) * Add support for `[PrimalSubstitute]` and `[PrimalSubstituteOf]`. * Fix * Fix. * Cleanup. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 09 March 2023, 05:52:34 UTC |
257733f | Yong He | 07 March 2023, 19:22:32 UTC | Reuse higher-order `ResolveInvoke` logic to resolve func refs in `[*DerivativeOf]` attribs. (#2688) * Reuse higher-order `ResolveInvoke` logic to resolve func refs in [*DerivativeOf] attribs. * Add diff implementation matrix versions of binary and ternary intrinsics. * Add diff impl for legacy intrinsics. * Fix diagnostics of using non-differentiable function in a diff operator. * Add diff implementation for `determinant`. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 07 March 2023, 19:22:32 UTC |
6f31eae | Yong He | 01 March 2023, 21:19:33 UTC | Implement derivatives for HLSL intrinsics. (#2684) * Implement derivatives for HLSL intrinsics. * Vector intrinsics. * Add all intrinsics. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 01 March 2023, 21:19:33 UTC |
6c26aa1 | Yong He | 01 March 2023, 20:59:51 UTC | Improve diagnostic on differentiablitiy check. (#2687) Co-authored-by: Yong He <yhe@nvidia.com> | 01 March 2023, 20:59:51 UTC |
3c32dd9 | Sai Praveen Bangaru | 01 March 2023, 02:24:24 UTC | AD: Fixed do-while loops (#2683) * WIP: Fix for do-while loops * Added a somewhat hacky fix for do-while loops * Redid the indexed region map builder step to fix issue with the nested loops test * rename * Used managed pointers | 01 March 2023, 02:24:24 UTC |
7eeda30 | Theresa Foley | 28 February 2023, 23:10:14 UTC | Fixes to GitHub Windows CI build (#2664) | 28 February 2023, 23:10:14 UTC |
f7db44a | Yong He | 28 February 2023, 23:09:19 UTC | Add `SourceBlobWithPathInfoArtifactRepresentation`. (#2682) | 28 February 2023, 23:09:19 UTC |
f23e362 | Yong He | 28 February 2023, 05:21:39 UTC | Diagnose on storing differentiable value into non-differentiable location. (#2681) | 28 February 2023, 05:21:39 UTC |
10e2d9c | Sai Praveen Bangaru | 28 February 2023, 04:42:06 UTC | More fixes for reverse-mode on complicated loops (#2675) * 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 | 28 February 2023, 04:42:06 UTC |
a3ba22b | Yong He | 27 February 2023, 23:18:07 UTC | Detect and deduplicate read-only resource access. (#2680) * Detect and deduplicate read-only resource access. * Fix tests. * Fix tests. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 27 February 2023, 23:18:07 UTC |
b1b76f0 | Sai Praveen Bangaru | 25 February 2023, 07:23:41 UTC | Added support for `kIROp_Select` (#2679) | 25 February 2023, 07:23:41 UTC |
85c1569 | Yong He | 24 February 2023, 22:33:32 UTC | Support dynamic dispatch a backward differentiable function. (#2678) Co-authored-by: Yong He <yhe@nvidia.com> | 24 February 2023, 22:33:32 UTC |
91694da | Yong He | 24 February 2023, 21:22:07 UTC | Fix differential type registration through non-differentiable type. (#2677) * Fix differential type registration through non-differentiable type. * More fix. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 24 February 2023, 21:22:07 UTC |
bd6306c | Yong He | 24 February 2023, 18:01:47 UTC | More control flow simplifications. (#2673) * More control flow and Phi param simplifications. * Fix. * Fix gcc error. * Fix. * More IR cleanup. * Fix bug in phi param dce + ifelse simplify. * Propagate and DCE side-effect-free functions. * Enhance CFG simplifcation to remove loops with no side effects. * Fix. * Fixes. * Fix tests. Add [__AlwaysFoldIntoUseSite] for rayPayloadLocation. * More cleanup. * Fixes. * Fix. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 24 February 2023, 18:01:47 UTC |
e8c08e7 | Sai Praveen Bangaru | 23 February 2023, 03:22:26 UTC | Fixed hoisting of intermediate array & context vars (#2674) Also added legalization for loops | 23 February 2023, 03:22:26 UTC |
6eb0b4d | Sai Praveen Bangaru | 23 February 2023, 00:33:42 UTC | Reverse-mode AD fixes for loops with non-trivial break region (#2671) * 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> | 23 February 2023, 00:33:42 UTC |
0ef7aa8 | Yong He | 21 February 2023, 20:51:46 UTC | Fix transposeCall. (#2669) * Modify control-flow test case * Update reverse-control-flow-3.slang * Fix `transposeCall`. * Fix. --------- Co-authored-by: Sai Praveen Bangaru <31557731+saipraveenb25@users.noreply.github.com> Co-authored-by: Yong He <yhe@nvidia.com> | 21 February 2023, 20:51:46 UTC |
6bca0ec | Sai Praveen Bangaru | 21 February 2023, 16:37:15 UTC | Added support for simple while loops (#2667) * Added support for simple while loops * Fix support for while loops by changing logic to grab the loop update block | 21 February 2023, 16:37:15 UTC |
e5bec2f | Ellie Hermaszewska | 21 February 2023, 03:46:21 UTC | Take into account the optimization level requested in TargetDesc (#2649) * Take into account the optimization level requested in TargetDesc Combines conflicting optimization levels by taking the maximum between them; as such they now each represent the lower bound on the eventual optimization level. Most of the changes are making this clear by s/optimizationLevel/minimumOptimizationLevel Closes https://github.com/shader-slang/slang/issues/2324 * Remove optimization level from TargetDesc | 21 February 2023, 03:46:21 UTC |
47715e6 | Yong He | 20 February 2023, 22:42:50 UTC | Miscellaneous backward autodiff fixes. (#2665) * 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> | 20 February 2023, 22:42:50 UTC |
8b05df4 | Yong He | 20 February 2023, 18:17:00 UTC | Add static for loop iteration inference. (#2659) | 20 February 2023, 18:17:00 UTC |
a8da735 | Sai Praveen Bangaru | 18 February 2023, 00:03:52 UTC | Allocate N+1 arrays instead of N to avoid out-of-bounds access when unzipping loops (#2663) | 18 February 2023, 00:03:52 UTC |
92ccc8f | Sai Praveen Bangaru | 17 February 2023, 23:02:58 UTC | AD: More legacy type handling cleanup + user-defined reverse-mode fix (#2662) * 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> | 17 February 2023, 23:02:58 UTC |
5cd39d1 | Sai Praveen Bangaru | 17 February 2023, 22:56:07 UTC | AD: Remove the original loop condition upon inversion (#2661) * Remove the original condition upon loop inversion (it's redundant, and causes out-of-bounds accesses) * minor fix (also removed the first loop check skip) * Cleanup unused insts * minor comment fix | 17 February 2023, 22:56:07 UTC |
0516073 | Yong He | 17 February 2023, 21:23:27 UTC | Fixed crash when lowering IR for no_diff struct member. (#2658) * Fixed crash when lowering IR for no_diff struct member. * Improve `setInsertBeforeOrdinaryInst` and `setInsertAfterOrdinaryInst`. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 17 February 2023, 21:23:27 UTC |
79049bc | Sai Praveen Bangaru | 17 February 2023, 21:22:47 UTC | Cleaned up legacy differential type handling + type casting bugfixes (#2660) | 17 February 2023, 21:22:47 UTC |
f253d15 | Sai Praveen Bangaru | 17 February 2023, 17:03:59 UTC | Proper reverse-mode loop handling with splitting + inversion steps (#2656) * Halfway to loop inversion * More progress towards proper loop inversion * More progress towards inverse insts. Only thing left is adding `counter>=0` at the right place * More fixes for inversion step. * Lots more fixes, added primal inst 'hoisting' mechanism as the central method that ensures primal values are placed in the right spot * Loop inversion is now functional * Cleaned up commented code * rename diffCounterVar -> diffCounterParam * minor update * removed some comments and commented code * Switch `IRBuilder(sharedIRBuilder)` to `IRBuilder(moduleInst)` | 17 February 2023, 17:03:59 UTC |
245466d | Yong He | 17 February 2023, 00:44:04 UTC | Remove `SharedIRBuilder`. (#2657) Co-authored-by: Yong He <yhe@nvidia.com> | 17 February 2023, 00:44:04 UTC |
4c4826d | Yong He | 16 February 2023, 21:55:32 UTC | Overhaul global inst deduplication and cpp/cuda backend. (#2654) * Overhaul global inst deduplication and cpp/cuda backend. * Update IR documentation. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 16 February 2023, 21:55:32 UTC |
eda88e5 | Yong He | 16 February 2023, 20:44:32 UTC | Design doc on IR deduplication and best practices when working with IR | 16 February 2023, 20:44:32 UTC |
8667515 | Yong He | 16 February 2023, 01:18:00 UTC | Treat user defined backward derivative function as non differentiable. (#2650) Co-authored-by: Yong He <yhe@nvidia.com> | 16 February 2023, 01:18:00 UTC |
266dc66 | jsmall-nvidia | 15 February 2023, 21:02:11 UTC | Upgrade to GLSLANG 12.0.0 binaries (#2652) * #include an absolute path didn't work - because paths were taken to always be relative. * Upgrade GLSLANG binaries to 12.0.0 | 15 February 2023, 21:02:11 UTC |
f13e080 | jsmall-nvidia | 15 February 2023, 19:32:50 UTC | Upgrade GLSLANG 12.0.0 (#2651) * #include an absolute path didn't work - because paths were taken to always be relative. * Update to glslang 12.0.0. Update SPIRV-Tools SPIRV-Headers. | 15 February 2023, 19:32:50 UTC |
598e07f | jsmall-nvidia | 14 February 2023, 23:30:04 UTC | Preliminary Shader Execution Reordering Doc (#2648) * #include an absolute path didn't work - because paths were taken to always be relative. * Add preliminary Shader Execution Reordering doc. Update target-compatibility docs. * Fix debugBreak. | 14 February 2023, 23:30:04 UTC |
b92a75d | jsmall-nvidia | 14 February 2023, 21:21:07 UTC | Preliminary debugBreak support (#2647) * #include an absolute path didn't work - because paths were taken to always be relative. * Preliminary support for debug break. * Add C++ debug break support. Add details about usage. * Improve debug break test details. * Make HLSL output a comment about no support. * Handle specialize for target assert, without a body if it has spv_instruction/target intrinsic | 14 February 2023, 21:21:07 UTC |
ec49215 | Yong He | 13 February 2023, 19:05:29 UTC | Various auto-diff bug fixes. (#2646) Co-authored-by: Yong He <yhe@nvidia.com> | 13 February 2023, 19:05:29 UTC |
977eb92 | Yong He | 13 February 2023, 18:39:12 UTC | Eliminate `continue` to allow unrolling any loops. (#2645) Co-authored-by: Yong He <yhe@nvidia.com> | 13 February 2023, 18:39:12 UTC |
4dbc74a | Yong He | 13 February 2023, 18:38:14 UTC | Add Loop Unrolling Pass. (#2644) Co-authored-by: Yong He <yhe@nvidia.com> | 13 February 2023, 18:38:14 UTC |
57af2c1 | Yong He | 12 February 2023, 06:48:18 UTC | Update README.md on auto diff feature. | 12 February 2023, 06:48:18 UTC |
77706b7 | Yong He | 11 February 2023, 21:30:18 UTC | Update 07-autodiff.md | 11 February 2023, 21:30:18 UTC |
5bab4ea | Yong He | 11 February 2023, 21:20:37 UTC | Update 07-autodiff.md | 11 February 2023, 21:20:37 UTC |
82c7c78 | Yong He | 11 February 2023, 21:19:15 UTC | Update 07-autodiff.md | 11 February 2023, 21:19:15 UTC |
94b2c67 | Ellie Hermaszewska | 11 February 2023, 05:57:35 UTC | Take into account existing initializer list type when performing coercions (#2641) Fixes https://github.com/shader-slang/slang/issues/2189 | 11 February 2023, 05:57:35 UTC |
c7f486c | Ellie Hermaszewska | 11 February 2023, 04:16:56 UTC | Comment call to vkCreateInstance with a potential pitfall (#2642) Although we could in principle write this explanatory message to stderr, that would entangle this call with the layer search above for what is probably a very unlikely possibility on any normal system. | 11 February 2023, 04:16:56 UTC |
aec57d8 | Yong He | 11 February 2023, 02:46:57 UTC | Fix several autodiff bugs. (#2643) | 11 February 2023, 02:46:57 UTC |
6e7b424 | Yong He | 10 February 2023, 17:01:59 UTC | Fix checking of `[BackwardDerivativeOf]` attribute. (#2640) * Fix checking of `[BackwardDerivativeOf]` attribute. * Fix crash in `canInstHaveSideEffectAtAddress`. * Fix. * Revert fix. * Fix. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 10 February 2023, 17:01:59 UTC |
df02f3f | Sai Praveen Bangaru | 09 February 2023, 22:40:20 UTC | Reverse-mode Loop Support (#2635) * 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 | 09 February 2023, 22:40:20 UTC |
d911e1b | Sai Praveen Bangaru | 09 February 2023, 22:19:55 UTC | Fixed derivatives for kIROp_Neg and kIROp_Div, added another test (#2639) | 09 February 2023, 22:19:55 UTC |
fbe31ad | Ellie Hermaszewska | 09 February 2023, 05:16:30 UTC | Use stable sort in generation of lookup tables (#2638) * Add Slang::List::stableSort * Use stable sort in generation of lookup tables * Disable newline translation when writing lookup tables | 09 February 2023, 05:16:30 UTC |
6bbd673 | Yong He | 08 February 2023, 21:35:43 UTC | Update 07-autodiff.md | 08 February 2023, 21:35:43 UTC |
f2e564c | Yong He | 08 February 2023, 21:20:17 UTC | Replace \cal with \mathbb (#2637) Co-authored-by: Yong He <yhe@nvidia.com> | 08 February 2023, 21:20:17 UTC |
80b1b37 | Yong He | 08 February 2023, 21:04:32 UTC | Update autodiff documentation with more precise math definitions. (#2636) Co-authored-by: Yong He <yhe@nvidia.com> | 08 February 2023, 21:04:32 UTC |
b1d7dc0 | winmad | 08 February 2023, 06:30:00 UTC | Add backward derivatives for functions in diff.meta.slang (#2633) * WIP: start adding backward derivatives * Overhaul `transposeParameterBlock` to support `inout` params. * Small bug fixes. * Bug fix on differentiable intrinsic specialization. * Fixes. * Run autodiff tests on CPU. * Clean up. * 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., * WIP: working on detach * Arithmetic simplifications and more IR clean up logic. * WIP: adding detach and abs * Fix detach and abs * Fix. * Add IR transform pass for cleaner code emit. * Fix test cases. * Fix type system logic for reference type. * Add backward derivatives for functions that already have forward derivatives * Fix changes --------- Co-authored-by: Yong He <yhe@nvidia.com> Co-authored-by: Lifan Wu <lifanw@nvidia.com> | 08 February 2023, 06:30:00 UTC |
4be623c | Yong He | 08 February 2023, 02:36:35 UTC | Arithmetic simplifications and more IR clean up logic. (#2632) | 08 February 2023, 02:36:35 UTC |
101f164 | Yong He | 07 February 2023, 00:11:49 UTC | Update documentation (#2631) Co-authored-by: Yong He <yhe@nvidia.com> | 07 February 2023, 00:11:49 UTC |
f12d422 | Yong He | 06 February 2023, 23:56:48 UTC | Fix documentation (#2630) Co-authored-by: Yong He <yhe@nvidia.com> | 06 February 2023, 23:56:48 UTC |
49d68bf | Yong He | 06 February 2023, 23:45:17 UTC | Update documentation TOC (#2629) Co-authored-by: Yong He <yhe@nvidia.com> | 06 February 2023, 23:45:17 UTC |
fe5cd24 | Yong He | 06 February 2023, 23:37:47 UTC | Improve autodiff documentation (#2628) Co-authored-by: Yong He <yhe@nvidia.com> | 06 February 2023, 23:37:47 UTC |
7ecdf54 | Yong He | 06 February 2023, 22:44:19 UTC | Fixup documentation (#2627) Co-authored-by: Yong He <yhe@nvidia.com> | 06 February 2023, 22:44:19 UTC |
e84c891 | Yong He | 06 February 2023, 22:35:27 UTC | Add documentation for autodiff feature (#2626) Co-authored-by: Yong He <yhe@nvidia.com> | 06 February 2023, 22:35:27 UTC |
5ede9a3 | Yong He | 06 February 2023, 22:34:19 UTC | GFX: make dispatch commands return error code. (#2625) * GFX: make dispatch commands return error code. * Fix cuda. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 06 February 2023, 22:34:19 UTC |
e893a83 | Yong He | 06 February 2023, 18:07:02 UTC | Fix crash when processing nested switch. (#2624) * Fix crash when processing nested switch. * Clean up. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 06 February 2023, 18:07:02 UTC |
a12c551 | Yong He | 05 February 2023, 04:07:14 UTC | Patch transcription of `inout` non differentiable params. (#2623) | 05 February 2023, 04:07:14 UTC |
228e71d | Yong He | 04 February 2023, 00:44:33 UTC | Overhaul `transposeParameterBlock` to support `inout` params. (#2621) * 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> | 04 February 2023, 00:44:33 UTC |
ee49a62 | jsmall-nvidia | 03 February 2023, 22:11:12 UTC | Small fixes around repro (#2622) * #include an absolute path didn't work - because paths were taken to always be relative. * Fix issues in repo due to C++ expression evaluation ordering is undefined. | 03 February 2023, 22:11:12 UTC |
1890836 | Ellie Hermaszewska | 03 February 2023, 04:19:10 UTC | Correct indentation in premake lua (#2620) No semantic change | 03 February 2023, 04:19:10 UTC |
a00dc69 | Ellie Hermaszewska | 03 February 2023, 04:18:49 UTC | Use SPIR-V opcode names rather than numbers (#2571) * s/emititng blobal/emitting global * Use SPIR-V opcode names rather than numbers * regenerate Visual Studio project files * Use names for extended SPIR-V GLSL instructions * Add missing operand for SPIR-V extended instruction * Add warning aginst modifying generated hashing files * Squash warnings on MSVC | 03 February 2023, 04:18:49 UTC |
bbd1e17 | Yong He | 01 February 2023, 22:18:57 UTC | Support `out` parameters in backward differentiation. (#2619) * Support `out` parameters in backward differentiation. * Fixes. * Fix cleanup. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 01 February 2023, 22:18:57 UTC |
c5895fb | Ellie Hermaszewska | 01 February 2023, 15:32:40 UTC | Use gmake2 as a premake target over gmake (#2587) The gmake generator has been deprecated by gmake2 https://premake.github.io/docs/Using-Premake/#using-premake-to-generate-project-files gmake2 has better dependency handling around our custom rules leading in fewer runs of slang-generate etc... | 01 February 2023, 15:32:40 UTC |
e312d5c | Sai Praveen Bangaru | 31 January 2023, 08:26:59 UTC | Patched support for multi-return and fallthrough if-else with break stmts (#2617) | 31 January 2023, 08:26:59 UTC |
77cdbb2 | Yong He | 31 January 2023, 04:03:46 UTC | Add transposition logic for constructor opcodes. (#2618) * Add transposition logic for constructor opcodes. * Fix. * Add language server regression test. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 31 January 2023, 04:03:46 UTC |
499b025 | Yong He | 31 January 2023, 03:24:09 UTC | Make ArrayExpressionType a DeclRefType and define its autodiff extension in stdlib. (#2615) * Allow array parameters in forward diff. * Use type canonicalization instead of coersion. * Reimplement array type. * Fix. * Update test case. --------- Co-authored-by: Yong He <yhe@nvidia.com> | 31 January 2023, 03:24:09 UTC |