Making it easier to work with shaders https://git.yummers.dev/slang.git/atom?h=master 2024-11-05T17:47:26+00:00 2024-11-05T17:47:26+00:00 Ellie Hermaszewska ellieh@nvidia.com 2024-11-05T17:47:26+00:00 urn:sha1:b118451e301d734e3e783b3acdf871f3f6ea851c * Move switch statement bodies to their own lines * format --------- Co-authored-by: Yong He <yonghe@outlook.com> 2024-10-29T06:49:26+00:00 Ellie Hermaszewska ellieh@nvidia.com 2024-10-29T06:49:26+00:00 urn:sha1:f65d756bff8d4c5cbc15bd0322a2ae8e6b896a21 * format * Minor test fixes * enable checking cpp format in ci 2019-09-18T20:20:22+00:00 Tim Foley tfoleyNV@users.noreply.github.com 2019-09-18T20:20:22+00:00 urn:sha1:a431d4f6da9e463c349c2e3819e87a83c8f2e043 Work on #1059 The `%` operator in the Slang implementation had several issues, and this change tries to address some of them: * Renamed most occurences of "mod" describing this operator to be "rem" for "remainder" to better match its semantics in HLSL * Split the operator into distinct integer and floating-point variants (`IRem` and `FRem`) to simplify having different codegen for the two * Added floating-point variants of `operator%` and `operator%=` to the stdlib. * Added custom C++ codegen for `kIROp_FRem` such that it maps to the standard C/C++ `remainder()` function * Added custom GLSL codegen so that `kIROp_FRem` maps to the GLSL `mod()` function (which isn't correct...) * Added a test case to confirm that D3D11, D3D12, and CPU targets all agree on the definition of floating-point `%` * Fixed `render-test-tool` to allow a negative integer in a `data=...` specification. This didn't end up being used in the final test, but still seems like a good fix. * Added a customized baseline for the Vulkan flavor of that test to confirm that we are *not* compiling correctly to SPIR-V just yet Addressing the correctness of the output for GLSL/SPIR-V will have to come as a later change given that the operation we want is not exposed directly by unextended GLSL. 2019-06-06T21:48:39+00:00 jsmall-nvidia jsmall@nvidia.com 2019-06-06T21:48:39+00:00 urn:sha1:fc083a75b94ac4b4e735b4a7ff566191b9123f74 * * Added SourceStyle to CLikeSourceEmitter, to limit cases to actual target types. * Made Impl methods _ prefixed * Small tidyup * * SourceStream -> SourceWriter * use slang-emit- prefix on SourceWriter file * * Remove EmitContext -> merge into CLikeSourceEmitter * slang-c-like-source-emitter -> slang-emit-source.cpp * ExtensionUsageTracker -> GLSLExtensionTracker slang-extension-usage-tracker.cpp/.h -> slang-emit-glsl-extension-tracker.cpp/.h * emit-source.cpp.h -> emit-c-like.cpp/.h * Small fix to move where some _ prefixed functions are declared in CLikeSourceEmitter. * * CLikeSourceEmitter::CInfo -> Desc * Functions to get and find CodeGenTarget by name * Split out empty language impls * Create an impl based on SourceStyle * * CodeGenTarget conversion to and from string * Move HLSL specific functions to HLSLEmitSource. * Emitting texture and image types. * Move move GLSL specific functionality to GLSLSourceEmitter * Split more out of slang-emit-c-like * Refactor more out of slang-emit-c-like * * tryEmitIRInstExprImpl(IRInst* inst, IREmitMode mode, const EmitOpInfo& inOuterPrec) * Fix bug around output of uintBitsToFloat * More work refactoring out target specifics from slang-emit-c-like * Move functions that are only implemented once in GLSL impl into their Impl method. * Move rate qualification out of slang-emit-c-like * * Added getEmitOpForOp - allows for table usage so different ops can be dealt with the same way * Moved vector comparison to slang-emit-glsl * * * Use EmitOpInfo to control output in slang-emit-c-like.cpp for unary ops * Move more functionality from CLikeSourceEmitter to HLSLSourceEmitter * Make output of parameters implementaion specific. * Extracted interpolation modifiers. * Remove IR from methods that don't need them. * Remove IR from method names. * Refactor handling of output of types - to make the impls implement the full path without lots of cases for specific impls * Add variable declaration modifiers and matrix layout to larget specific in slang-emit. * Make target specific internal functions _ prefixed. 2019-05-31T17:17:34+00:00 jsmall-nvidia jsmall@nvidia.com 2019-05-31T17:17:34+00:00 urn:sha1:b81ff3ef968d1cc4e954b31a1812b3c391d17b02 * WIP: Setting up C/Cpp source compilation targets. * WIP: Emitting C/CPP. * WIP: Split out SourceSink, and use it for source output on emit. * SourceSink -> SourceStream * * Made SourceStream use m_ prefixing of members. * Make all methods use lower camel * Removed methods from SourceStream interface that are not used externally (use _ prefixing) * Improvements to documentation * EmitContext is now effectively empty, so just use SharedEmitContext as EmitContext. * SharedEmitContext -> EmitContext * Methods to LowerCamel in emit.cpp * Split out EmitContext and ExtensionUsageTracker into separate files. * Split out EmitVisitor into slang-c-like-source-emitter files. * EmitVisitor -> CLikeSourceEmitter * Tidy up around CLikeSourceEmitter - simplify header. * Small tidy up - removing repeated comments that are in header. * Remove EmitContext paramter threading. * Small tidy up. Use prefixed macros for slang-c-like-source-emitter.h * Small tidy up in slang-c-like-source-emitter.cpp * First pass at splitting out UnmangleContext. * MangledNameParser -> MangledLexer. * WIP making EmitOp (EOp) enum available outside of cpp * Generating EmitOpInfo from macro. * Split out emit precedence handling. Don't use kOp_ style anymore, just use an array indexed by EmitOp. * Disable C simple test for now. * Keep g++/clang happy with token pasting. * Fix win32 narrowing warning.