aboutsummaryrefslogtreecommitdiffhomepage
path: root/ptx
AgeCommit message (Collapse)Author
2024-12-10Enable Geekbench 5 (#304)Andrzej Janik
2024-12-02Fix host code and update to CUDA 12.4 (#299)Andrzej Janik
2024-11-02Fix linking of AMD device libraries (#296)Andrzej Janik
It's weird that it fails without `-mno-link-builtin-bitcode-postopt`. I've tested it only on ROCm 6.2, might be broken on older or newer ROCm
2024-10-16Re-enable all failing PTX tests (#277)Andrzej Janik
Additionally remove unused compilation paths
2024-09-23Refactor compilation passes (#270)Andrzej Janik
The overarching goal is to refactor all passes so they are module-scoped and not function-scoped. Additionally, make improvements to the most egregiously buggy/unfit passes (so the code is ready for the next major features: linking, ftz handling) and continue adding more code to the LLVM backend
2024-09-13Connect new parser to LLVM bitcode backend (#269)Andrzej Janik
This is very incomplete. Just enough code to emit LLVM bitcode and continue further development
2024-09-04PTX parser rewrite (#267)Andrzej Janik
Replaces traditional LALRPOP-based parser with winnow-based parser to handle out-of-order instruction modifer. Generate instruction type and instruction visitor from a macro instead of writing by hand. Add separate compilation path using the new parser that only works in tests for now
2021-12-14Improve reporting of recovered unrecognized statement/directiveAndrzej Janik
2021-12-13Better reporting of unrecognized tokensAndrzej Janik
2021-12-13Make all user errors recoverableAndrzej Janik
2021-09-29Fix shared munging pass and add fix cuModuleLoadDataAndrzej Janik
2021-09-29Redo shared memory transformationAndrzej Janik
2021-09-27Add CUDA tests showing problems with .shared unificationAndrzej Janik
2021-09-26Finish fixing shared memory passAndrzej Janik
2021-09-24Start implementing .shared unificationAndrzej Janik
2021-09-19Fix minor bugsAndrzej Janik
2021-09-18Fix linkageAndrzej Janik
2021-09-18Have an implementation for vprintfAndrzej Janik
2021-09-18Minor fixesAndrzej Janik
2021-09-17Handle ld.volatile/st.volatileAndrzej Janik
2021-09-17Redo handling of sregsAndrzej Janik
2021-09-17Add early support for more sregsAndrzej Janik
2021-09-17Implement function pointers and activemaskAndrzej Janik
2021-09-16Add missing vray instructionsAndrzej Janik
2021-09-14Fix some unhandled cases in cvt instructionAndrzej Janik
2021-09-14Parse and test const buffersAndrzej Janik
2021-09-13Zero out buffer on creationAndrzej Janik
2021-09-13Continue HIP conversionAndrzej Janik
2021-09-09Update tests, disable OpenCL-style shared mem conversion, emit linking ↵Andrzej Janik
information
2021-09-08Merge commit 'e2fbdf7d7b78ee360fe5fcc13ed1a7cca727c921' into amdAndrzej Janik
2021-09-08Emit Execution mode for denorm flushAndrzej Janik
2021-09-07Make ptx unit tests run on AMD (except denormals)Andrzej Janik
2021-09-06Convert unit tests to HIPAndrzej Janik
2021-08-06Remove all use of L0Andrzej Janik
2021-08-03Hack enough functionality that AMD GPU code buildsAndrzej Janik
2021-08-02Use calls to OpenCL builtins when translating sregs, do SPIRV->LLVM ↵Andrzej Janik
conversion on every build
2021-08-01Change codegen for mul.wideAndrzej Janik
2021-07-25Tune generated code, add a workaround for geekbenchAndrzej Janik
2021-07-22Finish converting to OpenCLAndrzej Janik
2021-07-06Synchronize through barrierAndrzej Janik
2021-07-03Regenerate SPIR-V for ptx_impl and fix weird handling of ptr-ptr add or subAndrzej Janik
2021-07-02Be more correct when emitting brev, refactor inst->func call passAndrzej Janik
2021-06-28Bunch of tiny fixes and improvementsAndrzej Janik
2021-06-27Fix bugs related to replay on LinuxAndrzej Janik
2021-06-25Allow ptr offsets to non-scalar typesAndrzej Janik
2021-06-25Merge branch 'one_type_type2'Andrzej Janik
2021-06-25Clean up warningsAndrzej Janik
2021-06-25Update testsAndrzej Janik
2021-06-20Prepare level zero and our compiler for global addressingAndrzej Janik
2021-06-12Fix problems with non-dereferencing inline additionAndrzej Janik
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-26 01:29:13 +0000