llvm.org GIT mirror llvm / 03f06f0
[CUDA] Implemented _[bi]mma* builtins. These builtins provide access to the new integer and sub-integer variants of MMA (matrix multiply-accumulate) instructions provided by CUDA-10.x on sm_75 (AKA Turing) GPUs. Also added a feature for PTX 6.4. While Clang/LLVM does not generate any PTX instructions that need it, we still need to pass it through to ptxas in order to be able to compile code that uses the new 'mma' instruction as inline assembly (e.g used by NVIDIA's CUTLASS library https://github.com/NVIDIA/cutlass/blob/master/cutlass/arch/mma.h#L101) Differential Revision: https://reviews.llvm.org/D60279 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@359248 91177308-0d34-0410-b5e6-96231b3b80d8 Artem Belevich 1 year, 7 months ago
1 changed file(s) with 2 addition(s) and 0 deletion(s). Raw diff Collapse all Expand all
7474 "Use PTX version 6.1">;
7575 def PTX63 : SubtargetFeature<"ptx63", "PTXVersion", "63",
7676 "Use PTX version 6.3">;
77 def PTX64 : SubtargetFeature<"ptx64", "PTXVersion", "64",
78 "Use PTX version 6.4">;
7779
7880 //===----------------------------------------------------------------------===//
7981 // NVPTX supported processors.