From e5f3caf145b015d9a1acc155a8b7994fe98ec33e Mon Sep 17 00:00:00 2001 From: Lei Mao Date: Wed, 23 Oct 2024 09:52:43 -0700 Subject: [PATCH] Fix README (#1658) * Fix README * Improve README --------- Co-authored-by: Haicheng Wu <57973641+hwu36@users.noreply.github.com> --- README.md | 1 - media/docs/quickstart.md | 4 +++- 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index e61335f2..efe47872 100644 --- a/README.md +++ b/README.md @@ -41,7 +41,6 @@ and improves code composability and readability. More documentation specific to In addition to GEMMs, CUTLASS implements high-performance convolution via the implicit GEMM algorithm. Implicit GEMM is the formulation of a convolution operation as a GEMM thereby taking advantage of CUTLASS's modular GEMM pipeline. This allows CUTLASS to build convolutions by reusing highly-optimized GEMM components. - # What's New in CUTLASS 3.6 CUTLASS 3.6.0 is an update to CUTLASS adding: diff --git a/media/docs/quickstart.md b/media/docs/quickstart.md index 7faad445..97ed6a63 100644 --- a/media/docs/quickstart.md +++ b/media/docs/quickstart.md @@ -179,11 +179,13 @@ $ make test_unit_gemm_warp -j To minimize compilation time, specific GPU architectures can be enabled via the CMake command, selected by [CUDA Compute Capability.](https://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#compute-capabilities) -**NVIDIA Ampere Architecture.** +**NVIDIA Hopper Architecture.** ```bash $ cmake .. -DCUTLASS_NVCC_ARCHS=90a # compiles for NVIDIA Hopper GPU architecture ``` +**NVIDIA Ampere Architecture.** + ```bash $ cmake .. -DCUTLASS_NVCC_ARCHS=80 # compiles for NVIDIA Ampere GPU architecture ```