Skip to content

GitLab

Commit 4b4512fa authored by zhuwenwen's avatar zhuwenwen
Browse files

add gfx938

parent d4d251fe
Show whitespace changes
Inline Side-by-side
Showing with 2 additions and 2 deletions
CMakeLists.txt
View file @ 4b4512fa
...@@ -38,7 +38,7 @@ install(CODE "set(CMAKE_INSTALL_LOCAL_ONLY TRUE)" ALL_COMPONENTS) ...@@ -38,7 +38,7 @@ install(CODE "set(CMAKE_INSTALL_LOCAL_ONLY TRUE)" ALL_COMPONENTS)
set(PYTHON_SUPPORTED_VERSIONS "3.9" "3.10" "3.11" "3.12") set(PYTHON_SUPPORTED_VERSIONS "3.9" "3.10" "3.11" "3.12")
# Supported AMD GPU architectures. # Supported AMD GPU architectures.
set(HIP_SUPPORTED_ARCHS "gfx908;gfx90a;gfx942;gfx950;gfx1030;gfx1100;gfx1101;gfx1200;gfx1201;gfx906;gfx926;gfx928;gfx936") set(HIP_SUPPORTED_ARCHS "gfx908;gfx90a;gfx942;gfx950;gfx1030;gfx1100;gfx1101;gfx1200;gfx1201;gfx906;gfx926;gfx928;gfx936;gfx938")
# #
# Supported/expected torch versions for CUDA/ROCm. # Supported/expected torch versions for CUDA/ROCm.
......
vllm/utils/__init__.py
View file @ 4b4512fa
...@@ -86,7 +86,7 @@ POOLING_MODEL_MAX_NUM_BATCHED_TOKENS = 32768 ...@@ -86,7 +86,7 @@ POOLING_MODEL_MAX_NUM_BATCHED_TOKENS = 32768
MULTIMODAL_MODEL_MAX_NUM_BATCHED_TOKENS = 5120 MULTIMODAL_MODEL_MAX_NUM_BATCHED_TOKENS = 5120
GPU_ARCH = torch.cuda.get_device_properties("cuda").gcnArchName GPU_ARCH = torch.cuda.get_device_properties("cuda").gcnArchName
SUPPORT_TC = any(arch in GPU_ARCH for arch in ["gfx928", "gfx936"]) SUPPORT_TC = any(arch in GPU_ARCH for arch in ["gfx928", "gfx936" "gfx938"])
def _generate_random_int8( def _generate_random_int8(
tensor: torch.Tensor, tensor: torch.Tensor,
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment