@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / cd9d06fb8d
Fork 0

A high-throughput and memory-efficient inference and serving engine for LLMs

Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git

Synced Nov 16, 02:00 UTC

Allow hip sources to be directly included when compiling for rocm. (#12087)

tvirolai-amd committed Jan 15, 2025 at 23:46 UTC cd9d06fb8d1f89fc1bcc9305bc20d57c6d8b73d8
1 file changed +2 -2
cmake/utils.cmake
+2 -2
@@ -58,8 +58,8 @@ function (hipify_sources_target OUT_SRCS NAME ORIG_SRCS)
58 #
59 set(SRCS ${ORIG_SRCS})
60 set(CXX_SRCS ${ORIG_SRCS})
61 - list(FILTER SRCS EXCLUDE REGEX "\.(cc)|(cpp)$")
62 - list(FILTER CXX_SRCS INCLUDE REGEX "\.(cc)|(cpp)$")
61 + list(FILTER SRCS EXCLUDE REGEX "\.(cc)|(cpp)|(hip)$")
62 + list(FILTER CXX_SRCS INCLUDE REGEX "\.(cc)|(cpp)|(hip)$")
63
64 #
65 # Generate ROCm/HIP source file names from CUDA file names.