[libc-commits] [libc] 9a6517e - [libc][NFC] Do not emit init / fini kernels in NVPTX libc

[Joseph Huber via libc-commits]

Author: Joseph Huber
Date: 2023-11-13T09:24:02-06:00
New Revision: 9a6517e63a211080335f5226a53523c80d7168b8

URL: https://github.com/llvm/llvm-project/commit/9a6517e63a211080335f5226a53523c80d7168b8
DIFF: https://github.com/llvm/llvm-project/commit/9a6517e63a211080335f5226a53523c80d7168b8.diff

LOG: [libc][NFC] Do not emit init / fini kernels in NVPTX libc

Summray:
A recent patch upgrades the NVPTX ctor / dtor lowering pass to emit
kernels so other languages can call them. We do this manually in `libc`
so we do not need this. Use the provided flag to disable this step to
keep the created kernels cleaner.

Added: 
    

Modified: 
    libc/cmake/modules/LLVMLibCObjectRules.cmake

Removed: 
    


################################################################################
diff  --git a/libc/cmake/modules/LLVMLibCObjectRules.cmake b/libc/cmake/modules/LLVMLibCObjectRules.cmake
index cf5097e23a20453..d5df27a2dcb4341 100644
--- a/libc/cmake/modules/LLVMLibCObjectRules.cmake
+++ b/libc/cmake/modules/LLVMLibCObjectRules.cmake
@@ -89,6 +89,7 @@ function(get_nvptx_compile_options output_var gpu_arch)
   set(nvptx_options "")
   list(APPEND nvptx_options "-march=${gpu_arch}")
   list(APPEND nvptx_options "-Wno-unknown-cuda-version")
+  list(APPEND nvptx_options "SHELL:-mllvm -nvptx-emit-init-fini-kernel=false")
   if(${gpu_arch} STREQUAL "sm_35")
     list(APPEND nvptx_options "--cuda-feature=+ptx60")
   elseif(${gpu_arch} STREQUAL "sm_37")