8000 Update on "[AOTI] Embed cubin files into .so" · pytorch/pytorch@09b3516 · GitHub
[go: up one dir, main page]
Skip to content

Commit 09b3516

Browse files
desertfiredesertfire
committed
Update on "[AOTI] Embed cubin files into .so"
Summary: Embed cubin files so AOTI is one step closer to generate a single binary. Controlled by a flag and off as default. Differential Revision: [D72535357](https://our.internmc.facebook.com/intern/diff/D72535357) cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov [ghstack-poisoned]
2 parents 4792a4b + 6322889 commit 09b3516

File tree

1 file changed

+3
-0
lines changed

1 file changed

+3
-0
lines changed

‎torch/_inductor/codegen/cpp_wrapper_gpu.py

Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -58,6 +58,9 @@ class DeferredTritonCallWrapper:
5858
arg_types: list[Any]
5959

6060
def generate(self, wrapper: CppWrapperGpu):
61+
"""
62+
Generate the GPU kernel definition, as well as load and launch code.
63+
"""
6164
prefix = wrapper.prefix
6265
if self.kernel_name.startswith("multi_kernel_"):
6366
# MultiKernel will select one kernel after running the autotune block

0 commit comments

Comments
 (0)
0