-
Notifications
You must be signed in to change notification settings - Fork 536
Link xnn_executor_runner with optimized op library #8901
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
Stack from ghstack (oldest at bottom): |
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/8901
Note: Links to docs will display an error until the docs builds have been completed. ⏳ No Failures, 2 PendingAs of commit 18c64ad with merge base 09ad20a ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
Doesn't seem to be any reason not to allow optimized ops for this one. ghstack-source-id: aaf66178b1902763ebaaaa2f0b3a312505722ea8 ghstack-comment-id: 2695738759 Pull Request resolved: #8901
Thanks @swolchok! Ideally in OSS world we also have a executor_runner with optimized kernels, and use it as a test runner for all models. Maybe we can promote xnn_executor_runner to be this general runner. BTW this PR has diff from other commits? |
I'm having trouble exporting my stack; getting GitHub rate limits. Hopefully folks will review the bottom PR or two soon. |
Doesn't seem to be any reason not to allow optimized ops for this one. ghstack-source-id: 85e31ee40499f7858fb8ff4c3e42ec453ff88948 ghstack-comment-id: 2695738759 Pull Request resolved: #8901
fixed, please review |
7cbe7a1
to
7bc4529
Compare
7bc4529
to
764977b
Compare
backends/xnnpack/CMakeLists.txt
Outdated
@@ -139,7 +139,7 @@ if(NOT CMAKE_TOOLCHAIN_FILE MATCHES ".*(iOS|ios\.toolchain)\.cmake$") | |||
endif() | |||
|
|||
target_link_libraries( | |||
xnn_executor_runner gflags portable_ops_lib ${xnn_executor_runner_libs} | |||
xnn_executor_runner gflags optimized_native_cpu_ops_lib ${xnn_executor_runner_libs} |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you! So many perf cliffs because of this.
@@ -119,6 +118,12 @@ target_include_directories( | |||
target_compile_options(xnnpack_backend PUBLIC ${_common_compile_options}) | |||
target_link_options_shared_lib(xnnpack_backend) | |||
|
|||
if(EXECUTORCH_BUILD_KERNELS_OPTIMIZED) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
fixed build when XNNPACK is on but optimized lib is off. everything else (except corresponding change to line 149 below) is just cmake-format.
Doesn't seem to be any reason not to allow optimized ops for this one.
Doesn't seem to be any reason not to allow optimized ops for this one.