-
Notifications
You must be signed in to change notification settings - Fork 232
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
Attempt to switch everything to cmake #1659
Conversation
stack-info: PR: #1659, branch: drisspg/stack/33
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/1659
Note: Links to docs will display an error until the docs builds have been completed. ❌ 23 New Failures, 2 Unrelated FailuresAs of commit 1564578 with merge base 6ffe236 ( NEW FAILURES - The following jobs have failed:
FLAKY - The following job failed but was likely due to flakiness present on trunk:
BROKEN TRUNK - The following job failed but was present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
27274cc
to
1564578
Compare
|
||
# CUDA Setup | ||
if(CMAKE_CUDA_COMPILER) | ||
enable_language(CUDA) |
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.
isn't this redundant with setting CUDA in project?
|
||
# Set CUDA architectures if not already set | ||
if(NOT DEFINED CMAKE_CUDA_ARCHITECTURES) | ||
set(CMAKE_CUDA_ARCHITECTURES 70 75 80 86) |
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.
no ADA 89 and Hopper 90?
# Set CUDA flags | ||
# Set CUDA architectures and TORCH_CUDA_ARCH_LIST | ||
if(NOT DEFINED TORCH_CUDA_ARCH_LIST) | ||
set(TORCH_CUDA_ARCH_LIST "7.0;7.5;8.0;8.6;9.0") |
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.
no ADA?
if(CMAKE_BUILD_TYPE STREQUAL "Debug") | ||
add_compile_options(-g -O0) | ||
if(CMAKE_CUDA_COMPILER) | ||
add_compile_options($<$<COMPILE_LANGUAGE:CUDA>:-g>) |
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.
-lineinfo?
include_directories(${CMAKE_CURRENT_SOURCE_DIR}) | ||
|
||
# CUDA Setup | ||
if(CMAKE_CUDA_COMPILER) |
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.
what happens if python sets -DUSE_CUDA=OFF
, but cmake finds cuda, or the other way around?
) | ||
|
||
# Set Python limited API version | ||
target_compile_definitions(torchao_core PRIVATE Py_LIMITED_API=0x03090000) |
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.
Py_LIMITED_API
is already set for all targets somewhere above
cc @drisspg are you still planning on working on this? |
nope |
Stacked PRs:
Attempt to switch everything to cmake