-
Notifications
You must be signed in to change notification settings - Fork 564
Implement unary_ufunc functions using elementwise_util #9386
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
base: gh/swolchok/440/head
Are you sure you want to change the base?
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/9386
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New Failure, 1 Unrelated FailureAs of commit 7baa2f4 with merge base 71025df ( NEW FAILURE - The following job has failed:
FLAKY - The following job failed but was likely due to flakiness present on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
this will get shuffled later in the stack |
return out; | ||
} | ||
|
||
ET_SWITCH_REALHBBF16_TYPES(in.scalar_type(), ctx, op_name, CTYPE_IN, [&] { |
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.
this is a problem, right? the compute dtype here would be the input's scalar type,
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.
you need to switch over FLOATHBF16, right?
ScalarType compute_type = in.scalar_type() == ScalarType::Double | ||
? ScalarType::Double | ||
: ScalarType::Float; | ||
ET_SWITCH_FLOAT_TYPES(compute_type, ctx, op_name, CTYPE_COMPUTE, [&] { |
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.
this is the only ET_SWITCH change remaining -- we are switching over the possible compute types, not the input/output types.
[ghstack-poisoned]
One less set of independent implementations to worry about going forward (e.g., we don't have to vectorize these separately from elementwise_util and they get all benefits of elementwise_util).