-
Notifications
You must be signed in to change notification settings - Fork 71
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
Halevi-Shoup matvec kernel #1552
base: main
Are you sure you want to change the base?
Conversation
97d03da
to
6cce260
Compare
Add gitattributes for bazel-provided merge conflict resolution in the lockfile. Added a patch to re2 (transitive dep via or-tools) due to google/re2#525 Extracted from google#1552
14643ef
to
4a5006b
Compare
This commit also introduces tensor_ext.alignment which tracks data replication during packing.
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.
Before I step into the two pass changes... (the attribute changes lgtm)
values to pad at the end of that axis. The `paddingValue` attribute, | ||
defaulting to zero, describes the value used for padding. |
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.
Update here if you go the route of a possible unspecified padding - The paddingValue attr describes the value used for padding. If left unspecified, then the program will allow any value to be present in the padded slots.
DenseSet<long> insertedDimsSet(insertedDims.asArrayRef().begin(), | ||
insertedDims.asArrayRef().end()); | ||
if (insertedDimsSet.size() != insertedDims.size()) { | ||
return emitError() << "insertedDims must all be unique"; | ||
} |
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.
for the sake of verification, these probably should be within bounds as well? The code below that sets beforeReplication doesn't mind if they're out of range.
Rebased over #1428