-
Notifications
You must be signed in to change notification settings - Fork 30
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
Study type packing (VNNI/BFDOT/BFMMLA/etc) as a single operation #563
Labels
Comments
chelini
added a commit
to chelini/tpp-sandbox
that referenced
this issue
Nov 27, 2023
Until we have a better way to express the VNNI layout (see: libxsmm#563), it is up to the callee to specify the expected rank in the VNNI layout as the rank depends on the operations we are dealing with.
chelini
added a commit
to chelini/tpp-sandbox
that referenced
this issue
Nov 27, 2023
Until we have a better way to express the VNNI layout (see: libxsmm#563), it is up to the callee to specify the expected rank in the VNNI layout as the rank depends on the operations we are dealing with. The current check for rank < 3 is not correct; this is still ugly, but at least more explicit, as there are not magic assumptions on rank.
chelini
added a commit
to chelini/tpp-sandbox
that referenced
this issue
Nov 27, 2023
Until we have a better way to express the VNNI layout (see: libxsmm#563), it is up to the callee to specify the expected rank in the VNNI layout as the rank depends on the operations we are dealing with. The current check for rank < 3 is not correct; this is still ugly, but at least more explicit, as there are not magic assumptions on rank.
chelini
added a commit
to chelini/tpp-sandbox
that referenced
this issue
Nov 27, 2023
Until we have a better way to express the VNNI layout (see: libxsmm#563), it is up to the callee to specify the expected rank in the VNNI layout as the rank depends on the operations we are dealing with. The current check for rank < 3 is not correct; this is still ugly, but at least more explicit, as there are not magic assumptions on rank.
chelini
added a commit
to chelini/tpp-sandbox
that referenced
this issue
Nov 27, 2023
Until we have a better way to express the VNNI layout (see: libxsmm#563), it is up to the callee to specify the expected rank in the VNNI layout as the rank depends on the operations we are dealing with. The current check for rank < 3 is not correct; this is still ugly, but at least more explicit, as there are not magic assumptions on rank.
chelini
added a commit
that referenced
this issue
Nov 27, 2023
Until we have a better way to express the VNNI layout (see: #563), it is up to the caller to specify the expected rank in the VNNI layout as the rank depends on the operations we are dealing with. The current check for rank < 3 is not correct; this is still ugly, but at least more explicit, as there are not magic assumptions on rank.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Labels
Today we're working on a type packing for VNNI with the operation
tpp.vnni_pack
. But this isn't the only type of packing we may want, and they're all very similar, so we could probably come up with a single op (saytpp.type_pack
) with generic arguments that converts a standard packed tensor into a type packed tensor, given some shapes.We then map the existing types (VNNI/BFDOT/BFMMLA/etc) into this op and lower them as library calls when they match a recognised shape.
We shouldn't create this op if we don't know that the type packing makes sense (ex. by #561 calls), so we shouldn't need to lower this op to loop unless requested by the pipeline (
tpp-to-loops
).@KavithaTipturMadhu
References:
The text was updated successfully, but these errors were encountered: