[SME] Add scalable fp16->fp32 dense schedule #16981
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
11 tasks
This commit extends the functionality of the SME dense and matmul schedules to support operations with fp16 inputs and an fp32 output, where `transpose_a=False` and `transpose_b=True`. For convenience, it also adds a utility called `get_vscale_factor` which created the correct multiplier for `vscale` given a data type, reflecting ideas from an early design of the [SVE](apache/tvm-rfcs#104) RFC. Change-Id: I8c00bc6baf2df6015fa41200a238781126c73589
Change-Id: Ie7fb7a0a76119aa5c82e03ea0b2cc10de9f15f5e
lhutton1
force-pushed
the
sme-fp16-fp32-dense-schedule
branch
from
d2a164c to
1fe9bac
Compare
Change-Id: I0e9e45b285082b42676e53e74158e11d7e08608b
Change-Id: I32273241ae7569b65e082759e4f2ca4355ac6933
Change-Id: I237b4c5cb5ca22e33529d98cbd75177b94904857
lhutton1
force-pushed
the
sme-fp16-fp32-dense-schedule
branch
from
0d2be71 to
bc02e47
Compare
|
@tvm-bot rerun |
ekalda
approved these changes
May 28, 2024
|
Thanks @lhutton1 this is merged now! |
lhutton1
added a commit
to lhutton1/tvm
that referenced
this pull request
May 29, 2024
Fixes a merge conflict between apache#16981 and apache#17003. Change-Id: Ifcc983ef0b8c00250568a048fd682933adfdcde4
tqchen
pushed a commit
that referenced
this pull request
May 29, 2024
Anndrey24
added a commit
to Anndrey24/tvm
that referenced
this pull request
May 30, 2024
This commit extends the SME conv2d NHWC schedule to support convolutions with float16 inputs (data and kernel) and a float32 output using the tensor intrinsics added in apache#16981.
ekalda
pushed a commit
that referenced
this pull request
Jun 5, 2024
This commit extends the SME conv2d NHWC schedule to support convolutions with float16 inputs (data and kernel) and a float32 output using the tensor intrinsics added in #16981.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This commit extends the functionality of the SME dense and matmul schedules to support operations with fp16 inputs and an fp32 output, where
transpose_a=Falseandtranspose_b=True.For convenience, it also adds a utility called
get_vscale_factorwhich creates the correct multiplier forvscalegiven a data type, reflecting ideas from an early design of the SVE RFC.Note: this commit depends on #16921 so also contains the contents of #16921.