src: cpu: aarch64: Re-enable matmul static quantisation through ACL. #2308
+327
−4
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
renato-arantes
force-pushed
the
static_quant2
branch
2 times, most recently
from
88cf5e0 to
5cec04d
Compare
|
Hi @mgouicem, it's the same as before but more restrictive with Neoverse-N1 until the failing tests are fixed. |
theComputeKid
approved these changes
Dec 23, 2024
github-actions
bot
added
the
platform:cpu-aarch64
vpirogov
reviewed
Jan 8, 2025
vpirogov
reviewed
Jan 8, 2025
dzarukin
reviewed
Jan 8, 2025
mgouicem
reviewed
Jan 9, 2025
|
Hi @renato-arantes I see that you added a merge commit, please rebase your branch instead as we try to keep a linear history. |
renato-arantes
force-pushed
the
static_quant2
branch
4 times, most recently
from
ce7f10b to
66d9ea4
Compare
|
@renato-arantes, please resolve conflicts. |
renato-arantes
force-pushed
the
static_quant2
branch
from
66d9ea4 to
b322d6d
Compare
renato-arantes
changed the title
renato-arantes
force-pushed
the
static_quant2
branch
from
b322d6d to
38d8255
Compare
mgouicem
approved these changes
Jan 15, 2025
vpirogov
reviewed
Jan 15, 2025
| arm_compute::QuantizationInfo(*src_scale, -src_zero_point, true)); | ||
| acl_obj.wei_tensor.info()->set_quantization_info( | ||
| arm_compute::QuantizationInfo(*wei_scale, -wei_zero_point, true)); | ||
| // for efficiency reasons, OneDNN saves the inverse of the destination |
There was a problem hiding this comment.
It's oneDNN, not OneDNN. Fixup is here.
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Outdated suggestions cannot be applied.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.
Description
This is to re-enable
matmulstatic quantization operations through ACL. Currently, the supported data type combinations ares8:s8:s8andu8:s8:u8. Neoverse-N1 is skipped until failing tests are fixed.Previous PR: #2198
General
make testandmake test_benchdnn_*) pass locally for each commit?