[Snippets][CPU] Moved N_tail processing to the end in BrgemmCopyBKernel #28664
+18
−33
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
a-sidorova
force-pushed
the
feature/snippets/optimized_brgemm_copy_b_kernel_tail
branch
from
9a47d46 to
f5291d8
Compare
v-Golubev
approved these changes
Jan 28, 2025
...c/transformations/snippets/x64/pass/lowered/expressions/brgemm_copy_b_buffer_expressions.cpp
Outdated
Show resolved
Hide resolved
Comment on lines
79
to
80
| inline T compute_LDB(T n_block, const ov::element::Type& precision) { | ||
| return compute_repacked_n_dim(n_block, precision); |
There was a problem hiding this comment.
Why do we need 2 different functions that do the same thing?
Should we replace all compute_LDB calls with compute_repacked_n_dim then?
There was a problem hiding this comment.
I left the both them just to split logic into get_LDB and get_repacked_N by sense.
But the implementation is the same.
Anyway, I replaced compite_LDB with compute_repacked_n_dim in 15dadd5
| const auto& precision = parent_expr->get_node()->get_input_element_type(0); | ||
| m_allocation_size = std::max(n_blk, compute_inner_n_block(precision)); | ||
| } | ||
| m_allocation_size = compute_repacked_n_dim(n_blk, precision); |
There was a problem hiding this comment.
It's nice that we now reuse the dynamic values handling from ov::snippets::utils::rnd_up, and don't have to replicate this logic anywhere else.
a-sidorova
force-pushed
the
feature/snippets/optimized_brgemm_copy_b_kernel_tail
branch
from
c64f34f to
15dadd5
Compare
a-sidorova
force-pushed
the
feature/snippets/optimized_brgemm_copy_b_kernel_tail
branch
from
15dadd5 to
6ea5aff
Compare
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.
Details:
N_Tailprocessing should be at the end ofBrgemmCopyBKernel. The current PR moves tail processing from the beginning to the end in kernelTickets: