Skip to content
GitLab
Explore
Sign in
Register
Primary navigation
Search or go to…
Project
M
Microbenchmarks
Manage
Activity
Members
Labels
Plan
Issues
Issue boards
Milestones
Wiki
Requirements
Jira
Code
Merge requests
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Snippets
Locked files
Build
Pipelines
Jobs
Pipeline schedules
Test cases
Artifacts
Deploy
Releases
Package registry
Container registry
Model registry
Operate
Environments
Terraform modules
Monitor
Incidents
Analyze
Value stream analytics
Contributor analytics
CI/CD analytics
Repository analytics
Code review analytics
Issue analytics
Insights
Model experiments
Help
Help
Support
GitLab documentation
Compare GitLab plans
GitLab community forum
Contribute to GitLab
Provide feedback
Keyboard shortcuts
?
Snippets
Groups
Projects
Show more breadcrumbs
Mattia Mancini
Microbenchmarks
Commits
ac90b5bf
Commit
ac90b5bf
authored
May 14, 2024
by
Bram Veenboer
Committed by
Bram Veenboer
May 14, 2024
Browse files
Options
Downloads
Patches
Plain Diff
Remove helper variables
parent
78b92033
Branches
Branches containing commit
No related tags found
1 merge request
!4
Add matrixMultiplyAVX2b
Changes
1
Show whitespace changes
Inline
Side-by-side
Showing
1 changed file
code/matrix_multiplication.h
+4
-9
4 additions, 9 deletions
code/matrix_multiplication.h
with
4 additions
and
9 deletions
code/matrix_multiplication.h
+
4
−
9
View file @
ac90b5bf
...
...
@@ -125,15 +125,10 @@ void matrixMultiplyAVX2b(const std::complex<float>* a,
__m256
a_3
=
_mm256_permute_ps
(
a_m
,
_MM_SHUFFLE
(
1
,
1
,
1
,
1
));
__m256
a_4
=
_mm256_permute_ps
(
a_m
,
_MM_SHUFFLE
(
3
,
3
,
3
,
3
));
__m256i
b_1_ind
=
_mm256_set_epi32
(
3
,
2
,
1
,
0
,
3
,
2
,
1
,
0
);
__m256i
b_2_ind
=
_mm256_set_epi32
(
7
,
6
,
5
,
4
,
7
,
6
,
5
,
4
);
__m256i
b_3_ind
=
_mm256_set_epi32
(
2
,
3
,
0
,
1
,
2
,
3
,
0
,
1
);
__m256i
b_4_ind
=
_mm256_set_epi32
(
6
,
7
,
4
,
5
,
6
,
7
,
4
,
5
);
__m256
b_1
=
_mm256_permutevar8x32_ps
(
b_m
,
b_1_ind
);
__m256
b_2
=
_mm256_permutevar8x32_ps
(
b_m
,
b_2_ind
);
__m256
b_3
=
_mm256_permutevar8x32_ps
(
b_m
,
b_3_ind
);
__m256
b_4
=
_mm256_permutevar8x32_ps
(
b_m
,
b_4_ind
);
__m256
b_1
=
_mm256_permutevar8x32_ps
(
b_m
,
_mm256_set_epi32
(
3
,
2
,
1
,
0
,
3
,
2
,
1
,
0
));
__m256
b_2
=
_mm256_permutevar8x32_ps
(
b_m
,
_mm256_set_epi32
(
7
,
6
,
5
,
4
,
7
,
6
,
5
,
4
));
__m256
b_3
=
_mm256_permutevar8x32_ps
(
b_m
,
_mm256_set_epi32
(
2
,
3
,
0
,
1
,
2
,
3
,
0
,
1
));
__m256
b_4
=
_mm256_permutevar8x32_ps
(
b_m
,
_mm256_set_epi32
(
6
,
7
,
4
,
5
,
6
,
7
,
4
,
5
));
__m256
c_m
=
_mm256_mul_ps
(
a_1
,
b_1
);
c_m
=
_mm256_fmadd_ps
(
a_2
,
b_2
,
c_m
);
...
...
This diff is collapsed.
Click to expand it.
Preview
0%
Loading
Try again
or
attach a new file
.
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Save comment
Cancel
Please
register
or
sign in
to comment