Skip to content
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.

Sign up for GitHub

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

Jump to bottom

qgemm: optimize avxvnni QGEMM inner kernel for M=1 #22952

Open
wants to merge 2 commits into
base: main
Choose a base branch
from
Open

qgemm: optimize avxvnni QGEMM inner kernel for M=1 #22952

wants to merge 2 commits into from
+558 −99

Commits on Nov 26, 2024

  1. Add QGEMM benchmarks for M = 1 on a single thread

    @r-devulap
    r-devulap committed Nov 26, 2024
    Configuration menu
    Copy the full SHA
    1b35bb0 View commit details
    Browse the repository at this point in the history

  2. qgemm: optimize avxvnni QGEMM inner kernel for M=1 

    QGEMM Benchmarks when M = 1 on an 13th Gen Intel(R) Core(TM) i9-13900K
shows a 1.4x improvement on a single thread.

|--------------------------------------------------------------------+--------+---------+----------+----------+---------+---------|
| Benchmark                                                          | Time   | CPU     | Time Old | Time New | CPU Old | CPU New |
|--------------------------------------------------------------------+--------+---------+----------+----------+---------+---------|
| QGEMM/UnsignedAPackB/M:1/N:512/K:512/Batch:1/Threads:1/real_time   | -0.275 | -0.2756 | 4330     | 3137     | 4330    | 3136    |
| QGEMM/UnsignedAPackB/M:1/N:512/K:1024/Batch:1/Threads:1/real_time  | -0.292 | -0.2927 | 9027     | 6385     | 9027    | 6385    |
| QGEMM/UnsignedAPackB/M:1/N:1024/K:1024/Batch:1/Threads:1/real_time | -0.300 | -0.3005 | 17867    | 12499    | 17866   | 12498   |
| OVERALL_GEOMEAN                                                    | -0.289 | -0.2897 |          |          |         |         |
|--------------------------------------------------------------------+--------+---------+----------+----------+---------+---------|
    @r-devulap
    r-devulap committed Nov 26, 2024
    Configuration menu
    Copy the full SHA
    42a8d2d View commit details
    Browse the repository at this point in the history