Skip to content

igzip/riscv64: Optimize adler32_rvv for VLEN=128#390

Open
leiwen2025 wants to merge 2 commits intointel:masterfrom
leiwen2025:optimize_vlen128_new
Open

igzip/riscv64: Optimize adler32_rvv for VLEN=128#390
leiwen2025 wants to merge 2 commits intointel:masterfrom
leiwen2025:optimize_vlen128_new

Conversation

@leiwen2025
Copy link
Contributor

This PR optimizes the adler32_rvv implementation for vlen=128.

The optimization has been verified on the SG2044 platform:

SG2044:
        new: adler32_warm: runtime =    3062392 usecs, bandwidth 25988 MB in 3.0624 sec = 8486.24 MB/s
        old: adler32_warm: runtime =    3062471 usecs, bandwidth 23095 MB in 3.0625 sec = 7541.43 MB/s

Signed-off-by: WenLei <lei.wen2@zte.com.cn>
@pablodelara
Copy link
Contributor

@leiwen2025 could you update Release notes saying Adler32 has been optimized for RISCV?

Signed-off-by: WenLei <lei.wen2@zte.com.cn>
@leiwen2025
Copy link
Contributor Author

@leiwen2025 could you update Release notes saying Adler32 has been optimized for RISCV?

Done. I've updated the Release notes.

@pablodelara
Copy link
Contributor

@sunyuechi could you review this PR? Thanks!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants