Skip to content

Conversation

@pjbgf
Copy link
Owner

@pjbgf pjbgf commented Jul 11, 2025

AVO is a great way to build assembly code and worked quite nicely for the amd64 implementation of the sha1cdblock. The lack of support for arm64 meant that little could be reused for arm64, which then decreases the benefit of having it in the first place.

For UBC, the benefits of AVO are less clear as the Go code is largely a duplication of the assembler as it is just a very long number of statements.

This decision will be reviewed once AVO supports arm64.

pjbgf added 3 commits July 11, 2025 06:25
The assembler arm64 implementation is closely aligned with the Go
AVO version for amd64. However, given that there isn't support for
amd64 in AVO, it feels easier to drop the AVO dependency and have
a simpler assembly code for the time being.

In the future, when AVO adds support for arm64 this may be reviewed.

Signed-off-by: Paulo Gomes <[email protected]>
AVO is a great way to build assembly code and worked quite nicely
for the amd64 implementation of the sha1cdblock. The lack of support
for arm64 meant that little could be reused for arm64, which then
decreases the benefit of having it in the first place.

For UBC, the benefits of AVO are less clear as the Go code is largely
a duplication of the assembler as it is just a very long number of
statements.

This decision will be reviewed once AVO supports arm64.

Signed-off-by: Paulo Gomes <[email protected]>
@pjbgf pjbgf merged commit 351f559 into main Jul 11, 2025
14 checks passed
@pjbgf pjbgf deleted the refactor branch July 11, 2025 05:29
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.

1 participant