Skip to content

Fix alignment in acquire-release spec test - #9014

Open
stevenfontanella wants to merge 3 commits into
mainfrom
align
Open

Fix alignment in acquire-release spec test#9014
stevenfontanella wants to merge 3 commits into
mainfrom
align

Conversation

@stevenfontanella

@stevenfontanella stevenfontanella commented Aug 18, 2026

Copy link
Copy Markdown
Member

Instructions like i32.atomic.load8_u have a different alignment than the value of their argument. Add an extra align field to account for this.

The value_type was also wrong for a few load operations, but this didn't matter because loads don't use their value_type; it's only used to choose what param to pass as the value for stores.

cc @rmahdav

@stevenfontanella
stevenfontanella marked this pull request as ready for review August 18, 2026 21:06
@stevenfontanella
stevenfontanella requested a review from a team as a code owner August 18, 2026 21:06
@stevenfontanella
stevenfontanella requested review from tlively and removed request for a team August 18, 2026 21:06
The entire iterator represents a complete expression using the `template`. e.g.
(drop (i32.atomic.load (i32.const 42)))
"""
if template.op == "atomic.fence":

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can assert that the align is None here, too.

@stevenfontanella stevenfontanella Aug 18, 2026

Copy link
Copy Markdown
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I would lean towards not adding it since we’re not making use of the assumption here. It seems redundant because the Template is anyway the source of truth for the align, value_type, etc. e.g. we could also assert that value_type is None here, but we’d just be repeating what’s in the Template. I think the memory assertions make more sense here because they’re different arguments and it’s possible for the caller to use them incorrectly. Does that seem reasonable?

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