Skip to content

Clean-up code related to AlreadySignExtended register flag#8058

Open
r30shah wants to merge 1 commit intoeclipse-omr:masterfrom
r30shah:removeAlreadySignExtendedFlag
Open

Clean-up code related to AlreadySignExtended register flag#8058
r30shah wants to merge 1 commit intoeclipse-omr:masterfrom
r30shah:removeAlreadySignExtendedFlag

Conversation

@r30shah
Copy link
Contributor

@r30shah r30shah commented Nov 26, 2025

In #8035 we introduced a new node flag to notify the code-gen that the node was sign extended when evaluated. This removes the the use of AlreadySignExtended register flag which can be propagated to other nodes if the register is reused through clobber evaluation of the node.
This commit includes the clean-up related AlreadySignExtended register flag.

In eclipse-omr#8035 we introduced a new
node flag to notify the code-gen that the node was sign extended when
evaluated. This removes the the use of AlreadySignExtended register
flag which can be propagated to other nodes if the register is reused
through clobber evaluation of the node.
This commit includes the clean-up related AlreadySignExtended register
flag.

Signed-off-by: Rahil Shah <rahil@ca.ibm.com>
@r30shah r30shah force-pushed the removeAlreadySignExtendedFlag branch from eae5121 to f470de9 Compare November 26, 2025 15:15
@r30shah
Copy link
Contributor Author

r30shah commented Nov 26, 2025

As recommended in #8035 (review) , opening up this PR to clean-up AlreadySignExtended flag.

@joransiu / @hzongaro Can I request your review on this change ?

@0xdaryl
Copy link
Contributor

0xdaryl commented Jan 17, 2026

Jenkins build zlinux,zos

@0xdaryl 0xdaryl self-assigned this Jan 17, 2026
@0xdaryl 0xdaryl added the arch:z label Jan 17, 2026
@0xdaryl 0xdaryl self-requested a review January 17, 2026 03:33
@0xdaryl
Copy link
Contributor

0xdaryl commented Jan 17, 2026

By removing alreadySignExtended this way, you are asserting that its removal means that alreadySignExtended is always true on the register. Another way to look at it is that since nothing sets alreadySignExtended any longer that it should be considered false. Your interpretation in this PR is the former, but I just want to confirm these semantics.

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

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants