Skip to content

Backported fix for string.find recording - #10

Merged
EinsamHauer merged 1 commit into
luavela:masterfrom
igormunkin:master
Sep 27, 2019
Merged

EinsamHauer merged 1 commit into
luavela:masterfrom
igormunkin:master

Conversation

@igormunkin

Copy link
Copy Markdown
Member

This is a squashed changeset of the following commits from
tarantool/luajit repo:

Relates to LuaJIT gh-505

This is a squashed changeset of the following commits from
tarantool/luajit repo:
* 587532e
* 98b1cfc

Relates to LuaJIT gh-505

@igelhaus igelhaus left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

Looks good to me.

@igelhaus

Copy link
Copy Markdown
Contributor

Oh, by the way, @igormunkin, could you please provide a link to the CI results for your patch? I cannot find any.

Comment thread ChangeLog
* Added MOVTVPRI optimization, which extends MOVTV for recording-time nil, false and true values
* Extended information about contributors
* Excluded movtv and movtvpri from -O4
* Backported fix for string.find recording from tarantool/luajit repo

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

The fact that this is a backport is mentioned in a commit. Here, it would be more useful to describe what about string.find recording was fixed.

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.

IMHO, it's a bit excessive condering a conversation within the related LuaJIT issue.

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

The line in ChangeLog doesn't point to an issue, though

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.

Commit message does. The issue can be obtained by blaming exactly this ChangeLog line.

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

Then it's enough to write "Fix string.find fold machinery"

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.

No, it's a backporting of the specified commits. Yes, they are adjusted considering uJIT specific, but the original commits are from tarantool/luajit repo. The similar we have here. See no sense in the further conversation. Your move.

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

The commit hash is not specified in ChangeLog, which is why I commented on that. So, for message to be helpful, it should either have a commit hash in it, or shortly describe what was fixed.


$tester->run('find-fold-bug.lua', args => '-p-')
->exit_ok
->stdout_has_no(qr/TRACE.+?abort.?/)

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

Was the trace abort the only observable behaviour of this bug? I think there needs to be a check that fold transformations are not buggy (std_out_has_no(/* before fix output */))

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 guess it is enough. Previous implementation of string.find recording breaks the sematics while folding. Thereby we need to check whether a trace is assembled and the result is valid (see the assert in the corresponding lua chunk).

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

Was there a trace abort before the change?

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.

There were no aborts before (please see the original issue and commits for more info).

@igormunkin

igormunkin commented Sep 27, 2019

Copy link
Copy Markdown
Member Author

Oh, by the way, @igormunkin, could you please provide a link to the CI results for your patch?

@igelhaus, here you are: travis.

I cannot find any.

You can find them near green tick with "All check have passed" within details.

@EinsamHauer
EinsamHauer merged commit 8efe8cd into luavela:master Sep 27, 2019
@igormunkin

Copy link
Copy Markdown
Member Author

@EinsamHauer, thanks a lot.

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.

4 participants