Uh oh!
There was an error while loading. Please reload this page.
Prevent ghost breakpoints by making sure we can't have dangling trace points - #21
Conversation
vinistock
commented
Jul 31, 2023
Not sure why the tests are failing. They pass on upstream main. |
st0012
left a comment
There was a problem hiding this comment.
Is it possible to recreate the issue in console tests and write regression tests for the fixes?
I think the fix in on_load may be easier to recreate and then we can upstream the fix/open an issue later?
Uh oh!
There was an error while loading. Please reload this page.
15cfc7e to
ea383beComparevinistock
commented
Jul 31, 2023
Is there an existing test for |
st0012
commented
Jul 31, 2023
I think these tests were written to cover it. |
ea383be to
037ad7eComparevinistock
commented
Aug 1, 2023
Paired with Stan and we discussed a few things:
|
vinistock
commented
Aug 2, 2023
Okay, I paired with Peter and got a test working. The issue is the fact that some iseqs may be kept around even after a file is reloaded. One scenario where this happens is when The test reproduces the problem and the fix makes the test pass. |
vinistock
commented
Aug 2, 2023
Upstream PR ruby#1008. |
Any time we disable a breakpoint or re-set the
tpinstance variable, we need to make sure any previous trace points are disabled or else they might continue to suspend execution even after breakpoint removal.