jit: func_line alignment for line table #9003
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
CT Test Results 3 files 143 suites 49m 40s ⏱️ Results for commit 5295981. ♻️ This comment has been updated with latest results. To speed up review, make sure that you have read Contributing to Erlang/OTP and that all checks pass. See the TESTING and DEVELOPMENT HowTo guides for details about how to run test locally. Artifacts// Erlang/OTP Github Action Bot |
|
Github actions fail due to formatting errors, please run |
garazdawi
removed
the
testing
The changes from 5b4457e cause the line table to have extra entries when handling the last_error_offset. The line number from the subsequent function is mapped to the nop instruction following the error, and as a result, the line table suggests that the line is mapped in the former and current range. As an example, if you enable logging in jit-reader.c, and look for sets.erl, you would see something like: ``` Add range `sets:new/0-CodeInfoPrologue` (0x7f5fddf03f18, 0x7f5fddf03f48), 0 lines Add range `sets:new/0` (0x7f5fddf03f48, 0x7f5fddf04008), 3 lines sets.erl:170 sets.erl:171 sets.erl:177 Add range `sets:new/1-CodeInfoPrologue` (0x7f5fddf04008, 0x7f5fddf04038), 0 lines Add range `sets:new/1` (0x7f5fddf04038, 0x7f5fddf04118), 2 lines sets.erl:177 sets.erl:180 ``` which shows that line 177 is included twice, despite only being for sets:new/1. This can also be identified by using this GDB command: `maintenance info line-table sets.erl`. Move func_line back after the func label and emit the last error nop before the label. This ensures that the line is within the (possibly aligned) func range.
danielfinke
force-pushed
the
jit-func-line-alignment
branch
from
e5a9522 to
5295981
Compare
Fixed formatting with a typo fix & ran |
jhogberg
added
the
testing
|
Can this issue also affect function attribution in tools like perf? |
jhogberg
approved these changes
Nov 18, 2024
|
Merged, thanks for the PR! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The changes from 5b4457e cause the line table to have extra entries when handling the last_error_offset. The line number from the subsequent function is mapped to the nop instruction following the error, and as a result, the line table suggests that the line is mapped in the former and current range.
As an example, if you enable logging in jit-reader.c, and look for sets.erl, you would see something like:
which shows that line 177 is included twice, despite only being for sets:new/1. This can also be identified by using this GDB command:
maintenance info line-table sets.erl.Move func_line back after the func label and emit the last error nop before the label. This ensures that the line is within the (possibly aligned) func range.