-
Notifications
You must be signed in to change notification settings - Fork 12k
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Reapply "RegisterCoalescer: Add implicit-def of super register when c…
…oalescing SUBREG_TO_REG" This reverts commit e0f86ca. This was hitting some assertions which have since been relaxed.
- Loading branch information
Showing
5 changed files
with
668 additions
and
55 deletions.
There are no files selected for viewing
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
Oops, something went wrong.
ba385ae
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hey Matt, looks like this broke the MSan buildbot:
https://lab.llvm.org/buildbot/#/builders/74/builds/23298/steps/9/logs/stdio
I've reproduced it on my machine and verified the problem goes away (at
main
) with this patch reverted. I'll also revert 9832eb4 (the dependent tests) to make sure thatcheck-all
is green on main.You can find out more info on how to reproduce the bot at https://github.com/google/sanitizers/wiki/SanitizerBotReproduceBuild. In particular, the buildscript is
buildbot_bootstrap_msan.sh
. If you want to speed it up, comment out all the stuff withcheck_stage{2,3}_msan*
. You can just leave thebuild_stage1_clang
,check_stage1_msan
, andbuild_stage2_msan
to repro the problem.Thanks,
Mitch.
ba385ae
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This also broke the expensive checks bot on green dragon, which may be easier to repro: https://green.lab.llvm.org/green/job/clang-stage1-cmake-RA-expensive/26001/
ba385ae
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
So far no luck trying to reproduce this. I'm quite surprised there are aarch64 failures here, I thought this barely did anything for not-x86. I haven't managed to hit either the x86 or aarch64 cases. This also seems to be using the non-runtimes compiler-rt build?