1484888 - Avoid clang crash with a stack overflow while doing PGO + LTO on Linux

Status: RESOLVED FIXED

(Firefox Build System :: Toolchains, enhancement)

Comment 1

[Mike Hommey [:glandium]]

Attachment: Apply upstream patch that fixes a clang crash with stack overflow on PGO + LTO on Linux.

For some reason, clang 6 crashes with a stack overflow on PGO + LTO on
Linux 64 bits. Clang 7 doesn't, but has other problems.

After some bisecting, I found the following:
- r322684 is the first revision that is broken on the release_60 branch.
- that revision is a cherry pick of r322313 from trunk, which is
  similarly broken.
- trunk was fixed by r322325, which, funnily enough, predates when
  r322313 was cherry-picked.

While the change from r322325 is relatively large, mixing multiple
different changes in a single commit, there also haven't been
significant changes to the same file on trunk since (one macro name
change, one documentation change, and a change related to debug info),
which would tend to indicate the change is not going to break anything,
or at least not more than upgrading to clang 7 would.

The exact part that fixes the issue could probably be found in this
large commit, but I didn't feel like digging into it further considering
the above.

Comment 2

[Nathan Froyd [:froydnj]]

Comment on attachment 9002643 [details] [diff] [review]
Apply upstream patch that fixes a clang crash with stack overflow on PGO + LTO on Linux.

Review of attachment 9002643 [details] [diff] [review]:
-----------------------------------------------------------------

This patch is a little terrifying.

I would suggest filing a bug with LLVM pointing to this patch as worth a backport, but I guess the 6.x branch is probably dead at this point?

Comment 3

[(Away)]

(In reply to Mike Hommey [:glandium] from comment #1)
> While the change from r322325 is relatively large, mixing multiple
> different changes in a single commit, there also haven't been
> significant changes to the same file on trunk since (one macro name
> change, one documentation change, and a change related to debug info),
> which would tend to indicate the change is not going to break anything,
> or at least not more than upgrading to clang 7 would.

It's amusing and mildly troubling that the fix comes from a patch labeled as no functional change. :/

Comment 4

[Mike Hommey [:glandium]]

Filed https://bugs.llvm.org/show_bug.cgi?id=38663

Comment 5

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/mozilla-inbound/rev/737a89a8edb7
Apply upstream patch that fixes a clang crash with stack overflow on PGO + LTO on Linux. r=froydnj

Comment 6

[Andreea Pavel [:apavel]]

https://hg.mozilla.org/mozilla-central/rev/737a89a8edb7