+rnk who asked about this yesterday


ClangToTLinuxUBSanVptr is still red, but should go green after next V8 roll
This was https://bugs.chromium.org/p/v8/issues/detail?id=5628#c6
It's not a Clang problem though, so doesn't block the roll, but it could be
hiding other breakage.


CrWinAsan(dll) is red with a Clang assert.
https://build.chromium.org/p/chromium.fyi/builders/CrWinAsan%28dll%29/builds/5246
This is http://llvm.org/pr31029
Will hopefully be fixed very soon.


content_browsertsests failing some DevToolsProtocolTest. This has been happening
for a while, and also on the non-ToT bots, so maybe not a roll blocker?


There's a new component_unittests failure,  crbug.com/665951 

The bots are turning red faster than they're turning green :-(

ClangToTLinuxLLD failing blink_heap_unittests and webkit_unit_tests:
https://build.chromium.org/p/chromium.fyi/builders/ClangToTLinuxLLD%20tester/builds/304
But it's only one failure. Some kind of hickup?


ClangToTMac failing telemetry_perf_unittests:
https://build.chromium.org/p/chromium.fyi/builders/ClangToTMac%20tester/builds/5669
Also just one failure. Looks like the same thing happened 10 days ago for one build and went away. Flake?


ClangToTMacAsan failing browser_tests
https://build.chromium.org/p/chromium.fyi/builders/ClangToTMacASan%20tester/builds/3420
Also on a single build so far, but looks less flaky.
It's reporting a use-after-free running PolicyMakeDefaultBrowserTest.MakeDefaultDisabled


ClangToTWin(dbg) failed browser_tests
https://build.chromium.org/p/chromium.fyi/builders/ClangToTWin%28dbg%29%20tester/builds/3471
Single build so far.

I committed a new warning for sashab@chromium.org, so that's probably going to start showing up on the waterfall soon.

> I committed a new warning for sashab@chromium.org, so that's probably going to start showing up on the waterfall soon.

Here we go:
https://build.chromium.org/p/chromium.fyi/builders/ClangToTMac/builds/12116/steps/compile/logs/stdio

../../third_party/skia/include/private/SkRecords.h:195:49: error: enums in the Microsoft ABI are signed integers by default; consider giving the enum Op an unsigned underlying type to make this code portable [-Werror,-Wsigned-enum-bitfield]
    ClipOpAndAA(SkCanvas::ClipOp op, bool aa) : op(op), aa(aa) {}
                                                ^
1 error generated.

I'll commence the disable-new-warning dance.

Sorry - that's my fault. I was working on adding that flag to just skia:
https://codereview.chromium.org/2354453002

Where did you disable the warning? :) I'll work on adding it specifically for skia because it needs the warning disabled anyway.

> Where did you disable the warning? :) I'll work on adding it specifically for skia because it needs the warning disabled anyway.

Here:

https://codereview.chromium.org/2511613003/

There's still some red on the ToT bots, but none of it is believed to be due to Clang regressions, so let's try to roll.

Trying r287226 here: https://codereview.chromium.org/2508323003

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/9b87c91ac267d38736890d2b63e6e56a893ccbbb

commit 9b87c91ac267d38736890d2b63e6e56a893ccbbb
Author: hans <hans@chromium.org>
Date: Fri Nov 18 05:50:31 2016

Roll clang 284979:287226.

Ran `tools/clang/scripts/upload_revision.py 287226`.

BUG= 665063 

Review-Url: https://codereview.chromium.org/2508323003
Cr-Commit-Position: refs/heads/master@{#433124}

[modify] https://crrev.com/9b87c91ac267d38736890d2b63e6e56a893ccbbb/tools/clang/scripts/update.py

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/ece50209a690cba3f3450d9d11296be844a94ca1

commit ece50209a690cba3f3450d9d11296be844a94ca1
Author: thakis <thakis@chromium.org>
Date: Fri Nov 18 17:41:15 2016

Revert of Roll clang 284979:287226. (patchset #1 id:1 of https://codereview.chromium.org/2508323003/ )

Reason for revert:
Makes every link on Mac warn

clang: warning: libLTO.dylib relative to clang installed dir not found; using 'ld' default search path instead [-Wliblto]

Probably due to https://reviews.llvm.org/D25932. We need to talk to Mehdi and come up with a plan. Also, it's lame that that warning isn't affected by -Werror, else the bots would've caught this.

Original issue's description:
> Roll clang 284979:287226.
>
> Ran `tools/clang/scripts/upload_revision.py 287226`.
>
> BUG= 665063 
>
> Committed: https://crrev.com/9b87c91ac267d38736890d2b63e6e56a893ccbbb
> Cr-Commit-Position: refs/heads/master@{#433124}

TBR=hans@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG= 665063 

Review-Url: https://codereview.chromium.org/2520453002
Cr-Commit-Position: refs/heads/master@{#433230}

[modify] https://crrev.com/ece50209a690cba3f3450d9d11296be844a94ca1/tools/clang/scripts/update.py

https://reviews.llvm.org/D26944 should silence the warning, if Mehdi likes it.

We also want https://reviews.llvm.org/D26822 in the roll if we can get it, since that'll fix the DevToolsProtocolTest issue.

Looks like we have some new clang assertions:
https://build.chromium.org/p/chromium.fyi/builders/ClangToTWin64%28dbg%29/builds/7533/steps/compile/logs/stdio
Assertion failed: DefaultArg && "sema forgot to instantiate default args"

The clang revision range is 287628:287636, which doesn't include Peter's change. Otherwise, I would strongly suspect that r287600 caused the assertion in MicrosoftCXXABI.cpp.

Maybe it's the Chrome-side change "Enable pre-compiled headers for Blink on Windows" https://codereview.chromium.org/2520863002

https://build.chromium.org/p/chromium.fyi/console?category=win%20clang is green thou---oh right, those use goma and no pch.

We opened  bug 667891  for the pch issue, and disabled one of blink's pchs to hide the bug until we fix it.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/f7e19edb634c3a56e643b18145858fbd7bd9e0c3

commit f7e19edb634c3a56e643b18145858fbd7bd9e0c3
Author: thakis <thakis@chromium.org>
Date: Wed Nov 23 16:48:19 2016

Roll clang 284979:287685.

Ran `tools/clang/scripts/upload_revision.py 287685`.

BUG= 665063 

Review-Url: https://codereview.chromium.org/2523073002
Cr-Commit-Position: refs/heads/master@{#434185}

[modify] https://crrev.com/f7e19edb634c3a56e643b18145858fbd7bd9e0c3/tools/clang/scripts/update.py

Let's hope it sticks.