CrWinClangLLD should not use symbol_level = 0 |
||
Issue descriptionIn https://chromium-review.googlesource.com/c/chromium/src/+/757188#message-8f12a0d64277ad2d4efd4be3101c9d453acc1957, we discovered that https://build.chromium.org/p/chromium.clang/builders/CrWinClangLLD is using symbol_level = 0. Me: > Another side of this, though, is that the ability to generate debug info (including > proper .pdb files) is a pretty important part of the linker, so I find it a little > weird that your chromium-win-clang-lld tester turns off .pdb generation. Is that an > oversight? Reid: > It's a hold-over from when LLD's PDB writing code was new and less reliable. We > configured the LLD release bots to disable PDB generation and configured the debug bots > to generate PDBs. Otherwise, bugs in PDB writing would blind us to regressions in plain > executable linking. We could set this bot back to symbol_level=1 at this point. I’ll deal with the test failure at symbol_level = 0 in bug 782781 , but this builder should be reconfigured to not use symbol_level = 0.
,
Nov 10 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/edcc2158fce0f0d8bdb9992186ce4c613810a652 commit edcc2158fce0f0d8bdb9992186ce4c613810a652 Author: Reid Kleckner <rnk@google.com> Date: Fri Nov 10 00:39:47 2017 Use symbol_level=1 on non-debug LLD bots We needed symbol_level=0 as a workaround for http://crbug.com/706492 , but that's fixed upstream, so we don't need it anymore. Most bots used symbol_level=1, so the LLD release bots should too. R=dpranke@chromium.org, hans@chromium.org Bug: 706492 , 782784 Change-Id: Idc8596df997a77e278cea82745430adc4eb25644 Reviewed-on: https://chromium-review.googlesource.com/758702 Reviewed-by: Hans Wennborg <hans@chromium.org> Reviewed-by: Dirk Pranke <dpranke@chromium.org> Commit-Queue: Reid Kleckner <rnk@chromium.org> Cr-Commit-Position: refs/heads/master@{#515371} [modify] https://crrev.com/edcc2158fce0f0d8bdb9992186ce4c613810a652/tools/mb/mb_config.pyl
,
Nov 10 2017
|
||
►
Sign in to add a comment |
||
Comment 1 by jpar...@google.com
, Nov 9 2017