Chromium Code Reviews
|
|
Chromium Code Reviews
Issue
1949113002:
Roll clang 268373:268419 (Closed)
|
Created:
4 years, 7 months ago by Reid Kleckner Modified:
4 years, 7 months ago CC:
chromium-reviews, yunlian, eugenis+clang_chromium.org, glider+clang_chromium.org, dmikurube+clang_chromium.org, ukai+watch_chromium.org Base URL:
https://chromium.googlesource.com/chromium/src.git@master Target Ref:
refs/pending/heads/master Project:
chromium Visibility:
Public. |
DescriptionRoll clang 268373:268419
Picks up Windows fix for a miscompile of a virtual method call
on a final class in blink.
R=thakis@chromium.org,hans@chromium.org
BUG=608705
Committed: https://crrev.com/7f225037d3373bc6b4c37cd122b3235902e1602c
Cr-Commit-Position: refs/heads/master@{#391887}
Patch Set 1 #Patch Set 2 : Rebase, update description, this is just about devirtualization #Messages
Total messages: 21 (6 generated)
If you use tools/clang/scripts/upload_revision.py, it'll send try jobs to build and upload clang. (You can trigger them manually too since you already have this CL; unless you've already built and uploaded binaries.) Change lgtm once binaries exist and are in goma.
We already rolled in the pdata fix yesterday night (https://bugs.chromium.org/p/chromium/issues/detail?id=607312#c19), so you can remove that bug from the BUG= line
Description was changed from ========== Roll clang 267383:268419 Picks up Windows fixes for .pdata linker errors and a miscompile of a virtual method call on a final class in blink. R=thakis@chromium.org,hans@chromium.org BUG=608705,595702 ========== to ========== Roll clang 268373:268419 Picks up Windows fix for a miscompile of a virtual method call on a final class in blink. R=thakis@chromium.org,hans@chromium.org BUG=608705 ==========
Rebase, update description, this is just about devirtualization
looks like tests are broken on windows at that revision :-/ The failing test is new as of r268404 ******************** TEST 'lld :: ELF/reproduce-windows.s' FAILED ********************
Looks like the tests depends on "cpio" which isn't in our prebuilt binaries zip. So maybe that bot will succeed if we add that binary to the gnuwin zip at http://commondatastorage.googleapis.com/chromium-browser-clang/index.html?pat... and update update.py to pull that new zip
We could just wait longer to roll, unless you think the devirtualization miscompile is likely to hurt someone else soon. I wasn't aware that the .pdata fix was already in. I made https://codereview.chromium.org/1952433003/ so that won't be an issue going forwards.
Since we already have Mac and Linux binaries, and since we need to figure out cpio anyways, we might want to finish this roll. (Given that nobody ran into this miscompile for years, it's unlikely we'll hit it again soon though.) If you don't feel like fiddling with the gnuwin zip, that's understandable. If you want, I can try fixing that tomorrow and the trigger the win upload bot again. On May 4, 2016 1:33 PM, <rnk@chromium.org> wrote: > We could just wait longer to roll, unless you think the devirtualization > miscompile is likely to hurt someone else soon. > > I wasn't aware that the .pdata fix was already in. I made > https://codereview.chromium.org/1952433003/ so that won't be an issue > going > forwards. > > https://codereview.chromium.org/1949113002/ > -- You received this message because you are subscribed to the Google Groups "Chromium-reviews" group. To unsubscribe from this group and stop receiving emails from it, send an email to chromium-reviews+unsubscribe@chromium.org.
On 2016/05/04 17:36:59, Nico (hiding Wed May 4) wrote: > If you don't feel like fiddling with the gnuwin zip, that's understandable. > If you want, I can try fixing that tomorrow and the trigger the win upload > bot again. I think I have enough stuff for today, so if you want to fiddle with it tomorrow, that'd be great.
Ok, the Windows binary is built. Unless someone beats me to it, I'll push it to goma tomorrow morning.
The CQ bit was checked by thakis@chromium.org to run a CQ dry run
Dry run: CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/patch-status/1949113002/20001 View timeline at https://chromium-cq-status.appspot.com/patch-timeline/1949113002/20001
The CQ bit was unchecked by thakis@chromium.org
The linux_chromium_asan_rel_ng result is a bit inconclusive, but other than that lgtm. I'll hit cq.
The CQ bit was checked by thakis@chromium.org
CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/patch-status/1949113002/20001 View timeline at https://chromium-cq-status.appspot.com/patch-timeline/1949113002/20001
Message was sent while issue was closed.
Description was changed from ========== Roll clang 268373:268419 Picks up Windows fix for a miscompile of a virtual method call on a final class in blink. R=thakis@chromium.org,hans@chromium.org BUG=608705 ========== to ========== Roll clang 268373:268419 Picks up Windows fix for a miscompile of a virtual method call on a final class in blink. R=thakis@chromium.org,hans@chromium.org BUG=608705 ==========
Message was sent while issue was closed.
Committed patchset #2 (id:20001)
Message was sent while issue was closed.
Description was changed from ========== Roll clang 268373:268419 Picks up Windows fix for a miscompile of a virtual method call on a final class in blink. R=thakis@chromium.org,hans@chromium.org BUG=608705 ========== to ========== Roll clang 268373:268419 Picks up Windows fix for a miscompile of a virtual method call on a final class in blink. R=thakis@chromium.org,hans@chromium.org BUG=608705 Committed: https://crrev.com/7f225037d3373bc6b4c37cd122b3235902e1602c Cr-Commit-Position: refs/heads/master@{#391887} ==========
Message was sent while issue was closed.
Patchset 2 (id:??) landed as https://crrev.com/7f225037d3373bc6b4c37cd122b3235902e1602c Cr-Commit-Position: refs/heads/master@{#391887} |
