|
|
DescriptionFix webview preloading in zygote.
We weren't setting the process type, which is fetched via JNI during
JNI_OnLoad. Set it by constructing the LibraryLoader singleton, but
don't actually call any methods of LibraryLoader, which need to be on
the right thread, have a Context, have a CommandLine initialised, and
other setup we can't (or at least, don't currently) do in the zygote.
BUG=
Committed: https://crrev.com/db6be0f685e774c56eef7d03a8ab19736023d8bb
Cr-Commit-Position: refs/heads/master@{#439213}
Patch Set 1 #
Messages
Total messages: 19 (10 generated)
torne@chromium.org changed reviewers: + rsesek@chromium.org, tobiasjs@chromium.org
This is the change I think will make a difference (this may not be enough initialisation, but this is definitely required as a minimum, and it's probably only luck that standalone webview actually works without this). I haven't compiled or tested this yet, but I'm working on it. Uploading now so that you can try it yourself if you're in a better position than me (and in case I have to leave before I'm done).
Tested this and it appears to fix the crash on startup. Monochrome works in multiprocess for me with this patch (plus the fd whitelist fix for the framework)
LGTM. Verified that this fixes the issue we're seeing.
The CQ bit was checked by rsesek@chromium.org to run a CQ dry run
Dry run: CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...
The CQ bit was unchecked by commit-bot@chromium.org
Dry run: This issue passed the CQ dry run.
rsesek@chromium.org changed reviewers: + sgurun@chromium.org
LGTM
The CQ bit was checked by tobiasjs@chromium.org
CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...
CQ is committing da patch. Bot data: {"patchset_id": 1, "attempt_start_ts": 1481926100416430, "parent_rev": "903c721c6be668f508c325ad3bc3dfde8e794db9", "commit_rev": "752346c6d8df467e4d72101c40af3b615588535c"}
Message was sent while issue was closed.
Description was changed from ========== Fix webview preloading in zygote. We weren't setting the process type, which is fetched via JNI during JNI_OnLoad. Set it by constructing the LibraryLoader singleton, but don't actually call any methods of LibraryLoader, which need to be on the right thread, have a Context, have a CommandLine initialised, and other setup we can't (or at least, don't currently) do in the zygote. BUG= ========== to ========== Fix webview preloading in zygote. We weren't setting the process type, which is fetched via JNI during JNI_OnLoad. Set it by constructing the LibraryLoader singleton, but don't actually call any methods of LibraryLoader, which need to be on the right thread, have a Context, have a CommandLine initialised, and other setup we can't (or at least, don't currently) do in the zygote. BUG= Review-Url: https://codereview.chromium.org/2583913002 ==========
Message was sent while issue was closed.
Committed patchset #1 (id:1)
Message was sent while issue was closed.
Description was changed from ========== Fix webview preloading in zygote. We weren't setting the process type, which is fetched via JNI during JNI_OnLoad. Set it by constructing the LibraryLoader singleton, but don't actually call any methods of LibraryLoader, which need to be on the right thread, have a Context, have a CommandLine initialised, and other setup we can't (or at least, don't currently) do in the zygote. BUG= Review-Url: https://codereview.chromium.org/2583913002 ========== to ========== Fix webview preloading in zygote. We weren't setting the process type, which is fetched via JNI during JNI_OnLoad. Set it by constructing the LibraryLoader singleton, but don't actually call any methods of LibraryLoader, which need to be on the right thread, have a Context, have a CommandLine initialised, and other setup we can't (or at least, don't currently) do in the zygote. BUG= Committed: https://crrev.com/db6be0f685e774c56eef7d03a8ab19736023d8bb Cr-Commit-Position: refs/heads/master@{#439213} ==========
Message was sent while issue was closed.
Patchset 1 (id:??) landed as https://crrev.com/db6be0f685e774c56eef7d03a8ab19736023d8bb Cr-Commit-Position: refs/heads/master@{#439213}
Message was sent while issue was closed.
A revert of this CL (patchset #1 id:1) has been created in https://codereview.chromium.org/2586193002/ by tobiasjs@chromium.org. The reason for reverting is: Realised after the fact that we can't set the process type because the zygote may spawn renderers for Chrome.. |