PSA: Removing chrome://flags/#enable-site-per-process on desktop platforms

197 views
Skip to first unread message

Łukasz Anforowicz

unread,
Nov 5, 2018, 2:14:31 PM11/5/18
to embedd...@chromium.org, site-isol...@chromium.org
Hello embedd...@chromium.org,

PSA: https://crrev.com/c/1316649 will remove chrome://flags/#enable-site-per-process on desktop platforms (because on desktop platforms site-per-process has shipped to Chrome Stable channel in M67).  The removal of chrome://flags/#enable-site-per-process might matter for Chromium Embedders who re-use chrome://flags from //chrome layer and who haven't yet enabled Site Isolation on desktop.

Note that even after removal of the flag, embedders of //content can still control whether Strict Site Isolation (aka site-per-process) is enabled by either overriding ContentBrowserClient::ShouldEnableStrictSiteIsolation.  Additionally, individual users can still control site isolation via --site-per-process (forces strict site isolation) and --disable-site-isolation-trials (disables all site isolation) command line switches.

--
Thanks,

Lukasz

Łukasz Anforowicz

unread,
Nov 5, 2018, 2:16:06 PM11/5/18
to embedd...@chromium.org, site-isol...@chromium.org

PSA: https://crrev.com/c/1316649 will remove chrome://flags/#enable-site-per-process on desktop platforms (because on desktop platforms site-per-process has shipped to Chrome Stable channel in M67).  The removal of chrome://flags/#enable-site-per-process might matter for Chromium Embedders who re-use chrome://flags from //chrome layer and who haven't yet enabled Site Isolation on desktop.

Note that even after removal of the flag, embedders of //content can still control whether Strict Site Isolation (aka site-per-process) is enabled by either overriding ContentBrowserClient::ShouldEnableStrictSiteIsolation.  Additionally, individual users can still control site isolation via --site-per-process (forces strict site isolation) and --disable-site-isolation-trials (disables all site isolation) command line switches.

Thanks,

Lukasz

PS. My apologies if you get this email twice - I've initially sent it from a wrong email address... :-/
Reply all
Reply to author
Forward
0 new messages