On Wed, Apr 27, 2022 at 5:14 PM Chris Harrelson <chris...@chromium.org> wrote:
> > > On Wed, Apr 27, 2022 at 6:04 AM Lutz Vahl <v...@chromium.org> wrote: > >> Contact emails >> >> v...@chromium.org cl...@chromium.org >> >> Explainer >> >> >> https://docs.google.com/document/d/1zDlfvfTJ_9e8Jdc8ehuV4zMEu9ySMCiTGMS9y0GU92k >> >> Specification >> >> https://tc39.github.io/ecma262/#sec-sharedarraybuffer-objects >> >> Design docs Including the new security requirements >> >> >> https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/SharedArrayBuffer >> >> Discussion how and what to gate >> >> https://github.com/whatwg/html/issues/4732 >> >> Summary >> >> ‘SharedArrayBuffers’ (SABs) on desktop platforms are restricted to >> cross-origin isolated environments, matching the behavior we've recently >> shipped on Android and Firefox. We've performed that change in Chrome 92. A >> reverse OT was started to give developers the option to use SABs in case >> they are not able to adopt cross origin isolation yet. >> >> We’ve received lot’s of feedback that adopting COOP/COEP is hard (details >> below). Therefore I’m asking for your approval to extend the SAB reverse OT >> again from M103 until M113 (branch point 2023-03-23). This is an >> estimation - Can we come back to y'all in 6 months with a report on >> progress and usage to justify that extension and agree on the final >> milestone? >> >> Experimental timeline / plan for all new capabilities needed to replace >> the OT >> >> The SAB restriction in M92 went smoothly without any major issues in the >> wild because we offered the reverse OT. We’ve received lots of feedback >> that adopting COOP/COEP is hard and sometimes impossible. Therefore the >> reverse OT is currently the only way to enable SABs for some sites within >> Chromium. Chromestatus is showing that SABs in none COI context are being >> used on ~0.36% >> <https://chromestatus.com/metrics/feature/popularity#V8SharedArrayBufferConstructedWithoutIsolation> >> page loads. >> > > This seems off by a factor of 10. The real number seems to be 0.036% or so > <https://chromestatus.com/metrics/feature/timeline/popularity/3721>, > right? Can you highlight why it's important to extend for 10 more > milestones for such a small percentage of traffic? Will the sites in > question completely break for some reason, or just behave the same as in > non-chromium browsers? > That's on me: 0.036% <https://chromestatus.com/metrics/feature/timeline/popularity/3721> is correct! Some sites use SAB to gain extra performance on chromium based browsers in some cases 3P content is using SABs. Some might work without the OT others will break based on how they identify their code path to be used. The list of OT registrations is ~500 and most of them mentioned to be blocked by 3Ps to deploy COOP+COEP broadly. We're happy to extend the OT to give them time to adopt. Do you (and/or other API owners) think this is not required based on the low usage? > >> >> To overcome this limitation and make adoption possible more broadly (public >> feedback <https://github.com/WICG/proposals/issues/53>), we’re working >> on multiple solutions >> <https://github.com/camillelamy/explainers/blob/main/cross-origin-isolation-deployment.md> >> (all shared timelines are WIP): >> >> >> 1. >> >> COEP:credentialless <https://github.com/WICG/credentiallessness> - >> https://crbug.com/1218896 >> >> COEP:credentialless causes no-cors cross-origin requests not to include >> >> credentials (cookies, client certificates, etc...). Similarly to >> require-corp, it can be used to enable cross-origin-isolation. Some >> developers are blocked on a set of dependencies which don't yet assert that >> they're safe to embed in cross-origin isolated environments. >> >> This mechanism was shipped in M96. (Adoption is already at 0.02% >> <https://chromestatus.com/metrics/feature/popularity#CrossOriginEmbedderPolicyCredentialless> >> of main pages) >> >> >> 1. >> >> COI+popups (formally: COOP same-origin-allow-popups-plus-coep >> <https://github.com/camillelamy/explainers/blob/main/coi-with-popups.md> >> ) >> >> To allow crossOriginIsolated pages to use popup-based OAuth/payment >> flows, we plan to have COOP same-origin-allow-popups enable >> crossOriginIsolation when used in conjunction with COEP. Developers who >> depend on popups to 3P for e.g. identity or payment flows can’t currently >> deploy cross-origin-isolation. >> >> Spec work is ongoing and we’re targeting Q2 2022 for the OT and Q3 for >> the shipping. As soon as the spec is defined, we’ll kick off the intent >> process. Without this all sites need to migrate to FedCM and WebPayment for >> their flows to be able to use SABs. >> >> >> >> 1. >> >> Anonymous iframes <https://github.com/WICG/anonymous-iframe> >> >> Anonymous iframes are a generalization of COEP credentialless to support >> 3rd party iframes that may not deploy COEP. Like with COEP credentialless, >> we replace the opt-in of cross-origin subresources by avoiding to load >> non-public resources. This will remove the constraint and will unblock >> developers to adopt cross-origin-isolation as soon as they’re embedding 3P >> iframes. >> >> Based on the progress made for storage partitioning and CHIPs, which are >> needed to safely ship Anonymous iframes, we’re aiming to start the OT in Q2 >> 2022 (M106) and the rollout in Q3 2022 (M110). >> >> Blink component >> >> Blink>JavaScript >> <https://bugs.chromium.org/p/chromium/issues/list?q=component:Blink%3EJavaScript> >> >> Search tags >> >> SharedArrayBuffer >> <https://chromestatus.com/features#tags:SharedArrayBuffer>, SAB >> <https://chromestatus.com/features#tags:SAB> >> >> TAG reviewhttps://github.com/w3ctag/design-reviews/issues/471 >> TAG review statusClosed >> RisksInteroperability and Compatibility >> >> We expect this change to negatively impact developers using >> `SharedArrayBuffer` today. Chrome was the only platform where SABs have >> been available without COOP/COEP. Therefore we need to give developers the >> right capabilities and a clear path forward to ensure they’ve enough time >> to adopt. We aim to mitigate these risks by adopting a longer-than-usual >> depreciation period with console warnings/issues and a reverse origin >> trial. >> >> Good news is usage is down to ~0.36% >> <https://chromestatus.com/metrics/feature/popularity#V8SharedArrayBufferConstructedWithoutIsolation> >> page loads and that other browsers have or are shipping SABs again gated >> behind COOP/COEP. Bad news is that Chromium was the only browser that >> supported SABs without COI, therefore we need to provide a migration path >> to not break existing sites such as Zoom or Google Earth. >> >> Gecko: Shipped/Shipping ( >> https://bugzilla.mozilla.org/show_bug.cgi?id=1312446) >> >> WebKit: Added COOP/COEP and SAB support recently gated behind COOP/COEP >> >> Will this feature be supported on all six Blink platforms (Windows, Mac, >> Linux, Chrome OS, Android, and Android WebView)? >> >> No - This OT is only for desktop, as this was the only platform where >> SABs have been available without COOP/COEP. >> >> Android re-enabled SABs gated behind COOP/COEP: >> https://chromestatus.com/feature/5171863141482496 >> >> Tracking bug >> >> https://bugs.chromium.org/p/chromium/issues/detail?id=1144104 >> >> Launch bug >> >> https://bugs.chromium.org/p/chromium/issues/detail?id=1138860 >> >> Blink-dev Thread >> >> Planning isolation requirements (COOP/COEP) for SharedArrayBuffer >> <https://groups.google.com/a/chromium.org/g/blink-dev/c/_0MEXs6TJhg/m/QzWOGv7pAQAJ> >> >> I2S >> <https://groups.google.com/a/chromium.org/g/blink-dev/c/1NKvbIj3dq4/m/nLcgUst-BQAJ> >> >> Link to entry on the Chrome Platform Status >> >> https://chromestatus.com/feature/4570991992766464 >> >> -- >> You received this message because you are subscribed to the Google Groups >> "blink-dev" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to blink-dev+unsubscr...@chromium.org. >> To view this discussion on the web visit >> https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAH0ixBN2JhcYtpT4UYKcAfHt1e0Wz_Uxz0CkXcAntguhbmyNCA%40mail.gmail.com >> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAH0ixBN2JhcYtpT4UYKcAfHt1e0Wz_Uxz0CkXcAntguhbmyNCA%40mail.gmail.com?utm_medium=email&utm_source=footer> >> . >> > -- > You received this message because you are subscribed to the Google Groups > "blink-dev" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to blink-dev+unsubscr...@chromium.org. > To view this discussion on the web visit > https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAOMQ%2Bw_HkK7R3fA0pyGUm8MNjbqoBR54XrQZWKeD464qb6JNhA%40mail.gmail.com > <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAOMQ%2Bw_HkK7R3fA0pyGUm8MNjbqoBR54XrQZWKeD464qb6JNhA%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > -- You received this message because you are subscribed to the Google Groups "blink-dev" group. To unsubscribe from this group and stop receiving emails from it, send an email to blink-dev+unsubscr...@chromium.org. To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAH0ixBMg_ydJqgjAdPaWZKfv0Xqjj9xqs4kRENmFUgLLy2ZtaA%40mail.gmail.com.
