LGTM1 On Fri, May 2, 2025 at 11:16 AM François Beaufort <fbeauf...@google.com> wrote:
> > > On Fri, May 2, 2025 at 11:09 AM Yoav Weiss (@Shopify) < > yoavwe...@chromium.org> wrote: > >> >> >> On Fri, May 2, 2025 at 10:55 AM 'François Beaufort' via blink-dev < >> blink-dev@chromium.org> wrote: >> >>> Contact emails >>> >>> fbeauf...@google.com >>> >>> Explainer >>> >>> The GPUAdapter isFallbackAdapter boolean attribute is being removed as >>> it’s redundant with the GPUAdapterInfo isFallbackAdapter boolean attribute. >>> This attribute already serves a similar purpose, indicating if an adapter >>> has significant performance limitations in return for wider compatibility, >>> more predictable behavior, and improved privacy. >>> >>> Specification >>> >>> https://github.com/gpuweb/gpuweb/pull/5099 >>> >>> Summary >>> >>> Removes the GPUAdapter isFallbackAdapter boolean attribute from WebGPU, >>> which is redundant with the GPUAdapterInfo isFallbackAdapter boolean >>> attribute. >>> >>> This removal is a minor breaking change as support for fallback adapters >>> has not yet been implemented in any browser, thereby resulting in both >>> isFallbackAdapter attributes consistently returning a falsy value. >>> >>> Blink component >>> >>> Blink>WebGPU >>> <https://bugs.chromium.org/p/chromium/issues/list?q=component:Blink%3EWebGPU> >>> >>> Motivation >>> >>> Removing GPUAdapter isFallbackAdapter boolean attribute eliminates >>> unnecessary complexity and potential confusion by consolidating the >>> functionality within the existing GPUAdapterInfo isFallbackAdapter >>> boolean attribute. This change promotes cleaner code and a more >>> intuitive development experience. >>> >> >> How would breakage look like? if `isFallbackAdapter` turns from a boolean >> to an undefined, is it possible that developers won't notice it's gone? >> > > It is possible that developers won't notice it's gone if they use > something like "if (!adapter.isFallbackAdapter) { ... }". > Having said that. since WebGPU shipped only in Chromium and > isFallbackAdapter has always returned false since Day 1, I highly doubt > there are websites that use it for legitimate reasons yet. > OK! So the only way we'd see breakage is for sites that explicitly check if the attribute is undefined rather than false? That seems safe. > >> >> >> >>> >>> A search for the string "isFallbackAdapter" in HTTPArchive yielded no >>> results. >>> >> >> Yeah, the HTTPArchive run also doesn't show any sites triggering the >> usecounter. >> >> >>> >>> As of Apr 30, 2025, usage of the GPUAdapter isFallbackAdapter boolean >>> attribute reached a peak of 0.000338% of page loads according to the >>> ChromeStatus dashboard >>> https://chromestatus.com/metrics/feature/timeline/popularity/5460. >>> >> >> I'm seeing a significantly higher number in that URL (0.008%), but it >> seems to have a single data point.. Is the use counter recent? >> > > It was added Mar 28, 2025: > https://chromiumdash.appspot.com/commit/f0e55337ba45d45be08adb4cfbbd61860fa66d5e > OK, I suspect it may not be super stable.. > > > >> >> >>> Initial public proposal >>> >>> None >>> >>> TAG review >>> >>> None >>> >>> TAG review status >>> >>> Not applicable >>> >>> Risks >>> >>> Interoperability and Compatibility >>> >>> When WebGPU eventually launches in Safari and Firefox, websites will use >>> exclusively the GPUAdapterInfo isFallbackAdapter boolean attribute. >>> >>> We anticipate Safari and Firefox will soon support WebGPU, but won't >>> include the non-standard GPUAdapter isFallbackAdapter boolean attribute. >>> Therefore, the sooner Chromium implements the Deprecate and Remove process, >>> the less likely it is that content will work in Chromium but not in other >>> browsers. >>> >>> This removal has been approved by the GPU for the Web Community Group, >>> with representatives from Chrome, Firefox, and Safari. See minutes at >>> https://github.com/gpuweb/gpuweb/wiki/GPU-Web-2025-03-05#move-isfallbackadapter-into-gpuadapterinfo-4971 >>> and https://github.com/gpuweb/gpuweb/pull/5099#issuecomment-2737426600. >>> >>> Gecko: No signal - Mozilla members have requested during weekly >>> standardization meetings that we don't file standard positions anymore as >>> we reached Candidate Recommendation (CR) status. >>> https://github.com/mozilla/standards-positions/issues/1205 >>> >>> WebKit: No signal - >>> https://github.com/WebKit/standards-positions/issues/294#issuecomment-1877411933 >>> >>> Web developers: No signals >>> >>> Other signals: >>> >>> WebView application risks >>> >>> Does this intent deprecate or change behavior of existing APIs, such >>> that it has potentially high risk for Android WebView-based applications? >>> >>> None >>> >>> >>> Debuggability >>> >>> None >>> >>> Flag name on chrome://flags >>> >>> None >>> >>> Finch feature name >>> >>> WebGPUAdapterIsFallbackAdapter >>> >>> Rollout plan >>> >>> Will ship enabled for all users >>> >>> Non-finch justification >>> >>> None >>> >>> Requires code in //chrome? >>> >>> False >>> >>> Tracking bug >>> >>> https://issues.chromium.org/issues/409259074 >>> >>> Estimated milestones >>> >>> Shipping on desktop >>> >>> 138 >>> >>> Shipping on Android >>> >>> 138 >>> >>> >>> Link to entry on the Chrome Platform Status >>> >>> https://chromestatus.com/feature/4870179714236416 >>> >>> This intent message was generated by Chrome Platform Status >>> <https://chromestatus.com/>. >>> >>> -- >>> 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 visit >>> https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAPpwU5KaZ-bt0W3%3Dh6uzXDV6stkys3ed0qToB4d31r%2BEDXZmvA%40mail.gmail.com >>> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAPpwU5KaZ-bt0W3%3Dh6uzXDV6stkys3ed0qToB4d31r%2BEDXZmvA%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 visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAOmohSK8bjRMgRTs1CO5YbP9s3k5Yce4skyZxqE%3DOs0K%3D6P2cg%40mail.gmail.com.
