You mention 5 reviews, there's Security / Privacy / TAG but I'm not sure what the others are. We believe that the three reviews we know of are not applicable both because this feature doesn't impact them, and because any review would be a rehash of parts of the reviews for the original WebGPU release. Is there a way to mark this in the status entry?
On Wed, Oct 11, 2023 at 4:52 PM Chris Harrelson <chris...@chromium.org> wrote: > Please fill out the 5 other review categories (security, privacy, etc) in > chromestatus. Those will need to be started before the API owners can > review. > > On Thu, Oct 5, 2023 at 1:17 PM Ryan Harrison <rharri...@chromium.org> > wrote: > >> +cc cwallez, since he is the WebGPU lead, and can respond authoritatively >> if I am OOO >> >> Contact emailsrharri...@chromium.org >> >> ExplainerNone >> >> Specificationhttps://www.w3.org/TR/webgpu/#shader-f16 >> >> Summary >> >> Allows for the use of the half-precision floating-point type f16 in >> WebGPU shaders (WGSL). Developers can use the 'shader-f16' feature from the >> WebGPU spec and the 'f16' extension from the WGSL spec to access 16-bit >> floating point variables and APIs in their shaders. >> >> >> Blink componentInternals>GPU>Dawn >> <https://bugs.chromium.org/p/chromium/issues/list?q=component:Internals%3EGPU%3EDawn> >> >> TAG reviewNone >> >> TAG review statusNot applicable >> >> Risks >> >> >> Interoperability and Compatibility >> >> None >> >> >> *Gecko*: Positive Mozilla has already implemented this feature. >> >> *WebKit*: Positive (https://bugs.webkit.org/show_bug.cgi?id=254668) >> >> *Web developers*: Positive f16 support in general has been requested >> feature in multiple different web runtimes/frameworks, e.g. >> https://github.com/microsoft/onnxruntime/issues/9758 and >> http://tc39.es/proposal-float16array/ From the initial proposal, >> https://github.com/gpuweb/gpuweb/issues/658, there were multiple >> benchmarks/demoes that indicate a substantial performance benefit shown in >> demos/tests, ~25% if ALU bound, and upto 50% if memory bound. TF.js has >> been investigating using f16 in WebGPU, >> https://github.com/gpuweb/gpuweb/issues/658#issuecomment-606492757 and >> already has it available in the WebGL implementation TF.js has significant >> improvements from forcing using Half Float (the GL equivalent feature), >> https://github.com/tensorflow/tfjs/issues/1047 The Dawn team has >> interacted with multiple internal and external partners that have indicated >> that f16 support is an important feature for them to get the performance >> needed to adopt WebGPU >> >> *Other signals*: Intel has dedicated significant resources to help >> implement f16 in Dawn/Tint, and write proper testing coverage in the CTS. >> >> 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 at the moment, WebGPU currently does not ship on Android WebView. >> Parallel work is occuring to launch WebGPU on Android. >> >> >> Debuggability >> >> None >> >> >> Will this feature be supported on all six Blink platforms (Windows, Mac, >> Linux, Chrome OS, Android, and Android WebView)?No >> >> All platforms will eventually have support. Will immediately be available >> on Mac and Chrome OS, since those are platforms already support WebGPU. >> Windows support depends on internal changes to use DXC, which are in >> progress, so f16 will become availabe once that is finished. Linux and >> Android are planned to have WebGPU support in the future, so feature will >> become available when WebGPU does. >> >> >> Is this feature fully tested by web-platform-tests >> <https://chromium.googlesource.com/chromium/src/+/main/docs/testing/web_platform_tests.md> >> ?No >> >> WebGPU/WGSL have a conformance test suite (https://github.com/gpuweb/cts) >> that is regularly pulled into Chromium and part of the testing of Dawn/Tint >> in Chromium. Test coverage for execution of f16 operations has been >> implemented, https://github.com/gpuweb/cts/issues/1248. Additional >> testing is in development. >> >> >> Flag name on chrome://flagsNone >> >> Finch feature nameWebGPUExposeShader16 >> >> Requires code in //chrome?False >> >> Availability expectationFeature is available only in Chromium browsers >> for the near future, on the order of months. Other browsers intend to ship >> WebGPU and f16 support, but don't have specified timelines. >> >> Adoption expectationFeature is used by specific partner(s) to provide >> functionality within 12 months of launch in Chrome. >> >> Adoption planMultiple partners are actively developing products/features >> that will use f16 in WebGPU. >> >> Non-OSS dependencies >> >> Does the feature depend on any code or APIs outside the Chromium open >> source repository and its open-source dependencies to function? >> No >> >> Estimated milestones >> Shipping on desktop 120 >> >> Anticipated spec changes >> >> Open questions about a feature may be a source of future web compat or >> interop issues. Please list open issues (e.g. links to known github issues >> in the project for the feature specification) whose resolution may >> introduce web compat/interop risk (e.g., changing to naming or structure of >> the API in a non-backward-compatible way). >> None, f16 has been landed in the WebGPU and WGSL specs >> https://github.com/gpuweb/gpuweb/pull/2696 (There is some discussion >> about whether or not explicit enabling of extensions should be required. >> Currently it is, so If this changes existing shaders would be fine because >> the desire is to make enabling optional, not removing the mechanism >> entirely). >> >> Link to entry on the Chrome Platform Status >> https://chromestatus.com/feature/5180552617656320 >> >> 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 on the web visit >> https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAM4DYDaCcxwhdQb%2B%3DoZcAyd1-o3tU3qNu2Rbv5p0V95nGMeKQg%40mail.gmail.com >> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAM4DYDaCcxwhdQb%2B%3DoZcAyd1-o3tU3qNu2Rbv5p0V95nGMeKQg%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/CAGdfWNPxpUF5_x-QAkH4_nb1U1WUE9Xi54%3DN%3D%3DJZoYsNsxYwEw%40mail.gmail.com.
