[webkit-reviews] review granted: [Bug 225896] [WebXR] Attach IOSurface to WebXROpaqueFramebuffer : [Attachment 430116] Patch

[bugzilla-daemon at webkit.org]

Sam Weinig <sam at webkit.org> has granted Dean Jackson <dino at apple.com>'s request
for review:
Bug 225896: [WebXR] Attach IOSurface to WebXROpaqueFramebuffer
https://bugs.webkit.org/show_bug.cgi?id=225896

Attachment 430116: Patch

https://bugs.webkit.org/attachment.cgi?id=430116&action=review




--- Comment #6 from Sam Weinig <sam at webkit.org> ---
Comment on attachment 430116
  --> https://bugs.webkit.org/attachment.cgi?id=430116
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=430116&action=review

> Source/WebCore/Modules/webxr/WebXROpaqueFramebuffer.cpp:189
> +    if (m_ioSurfaceTextureHandle) {
> +	   auto gCGL = static_cast<GraphicsContextGLOpenGL*>(&gl);
> +	   gCGL->destroyPbufferAndDetachIOSurface(m_ioSurfaceTextureHandle);
> +    }

Is there a smart pointer we can use for this? This seems extremely easy to get
wrong.

> Source/WebCore/Modules/webxr/WebXROpaqueFramebuffer.cpp:248
> +    bool supportsPackedDepthStencil = true;

Seems like this should probably only be true is useDepthStencil is true? (all
the other #if sections check useDepthStencil here.

> Source/WebCore/Modules/webxr/WebXROpaqueFramebuffer.cpp:267
> +	   // Cap the maxiumum multisample count at 4. Any more than this is
likely overkill and will impact performance.

Is this observable and/or in the spec?