[Webkit-unassigned] [Bug 199402] [GTK] Crash in webkitWebViewBaseRenderHostFileDescriptor

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Thu Jul 18 04:10:44 PDT 2019


https://bugs.webkit.org/show_bug.cgi?id=199402

Michael Catanzaro <mcatanzaro at igalia.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
 Attachment #374372|review?                     |review+
              Flags|                            |

--- Comment #5 from Michael Catanzaro <mcatanzaro at igalia.com> ---
Comment on attachment 374372
  --> https://bugs.webkit.org/attachment.cgi?id=374372
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=374372&action=review

> Source/WebKit/UIProcess/gtk/AcceleratedBackingStoreWayland.cpp:76
> -std::unique_ptr<AcceleratedBackingStoreWayland> AcceleratedBackingStoreWayland::create(WebPageProxy& webPage)
> +bool AcceleratedBackingStoreWayland::checkRequirements()
>  {
>  #if USE(WPE_RENDERER)
>      if (!glImageTargetTexture2D) {
>          if (!wpe_fdo_initialize_for_egl_display(PlatformDisplay::sharedDisplay().eglDisplay()))
> -            return nullptr;
> +            return false;
>  
>          std::unique_ptr<WebCore::GLContext> eglContext = GLContext::createOffscreenContext();
>          if (!eglContext)
> -            return nullptr;
> +            return false;

Hmm, you sure this is a good idea? This does a lot of serious work under the guise of a checkRequirements function. I don't think a reasonable programmer calling AcceleratedBackingStore::checkRequirements would expect a function with this name to be setting up graphics resources like this. I think it's only a naming problem; you could call it AcceleratedBackingStore::setup or something like that, for example. (Although that too is an odd name for a static method....)

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20190718/8f84794b/attachment-0001.html>


More information about the webkit-unassigned mailing list