[webkit-reviews] review granted: [Bug 219586] [Concurrent Display Lists] GPU process should not immediately sleep after reading all available display list items : [Attachment 417008] Patch

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Tue Jan 5 09:04:42 PST 2021


Chris Dumez <cdumez at apple.com> has granted Wenson Hsieh
<wenson_hsieh at apple.com>'s request for review:
Bug 219586: [Concurrent Display Lists] GPU process should not immediately sleep
after reading all available display list items
https://bugs.webkit.org/show_bug.cgi?id=219586

Attachment 417008: Patch

https://bugs.webkit.org/attachment.cgi?id=417008&action=review




--- Comment #14 from Chris Dumez <cdumez at apple.com> ---
Comment on attachment 417008
  --> https://bugs.webkit.org/attachment.cgi?id=417008
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=417008&action=review

r=me with a few changes.

> Source/WebKit/GPUProcess/graphics/RemoteRenderingBackend.cpp:234
> +		   RELEASE_ASSERT_NOT_REACHED();

I don't think we should crash the GPU Process here, at least not in release. It
is never OK for a WebProcess to send bad IPC to the GPU Process and cause it to
crash (since the GPUProcess is shared with other processes).

> Source/WebKit/GPUProcess/graphics/RemoteRenderingBackend.cpp:241
> +		   RELEASE_ASSERT_NOT_REACHED();

ditto.

> Source/WebKit/GPUProcess/graphics/RemoteRenderingBackend.cpp:249
> +		   RELEASE_ASSERT_NOT_REACHED();

ditto.

> Source/WebKit/GPUProcess/graphics/RemoteRenderingBackend.h:45
> +#include <wtf/cocoa/MachSemaphore.h>

Seems like this could be forward-declared.

> Source/WebKit/Shared/SharedDisplayListHandle.h:59
> +	   NotWaiting = 0,

= 0 seems redundant


More information about the webkit-reviews mailing list