1017425 - Consider making window.performance StoreInSlot

Status: RESOLVED FIXED

(Core :: DOM: Core & HTML, defect)

Description

[Boris Zbarsky [:bzbarsky]]

To make performance.now() faster.

Comment 1

[Boris Zbarsky [:bzbarsky]]

Attachment: Fix, for when we can actually land this

Comment 2

[Peter Van der Beken [:peterv]]

Comment on attachment 8430541 [details] [diff] [review]
Fix, for when we can actually land this

Review of attachment 8430541 [details] [diff] [review]:
-----------------------------------------------------------------

::: dom/base/nsGlobalWindow.cpp
@@ +2346,5 @@
>    // document.
>    mDoc = aDocument;
>    if (IsInnerWindow() && IsDOMBinding()) {
>      WindowBinding::ClearCachedDocumentValue(cx, this);
> +    WindowBinding::ClearCachedPerformanceValue(cx, this);

Might be worth extracting this to a nsGlobalWindow::ClearSlots(...)?

Comment 3

[Boris Zbarsky [:bzbarsky]]

> Might be worth extracting this to a nsGlobalWindow::ClearSlots(...)?

How about ClearDocumentDependentSlots()?

Comment 4

[Peter Van der Beken [:peterv]]

Sure.

Comment 5

[Boris Zbarsky [:bzbarsky]]

Olli, do this bug we need to make the "performance" getter non-throwing.  That basically means backing out bug 1025078 and replacing it with the changes in this bug (to return null and null-check in console code).  Any objections?

Comment 6

[Olli Pettay [:smaug][[e-mail]]]

Returning null in those very odd cases sounds ok to me.

Comment 7

[Boris Zbarsky [:bzbarsky]]

Attachment: Patch updated to review comments and tip of tree

Comment 8

[Boris Zbarsky [:bzbarsky]]

I filed bug 1063879 on making this faster on Worker too.

Comment 9

[Boris Zbarsky [:bzbarsky]]

Attachment: Patch merged to tip again

Comment 10

[Boris Zbarsky [:bzbarsky]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/fc2a4c5ac088

Comment 11

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/mozilla-central/rev/fc2a4c5ac088