ba8cc57e0848818df689149c4eed99c4c0d7b5e9: Transform RegExp.exec to RegExp.test in native call IC, bug 645889. r=dmandelin
Brian Hackett <bhackett1024@gmail.com> - Wed, 30 Mar 2011 17:43:36 -0700 - rev 64556
Transform RegExp.exec to RegExp.test in native call IC, bug 645889. r=dmandelin
76d04b5e5e75345190149c1cfb015b668ccaae6f: Bug 645184: normalize id in addprop IC handler, r=dvander
David Mandelin <dmandelin@mozilla.com> - Wed, 30 Mar 2011 16:57:44 -0700 - rev 64555
Bug 645184: normalize id in addprop IC handler, r=dvander
aba14034d8197dbc28a106680c8cd3e88ba0d7f1: Fix atomsCompartment assertion. No Bug, r=billm
Gregor Wagner <anygregor@gmail.com> - Wed, 30 Mar 2011 14:02:21 -0700 - rev 64554
Fix atomsCompartment assertion. No Bug, r=billm
b55ca8a7c0ade1a42e8460f3156062e542daf5f1: Bug 645826: Include standard output and standard error in exception text when testing manifest conditions. r=dmandelin
Jim Blandy <jimb@mozilla.com> - Wed, 30 Mar 2011 12:49:47 -0700 - rev 64553
Bug 645826: Include standard output and standard error in exception text when testing manifest conditions. r=dmandelin
1e5ded29dd9f2c15c9fa2453e844832217417b28: Bug 643222: Don't assert that we never cached adds to call objects. r=jorendorff
Jim Blandy <jimb@mozilla.com> - Wed, 30 Mar 2011 12:48:24 -0700 - rev 64552
Bug 643222: Don't assert that we never cached adds to call objects. r=jorendorff This assertion should not have been added. Although most adds to call objects are done by DEFVAR operations, which don't create property cache entries, the test case shows a situation in which a SETNAME, which is cached, does the add. Since the object whose property should receive the value of an assignment's RHS is chosen before the RHS is evaluated, it is correct to put the value of the 'delete' expression on the call. It would be nice to ensure that this call object, which we know we are adding bindings to, had been recognized as extensible at compile time by the analysis added in bug 554955. However, we only set "extensible parents" flag on the bindings of calls whose *parents* are extensible, not the extensible calls themselves. So there isn't anything convenient to check here.
d10c089a6888f29f4e097990b49afb1a5c16396b: Bug 642274: Add loose equality to JSAPI. (r=Waldo)
Chris Leary <cdleary@mozilla.com> - Mon, 21 Mar 2011 10:04:43 -0700 - rev 64551
Bug 642274: Add loose equality to JSAPI. (r=Waldo)
8359361f8a7b64fe63ea20d9ebf6ffb770d42e7b: Expose public APIs for converting doubles to signed/unsigned int per ECMA spec (bug 631132, r=bzbarsky).
Andreas Gal <gal@mozilla.com> - Tue, 29 Mar 2011 19:19:15 -0700 - rev 64550
Expose public APIs for converting doubles to signed/unsigned int per ECMA spec (bug 631132, r=bzbarsky).
df211d75bddcfb9990841fd89353962ca6b41161: Bug 646304: Fix MSVC warning, r=bent
David Mandelin <dmandelin@mozilla.com> - Tue, 29 Mar 2011 19:05:12 -0700 - rev 64549
Bug 646304: Fix MSVC warning, r=bent
839c87b7e4bb175488843abfcbf1470ad82c6c7b: Bug 644045 - Handle Theora granulepos 'pframe' of 0 when decrementing granulepos. r=derf
Chris Pearce <chris@pearce.org.nz> - Fri, 01 Apr 2011 11:10:49 +1300 - rev 64548
Bug 644045 - Handle Theora granulepos 'pframe' of 0 when decrementing granulepos. r=derf
44d43f095a4f34ae8c4cb85bdb29baf9f6c9717d: Bug 641718 - Convert to using microseconds for internal video timestamps. r=kinetik
Chris Pearce <chris@pearce.org.nz> - Fri, 01 Apr 2011 11:10:49 +1300 - rev 64547
Bug 641718 - Convert to using microseconds for internal video timestamps. r=kinetik
56c1a338a41f4a52d335a917b29187fbd99cd0fe: Bug 640935 - Fix data race on global var vp8_build_intra_predictors_mby_ptr. r=cpearce
Timothy B. Terriberry <tterribe@vt.edu> - Fri, 01 Apr 2011 11:10:49 +1300 - rev 64546
Bug 640935 - Fix data race on global var vp8_build_intra_predictors_mby_ptr. r=cpearce
d6da3f26daea892f71c1246daaa6e4a8a4fd4b0e: Bug 646819 - Fix incorrect position calculation in nsWaveDecoder::DecodeAudioData causing OOM - r=kinetik
Chris Double <chris.double@double.co.nz> - Fri, 01 Apr 2011 03:49:07 +1300 - rev 64545
Bug 646819 - Fix incorrect position calculation in nsWaveDecoder::DecodeAudioData causing OOM - r=kinetik
d2a5747b10b5cb794e18e7d1b2030991e8f6f0a1: Bug 646766 - Make test_info_leak use MediaTestManager so it can be parallelized. r=roc
Chris Pearce <chris@pearce.org.nz> - Fri, 01 Apr 2011 11:10:48 +1300 - rev 64544
Bug 646766 - Make test_info_leak use MediaTestManager so it can be parallelized. r=roc
d13ba0674a663308d86da4422860905843985ea2: Bug 646682 - Fix range check in test_wave_data media tests - r=kinetik
Chris Double <chris.double@double.co.nz> - Thu, 31 Mar 2011 14:08:49 +1300 - rev 64543
Bug 646682 - Fix range check in test_wave_data media tests - r=kinetik
69a9aa30f2eff07d0dbf973fd72931a50b3dbb83: Bug 622072. Part 3: Support empty transactions for canvas layers. r=tnikkel
Robert O'Callahan <robert@ocallahan.org> - Mon, 28 Mar 2011 12:59:47 +1300 - rev 64542
Bug 622072. Part 3: Support empty transactions for canvas layers. r=tnikkel The basic idea is that whenever a layer transaction updates the window, we clear out the invalidation state for the canvas rendering context, using a DidTransactionCallback registered on the layer(s) for the canvas, which calls MakeContextClean. The DidTransactionCallbacks are directed to the user data attached to the Layer, which holds a strong reference to the canvas element. This ensures that the element lives as long as the layer. Layers are destroyed when the presentation is torn down (including if the frame is destroyed), so we can't have a leak here. The reference to the canvas element is only strong because the layer might briefly outlive the frame (the layer won't be destroyed until the next paint of the window). This patch moves responsibility for calling CanvasLayer::Updated and nsFrame::MarkLayersActive from the canvas context to nsHTMLCanvasElement::InvalidateFrame. We call Updated on the retained CanvasLayer, if there is one; any other CanvasLayers created for this canvas would only be used once, and have Updated called on them in BuildLayer when created.
d3481745c067a9df6c8446079db1aeb26d257b1a: Bug 622072. Part 2: Add DidTransactionCallback to CanvasLayer. r=bas, sr=joe
Robert O'Callahan <robert@ocallahan.org> - Mon, 28 Mar 2011 12:59:46 +1300 - rev 64541
Bug 622072. Part 2: Add DidTransactionCallback to CanvasLayer. r=bas, sr=joe
f1ea5b6264d0a150efe7275a526a1c567dca8f23: Bug 622072. Part 1: Remove rect parameter from Updated() and change implementations to defer updates to render time. r=bas
Robert O'Callahan <robert@ocallahan.org> - Mon, 28 Mar 2011 12:59:46 +1300 - rev 64540
Bug 622072. Part 1: Remove rect parameter from Updated() and change implementations to defer updates to render time. r=bas The rectangle parameter is currently not used --- all callers always pass the full canvas bounds. In the long term, we probably won't want this parameter since all implementations should be doing accelerated drawing direct to buffers with no intermediate copies, hence there will be no need to optimize the size of those copies. Plus, performance-sensitive testcases tend to paint most or all of the canvas on every frame anyway.
bec3f251f53a43a08a13d31f190cf2877bc99f23: Bug 635373. When resampling a ThebesLayer, ensure that D3D9 and OpenGL layers paint the entire layer as one quad to avoid unnecessary aliasing. r=bas
Robert O'Callahan <robert@ocallahan.org> - Fri, 01 Apr 2011 10:33:46 +1300 - rev 64539
Bug 635373. When resampling a ThebesLayer, ensure that D3D9 and OpenGL layers paint the entire layer as one quad to avoid unnecessary aliasing. r=bas
844579d342003efcb2baaf3ec4d5bf0d63cae619: Bug 635373. ThebesLayerOGL needs to make sure we only sample valid pixels too. r=mattwoodrow
Robert O'Callahan <robert@ocallahan.org> - Fri, 01 Apr 2011 10:33:46 +1300 - rev 64538
Bug 635373. ThebesLayerOGL needs to make sure we only sample valid pixels too. r=mattwoodrow
c686a9fb1b50ca9faf055b3edf79e48b7578ee6c: Bug 635373. Tests
Robert O'Callahan <robert@ocallahan.org> - Fri, 01 Apr 2011 10:33:46 +1300 - rev 64537
Bug 635373. Tests
(0) -30000 -10000 -3000 -1000 -300 -100 -50 -20 +20 +50 +100 +300 +1000 +3000 +10000 +30000 +100000 tip