b35d742f3d6b: Bug 1374774: P10. Attempt to reduce SPS change false positives. r=gerald a=jcristau
Jean-Yves Avenard <jyavenard@mozilla.com> - Fri, 23 Jun 2017 03:01:57 +0200 - rev 412029
Push 7524 by jyavenard@mozilla.com at Thu, 20 Jul 2017 12:54:10 +0000
Bug 1374774: P10. Attempt to reduce SPS change false positives. r=gerald a=jcristau We now compare the decoded data rather than the raw data, otherwise as seen in video from bug 1372766, we keep draining the decoder. For some reasons the SPS NALs only differ by 1 byte at a time. MozReview-Commit-ID: LdXinUZHjD4
efc2ceb403ae: Bug 1374774: P9. Only look for SPS changes in stream format change detection. r=gerald a=jcristau
Jean-Yves Avenard <jyavenard@mozilla.com> - Thu, 22 Jun 2017 19:46:27 +0200 - rev 412028
Push 7524 by jyavenard@mozilla.com at Thu, 20 Jul 2017 12:54:10 +0000
Bug 1374774: P9. Only look for SPS changes in stream format change detection. r=gerald a=jcristau All decoders appear capable of handling content change when just new PPS appears. So we restrict the test to SPS changes. MozReview-Commit-ID: LPSfMaTIj6C
66bb2be5e144: Bug 1374774: P8. Use new SPSNALIterator class. r=gerald a=jcristau
Jean-Yves Avenard <jyavenard@mozilla.com> - Fri, 23 Jun 2017 02:08:41 +0200 - rev 412027
Push 7524 by jyavenard@mozilla.com at Thu, 20 Jul 2017 12:54:10 +0000
Bug 1374774: P8. Use new SPSNALIterator class. r=gerald a=jcristau MozReview-Commit-ID: 1nrvInxgugz
d7e7edd0f8ed: Bug 1374774: P7. Add SPSNAL and SPSNALIterator classes. r=gerald
Jean-Yves Avenard <jyavenard@mozilla.com> - Fri, 23 Jun 2017 01:51:42 +0200 - rev 412026
Push 7524 by jyavenard@mozilla.com at Thu, 20 Jul 2017 12:54:10 +0000
Bug 1374774: P7. Add SPSNAL and SPSNALIterator classes. r=gerald We will use them to simplify the parsing of the extradata. MozReview-Commit-ID: 5M5uGXAkkFb
babdaefa1a98: Bug 1374774: P6. Fix ByteReader constness. r=gerald
Jean-Yves Avenard <jyavenard@mozilla.com> - Thu, 22 Jun 2017 22:50:08 +0200 - rev 412025
Push 7524 by jyavenard@mozilla.com at Thu, 20 Jul 2017 12:54:10 +0000
Bug 1374774: P6. Fix ByteReader constness. r=gerald MozReview-Commit-ID: 8bQA189OOVF
aa9660e02f66: Bug 1374774: P5. Ignore duplicated SPS when reading NALs in-band. r=gerald a=jcristau
Jean-Yves Avenard <jyavenard@mozilla.com> - Thu, 22 Jun 2017 19:41:33 +0200 - rev 412024
Push 7524 by jyavenard@mozilla.com at Thu, 20 Jul 2017 12:54:10 +0000
Bug 1374774: P5. Ignore duplicated SPS when reading NALs in-band. r=gerald a=jcristau With some H264 streams, we find that the SPS/PPS NALs are duplicated on the stream. This caused us to treat it as if the content changed due to the discrepency between the extradata found in the MP4 metadata and what found inband. When scanning for SPS NALs, we now attempt to detect duplicates, and if so ignore them. MozReview-Commit-ID: D8OVOXSwEkY
00bd2e527ae0: Bug 1374774: P4. Correct H264 telemetry reporting. r=gerald a=jcristau
Jean-Yves Avenard <jyavenard@mozilla.com> - Thu, 22 Jun 2017 14:52:07 +0200 - rev 412023
Push 7524 by jyavenard@mozilla.com at Thu, 20 Jul 2017 12:54:10 +0000
Bug 1374774: P4. Correct H264 telemetry reporting. r=gerald a=jcristau The code couldn't have worked, mNeedSPSForTelemetry can only be set if we didn't get an H264 extradata in the MP4 metadata. So any calls to HasSPS(sample->mExtraData) would have always returned false. MozReview-Commit-ID: 3PcRXJHBJyC
ffcb435fd7b3: Bug 1374774: P3. Remove HasSPS method. r=gerald a=jcristau
Jean-Yves Avenard <jyavenard@mozilla.com> - Thu, 22 Jun 2017 14:44:21 +0200 - rev 412022
Push 7524 by jyavenard@mozilla.com at Thu, 20 Jul 2017 12:54:10 +0000
Bug 1374774: P3. Remove HasSPS method. r=gerald a=jcristau It was only used in one spot, and incorrectly at that. MozReview-Commit-ID: EWkkrAlYT7W
01f5e7fc87f5: Bug 1374774: P2. Add comment for HasSPS methods. r=gerald a=jcristau
Jean-Yves Avenard <jyavenard@mozilla.com> - Thu, 22 Jun 2017 14:42:46 +0200 - rev 412021
Push 7524 by jyavenard@mozilla.com at Thu, 20 Jul 2017 12:54:10 +0000
Bug 1374774: P2. Add comment for HasSPS methods. r=gerald a=jcristau MozReview-Commit-ID: Irvcx5XnPyp
c079a04f1046: Bug 1374774: P1. Move H264 methods into H264 code. r=gerald a=jcristau
Jean-Yves Avenard <jyavenard@mozilla.com> - Thu, 22 Jun 2017 14:36:11 +0200 - rev 412020
Push 7524 by jyavenard@mozilla.com at Thu, 20 Jul 2017 12:54:10 +0000
Bug 1374774: P1. Move H264 methods into H264 code. r=gerald a=jcristau HasSPS, ExtractExtraData and CompareExtraData have nothing to do with the handling of annex B format. They are raw H264 related methods. It will also prevent in the following change to have cycling references between two headers. MozReview-Commit-ID: FCs5aU4GcTU
30e0585ef763: Bug 1379762 - Don't call MediaFeaturesChanged if our override device pixel ratio is set to its current value. r=dbaron, a=jcristau
Boris Zbarsky <bzbarsky@mit.edu> - Mon, 17 Jul 2017 23:21:41 -0400 - rev 412019
Push 7523 by ryanvm@gmail.com at Wed, 19 Jul 2017 23:02:11 +0000
Bug 1379762 - Don't call MediaFeaturesChanged if our override device pixel ratio is set to its current value. r=dbaron, a=jcristau
28488519e000: Bug 1374556 - Use Android-like signal handlers instead of SIG_DFL. r=glandium, a=jcristau
John Lin <jolin@mozilla.com> - Mon, 10 Jul 2017 14:44:50 +0800 - rev 412018
Push 7523 by ryanvm@gmail.com at Wed, 19 Jul 2017 23:02:11 +0000
Bug 1374556 - Use Android-like signal handlers instead of SIG_DFL. r=glandium, a=jcristau For some reason the remote decoder process will crash on some x86/Android M devices when SIGSEGV is handled by SIG_DFL. To avoid this issue and continue suppressing crash dialog, capture the signals with Android like handlers and strip debuggerd related code. MozReview-Commit-ID: 3diOGc3OInD
ecafc69c34da: Bug 1367226 - Remove unused pref "plugins.navigator_hide_disabled_flash". r=bsmedberg, a=jcristau
Chris Peterson <cpeterson@mozilla.com> - Thu, 13 Jul 2017 23:18:19 -0700 - rev 412017
Push 7523 by ryanvm@gmail.com at Wed, 19 Jul 2017 23:02:11 +0000
Bug 1367226 - Remove unused pref "plugins.navigator_hide_disabled_flash". r=bsmedberg, a=jcristau MozReview-Commit-ID: Bbkh0p1e3Co
b3eff505c8d1: Bug 1381438 - Fix boxDouble to not clobber the source register on x86 CPUs without SSE 4.1. r=evilpie, a=jcristau
Jan de Mooij <jdemooij@mozilla.com> - Mon, 17 Jul 2017 16:29:52 +0200 - rev 412016
Push 7523 by ryanvm@gmail.com at Wed, 19 Jul 2017 23:02:11 +0000
Bug 1381438 - Fix boxDouble to not clobber the source register on x86 CPUs without SSE 4.1. r=evilpie, a=jcristau
84b80aceba4c: Bug 1312623 - Don't attempt to shutdown a thread in the dtor of a GraphDriver, since it might be called during stable state. r=smaug, a=jcristau
Paul Adenot <paul@paul.cx> - Tue, 04 Jul 2017 09:21:23 +0200 - rev 412015
Push 7523 by ryanvm@gmail.com at Wed, 19 Jul 2017 23:02:11 +0000
Bug 1312623 - Don't attempt to shutdown a thread in the dtor of a GraphDriver, since it might be called during stable state. r=smaug, a=jcristau MozReview-Commit-ID: HA67HQQrfow
a05565c8035c: Bug 1372824 - Work around race conditions on startup enabling e10s. r=Felipe, a=jcristau
Blake Kaplan <mrbkap@gmail.com> - Thu, 13 Jul 2017 12:16:42 -0700 - rev 412014
Push 7523 by ryanvm@gmail.com at Wed, 19 Jul 2017 23:02:11 +0000
Bug 1372824 - Work around race conditions on startup enabling e10s. r=Felipe, a=jcristau There are two race conditions that happen on startup because we read an install.rdf with a DOM parser before addons have a chance to run. The current code caches the first value that it computes, meaning that if it runs before e10srollout does, it will cache the default values for its channel (off on Beta and Release). This patch fixes the races by never caching the value of the MultiprocessBlockPolicy (which is fine, it's only ever called in two places) and by letting e10srollout tell BrowserTabsRemote to update its cached value. MozReview-Commit-ID: HsUIrPzqwl0
20b0f0b4a612: Bug 1380909 - Try to solve crash: move GetElement out of lambda. r=michal, a=jcristau
Junior Hsu <juhsu@mozilla.com> - Fri, 14 Jul 2017 00:13:00 -0400 - rev 412013
Push 7523 by ryanvm@gmail.com at Wed, 19 Jul 2017 23:02:11 +0000
Bug 1380909 - Try to solve crash: move GetElement out of lambda. r=michal, a=jcristau
64f34d964fa3: Bug 1327947 - Remove id's from about:preferences#containers as it breaks sorting for applications. r=jaws, a=jcristau
Jonathan Kingston <jkt@mozilla.com> - Wed, 05 Jul 2017 17:45:27 -0700 - rev 412012
Push 7523 by ryanvm@gmail.com at Wed, 19 Jul 2017 23:02:11 +0000
Bug 1327947 - Remove id's from about:preferences#containers as it breaks sorting for applications. r=jaws, a=jcristau MozReview-Commit-ID: H39D5kT4lc5
fe847bf98f6a: Bug 1379444 - Use generic cmap-based font fallback if platform-specific code fails to find a usable font. r=lsalzman, a=jcristau
Jonathan Kew <jkew@mozilla.com> - Thu, 13 Jul 2017 09:40:33 +0100 - rev 412011
Push 7523 by ryanvm@gmail.com at Wed, 19 Jul 2017 23:02:11 +0000
Bug 1379444 - Use generic cmap-based font fallback if platform-specific code fails to find a usable font. r=lsalzman, a=jcristau
cbaff692f6fe: Bug 1378136 - Redirect user to the general pane for search settings on old preferences page. r=jaws, a=jcristau
Timothy Guan-tin Chien <timdream@gmail.com> - Mon, 10 Jul 2017 17:01:41 +0800 - rev 412010
Push 7523 by ryanvm@gmail.com at Wed, 19 Jul 2017 23:02:11 +0000
Bug 1378136 - Redirect user to the general pane for search settings on old preferences page. r=jaws, a=jcristau MozReview-Commit-ID: Akb4yDPhWTF
(0) -300000 -100000 -30000 -10000 -3000 -1000 -300 -100 -50 -20 +20 +50 +100 +300 +1000 +3000 +10000 +30000 +100000 tip