Closed
Bug 1191334
Opened 9 years ago
Closed 9 years ago
High CPU usage on Nightly42.0a1 than Aurora41.0a2
Categories
(Core :: Audio/Video: Playback, defect)
Tracking
()
RESOLVED
DUPLICATE
of bug 1188131
People
(Reporter: alice0775, Assigned: jya)
Details
(Keywords: power, regression)
Attachments
(2 files)
+++ This bug was initially created as a clone of Bug #1190258 +++ Steps to reproduce 1. Open https://www.youtube.com/watch?v=eyU3bRy2x44 Actual Results: plugin-container.exe consumes 10-15% CPU (C2Q Q8300). And "Page fault delta" in taskmanager indicates 2000-2500. (Firefox.exe consumes 1-2% CPU) Expected Results: plugin-container.exe consumes around 5% CPU. And "Page fault delta" in taskmanager indicates 0-200. (Firefox.exe consumes 1-2% CPU) Regression window: https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=f8051d507461&tochange=de0a5cf8c4f9 Suspect: a465aecfd856 Jean-Yves Avenard — Bug 1171379: P1. Enable MediaSourceDemuxer by default. r=kentuckyfriedtakahe And I confirmed that setting media.mediasource.format-reader = false fixes the problem
Reporter | ||
Comment 1•9 years ago
|
||
Tested with: Nightly42.0a1 https://hg.mozilla.org/mozilla-central/rev/f3b757156f69 Mozilla/5.0 (Windows NT 6.1; WOW64; rv:42.0) Gecko/20100101 Firefox/42.0 ID:20150805030208 Aurora41.0a2 https://hg.mozilla.org/releases/mozilla-aurora/rev/2b17edd9c297 Mozilla/5.0 (Windows NT 6.1; WOW64; rv:41.0) Gecko/20100101 Firefox/41.0 ID:20150805004014
Assignee | ||
Updated•9 years ago
|
Assignee: nobody → jyavenard
Assignee | ||
Comment 2•9 years ago
|
||
I can't reproduce this here. Running under a profiler actually shows that the CPU usage is significantly lower (as far as MediaSource is concerned) than the old architecture. A drop in 20% CPU time in fact spent less in the mediasource+demuxing on average over a 2 minutes youtube playback (TrackBuffersManager+MediaSourceDemuxer vs TrackBuffer+MediaSourceReader). And the old MSE architecture the entire time is spent on the main thread while the new one is highly parallelized resulting in the UI being more responsive. Could it be related to bug 1188131, and what Alice is seeing is that it's now using software decoding. I don't see how that's possible. Matt, how can we tell if DXVA is in used or not for a particular video and have the user reports this info?
Flags: needinfo?(matt.woodrow)
Reporter | ||
Comment 3•9 years ago
|
||
Profiler log http://people.mozilla.org/~bgirard/cleopatra/#report=5965ed8c77b900e67ab70b449a617e4a7d58b2d2
Reporter | ||
Comment 4•9 years ago
|
||
Profiler log with media.mediasource.format-reader=false http://people.mozilla.org/~bgirard/cleopatra/#report=c50edeb0685f2b9476b09e6cfee4bbd6d7b6a267
Reporter | ||
Comment 5•9 years ago
|
||
Reporter | ||
Comment 6•9 years ago
|
||
Reporter | ||
Comment 7•9 years ago
|
||
Application Basics ------------------ Name: Firefox Version: 42.0a1 Build ID: 20150805030208 Update Channel: nightly User Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:42.0) Gecko/20100101 Firefox/42.0 Multiprocess Windows: 1/1 (default: true) Safe Mode: false Crash Reports for the Last 3 Days --------------------------------- All Crash Reports (including 1 pending crash in the given time range) Extensions ---------- Name: about:media Version: 1 Enabled: true ID: aboutmedia@gavinsharp.com Graphics -------- Adapter Description: AMD Radeon HD 6450 Adapter Drivers: aticfx64 aticfx64 aticfx64 aticfx32 aticfx32 aticfx32 atiumd64 atidxx64 atidxx64 atiumdag atidxx32 atidxx32 atiumdva atiumd6a atitmm64 Adapter RAM: 1024 Asynchronous Pan/Zoom: wheel input enabled ClearType Parameters: Gamma: 2200 Pixel Structure: R ClearType Level: 100 Enhanced Contrast: 300 Device ID: 0x6779 Direct2D Enabled: true DirectWrite Enabled: true (6.2.9200.17292) Driver Date: 6-22-2015 Driver Version: 15.200.1046.0 GPU #2 Active: false GPU Accelerated Windows: 1/1 Direct3D 11 (OMTC) Subsys ID: 23111787 Supports Hardware H264 Decoding: true Vendor ID: 0x1002 WebGL Renderer: Google Inc. -- ANGLE (AMD Radeon HD 6450 Direct3D11 vs_5_0 ps_5_0) windowLayerManagerRemote: true AzureCanvasBackend: direct2d 1.1 AzureContentBackend: direct2d 1.1 AzureFallbackCanvasBackend: cairo AzureSkiaAccelerated: 0 Important Modified Preferences ------------------------------ browser.cache.disk.capacity: 358400 browser.cache.disk.filesystem_reported: 1 browser.cache.disk.smart_size.first_run: false browser.cache.frecency_experiment: 3 browser.download.importedFromSqlite: true browser.places.smartBookmarksVersion: 7 browser.sessionstore.upgradeBackup.latestBuildID: 20150805030208 browser.startup.homepage_override.buildID: 20150805030208 browser.startup.homepage_override.mstone: 42.0a1 browser.tabs.drawInTitlebar: false browser.tabs.warnOnClose: false dom.apps.reset-permissions: true dom.mozApps.used: true extensions.lastAppVersion: 42.0a1 gfx.direct3d.last_used_feature_level_idx: 0 gfx.driver-init.appVersion: 42.0a1 gfx.driver-init.deviceID: 0x6779 gfx.driver-init.driverVersion: 15.200.1046.0 gfx.driver-init.feature-d2d: true gfx.driver-init.feature-d3d11: true gfx.driver-init.status: 2 media.gmp-eme-adobe.lastUpdate: 1438780512 media.gmp-eme-adobe.version: 12 media.gmp-gmpopenh264.lastUpdate: 1438780514 media.gmp-gmpopenh264.version: 1.4 media.gmp-manager.buildID: 20150805030208 media.gmp-manager.lastCheck: 1438812354 media.hardware-video-decoding.failed: false network.cookie.prefsMigrated: true network.predictor.cleaned-up: true places.database.lastMaintenance: 1438780850 places.history.expiration.transient_current_max_pages: 104858 plugin.disable_full_page_plugin_for_types: application/pdf plugin.importedState: true privacy.cpd.downloads: false privacy.cpd.history: false privacy.cpd.offlineApps: true privacy.sanitize.migrateFx3Prefs: true privacy.sanitize.timeSpan: 0 security.sandbox.content.tempDirSuffix: {36af7824-1ec4-4bd7-8ec9-5466a519be4c} storage.vacuum.last.index: 0 storage.vacuum.last.places.sqlite: 1438780850 Important Locked Preferences ---------------------------- JavaScript ---------- Incremental GC: true Accessibility ------------- Activated: false Prevent Accessibility: 0 Library Versions ---------------- NSPR Expected minimum version: 4.10.9 Beta Version in use: 4.10.9 Beta NSS Expected minimum version: 3.19.2 Basic ECC Version in use: 3.19.2 Basic ECC NSSSMIME Expected minimum version: 3.19.2 Basic ECC Version in use: 3.19.2 Basic ECC NSSSSL Expected minimum version: 3.19.2 Basic ECC Version in use: 3.19.2 Basic ECC NSSUTIL Expected minimum version: 3.19.2 Version in use: 3.19.2 Experimental Features ---------------------
Assignee | ||
Comment 8•9 years ago
|
||
Comment on attachment 8644250 [details]
Screenshot2, what I see CPU usage with media.mediasource.format-reader=false
156% in the plugin container is almost certainly due to software decoding...
What resolution was the video in 1080P ?
Reporter | ||
Comment 9•9 years ago
|
||
Right click video stat: Video ID: eyU3bRy2x44 Dimensions: 640 x 360 Resolution: 640 x 360@30 Volume: 100% Stream Host: r1---sn-3pm7sn7z Stream Type: https CPN: -CnjSvW8B26TJnFO Mime Type: video/mp4; codecs="avc1.4d401e" DASH: yes (134/140)
Assignee | ||
Comment 10•9 years ago
|
||
Okay, found the issue, it is indeed bug 1188131. The hardware layer isn't initialized and as such HW acceleration is never active. The reason is that MediaDecoderReader::InitLayersBackendType() is called way too early, before we know that the media is video or audio only.
Assignee | ||
Updated•9 years ago
|
Assignee | ||
Comment 11•9 years ago
|
||
Alice, let me know if the build from https://treeherder.mozilla.org/#/jobs?repo=try&revision=b37aaa47a155 you see any improvement. Was watching a 4K video and my debug build on windows 8.1 went from 74% to 12% (nvidia quadro card). An opt build will use much less.
Reporter | ||
Comment 12•9 years ago
|
||
(In reply to Jean-Yves Avenard [:jya] from comment #11) > Alice, let me know if the build from > https://treeherder.mozilla.org/#/jobs?repo=try&revision=b37aaa47a155 you see > any improvement. > Yes, it improved. CPU usage are less than 5% (360p). https://hg.mozilla.org/try/rev/b37aaa47a155 Mozilla/5.0 (Windows NT 6.1; WOW64; rv:42.0) Gecko/20100101 Firefox/42.0 ID:20150806042142
Assignee | ||
Comment 13•9 years ago
|
||
awesome. I'm going to mark this bug as duplicate of 1188131. Thanks again for your outstanding bug reports. Always easy to reproduce thanks to the amazingly detailed STR; all informations always there and prompt to answer.
Status: NEW → RESOLVED
Closed: 9 years ago
Flags: needinfo?(matt.woodrow)
Resolution: --- → DUPLICATE
Comment 14•9 years ago
|
||
Not tracked, flag cleared, as this is a duplicate of bug 1188131.
tracking-firefox42:
? → ---
Updated•9 years ago
|
You need to log in
before you can comment on or make changes to this bug.
Description
•