I'm seeing vine videos in my twitter stream get real jittery while I'm panning. The jitter only happens on the first play cycle though, not once it repeats.

Application Basics ------------------ Name: Firefox Version: 48.0a1 Build ID: 20160328030215 Update Channel: nightly User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:48.0) Gecko/20100101 Firefox/48.0 OS: Windows_NT 10.0 x86-64 Multiprocess Windows: 2/2 (Enabled by user) Safe Mode: false Crash Reports for the Last 3 Days --------------------------------- All Crash Reports (including 12 pending crashes in the given time range) Extensions ---------- Name: activity-streams-experiment Version: 1.0.4 Enabled: true ID: @activity-streams Name: ADB Helper Version: 0.8.7 Enabled: true ID: adbhelper@mozilla.org Name: BugzillaJS Version: 3.5.0 Enabled: true ID: jid0-NgMDcEu2B88AbzZ6ulHodW9sJzA@jetpack Name: Firefox Hello Beta Version: 1.2.2 Enabled: true ID: loop@mozilla.org Name: Multi-process staged rollout Version: 1.0 Enabled: true ID: e10srollout@mozilla.org Name: Pocket Version: 1.0 Enabled: true ID: firefox@getpocket.com Name: Valence Version: 0.3.5 Enabled: true ID: fxdevtools-adapters@mozilla.org Graphics -------- Adapter Description: Intel(R) HD Graphics 530 Adapter Description (GPU #2): NVIDIA GeForce GTX 960M Adapter Drivers: igdumdim64 igd10iumd64 igd10iumd64 igd12umd64 igdumdim32 igd10iumd32 igd10iumd32 igd12umd32 Adapter Drivers (GPU #2): nvd3dumx,nvwgf2umx,nvwgf2umx,nvwgf2umx nvd3dum,nvwgf2um,nvwgf2um,nvwgf2um Adapter RAM: Unknown Adapter RAM (GPU #2): 2048 Asynchronous Pan/Zoom: wheel input enabled; touch input enabled Device ID: 0x191b Device ID (GPU #2): 0x139b Direct2D Enabled: true DirectWrite Enabled: true (10.0.10586.0) Driver Date: 11-20-2015 Driver Date (GPU #2): 3-7-2016 Driver Version: 20.19.15.4331 Driver Version (GPU #2): 10.18.13.6451 GPU #2 Active: false GPU Accelerated Windows: 2/2 Direct3D 11 (OMTC) Subsys ID: 06e41028 Subsys ID (GPU #2): 06e41028 Supports Hardware H264 Decoding: Yes; Using D3D11 API Vendor ID: 0x8086 Vendor ID (GPU #2): 0x10de WebGL Renderer: Google Inc. -- ANGLE (Intel(R) HD Graphics 530 Direct3D11 vs_5_0 ps_5_0) windowLayerManagerRemote: true AzureCanvasAccelerated: 0 AzureCanvasBackend: direct2d 1.1 AzureContentBackend: direct2d 1.1 AzureFallbackCanvasBackend: cairo (#0): CP+[GFX1-]: Failed to allocate a surface due to invalid size (CDT) Size(0,0) Important Modified Preferences ------------------------------ accessibility.lastLoadDate: 1451314545 accessibility.loadedInLastSession: true accessibility.typeaheadfind.flashBar: 0 browser.cache.disk.capacity: 358400 browser.cache.disk.filesystem_reported: 1 browser.cache.disk.hashstats_reported: 1 browser.cache.disk.smart_size.first_run: false browser.cache.disk.smart_size.use_old_max: false browser.cache.frecency_experiment: 4 browser.download.importedFromSqlite: true browser.places.smartBookmarksVersion: 7 browser.sessionstore.upgradeBackup.latestBuildID: 20160328030215 browser.startup.homepage_override.buildID: 20160328030215 browser.startup.homepage_override.mstone: 48.0a1 browser.tabs.crashReporting.email: blassey@mozilla.com browser.tabs.crashReporting.emailMe: true browser.tabs.crashReporting.includeURL: true browser.tabs.remote.autostart: true browser.tabs.remote.disabled-for-a11y: true browser.tabs.remote.force-enable: true browser.urlbar.userMadeSearchSuggestionsChoice: true dom.apps.reset-permissions: true dom.ipc.processCount: 10 dom.mozApps.used: true dom.push.userAgentID: 3355fc5ced8a460e884f2007cc4d1cae extensions.lastAppVersion: 48.0a1 font.internaluseonly.changed: true gfx.crash-guard.d3d11layers.appVersion: 44.0 gfx.crash-guard.d3d11layers.deviceID: 0x191b gfx.crash-guard.d3d11layers.driverVersion: 20.19.15.4331 gfx.crash-guard.d3d11layers.feature-d2d: true gfx.crash-guard.d3d11layers.feature-d3d11: true gfx.crash-guard.status.d3d11layers: 2 gfx.crash-guard.status.d3d9video: 2 gfx.direct3d.last_used_feature_level_idx: 0 gfx.driver-init.appVersion: 42.0 gfx.driver-init.deviceID: 0x191b gfx.driver-init.driverVersion: 20.19.15.4331 gfx.driver-init.feature-d2d: true gfx.driver-init.feature-d3d11: true gfx.driver-init.status: 2 keyword.enabled: false media.benchmark.vp9.fps: 392 media.gmp-eme-adobe.abi: x86_64-msvc-x64 media.gmp-eme-adobe.lastUpdate: 1456286240 media.gmp-eme-adobe.version: 17 media.gmp-gmpopenh264.abi: x86_64-msvc-x64 media.gmp-gmpopenh264.lastUpdate: 1453999021 media.gmp-gmpopenh264.version: 1.5.3 media.gmp-manager.buildID: 20160328030215 media.gmp-manager.lastCheck: 1459440777 media.hardware-video-decoding.failed: false network.cookie.prefsMigrated: true network.predictor.cleaned-up: true places.database.lastMaintenance: 1459368213 places.history.expiration.transient_current_max_pages: 104858 plugin.disable_full_page_plugin_for_types: application/pdf plugin.importedState: true print.printer_Brother_MFC-J485DW_Printer.print_bgcolor: false print.printer_Brother_MFC-J485DW_Printer.print_bgimages: false print.printer_Brother_MFC-J485DW_Printer.print_duplex: -437918235 print.printer_Brother_MFC-J485DW_Printer.print_edge_bottom: 0 print.printer_Brother_MFC-J485DW_Printer.print_edge_left: 0 print.printer_Brother_MFC-J485DW_Printer.print_edge_right: 0 print.printer_Brother_MFC-J485DW_Printer.print_edge_top: 0 print.printer_Brother_MFC-J485DW_Printer.print_evenpages: true print.printer_Brother_MFC-J485DW_Printer.print_in_color: true print.printer_Brother_MFC-J485DW_Printer.print_margin_bottom: 0.5 print.printer_Brother_MFC-J485DW_Printer.print_margin_left: 0.5 print.printer_Brother_MFC-J485DW_Printer.print_margin_right: 0.5 print.printer_Brother_MFC-J485DW_Printer.print_margin_top: 0.5 print.printer_Brother_MFC-J485DW_Printer.print_oddpages: true print.printer_Brother_MFC-J485DW_Printer.print_orientation: 0 print.printer_Brother_MFC-J485DW_Printer.print_page_delay: 50 print.printer_Brother_MFC-J485DW_Printer.print_paper_data: 1 print.printer_Brother_MFC-J485DW_Printer.print_paper_height: -1.00 print.printer_Brother_MFC-J485DW_Printer.print_paper_name: print.printer_Brother_MFC-J485DW_Printer.print_paper_size_unit: 1 print.printer_Brother_MFC-J485DW_Printer.print_paper_width: -1.00 print.printer_Brother_MFC-J485DW_Printer.print_resolution: 600 print.printer_Brother_MFC-J485DW_Printer.print_reversed: false print.printer_Brother_MFC-J485DW_Printer.print_scaling: 1.00 print.printer_Brother_MFC-J485DW_Printer.print_shrink_to_fit: true print.printer_Brother_MFC-J485DW_Printer.print_to_file: false print.printer_Brother_MFC-J485DW_Printer.print_unwriteable_margin_bottom: 0 print.printer_Brother_MFC-J485DW_Printer.print_unwriteable_margin_left: 0 print.printer_Brother_MFC-J485DW_Printer.print_unwriteable_margin_right: 0 print.printer_Brother_MFC-J485DW_Printer.print_unwriteable_margin_top: 0 print.printer_HP732A35_(HP_Photosmart_D110_series).print_bgcolor: false print.printer_HP732A35_(HP_Photosmart_D110_series).print_bgimages: false print.printer_HP732A35_(HP_Photosmart_D110_series).print_duplex: -437918235 print.printer_HP732A35_(HP_Photosmart_D110_series).print_edge_bottom: 0 print.printer_HP732A35_(HP_Photosmart_D110_series).print_edge_left: 0 print.printer_HP732A35_(HP_Photosmart_D110_series).print_edge_right: 0 print.printer_HP732A35_(HP_Photosmart_D110_series).print_edge_top: 0 print.printer_HP732A35_(HP_Photosmart_D110_series).print_evenpages: true print.printer_HP732A35_(HP_Photosmart_D110_series).print_footercenter: print.printer_HP732A35_(HP_Photosmart_D110_series).print_footerleft: &PT print.printer_HP732A35_(HP_Photosmart_D110_series).print_footerright: &D print.printer_HP732A35_(HP_Photosmart_D110_series).print_headercenter: print.printer_HP732A35_(HP_Photosmart_D110_series).print_headerleft: &T print.printer_HP732A35_(HP_Photosmart_D110_series).print_headerright: &U print.printer_HP732A35_(HP_Photosmart_D110_series).print_in_color: true print.printer_HP732A35_(HP_Photosmart_D110_series).print_margin_bottom: 0.5 print.printer_HP732A35_(HP_Photosmart_D110_series).print_margin_left: 0.5 print.printer_HP732A35_(HP_Photosmart_D110_series).print_margin_right: 0.5 print.printer_HP732A35_(HP_Photosmart_D110_series).print_margin_top: 0.5 print.printer_HP732A35_(HP_Photosmart_D110_series).print_oddpages: true print.printer_HP732A35_(HP_Photosmart_D110_series).print_orientation: 0 print.printer_HP732A35_(HP_Photosmart_D110_series).print_page_delay: 50 print.printer_HP732A35_(HP_Photosmart_D110_series).print_paper_data: 1 print.printer_HP732A35_(HP_Photosmart_D110_series).print_paper_height: -1.00 print.printer_HP732A35_(HP_Photosmart_D110_series).print_paper_name: print.printer_HP732A35_(HP_Photosmart_D110_series).print_paper_size_type: 0 print.printer_HP732A35_(HP_Photosmart_D110_series).print_paper_size_unit: 1 print.printer_HP732A35_(HP_Photosmart_D110_series).print_paper_width: -1.00 print.printer_HP732A35_(HP_Photosmart_D110_series).print_resolution: 600 print.printer_HP732A35_(HP_Photosmart_D110_series).print_reversed: false print.printer_HP732A35_(HP_Photosmart_D110_series).print_scaling: 1.00 print.printer_HP732A35_(HP_Photosmart_D110_series).print_shrink_to_fit: true print.printer_HP732A35_(HP_Photosmart_D110_series).print_to_file: false print.printer_HP732A35_(HP_Photosmart_D110_series).print_unwriteable_margin_bottom: 0 print.printer_HP732A35_(HP_Photosmart_D110_series).print_unwriteable_margin_left: 0 print.printer_HP732A35_(HP_Photosmart_D110_series).print_unwriteable_margin_right: 0 print.printer_HP732A35_(HP_Photosmart_D110_series).print_unwriteable_margin_top: 0 print.printer_Microsoft_Print_to_PDF.print_bgcolor: false print.printer_Microsoft_Print_to_PDF.print_bgimages: false print.printer_Microsoft_Print_to_PDF.print_duplex: -437918235 print.printer_Microsoft_Print_to_PDF.print_edge_bottom: 0 print.printer_Microsoft_Print_to_PDF.print_edge_left: 0 print.printer_Microsoft_Print_to_PDF.print_edge_right: 0 print.printer_Microsoft_Print_to_PDF.print_edge_top: 0 print.printer_Microsoft_Print_to_PDF.print_evenpages: true print.printer_Microsoft_Print_to_PDF.print_footercenter: print.printer_Microsoft_Print_to_PDF.print_footerleft: &PT print.printer_Microsoft_Print_to_PDF.print_footerright: &D print.printer_Microsoft_Print_to_PDF.print_headercenter: print.printer_Microsoft_Print_to_PDF.print_headerleft: &T print.printer_Microsoft_Print_to_PDF.print_headerright: &U print.printer_Microsoft_Print_to_PDF.print_in_color: true print.printer_Microsoft_Print_to_PDF.print_margin_bottom: 0.5 print.printer_Microsoft_Print_to_PDF.print_margin_left: 0.5 print.printer_Microsoft_Print_to_PDF.print_margin_right: 0.5 print.printer_Microsoft_Print_to_PDF.print_margin_top: 0.5 print.printer_Microsoft_Print_to_PDF.print_oddpages: true print.printer_Microsoft_Print_to_PDF.print_orientation: 0 print.printer_Microsoft_Print_to_PDF.print_page_delay: 50 print.printer_Microsoft_Print_to_PDF.print_paper_data: 1 print.printer_Microsoft_Print_to_PDF.print_paper_height: -1.00 print.printer_Microsoft_Print_to_PDF.print_paper_name: print.printer_Microsoft_Print_to_PDF.print_paper_size_type: 0 print.printer_Microsoft_Print_to_PDF.print_paper_size_unit: 1 print.printer_Microsoft_Print_to_PDF.print_paper_width: -1.00 print.printer_Microsoft_Print_to_PDF.print_resolution: 600 print.printer_Microsoft_Print_to_PDF.print_reversed: false print.printer_Microsoft_Print_to_PDF.print_scaling: 1.00 print.printer_Microsoft_Print_to_PDF.print_shrink_to_fit: true print.printer_Microsoft_Print_to_PDF.print_to_file: false print.printer_Microsoft_Print_to_PDF.print_unwriteable_margin_bottom: 0 print.printer_Microsoft_Print_to_PDF.print_unwriteable_margin_left: 0 print.printer_Microsoft_Print_to_PDF.print_unwriteable_margin_right: 0 print.printer_Microsoft_Print_to_PDF.print_unwriteable_margin_top: 0 print.printer_MTV2-49ers.print_bgcolor: false print.printer_MTV2-49ers.print_bgimages: false print.printer_MTV2-49ers.print_duplex: -437918235 print.printer_MTV2-49ers.print_edge_bottom: 0 print.printer_MTV2-49ers.print_edge_left: 0 print.printer_MTV2-49ers.print_edge_right: 0 print.printer_MTV2-49ers.print_edge_top: 0 print.printer_MTV2-49ers.print_evenpages: true print.printer_MTV2-49ers.print_in_color: true print.printer_MTV2-49ers.print_margin_bottom: 0.5 print.printer_MTV2-49ers.print_margin_left: 0.5 print.printer_MTV2-49ers.print_margin_right: 0.5 print.printer_MTV2-49ers.print_margin_top: 0.5 print.printer_MTV2-49ers.print_oddpages: true print.printer_MTV2-49ers.print_orientation: 0 print.printer_MTV2-49ers.print_page_delay: 50 print.printer_MTV2-49ers.print_paper_data: 1 print.printer_MTV2-49ers.print_paper_height: -1.00 print.printer_MTV2-49ers.print_paper_name: print.printer_MTV2-49ers.print_paper_size_unit: 1 print.printer_MTV2-49ers.print_paper_width: -1.00 print.printer_MTV2-49ers.print_resolution: 600 print.printer_MTV2-49ers.print_reversed: false print.printer_MTV2-49ers.print_scaling: 1.00 print.printer_MTV2-49ers.print_shrink_to_fit: true print.printer_MTV2-49ers.print_to_file: false print.printer_MTV2-49ers.print_unwriteable_margin_bottom: 0 print.printer_MTV2-49ers.print_unwriteable_margin_left: 0 print.printer_MTV2-49ers.print_unwriteable_margin_right: 0 print.printer_MTV2-49ers.print_unwriteable_margin_top: 0 print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_bgcolor: false print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_bgimages: false print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_duplex: -437918235 print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_edge_bottom: 0 print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_edge_left: 0 print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_edge_right: 0 print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_edge_top: 0 print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_evenpages: true print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_footercenter: print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_footerleft: &PT print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_footerright: &D print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_headercenter: print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_headerleft: &T print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_headerright: &U print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_in_color: true print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_margin_bottom: 0.5 print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_margin_left: 0.5 print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_margin_right: 0.5 print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_margin_top: 0.5 print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_oddpages: true print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_orientation: 0 print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_page_delay: 50 print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_paper_data: 1 print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_paper_height: -1.00 print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_paper_name: print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_paper_size_unit: 1 print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_paper_width: -1.00 print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_resolution: 600 print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_reversed: false print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_scaling: 1.00 print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_shrink_to_fit: true print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_to_file: false print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_unwriteable_margin_bottom: 0 print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_unwriteable_margin_left: 0 print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_unwriteable_margin_right: 0 print.printer_MTV2-Raiders_(HP_Color_LaserJet_CM2320nf_MFP).print_unwriteable_margin_top: 0 print.printer_MTV2-Warriors.print_bgcolor: false print.printer_MTV2-Warriors.print_bgimages: false print.printer_MTV2-Warriors.print_duplex: -437918235 print.printer_MTV2-Warriors.print_edge_bottom: 0 print.printer_MTV2-Warriors.print_edge_left: 0 print.printer_MTV2-Warriors.print_edge_right: 0 print.printer_MTV2-Warriors.print_edge_top: 0 print.printer_MTV2-Warriors.print_evenpages: true print.printer_MTV2-Warriors.print_in_color: true print.printer_MTV2-Warriors.print_margin_bottom: 0.5 print.printer_MTV2-Warriors.print_margin_left: 0.5 print.printer_MTV2-Warriors.print_margin_right: 0.5 print.printer_MTV2-Warriors.print_margin_top: 0.5 print.printer_MTV2-Warriors.print_oddpages: true print.printer_MTV2-Warriors.print_orientation: 0 print.printer_MTV2-Warriors.print_page_delay: 50 print.printer_MTV2-Warriors.print_paper_data: 1 print.printer_MTV2-Warriors.print_paper_height: -1.00 print.printer_MTV2-Warriors.print_paper_name: print.printer_MTV2-Warriors.print_paper_size_unit: 1 print.printer_MTV2-Warriors.print_paper_width: -1.00 print.printer_MTV2-Warriors.print_resolution: 600 print.printer_MTV2-Warriors.print_reversed: false print.printer_MTV2-Warriors.print_scaling: 1.00 print.printer_MTV2-Warriors.print_shrink_to_fit: true print.printer_MTV2-Warriors.print_to_file: false print.printer_MTV2-Warriors.print_unwriteable_margin_bottom: 0 print.printer_MTV2-Warriors.print_unwriteable_margin_left: 0 print.printer_MTV2-Warriors.print_unwriteable_margin_right: 0 print.printer_MTV2-Warriors.print_unwriteable_margin_top: 0 print.printer_TOR5-DrClaw.print_bgcolor: false print.printer_TOR5-DrClaw.print_bgimages: false print.printer_TOR5-DrClaw.print_duplex: -437918235 print.printer_TOR5-DrClaw.print_edge_bottom: 0 print.printer_TOR5-DrClaw.print_edge_left: 0 print.printer_TOR5-DrClaw.print_edge_right: 0 print.printer_TOR5-DrClaw.print_edge_top: 0 print.printer_TOR5-DrClaw.print_evenpages: true print.printer_TOR5-DrClaw.print_footercenter: print.printer_TOR5-DrClaw.print_footerleft: &PT print.printer_TOR5-DrClaw.print_footerright: &D print.printer_TOR5-DrClaw.print_headercenter: print.printer_TOR5-DrClaw.print_headerleft: &T print.printer_TOR5-DrClaw.print_headerright: &U print.printer_TOR5-DrClaw.print_in_color: true print.printer_TOR5-DrClaw.print_margin_bottom: 0.5 print.printer_TOR5-DrClaw.print_margin_left: 0.5 print.printer_TOR5-DrClaw.print_margin_right: 0.5 print.printer_TOR5-DrClaw.print_margin_top: 0.5 print.printer_TOR5-DrClaw.print_oddpages: true print.printer_TOR5-DrClaw.print_orientation: 0 print.printer_TOR5-DrClaw.print_page_delay: 50 print.printer_TOR5-DrClaw.print_paper_data: 1 print.printer_TOR5-DrClaw.print_paper_height: -1.00 print.printer_TOR5-DrClaw.print_paper_name: print.printer_TOR5-DrClaw.print_paper_size_unit: 1 print.printer_TOR5-DrClaw.print_paper_width: -1.00 print.printer_TOR5-DrClaw.print_resolution: 600 print.printer_TOR5-DrClaw.print_reversed: false print.printer_TOR5-DrClaw.print_scaling: 1.00 print.printer_TOR5-DrClaw.print_shrink_to_fit: true print.printer_TOR5-DrClaw.print_to_file: false print.printer_TOR5-DrClaw.print_unwriteable_margin_bottom: 0 print.printer_TOR5-DrClaw.print_unwriteable_margin_left: 0 print.printer_TOR5-DrClaw.print_unwriteable_margin_right: 0 print.printer_TOR5-DrClaw.print_unwriteable_margin_top: 0 privacy.sanitize.migrateClearSavedPwdsOnExit: true privacy.sanitize.migrateFx3Prefs: true privacy.trackingprotection.enabled: true privacy.trackingprotection.introCount: 20 security.sandbox.content.tempDirSuffix: {4da67282-f63a-4962-92a0-b1a2adb8331f} services.sync.declinedEngines: services.sync.engine.addons: false services.sync.engine.prefs: false services.sync.engine.prefs.modified: false services.sync.lastPing: 1459441265 services.sync.lastSync: Thu Mar 31 2016 14:49:09 GMT-0400 (Eastern Standard Time) services.sync.numClients: 7 storage.vacuum.last.index: 1 storage.vacuum.last.places.sqlite: 1457962171 ui.osk.debug.keyboardDisplayReason: IKPOS: Rotation sensor not found. Important Locked Preferences ---------------------------- JavaScript ---------- Incremental GC: true Accessibility ------------- Activated: true Prevent Accessibility: 0 Library Versions ---------------- NSPR Expected minimum version: 4.12 Version in use: 4.12 NSS Expected minimum version: 3.23 Basic ECC Version in use: 3.23 Basic ECC NSSSMIME Expected minimum version: 3.23 Basic ECC Version in use: 3.23 Basic ECC NSSSSL Expected minimum version: 3.23 Basic ECC Version in use: 3.23 Basic ECC NSSUTIL Expected minimum version: 3.23 Version in use: 3.23 Experimental Features --------------------- Name: DisplayPort Size Tuning ID: displayport-tuning-nightly@experiments.mozilla.org Description: Measuring the effect of different displayport sizes on checkerboarding. Active: false End Date: 1457971574628 Homepage: Branch: group2

Thank you for the profile. I'm still unsure what the problem might be here. The profile shows the main thread staying busy trying to paint an animation at 60fps. This might be the video timeline scrubber... not sure. However, the profile doesn't show whether there was any APZ-induced scrolling. There are a few cases where the main thread janks while painting - during those paints I would expect additional composites, triggered by APZ scrolling. But in the profile there aren't any composites happening during the jank periods, as far as I can tell; the compositor stays idle during those periods.

(In reply to Brad Lassey [:blassey] (use needinfo?) from comment #0) > I'm seeing vine videos in my twitter stream get real jittery while I'm > panning. The jitter only happens on the first play cycle though, not once it > repeats. Is it reproducible on nightly?

Yes, this is on nightly. To be clear, its the video that is janking/flickering during a scroll, not the scroll that is janking.

hopefully this illustrates the issue https://www.dropbox.com/s/qt5rbbi6i2iakcw/%284%29%20Twitter%20-%20Nightly%204_6_2016%209_28_17%20AM.mp4?dl=0

Oooh! This video made the problem much clearer to me. I thought the video was pausing, but it's really alternating between random video frames. Jim, are we showing an outdated snapshot for the plugin?

(In reply to Markus Stange [:mstange] from comment #7) > Oooh! This video made the problem much clearer to me. I thought the video > was pausing, but it's really alternating between random video frames. > > Jim, are we showing an outdated snapshot for the plugin? I seriously doubt that, we only trigger captures on first load and after a scroll completes. Also, captures are broken under the sandbox currently, which is on in nightly. (blassey, are you running nightly?) There's a debug log dump you can turn on to see if this code is interacting with this. Should show up in release builds with a debugger attached to the content process. http://mxr.mozilla.org/mozilla-central/source/dom/plugins/ipc/PluginInstanceParent.cpp#1215

If this is a <video>, and not a plugin, then Matt might know what's going on here.

I don't use twitter, can someone find out if it's actually a video or plugin?

Brad - can you make life easy for Matt and add a link and STR?

(In reply to Matt Woodrow (:mattwoodrow) from comment #10) > I don't use twitter, can someone find out if it's actually a video or plugin? I just checked, it's a <video>. At least on Mac.

(In reply to Anthony Jones (:kentuckyfriedtakahe, :k17e) from comment #11) > Brad - can you make life easy for Matt and add a link and STR? STR: 1) Navigate to twtter.com 2) scroll an auto-play video into view. Following a video heavy account like https://twitter.com/AwardsDarwin might help

Brad - thanks, that is helpful. I can see what looks like a left to right wobble. Looks like out of order frames. I think media.windows-media-foundation.allow-d3d11-dxva=false makes the problem go away. Can you try it to see if it looks better for you?

It does sound similar to bug 1236112. That should definitely be fixed on Nightly though (either by that bug or bug 1257013), since it was all landed by 3/22. I can't reproduce this fwiw, scrolling on that twitter feed plays the videos fine. :(

Is this a regression? If so, can you please get a regression range? Testing the landing of bug 1257013 might be a good starting point.

I'm away for two weeks and don't have my laptop with me. Let me know if you'd still like me to do annoying when I get back

STR: * Navigate to http://people.mozilla.org/~ajones/frame-drop-test/bbb-11.mp4 * Refresh * Click on full-screen Expected Results: Normal playback Actual Results: Jumpy playback Regression Range: https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=2678b91f4165e884f043df42bcddc91f91d2bb69&tochange=6e705888d2a854f60deadc23d1382fbba2ef7dc2

It looks like TextureClientRecycleAllocator doesn't wait for the compositor to release textures, but instead recycles them as soon as we drop the ref on the client side. My understanding is that Sotaro's work in bug 1252835 will fix this. Is that correct Sotaro?

Anthony, there's an old-ish try build in that bug, do you want to give that a shot? https://treeherder.mozilla.org/#/jobs?repo=try&revision=6c9baec83d78

(In reply to Matt Woodrow (:mattwoodrow) from comment #19) > It looks like TextureClientRecycleAllocator doesn't wait for the compositor > to release textures, but instead recycles them as soon as we drop the ref on > the client side. > > My understanding is that Sotaro's work in bug 1252835 will fix this. Is that > correct Sotaro? Yea, it is correct. In current gecko's TextureClient recycling, Host side's reference is not cared enough. But we might need additional fix to handle direct buffer binding case. It is not a problem on gonk, because we use android fence to wait end of its usage.

We might want something simpler so we can uplift to beta. Here's a couple of options: https://treeherder.mozilla.org/#/jobs?repo=try&revision=b13eaaa45bd7

https://treeherder.mozilla.org/#/jobs?repo=try&revision=0e739bc67de5 Anthony, can you test these and make sure they work? Thanks!

[Tracking Requested - why for this release]: Regression from bug 1257013, probably earlier for d3d9 video. Out of order video frames presented sometimes.

(In reply to Matt Woodrow (:mattwoodrow) from comment #22) > force-compositor-recycle-video > https://treeherder.mozilla.org/#/jobs?repo=try&revision=b13eaaa45bd7 The problem is still reproducible with force-compositor-recycle-video > use-a-queue-for-recycling > https://treeherder.mozilla.org/#/jobs?repo=try&revision=0e739bc67de5 I could not preproduce the problem with use-a-queue-for-recycling. I will continue to test it.

(In reply to Anthony Jones (:kentuckyfriedtakahe, :k17e) from comment #25) > (In reply to Matt Woodrow (:mattwoodrow) from comment #22) > > force-compositor-recycle-video > > https://treeherder.mozilla.org/#/jobs?repo=try&revision=b13eaaa45bd7 > > The problem is still reproducible with force-compositor-recycle-video I re-pushed this one with the obvious bug fixed: https://treeherder.mozilla.org/#/jobs?repo=try&revision=ba565fc75bb9

(In reply to Matt Woodrow (:mattwoodrow) from comment #26) > https://treeherder.mozilla.org/#/jobs?repo=try&revision=ba565fc75bb9 Fixes nothing, crashes a lot. https://crash-stats.mozilla.com/report/index/ba0d1f62-6a99-4f87-9836-234fa2160428

Ok, I think I have a good idea of what is happening here. WaitForCompositorRecycle only waits for the TextureHost to get all refs dropped, it doesn't wait for the GPU to have finished reading from the underlying texture data. Sotaro, do you have any plans to fix this as part of your recycling fix?

Yes, before this bug, I was originally thought to address the problem as a follow up of bug 1252835. But because of this bug, it might be better to change the plan. If we want to uplift the fix, it seems better to address this problem independently from bug 1252835, since bug 1252835 is going to change a lot. Do we need to uplift the fix?

(In reply to Sotaro Ikeda [:sotaro ouf of office 29/Apr - 8/May] from comment #29) > Yes, before this bug, I was originally thought to address the problem as a > follow up of bug 1252835. But because of this bug, it might be better to > change the plan. > > If we want to uplift the fix, it seems better to address this problem > independently from bug 1252835, since bug 1252835 is going to change a lot. > Do we need to uplift the fix? This is a regression, so being able to uplift would be really nice. That said, the patch on this bug ' use-a-queue-for-recycling' does seem to fix thing (even if it is a horrible hack), so we have that as a last resort if necessary. I'd much rather have an upliftable real fix though, which I think is possible. Have you started on anything, and do you want me to?

Ritu - can I have this bug tracking on 47? It is the most important bug to fix for media playback in 47. It affects Widevine playback on fast machines and also prevents us from rolling out ffvp9 on accelerated layers.

(In reply to Anthony Jones (:kentuckyfriedtakahe, :k17e) from comment #32) > Ritu - can I have this bug tracking on 47? > > It is the most important bug to fix for media playback in 47. It affects > Widevine playback on fast machines and also prevents us from rolling out > ffvp9 on accelerated layers. Absolutely! Thank you for flagging this one. I have marked this blocking for Fx47 release so it gets max attention from relman, devs, etc.

https://treeherder.mozilla.org/#/jobs?repo=try&revision=e84d0fe8fdc7 Anthony, can you give this one a try please when it finishes building.

Apologies, hopefully this one will compile: https://treeherder.mozilla.org/#/jobs?repo=try&revision=ccb9928bf693

(In reply to Matt Woodrow (:mattwoodrow) from comment #35) > Apologies, hopefully this one will compile: > https://treeherder.mozilla.org/#/jobs?repo=try&revision=ccb9928bf693 This seems to fix the issue. I will need to file a separate bug for the Choice TV issue.

Comment on attachment 8749062 [details] [diff] [review] Part 2: Block CompositorD3D11 after presenting until the previous frame is done Review of attachment 8749062 [details] [diff] [review]: ----------------------------------------------------------------- ::: gfx/layers/d3d11/CompositorD3D11.cpp @@ +1276,5 @@ > > + // Spin and block until the previous frames work has been completed. > + if (mQuery) { > + BOOL result; > + while (mContext->GetData(mQuery, &result, sizeof(BOOL), 0) != S_OK) {} We should consider yielding inside this while loop. We perhaps should also consider a timeout of say, a second or two, I don't know how this will play with device resets.

I backed to work today. But my laptop had a problem today and I did not have time. I am going to check the patch tomorrow.

(In reply to Matt Woodrow (:mattwoodrow) from comment #30) > I'd much rather have an upliftable real fix though, which I think is > possible. > > Have you started on anything, and do you want me to? Thanks, you are already working for it during my vacation.

Comment on attachment 8749061 [details] [diff] [review] Part 1: Opt-in to waiting for compositor recycling for TextureClientRecycleAllocator Review of attachment 8749061 [details] [diff] [review]: ----------------------------------------------------------------- This is simple enough as a short term workaround, though this is a bit hacky. Just applying the patch addressed the problem of Comment 18 on my laptop.

I had a look at what ANGLE does, and this seems to work for them.

Comment on attachment 8749063 [details] [diff] [review] Part 3: Hold on to TextureHosts until the following composite is done Review of attachment 8749063 [details] [diff] [review]: ----------------------------------------------------------------- The patch works well with part 1 patch. It put off the TextureClient recycling. It is good as a short term solution. You might want to call mPreviousHeldTextureHosts.Clear() in LayerManagerComposite::Destroy(). ::: gfx/layers/composite/ImageHost.cpp @@ +331,5 @@ > > TimedImage* img = &mImages[imageIndex]; > img->mTextureHost->SetCompositor(GetCompositor()); > + if (img->mTextureHost->GetFlags() & TextureFlags::RECYCLE) { > + aLayer->GetLayerManager()->HoldTextureUntilNextComposite(img->mTextureHost); You might want to add a comment here why it is necessary.

Comment on attachment 8750572 [details] [diff] [review] Part 2: Block CompositorD3D11 after presenting until the previous frame is done Review of attachment 8750572 [details] [diff] [review]: ----------------------------------------------------------------- ::: gfx/layers/d3d11/CompositorD3D11.cpp @@ +1280,5 @@ > + while (mContext->GetData(mQuery, &result, sizeof(BOOL), 0) != S_OK) { > + if (mDevice->GetDeviceRemovedReason() != S_OK) { > + break; > + } > + Sleep(0); I still wonder whether we should simply bail if say, 5 seconds have passed. It's really bad if by some reason this would hang as it's on the compositor thread and I'm not convinced our crash reporter will catch that. If you're going to push this code without having a break-out here. Please check that our crash reporter catches this. We don't want to be stuck with people in the wild having a frozen compositor that we won't find out about easily.

https://hg.mozilla.org/integration/mozilla-inbound/rev/787b1383d901 https://hg.mozilla.org/integration/mozilla-inbound/rev/666c10f1dd52 https://hg.mozilla.org/integration/mozilla-inbound/rev/b1c6143527c0

https://hg.mozilla.org/mozilla-central/rev/787b1383d901 https://hg.mozilla.org/mozilla-central/rev/666c10f1dd52 https://hg.mozilla.org/mozilla-central/rev/b1c6143527c0

Matt, is this fix safe enough to uplift to Aurora 48 or even Beta 47?

(In reply to Chris Peterson [:cpeterson] from comment #49) > Matt, is this fix safe enough to uplift to Aurora 48 or even Beta 47? That is the hope, yes. Lets give it a few days and Nightly first though.

Any particular testing that should happen in the meantime on Nightly? See https://bugzilla.mozilla.org/show_bug.cgi?id=1260611#c32 for more description of the issues (in addition to jitteriness on vine/twitter) Andrei is this something your team can look at?

(In reply to Liz Henry (:lizzard) (needinfo? me) from comment #51) > Any particular testing that should happen in the meantime on Nightly? > See https://bugzilla.mozilla.org/show_bug.cgi?id=1260611#c32 for more > description of the issues (in addition to jitteriness on vine/twitter) > > Andrei is this something your team can look at? The best way to reproduce it is a combination of: * Low quality/resolution video * Busy GPU This causes the video decoding to get ahead of the compositing. I really want to see this fix get into beta because D3D11 DXVA should put us slightly ahead on playback performance and it would be a shame if it were to be problematic for some users or use cases. Brad - can you see if you can still reproduce the issue?

Unable to reproduce using all the available guidance (comment 13, comment 18 and comment 52) on latest Nightly 49.0a1, under Windows 10 64-bit and Mac OS X - everything is looking good. Also tried to replicate this issue with an older Nightly (from 2016-04-18), but with no success.

Doesn't look like it reproduces for me anymore

(In reply to Matt Woodrow (:mattwoodrow) from comment #50) > (In reply to Chris Peterson [:cpeterson] from comment #49) > > Matt, is this fix safe enough to uplift to Aurora 48 or even Beta 47? > > That is the hope, yes. > > Lets give it a few days and Nightly first though. Matt -- Can you nominate this? I think it's had enough time, and it's currently being tracked by relman as a blocker for Fx47. Thanks!

Comment on attachment 8749061 [details] [diff] [review] Part 1: Opt-in to waiting for compositor recycling for TextureClientRecycleAllocator Approval Request Comment [Feature/regressing bug #]: Bug 1257013 [User impact if declined]: Out of order frames on some videos [Describe test coverage new/current, TreeHerder]: Manually tested, verified by bug reporter. [Risks and why]: Fairly low risk, just stops us from recycling video frames too early. [String/UUID change made/needed]: None

Comment on attachment 8749061 [details] [diff] [review] Part 1: Opt-in to waiting for compositor recycling for TextureClientRecycleAllocator Recent regression in Widevine CDM video playback, fix was verified by a few people. Aurora48+, Beta47+

Part 1 doesn't apply cleanly to beta, can we get rebased patches if this is needed? hg graft -r 344895:344897 grafting 344895:4a4cf532f07c "Bug 1260611 - Part 1: Opt-in to waiting for compositor recycling for TextureClientRecycleAllocator. r=sotaro a=ritu" merging gfx/layers/client/TextureClient.cpp merging gfx/layers/client/TextureClient.h merging gfx/layers/client/TextureClientRecycleAllocator.cpp warning: conflicts while merging gfx/layers/client/TextureClientRecycleAllocator.cpp! (edit, then use 'hg resolve --mark') abort: unresolved conflicts, can't continue

remote: https://hg.mozilla.org/releases/mozilla-aurora/rev/4a4cf532f07c remote: https://hg.mozilla.org/releases/mozilla-aurora/rev/e5dc1c7fcb08 remote: https://hg.mozilla.org/releases/mozilla-aurora/rev/b91fad4c1ea7

All builds were broken when this landed: https://treeherder.mozilla.org/logviewer.html#?job_id=2665813&repo=mozilla-aurora Backed out in https://hg.mozilla.org/releases/mozilla-aurora/rev/4f7c3a2567f8

remote: https://hg.mozilla.org/releases/mozilla-aurora/rev/6d2e4eb24c2e remote: https://hg.mozilla.org/releases/mozilla-aurora/rev/0ecc04efa742 remote: https://hg.mozilla.org/releases/mozilla-aurora/rev/33693d8bdec1

remote: https://hg.mozilla.org/releases/mozilla-beta/rev/f9dd2fb1b991 remote: https://hg.mozilla.org/releases/mozilla-beta/rev/fa709c256861 remote: https://hg.mozilla.org/releases/mozilla-beta/rev/7c3ef5d06736