User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:53.0) Gecko/20100101 Firefox/53.0 Build ID: 20161223030226 Steps to reproduce: 1. Load https://www.google.com/ Actual results: See attached screenshot

2016-12-24T12:41:26: INFO : application_vendor: Mozilla 2016-12-24T12:41:26: INFO : application_version: 53.0a1 2016-12-24T12:41:26: INFO : platform_buildid: 20161222010952 2016-12-24T12:41:26: INFO : platform_changeset: 66b31b9d9a1a636ef8855e578867bdb34c435546 2016-12-24T12:41:26: INFO : platform_repository: https://hg.mozilla.org/integration/mozilla-inbound 2016-12-24T12:41:26: INFO : platform_version: 53.0a1 2016-12-24T12:41:36: INFO : Narrowed inbound regression window from [3fe0ff6b, 96cf05fa] (4 revisions) to [66b31b9d, 96cf05fa] (2 revisions) (~1 steps left) 2016-12-24T12:41:36: DEBUG : Starting merge handling... 2016-12-24T12:41:36: DEBUG : Using url: https://hg.mozilla.org/integration/mozilla-inbound/json-pushes?changeset=96cf05faacfe6ba130553323bde9f0bb80298c8a&full=1 2016-12-24T12:41:37: DEBUG : Found commit message: Bug 1325199: Only read system parameters for ClearType when we receive a paint event. r=jrmuizel MozReview-Commit-ID: EjQyCcdWXEc With Multiprocess disabled everything is fine

Application Basics ------------------ Name: Firefox Version: 53.0a1 Build ID: 20161222030652 Update Channel: default User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:53.0) Gecko/20100101 Firefox/53.0 OS: Windows_NT 10.0 Multiprocess Windows: 1/1 (Enabled by default) Safe Mode: false Crash Reports for the Last 3 Days --------------------------------- Report ID: bp-6c4735b0-67fc-451c-a22a-1b6b62161222 Submitted: 2 days ago Report ID: bp-cd2eb9df-fb0e-4f17-b0c0-9a73e2161221 Submitted: 2 days ago All Crash Reports Extensions ---------- Name: Application Update Service Helper Version: 1.0 Enabled: true ID: aushelper@mozilla.org Name: FlyWeb Version: 1.0.0 Enabled: true ID: flyweb@mozilla.org Name: Form Autofill Version: 1.0 Enabled: true ID: formautofill@mozilla.org Name: Multi-process staged rollout Version: 1.6 Enabled: true ID: e10srollout@mozilla.org Name: Pocket Version: 1.0.5 Enabled: true ID: firefox@getpocket.com Name: Presentation Version: 1.0.0 Enabled: true ID: presentation@mozilla.org Name: Web Compat Version: 1.0 Enabled: true ID: webcompat@mozilla.org Graphics -------- Features Compositing: Direct3D 11 Asynchronous Pan/Zoom: wheel input enabled; touch input enabled WebGL Renderer: Google Inc. -- ANGLE (AMD Radeon R9 200 Series Direct3D11 vs_5_0 ps_5_0) WebGL2 Renderer: Google Inc. -- ANGLE (AMD Radeon R9 200 Series Direct3D11 vs_5_0 ps_5_0) Hardware H264 Decoding: Yes; Failed to create D3D11 device for decoder; Using D3D9 API Audio Backend: wasapi Direct2D: true DirectWrite: true (10.0.14393.351) GPU #1 Active: Yes Description: AMD Radeon R9 200 Series Vendor ID: 0x1002 Device ID: 0x6810 Driver Version: 21.19.137.1 Driver Date: 9-16-2016 Drivers: aticfx64 aticfx64 aticfx64 amdxc64 aticfx32 aticfx32 aticfx32 amdxc32 atiumd64 atidxx64 atidxx64 atiumdag atidxx32 atidxx32 atiumdva atiumd6a atitmm64 Subsys ID: 00000000 RAM: 2048 Diagnostics ClearType Parameters: Gamma: 2,2 Pixel Structure: RGB ClearType Level: 50 Enhanced Contrast: 50 AzureCanvasAccelerated: 0 AzureCanvasBackend: skia AzureContentBackend: skia AzureFallbackCanvasBackend: cairo GPUProcessPid: 7996 GPUProcess: Terminate GPU Process ClearType Parameters: Gamma: 2,2 Pixel Structure: RGB ClearType Level: 50 Enhanced Contrast: 50 Decision Log D3D9_COMPOSITING: disabled by default: Disabled by default 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: 1 browser.download.importedFromSqlite: true browser.places.smartBookmarksVersion: 8 browser.sessionstore.resume_from_crash: false browser.startup.homepage_override.buildID: 20161222030652 browser.startup.homepage_override.mstone: 53.0a1 browser.tabs.warnOnClose: false browser.urlbar.daysBeforeHidingSuggestionsPrompt: 3 browser.urlbar.lastSuggestionsPromptDate: 20161224 dom.gamepad.extensions.enabled: true extensions.lastAppVersion: 53.0a1 media.gmp-gmpopenh264.abi: x86_64-msvc-x64 media.gmp-gmpopenh264.lastUpdate: 1482560325 media.gmp-gmpopenh264.version: 1.6 media.gmp-manager.buildID: 20161222030652 media.gmp-manager.lastCheck: 1482560324 media.gmp-widevinecdm.abi: x86_64-msvc-x64 media.gmp-widevinecdm.lastUpdate: 1482560326 media.gmp-widevinecdm.version: 1.4.8.903 media.gmp.storage.version.observed: 1 media.hardware-video-decoding.failed: false network.cookie.prefsMigrated: true network.predictor.cleaned-up: true places.history.expiration.transient_current_max_pages: 122334 plugin.disable_full_page_plugin_for_types: application/pdf security.sandbox.content.tempDirSuffix: {207e0df9-2350-4d9e-aa3d-9f14c553c9f8} ui.osk.debug.keyboardDisplayReason: IKPOS: Touch screen not found. user.js Preferences ------------------- Your profile folder contains a user.js file, which includes preferences that were not created by Nightly. Important Locked Preferences ---------------------------- Places Database --------------- JavaScript ---------- Incremental GC: true Accessibility ------------- Activated: false Prevent Accessibility: 0 Library Versions ---------------- NSPR Expected minimum version: 4.13.1 Version in use: 4.13.1 NSS Expected minimum version: 3.29 Beta Version in use: 3.29 Beta NSSSMIME Expected minimum version: 3.29 Beta Version in use: 3.29 Beta NSSSSL Expected minimum version: 3.29 Beta Version in use: 3.29 Beta NSSUTIL Expected minimum version: 3.29 Beta Version in use: 3.29 Beta Experimental Features --------------------- Sandbox ------- Content Process Sandbox Level: 2

[Tracking Requested - why for this release]: I can't reproduce it with my Nvidia GPU.

Tracking 53+ - we should find out if other Radeon GPUs are affected as well. ni on Andrei to see if SV has any such machines available to test.

Got some free time for some testing 1. Updated video drivers to latest version (16.12.1). Same result 2. I can reproduce this problem on every wabpage, not just google.com, except https://www.mozilla.org/ :) 3. With hardware acceleration disabled everything is fine to Any suggestion?

Really surprising the change flagged in comment 0 would be responsible for this. Is about:support drawn incorrectly as well? Does disabling "multi-process Nightly" in about:preferences make a difference?

(In reply to Milan Sreckovic [:milan] from comment #6) > Really surprising the change flagged in comment 0 would be responsible for > this. Is about:support drawn incorrectly as well? UI in browser looks fine, checked some pages from about:about > Does disabling "multi-process Nightly" in about:preferences make a difference? It does, as i noted in https://bugzilla.mozilla.org/show_bug.cgi?id=1325743#c1 with Multiprocess disabled everything is fine Tried to run mozregression tool several times and always hit bug https://bugzilla.mozilla.org/show_bug.cgi?id=1325199 in the end. I can make video, but whole lot of pages looks the same. I can see some images and rarely text, when hovering cursor over page elements those i can't see at first they appear. for example on youtube hovering over video preview make some other preview disappear. Switching from tab to tab make some or all content on the page to disappear completely

I didn't managed to reproduce this issue with e10s and hardware acceleration enabled, using latest Nightly 53.0a1 (2017-01-02). I've tried on 3 different Radeon GPUs (see below) but, the sites mentioned in this report and also others are always rendered correctly on my side. - AMD Radeon HD 6450 under Win 7 x86 and 10 x64 - ATI Radeon HD 3000 under Win 10 x64 - AMD Radeon R7 M340 under Win 10 x64 If I can help with more info please ni? me!

I can reproduce this issue on a laptop with AMD FirePro M4000 graphic card, but only when I connect to it via Remote Desktop client (and as long as I have e10s enabled). Bisection points to the same commit as mentioned previously: INFO : Narrowed inbound regression window from [3fe0ff6b, 96cf05fa] (4 revisions) to [66b31b9d, 96cf05fa] (2 revisions) (~1 steps left) DEBUG : Starting merge handling... DEBUG : Using url: https://hg.mozilla.org/integration/mozilla-inbound/json-pushes?changeset=96cf05faacfe6ba130553323bde9f0bb80298c8a&full=1 DEBUG : Found commit message: Bug 1325199: Only read system parameters for ClearType when we receive a paint event. r=jrmuizel MozReview-Commit-ID: EjQyCcdWXEc INFO : The bisection is done. But at least in my case, it could be caused by bug 1285563, autoland changeset 3d4f10649284. That change enabled hardware acceleration when using RDP and caused "RDPDD Chained DD" to be reported as inactive GPU for me. Relevant about:support data before changeset 3d4f10649284: Application Basics ------------------ Name: Firefox Version: 53.0a1 Build ID: 20161216090449 Update Channel: default User Agent: Mozilla/5.0 (Windows NT 6.1; Win64; x64; rv:53.0) Gecko/20100101 Firefox/53.0 OS: Windows_NT 6.1 Graphics -------- Features Compositing: Basic Asynchronous Pan/Zoom: none WebGL Renderer: Google Inc. -- ANGLE (Software Adapter Direct3D11 vs_5_0 ps_5_0) WebGL2 Renderer: Google Inc. -- ANGLE (Software Adapter Direct3D11 vs_5_0 ps_5_0) Hardware H264 Decoding: No; Hardware video decoding disabled or blacklisted Audio Backend: wasapi Direct2D: Blocked for your graphics card because of unresolved driver issues. DirectWrite: true (6.2.9200.21976) GPU #1 Active: Yes Description: RDPDD Chained DD Vendor ID: 0x0000 Device ID: 0x0000 Drivers: RDPDD Subsys ID: 00000000 RAM: Unknown GPU #2 Active: No Description: AMD FirePro M4000 Vendor ID: 0x1002 Device ID: 0x682d Driver Version: 16.200.1021.0 Driver Date: 6-21-2016 Drivers: aticfx64 aticfx64 aticfx64 aticfx32 aticfx32 aticfx32 atiumd64 atidxx64 atidxx64 atiumdag atidxx32 atidxx32 atiumdva atiumd6a atitmm64 Subsys ID: 176c103c RAM: 1024 Diagnostics AzureCanvasAccelerated: 0 AzureCanvasBackend: skia AzureContentBackend: skia AzureFallbackCanvasBackend: cairo Decision Log D3D11_COMPOSITING: Blocklisted; failure code BLOCKLIST_FEATURE_FAILURE_UNKNOWN_DEVICE_VENDOR D3D9_COMPOSITING: disabled by default: Disabled by default Blocklisted; failure code BLOCKLIST_FEATURE_FAILURE_UNKNOWN_DEVICE_VENDOR DIRECT2D: unavailable by default: Direct2D requires Direct3D 11 compositing D3D11_HW_ANGLE: unavailable by default: D3D11 compositing is disabled disabled by env: D3D11 compositing is disabled And after: Application Basics ------------------ Name: Firefox Version: 53.0a1 Build ID: 20161216091149 Update Channel: default User Agent: Mozilla/5.0 (Windows NT 6.1; Win64; x64; rv:53.0) Gecko/20100101 Firefox/53.0 OS: Windows_NT 6.1 Graphics -------- Features Compositing: Direct3D 11 Asynchronous Pan/Zoom: none WebGL Renderer: Google Inc. -- ANGLE (AMD FirePro M4000 Direct3D11 vs_5_0 ps_5_0) WebGL2 Renderer: Google Inc. -- ANGLE (AMD FirePro M4000 Direct3D11 vs_5_0 ps_5_0) Hardware H264 Decoding: No; D3D9 blacklisted with DLL atiumd64.dll (9.14.10.1197) Audio Backend: wasapi Direct2D: true DirectWrite: true (6.2.9200.21976) GPU #1 Active: Yes Description: AMD FirePro M4000 Vendor ID: 0x1002 Device ID: 0x682d Driver Version: 16.200.1021.0 Driver Date: 6-21-2016 Drivers: aticfx64 aticfx64 aticfx64 aticfx32 aticfx32 aticfx32 atiumd64 atidxx64 atidxx64 atiumdag atidxx32 atidxx32 atiumdva atiumd6a atitmm64 Subsys ID: 176c103c RAM: 1024 GPU #2 Active: No Description: RDPDD Chained DD Vendor ID: 0x0000 Device ID: 0x0000 Drivers: RDPDD Subsys ID: 00000000 RAM: Unknown Diagnostics AzureCanvasAccelerated: 0 AzureCanvasBackend: direct2d 1.1 AzureContentBackend: direct2d 1.1 AzureFallbackCanvasBackend: cairo Decision Log D3D9_COMPOSITING: disabled by default: Disabled by default

(In reply to Ziga Seilnacht from comment #13) > I can reproduce this issue on a laptop with AMD FirePro M4000 graphic card, > but only when I connect to it via Remote Desktop client (and as long as I > have e10s enabled). My situation in bug 1328496 comment 3 is also remote desktop. I'll have to check the screen when I get in to work.

ignus, ziega, did the regression test (or did you manually try) the 2016-12-18 build, to totally rule out bug 1285563 as being the cause?

I used mozregression to bisect both problems mentioned in my comment; bug 1285563 (changeset 3d4f10649284) enabled hardware acceleration for Remote Desktop (Windows 7's RDPDD Chained DD), which I think was not intended. Bug 1328496 (changeset 96cf05faacfe) is where I start seeing rendering glitches as visible in ignus' screenshots - bug 1325944 could be another report of the same issue, this time with Nvidia graphics. The rendering glitches are not visible in safe mode (which disables hardware acceleration) for me, so it seems that the combination of both changesets is responsible for this bug to be visible in Remote Desktop sessions.

I can't reproduce this issue with nightly build 2016-12-18 https://archive.mozilla.org/pub/firefox/nightly/2016/12/2016-12-18-03-02-13-mozilla-central/firefox-53.0a1.en-US.win64.zip

Reporter indicates this is fixed on nightly.

(In reply to Bas Schouten (:bas.schouten) from comment #18) > Reporter indicates this is fixed on nightly. Sorry, but no. I only checked build 2016-12-18 as asked by Wayne Mery

It's still broken in latest nightly

(In reply to Wayne Mery (:wsmwk, NI for questions) from comment #14) > (In reply to Ziga Seilnacht from comment #13) > > I can reproduce this issue on a laptop with AMD FirePro M4000 graphic card, > > but only when I connect to it via Remote Desktop client (and as long as I > > have e10s enabled). > > My situation in bug 1328496 comment 3 is also remote desktop. I'll have to > check the screen when I get in to work. Just checked - direct viewing without remote desktop is fine. Then ooked again with Remote desktop - broken.

(In reply to Wayne Mery (:wsmwk, NI for questions) from comment #21) > (In reply to Wayne Mery (:wsmwk, NI for questions) from comment #14) > Just checked - direct viewing without remote desktop is fine. > Then ooked again with Remote desktop - broken. I can see this bug without any RDP session to my computer...

(In reply to Wayne Mery (:wsmwk, NI for questions) from comment #21) > (In reply to Wayne Mery (:wsmwk, NI for questions) from comment #14) > > (In reply to Ziga Seilnacht from comment #13) > > > I can reproduce this issue on a laptop with AMD FirePro M4000 graphic card, > > > but only when I connect to it via Remote Desktop client (and as long as I > > > have e10s enabled). > > > > My situation in bug 1328496 comment 3 is also remote desktop. I'll have to > > check the screen when I get in to work. > > Just checked - direct viewing without remote desktop is fine. > Then ooked again with Remote desktop - broken. Hrm, I can't reproduce this either with, or without RDP. Could you post an about:graphics from when you're running using RDP.

(In reply to ignus from comment #22) > (In reply to Wayne Mery (:wsmwk, NI for questions) from comment #21) > > (In reply to Wayne Mery (:wsmwk, NI for questions) from comment #14) > > Just checked - direct viewing without remote desktop is fine. > > Then ooked again with Remote desktop - broken. > I can see this bug without any RDP session to my computer... The patch above should get us back the "no acceleration with RDP", but that's a recent regression; :ignus' issue would be something else.

I believe I have found a way to reproduce this issue, when I disable cleartype in the control panel -while running- firefox, we appear to run into this issue. I'm looking into it.

(In reply to Milan Sreckovic [:milan] from comment #25) > (In reply to ignus from comment #22) > > (In reply to Wayne Mery (:wsmwk, NI for questions) from comment #21) > > > (In reply to Wayne Mery (:wsmwk, NI for questions) from comment #14) > > > Just checked - direct viewing without remote desktop is fine. > > > Then ooked again with Remote desktop - broken. > > I can see this bug without any RDP session to my computer... > > The patch above should get us back the "no acceleration with RDP", but > that's a recent regression; :ignus' issue would be something else. The cause for this is essentially that in the content process OnPaint events are never received and we're not appropriately updating the drawing mode.

My remote system (desktop running Firefox) and my local system (laptop running RDP)

Comment on attachment 8823760 [details] Bug 1325743: Part 1. When RDP is in play, do not swap the GPUs. .schouten https://reviewboard.mozilla.org/r/102274/#review102624 This won't fix the bug, and I even doubt it is a good idea. I have a patch upcoming.

(In reply to Bas Schouten (:bas.schouten) from comment #29) > Comment on attachment 8823760 [details] > Bug 1325743: Part 1. When RDP is in play, do not swap the GPUs. .schouten > > https://reviewboard.mozilla.org/r/102274/#review102624 > > This won't fix the bug, and I even doubt it is a good idea. I have a patch > upcoming. This is backing out part of the change from bug 1285563, where running RDP now reports as running the discrete card. It could be we want to keep that though.

Passing the review to Mason.

(In reply to Milan Sreckovic [:milan] from comment #24) > Created attachment 8823760 [details] > Bug 1325743: Part 1. When RDP is in play, do not swap the GPUs. .schouten > > Review commit: https://reviewboard.mozilla.org/r/102274/diff/#index_header > See other reviews: https://reviewboard.mozilla.org/r/102274/ > > + bool primaryIsRDP = false; > if (mWindowsVersion >= kWindows8 && > mDeviceID[0].Length() == 0 && > mDeviceString[0].EqualsLiteral("RDPUDD Chained DD")) > { > WCHAR sysdir[255]; > UINT len = GetSystemDirectory(sysdir, sizeof(sysdir)); > if (len < sizeof(sysdir)) { > + primaryIsRDP = true; This looks wrong; primaryIsRDP would get set to true only in Windows 8+, where RDP hardware acceleration was enabled even before bug 1285563. The questionable part of bug 1285563 is that it enabled RDP hardware acceleration on Windows 7, where it wasn't enabled previously.

Comment on attachment 8823775 [details] Bug 1325743: Cache Moz2D ClearType usage information in a way similar to Thebes. https://reviewboard.mozilla.org/r/102290/#review103022 ::: gfx/2d/HelpersWinFonts.h:22 (Diff revision 1) > + > +// Cleartype can be dynamically enabled/disabled, so we have to allow for dynamically > +// updating it. > +static void > +UpdateSystemTextQuality() > +{ nit: Are we sure this is always accessed only by one thread? Can we assert that thread here please?

Comment on attachment 8823775 [details] Bug 1325743: Cache Moz2D ClearType usage information in a way similar to Thebes. https://reviewboard.mozilla.org/r/102290/#review104036 ::: gfx/2d/HelpersWinFonts.h:22 (Diff revision 1) > + > +// Cleartype can be dynamically enabled/disabled, so we have to allow for dynamically > +// updating it. > +static void > +UpdateSystemTextQuality() > +{ This is actually really tricky to do in Moz2D. I'll look into this but for now I want to fix the bug.

Pushed by bschouten@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/84f4a96a130b Cache Moz2D ClearType usage information in a way similar to Thebes. r=mchang

I had to back this out for being the likely cause of a bunch of reftest failures on Windows 7 VM: https://treeherder.mozilla.org/#/jobs?repo=autoland&fromchange=10841f767eaf3f85cd1804016553fdeffc07b82e&group_state=expanded&filter-resultStatus=testfailed&filter-resultStatus=busted&filter-resultStatus=exception&filter-resultStatus=success&filter-classifiedState=unclassified&filter-searchStr=3fd70802c3f5b4e1bc35a83215b433bd10b2bae0 https://hg.mozilla.org/mozilla-central/rev/20094311ab5ec56d5be7afc6d5c2e2b09e656fb2

Hmm, the failures kept happening on the backout push. I guess this is safe to reland, then?

(In reply to Wes Kocher (:KWierso) from comment #41) > Hmm, the failures kept happening on the backout push. I guess this is safe > to reland, then? It should be, can you reland it or should I?

I'm on time off until mid next week, would be faster for you to reland it. You should just have to reopen the review request and retrigger autoland.

I tried reopening but don't have the permission; seems like the author has to (or somebody with magical powers I don't have.)

Pushed by bschouten@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/279ce8dbf8f6 Cache Moz2D ClearType usage information in a way similar to Thebes. r=mchang

v.fixed daily build

Windows 10 Pro 64-bit,Direct X 12, Nvidia GeForce GTX 550 TI, Nightly 54.0a1 (2017-02-15) (64-bit) I could not replicate the issues.

Updating status flags based on Comment 47 and Comment 49. Release QA will cover this through web compatibility, during Beta 53.

The initial issue is reproducible again starting with the 59.0a1 (2017-11-24) build: * last good revision: 4cdfb92b93e1c51402aac4a4637c4798f5358d25 * first bad revision: e02699fd3129bef049f848e2f71189182eabe555 * pushlog: https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=4cdfb92b93e1c51402aac4a4637c4798f5358d25&tochange=e02699fd3129bef049f848e2f71189182eabe555 [Affected versions]: - 61.0a1 (2018-03-14) - 60.0b3 build1 (20180312152746) - 59.0 build5 (20180310025718) [Affected platforms]: - Windows 10 x64 [Affected websites]: - https://www.reddit.com/ - http://store.steampowered.com/ [Note]: - The used GPU is Intel HD Graphics 4600

we are fixing the reoccurring issue in bug 1435472