919454 - Aurora (26.0a2 (2013-09-22)) whole display blur on scroll (mouse or pageDown)

Status: VERIFIED FIXED

(Core :: Graphics, defect)

Description

[felix.gerzaguet]

Attachment: Aurora_bugs_scroll_blur.PNG

User Agent: Mozilla/5.0 (Windows NT 5.1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/28.0.1500.95 Safari/537.36

Steps to reproduce:

1. Open any web page long enough to have a scroll bar (for example: https://bugzilla.mozilla.org/buglist.cgi?query_format=specific&order=relevance%20desc&bug_status=__open__&comments=0&list_id=7962787 )
2. Scroll down / up using the mouse the scrollbar or the keybord (pageDown/pageUp)



Actual results:

The page become very very difficult to read. Like if it was stretched... See the attached screenshot.


Expected results:

The page should remain readable.

Comment 1

[felix.gerzaguet]

After further tests, it appears that this issue does not happens in safe mode.
But it happens in normal mode with all extension manually disabled (I just use noscript and Microsoft .NET Framework assistant 1.0 which is incompatible with 26.0a2, so always disabled and cannot be enabled).

Comment 2

[Loic]

Common issue.

Comment 3

[Alice0775 White]

(In reply to Loic from comment #2)
> Common issue.
> 
> *** This bug has been marked as a duplicate of bug 812695 ***

This bug is for windows XP.
I think this is different bug.

Comment 4

[Loic]

SO HWA is involved. Disable HWA in advanced options and it's fixed.

Comment 5

[Alice0775 White]

Could you post contents of the Graphics section of about:support here?

Comment 6

[felix.gerzaguet]

Hi,
I tried to disable HWA (through the gfx.direct2d.disabled flag): it doesn't works.

Another piece of info: I cannot reproduce the bug with a textarea scroll bar. Only with the entire page scroll bar.

Here is the content of my about:support

{
  "application": {
    "name": "Firefox",
    "version": "26.0a2",
    "userAgent": "Mozilla/5.0 (Windows NT 5.1; rv:26.0) Gecko/20100101 Firefox/26.0",
    "supportURL": "https://support.mozilla.org/1/firefox/26.0a2/WINNT/en-US/"
  },
  "modifiedPreferences": {
    "accessibility.typeaheadfind.flashBar": 0,
    "browser.cache.disk.capacity": 358400,
    "browser.cache.disk.smart_size.first_run": false,
    "browser.cache.disk.smart_size.use_old_max": false,
    "browser.cache.disk.smart_size_cached_value": 358400,
    "browser.places.smartBookmarksVersion": 4,
    "browser.sessionstore.upgradeBackup.latestBuildID": "20130924004001",
    "browser.startup.homepage_override.mstone": "26.0a2",
    "browser.startup.homepage": "www.google.com",
    "browser.startup.homepage_override.buildID": "20130924004001",
    "dom.mozApps.used": true,
    "extensions.lastAppVersion": "26.0a2",
    "network.cookie.prefsMigrated": true,
    "network.negotiate-auth.trusted-uris": "airliquide.com",
    "network.automatic-ntlm-auth.trusted-uris": "airliquide.com",
    "places.history.expiration.transient_current_max_pages": 92997,
    "places.database.lastMaintenance": 1380095392,
    "plugin.importedState": true,
    "plugin.disable_full_page_plugin_for_types": "application/pdf",
    "print.printer_ScanSoft_PDF_Create!.print_downloadfonts": false,
    "print.printer_ScanSoft_PDF_Create!.print_to_file": false,
    "print.printer_ScanSoft_PDF_Create!.print_reversed": false,
    "print.printer_ScanSoft_PDF_Create!.print_headerleft": "&T",
    "print.printer_ScanSoft_PDF_Create!.print_paper_size_unit": 1,
    "print.printer_ScanSoft_PDF_Create!.print_bgimages": false,
    "print.printer_ScanSoft_PDF_Create!.print_unwriteable_margin_right": 0,
    "print.printer_ScanSoft_PDF_Create!.print_evenpages": true,
    "print.printer_ScanSoft_PDF_Create!.print_headercenter": "",
    "print.printer_ScanSoft_PDF_Create!.print_shrink_to_fit": true,
    "print.printer_ScanSoft_PDF_Create!.print_paper_name": "",
    "print.printer_ScanSoft_PDF_Create!.print_unwriteable_margin_left": 0,
    "print.printer_ScanSoft_PDF_Create!.print_edge_right": 0,
    "print.printer_ScanSoft_PDF_Create!.print_edge_left": 0,
    "print.printer_ScanSoft_PDF_Create!.print_paper_width": "  8,50",
    "print.printer_ScanSoft_PDF_Create!.print_unwriteable_margin_bottom": 0,
    "print.printer_ScanSoft_PDF_Create!.print_colorspace": "",
    "print.printer_ScanSoft_PDF_Create!.print_duplex": 2490417,
    "print.printer_ScanSoft_PDF_Create!.print_edge_bottom": 0,
    "print.printer_ScanSoft_PDF_Create!.print_headerright": "&U",
    "print.printer_ScanSoft_PDF_Create!.print_resolution_name": "",
    "print.printer_ScanSoft_PDF_Create!.print_page_delay": 50,
    "print.printer_ScanSoft_PDF_Create!.print_in_color": true,
    "print.printer_ScanSoft_PDF_Create!.print_resolution": 3145781,
    "print.printer_ScanSoft_PDF_Create!.print_margin_top": "0.5",
    "print.printer_ScanSoft_PDF_Create!.print_scaling": "  1,00",
    "print.printer_ScanSoft_PDF_Create!.print_bgcolor": false,
    "print.printer_ScanSoft_PDF_Create!.print_paper_size_type": 0,
    "print.printer_ScanSoft_PDF_Create!.print_command": "",
    "print.printer_ScanSoft_PDF_Create!.print_footerleft": "&PT",
    "print.printer_ScanSoft_PDF_Create!.print_paper_data": 9,
    "print.printer_ScanSoft_PDF_Create!.print_margin_left": "0.5",
    "print.printer_ScanSoft_PDF_Create!.print_unwriteable_margin_top": 0,
    "print.printer_ScanSoft_PDF_Create!.print_footerright": "&D",
    "print.printer_ScanSoft_PDF_Create!.print_paper_height": " 11,00",
    "print.printer_ScanSoft_PDF_Create!.print_orientation": 0,
    "print.printer_ScanSoft_PDF_Create!.print_edge_top": 0,
    "print.printer_ScanSoft_PDF_Create!.print_footercenter": "",
    "print.printer_ScanSoft_PDF_Create!.print_margin_right": "0.5",
    "print.printer_ScanSoft_PDF_Create!.print_oddpages": true,
    "print.printer_ScanSoft_PDF_Create!.print_margin_bottom": "0.5",
    "print.printer_ScanSoft_PDF_Create!.print_plex_name": "",
    "privacy.sanitize.migrateFx3Prefs": true,
    "storage.vacuum.last.places.sqlite": 1378109523,
    "storage.vacuum.last.index": 1
  },
  "graphics": {
    "numTotalWindows": 1,
    "numAcceleratedWindows": 1,
    "windowLayerManagerType": "Direct3D 9",
    "windowLayerManagerRemote": false,
    "adapterDescription": "Intel(R) 4 Series Internal Chipset",
    "adapterVendorID": "0x8086",
    "adapterDeviceID": "0x2e12",
    "adapterRAM": "Unknown",
    "adapterDrivers": "igxprd32",
    "driverVersion": "6.14.10.5082",
    "driverDate": "6-25-2009",
    "adapterDescription2": "",
    "adapterVendorID2": "",
    "adapterDeviceID2": "",
    "adapterRAM2": "",
    "adapterDrivers2": "",
    "driverVersion2": "",
    "driverDate2": "",
    "isGPU2Active": false,
    "direct2DEnabled": false,
    "directWriteEnabled": false,
    "directWriteVersion": "0.0.0.0",
    "direct2DEnabledMessage": [
      "blockedDriver"
    ],
    "webglRenderer": "Google Inc. -- ANGLE (Intel(R) 4 Series Internal Chipset Direct3D9 vs_3_0 ps_3_0)",
    "info": {
      "AzureCanvasBackend": "skia",
      "AzureSkiaAccelerated": 0,
      "AzureFallbackCanvasBackend": "cairo",
      "AzureContentBackend": "none"
    }
  },
  "javaScript": {
    "incrementalGCEnabled": true
  },
  "accessibility": {
    "isActive": false,
    "forceDisabled": 0
  },
  "libraryVersions": {
    "NSPR": {
      "minVersion": "4.10.1",
      "version": "4.10.1"
    },
    "NSS": {
      "minVersion": "3.15.2 Basic ECC Beta",
      "version": "3.15.2 Basic ECC Beta"
    },
    "NSSUTIL": {
      "minVersion": "3.15.2 Beta",
      "version": "3.15.2 Beta"
    },
    "NSSSSL": {
      "minVersion": "3.15.2 Basic ECC Beta",
      "version": "3.15.2 Basic ECC Beta"
    },
    "NSSSMIME": {
      "minVersion": "3.15.2 Basic ECC Beta",
      "version": "3.15.2 Basic ECC Beta"
    }
  },
  "userJS": {
    "exists": false
  },
  "extensions": [
    {
      "name": "Microsoft .NET Framework Assistant",
      "version": "1.0",
      "isActive": false,
      "id": "{20a82645-c095-46ed-80e3-08825760534b}"
    },
    {
      "name": "NoScript",
      "version": "2.6.8.1",
      "isActive": false,
      "id": "{73a6fe31-595d-460b-a920-fcc0f8843232}"
    }
  ]
}

Comment 7

[Alice0775 White]

Please check if the issue occurs using Firefox on a new empty profile:
http://support.mozilla.org/en-US/kb/Managing-profiles#w_starting-the-profile-manager .

Your graphic driver is blocked, however layers acceleration seems to be enabled.(I do not know why)

Comment 8

[felix.gerzaguet]

Attachment: 919454_about_support.txt

Comment 9

[felix.gerzaguet]

Attachment: 919454_new_empty_profile_about_support.txt

Comment 10

[felix.gerzaguet]

Attachment: 919454_ Aurora_bugs_scroll_blur_even_on_new_empty_profile.PNG

Comment 11

[felix.gerzaguet]

I reproduced the issue even on brand new (empty) profile. You can look at the [1] screenshot. I also provided the about:support of the empty profile [2].

[1] 919454_ Aurora_bugs_scroll_blur_even_on_new_empty_profile.PNG
[2] 919454_new_empty_profile_about_support.txt

Comment 12

[felix.gerzaguet]

Attachment: 919454_about_support_in_safe_mode_BUG_not_reproduced.txt

Comment 13

[felix.gerzaguet]

As per the new attachment [1], is seems to have something to do with "numAcceleratedWindows" ...

[1] 919454_about_support_in_safe_mode_BUG_not_reproduced.txt

Comment 14

[Alice0775 White]

I can reproduce the graphical glitch with the following driver spoofing on WindowsXP sp3.
SET MOZ_GFX_SPOOF_WINDOWS_VERSION=50001
SET MOZ_GFX_SPOOF_VENDOR_ID=0x8086
SET MOZ_GFX_SPOOF_DEVICE_ID=0x2e12
SET MOZ_GFX_SPOOF_DRIVER_VERSION=6.14.10.5082


Regression window(m-c)
Good:
http://hg.mozilla.org/mozilla-central/rev/a8daa428ccbc
Mozilla/5.0 (Windows NT 5.1; rv:26.0) Gecko/20100101 Firefox/26.0 ID:20130814141018
Bad:
http://hg.mozilla.org/mozilla-central/rev/31c08ca022b3
Mozilla/5.0 (Windows NT 5.1; rv:26.0) Gecko/20100101 Firefox/26.0 ID:20130815035619
Pushlog:
http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=a8daa428ccbc&tochange=31c08ca022b3

Regression window(cached m-i)
Good:
http://hg.mozilla.org/integration/mozilla-inbound/rev/ac3f07dc2ff3
Mozilla/5.0 (Windows NT 5.1; rv:26.0) Gecko/20100101 Firefox/26.0 ID:20130814083322
Bad:
http://hg.mozilla.org/integration/mozilla-inbound/rev/1fa1f28c6edf
Mozilla/5.0 (Windows NT 5.1; rv:26.0) Gecko/20100101 Firefox/26.0 ID:20130814101421
Pushlog:
http://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=ac3f07dc2ff3&tochange=1fa1f28c6edf

Suspected:
2241fd070f12	Jeff Muizelaar — Bug 904266. Relax intel blacklist on D3D9 more. r=bjacob We've been using the 7.15.10.1666 IntelGMAX3000 driver on Vista so let's use that everywhere. This should cover the majority of users.

Comment 15

[Benoit Jacob [:bjacob] (mostly away)]

Jeff, should we re-blacklist?

Comment 16

[Benoit Jacob [:bjacob] (mostly away)]

Also worth asking Bas.

Comment 17

[Bas Schouten (:bas.schouten)]

I don't think we should immediately reblacklist if we can't reproduce this on other, similar hardware. This might indicate a hardware bug with this specific instance of the hardware considering how exotic the symptoms are.. The affected user should be able to disable HWA completely through Tools->Options->Advanced and be able to fix the problem for him.

Comment 18

[felix.gerzaguet]

Indeed, if I disable the HWA through the above stated option, I do not have the BUG  anymore.
This solution is fine by me. It is perfectly possible that my workstation is corrupted somehow, but since I use Aurora, I suspected the FF software (sorry If I was wrong).
I will use the workaround for now.
Thanks.

Comment 19

[Lukas Blakk [:lsblakk] use ?needinfo]

Milan - can you find someone to take this?

Comment 20

[Milan Sreckovic [:milan] (needinfo for best results)]

Based on comment 14, we should back out 904266 until we figure this out.  Bas, did your comment 17 consider comment 14? Is there a way to get the statistics as to how often the comment 14 setup shows up in the wild?

If we are to back it out, Benoit can you get the patch for it or ask for it?  As a reviewer :), since Jeff is gone for another week and a half.

Comment 21

[Bas Schouten (:bas.schouten)]

(In reply to Milan Sreckovic [:milan] from comment #20)
> Based on comment 14, we should back out 904266 until we figure this out. 
> Bas, did your comment 17 consider comment 14? Is there a way to get the
> statistics as to how often the comment 14 setup shows up in the wild?
> 
> If we are to back it out, Benoit can you get the patch for it or ask for it?
> As a reviewer :), since Jeff is gone for another week and a half.

I didn't completely understand that comment. If we spoof a driver version sure we'll see bugs.. The question is whether there's more machines out there that experience this bug -without- having spoofed a driver version. So far the bug reporter has been the only person to report this on a clean profile without spoofing. I believe that by itself is not sufficient reason to turn -many- people off.

I do believe we should put some effort into finding out if other people are having this issue and simply not reporting it. I reiterate my belief that we should simply try and see if the bug reproduces on a setup similar to that of the reporter.

This looks a lot like a hardware bug or something we can workaround. In the latter case it should be more common and we should be able to find a machine that reproduces it and see if it's feasible to address, in the former case we should not do anything to the blacklist.

Alternatively we could backout the blacklisting change and see how fixed function D3D9 OMTC layers do on these machines.

Comment 22

[Jeff Muizelaar [:jrmuizel]]

(In reply to Alice0775 White from comment #14)
> I can reproduce the graphical glitch with the following driver spoofing on
> WindowsXP sp3.
> SET MOZ_GFX_SPOOF_WINDOWS_VERSION=50001
> SET MOZ_GFX_SPOOF_VENDOR_ID=0x8086
> SET MOZ_GFX_SPOOF_DEVICE_ID=0x2e12
> SET MOZ_GFX_SPOOF_DRIVER_VERSION=6.14.10.5082


Alice, what's the actual vendor/device id and driver version on your machine that reproduces this?

Comment 23

[Jeff Muizelaar [:jrmuizel]]

We need to figure out how widespread this. QA can you reproduce this on any similar hardware that you have?

Comment 24

[Alice0775 White]

(In reply to Jeff Muizelaar [:jrmuizel] from comment #22)
> (In reply to Alice0775 White from comment #14)
> > I can reproduce the graphical glitch with the following driver spoofing on
> > WindowsXP sp3.
> > SET MOZ_GFX_SPOOF_WINDOWS_VERSION=50001
> > SET MOZ_GFX_SPOOF_VENDOR_ID=0x8086
> > SET MOZ_GFX_SPOOF_DEVICE_ID=0x2e12
> > SET MOZ_GFX_SPOOF_DRIVER_VERSION=6.14.10.5082
> 
> 
> Alice, what's the actual vendor/device id and driver version on your machine
> that reproduces this?

WindowsXP xp3 on VMwarePlayer hosted Windows7x64.

Graphics
--------
Adapter Description: VMware SVGA II
Adapter Drivers: vmx_fb
Adapter RAM: Unknown
Device ID: 0x0405
Direct2D Enabled: Blocked for your graphics card because of unresolved driver issues.
DirectWrite Enabled: false (0.0.0.0)
Driver Date: 7-15-2013
Driver Version: 12.0.25.0
GPU #2 Active: false
GPU Accelerated Windows: 0/1 Basic Blocked for your graphics card because of unresolved driver issues.
Vendor ID: 0x15ad
WebGL Renderer: Blocked for your graphics card because of unresolved driver issues.
windowLayerManagerRemote: false
AzureCanvasBackend: skia
AzureContentBackend: cairo
AzureFallbackCanvasBackend: cairo
AzureSkiaAccelerated: 0

Comment 25

[Catalin Varga [QA][:VarCat]]

The bug is not reproducible using the following environment:

Latest Aurora:
Build identifier: Mozilla/5.0 (Windows NT 5.1; rv:26.0) Gecko/20100101 Firefox/26.0
Build Id:20131028004004


Latest Nightly:
Build identifier: Mozilla/5.0 (Windows NT 5.1; rv:27.0) Gecko/20100101 Firefox/27.0
Build Id:20131028030205


OS: Windows Xp x86 SP3

Graphics:

Adapter Description	Intel(R) G41 Express Chipset
Adapter Drivers	igxprd32
Adapter RAM	Unknown
Device ID	0x2e32
Direct2D Enabled	Blocked for your graphics driver version.
DirectWrite Enabled	false (0.0.0.0)
Driver Date	4-21-2010
Driver Version	6.14.10.5259
GPU #2 Active	false
GPU Accelerated Windows	1/1 Direct3D 9
Vendor ID	0x8086
WebGL Renderer	Google Inc. -- ANGLE (Intel(R) G41 Express Chipset Direct3D9 vs_3_0 ps_3_0)
windowLayerManagerRemote	false
AzureCanvasBackend	skia
AzureContentBackend	none
AzureFallbackCanvasBackend	cairo
AzureSkiaAccelerated	0

Comment 26

[Jeff Muizelaar [:jrmuizel]]

(In reply to Alice0775 White from comment #24)

> 
> WindowsXP xp3 on VMwarePlayer hosted Windows7x64.
> 

What's the hardware on your Windows 7 machine?

Comment 27

[Alice0775 White]

 VMware Player6.0.1 virtual machine

Comment 28

[Jeff Muizelaar [:jrmuizel]]

(In reply to Alice0775 White from comment #27)
>  VMware Player6.0.1 virtual machine

I meant "what hardware is the vm running on?"

Comment 29

[Alice0775 White]

(In reply to Jeff Muizelaar [:jrmuizel] from comment #28)
> (In reply to Alice0775 White from comment #27)
> >  VMware Player6.0.1 virtual machine
> 
> I meant "what hardware is the vm running on?"
ATI HD4350 ?

Comment 30

[Paul Silaghi, QA [:pauly]]

FWIW, this is also not reproducible on AMD Radeon HD3000, nVIDIA GeForce 8400 GS, Win XP, latest nightly

Comment 31

[Jeff Muizelaar [:jrmuizel]]

(In reply to Catalin Varga [QA][:VarCat] from comment #25)
> The bug is not reproducible using the following environment:

> Driver Version	6.14.10.5259


Can you downgrade this driver to 6.14.10.5082 and try again?

Comment 32

[Alice0775 White]

I think this is reproducible on blocked driver but enabled D3D9 accentuation.

Comment 33

[Catalin Varga [QA][:VarCat]]

(In reply to Jeff Muizelaar [:jrmuizel] from comment #31)
> Can you downgrade this driver to 6.14.10.5082 and try again?

I've performed the downgrade to 6.14.10.5082 and I managed to reproduce the bug on the Latest Aurora and Latest Nightly.

Environment used:

Latest Aurora

Build identifier: Mozilla/5.0 (Windows NT 5.1; rv:27.0) Gecko/20100101 Firefox/27.0
Build Id:20131031004003

Latest Nightly
Build identifier: Mozilla/5.0 (Windows NT 5.1; rv:28.0) Gecko/20100101 Firefox/28.0
Build Id:20131031030203

OS: Windows Xp x86 SP3

Graphics
Adapter Description	Intel(R) G41 Express Chipset
Adapter Drivers	igxprd32
Adapter RAM	Unknown
Device ID	0x2e32
Direct2D Enabled	Blocked for your graphics driver version.
DirectWrite Enabled	false (0.0.0.0)
Driver Date	6-25-2009
Driver Version	6.14.10.5082
GPU #2 Active	false
GPU Accelerated Windows	1/1 Direct3D 9
Vendor ID	0x8086
WebGL Renderer	Google Inc. -- ANGLE (Intel(R) G41 Express Chipset Direct3D9 vs_3_0 ps_3_0)
windowLayerManagerRemote	false
AzureCanvasBackend	skia
AzureContentBackend	cairo
AzureFallbackCanvasBackend	cairo
AzureSkiaAccelerated	0

Comment 34

[Jeff Muizelaar [:jrmuizel]]

(In reply to Catalin Varga [QA][:VarCat] from comment #33)
> (In reply to Jeff Muizelaar [:jrmuizel] from comment #31)
> > Can you downgrade this driver to 6.14.10.5082 and try again?
> 
> I've performed the downgrade to 6.14.10.5082 and I managed to reproduce the
> bug on the Latest Aurora and Latest Nightly.

Awesome.

Can you bisect to find out which driver version fixes the problem?

Comment 35

[Catalin Varga [QA][:VarCat]]

Unfortunately very few older version of the driver are available on the internet. Best window I could find with the available drivers is:

Last Bad:
6.14.10.5157

First Good:
6.14.10.5218

Comment 36

[Jeff Muizelaar [:jrmuizel]]

Ok, we should probably move the black list back to 5218. It would also be good to see if we can work around the problem.

Catlin or felix what are the manufacturers and models of the systems you can reproduce this on?

Comment 37

[felix.gerzaguet]

I will not have access to the faulty computer until next Wednesday. I will let you know then.

Comment 38

[Benoit Jacob [:bjacob] (mostly away)]

I'll un-assign myself to reflect the reality that I haven't been working on this, so I won't get release tracking alerts anymore. Jeff, since apparently you're the only one to have been looking into this, do you want to self-assign?

Comment 39

[Milan Sreckovic [:milan] (needinfo for best results)]

Jeff tracked down the hardware as well.

Comment 40

[Catalin Varga [QA][:VarCat]]

(In reply to Jeff Muizelaar [:jrmuizel] from comment #36)
> Ok, we should probably move the black list back to 5218. It would also be
> good to see if we can work around the problem.
> 
> Catlin or felix what are the manufacturers and models of the systems you can
> reproduce this on?

System Information
------------------
Operating System: Windows XP Professional (5.1, Build 2600) Service Pack 3 (2600.xpsp_sp3_qfe.130704-0421)
Language: English (Regional Setting: English)
System Manufacturer: Gigabyte Technology Co., Ltd.
System Model: G41MT-S2P
BIOS: Award Modular BIOS v6.00PG
Processor: Intel(R) Core(TM)2 Duo CPU     E7500  @ 2.93GHz (2 CPUs)
Memory: 3292MB RAM
Page File: 568MB used, 4607MB available
Windows Dir: D:\WINDOWS
DirectX Version: DirectX 9.0c (4.09.0000.0904)
DX Setup Parameters: Not found
DxDiag Version: 5.03.2600.5512 32bit Unicode


Display Devices
---------------
Card name: Intel(R) G41 Express Chipset
Manufacturer: Intel Corporation
Chip type: Intel(R) 4 Series Express Chipset Family
DAC type: Internal
Device Key: Enum\PCI\VEN_8086&DEV_2E32&SUBSYS_D0001458&REV_03
Display Memory: 128.0 MB
Current Mode: 1680 x 1050 (32 bit) (60Hz)
Vendor ID: 0x8086
Device ID: 0x2E32

Comment 41

[felix.gerzaguet]

System Information
------------------
Time of this report: 11/6/2013, 14:35:38
       Machine name: FR-W-EPS0055611
   Operating System: Windows XP Professional (5.1, Build 2600) Service Pack 3 (2600.xpsp_sp3_qfe.120504-1617)
           Language: English (Regional Setting: French)
System Manufacturer: Dell Inc.
       System Model: OptiPlex 780                 
               BIOS: Phoenix ROM BIOS PLUS Version 1.10 A08
          Processor: Pentium(R) Dual-Core  CPU      E5800  @ 3.20GHz (2 CPUs)
             Memory: 3548MB RAM
          Page File: 2005MB used, 3424MB available
        Windows Dir: C:\WINDOWS
    DirectX Version: DirectX 9.0c (4.09.0000.0904)
DX Setup Parameters: Not found
     DxDiag Version: 5.03.2600.5512 32bit Unicode


Display Devices
---------------
        Card name: Intel(R) 4 Series Internal Chipset
     Manufacturer: Intel Corporation
        Chip type: Intel(R) 4 Series Express Chipset Family
         DAC type: Internal
       Device Key: Enum\PCI\VEN_8086&DEV_2E12&SUBSYS_04201028&REV_03
   Display Memory: 1024.0 MB
     Current Mode: 1280 x 1024 (32 bit) (60Hz)
          Monitor: Plug and Play Monitor
  Monitor Max Res: 1600,1200
      Driver Name: igxprd32.dll
   Driver Version: 6.14.0010.5082 (English)
      DDI Version: 9 (or higher)
Driver Attributes: Final Retail
 Driver Date/Size: 6/25/2009 08:09:14, 57344 bytes
         Mini VDD: igxpmp32.sys
    Mini VDD Date: 6/25/2009 08:09:16, 6316160 bytes
        Vendor ID: 0x8086
        Device ID: 0x2E12
        SubSys ID: 0x04201028
      Revision ID: 0x0003
      Revision ID: 0x0003

Comment 42

[Jeff Muizelaar [:jrmuizel]]

My guess is that this is caused by our StretchRect call. We've been bitten by StretchRect bugs before it might be good for us to avoid using it. Any thoughts Bas?

Comment 43

[Jeff Muizelaar [:jrmuizel]]

Seems like we can use UpdateSurface instead.

Comment 44

[Jeff Muizelaar [:jrmuizel]]

Nope. That won't work because both surfaces are D3DPOOL_DEFAULT

Comment 45

[Jeff Muizelaar [:jrmuizel]]

I was able to narrow the range down some:
Bad: 6.14.10.5157
Good: 6.14.10.5160

Comment 46

[Jeff Muizelaar [:jrmuizel]]

Drivers before 6.14.10.5076 also seem to work.

Comment 47

[Jeff Muizelaar [:jrmuizel]]

Attachment: Add a rule to catch broken StretchRect()

Comment 48

[Jeff Muizelaar [:jrmuizel]]

Comment on attachment 828392 [details] [diff] [review]
Add a rule to catch broken StretchRect()

[Approval Request Comment]
Bug caused by (feature/regressing bug #): 904266
User impact if declined: very broken rendering
Testing completed (on m-c, etc.): This removes some block list allowences.
Risk to taking this patch (and alternatives if risky): Very low

Comment 49

[Benoit Jacob [:bjacob] (mostly away)]

Comment on attachment 828392 [details] [diff] [review]
Add a rule to catch broken StretchRect()

Review of attachment 828392 [details] [diff] [review]:
-----------------------------------------------------------------

::: widget/windows/GfxInfo.cpp
@@ +909,5 @@
> +    // StrechRect seems to suffer from precision issues which leads to artifacting
> +    // during content drawing starting with at least version 6.14.10.5082
> +    // and going until 6.14.10.5160. See bug 919454 for more info.
> +    APPEND_TO_DRIVER_BLOCKLIST_RANGE(DRIVER_OS_WINDOWS_XP,
> +      (nsAString&) GfxDriverInfo::GetDeviceVendor(VendorIntel), (GfxDeviceFamily*) GfxDriverInfo::GetDeviceFamily(IntelGMAX4500HD),

The (GfxDeviceFamily*) should be a const_cast<GfxDeviceFamily*>(...). That would be a lot safer.

Looking at GfxDriverInfo.h, the only reason for this casting business seems to be that the GfxDriverInfo ctor takes pointers/refs to non-const, and it should instead take pointers/refs to const. Please consider fixing it! R+!

Comment 50

[Jeff Muizelaar [:jrmuizel]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/5c4ae2a331c4

Comment 51

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/5c4ae2a331c4

Comment 52

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/releases/mozilla-aurora/rev/724a412d965d
https://hg.mozilla.org/releases/mozilla-beta/rev/be23143070b8

Comment 53

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/releases/mozilla-b2g26_v1_2/rev/be23143070b8

Comment 54

[Bas Schouten (:bas.schouten)]

(In reply to Jeff Muizelaar [:jrmuizel] from comment #42)
> My guess is that this is caused by our StretchRect call. We've been bitten
> by StretchRect bugs before it might be good for us to avoid using it. Any
> thoughts Bas?

There's somewhat of a lack of alternatives :s In D3D9 that is. We could create our own version that draws explicitly but it'll very likely be slower.

Comment 55

[Catalin Varga [QA][:VarCat]]

I managed to reproduce this bug in the past by using a Intel(R) G41 Express Chipset graphics card.

This bug can be reproduced by using one of the following graphics cards:  Intel® G31, G33, G35, G41, G43, G45, Q33, Q35, Q43, and Q45.

My graphics card was upgraded to ATI Radeon HD 5450 so I can't verify this bug anymore.
No one from our team is currently using a graphics card suitable for this regression.
If the reporter or anyone that has a suitable graphics card can verify this bug, that would be great.

Comment 56

[felix.gerzaguet]

Hi,
I can verify that my faulty driver driver is correctly blacklisted. Here is en excerpt of my about:support:

Adapter Description: Intel(R) 4 Series Internal Chipset
Adapter Drivers: igxprd32
Adapter RAM: Unknown
Device ID: 0x2e12
Direct2D Enabled: Blocked for your graphics driver version.
DirectWrite Enabled: false (0.0.0.0)
Driver Date: 6-25-2009
Driver Version: 6.14.10.5082
GPU #2 Active: false
GPU Accelerated Windows: 0/1 Basic Blocked for your graphics driver version. Try updating your graphics driver to version 6.14.10.5160 or newer.
Vendor ID: 0x8086
WebGL Renderer: Blocked for your graphics driver version. Try updating your graphics driver to version 6.14.10.5160 or newer.
windowLayerManagerRemote: false
AzureCanvasBackend: skia
AzureContentBackend: cairo
AzureFallbackCanvasBackend: cairo
AzureSkiaAccelerated: 0

For me the issue is fixed.

Comment 57

[Ioana (away)]

Removing the keyword since QA can't do anything here.

Thanks for verifying this felix! Could you please let us know what Firefox version you did the verification on?

Comment 58

[felix.gerzaguet]

I used the Aurora version of that day (so probably 27.0a2 (2013-11-25)).

Comment 59

[Ioana (away)]

(In reply to felix.gerzaguet from comment #58)
> I used the Aurora version of that day (so probably 27.0a2 (2013-11-25)).

Thanks. Updating flags per this information.

Comment 61

[u279076]

Felix, can you please check this is fixed for you in Firefox 28 (aurora) and Firefox 26 (release)?

Comment 62

[Loic]

Is it supposed to be fixed in FF26?
We have 3 reports about that issue on the French support forum.
Both with Intel chipsets:
Intel(R) G41 Express Chipset (driver 1-12-2010)
Intel(R) HD Graphics 4000 (driver 10-1-2013)

Comment 63

[Jeff Muizelaar [:jrmuizel]]

(In reply to Loic from comment #62)
> Is it supposed to be fixed in FF26?
> We have 3 reports about that issue on the French support forum.
> Both with Intel chipsets:
> Intel(R) G41 Express Chipset (driver 1-12-2010)
> Intel(R) HD Graphics 4000 (driver 10-1-2013)

It looks like there are some driver versions that are still affected. Can you get driver versions for the reporters of this issue?

Comment 64

[felix.gerzaguet]

(In reply to Anthony Hughes, QA Mentor (:ashughes) from comment #61)
> Felix, can you please check this is fixed for you in Firefox 28 (aurora) and
> Firefox 26 (release)?

I confirm that for me, on my faulty hardware (driver 6.14.10.5082), the bug is fixed with Firefox 26 and with Firefox 28.0a2 (2013-12-17)

Comment 65

[Loic]

One of users said me upgrading the driver of his Intel chipset has fixed the issue.

Comment 66

[u279076]

I would say let's call this verified fixed since it's been proven to resolve the issue (comment 64). We should enlist SUMO to assist users who still encounter this in updating their drivers. We should address any edge cases which are not resolved by either in follow-up bugs.

Comment 67

[Loic]

(In reply to Jeff Muizelaar [:jrmuizel] from comment #63)
> (In reply to Loic from comment #62)
> > Is it supposed to be fixed in FF26?
> > We have 3 reports about that issue on the French support forum.
> > Both with Intel chipsets:
> > Intel(R) G41 Express Chipset (driver 1-12-2010)
> > Intel(R) HD Graphics 4000 (driver 10-1-2013)
> 
> It looks like there are some driver versions that are still affected. Can
> you get driver versions for the reporters of this issue?

For the record, this user upgraded the driver but that didn't solve the issue.
His about:support details:

Date du pilote 10-1-2013
Description de la carte Intel(R) HD Graphics 4000
Direct2D activé true
DirectWrite activé true (6.3.9600.16384)
Fenêtres avec accélération graphique 1/1 Direct3D 10
GPU #2 active false
ID du périphérique 0x0166
ID du vendeur 0x8086
Pilotes de la carte igdumdim64 igd10iumd64 igd10iumd64 igdumdim32 igd10iumd32 igd10iumd32
RAM de la carte Unknown
Rendu WebGL Google Inc. -- ANGLE (Intel(R) HD Graphics 4000 Direct3D9Ex vs_3_0 ps_3_0)
Version du pilote 10.18.10.3316
windowLayerManagerRemote false
AzureCanvasBackend direct2d
AzureContentBackend direct2d
AzureFallbackCanvasBackend cairo
AzureSkiaAccelerated 0