Intel® Graphics Media Accelerator
Production Version 14.32.3.4906
Release Notes
December 28, 2007
Microsoft Windows* XP
INTEL
CONFIDENTIAL
DISCLAIMER: INFORMATION IN THIS DOCUMENT IS PROVIDED IN CONNECTION WITH
INTEL® PRODUCTS. NO LICENSE, EXPRESS OR IMPLIED, BY ESTOPPEL OR OTHERWISE, TO
ANY INTELLECTUAL PROPERTY RIGHTS IS GRANTED BY THIS DOCUMENT. EXCEPT AS
PROVIDED IN INTEL'S TERMS AND CONDITIONS OF SALE FOR SUCH PRODUCTS, INTEL
ASSUMES NO LIABILITY WHATSOEVER, AND INTEL DISCLAIMS ANY EXPRESS OR IMPLIED
WARRANTY, RELATING TO SALE AND/OR USE OF INTEL PRODUCTS INCLUDING LIABILITY OR
WARRANTIES RELATING TO FITNESS FOR A PARTICULAR PURPOSE, MERCHANTABILITY, OR
INFRINGEMENT OF ANY PATENT, COPYRIGHT OR OTHER INTELLECTUAL PROPERTY RIGHT.
UNLESS OTHERWISE AGREED IN WRITING BY INTEL, THE INTEL PRODUCTS ARE NOT DESIGNED NOR INTENDED FOR ANY APPLICATION IN WHICH THE FAILURE OF THE INTEL PRODUCT COULD CREATE A SITUATION WHERE PERSONAL INJURY OR DEATH MAY OCCUR.
Intel may make changes to specifications and product descriptions at any time, without notice. Designers must not rely on the absence or characteristics of any features or instructions marked "reserved" or "undefined." Intel reserves these for future definition and shall have no responsibility whatsoever for conflicts or incompatibilities arising from future changes to them. The information here is subject to change without notice. Do not finalize a design with this information.
The products described in this document may contain design defects or errors known as errata which may cause the product to deviate from published specifications. Current characterized errata are available on request.
Contact your local Intel sales office or your distributor to obtain the latest specifications and before placing your product order.
This document contains information on products in the design phase of development.
All products, platforms, dates, and figures specified are preliminary based on current expectations, and are subject to change without notice. All dates specified are target dates, are provided for planning purposes only and are subject to change.
This document contains information on products in the design phase of development. Do not finalize a design with this information. Revised information will be published when the product is available. Verify with your local sales office that you have the latest datasheet before finalizing a design.
Code names featured are used internally within Intel to identify products that are in development and not yet publicly announced for release. Customers, licensees and other third parties are not authorized by Intel to use code names in advertising, promotion or marketing of any product or services and any such use of Intel's internal code names is at the sole risk of the user.
Intel is a trademark of Intel
Corporation in the
* Other names and brands may be claimed
as the property of others.
Copyright © 2007, Intel Corporation. All rights reserved.
Date |
Driver Revision |
Description |
Build Number |
December 31, 2007 |
14.32.3.4906 |
Production Version; fully validated |
4906 |
November 20, 2007 |
14.32.2.4897 |
Production Version; fully validated |
4897 |
November 8, 2007 |
14.32.1.4889 |
Production Version; fully validated |
4889 |
November 8, 2007 |
14.32.0.4885 |
Production Version; fully validated |
4885 |
October 18, 2007 |
14.32.0.4879 |
Production Candidate; not fully validated |
4879 |
October 8, 2007 |
14.32.0.4876 |
Beta engineering drop; not fully validated |
4876 |
September 13, 2007 |
14.32.0.4869 |
Beta engineering drop; not fully validated |
4869 |
August 31, 2007 |
14.32.0.4861 |
Beta engineering drop; not fully validated |
4861 |
Intel® Graphics Media Accelerator Driver
The Intel® Graphics Media Accelerator Driver contains support for the following Intel Chipsets:
·
Intel® 945G Express Chipset
·
Intel® 945GZ Express Chipset
·
Intel® 946GZ Express Chipset
·
Intel® G33 Express Chipset
·
Intel® G35 Express Chipset
·
Intel® G965 Express Chipset
·
Intel® Q33 Express Chipset
·
Intel® Q33/Q35/G33 Express Chipsets
·
Intel® Q35 Express Chipset
·
Intel® Q963 Express Chipset
·
Intel® Q965 Express Chipset
·
Mobile Intel® 940GML Express Chipset
·
Mobile Intel® 945GM Express Chipset
·
Mobile Intel® 945GME Express Chipset
·
Mobile Intel® 945GMS Express Chipset
·
Mobile Intel® GL960 Express Chipset
·
Mobile Intel® GLE960 Express Chipset
·
Mobile Intel® GM965 Express Chipset
· Mobile Intel® GME965 Express Chipset
New Features
Features Added in Baseline 14.32 |
|||
Reference No. |
Description |
Affected Component(s) |
Build |
1003209 |
Provide 'scan <interlaced/progressive>' select option for mode removal in VBT |
Soft BIOS |
4861 |
1006224 |
Extending current Windows* XP mode persistence and first boot requirements to Windows Vista* and desktop chipsets |
CUI2 NT-miniport |
4869 |
1010359 |
Adding Open GL extension, Texture Rectangle to OpenGL 1.5 |
Open GL* |
4879 |
Known Issues
Known Issues in Build 4906 |
||||
Reference No. |
Description |
Affected Component(s) |
Affected OS(s) |
Affected Project(s) |
2554877 |
InterVideo WinDVD8
- The error message is displayed on InterVideo WinDVD8
application while playing the DVD Test Annex 3.0 DVD movie. |
N/A |
|
Intel® 3 Series Chipset Family B |
2554668 |
TVWSetup - The
warning message is not displayed after launching the TV Wizard application via
WINDOWS MEDIA* CENTER with CRT+Digital TV DDC mode. |
TV Wizard |
Windows |
Intel® 3 Series Chipset Family B |
N/A |
The blank screen is displayed on TV with CRT+TV EXTD mode
after hot plugging TV device. |
Resource Manager |
Windows* XP |
Intel® 3 Series Chipset Family-B |
2554440 |
The blank screen is displayed on all devices after
executing the command as dobios 5f64 0 8. |
Video BIOS |
DOS 6.22 |
Cantiga |
2553493 |
cmd, Ruby_Setup32,
eTestSDK_2_3_32-bit, dotnetfx, DOTNETFX1, Install.bat - The result of log file is failed after
executing the SDVOPowerDown_VBTDisable.bat file. |
Soft BIOS |
Windows Vista* Windows |
Cantiga |
2553179 |
Fprog, BMP, Mergetool,
dotnetfx3, TVWSetup, |
TV Wizard |
|
Cantiga |
2552995 |
BMP, Mergetool, fwhflash - Display
switch very slowly after resuming the system at LFP+TV DDC then pressing
Ctrl+Alt+Shift+F1 combinations keys. |
NT - miniport |
Windows* XP |
Intel® GM965 Express Chipset |
2550588 |
dotnetfx3, TVWSetup, |
TV Wizard |
Windows* XP |
Intel® 3 Series Chipset Family |
2554877 |
InterVideo WinDVD8
- The error message is displayed on InterVideo
WinDVD8 application while playing the DVD Test Annex 3.0 DVD movie. |
N/A |
|
Intel® 3 Series Chipset Family B |
2554668 |
TVWSetup - The
warning message is not displayed after launching the TV Wizard application
via WINDOWS MEDIA* CENTER with CRT+Digital TV DDC
mode. |
TV Wizard |
Windows |
Intel® 3 Series Chipset Family B |
|
The blank screen is displayed on TV with CRT+TV EXTD mode
after hot plugging TV device. |
Resource Manager |
Windows* XP |
Intel® 3 Series Chipset Family-B |
2554440 |
The blank screen is displayed on all devices after
executing the command as dobios 5f64 0 8. |
Video BIOS |
DOS 6.22 |
Cantiga |
Issues Resolved
Issues Resolved in Build 4906 |
|||||||||||||||||||||||||||||
Reference No. |
Description |
Affected Component(s) |
Affected OS(s) |
Affected Project(s) |
|||||||||||||||||||||||||
BugID: N/A |
Functional Failure / SoftBIOS - Display goes black when
switching to ED TV+CRT or DFP+CRT in Windows* 2000 |
NT - miniport, |
Windows* 2000 |
Intel® 945GM Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: The issue was happening as a side effect of bug fix for 2140130.
Basically with that code change we had removed all the code that was
modifying OS referred registry keys to update the devices' optimal modes.
This was because once system comes up, at desktop, CUI will set the optimal
mode. However, in these modifications, there was one modification of registry
key related to 'Attach.toDesktop' which was necessary in Windows2K platform
for EDS display switching to work properly through CUI. Since this got
removed with the changes, issue was happening. FIX:The fix is to set this key
during boot on Windows 2K only. Apart from this another registry modification
for Windows* XP related to rotation and upgrade(old bug) has also been put
back though we don't have a bug raised on it yet. ETM was done for a driver from 14.32 branch
itself though since we did not get checkin permission for this, we used
another DCN to checkin to main and that DCN was updated as |
|||||||||||||||||||||||||||||
BugID: 2551012 |
LCD resolution change when monitor move to clone mode by
Fn+F3 hotkey. |
NT - miniport |
Tablet Pc |
Intel® 945GM |
|||||||||||||||||||||||||
Resolution Description: Root Cause The issue is happening inspite of driver reporting Clone
EDID properly. Looks like an OS issue and a WA is being provided currently
for 14.32 branch. For main, more investigation will be done before porting
the fix. FIX: To fix the issue in 14.32
we are updating OS registry keys for Clone device with the optimal mode as
and when the event triggering the switch is ACPI Hotkeys or lid close or
open. For boot, CUI would anyway override the OS set mode. |
|||||||||||||||||||||||||||||
BugID: 2553138 |
Visual Anomaly / Direct3D* - Rainbow Six Vegas - Corruption
during gameplay in the form of flickering walls |
GHAL3D |
Windows Vista* |
Intel® G965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Fixed the way the ReverseBindings table is created: 1.
First we look for a binding that is a declared Pixel Shader input 2. If none of
found bindigs is a declared Pixel Shader input we take any bindig that we
found in VertexElementBinding table 3. If we didn't find and binding in VEB
we set a binding value to be equal to the table index (that points to this
value - i.e. VertexBinding[i] = i. In
that way we handle cases when: 1. There is more then one binding for clipper
input (VS output) in VertexElementBinding table and only one of them is a
valid PS input. 2. Some clipper inputs do not have corresponding PS input
(e.g. Position) 3. There is no binding at all the the given element number |
|||||||||||||||||||||||||||||
BugID: 2552086 |
Miniport - Test ID 11204 - BMP, Mergetool, Fwhflash -
System does not boot up in the OEM defined mode on legacy CRT with modified
BIOS. |
CUI2 |
Windows* XP |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: With Fresh Install of the driver with OEM defaul values set
in VBT, the display was not coming up in OEM default values. It was coming
with default values 800X600. Rootcause : this was happening because in CUI
the flag CUI_FIRST_BOOT was set before zeromory of Validate Advise config
structure. This flag was never getting set. Also in the COM layer , when
ValAdvcfgMode is called , the above mentioned flag was not due to the reason
, it was checking whether Resolution XX was Zero , which Will be true in this case, Since it’s a fresh
install and No history was saved to load. Fix: Flag set is done after ZeroMemory of
ValidateAdsviseCfg. Flag set is also changed in GetvalidateAdviseCfgMode ( ) |
|||||||||||||||||||||||||||||
BugID: 2482563 |
The display corruption on LFP while applying different
video std. |
NT - miniport |
Windows* XP |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: |
|||||||||||||||||||||||||||||
BugID: N/A |
Windows* XP : Desktop ICON/Font is not sharp (blurred) on
1680x1050 resolution. |
Soft BIOS, |
Windows* XP |
Cantiga |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Panel fitter was not getting disabled, as pipe reprogram
optimization is applied Fix: Now the check is done in function IsPipeDPLLReprogramRequired()
in Gen4pipe.c, if the applied mode is same as native mode of the LFP panel
and the optimization is suppressed, so panel fitter gets disabled. The
GDIModeInfo struct is now passed as an argument to the
IsPipeDPLLReprogramRequired(). File
Modified: Gen4pipe.c IntLvdsEncoder.c
Pipebase.c Pipebase.h Config
Spec: element * CHECKEDOUT element *
.../DEV_panel_fitter_bug/LATEST element * /main/LATEST -mkbranch
DEV_panel_fitter_bug load
\gfx_Development |
|||||||||||||||||||||||||||||
BugID: N/A |
Functional Failure / Miniport - Test ID 36774 - Desktop
resolution is 1024x768x32bit@ 60Hz after installing the graphics driver using
the command : \setup -res 800x600x32x60 and then restart system. |
CUI2,NT - miniport |
Windows 2003 |
Intel® 945GM Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: The issue was happening because on a factory install,
installer app was writing the factory keys to normal device0 location of
driver. After system boot, CUI would send out a requrest for reading these
registries. For Driver, since this is a standard path which OS has given,
driver will use OS VideoPort calls to read and write registries supplying the
key name without giving any path information. This is because VideoPort is
aware of the path being assigned to the driver. In 32 bit, Videoport calls
also refer to the standard Device0 path and so we saw no issues. However in
64 bit systems, VideoPort refers to guid based paths to access registries
which are not conventional. So, CUI request to read factory keys were failing
in 64 bit and so was the issue. FIX: The fix is CUI instead of
calling driver to read registry, will issue an escape call to get factory
information. Driver will read from the hardcoded path and will fill in the
factory information. Since Driver knows that this call is for factory keys,
it will not use Videoport calls for reading\writing registries but will use
KCH exposed standard DDIs to read and write.
CUI Design: CUI will now get the information about factory Install
from Driver through a Esc call. Reset of the registry keys will be handled by
driver. |
|||||||||||||||||||||||||||||
BugID: N/A |
1.Applying 1440x900@60 HZ on HDMI display results in a
blank out.(2) G35 chipset, It will show black screen when using AOC 2212ph
monitor and change to 1440x1050@75hz |
AIM Module |
Windows* XP |
Intel® 3 Series Chipset Family-B |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Applying 1440x900@60 HZ on HDMI display results in a blank
out. |
|||||||||||||||||||||||||||||
BugID: N/A |
DVI is not recognized when Hot Docking. |
Soft BIOS |
Windows Vista* |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: |
|||||||||||||||||||||||||||||
BugID: N/A |
CLONE from pcgsw: [Intel_Vist32]Can't run ContentStreaming
in D3D9 mode. |
Direct3D* |
Windows Vista* |
Intel® 3 Series Chipset Family-B |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Lack of indexes range checking if invalid data are sent by
the application Implemented index
range checking for indexed draw calls (Gen3 and Windows Vista*), taking care
of negative vertex offsets, zero stride case and not stride aligned vertex
offsets. Also changed vertex offset
type from DWORD to INT in draw calls - this value can be negative in draw
calls coming from an application, and the casting from INT to DWORD that was
present in the code could potentialy cause some problems. checked the solution using 3DMark06 and
SDK sample applications: ContentStreaming, Instancing, LocalDeformablePRT,
SkinnedMesh, MultiAnimation. Ran this
application on Gen4 (Intel® G965 Express Chipset) and Gen3 (Berlake) on
Windows* XP32, Windows Vista*32 and 64 systems using both hardware and
software vertex processing where applicatble, no regression. There are problems running
ContentStreaming application in SWVP mode on Gen4 under Windows* XP (corruption)
and Windows Vista*(application not responding) - but this behaviour is
independent on the changes for this DCN and might be application issue
(anyway might be worth investigating) |
|||||||||||||||||||||||||||||
BugID: 2448498 |
Visual Anomaly/Direct3D* ? DX9c_StatePermuter; Alpha Blend,
Tiled Alpha Blend, and Tiled4 Alpha Blend Tests are failing because textures
are being displayed when they are not supposed to be |
Direct3D* |
Windows* XP |
Intel® G965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause Process_BlendOpAlpha was using the same flag as
Process_BlendOp() saparate flag was added for Process_BlendOpAlpha(). |
|||||||||||||||||||||||||||||
BugID: N/A |
The system somtimes shows blank screen with CH7315 by changed
the resoluton |
AIM Module |
Windows Vista* |
Intel® G965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause At certain temperature range (most likely at case temperature
35 – 55 degree C), when Fujitsu Spica system is doing display mode switching,
the CH7315A/B’s DVI output driver may not function properly even there are
SDVO inputs to the device. This will cause no-display (blackout) to the DVI
panel. The no-display failure/mode-changes rate is device dependent and can
be from 0% to 5% (1 out of 20 mode changes).
Adding a software “datapath reset” to the current Intel driver
releases after any mode change will solve this issue. The “datapath reset”
will reset the circuit blocks of DVI PLL and Clock-tree. It will force the
switching mode condition to be uniform. The details of the software
modification are described as follows: The original Intel SDVO EDS
mode-change sequence does not need any change. Two CH7315A/B and CH7319/20A
register writes with DID and DRID check are required after all mode-change
Opcodes are done. |
|||||||||||||||||||||||||||||
BugID: 2553167,
2553463 |
Mode Persistance broken in 14.32 4879 & 4889 |
NT - miniport |
Windows* XP |
Intel® 3 Series Chipset Family-B |
|||||||||||||||||||||||||
Resolution Description: Root Cause: With existing code,
while deciding the boot type (i.e. First boot, Driver upgrade or Driver boot),
driver tries to read the first boot registry key, if it gets successful, it
sets the boot event as first boot. This logic is wrong because, if the first
boot registry key is already present on a system (For example systems with
older driver versions will already have it), on every reboot, driver will
consider every boot event as first boot and hence the issue. FIX: Read the first boot registry key, if it is
NOT successful then consider this as first bootJ. Next set mode will create
this registry key and make the boot event as driver boot. |
|||||||||||||||||||||||||||||
BugID: N/A |
GMBus hang when the DDC bus being stuck low by a Display |
AIM Module |
Windows Vista* |
Intel® G965 Express Chipset, |
|||||||||||||||||||||||||
Resolution Description: Root Cause: there is a hardware issue with ASUS panel and they have
requested for a workaround for the issue with ASMEDIA Encoder to disable EDID
reads for SDVO based CRT. This is controlled through inf |
|||||||||||||||||||||||||||||
BugID: N/A |
To remove the changes made to fix the bug 2482563 as it was
creating a critical issue. |
NT - miniport |
Windows* XP |
Intel® 945GM Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: The changes were causing the EDID to be corrupted for
panels having refresh rate < 60Hz in which case OS will not prune the modes
and CUI and OS page would show up modes till 20x15. Done ULT to see that 12x8
Panel is not seen as Default monitor with OS properly pruning modes. |
|||||||||||||||||||||||||||||
BugID: N/A |
Disabling HW Binning for Intel® 945G Express Chipset GZ |
Direct3D* |
Windows Vista* |
Intel® 945G Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Disabling HW Binning for Intel® 945G Express Chipset GZ to workaround
a potential platform issue. |
|||||||||||||||||||||||||||||
BugID: N/A |
Visual Anomaly/Direct3D* ? DX9c_StatePermuter; Alpha Blend,
Tiled Alpha Blend, and Tiled4 Alpha Blend Tests are failing because textures are
being displayed when they are not supposed to be. |
Direct3D* |
Windows* XP |
Intel® G965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Process_BlendOpAlpha was using the same flag as
Process_BlendOp() saparate flag was added for Process_BlendOpAlpha(). |
|||||||||||||||||||||||||||||
BugID: N/A |
Code changes done for fixing the Toshiba mode removal
issues. |
INF File, Soft BIOS |
Windows* XP |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Need for the code
change: -Need a solution from driver side which will disable the interlaced mode
removal and will show the original behavior of removing only progressive
modes. Fix: - An INF based solution
has been made. In INF, we will be having a flag which will indicate whether
we need to enable interlaced mode removal. By default, this will be enabled
(one). Customers who doesn’t need this feature can disable that by writing a
zero to this field in INF. |
|||||||||||||||||||||||||||||
Issues Resolved in Build 4897 |
|||||||||||||||||||||||||||||
Reference No. |
Description |
Affected Component(s) |
Affected OS(s) |
Affected Project(s) |
|||||||||||||||||||||||||
BugID: 102670, 2550192 |
Intel® 945GZ have some issue w website of China IM app |
GDI, |
Windows* XP, Windows*
XP-64, Windows Vista*, Windows |
Intel® GM965 Express Chipset, |
|||||||||||||||||||||||||
Resolution Description: Root Cause: While visiting the website and moving the mouse from center
to left, We are getting the various SetPointerShape() calls.At the very left
side, we are getting the wrong argument values for xHot and YHot . This xHot and yHot arguments are the
relative position of pointer with respect to the Upper-left pixel co-ordinates
of the pointer and this should not
exceed the limit of the max size of the cursor(which is 64x64) . But while
moving the cursor pointer, these values are 480x480 which is not acceptable
that is why corruption is happening.
This App is doing something wrong due to which we are getting wrong
values. It is work around for that. Fix
Description: Solution is to punt the
call in case we get wrong hot spot values for cursor. The ETM has been completed and the test
passed with issues. The ETM has been completed and the test passed with
issues. Sighting 102670 has been
attached for reference. |
|||||||||||||||||||||||||||||
Issues Resolved in Build 4889 |
|||||||||||||||||||||||||||||
Reference No. |
Description |
Affected Component(s) |
Affected OS(s) |
Affected Project(s) |
|||||||||||||||||||||||||
BugID: 2549608 |
3D CAD software
bluescreen issue |
OGL |
Windows* XP |
Intel® 946GZ Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: The problem happens when the command buffer is overwritten by the driver which would cause the hardware hang. It is caused by flushing the pipeline in one part of the openGL product (OGL) without letting the other part (Ghal3D) know. |
|||||||||||||||||||||||||||||
Issues Resolved in Build 4885 |
|||||||||||||||||||||||||||||
Reference No. |
Description |
Affected Component(s) |
Affected OS(s) |
Affected Project(s) |
|||||||||||||||||||||||||
BugID: 2484714, |
Game- Snowy corruption is displayed in the Main Menu of the race car game |
Direct3D* |
Windows* XP, Windows* XP-64 |
Intel® G965 Express
Chipset, |
|||||||||||||||||||||||||
Resolution
Description: Root Cause: Incorrect swizzle settings in lit instruction |
|||||||||||||||||||||||||||||
BugID: 2552211, |
DXVA encryption workaround for player app |
DVD |
Windows* 2000, Windows* XP, Windows Media* Center |
Intel® 915GM Express
Chipset, |
|||||||||||||||||||||||||
Resolution
Description: Root Cause: Player App does not send in correct parameters for DXVA encryption, but because the product is already released we have to work around the problem in the driver. |
|||||||||||||||||||||||||||||
Issues Resolved in Build 4879 |
|||||||||||||||||||||||||||||
Reference No. |
Description |
Affected Component(s) |
Affected OS(s) |
Affected Project(s) |
|||||||||||||||||||||||||
BugID: 2436845, |
1280x720 resolution mode is not on the IGT page. |
Soft BIOS |
Windows* XP |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: bModeCurrentlyNotSupported is set for 1280x720 in
AIMENCODER_AddModes(). As a result it is being pruned in
MODETABLE_GetGDIModes() |
|||||||||||||||||||||||||||||
BugID: 2448290, |
BSOD when running ClickToMeet application. |
GMM |
Windows* XP, |
Intel® 945G Express
Chipset, |
|||||||||||||||||||||||||
Resolution Description: Root Cause: GMM is running out of nodes in their page cache data
structure and not failing gracefully causing the BSOD. GMM is running out of
nodes in their page cache data structure when allocating some pages. They are not failing gracefully and so when
the pages are freed, the BSOD occurs because they have corrupted the page
pool. Increasing the number of nodes
in their page cache fixes the issue. GMM has an AR to investigate why they
didn't fail gracefully. |
|||||||||||||||||||||||||||||
BugID: 2450838 |
Some CRT refresh rate
cannot be set with specific resolution |
Soft BIOS |
Windows* XP, |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: While applying timing for Maintain Aspect Ratio for modes,
whose AR doesnt match that of the display, it was not taken into
consideration the RR of the timing which was getting applied. customer wanted
the RR of the timing has to be the same as that of the mode. Code changes
have been done for finding a timing with the same RR as that of the mode. |
|||||||||||||||||||||||||||||
BugID: 2485841, |
Intel Display Driver unable to save monitor positions using custom schemes by CUI. |
CUI2 |
Windows* XP, |
Intel® 945GM Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: The issue will happen only if the Apply button in maindlg is
clicked and not the OK button whenever we change the positions in extended
desktop. This is because the correct values to BackupBuilder structure which
will be used by READ_SCHEME when a scheme is to be applied was not being
stored. Fix Description: Have
added a function which compares the builder for secondary position and
returns a int val(return from memcmp) .
Based on the return from the above , the Builder structure will
updated. ETM Execution for this Bug Stands complete now. |
|||||||||||||||||||||||||||||
BugID: 2517302, |
Cannot toggle any more after enable Extended mode via Graphics Options |
NT - miniport |
Windows Vista*, |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: This bug is primarily caused as CUI is issuing more than
one RFVN for same config & mode for MDS if trying to apply it through
Graphics Options>>choose Output To>>choose Extended Desktop.
Moreover we have a WA which is provided by CUI & Miniport for ACPI hotkey
& due to CUI misbehavior & this WA, Event Manger flags are getting
inconsistent and causing the issue. |
|||||||||||||||||||||||||||||
BugID: 2518893 |
The display mode is CRT only mode after updating the graphics driver and reboot up to the system. |
NT - miniport |
Windows* 2000 Windows* XP Windows* XP-64 |
Intel® 3 Series Chipset Family |
|||||||||||||||||||||||||
Resolution Description: Root Cause: The issue is that after driver upgrade in Extended Desktop,
since driver sees that next configuration is Clone different from previous
one, in the function EM_IsExitEDT() in Miniport, we set a registry key which
makes CUI to issue a call to come out of Extended Desktop and come in Single
Display. Since it also has the logic to request driver for advise of
configuration, both the threads execute and which ever comes late, the
configuration corresponding to it persists and in the failure case the Single
display switch call is coming later. Fix is to not set the registry key in
driver as anyway CUI will request for advise of configuration everytime after
driver upgrade. The issue happens always. After looking into CUI and
discussing with Indranil, CUI does the job of setting Clone configuration and
calling wa_exitEDT() in a single thread and so there is no sporadicity in
this case. Basically, driver sets switchtoSPSD bit for CRT during boot. CUI
when coming up, asks driver for configuration and mode since it is Clone and
applies it. In the same thread later calls wa_exitEDT() where it finds that
the key is set and so switches to SPSD on CRT. |
|||||||||||||||||||||||||||||
BugID: 2519302, |
Generic pnp string missing after selecting TV in information page and going back to CRT/DFP/Notebook |
CUI2 |
Windows* XP, |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: String for CRT/DFP/LFP missing after selecting TV in the
Information Page. OS returns a NULL
string for TV. So the Control eas hidden for showing the name as the length
of the stringis NULL. Hence for all consecutive clicks on the tabs for
CRT/LFP/DFP, no name is displayed. Fix Description:
Instead of removing the customization, added a else loop to take care to
showing the strings for devices other than TV. |
|||||||||||||||||||||||||||||
Issues Resolved in Build 4876 |
|||||||||||||||||||||||||||||
Reference No. |
Description |
Affected Component(s) |
Affected OS(s) |
Affected Project(s) |
|||||||||||||||||||||||||
BugID: 2102412,2328999, |
Game in OpenGL* mode: object disappears |
OpenGL* |
Windows* XP, |
Intel® G965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Fragment program invalidated forever after using bad
texture. |
|||||||||||||||||||||||||||||
BugID: 2445048, |
"Mouse cursor" is hidden for an instant after
drag window. |
GDI |
Windows Vista*, |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root
Cause: Fix Description: SR
needs to be disabled after making any updates to cursor status register. Then
after a Vblank period, re-enable SR back. We need to wait for a Vblank
period, because update to SR register will take effect only in the next
Vblank period. A wait for Vblank will
have benchmark hit. So, to avoid this, the sequence will be • Make cursor
related changes. (Enable/disable cursor) • Disable SR • In the next VBI
handler, enable SR. Enabling/Disabling
SR is handled by PC. With Adaptive
Vsync feature in Windows Vista* SP1, there can be a case in which the VBI is
disabled by OS when cursor call is being handled. In this case, the logic
will be enable VBI internally and disable it after one Vblank period in the
next VBI handler. |
|||||||||||||||||||||||||||||
BugID: 2446955, |
Add VertexProcessingCaps to reported caps for Windows
Vista* |
Direct3D* |
Windows Vista*, Windows |
Intel® G965 Express Chipset, |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Received confirmation from Microsoft that the driver should
be reporting the VertexProcessingCaps even though, on Windows Vista*, the
d3d9 runtime emulates the support represented by these caps (fixed function
TnL). When an application calls
IDirect3D9Device::GetDeviceCaps, Microsoft will fill in the caps for the emulated
FFTnL features (VertexProcessingCaps). If the application instead calls
IDirect3D9::GetDeviceCaps, they will get back the caps that the driver
returns which don't report any VertexProcessingCaps. This will cause the app to think the
hardare doesn't support FFTnL and it will create a software device. Fix Description: To
handle both cases correctly, the caps were added even though Microsoft is
still emulating the features. This
will allow some applications which previously created a software device to
now create a hardware accelerated device. |
|||||||||||||||||||||||||||||
BugID: 2483142, |
Fn-F5 works behind Command Prompt full screen. |
CUI COM |
Windows* XP, |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: In FSDOS mode display switiching was happening when the
same is disabled through the VBT. The
reason for this was validation logic was not sufficient enough to handle the scenario,
in CUI aswell as Driver. Fix Description: changed
the part in CUI SDK to do the first level of validation which checks with
current and requested version. |
|||||||||||||||||||||||||||||
BugID: 2483738, |
After Undock/Dock operation, Display resolution changes to
1024x768 unexpectedly. |
NT - miniport, CUI2 |
Windows* 2000, |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: The reason for the issue is two fold - i) OS maintains two profiles for Dock and
Undock with information of modes for each devices that is set in each
profiles separately. Whenever we are switching over to Docked/Undocked state,
OS picks mode from that profile and requests driver to set. But in the buggy
case, we see that Dock profile is corrupt and has a different mode as
compared to the mode that was set previously in Docked state. ii) Though OS picks up wrong mode, CUI
should persist the right mode back, however in the buggy case, it seemsthat a
wrong set mode from OS is happening without any event link and is causing CUI
to corrupt its database. So, when CUI reapplies it back, it would be same as
what OS had applied. Fix Description: Basically, all the changesare to lock the
CUI data base from corruption without causing any noticeable change in
behaviour for end user. |
|||||||||||||||||||||||||||||
BugID: 2485561, |
ArrayQPitch Offset Fix for Compressed textures DX/FX |
GMM, GHAL3D, KMD Render |
Windows Vista*, |
Intel® 945G Express Chipset, |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Array Q Pitch calculation for compressed textures is
incorrect. Array Q Pitch calculation for compressed textures was clarified in
BUN 07ww32#1. This BUN also revealed
another requirement that 1D surfaces always use Layout_Down. Both this fixes are in this DCN branch. The
DX10 ULT Frame Work Suite was failing due to an error related to the 1D
surface Layout_Below requirement. Fix Description: 1)
Update Array QPitch calculation for compressed textures 2) Force 1D textures
to be Layout_Below 3) Fix MinimumArrayElement Bit field in KmRender, GHAL,
GHAL3D 4) UMD/Render Interface correction. |
|||||||||||||||||||||||||||||
BugID: 2485678, |
Flicker is displayed while playing DVD. |
DVD |
Windows* XP, |
Intel® 945GM Express Chipset, Intel® GM965 Express Chipset,
Intel® 3 Series Chipset Family |
|||||||||||||||||||||||||
Resolution Description: Root Cause: In a fix for another issue, changed the reference frame checking
from a threshold check to checking if the frame was the true previous frame
because this was how reference frames were handled in Windows* XP. However,
this changed caused choppiness in the certain videos due to the frequent
shifting between BOB and Advanced Deinterlacing. Fix Description: the
check back was changed to a threshold check in both Windows* XP and Windows
Vista*. |
|||||||||||||||||||||||||||||
BugID: 2516050, |
BSOD(116) if Hot-plug 2nd HDD with PW and resume from S3 |
NT - miniport |
Windows Vista*, |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Driver changes which are done for safety purpose. Fix Description: Removing
the bSystemInTdr check for SetPowerStateDDI. |
|||||||||||||||||||||||||||||
BugID: 2516355, |
Not able to change the aspect ratio after rotation |
CUI2 |
Windows Vista*, |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Fix Description:
Values are not initialized properly for 90 and 270. Fixed the same. |
|||||||||||||||||||||||||||||
BugID: 2516360, |
Aspect Ratio Options button displayed for TV |
CUI2 |
Windows* XP, |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Fix Description: ZeroMemory
pBuilder.Settings[1].aspect structure. |
|||||||||||||||||||||||||||||
BugID: 2516547, |
Windows Logon screen is upside down after resume from
suspend on independent rotation condition. |
Rotation |
Windows Vista*, |
Intel® 945GM Express Chipset, Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Issue happens because at the time of resume from S3, we do
not base our checking for the eOrientation for Primary and Secondary based on
Importance Ordinal. Since there is OS limitation of not comprehending the
Secondary device orientation. Fix Description: Fix
resolves the ambiguity of Primary/Secondary devices and updates registry and
flips secondary angle only if OS doesn’t comprehend it. The validattion
hasbeen completed and the test has passed with issues. The issues were related
to CUI and display switching. Sightings are attached. |
|||||||||||||||||||||||||||||
BugID: 2516822, |
Video froze but audio was normal during play the sample
video repeatedly with WMP with IGD |
DVD |
Windows Vista*, Windows |
Intel® 945G Express Chipset, |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Surface Format is used as Source, then we use the hTempHybridPlanarBuffer
to convert NV12 to YUY2 (or IMC3). However, this temp buffer was never freed,
so Gfx Memory usage goes up over time.
This DCN is used to add a free for the temp buffer. |
|||||||||||||||||||||||||||||
BugID: 2516878, |
Confirmation window appear twice |
CUI2 |
Windows |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Issue happens because after applying the “Enable rotation” check-box,
m_builder.settings[0].isServiceReq is still not reset in OnApply(). Unlike in
other controls where there is a COM call, and the variable gets reset to
zero. But when we click OK in the main CUI window, the OnCancel function is
called and the m_builder.settings[0].isServiceReq variable is reset to zero.
ETM execution complete for this BUG. The actual bug stands fixed now. |
|||||||||||||||||||||||||||||
BugID: 2517077, |
Texture Blending IsValidStage Fix |
Direct3D* |
Windows* XP, |
Intel® G965 Express Chipset, |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Texture Blending TFACTOR operation was not implemented
correctly in Direct3D* and instead called an unimplemented GHAL3D state
processor function. Tested Lego Wars Pix Capture and TextureStage DCT tests
on Windows* XP32 |
|||||||||||||||||||||||||||||
BugID: 2517272, |
Corruption is displayed while playing DVD movie via player
application with modified registry keys. |
DVD |
Windows* XP, |
Intel® 3 Series Chipset Family |
|||||||||||||||||||||||||
Resolution Description: Root Cause: |
|||||||||||||||||||||||||||||
BugID: 2517731, |
Color quality column disappear when operating CUI. |
CUI2 |
Windows* XP, |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Fix Description: Added
notification. This notificatin will be sent when the listbox of the combo box
is closed. At this condition we call Invalidate for painting. |
|||||||||||||||||||||||||||||
BugID: 2518020, |
Gamma, Brightness and Contrast values do not match values
selected for scheme. |
CUI2 |
Windows Vista*, |
Intel® G965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Actual values are getting set but the UI is not reflected
with new values. This is because a Display change message is not received. Unless
this is received, other dialogs will not be updated with the latest
values. Fix Description:
Issued a postmessage with User defined Display change notification. Etm
execution for this issue stands complete now. The Original bug is fixed. The
Grid was executed on Windows Vista* 32 and Windows* XP32. No other issues
found. Updating the DCN as " |
|||||||||||||||||||||||||||||
BugID: 2518153, |
TERC4 is transmited from DVI-D port (DVI-HDMI cable issue) |
NT - miniport |
Windows |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: TERC4 is transmited from DVI-D port (DVI-HDMI cable issue).
Root cause: OS DDI DxgkDdiQueryDeviceDescriptor is requesting multiple blocks
of EDID data (i.e. Descriptor Length > 128). Currently, we return only one block of EDID
data on request for OS. In this
case,OS is requesting for 256 bytes of EDID data starting from offset =
0. We call Softbios and get block 0
(since offset = 0), Softbios returns first block i.e. block num = 0 with 128
bytes and we provide the same to OS. Fix Description: Miniport will get all the blocks
requested by OS (depending on the Descriptor Length) and provide the same to
the OS |
|||||||||||||||||||||||||||||
BugID: 2519341, |
enable offscreen plain for A32B32G32R32f |
Direct3D*, DirectDraw* |
Windows* XP, |
Intel® G965 Express Chipset, |
|||||||||||||||||||||||||
Resolution Description: Root Cause: This branch enables A32B32G32R32f as an offscreen plain in Direct3D*
component, and parses it in the DirectDraw* component. This branch adds the
following offscreen plain formats:
A32B32G32R32f A16B16G16R16f A16B16G16R16 G32R32 A2R10G10B10 G16R16
R32f Q8W8V8U8 V16U16 A2W10V10U10 R16f
The paths to use these formatsis different in Windows* XP and Windows
Vista*. Windows* XP has code paths in
GDI and DirectDraw* that are affected, and as a result the changes can affect
the results on Gen 3.5 architectures like Intel® 3 Series Chipset
Family. On Windows Vista*, the ghal2d
path is taken instead. |
|||||||||||||||||||||||||||||
Issues Resolved in Build 4869 |
|||||||||||||||||||||||||||||
Reference No. |
Description |
Affected Component(s) |
Affected OS(s) |
Affected Project(s) |
|||||||||||||||||||||||||
BugID: 2326928,2329502 |
OpenGL driver bug for 945GM and 82852/82855 (Black Triangles Bug) |
OpenGL* |
Windows* XP, |
Intel® 855GM Express
Chipset, |
|||||||||||||||||||||||||
Resolution Description: Root Cause: This will fix the subproblem #3 (Markers lose color in Fog): Raw eye.z coordinate is used instead of absolute eye.z value (eye-to-fragment distance) to calculate fog factor. In some cases it results in negative fog factor, which is next clamped to 0 (markers have 100% fog color). Fix affects both Windows* XP and Windows Vista*. |
|||||||||||||||||||||||||||||
BugID: 2329502 |
Molsoft/Novartis OpenGL driver bug for 945GM and 82852/82855 (markers loosing color in fog) |
OpenGL* |
Windows* XP, |
Intel® 855GM Express
Chipset, |
|||||||||||||||||||||||||
Resolution Description: Root Cause: This will fix the subproblem #1 (Black Triangles Bug): Fix affects both Windows* XP and Windows Vista*. |
|||||||||||||||||||||||||||||
BugID: 2439409,2447981 |
49% degradation between 15.4 PV and 15.4.x driver on Quake III* demo score |
OpenGL* |
Windows Vista*, Windows |
Intel® 3 Series Chipset Family |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Performance degradation fix. Handles caching mechanism implemented; performance improvement is about 55%. |
|||||||||||||||||||||||||||||
BugID: 2329502 |
Molsoft/Novartis OpenGL driver bug for 945GM and 82852/82855 (Missing Triangles Bug) |
OpenGL* |
Windows* XP, |
Intel® 855GM Express
Chipset, |
|||||||||||||||||||||||||
Resolution Description: Root Cause: This will fix the subproblem #2 (Missing Triangles): flushCountHi have to be a multiple of 12, to allow simple partitioning of triangles and quads in __glDrawArraysLarge() routine. Fix affects both Windows* XP and Windows Vista*. |
|||||||||||||||||||||||||||||
BugID: 2451063 |
LVDS Programming update for fixed frequency Panel |
Soft BIOS |
Windows* XP, |
Intel® GM965 Express Chipset, |
|||||||||||||||||||||||||
Resolution Description: Root Cause: LVDS Programming update for fixed frequency Panel. |
|||||||||||||||||||||||||||||
BugID: 2484264,2517744,2518453, |
Clipper Vertex Element Descriptor updates (DX9,10) |
Direct3D*, OpenGL*, GHAL3D |
Windows* XP, |
Intel® G965 Express Chipset, Intel® GM965 Express Chipset, |
|||||||||||||||||||||||||
Resolution Description: Root Cause: GHAL3D Clipper Template DCN. Updating Vertex Element Descriptors in Clipper Cache Key - Fix binding - Reduce size (eliminate redundancy) - etc. |
|||||||||||||||||||||||||||||
BugID: 2485678,2517472, |
Flicker is displayed while playing Title 19, Chapter 1 of Video Essential DVD. |
DVD |
Windows Vista*, Windows |
Intel® 945GM Express Chipset, Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Transitioning from successive progressive frames to interlaced frames (specifically those deinterlaced using adaptive DI) resulted in a flicker on Windows Vista* using PowerDVD due to a discontinuity in the timestamps between interlaced frames. So this DCN fixes the problem by changing the deinterlacing format to BOB when the discontinuity occurs, eliminating the need for a reference surface. |
|||||||||||||||||||||||||||||
BugID: 2451401, |
GHAL3D support for D3DRS_WRAP8 -15 in DX-9. |
Direct3D*, GHAL3D |
Windows* XP, |
Intel® G965 Express Chipset, Intel® G965 Express Chipset, Intel® GM965 Express Chipset, |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Adding support to GHAL3D setup compiler to support D3D-9 D3DRS_WRAP8 to D3DRS_WRAP15 texture wrap modes. |
|||||||||||||||||||||||||||||
BugID: 2450349, |
Windows Vista* persistence implementation |
CUI2 |
Windows Vista*, Windows |
Intel® 945GM Express Chipset, Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Implemented persistence on Windows Vista*. |
|||||||||||||||||||||||||||||
BugID: 2448894, |
There is no cube while test GL 1.4 and 1.5. |
OpenGL* |
Windows* XP, |
Intel® G965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: User Clip Plane has not been added to __glValidateVertexProgramMatrixBindings. |
|||||||||||||||||||||||||||||
BugID: 2517472, |
Update selection of optimal Gen4 Kernel for different usage models on Windows Vista* |
DVD |
Windows* XP, |
Intel® G965 Express Chipset, Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description:
Root Cause: Update gen4 kernel selection for different usage models on Windows* XP and Windows Vista*, based on kernel team recommendations. |
|||||||||||||||||||||||||||||
BugID: 2441515, |
Doom 3* benchmark Application Error when running on high video setting @ 1280x1024 resolution |
OpenGL* |
Windows* XP |
Intel® G965 Express Chipset, Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: The problem ocurs in gfxTexSubImage() function when mip->curimg->base (NULL ptr) is used to set the value of dstSliceAddr. mip->curimg points to &mip->sysimg, which base is NULL. CAUSE: mip->curimg->base is NULL after gfxCopyTexImageGetLevel is called. Look at gfxCopyTexImageGetLevel() function - VMM_LEVEL_FREE(gc, texture, texFace, lod) is called to free VMM memory (mip->sysimg.base != NULL and __GL_LEVEL_IS_RESIDENT(texFace, lod) is true before that). We try to allocate HW memory for this texture. This allocation fails and we return from gfxCopyTexImageGetLevel() and from gfxTexSubImageBlitGen4() with GL_FALSE. Than gfxTexSubImage() is called and Application crashes because of mip->curimg->base NULL ptr when dstSliceAddr is counted. There is assumption that if mip->curimg is not NULL one of memory location (devimg or sysimg) is allocated. gfxCopyTexImageGetLevel() function leaves mip structure with devimg and sysimg as NULL, and currimg set to sysimg, which is unacceptable state. Fix Description: In gfxCopyTexImageGetLeve() after HW allocation fails set mip->curimg as NULL before return (withGL_FALSE) to indicate that HW memory is not allocated. |
|||||||||||||||||||||||||||||
BugID: 2451401 |
Shader Memory Optimizations |
Direct3D* |
Windows* XP, |
Intel® 945G Express Chipset, Intel® 945GM Express Chipset, Intel® G965 Express Chipset, Intel® G965 Express Chipset, Intel® GM965 Express Chipset, Intel® 3 Series Chipset Family |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Made a change to the CConstantArray class to allow for compression of data. This has been enabled for local constants stored within the pixel and vertex shaders. I've also addressed an issue that could leadus to send down too many vertex elements on Gen3/Windows Vista* parts. Successfully completed benchmark testing and DTM testing on Windows* XP. Addressed issues seen in previously failed ETM. Please confirm on Windows Vista*. |
|||||||||||||||||||||||||||||
BugID: 2484022,2516355 |
Rotation Orientation for Secondary as 270 Degrees is disappeared in DTCM. |
CUI2 |
Windows Vista*, Windows |
Intel® GM965 Express Chipset, Intel® 3 Series Chipset Family |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Rotation angles for Secondary device are not updated properly in DTCM Menu when we are 90-270 Independent Rotation. Instead of showing 90-270 for secondary when primary is either 90 or 270, DTCM menu ends up showing 0-180 for secondary in all cases [0-180, 90-270, 180-0, 270-90]. Rootcause: When we are in DDC independent rotation, we update the secondary rotation caps based on primary angle. In the current code, we update secondary caps before updating the correct primary angle. As a result, we always set secondary rotation caps to 0-180 in all the cases of Independent rotation. So no issue w.r.t 0-180 and 180-0 but with 90-270 and 270-90, we need to update it correctly based on correct primary angle. Fix Description: Moved the part of the code which updates the primary angle and secondary angle properly before updating secondary rotation caps. |
|||||||||||||||||||||||||||||
BugID: 2450309 |
The display does not switch to SDTV for hotplug event. |
Soft BIOS |
Windows* XP, |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: When TV is active and the only device connected, unplugging TV will result in TV encoder remaining active. |
|||||||||||||||||||||||||||||
BugID: 2449299, |
Windows* XP 3DMark2006 intermittently crashes on launch of benchmark on Gen 3.5 |
Direct3D*, GMM |
Windows* XP, |
Intel® 945G Express Chipset, Intel® 945GM Express Chipset, Intel® G965 Express Chipset, Intel® G965 Express Chipset, Intel® GM965 Express Chipset, Intel® 3 Series Chipset Family, |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Certain types of memory is being relocated into special GPU_ONLY heap in GMM. This heap is only available on BL-B. This addresses the 3DMark '06 issue where the application would fault if it could not get enough memory to allocate a large cube map. The Gen3 internal vertex buffer and hardware binning batch buffers are being relocated to GPU_ONLY. These have the advantage of not being lockable by the application and linear. |
|||||||||||||||||||||||||||||
BugID: 2482692, |
Brightness hotkey cannot work correctly after change setting by scheme option on CUI |
GDI |
Windows* XP |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Issue happens because Software control over PWM backlight is disabled after applying any scheme. Before mode change we disable specific power conservation features and enable it back after mode change. So SW control for brightness is disabled during premodechange event for SB_SET_COMPENSATION and enabled back after mode change is successful. But one of the calls to SB_SET_COMPENSATION fails and postmodechange event is not called to enable the PW features(BLC) again. In CUIDriverFunction GdiRmNotifyEvent(FALSE) should be called even when SB_SET_COMPENSATION fails. |
|||||||||||||||||||||||||||||
BugID: 2484074, |
Fixing the last failure in WDDM Passthrough Lock |
Direct3D*, OpenGL*, GHAL3D |
Windows Vista*, Windows |
Intel® G965 Express Chipset, Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: It will reset the command buffer and all the internal variables to their initialized state. This will prevent GHAL3D clients (like GHAl2D, Direct3D*, and OpenGL*) from spinning infinitely trying to send commands to GHAl3D. |
|||||||||||||||||||||||||||||
BugID: 2485424, |
Fix Memory Leak with HWVP and mixed internal and extrnal vertex buffers |
Direct3D* |
Windows* XP |
Intel® G965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: Fix Memory Leak with HWVP and mixed internal and extrnal vertex buffers. During a transition between external and internal vertex buffers, the internal d3dresource was being deleted and recreated multiple times. When running multiple SDK sample applications, the system would eventually run out of memory. The fix will only create one internal d3dresource for the internal HW vertex buffer. |
|||||||||||||||||||||||||||||
BugID: 2516354, |
Corruption displayed while playing the Video Essential DVD movie and closing CyberLink Power DVD 7 application then re-play the Video Essential DVD movie again. |
DVD |
Windows* XP-64 |
Intel® 945G Express Chipset, Intel® 945GM Express Chipset, Intel® G965 Express Chipset, Intel® GM965 Express Chipset, Intel® 3 Series Chipset Family |
|||||||||||||||||||||||||
Resolution Description: Root Cause: This issue was specific to Windows* XP64. When a 32-bit application is used on 64-bit OS, the struct we get is 32-bit size, so a thunk is needed to convert from 32-bit to 64-bit. The thunk was being done incorrectly and has been fixed. |
|||||||||||||||||||||||||||||
BugID: 2516099, |
Some of the TV standards are not getting applied |
Soft BIOS |
Windows* XP, |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description: Root Cause: The mode in concern is 848 * 480 for the standard 720p60. In fact, this mode needs to be there for all the standards except 576i/p. The current code had only this mode available for 480i/p. The necessary code changes have been made to enumerate this mode for all the standards except 576i/p. |
|||||||||||||||||||||||||||||
Issues Resolved in Build 4861 |
|||||||||||||||||||||||||||||
Reference No. |
Description |
Affected Component(s) |
Affected OS(s) |
Affected Project(s) |
|||||||||||||||||||||||||
BugID: 2440143, 2485836 |
Windows Vista* 6 Flash to 2 Flash Reduction changes when Aspect Scaling, Independent Rotation, Configuration and TV standard are applied together. |
Soft BIOS Rotation CUI2 |
Windows Vista*, |
Intel® 945GM Express Chipset Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description:
Root Cause: Problem: Multiple flashes when all AS, TV Standard, Rotation & Configuration change together, 6 separate invalidates will result. This leads to 6 flashes. Solution: To combine all DDC independent rotation, rotation, aspect scaling and TV format into 2 invalidates with TV still needing 2 escapes prior to it. |
|||||||||||||||||||||||||||||
BugID: 2440922 |
LCD monitor can't resume from power saving mode |
Soft BIOS |
Windows* XP, |
Intel® G965 Express Chipset Intel® G965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description:
Root Cause: Monitor does not resume from Monitor OFF mode, in case of Intel® G965 Express Chipset. It happens only with particular CRTs. This is due to a hw. issue wherein, if port multiplier is ON and CRT is the only port on the pipe, turning ON DPMS during resume causes this issue. |
|||||||||||||||||||||||||||||
BugID: 2443806 |
Intel® 945G Express Chipset Fix for "System hang running graphics stress apps" Siemens related |
Direct3D* |
Windows Vista*, |
Intel® 945G Express Chipset Intel® 945GM Express Chipset |
|||||||||||||||||||||||||
Resolution Description:
Root Cause: This fix addresses an issue where the driver is respoonsible for writing vertex buffer data on new cache-line boundaries in order to stop a possible scenario when the CPU and GPU access the same cache-line at the same time causing a hang in either the GPU, or the CPU. |
|||||||||||||||||||||||||||||
BugID: 2445134 |
Takes a while to change display destination by CUI or Scheme option |
CUI2 |
Windows* XP, Windows* XP-64, Windows Vista*, Windows |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description:
Root Cause: Bug :The bug is User is not able to select the OK and Cancel button , it takes some time. RootCause :The reason is we are getting every time wm_timer message, so the user will not be able to select. Fix Description: Added a global variable which holds the value returned from SetTimer and the same is checked to do the further processing depending on whether the timer value is >0 or not. |
|||||||||||||||||||||||||||||
BugID: 2449441, |
The screen color changed by Color Correction is restored to default after rebooting |
CUI2 |
Windows |
Intel® 3 Series Chipset Family |
|||||||||||||||||||||||||
Resolution Description:
Root Cause: ETM Execution for Round 2 is over. Thr original Bug is fixed now. No issues with that and all the relevant scenarios related to Gamma PErsistence after s3 & S4 are working fine. |
|||||||||||||||||||||||||||||
BugID: 2449671 |
User able to select 90 & 180 rotations from Intel taskbar icon & desktop Windows Vista*&Windows* XP |
CUI2 |
Windows* XP, Windows* XP-64, Windows Vista*, Windows |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description:
Root Cause: Added check for verifying whether the string length is not "0" , and then only Append the string to the Menu. The same is taken care for all the 4 rotation angles appending in the menu. |
|||||||||||||||||||||||||||||
BugID: 2451332, 2451475, 2482688 |
Addition of full overlay support to Windows Vista* driver with Windows* XP parity - 15.6 branch |
INF File Miniport |
Windows Vista*, Windows |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description:
Root Cause: Overlay creation in Windows Vista* is not POR. This RCR has been created to bring discussions regarding the addition of full overlay support to the Windows Vista* driver. The request is for Parity with Windows* XP functionality except where architecturally limited by Windows Vista* OS. |
|||||||||||||||||||||||||||||
BugID: 2451368 |
HDMI1.2a compliance test failure for Hot Plug |
Soft BIOS |
Windows |
Intel® G965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description:
Root Cause: The issue here was that the Sony HDMI CEC Compliance test was failing for the hotplug /unplug test. The time it should take to complete the device detection after an interrupt should not exceed 125ms and our driver was taking much more then a sec's time.Root casued the issue in which we had put a delay to fix the DTM issue |
|||||||||||||||||||||||||||||
BugID: 2483884, |
Title of Horizontal slider bar in Aspect options for Digital TV |
CUI2 |
Windows* XP |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description:
Root Cause: The title "Horizontal" for the Horizontal slider bar under custom settings for Digital TV (Aspect Ratio Options) is a little lower than expected. e.g. if we compare it to the vertical slider bar on the same page. |
|||||||||||||||||||||||||||||
BugID: 2483884, |
Digital Television - Aspect Ratio Options - Layout issues |
CUI2 |
Windows* XP |
Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description:
Root Cause: Issue happens due to location of the Min and Max labels. Changing their location exactly as per upper sliders resolves the issue. |
|||||||||||||||||||||||||||||
BugID: 2484129, 2484243 |
Corruption is displayed on root menu while playing the Twister DVD movie with InterVideo WinDVD 8 application. |
DVD |
Windows Vista*, Windows |
Intel® 945G Express Chipset Intel® 945GM Express Chipset Intel® 3 Series Chipset Family |
|||||||||||||||||||||||||
Resolution Description:
Root Cause: The destination surface size has been fixed for Adaptive Deinterlacing on Windows Vista*. |
|||||||||||||||||||||||||||||
BugID: 2484426 |
DVD screen stopped while playing Interlaced titles |
DVD |
Windows Vista*, Windows |
Intel® 945G Express Chipset Intel® 945GM Express Chipset Intel® 3 Series Chipset Family |
|||||||||||||||||||||||||
Resolution Description:
Root Cause: Add back the missing () to avoid creating resource every frame. |
|||||||||||||||||||||||||||||
BugID: 2485941 |
Media Support for CUI 2.5 |
DVD |
Windows* XP Windows* XP-64 Windows Windows |
Intel® 945G Express Chipset Intel® 945GM Express Chipset Intel® G965 Express Chipset Intel® GM965 Express Chipset Intel® 3 Series Chipset Family |
|||||||||||||||||||||||||
Resolution Description:
Root Cause: This DCN adds Media Support for CUI 2.5, to allow CUI to control Advanced Media Post-Processing Features. |
|||||||||||||||||||||||||||||
BugID: 2516130 |
Allow DPST to work in clone, or extended mode |
Power Conservation |
Windows* 2000 Windows* XP Tablet Pc Windows* XP-64 Windows Windows |
Intel® 945GM Express Chipset Intel® GM965 Express Chipset |
|||||||||||||||||||||||||
Resolution Description:
Root Cause: Allow DPST in Dual-Display, Extended and Twin-mode |
|||||||||||||||||||||||||||||