|
November 10,
2005
Microsoft* Windows* 2000
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. Intel products are not intended for use
in medical, lifesaving, or life-sustaining applications.
Intel may make changes to specifications and product
descriptions at any time, without notice.
Contact your local Intel sales office or your distributor
to obtain the latest specifications and before placing your product order.
* Other names and brands may be claimed as the property of
others.
Copyright © Intel Corporation 2005
This document contains information on products in the
design phase of development. The information herein is subject to change
without notice. Do not finalize a design with this information.
Revision
History
|
Date
|
Driver Revision
|
Description
|
Build Number
|
|
November 10, 2005
|
Production Version 14.18.0.4421
|
Production Version; fully
validated
|
4421
|
|
November 4, 2005
|
Production Candidate 14.18.0.4421
|
Production Candidate; fully
validated
|
4421
|
|
October 27, 2005
|
Beta 14.18.0.4414
|
Weekly engineering drop; not
fully validated
|
4414
|
|
October 13, 2005
|
Beta 14.18.0.4407
|
Weekly engineering drop; not
fully validated
|
4407
|
|
October 6, 2005
|
Beta 14.18.0.4398
|
Weekly engineering drop; not
fully validated
|
4398
|
|
September 29, 2005
|
Beta 14.18.0.4397
|
Weekly engineering drop; not
fully validated
|
4397
|
|
September 22, 2005
|
Beta 14.18.0.4395
|
Weekly engineering drop; not
fully validated
|
4395
|
|
September 8, 2005
|
Beta 14.18.0.4388
|
Weekly engineering drop; not
fully validated
|
4388
|
|
September 1, 2005
|
Beta 14.18.0.4386
|
Weekly engineering drop; not
fully validated
|
4386
|
|
August 25, 2005
|
Beta 14.18.0.4382
|
Weekly engineering drop; not
fully validated
|
4382
|
|
August 11, 2005
|
Beta 14.18.0.4374
|
Weekly engineering drop; not
fully validated
|
4374
|
|
August 04, 2005
|
Beta 14.18.0.4371
|
Weekly engineering drop; not
fully validated
|
4371
|
|
July 28, 2005
|
Beta 14.18.0.4365
|
Weekly engineering drop; not fully
validated
|
4365
|
|
July 21, 2005
|
Beta 14.18.0.4360
|
Weekly engineering drop; not
fully validated
|
4360
|
|
July 14, 2005
|
Beta 14.18.0.4349
|
Weekly engineering drop; not
fully validated
|
4349
|
|
June 30, 2005
|
Beta 14.18.0.4345
|
Weekly engineering drop; not
fully validated
|
4345
|
|
June 23, 2005
|
Beta 14.18.0.4338
|
Weekly engineering drop; not
fully validated
|
4338
|
|
June 16, 2005
|
Beta 14.17.0.4336
|
Weekly engineering drop; not
fully validated
|
4336
|
|
June 9, 2005
|
Beta 14.17.0.4330
|
Weekly engineering drop; not
fully validated
|
4330
|
|
June 2, 2005
|
Beta 14.17.0.4326
|
Weekly engineering drop; not
fully validated
|
4326
|
|
May 26, 2005
|
Beta 14.17.0.4324
|
Weekly engineering drop; not
fully validated
|
4324
|
|
May 19, 2005
|
Beta 14.17.0.4321
|
Weekly engineering drop; not
fully validated
|
4321
|
|
May 12, 2005
|
Beta 14.17.0.4318
|
Weekly engineering drop; not
fully validated
|
4318
|
|
May 5, 2005
|
Beta 14.17.0.4315
|
Weekly engineering drop; not
fully validated
|
4315
|
|
April 29, 2005
|
Alpha 14.17.0.4310
|
Weekly engineering drop; not
fully validated
|
4310
|
|
April 21, 2005
|
Alpha 14.17.0.4306
|
Weekly engineering drop; not
fully validated
|
4306
|
|
April 14, 2005
|
Alpha 14.17.0.4302
|
Weekly engineering drop; not
fully validated
|
4302
|
|
March 31, 2005
|
Alpha 14.17.0.4293
|
Weekly engineering drop; not
fully validated
|
4293
|
|
March 18, 2005
|
Alpha 14.17.0.4281
|
Weekly engineering drop; not
fully validated
|
4281
|
|
March 3, 2005
|
Pre-Alpha 14.17.0.4271
|
Weekly engineering drop; not
fully validated
|
4271
|
|
February 24, 2005
|
Pre-Alpha 14.17.0.4254
|
Weekly engineering drop; not
fully validated
|
4264
|
|
February 18, 2005
|
Pre-Alpha 14.17.0.4259
|
Weekly engineering drop; not fully
validated
|
4259
|
|
February 09, 2005
|
Pre-Alpha 14.17.0.4251
|
Weekly engineering drop; not
fully validated
|
4251
|
|
February 3, 2005
|
Pre-Alpha 14.17.0.4245
|
Weekly engineering drop; not
fully validated
|
4245
|
|
January 27, 2005
|
Pre-Alpha 14.17.0.4240
|
Weekly engineering drop; not
fully validated
|
4240
|
|
January 20, 2005
|
Pre-Alpha 14.17.0.4017
|
Weekly engineering drop; not
fully validated
|
4017
|
|
January 13, 2005
|
Pre-Alpha 14.17.0.4005
|
Weekly engineering drop; not fully
validated
|
4005
|
|
January 06, 2005
|
Pre-Alpha 14.17.0.3998
|
Weekly engineering drop; not
fully validated
|
3998
|
|
December 22, 2004
|
Pre-Alpha 14.17.0.3993
|
Weekly engineering drop; not
fully validated
|
3993
|
|
December 16, 2004
|
Pre-Alpha 14.17.0.3987
|
Weekly engineering drop; not
fully validated
|
3987
|
Intel®
Graphics Media Accelerator Driver
The Intel® Graphics Media
Accelerator Driver contains support for the following Intel Chipsets:
- Intel(R) 852GM Chipset
- Intel(R) 852GME Chipset
- Intel(R) 855GM Chipset
- Intel(R) 855GME Chipset
- Intel(R) 915G Express Chipset
Family
- Intel(R) 915GV Express
Chipset Family
- Intel(R) 915GM/GMS Express
Chipset Family
- Intel(R) 910GML Express Chipset
Family
- Intel(R) 945G Express Chipset
Family
- Intel(R) 945GM Express
Chipset Family
New
Features
|
Features
Added in Baseline 14.18.0
|
|
Reference
No.
|
Description
|
Affected
Component(s)
|
Build
|
|
N/A
|
Silicon Image
HDCP parts require additional opcodes to function correctly. The
parts will not work with the original implementation described in ver. 0.9
of the HDCP SAS. This change will add the optional opcodes required by the
Si Image parts
|
DVD
|
4361
|
|
N/A
|
Method
to remove/prevent any specified TV Output Format from being available
|
AIM
|
4361
|
Known Issues
|
Known
Issues in Build Number 4421
|
|
Reference
No.
|
Description
|
Affected
Component(s)
|
Affected
OS(s)
|
Affected
Project's)
|
|
82222
|
The
TestPowerAPI error message appears after opening the IGT page and then
launching TestPowerAPI application with the modified BIOS.
|
N/A
|
Windows*
2000
|
Intel(R)
945GM Express Chipset
|
|
82327
|
No FS DOS
in HDCP or Non-HDCP DFP with Chrontel CH7312 HDCP video card
|
N/A
|
Windows*
Media Center
|
Intel(R)
945GM Express Chipset
|
|
82418
|
Serious Sam The Second Encounter
- The system lags while playing the Serious Sam game under Windows XP.
|
N/A
|
Windows*
XP
|
Intel(R)
945GM Express Chipset
|
|
82613
|
CUI2 /
CUI (tray icon, right click, hotkey) not accessible after reinstall w/ OS
HW acceleration set to none.
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(R)
945GM Express Chipset
|
|
82511
|
Running
selected conformance lighting tests on MGM would result in tests failing
|
OGL
|
Windows*
XP
|
Intel(R)
955GM Express Chipset
|
|
82218
|
The
created schemes disappears in DTCM \ Scheme submenu after setting CRT+EFP
EXTD scheme with CRT at 90 degree view and then setting CRT+EFP DDC 90
degree scheme.
|
CUI2
|
Windows*
XP
|
Intel(R)
945GM Express Chipset, Intel(R) 945G Express Chipset
|
|
74079
|
The
Program Error error message is displayed and the application halted after running
the cui_sdk_devattach.rb application with CRT, TV and EFP.
|
CUI2
|
Windows*
XP
|
Intel(R)
915GM Express Chipset , Intel(R) 945GM Express Chipset, Intel(R) 945G
Express Chipset
|
|
80374
|
If Intel®
Smart 2D Display Technology (a.k.a. Frame Buffer Compression or FBC) is
enabled then the monitor may not resume its display when returning from
Power Off state as defined by the operating system Control Panel / Power
Options setting, when Monitor Power Option is enabled and System is left
idle and the system enters a power savings state
|
N/A
|
Windows*
XP
|
Intel(R)
945GM Express Chipset
|
Issues Resolved
|
Issues Resolved in 4421
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1699387
|
In
720p with scaling mode, the picture is not centered precisely.
|
AIM
|
Windows* XP, Windows* Media Center,
Windows* 2000
|
Intel(R)
945G Express Chipset
|
|
Resolution Description:
Root Cause: This is a CH hardware bug related to the default
position values for 720p format. We have put a workaround for this hardware
issue.
Fix Description:
|
|
BugID: 1669185
|
Jittery/jerky
DVD playback when Adaptive Deinterlacing is enabled.
|
DVD
|
Windows* XP, Windows* Media Center
|
Intel(R)
945GM Express Chipset, Intel(R) 915G Express Chipset, Intel(R) 945G
Express Chipset
|
|
Resolution Description:
Root Cause: This bug is a side effect of a VRM-9 bug (bad forward reference
frames), which causes A-DI to receive bad reference frames for
deinterlacing.
Fix Description: Introduced a workaround for the VMR-9 bug, which
includes detection and handling of progressive/interlaced discontinuities.
|
|
BugID: 1703367
|
Dithering
does not work after resume from S3.
|
AIM
|
Windows* XP
|
Intel(R)
945G Express Chipset
|
|
Resolution Description:
Root Cause: Get Dither opcode returns current value and Default
value, but this is interpreted as max Value and default value in AIM
module.
Fix Description: Corrected this behavior appropriately.
|
|
BugID: 1701536
|
IGDI
(Intel Graphics Driver Installer ) Command Line Option does not work
correctly.
|
MINIPORT
|
Windows* XP, Windows* 2000
|
Intel(R)
915GM Express Chipset
|
|
Resolution Description:
Root Cause: The installer creates the following keys: RequestedBPP,
RequestedRate, RequestedXRes, and RequestedYRes. The driver reads them
correctly. It has to store the values in the registry at a) “HKLM\machine\System\CurrentControlSet\HardwareProfiles\Current\System\Current
ControlSet\Services\iAlm\Device0" and
b)
“HKLM\System\CurrentControlSet\Hardware
Profiles\Current\System\Current-ControlSet\Services\iAlm\Device0\Mon(display
device number)"
Driver
then reads from (b) to (a) to apply the display settings. The current
driver is not writing to (b) in the first place. In the working version
(4352), the installer key values were written to all enumerated devices.
This portion of the code was removed due to an earlier fix.
Fix Description: The registry in (b) is written to for all devices
that were active at boot time and the display configuration registry is
also appropriately updated.
|
|
BugID: 1741544
|
Desktop blacks
out after HW accelerator is disabled with MS property.
|
GDI
|
Windows* XP
|
Intel(R)
945GM Express Chipset
|
|
Resolution Description:
Root Cause: When we reduce Hardware Acceleration to none, we get DrvCopyBits
call with source PDEV not activated. The issue happens because DrvCopyBits
returns TRUE without actually doing anything since source Pdev is not
activated. We can punt the call to GDI if we could not handle. Issue does
not happen when we punt the call instead of returning TRUE.
Fix Description:
|
|
Issues Resolved in 4414
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1616430
|
DCT5.2/DCT5.3 - ALV/GDG - Display Stress test & HCT12.1 System
Stress test failed. (*.exe tests reported as "DIED")
|
GMM, MINIPORT
|
Windows* XP, Windows* XP-64
|
Intel(R)
855GM Chipset
Intel(R) 865G Chipset
Intel(R) 915G Express Chipset
Intel(R)
915GM Express Chipset
Intel(R)
945G Express Chipset
Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: VideoportUnmapMemory is
called with undocumented “-1” as current process handle which is a
different handle process from VideoportMapMemory. As a result, it caused “access
violation.” (i.e. Process died.) Fix
Description: Calling
undocumented ZwOpenProcess to translate current process ID to handle and
unmap it with original process handle.
Map and Unmap memory are always the same process.
|
|
BugID: 1619764
|
DVD - PowerDVD 6– Contrast & Saturation Video Overlay
Settings are inconsistent when enabling hardware acceleration.
|
DD
|
Windows* 2000, Windows* XP
|
Intel(R)
915G Express Chipset, Intel(R) 945GM
Express Chipset, Intel(R) 945G Express Chipset
|
|
Resolution Description:
Root Cause: In contrast and saturation,
a variable used to decide an upper or lower range of value conversion from
DD to hardware was incorrect. With the wrong range of conversion, negative
numbers are produced with the CUI slide bar moved to the scale less than
40. As a result, the contrast reset the bright color after the CUI slide
bar passed 40.
Fix Description: This variable should
be replaced by another variable which actually reflects the DD value
adjusted from CUI.
|
|
BugID: 1698861
|
The butterfly on secondary display is displayed at 1024x768 desktop
resolution with CRT+EFP DDC mode.
|
DD
|
Windows* 2000, Windows* XP
|
Intel(R)
945GM Express Chipset, Intel(R) 945G Express Chipset
|
|
Resolution Description:
Root Cause: The VBT was not supported
before a certain version of desktop system. In DDraw driver, the desktop
system did not take any VBT info set from BIOS. That is why the display did
not follow the VBT bit set in BIOS.
Fix Description:
SoftBios validates the VBT info and reset bEnableSpriteInDPSIM. If it is
not supported in VBT, then it keeps the value as retrieved from VBT block.
|
|
Issues Resolved in 4407
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1655899
|
DVD
playback only on Primary Display in Clone Mode.
|
DD
|
Windows* 2000, Windows*
XP
|
Intel(R)
915GM Express Chipset
|
|
Resolution Description:
Root Cause: If we are in double-wide mode, then
reject overlay surface creation in CanCreateSurface.
Fix Description:
|
|
BugID: 1682979
|
Display
does not come on TV even though ACPI static bit is set.
|
TVOUT
|
Windows* XP
|
Intel(R)
915GM Express Chipset
|
|
Resolution Description:
Root Cause: The dac state was not initialized hence
IntTV_IsModeSupported was returning failure.
Fix Description: Added the check
for hotkey bit in IntTV_IsOutputDeviceAttached, and initialize the dac
state.
|
|
BugID: 1696752
|
<PC14.19>Functional
Failure / GDI - The blank screen is displayed on CRT when switching it via
Ctrl+Alt+F1 keys after disconnecting CRT,EFP,TV on EFP+TV DDC mode when
system enters Suspend state then reconnecting CRT,EFP,TV back after
resuming.
|
SOFTBIOS
|
Windows* XP
|
Intel(R)
945GM Express Chipset
|
|
Resolution Description:
Root cause: m_eCurrentDisplayPowerState is not getting updated
during resume from standby when CRT is unplugged.
Fix Description: Modified
IntCRTEncoder.c to fix the problem.
|
|
BugID: 1701387
|
Cannot
change to extended display mode when TV is hot plugged.
|
CUI2
|
Windows* XP
|
Intel(R)
945GM Express Chipset
|
|
Resolution Description:
Root Cause: The global variable
g_InvalidPrimaryDevices was still showing TV in it (the hotplugged device
in this case), fixed this appropriately by removing the global variable and
adding a local variable dwInvalidPrimaryDevices and calling the
GetInValidPrimaryDevi
Fix Description:
|
|
BugID: 1684700
|
Persistance
settings are not working correctly when changing displays through CUI.
|
CUI2
|
Windows* XP
|
Intel(R)
945GM Express Chipset
|
|
Resolution Description:
Root Cause:
Unabled to load the data from registry because of global key. Storage
of persistance data in registry is cleaned. ETM was completed in second
attempt on 10/07/2005.
Fix Description:
|
|
BugID: 1660593, 1702534
|
CUI2
- The EXTD saved scheme cannot be displayed on DTCM and TICM, and the Apply
Button is grayed out while selecting the EXTD saved scheme.
|
CUI2
|
Windows* XP
|
Intel(R)
915GM Express Chipset, Intel(R) 945GM Express Chipset, Intel(R) 945G
Express Chipset
|
|
Resolution Description:
Root Cause: The settings does not apply before saving the scheme.
Fix Description:: Fixed the bug appropraitely.
|
|
Issues Resolved in 4398
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1684540
|
When
TV / CRT connected at the same time and then switched to TV only, the
screen flicked.
|
SOFTBIOS
|
Windows* XP
|
Intel(R)
945GM Express Chipset
|
|
Resolution Description:
Root Cause: The dotclock programmed into the GMCH registers was different
from the dotclock returned to KCH for watermark calculation. This resulted
in wrong watermark values and sever flicking on TV.
Fix Description: Sent the correct dotclock from SB_GetTimingInfo to
KCH. Added functions util_Get_CurrentTimingDetailsFromMMIO and
clk_Napa_GetDotClockFromMMIO to be used in the future to obtain the
dotclock from GMCH registers.
|
|
BugID: 1686423
|
PM - The
640x480x8bit resolution is displayed in IGT after changing the rotation
view from 0 degree to 90 degrees at 480x640x8bit desktop resolution with
CRT+EFP DDC mode.
|
CUI
COM
|
Windows* 2000
|
Intel(R)
945GM Express Chipset,Intel(R) 945G Express Chipset
|
|
Resolution Description:
Root Cause: X and Y modes are reversed in combinedmodelist().Root
Cause:
Fix Description: Fixed the X and Y modes are reversed in
combinedmodelist().
|
|
BugID: 1642597
|
When
trying to apply a saved scheme with Full Screen TV Option checked, TV
display with TV option is unchecked.
|
CUI2
|
Windows* XP-64
|
Intel(R)
915GM Express Chipset
|
|
Resolution Description:
Root Cause: Schemes was not comprehending Full Screen Option and TV
standard. The scheme gets applied, but the Full Screen checkbox remains
unchecked
Fix Description: Made the changes appropriately.
|
|
BugID: 1684696
|
W2K/WXP
: OEMDSPIF.DLL does not support DVI
devices on Davinci system.
|
CUI
SDK
|
Windows* 2000, Windows*
XP
|
Intel(R)
945GM Express Chipset
|
|
Resolution Description:
Root Cause: DFP support was not added to OEMDLL when it was written
originally.
Fix Description: Added DFP support to the code. Basically, BIT3
which stands for DFP was not being processed by the problematic functions.
|
|
Issues Resolved in 4397
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1643277
|
PREfast
erros on DDRAW/GHAL.
|
DD,
DVD, GDI
|
Windows* XP
|
Intel(R)
915GM Express Chipset
|
|
Resolution Description:
Root Cause: Warning message appeared from
compiling with Prefast. The causes are from ambiguous and redundant
variable declarations, wrong memory copy size, and uninitialized variables.
Fix Description: Optimized the defected codes to
pass the Prefast detection.
|
|
BugID: 1681665
|
Fixed
3DMark '03 issues during Mother Nature Test with 64MB DVMT.
|
GMM
|
Windows* XP
|
Intel(R)
915GM Express Chipset
|
|
Resolution Description:
Root Cause: Only on DVMT 64 setting: When we ran out of tiled memory
in lockable space, we switched to allocating linear from lockable space. As
a result, we ran out of linear quickly.
Fix Description: Prevented linear allocations from lockable space.
|
|
BugID: 1664872
|
WXP
: Extended desktop is enabled after suspend/resume even though CRT is not
attached to the system.
|
ACPI,
POWER MANAGEMENT, CUI, CORE DRIVER
|
Windows* 2000, Windows*
XP
|
Intel(R)
915GM Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: This issue happens when a system is in EDT, and the
system is suspended with one device disconnected. When it resumes, the
system stays in EDT. Since there is only one device available, the other
device becomes a NULL device. OS (and driver) does not know when to remove
the system out of EDT.
Fix Description: (i) The driver logic set the registry value
to SPSD when resuming from S3/S4/S5 if the primary/secondary/both display
devices are not connected. The logic has been changed, so the registry
value is set only if the earlier display configuration was not Extended
Desktop. (ii) The workaround in driver code forces the OS out of EDT when
entering default single pipe mode when resuming from hibernate has been
removed. 2) A workaround is provided in CUI to bring the system to SPSD
configuration on resume from standby if the system is in MDS and the
primary device / secondary device / both devices are not connected.
|
|
BugID: 1670106
|
Video
files cannot be displayed when rotate from 90 degrees to 180 degrees or 270
degrees to 180 degrees.
|
ROTATION
|
Windows* XP
|
Intel(R)
945G Express Chipset
|
|
Resolution Description:
Root Cause: At some point, our driver gets a delayed call for DrvDisableDirectDraw,
because the presence of 3D pipe screen saver preview. The rotation buffers
and the data structure( Surface descriptors) needed would have created by
that time. This rotation specific data structures get destroyed in Rotation
driver’s DrvDisableDirectDraw. Before destroying pDDSurfDesc, rotation
driver does not do a check whether it is belonging to pdev for which
DrvDisableDirectDraw call has come. Root Cause:
Fix Given: Before destroying pDDSurfDesc, rotation driver now does
a check whether it is belonging to pdev for which DrvDisableDirectDraw call
has come.
|
|
BugID: 1685671
|
Missing
Functionality / Miniport - There is no LogResumeTime key in Registry after
selecting "Delete existing settings" in Driver Verifier Manager
and then installing the graphics driver.[REG]
|
INF
|
Windows* XP
|
Intel(R)
945GM Express Chipset
|
|
Resolution Description:
Root Cause: Missing Functionality /
Miniport - There is no LogResumeTime key in Registry. Fix Description: Add
registry to INF.
Fix Description:
|
|
BugID: 1684540
|
When
TV / CRT connected at the same time and then switched to TV only, the
screen flicked.
|
SOFTBIOS
|
Windows* XP
|
Intel(R)
945GM Express Chipset
|
|
Resolution Description:
Root Cause: The dotclock programmed into the GMCH registers was
different from the dotclock returned to KCH for watermark calculation. This
resulted in wrong watermark values and sever flicking on TV. Fix Given:
Sent the correct dotclock from SB_GetTimingInfo to KCH. Added functions
util_Get_CurrentTimingDetailsFromMMIO and clk_Napa_GetDotClockFromMMIO to
be used in the future to obtain the dotclock from GMCH registers.
Fix Description:
|
|
BugID: 1642597
|
When
trying to apply a saved scheme with Full Screen TV Option checked, TV
display with TV option is unchecked.
|
CUI2
|
Windows* XP, Windows*
XP -64
|
Intel(R)
915GM Express Chipset
|
|
Resolution Description:
Root Cause: schemes was not comprehending Full
Screen Option and TV standard. Fix Given: Made the changes appropriately.
Fix Description:
|
|
BugID: 1698250
|
Visual
Anomaly / CUI2 - Dual Independant Displays - "y" in
"Display" clipped in CUI2 when two DFPs are used in DDC/DFP2-DFP
mode.
|
CUI2
|
XP
|
Intel(R)
915GM Express Chipset
|
|
Resolution Description:
Root Cause: Fixed alignment for tab text.
Fix Description:
|
|
BugID: 1571077
|
D1
(HDTV_480i59) cannot set to another mode under Device Settings Property.
|
CUI2
|
XP,
MCE
|
Intel(R)
915G Express Chipset
|
|
Resolution Description:
Root Cause: This is added as a new feature. The Integrated TV-Out
does not cover this even after HDTV RCR is done. Fix Given: Flipper logic now
implemented for both Primary / Secondary devices as HDTV. Also fixed
Hidemodes Unchecked issue due to HDTV workaround, Fixed 15 Sec message box
not appearing issue.
Fix Description:
|
|
Issues Resolved in 4395
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1686121,
1686114
|
Fix for Build# 4392 which addresses the two following Bugs,
1686121 and 1686114.
|
OGL
|
Windows*
XP
|
Intel(R)
855GM Chipset
Intel(R) 865G Chipset
Intel(R) 915G Express Chipset
Intel(R)
915GM Express Chipset
Intel(R)
945G Express Chipset
Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: This fix checks if buffer in
__glSrvBufferCreate() in /glsrv/srvdraw.c(600) is not NULL before
dereferencing. If a non-OpenGL app
is in exclusive mode and the buffer is NULL, it causes a crash when it is
dereferenced.
Fix Description:
|
|
BugID: 1682981
|
A white line appears after Windows Logo.
|
SOFTBIOS
|
Windows*
XP
|
Intel(R)
915GM Express Chipset
|
|
Resolution Description:
Root Cause: The LVDS port control
register was overwritten with the VBT register value, this resulted in 31st
bit getting reseted and flicking.
Fix Description: Added a mask for port
control and panel fitter register for bits that should not be taken from
the VBT.
|
|
BugID: 1667329
|
Corruption / Miniport - The corruption is displayed on IGT page
after resuming from suspend state (S3) on DDT mode.[64-bit REG]
|
GDI
|
Windows*
2000, Windows* XP, Windows* XP-64
|
Intel(R)
945G Express Chipset
|
|
Resolution Description:
Root Cause: When the COM_GET_MODE_TABLE
call comes for twin, GDI receives two UIDs and compares these UIDs with the
UIDs stored in pSharedInfo to check wether it is current config are not.
The assumption is that UID1 is the primary device and UID2 is the
secondary. This assumption is wrong. There is no concept of
primary/secondary in case of clone/twin. We fail the call by sending NULL
mode list, and CUI was displaying no configuration.
Fix Description:
|
|
BugID: 1682762
|
[Clone mode]Can not change the refresh rate when the device is
replaced.
|
CUI2
|
Windows*
XP
|
Intel(R)
915GM Express Chipset
|
|
Resolution Description:
Root Cause: This is a data
initialization issue.
Fix Description:
|
|
BugID: 1637928
|
Integrated: IGT page displays Monitor item at
800x600x32bit@60Hz in the Display Devices when connecting with only TV and
LFP.
|
SOFTBIOS
|
Windows*
XP
|
Intel(R)
945GM Express Chipset
|
|
Resolution Description:
Root Cause: 1)
util2_IsCRTHotPlugEnabled() reads the PORT_HOTPLUG_ENABLE_REG register and
checks 10th bit (BIT9) to see if integrated CRT hotplug is enabled. It
assigned to a BOOLEAN variable, bIsCRTHotplugEnabled. This BOOLEAN is
defined as UCHAR and the abovebit check results in the 10th bit getting
set, it always gets assigned to FALSE. 2) A known limitation that Legacy
CRT detection fails when done on the same pipe on which TV is being driven
at 800 X 600 mode, due to some timing limitations. Hence, a workaround was
introduced as part of the routine util_Analog_CRT_Sense_Pipe_A(), which
uses a pipe other than the one on which TV is being driven for legacy CRT
detection. However, this section of code is not enabled for Calistoga. Root
Cause:
Fix Description: Initialized local
variable ulReg in util2_IsCRTHotPlugEnabled(). 2) Assign
bIsCrtHotplugEnabled with an explicit check – if the bit check has a value
zero, assign FALSE. For any other value, assign TRUE. 3) Enabled the
workaround mentioned above for Calistoga.
|
|
Issues Resolved in 4388
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1643271
|
PREfast Errors on D3D.
|
D3D
|
Windows*
XP
|
Intel(R)
915GM Express Chipset
|
|
Resolution Description:
Root Cause: 1) Warning 11:
Dereferencing NULL pointer ' pRenderTarget.' 2) Warning 8104: Possiblely
leaking memory 'lpDDTermProcData.' 3) Waring 8110: "Driver must protects
floating point hardware state."
Fix Description: 1) Fixed in
clear.h. 2) There is no memory leaking regarding 'lpDDTermProcData'. 3) we covered all
floating point operation and verified by debugging into each function call.
Floating point hardware state has been save/restore correctly before/after
floating point operation. Remaining 8110 warnings are consider as
"Noise". We may need to contact Microsoft.
|
|
BugID: 1682149
|
2XXP -- Closing out of CUI does not allow scan codes through to
the system.
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(R)
915GM Express Chipset
|
|
Resolution Description:
Root Cause: Windows passes
window message to window even after window is destroyed. Root Cause:
Fix Description: Even if WIndows passes
window message to window after window is destroyed, a check,
IsWindowVisible(), is made and set focus is called.
|
|
BugID: 1621623
|
Functional Failure/Rotation still enabled on LPT after BMP.
|
CUICOM
|
Windows*
XP
|
Intel(R)
945G Express Chipset
|
|
Resolution Description:
Root Cause: When changing
the BMP option, if the same Bios is used, the driver needs to be updated.
If a different bios version is used, everything works properly. This is a
limitation due to display1_enableRotation key will not get deleted, and
there is no way that CUI can identify this situation.
Fix
Description: Migrates
bug fix from 14.x to CUI2.
|
|
BugID: 1642750
|
IN ETM, Configuring Desktop in EDS, and changing the resolution
of sec on closing lid to DN display on LFP is seen.
|
CUI2
|
Windows*
XP
|
Intel(R)
915GM Express Chipset
|
|
Resolution Description:
Root Cause: In one path, CUI COM was not
following the MS specified call sequence to set EDT.
Fix Description:
Implemented MS specified call sequence to set EDT.
|
|
Issues Resolved in 4386
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1619161
|
System
reboots with BSOD when the video game, Mabinogi, is loaded.
|
D3D
|
Windows*
XP
|
Intel(R)
855GM Chipset
Intel(R) 865G Chipset
Intel(R) 915G Express Chipset
Intel(R)
915GM Express Chipset
Intel(R)
945G Express Chipset
Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: DXT5 surface compression is not handle properly in
function, TextureAliasDestAndBlt.
Fix Description: Disabled compression of DXT5 surfaces.
|
|
BugID: 1652871, 1655806
|
Aspen-U:
FN+F4 does not work after resuming from S3 intermittently. (1652871)/ 2KXP
FN+F8 will not exit Extended desktop.
Extended Desktop display locations are not restored after reboot or
PM event. (1655806)
|
CUI2
|
Windows*
XP
|
Intel(R)
915GM Express Chipset
|
|
Resolution Description:
Fix Description: (1) In HotKeyImpl.cpp, function,
ChangeActiveState(), is modified to address this issue and also to switch
display only when it is not already in the target display. When registry
entry is not available for the arget device, and persistence is enabled,
CUI queries driver for the optimal mode and sets the optimal mode. (2) When
a ACPI/SMI hotkey event is received by CUI, if persistence is enabled, if
registry entry is not available for the target device, CUI queries driver
for the optimal mode and sets the optimal mode. (3) This fix is implemented
currently only for hotkey event, and for SPSD mode. Dual display mode will
be handled later, under the first boot RCR.
|
|
BugID: 1662198
|
DspConfigSwap_Prisec
- SoftBIOS - Swapping primary and secondary display device for Device1 +
Device2 EDT with 180 degrees and 90 degrees will change the resolution for
Device 1.
|
CUI2
|
Windows*
XP
|
Intel(R)
915GM Express Chipset, LAKEPORT
|
|
Resolution Description:
Root Cause: In UpdateRefreshRate(), mode swapping is not done if
devices are swapped in the devices page.
Fix Description: In UpdateRefreshRate(), the mode swapping is done
if the Devices are swapped in the devices page now.
|
|
BugID: 1666378
|
After
switch to TV-Out and attach to a TV later, TV does not turn on.
|
SOFTBIOS
|
Windows*
XP
|
Intel(R)
915GM Express Chipset
|
|
Resolution Description:
Root Cause: In SB_GetCloneModeListFromEdid, when a request for
TV+<any device> comes, it fails because there is no mode table
representing the TV. In passing drivers, it uses to create a combined mode
table and the issue was not seen.
Fix Description: Consider the static bit when
IntTV_IsOutputDeviceAttached is called.
|
|
BugID: 1666735
|
Cineplayer
3.0 - BSOD when DXVA is activated.
|
DVD
|
Windows*
XP
|
Intel(R)
945G Express Chipset
|
|
Resolution Description:
Root Cause: The subtitle-blending function terminated due to the
wrong passed-in data. Also the batch buffer was sent out by another function
with a wrong HW context bound unexpectedly.
Fix Description: An enhancement was done in the subtitle-blending
function. It will send out the batch buffer whenever it is not empty.
|
|
BugID: 1669357
|
In
ALVISO, corruption seen on SDVO-LVDS when TWIN+full centering is selected.
|
SOFTBIOS
|
Windows*
XP
|
Intel(R)
915GM Express Chipset
|
|
Resolution Description:
Root Cause: For SDVO-LVDS, TWIN mode should not be enabled with
scalar enabled
Fix Description As in desktop, do not pass the stretch flag to AIM
irrespective of whether user selects full screen option or not.
|
|
Issues Resolved in 4382
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1642437
|
ACP
signal is not seen when you change the standard in TV.
|
AIM
|
Windows*
XP
|
Intel(R) 865G Chipset
|
|
Resolution Description:
Root Cause: While setting timing, the
encoder is initialized without verifying whether macrovision is to be
turned off or turned on.
Fix Description: Turn on the macrovision signal
after initialization of encoder.
|
|
BugID: 1655977
|
Changing
DDC:LFP+DFP In FSDOS with lid closed and DoNothing display switch to CRT
and lid-open in DDC:CRT+LFP.
|
MINIPORT
|
Windows*
XP
|
Intel(R) 915GM Express Chipset
|
|
Resolution Description:
Root Cause: When we go to FSDOS in DDC, LFP
+ DFP, the display comes on LFP and the active device list contains LFP
only. When lid closed, expected
fbehavior is that display should come on DFP since it is one of the active
devices but DFP is not present in the active device list. Since there is no active device other
than LFP and the lid is closed, next device is selected based on the priority in this case, the display goes
on CRT but not DFP. When lid is
opened, the display comes on CRT + LFP, because CRT is now one of the
active devices:
Fix Description: In FSDOS scenario, the active device list from
SoftBios does not give a true picture of the active devices. Instead of looking into the active device
list from SoftBios, we are looking into the registry for the active
devices. In FSDOS (LFP + DFP), even though the display is seen in LFP only,
the registry will still have LFP as primary device and DFP as secondary
device. When lid is closed, in FSDOS
(DDC:LFP + DFP), the registry shows DFP as active device and switch to
DFP. When lid is opened, since DFP
was one of the active devices, we come up in DFP + LFP.
|
|
BugID: 1664496
|
CRT
blanks out when switching from extended desktop to DDC with 180 degrees
rotation and movie running.
|
GDI
|
Windows*
XP
|
Intel(R)
915GM Express Chipset
Intel(R)
945G Express Chipset
|
|
Resolution Description:
Root Cause: The occurrence of
the issue depends on sequence of EnableSurface and DisableSurface call that
OS sends. So the issue is sporadic.
Fix Description:
|
|
BugID: 1662745
|
The
windows pop up message could not show in "Launch Zoon" window by VGA
utility, but it is OK by windows xp built-in Zoom utility.
|
CUI2
|
Windows*
XP
|
Intel(R) 915GM Express Chipset
|
|
Resolution Description:
Root Cause: Fixed by adding
CAPTUREBLT to stretchblt
Fix Description:
|
|
BugID: 1658248
|
SMI
090D is not called in a particular case.
|
CUI2
|
Windows*
XP
|
Intel(R) 915GM Express Chipset
|
|
Resolution Description:
Root Cause: Send the NotifySystemBios 090D call even when the Get
Current Mode fails (as in CUI1)
Fix Description:
|
|
BugID: 1655925
|
WM_DISPLAYCHANGE
message is sent just before the system goes on S3 or S4.
|
CUI2
|
Windows*
XP
|
Intel(R) 915GM Express Chipset
|
|
Resolution Description:
Root Cause: Currently, WM_DISPLAYCHANGE message is send from
igfxpers.exe to igfxcfg.exe and updating itself after events like Hot plug, Lid switch,
Power ,and etc. But many OEMs are complaining their apps are affected by
the WM_DISPLAYCHANGE message broadcasted by igfxpers.exe.
Fix Description: Instead of broadcasting windows
message WM_DISPLAYCHANGE, igfxpers.exe will broadcast a CUI defined message
which is registered with Windows. Igfxcfg will process the message just
like WM_DISPLAYCHANGE message. Also it will send the message to advanced
windows.:
|
|
BugID: 1626358
|
CUI2 -
In limited user accounts, empty tabs in CUI causes CUI to misreport display
device configurations.
|
CUI2
|
Windows*
2000, Windows* XP
|
Intel(R)
855GM Chipset
Intel(R) 865G Chipset
Intel(R) 915G Express Chipset
Intel(R)
915GM Express Chipset
Intel(R)
945G Express Chipset
Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: Made the selection of empty list
box items non-selectable as is with any window control in the custom
control code.
Fix Description:
|
|
BugID: 1556152
|
Display
corruption or blank screen after lid close and open in extended desktop
with CRT+TV
|
MINIPORT
|
Windows*
XP
|
Intel(R)
915GM
|
|
Resolution Description:
Root Cause: Miniport function MD_Lid_Validation (mutidisp.c) allowed
display switches on all lid close events, even when the system was in
Extended Desktop with two non-LFP displays.
This would result in the existing configuration being disturbed and
possible loss of primary display. Root Cause:
Fix Description: Closing the lid will no longer force a display
switch when the system is in Extended Desktop using two non-LFP displays.
|
|
Issues Resolved in 4374
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1623449
|
System
turns to blue screen when playback MediaPlayer and set to Dual Mode with
Same display configuration device on both displays.
|
GDI
|
Windows*
XP, Tablet PC
|
Intel(R)
915GM
|
|
Resolution Description:
Root Cause: psharedinfo->dwconfiguration gets set to CONFIG_MDS
but not CONFIG_DPSM due to a check of
(pdevicesharedinfo[0]->!=NULL) in DrvEnableSurface, which sets
the pshaedinfo->dwconfiguration field to CONFIG_MDS if the condition is
true.
Fix Description: The fix is to remove the wrong check-in
DrvenableSurface()
|
|
BugID: 1643280
|
Aspen
UMA PV1 M4, 4325: Res of first CRT attached is used for subsequent CRTs
attached.
|
CUI2,
GDI, MINIPORT, SOFTBIOS
|
Windows*
XP
|
Intel(R)
915GM
|
|
Resolution Description:
Root Cause: This behavior is expected from current design.
Fix Description: 1) In HotKeyImpl.cpp function ChangeActiveState()
is modified to address this issue and switch display only when it is not in
the target display. When registry entry is not available for the target
device and persistence is enabled, CUI queries driver for the optimal mode
and sets the optimal mode. 2) When a ACPI/SMI hotkey event is received by
CUI, if persistence is enabled but registry entry is not available for the
target device, CUI queries driver for the optimal mode and sets the optimal
mode. 3) This fix is implemented currently only for hotkey event and for
SPSD mode. Dual display mode will be handled later, under the first boot
RCR.
|
|
BugID: 1660162
|
After
resume from S3/S4, TV setting is not restored.
|
TVOUT
|
Windows*
XP
|
Intel(R)
945GM
|
|
Resolution Description:
Root Cause: In IntTV_GetEXTVParameters, the parameter pEXTV is
memset to 0. This resulted in aim_GetSetParameters to overwrite the GUID with
zero. When we resume from S3/S4, the GUID is corrupted, the correct value
is not obtained, and the default value is restore.
Fix Description: Corrected the function IntTV_GetEXTVParameters.
|
|
BugID: 1657302
|
CUI
customization with language specific does not work.
|
CUI2
|
Windows*
XP
|
Intel(R)
915GM
|
|
Resolution Description:
Root Cause: The string, IDS_LANGUAGE, was not added to the igfxres
resource file, so the language was not picked in the loadstring method.
Fix Description: Added the string IDS_LANGUAGE to the igfxres
resource file.
|
|
Issues Resolved in 4371
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID:
1642585
|
3D Screen
saver does not work in LID closed extended desktop environment.
|
GDI
|
Windows*
XP, Tablet PC
|
Intel(R)
915GM Express Chipset, Intel(R) 915G Express Chipset, Intel(R) 855GM
Chipset
|
|
Resolution Description:
Root Cause: This happens when display config. is MDS with LFP+CRT
with lid closed and screen launched. When the screen saver tries to set the
secondary/primary mode by calling the Enable Surface, the Enable Surface
call fails because there are no devices attached for the secondary. Then OS
takes system out of MDS by calling disablesurface(), but screen saver still
thinks that it is in extended desktop. So screen saver does not launch.
Either a blank screen or screen saver not launching but shows the
background desktop flicking
Fix Description: The fix is to pass the enablesurface() call for
secondary, even if we do not have any displays, provided we are already in
extended desktop. If we are already in extended desktop and getting calls
for enablesurface(), it means a change in the mode is needed. Select
“Virtual Display,” if we do not have any devices and pass the call. If we
are not already in the extended desktop, the call will be failed. This
check is to avoid going to MDS from SPSD or any such configuration if we do
not have any displays.
|
|
BugID: 1579540
|
Remove
WA for Calistoga A1 and up. GDG system hangs when running multiple SDK
apps. (Billboard in classic & Text3D in inline mode)
|
D3D
|
Windows*
XP, Windows* XP -64
|
Intel(R)
855GM Chipset
Intel(R) 865G Chipset
Intel(R) 915G Express Chipset
Intel(R)
915GM Express Chipset
Intel(R)
945G Express Chipset
Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause:
Fix Description: Fix SyncFlushOp and add hardware workaround. Hardware WA
added to stop ring prior to issue a SyncFlush, then resume ring after
SyncFlush.
|
|
BugID:
1655439
|
TV
contrast setting is only 10 steps.
|
SOFTBIOS
|
Windows*
XP
|
Intel(R)
915GM Express Chipset
|
|
Resolution Description:
Root Cause: The hardware does not have the capability to do floating
point calculation depending on the no contrast level that we selected.
Current implementation statically increses the contrast in steps of 10,
although the capability reported to CUI states the stepping is 1.
Fix Description: : In IntTV_SetContrast, we changed the
implementation to dynamic calculation of contrast value based upon the
contrast stepping requested from CUI. The range for constrast is still
between 0.2 to 2.
|
|
BugID: 1660291
|
CUI2 -
The CRT screen position moves down after clicking the up position button
with MCCS Monitor.
|
CUI2
|
Windows*
XP
|
Intel(R)
915GM Express Chipset
|
|
Resolution Description:
Root Cause: Wrong flags were used.
Fix Description: Fixed it by changing to the correct flags.
|
|
BugID: 1641750
|
In
Extended Desktop, display goes blank when DFP hotplugs.
|
CUI2
|
Windows*
XP
|
Intel(R)
915GM Express Chipset
|
|
Resolution Description:
Root Cause: EnumDisplaySettings() OS API call fails or passes after
enough time delay after a device hot plugged.
Fix Description: Retry EnumDisplaySettings () until it passed, this
is a pure workaround and need to be removed when the driver becomes stable
on hot plug.
|
|
Issues Resolved in 4365
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1640291, 1654192
|
TV-Out
shows black screen by some TV setting operations.
|
AIM
|
XP
|
Intel(R) 945G Express Chipset
|
|
Resolution Description:
Root Cause: 7021 card had hardware issues with overscan in 1080i/p
with 17x9 and 19x10, and 2.)When display is in 720p and overscan
option is checked, TV blanks out
when Restore Defaults button is clicked.
Fix Description: Disable overscan in 1080i/p with 17x9 and 19x10
resolutions
|
|
BugID: 1640528
|
Transient
green grabage displays when Live TV, Video Playabck, and DVD is launched in
Media Center.
|
DD, DVD
|
XP, MCE
|
Intel(R) 945GM Express Chipset, Intel(R) 945G Express Chipset
|
|
Resolution Description:
Root Cause: OS is using the YUY2 texture surfaces without initialize
them. During LiveTV, DVD, or pre-corded video playback in MCE 2005, the
transient green (sometimes also pink squares) display for second when
switching from window viewing to full screen. Expectation is a solid black
screen.
Fix Description: We decided not to fix the bug because the OEM that
wanted it is happy enough that the next version of MCE does not show this
issue. This issue was a MS problem
becuase they did not choose to initialize their memory before they
displayed it to the screen.
|
|
BugID: 1646770
|
Garbage shows on screen while testing Coredraw with Pivot
mode.
|
ROTATION
|
Windows*
2000, Windows* XP
|
Intel(R)
945G Express Chipset
|
|
Resolution Description:
Root Cause: This is introduced due to DrvStrokePath optimization in
rotation. The optimization takes each line in path object, rotate the
coordinates and draws the line to the Frame buffer. While converting the
float coordinates to long, there is data lost from the conversion. One
pixel difference in calculation changes the slope of the line. A path drawn
with non optimized and optimized path can have difference.
Fix Description: DrvStrokepath optimization in Rotation is
disabled.
|
|
BugID: 1625842
|
The
position arrows in TV/Advanced Settings are not in accordance with the
orientation chosen.
|
CUI2
|
Windows*
XP
|
Intel(R)
915GM Express Chipset
|
|
Resolution Description:
Root Cause: Scaling issue reported by customers.
Fix Description: The calling sequence changes for
an active detect was done before getting the current configuration.
|
|
BugID: 1649176
|
FBC stride
using bpp that corresponds to current mode, C3SR wmark to use 32bpp always
for calculation.
|
RESOURCE
MANAGER
|
Windows*
XP
|
Intel(R)
915GM Express Chipset, Intel(R) 945GM Express Chipset, Intel(R) 855GM
Chipset
|
|
Resolution Description:
Root Cause: Past directions from HW to use 32bpp for FBC stride
calculation for all bpp modes no longer valid. The larger than expected stride threshold
for 16bpp modes caused display corruption.
Using the bpp of the current mode fixes the issue. Then, for C3SR, 16bpp modes were seeing
flicker on Calistoga. There is a HW
arch issue for all platforms for C3SR.
Using 32bpp for watermark calculations is the workaround
Fix Description: Changed FBC stride calculation to
use the actual bpp of the current mode instead of a constant 32bpp. ALso changed C3SR watermark calculations
to always use 32bpp. This is a
workaround for a HW issue in lower bpp modes. We will use 32bpp for all
watermark calculations.
|
|
BugID: 1624597
|
CUI2_Some
help tips can not be reached by right clicking, but can be reached by the
question mark
|
CUI2
|
Windows*
XP
|
Intel(R)
945G Express Chipset
|
|
Resolution Description:
Root Cause: Notify property is missing few controls.
Fix Description: Added Notify property into the resource
file and also added controls which are not having strings IDs in
commonutils.cpp.
|
|
Issues Resolved in 4360
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1640304
|
Monitor information
is overwritten when we install display driver to the preferred timing of
monitor which is attached when install driver.
|
Miniport
|
Windows*
XP
|
Intel(R)
915G Express Chipset
|
|
Resolution Description:
Root Cause: Monitor information is overwritten when install display
driver to the preferred timing of monitor which is attached when install
driver.
Fix Description:
|
|
BugID: 1647479
|
Disable
FBC in interlaced modes as per specification.
|
Resource
Manager
|
Windows*
XP
|
Intel(R)
915GM Express Chipset
Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: FBC not supported in interlaced modes.
Fix Description: Certain modes in IGD are treated
as interlaced, such as 1776x1000 in 1080i59. FBC is not supported in such
modes according to B-spec. Forced to directly read Pipe Config Register to
determine interlaced vs progressive because ScanType flags in SharedInfo
which should provide this information are not properly maintained. Contact
has been made to the appropriate parties to resolve that issue. (GDI,
SoftBIOS, UAIM, RM)
|
|
BugID: 1634830
|
Fix
snowy screen issue for HDCP
|
DVD
|
Windows*
XP, Windows* MCE
|
Intel(R)
855GM Chipset
Intel(R) 865G Chipset
Intel(R) 915G Express Chipset
Intel(R)
915GM Express Chipset
Intel(R)
945G Express Chipset
Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: The HDCP snowy screen issue appears to be caused by the HDCP
transmitter and HDCP receiver getting out of sync.
Fix Description: To help keep the transmitter and
receiver in sync, the driver will not turn on encryption if it is already
on, and anytime the driver is going to change the HDCP encryption level, it
always sets encryption level to OFF first to reset everything. Then it
proceeds to set it to the desired level. This fix reduces the frequency of
the snowy screen dramatically, but it is still seen very sporadically and a
new defect was cloned from this one to track the issue.
|
|
BugID: 1644026
|
System
blooms or is missing display upon resume from S3
|
SoftBIOS
|
Win2000,
Windows* XP
|
Intel(R)
915GM Express Chipset
|
|
Resolution Description:
Root Cause: With VBT settings changed to have a different panel
type, the VBT is read from that point onwards. The update of the internally
cached data was getting corrupted because of the wrong interpretation of
the VBT data.
Fix Description: In lvds_CreateLfpDataFromEdid()
now get the lfp parameter registers from the panel type, that user has set
in VBT settings. Use that set of register values to update all the VBT
tables in a per DTD timings basis. That is all the DTD’s are added to the
cached VBT data and is used later on for setting those modes.
|
|
BugID: 1644026
|
Displays
turn to TWIN mode when FSDOS mode is selected on Digital display-2.
|
SoftBIOS
|
Windows*
XP
|
Intel(R)
945G Express Chipset,
Intel(R)
915G Express Chipset,
Intel(R)
915GM Express Chipset
|
|
Resolution Description:
Root Cause: After the driver is installed and Digital display-2 is
selected under FSDOS mode, the displays show TWIN configuration on both
digital display and digital display-2 with Dual DVI-7307-card connected.
Fix Description:
|
|
BugID: 1628299
|
LFP+TV
in clone resolution of 640x480 @ 8 bpp is applied.
|
CUICOM
|
Windows*
XP
|
Intel(R)
855G Express Chipset
|
|
Resolution Description:
Root Cause: Currently the min mode supported on TV is 640x480 x8 bpp,
however this is only valid for HDTV. For regular TVs, CUI needs to show
proper mode list. For regular TVs, they should be 8x6 in case of XP.
Fix Description: While getting the mode list, if
the device is HDTV, added code to modify the min xx and yy values. Added
ISHDTV function to find out whether the device is HDTV or not.
|
|
Issues Resolved in 4349
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1646688
|
DXVA
Deinterllacing Test - Fail
|
DVD
|
Windows*
XP
|
Intel(R)
945G Express Chipset
|
|
Resolution Description:
Root Cause: Extensive tests were conclusive: the NEAREST sampler has
a different behavior during DVD decode causing a 1 pixel shift.
Fix Description: Modified the field copy
operation (2nd phase of the DI algorithm) originally using NEAREST sampler.
The field copy is now performed using LINEAR sampler, the same used by the
Adaptive DI pixel shader. Using only LINEAR samplers, the deinterlacing
works in both environments: DVD playback (HW decode + DI) and WHQL (only
DI).
|
|
BugID: 1645795
|
Visual
Anomaly / 3DScreenSaver_multiconfig -
Miniport - 3D screensaver flickers in "Extended Desktop"
|
DD, GDI
|
Windows*
XP
|
Intel(R)
945G Express Chipset
|
|
Resolution Description:
Root Cause: When the user sets SPSD[DFP], PipeA and PlaneA are
allocated for DFP. When user extends onto CRT, pipe-plane-swap happens.
Because of the conditional errors in the IF and ELSE statements, CRT gets PlaneA
and PipeA, but DFP gets PlaneB and PlaneB.
Fix Description: Fix the conditional errors in
the IF and ELSE statements, so pOtherSharedInfo->dwDisplayConfiguration
updates properly for (ulOtherPlaneIndex == PLANE_B).
|
|
BugID: 1644340
|
FIxes for
Prefast warnings and for a bug in GfxBugcheckCallback() function
|
MINIPORT
|
Windows*
XP
|
Intel(r)
915GM
|
|
Resolution Description:
Root Cause: If the driver tries to read the contents of PRB or HWB
ring buffers when the ring buffers are not enabled, the memcpy() fails and
causes this issue.
Fix Description: Read the contents of ring
buffers only if the ring buffer is enabled.
|
|
BugID: 1642546
|
720x480
mode is seen when you configure TV+CRT in Clone
|
SOFTBIOS,
GDI, MINIPORT.
|
Windows*
XP
|
Intel(r)
915G
|
|
Resolution Description:
Root Cause: Current mode enumerations for primary and secondary to
CUI lists all modes from individual displays mode table to the maximum
Clone mode. CUI lists X, Y of the primary. If a mode is in the primary but
not in the secondary (like GTF modes or underscan modes), then in certain
combinations, these modes show up in reversed combinations and are not
shown in the CUI. OS will always list these modes as long as they are less
than the maximum clone mode reported in the Fake EDID.
Fix Description:: --In SoftBIOS: While building
clone mode table in function mode_GetCommonModes(), modes that are present
in individual displays mode table but are less than maximum common mode are
added. --In GDI: FilterModeList() function in cuicom_esc.c logic is changed
to include modes from clone mode table so that modes not present in single
display list get populated to CUI with a default RR of 60Hz. --In Miniport:
The Clone enumeration logic in function GfxGetChildDescriptor() in
vidmini.c did not take care of enumerating Clone only for Multi-pipe
chipsets. Changes have been made to make sure we do not enumerate Clone for
chipsets with single pipe like BDG and SDG.
|
|
BugID: 1651153
|
Functional
Failure / CUI2 - The LFP resolution is changed to 10x7 when switching the
display mode to CRT and then back to LFP via hot key after flashing the
modified BIOS.
|
CUI2
|
WIN2000,
Windows* XP
|
Intel(r)
915GM Chipset, Intel(r) 915G Chipset, Intel(R) 945G Express Chipset
|
|
Resolution Description:
Root Cause: Persistence options were not initialized in hkcmd.exe
for EDID Persistence.
Fix Description: In HotKeyImpl.cpp, function
ChangeActiveState() is modified to address this issue and switches the
display only when it is not in the target display.
|
|
Issues Resolved in 4345
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1647885
|
COPP
updates for integrated TV-Out.
|
DVD
|
MCE
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: The driver did not differentiate
between CGMS-A disabled and CGMS-A copy freely.
Fix Description: As part of enabling CGMS-A on
integrated TV-Out, this differentiation was required, so it was implemented
as part of that change.
|
|
BugID: 1621006, 1642750
|
DCN for prefast
related changes in Miniport for perf (resume time ) files.
|
MINIPORT
|
Windows*
XP
|
Intel(r)
915G Chipset
|
|
Resolution Description:
Root Cause: Connect CRT & LFP to the
system and then go to CRT only using CUI property pages. Then select 8x6 at
56Hz and apply on CRT. Soon select Extended Desktop from CUI property pages
with CRT as secondary device. Afterward, click on Display Settings tab and
select CRT, but CRT shows blank combos. "Defect Prospect" (as an
existing bug PCGSW 1642750) for NT-miniport.
Fix Description: Bug fix for 1441170 may have
solved the issue of "not turning off" the LFP but the issue of
"CRT being in secondary in ED mode when lid is closed" is not
fixed.
|
|
BugID: 1625330
|
Display
becomes unusual in GL_Excess.
|
OGL
|
Windows*
XP
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: Buffer_Info has set a wrong
pitch at __glSrvFlipToPrimary().
Fix Description: GDI has already implicitly
flipped so that "pSharedInfo->ulDispPitch" hold the correct
pitch. We save dwModeStateSeqTag at buffer creation time and check to see
if it is changed before flipping in __glSrvFlipToPrimary().
|
|
BugID: 1627663
|
Fixing
initialization watermark register programming.
|
RESOURCE
MANAGER
|
Windows*
XP
|
Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: Driver install process has
changed where a reboot is no longer required. This causes C3SR to be enabled without
being properly initialized. All
components of the driver are not yet loaded. Could be issues in other components as
well.
Fix Description: No longer enabling C3SR during
initialization. We will wait for an
event to enable it, such as a mode change.
|
|
BugID: 1621006
|
In
Extended desktop, CRT as secondary CUI shows blank combos and CRT is at 8x6
56Hz.
|
GDI
|
Windows*
XP
|
Intel(r)
915GM Chipset
|
|
Resolution Description:
Root Cause: CUI was sending an invalid mode
for LFP whenever querying for modes for ED-CRT secondary caused
COMTestConfig() failure. Because of that, GDI returned NULL mode table, and
we were seeing empty mode table for secondary device in CUI page.
Fix Description: Do not need to ask SoftBios to
validate modes via COMTestConfig() instead we are limiting COMTestConfig()
to call SoftBios with a flag to tests the devices.
|
|
BugID: 1628124
|
Functional
Failure / Miniport - cmd - The display switches to LFP+ CRT DDC mode and
the blank screen is displayed on LFP after closing and opening the lid
switch on CRT mode with opening cmd.[REG]
|
MINIPORT
|
Windows*
2000
|
Intel(r)
915GM Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: Currently when the lid closed,
the LVDS port is turned off if the display switch is blocked since FSDOS is
actived, or Overlay is playing. Else if the lid is open, this logic depends
on the Set-Mode call to turn on the LVDS port. But after fixed BUG ID
1562111, during the Set-Mode call, the LVDS port will not turned on if it
is off at that point. Because of this LVDS port is off, this issue happens.
Fix Description: When the lid closes, we turn off
the LVDS port. When the lid opens, we turn on the LVDS port. If the OS
sends the Power On call of LFP when the Lid switches is closed, then just
return without actually turning on the power of LFP.
|
|
BugID: 1634391
|
Bad
behavior with some CRTs.
|
SOFTBIOS
|
Windows*
XP
|
Intel(r)
915GM Chipset
|
|
Resolution Description:
Root Cause: This is very random in nature,
so the issue that customers saw was page fault error in
mode_ConsolidateAndSortModeTable(…). This was due to the improper boundary verification
processed in the function. Although the issue is sporadic which the CRT's
uid stated that it is a CRT with EDID, but there are no modes listed from
the EDID. Base on the improper boundary checking, the OS reports page fault
error.
Fix Description:
Initialized
all the local variables and added a null check to the pModeTable. Then add
a check: if(ulModeEnd != 0) { ……}
else { print critical message}.
|
|
BugID: 1645144
|
Functional
Failure / NewCUI - After disabling CUI Hot Keys with DTCM they cannot be
re-enabled.
|
CUI2
|
Windows*
XP
|
Intel(r)
915GM Chipset
|
|
Resolution Description:
Root Cause: State read from different places
in registry.
Fix Description: Hotkey Enable/Disable information
was stored at two different parts (HKLM and HKCU) of registry. While
enabling and disabling Hotkeys, previous state was read from different
places in registry. By changing everything to read from HKLM, the problem
is solved.
|
|
BugID: 1581099
|
No TAB
Key functionality on Display Devices page of CUI. (Generic Tab key
functionality defect.)
|
CUI2
|
Windows*
XP
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset,
Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: Main Dialog: Here the TAB key
navigation is functioning, but not visible because custom controls like skin
button, skin radio, and etc do not handle focus Draw and Erase
functionality. TAB Child Dialogs: Here (PageDevices, Settings …) the
controls inside each of the TABS doesn’t get any windows message to set or
Erase focus.
Fix Description: 1. All skinned controls are made
to handle set focus windows message and kill focus windows message and the
focus rectangle is drawn according to the controls focus state. 2. For
child dialogs inside the static box, keyboard messages are sent by the OS.
Using <TAB> keys user can now navigate through controls and after
settings focus can press <SPACE BAR> to select, press <UP>
<DOWN> keys to select different options in combo box control and can
press <LEFT> <RIGHT> keys to move slider controls, <ARROW
KEYS> for navigating radio buttons
|
|
BugID: 1623473, 1637514
|
Power
Setting of CUI disappear after CUI closed and opened.
|
CUI2
|
Windows*
XP
|
Intel(r)
915GM Chipset
|
|
Resolution Description:
Root Cause: Restore Default option was not updating
the position buttons.
Fix Description: Called the appropriate functions
to restore Default.
|
|
BugID: 1635852, 1631082
|
In Clone
mode, you can apply "not supported modes" for 720p59 standard on
HDTV.
|
AIM
|
Windows*
XP
|
Intel(r)
915G Chipset
|
|
Resolution Description:
Root Cause: There is an inconsistency in the
modes that we can pass with fake EDID signal and reach getHDTVCaps.
Fix Description: Removed the inconsistency codes.
|
|
Issues Resolved in 4338
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1645180
|
The
flicker screen is displayed while playing the LiveTV.
|
DVD
|
Windows*
XP, MCE
|
Intel(r)
945G Chipset
|
|
Resolution Description:
Root Cause: When A-DI is enabled, using
vertical line stride with a destination surface in Tiled memory produces
incorrect addressing, which causes DVD playback and live TV corruption.
Fix Description: Always generate an intermediate
DI output in Linear memory
|
|
BugID: 1574941
|
Functional
Failaure / OGL - Serious Sam The Second Encounter - The game resolution
changes from 1280x1024 to 640x480 and still on full screen mode after
pressing Alt + Enter keys.
|
GDI
|
Windows*
XP, Windows* 2000, Windows* XP-64
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: If the GCAPS2_ICD_MULTIMON
capability is not specified in GDI, and the game region involved intersects
more than one device, no driver is called in Extended desktop and the game
will stay in the full screen resolution.
Fix Description: Need to set GCAPS2_ICD_MULTIMON
in hwenable.c
|
|
BugID: 1640863
|
Novell
GW causes a hang or blue screen on GX280/270
|
GDI
|
Windows*
XP
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: XY_TEXT_IMMEDIATE_BLT outside
clipping rectangle cause ring buffer hang.
Fix Description: Filter out Immediate_text_blt
commands that do not fall within the clipping rectangle. Keep destination
rectangle or glyph rectangle to be within 16bit signed number range (short) number range.
|
|
BugID: 1643053
|
New COPP
cert / Integrated TV updates
|
DVD
|
Windows*
XP
|
Intel(r)
915GM Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: Disable CGMS-A for integrated TV out (code
change includes fix for correct programming for the Int TV out registers
for sending CGMS-A data over line 20 and line 21, but Line20/21 calls are
not available due to CGMS-A disabling).
Include the new COPP ce
Fix Description: Adjust for the documentation error and add the start and
end packets. Additionally, stop
calling AIM from the COPP timer if there is no protection active. This improves performance and works
around a possible hang in full screen DOS while doing line 21 CGMS-A on
integrated TV out. In order to do
that the driver also has to start differentiating between CGMS-A disabled
and CGMS-A copy freely.
|
|
BugID: 1628124
|
Functional
Failure / Miniport - cmd - The display switches to LFP+ CRT DDC mode and the
blank screen is displayed on LFP after closing and opening the lid switch
on CRT mode with opening cmd.
|
MINIPORT
|
Windows*
2000
|
Intel(r)
915GM Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: On Lid events, the LVDS port is turned off if the
display switch is blocked because of FSDOS is active or Overlay is playing.
As a result, the driver will not turn on LVDS port during modeset call for
Lid open due to the above logic
Fix Description: On Lid close turn off LVDS port irrespective of any
condition and on Lid open turn ON LVDS port irrespective of any condition.
If the OS sends the Power On call for LFP when the Lid switches is closed
then just return without actually turning on the power for LFP
|
|
BugID: 1641217
|
Display
goes blank when you change the configuration from Extended to TWIN
|
SoftBios
|
Windows*
XP
|
Intel(r)
915GM Chipset
|
|
Resolution Description:
Root Cause: On a pipe swap, when enable LVDS
on PipeB when the driver enables the panel fiting, the LVDS screenwill
blank because VGA centering is disabled.
Fix Description:
|
|
BugID: 1629016
|
W HCT 12.1 ACPI Stress Extended - sporadic blank screen hard
hang
|
SoftBios
|
Windows*
XP
|
Intel(r)
915GM Chipset
|
|
Resolution Description:
Root Cause: Even though we were disabling the
pipe before entering into standby, the Save function has the Enabled pipe
value. On the resume, the Restore did not comprehend the enable sequence of
pipes and PLLs.
Fix Description: Don't
save/restore the pipe value while going to standby anyhow set mode call
will enable the PLL and Pipe in a right sequence
|
|
BugID: 1639150
|
Apply
button does not get highlighted when you select Dot Crawl in TV-setting
|
CUI2
|
Windows*
XP
|
Intel(r)
915G Chipset
|
|
Resolution Description:
Root Cause: Enable apply button is not
getting called when dotcrawl option is selected.
Fix Description: Call apply button when dotcrawl
option is selected.
|
|
BugID: 1581099
|
No TAB Key
functionality on display devices page of CUI (Generic Tab key functionality
defect)
|
CUI2
|
Windows*
XP
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: The controls inside each of the
TABS do not get any windows message to set or erase focus.
Fix Description: All skinned controls will be
handling set/kill focus windows messages
and the focus rectangle is drawn according to the controls focus
|
|
BugID: 1625339, 1634698
|
The
1024x768 mode display 640x480mode when you rotate the display to 90 or 270
degree on TV
|
CUI2
|
Windows*
XP
|
Intel(r)
945G Chipset
|
|
Resolution Description:
Root Cause: Rotation modes are not handled in
DDC re-write.
Fix Description: Added support for rotation in
CUI for HDTV and Modes.
|
|
Issues Resolved in 4336
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1640670
|
Display
goes blank on DFP & LFP when you apply Extended desktop
|
RM
|
Windows*
XP, Windows* 2000
|
Intel(r) 865G, Intel(r) 915GM Chipset
|
|
Resolution Description:
Root Cause: When RM call to provide
plane/pipe to LFP and DFP was already enabled and using PipeB/PlaneB, the
RM assigned PipeA/PlaneA to LFP. Since LFP can only be enabled on PipeB,
display went blank.
Fix Description: Include in the plane/pipe
selection for each display type algorithm the restriction check first, then
the capabilities before returning plane/pipe to be assigned to device.
|
|
BugID: 1644007, 1643863
|
On a
system with TV chip on SDVO_C, we can not see TV Icon on Intel CUI Tab.
|
AIM
|
Windows*
XP, Windows* 2000
|
Intel(r)
915G Chipset, Intel(r) 945G Chipset
|
|
Resolution Description:
Root Cause: The ADD card was programmed as
multifunction device and due to incorrect mapping the TV detection fails.
Fix Description: During device type detection,
enable the encoder as single function, not multifunction
|
|
BugID: 1611936
|
Live HD
TV on MCE in VLD mode not able to be
displayed correctly and system hangs.
|
DxVA
|
Windows*
XP, Windows* 2000
|
Intel(r)
945GM Chipset
|
|
Resolution Description:
Root Cause: Between frames, HW hangs.
Fix Description: Add workaround for VLD buffer
slice per HW DE recommendation. Note: Not yet
verified on Silicon, will be verified on Calistoga A2 HW ECO 1629920.
|
|
BugID: 1625839, 1627629, 1642161
|
Display
comes in single display whereas it should have come in extended desktop
|
miniPort
|
Windows*
XP, Windows* 2000
|
Intel(r) 865G, Intel(r) 915GM Chipset, Intel(r) 945GM
Chipset
|
|
Resolution Description:
Root Cause: The driver does not get the call for enumeration on Lid
event before getting call for display change. (OS does not consistently
enumerate on Lid event). The driver depends on enumeration to update the
internal data structures which contain the display UIDs, and when no
enumeration the registry for LFP UID will be 0.
Fix Description:
Always include LFP UID flag done while doing the passive detection
for next configuration, so we always get the LFP UID irrespective of Lid
status.
|
|
BugID: 1619159
|
corruption
is seen under Taskbar when you apply 1080i standard.
|
softBIOS
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM Chipset
|
|
Resolution Description:
Root Cause: in HD when vertical scalar is by-passed
and only X size is set, extra video data is displayed before the vertical
retrace causing corruption.
Fix Description: Chipset requires programming of
both X and Y Size on the display plane, which is now done in the Int TV get
mode values function; these values will
be used for programming the sourceregister of the corresponding
pipe.
|
|
BugID: 1642775
|
Failure
of DDC/Ci communication
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: IGFXEXT was using an invalid
pointer (null) to access the service module and functions of MCCS controls
returned failure.
Fix Description: Include additional AddRef() to
the IMCCS interface. From the caller, one reference will be released, and
the first reference obtained during the first invocation will be released
only when the MCCS handle is Closed.
|
|
BugID: 1636548
|
OGL:
conformance multitex.c test resulted in BSOD
|
OGL
|
Windows*
XP, Windows* 2000
|
Intel(r)
915G Chipset, Intel(r) 915GM Chipset, Intel(R) 945G Express Chipset,
Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: OGL device code was generating a
stipple shader instead of the non-stipple version and when it was executed,
it caused BSOD.
Fix Description: Use the local copy of the shader,
instead of the one pointed to by the context. The context no longer
contains this variable that is compute locally in the function.
|
|
Issues Resolved in 4330
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1612280
|
DVD Test
Annex 2.0 Menu on MCE with VLD mode enabled causes system hangs.
|
DxVA
|
Windows*
XP, MCE
|
Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: The batch buffer was over-written before HW finishes it.
Fix Description: We need to wait for the previous commands done by
HW before re-use a batch buffer.
|
|
BugID: 1635307
|
“Intel'' folder in root directory after
installation of graphics driver
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: Two executables (hkcmd.exe and igfxtray.exe) are sharing
the same language resource file igfxres.dll from System Directory for Admin
Mode. At boot time, the second one to initialize will
incorrectly create the resource file in the User Space under \Intel\ExtremeGraphics.
Fix Description: If the file copy in second call fails, load
igfxres.dll from System Directory instead of creating User directory. Secondary issue is fixed, for hot keys
not working in Limited User.
|
|
BugID: 1635574
|
Cannot
apply Clone for Monitor2+monitor configuration from DTCM
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset,
Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: Software\\intel\\display\\igfxcui\\schemes is limited to
4, instead to 8 for the maximum
number of devices.
Fix Description: updated the # define value from 4 to 8.
|
|
Issues Resolved in 4326
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1634826
|
COPP
driver update
|
AIM,
miniPort
|
Windows*
XP
|
Intel(r)
915G Chipset, Intel(r) 915GM Chipset, Intel(R) 945G Express Chipset
|
|
Resolution Description:
Root Cause: The tamper resistance module assigns an ID to each COPP
session. The first session is being
assigned ID zero, which is the same as the initialization value. This causes the second COPP session to be
aborted before getting the ID and corrupting the original valid COPP
session.
Fix Description: Initialize the tamper resistance module ID for
each COPP session to something other than zero (-1).
|
|
BugID: 1631670, 1634692
|
The BMP
setting is ignored about spread spectrum.
|
AIM
|
Windows*
XP, Windows* 2000
|
Intel(r)
915G Chipset, Intel(R) 945G Express Chipset
|
|
Resolution Description:
Root Cause: Driver was not reading the BMP setting for SSC for the SDVO
LVDS and was programming the default value, which would cause EMI issues
with LVDS.
Fix Description: Changed the driver code to read VBT settings and
programm it accordingly
|
|
BugID: 1628412
|
Corruption\DVD
- The corruption is displayed while playing the Planetary Traveler DVD
movie
|
DxVA
|
Windows*
XP, MCE
|
Intel(R)
945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: In VLD decode mode, the pass-in TopFieldFirst info is
incorrect.
Fix Description: We need to use our own algorithm to tell if Top
Field is First
|
|
BugID: 1637989
|
ACP signal
disappear when opening second DVD player and closing first DVD player
|
AIM
|
Windows*
XP, Windows* 2000, MCE
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset,
Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: Enabled support for 64 Multiple Session handling, which fixed
the issue of losing MV signals when one of the two active sessions was
closed. This includes miniPort change to report NO_ERROR status in
VIDEOPARAMETERS ioctrl, (revert back to same as build 4316). The error code
was ERROR_INVALID_FUNCTION, and DCT
test.
Fix Description: The fix is ported to all DVO and SDVO TV out
modules.
|
|
BugID: 1629992
|
S/Composite
Video shift to right side after S3 resume issue
|
AIM
|
Windows*
XP, Windows* 2000
|
Intel(r)
915G Chipset, Intel(R) 945G Express Chipset
|
|
Resolution Description:
Root Cause: The TV position/size is not persisted over reboot or
suspend/resume, because the checking of the current value is incorrect
andthe max values will be saved instead.
Fix Description: The current position value should be checked
against max value instead max value is checked against pAim max value which
is always zero for the first time while resuming.
|
|
BugID: 1628162
|
S/Composite
Video shift to right side after S3 resume issue
|
AIM,
softBIOS, CUI
|
Windows*
XP, Windows* 2000
|
Intel(r)
915G Chipset, Intel(R) 945G Express Chipset
|
|
Resolution Description:
Root Cause: Displaying same encoder twice is eliminated based on
vendor Id and device ID, but for dual dvi card the vendor Id and displayed are
same.
Fix Description: Eliminating duplicates based on I2C address
instead of vendor Id and Device ID.
|
|
BugID: 1635564
|
S/Composite
Video shift to right side after S3 resume issue
|
AIM,
softBIOS
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: The TV position persistence is
maintained using registry keys; when CUI is not used for changing
positions, there are no registry keys and registry key read fails.
Fix Description: If registry read fails, assign
default values to position.
|
|
BugID: 1557688
|
Change
the screen resolution and color quality, then run screen saver 3D Pipes.
Garbage is displayed.
|
D3D
|
Windows*
XP, Windows* 2000
|
Intel(r)
915G Chipset, Intel(r) 915GM Chipset, Intel(R) 945G Express Chipset, Intel(R)
945GM Express Chipset
|
|
Resolution Description:
Root Cause: Driver destroyed the DX context
that was still being used by the 3D pipes scenario.
Fix Description:
In DestroyDirectDrawContext(), only allow a
context release if the current process is the owner, or if the context has
no owner.
|
|
BugID: 1621006
|
In
Extended desktop, CRT as secondary CUI shows blank combos and CRT is at 8x6
56Hz
|
GDI
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: CUI query for available modes in
Extended Desktop with CRT secondary was failing and returning null mode
table.
Fix Description: Do not call softbios to validate
modes via COMTestConfig(), since
softbios will not do any bandwidth considerations and instead call softbios
with a flag to tests for devices only.
|
|
BugID: 1638580
|
Abnormal
display or driver fails when playing DVD in Clone Panning mode
|
GDI
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: In Clone, the X,Y overlay panning
window variables for secondary are initialized with primary X,Y values
because of incorrect pipe check.
Fix Description: Check for correct pipe when
initializing secondary pan window in Clone.
|
|
BugID: 1624432
|
Display
on LFP is blank on resuming from Standby ( lidswitch), in FSDOS mode.
|
softBIOS
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: Power Management Cleanup related
to Only SoftBios. Implemented a new interface between SoftBios & AIM
devices which will set Display state of a device independently. Implemented
Display States for IntegratedLVDS & Integrated TV Out.
|
|
BugID: 1639394
|
ACP
signal disappear when opening second DVD player and closing first DVD
player
|
softBIOS
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: When multiple DVDs are playing on
integrated TV out, if we turn off the first DVD app while 2nd DVD app is playing
MV ACP signals were cleared and still 2nd DVD app was playing CP DVD
without MV enabled.
Fix Description: Generate unique key for each of
the 64 DVD apps when is requested for activation. We will turn off the MV
signals only when the last of app is requested for deactivate command to
close the e MV session.
|
|
BugID: 1638708
|
Unable
to apply 848x480 HDTV mode
|
softBIOS
|
Windows*
XP, Windows* 2000
|
Intel(R)
945GM Express Chipset
|
|
Resolution Description:
Root Cause:The driver reports that for 848 x
480 it supports both 720 and 1080
video standards, instead of supporting only 480 Video Standard.
Fix Description: Fix the mode set logic for this
mode to allow switching to 480 Video standard.
|
|
BugID: 1634762, 1623466
|
Integrated
- The Vertical Size and Horizontal Size values of TV settings restore to default
settings after switching TV to LFP and then switching back to TV.
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: When ever a TV parameter is changed, even though it is
not full screen option, we set the full screen parameter which is incorrect
Fix Description: Check if the status changed for
full screen option, if so then set the TV parameter
|
|
BugID: 1636352
|
When
closing CUI, an desktop icon flickers.
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: CUI controls are initialized when
closing CUI.
Fix Description: Initialize the controls only if
the dialog is not exiting.
|
|
BugID: 1616518
|
W2K/XP:
The return value of SMI 090Dh is not valid.
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
855GME Chipset, Intel(r) 915GM Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: On every display change, CUI
should call system bios method Notify. This method currently is not called by
Persistence module.
Fix Description: Add system notification method,
functions and GUID and call the function for all display changes. When a
display configuration change is detected, call the NotifySystemBios method
of the service layer and update its previous configuration with the new
configuration.
|
|
BugID: 1639737
|
WM_DISPLAYCHANGE
message is sent with WM_POWERBROADCAST.
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
855GME Chipset, Intel(r) 915GM Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause:Whenever a power notification
coming to CUI, it is initiating a Display change message, which should be avoided
for all the cases except for resume & Suspend operations. This results
in unneceesary display change message for all platform apps and OS.
Fix Description:
Whenever a power broadcast message comes to the system, before
sending a BroadCastDisplayChange message check for resume/suspend
operations before broadcasting.
|
|
Issues Resolved in 4324
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1627901
|
Case of
mouse cursor display and the pointing coordinate are difference on Rotated
Extended Desktop
|
Rotation
|
Windows*
XP
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: When pointer is moved to media player window in the
secondary DrvSetPointerShape call for the secondary, the x,y coordinates
from primary and wrong hot spot (xhot, yHot) values gets stored in the
primary pdev. When pointer is moved to primary, cursor's hot spot is
incorrect.
Fix Description: Since mis-calculation is due to driver not
calculating hot-spot in DrvSetPointerShape for negative values, add code to
do so for this case also (x <0, y<0 ).
|
|
BugID: 1615920
|
A monitor
off function does not work at 270 degree
|
softBIOS,
AIM
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: UIAM driver does not have state information of the
current power state of the display when screen saver is followed by Monitor
Off in Rotated mode. As opposed to Landscape mode, DRV Assert and
EnablePDev do not get called in this sequence.
Fix Description: A new
variable is defined to store the power state of the Display for SDVO. SDVO
devices use setPowerstate with MD3 power state. This ensures that encoder
power state is not changed when there is call to change display power
state.
|
|
BugID: 1595205
|
TV
display distortion (jump) while set res. 1280by600 in Dual Mode.
|
softBIOS
|
Windows*
XP, Windows* 2000
|
Intel(r)
915G Chipset, Intel(R) 945GM Express
Chipset
|
|
Resolution Description:
Root Cause: IntegratedTV
Scaler not capabable to scale the centered timings with a high blanking,
and we use to observe flicker on TV. This issue is seen because encoder
scaler is not working on non GTF timings. Fix Description:
Added static GTF timings for non-standard modes 1024x600 and 1024x720, in
addition to the generic centering logic which is not applicable for the
Integrated TV encoder. Use these new timings to program the GMCH register in
util_set_controller_timing.
|
|
BugID: 1634578, 1635813
|
CUI2 -
Calistoga UI displaying wrong bitmaps.
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(R)
945GM Express Chipset
|
|
Resolution Description:
Root Cause: Calistoga chipset ID check has not been done in some places
of the UI, and hence by default it goes to the default chipset bitmaps.
Fix Description: Add check for Calistoga chipset ID for secondary
dialog. For defect 1635813, Strings are changed in device settings page to
show "Color Quality" and
"Screen Resolution" instead of "Color quality" and
"Screen resolution"
|
|
BugID: 1628412
|
The
corruption is displayed while playing the Planetary Traveler DVD movie with
Media Center application.
|
DxVA
|
Windows*
XP
|
Intel(r)
915G Chipset, Intel(R) 945GM Express
Chipset
|
|
Resolution Description:
Root Cause: In VLD decode mode, the pass-in TopFieldFirst info is
incorrect.
Fix Description: We need to use our own algorithm to tell if Top Field
is First
|
|
BugID: 1628195, 1636592
|
Integrated
- The HDTV video standard is changed after restarting the system with HDTV.
|
softBIOS
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM Chipset, Intel(R) 945G Express Chipset
|
|
Resolution Description:
Root Cause: Driver is setting the SDTV bits when Playback app (e.g
WinDVD) starts request for the current standard. On resume from standby/hibernation/restart,
the TV standard is incorectly restored from registry as SDTV.
Fix Description: While initializing the TV, initialize the
connector type based on the connector type mask, and set the correct TV
standard. In addition, include fix
for launching DOS session in 480p59 results in TV std as 480i59.
|
|
BugID: 1616907
|
First
boot after gfx installation, Windows 2000 Display Properties will report 1280x1024
as the maximum resolution despite having 16x12 capable devices hooked up to
system
|
softBIOS
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: Driver was adding modes (even GTF) in ascending order
one by one in Multi-Monitor Clone Edid. If the number of modes is more than
12 after Est Timings have been filled, then higher modes do not get
included.
Fix Description: Optimized Edid to definitely add largest common
mode and mode with Max RR at a particular (X,Y).
|
|
BugID: 1619195
|
Blank LCD
display after LID close/open operation.
|
miniPort
|
Windows*
XP, Tablet PC
|
Intel(r)
915GM Chipset, Intel(r) 855GME Chipset
|
|
Resolution Description:
Root Cause: Driver workaround for Tablet PC issue.
Fix Description: Remove the Logic in driver code which Queues a DPC
for enumeration on Lid status change; always enumerate the LFP as attached
irrespective of the Lid status. This causes the driver to always enumerate
the LFP and hence the OS does not remove the LFP device from device
manager.
|
|
BugID: 1635151
|
CRT does
not invoke preferred mode when switched to it for the first time
|
miniPort
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM Chipset, Intel(r) 855GME Chipset
|
|
Resolution Description:
Root Cause: Driver does not have the support to drive the CRT
preferred mode, when it switched via ACPI hot keys the very first time.
Fix Description: On ACPI display switch, identify if the next
display mode is SPSD, if yes then identify if the display is being driven
for the first time by checking if OS has already created a key for this or
not. If not then identify the preferred mode and write to registry after
creating the key.
|
|
BugID: 1638580
|
Abnormal
display or driver fails when playing DVD in Clone Panning mode
|
GDI
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause:The overlay pan window variable is not properly
initialized for secondary display in Clone configuration.
Fix Description: Initialize
the variable for secondary also.
|
|
BugID: 1633757
|
First boot
after gfx installation, Windows 2000 Display Properties will report
1280x1024 as the maximum resolution despite having 16x12 capable devices
hooked up to system
|
GDI
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause Hang / D3D - Bend - The system hangs
and the blue screen is displayed after switching the DOS command and Bend application in
the full screen mode via Alt+Tab.
Fix Description: Request OS GDI to create bitmap if driver fails to
create one in DrvCreateDeviceBitmap().
|
|
BugID: 1623800
|
One
wallpaper color turns to red after switching users
|
GDI
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset,
Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: On a Fast User switch, the wall
paper is about to be rendered, OS calls DrvCopyBits call with source Palette
in BGR format. But the destination palette will be set RGBformat in a call
to enable PDEV during Fast User Switching. So source format which is BGR is
different from destination format which is RGB. When this is rendered as it
is the “blue” wallpaper turns “Red”.
Fix Description: When source is BGR format and
Destination is RGB565, punt blitter call to the OS.
|
|
BugID: 1578185
|
Hide
modes checked issue
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: HideModesChecked() method was not working properly.
Fix Description: Fixed ishidemodeschecked method in CUICOM.
|
|
BugID: 1618863
|
Bug fixes for Tibet ID 1618863
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: On a resume from S3 event, the persistence module
receives a hot-plug event upon which is searching the best suited
resolution setting. This algorithm did not comprehend rotated modes.
Fix Description: Define a specific function to get the
"best" supported display mode and call this function when the
modes needs to be validated on the display hot plug event. The new function
also takes BPP and Rotation into consideration.
|
|
Issues Resolved in 4321
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1623875
|
System
loses display and can crash when using OS controls to control monitors
|
GDI
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause:The SetRegistryDisplayDevice function was always
selecting the Display1_UID1 to update the secondary displays UID causing
the displays to be swapped with every mode set call on secondary controller
in extended mode.
Fix Description: Mode set
call for device 1 can come only through OS display properties pages while
setting the extended mode. Mode set call on Device 1 to use Display2_UID1
to read/write the display id, while mode set call for device 0 use
Display1_UID1, with the exception for TWIN/DDC mode.
|
|
BugID: 1593171
|
Demo
crash
|
OGL
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: Persistent buffer was being allocated in one context for
fragment programs and freed in a second context. In the second context, it was
attempting to free the already freed buffer, causing the blue screen.
Fix Description: Manage persistent buffers on a per-process basis
instead of on a per-context basis, similar to textures. Persistent buffers can be shared across
contexts, similar to textures.
|
|
BugID: 1632073
|
CUI2:
Power Settings button doesn’t prompt the Power Settings window after the
first time
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: In Service layer, when releasing the mutex,
m_hPowerAPILock is not set to NULL, and based on this variable, on next CUI
launch, Power Setting will not be opened.
When ReleaseMutex is successful initialized, reset m_hPowerAPILock
to NULL.
|
|
BugID: 1624597
|
CUI2_Some
help tips can not be reached by right clicking, but can be reached by the
question mark
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: In Resource file, all the static controls when right
click is not providing what’s this? Menu. But click on these controls with
arrow will work. In Resource file, the Notify property is not selected,
hence the right click message is not coming to the control.
Fix Description: Select the
notify property in resource file. Also, changed help tips per Localization
request.
|
|
Issues Resolved in 4318
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID:
1628837
|
CUI PAGE is not refreshed if TV unplugged in DDC
|
softBIOS
|
Windows* XP, Windows* 2000
|
Intel(r) 915GM Chipset, Intel(R) 945GM Express
Chipset
|
|
Resolution Description:
Root Cause: When TV is unplugged and then plugged back, the driver returns no devide on CUI query
of devices, and because of this CUI is not showing TV icon.
Fix Description: When the TV encoder is enabled, always do a TV
detect, do not rely on DAC reporting no device.
|
|
BugID:
1629826
|
ACP signal disappear when opening second DVD player
and closing (doesn't matter which DVD player we use).
|
AIM, miniPort
|
Windows* XP, Windows* XP-64, MCE
|
Intel(r) 865G, Intel(r) 915GM Chipset, Intel(r)
915G Chipset, Intel(r) 855GME Chipset, Intel(R) 945G Express Chipset,
Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: When opening second DVD player, driver did not return
proper error code for IOCTL handling VideoParameters. This is causing the second
session to run MV protected content using the same Key. Fix Description:
Do not enable second MV protected session, by returning unsuccessful code
and not send MV code to SDVO.
|
|
BugID:
1612637
|
Movie-screen is monochrome.
|
DD
|
Windows*
2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: The Indio codec neglects the pitch when
writing chroma data to I420/IYUV surfaces.
Fix Description: Extend existing IYUV/I420
workaround for Indeo codec to non-overlay surfaces; for the workaround, we
have created a surface that is twice as big, half for the driver, half used
for the application.
|
|
BugID:
1619828
|
Workaround/Temp patch for Presentation director
Hang issue
|
CUI2
|
Windows* XP, Windows* 2000
|
Intel(r) 915GM Chipset, Intel(R) 945GM Express
Chipset
|
|
Resolution Description:
Root Cause: The CoCreateInstance hangs in OEMDSPIF
dll, due to App power manager OCX. Fix Description: Modified OEMDSPIF dll to
directly call CUI COM layer instead of CUI service to avoid the OCX control
causing hang.
|
|
Issues Resolved in 4315
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1598883
|
Right
arrow in the TV-setting is not getting grayed out and also not able move
the display screen to right
|
TVout
|
Windows*
XP
|
Intel(r)
915GM
|
|
Resolution Description:
Root Cause:
When we change the Position By pressing the Right Arrow we get a new
position X and Y with Position Flag Set.
Fix Description:
We calculate the Current Position in IntTV_GetSTDTVParameters
function based on the following formula:
PositionX =
((pAimObj->intTVMode).dwWindowsPosReg >> 16) & 0x1FFF;
PositionY =
(pAimObj->intTVMode).dwWindowsPosReg & 0x0FFF;
With
this change there is sync between IntTV_GetSTDTVParameters &
IntTV_GetVideoParameters function. Both calculate the Current Position
using the same above formula.
|
|
BugID: 1601574
|
Display
Blanks and No-Signal is displayed on
screen when you apply twin for Monitor and LFP
|
Softbios
|
Windows*
XP
|
Intel(r)
915GM
|
|
Resolution Description:
Root Cause: The twin mode behavior should take
into consideration any the capabilities of CRT with respect to the AIM
device. LFP is a special case where the EDID is faked but the final timing
programmed is of the DTD timings.
Fix Description: Reverted back all the changes in
lvds.c.
In
mode_GetModeSupportPipe() we check for the AIM device is it LFP. If LFP for
any mode requested we take the max resolution of the LFP(with current
implementation that is the DTD timings) and find the corresponding mode
entry and check whether that mode is supported by CRT
|
|
BugID:
1619751
|
Hardware
accelerated stretching not working properly (neither RGB BitBLT with
stretching, nor YUV BitBLT with stretching).
|
DD
|
Windows*
XP
|
Intel(r) 865G, Intel(r) 855GME Chipset
|
|
Resolution Description:
Root Cause: App uses mmx instructions without
clearing mmx state. This causes
floating point errors.
Fix Description: Clear mmx state in stretch blit before
we execute floating point operations.
|
|
BugID: 1617046
|
DualDisplayClone
Preferred Mode Issue]Can't change resolution to WXGA(1280x768)
|
GDI
|
Windows*
XP
|
Intel(r)
855GME Chipset
|
|
Resolution Description:
Root Cause: In
the new DDC design, modes are enumerated based on the common x, y, bpp in
both participating display’s mode list when the hide mode is checked. So, even
preferred mode gets pruned if it does not have a same x, y, bpp entry in
the other display’s mode table. If the EDID reports a preferred
timing that preferred timing fits into any available timing on the other
monitor without panning, it will included in the “Multi Monitor EDID” for
DDC.
Fix Description:
|
|
BugID: 1628448
|
Cannot
install Rotation utility
|
CUISDK
|
Windows*
XP
|
Intel(r)
915GM
|
|
Resolution Description:
Root Cause:
Rotation utility was checking for RotationFlag from
GetConfiguration, this flag was returning false due to the bug in the CUI
SDK code.
Rotation interface is retrieved using queryinterface. This does not work
with CUIService since each interface is implemented by separate objects
with clsids.
Fix Description:
|
|
Issues Resolved in 4306
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1598883
|
Right
arrow in the TV-setting is not getting grayed out and also not able move
the display screen to right
|
softBIOS
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: The Set and Get TV parameters
functions were not in sync, and as a result created mis-match for ( X, Y)
in CUI and driver.
Fix Description: The Get function is now using the
same algorithm to calculate X,Y as the Set function.
|
|
BugID: 1601316
|
CFLUSH
fix + Smithfield FourCC BSOD fix + Allocation Clean-Up
|
GMM, D3D
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: Cache coherency with dual core causing
0's starting at cacheline-aligned addresses part of the way through the
batch. Corrupted instructions have included LSI1 and variable length
indirect tri-lists.
Fix Description: Dual core platform require Cache
line flush via CLFUSH instead of normal WBINVD since using WBIND will not
flush cache of both CPUs. This change is using CLFLUSH for all platforms,
if CPU will support it (in PC14.12 the change was for Lakeport only).
|
|
BugID: 1618895
|
CFLUSH
fix + Smithfield FourCC BSOD fix + Allocation Clean-Up
|
GMM,
D3D, MiniPort
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: When a client requests an
allocation that can be Linear OR X-Tiled OR Y-Tiled, we can hit a case
where the pitch does not get re-aligned to the correct tile size after the
tiling is switched. This probably
never showes up before since it is only since Lakeport that the tiling
widths are different.
Fix Description: The Allocation Clean-Up code
takes this into account in the __GmmAllocVaryPitch() and __GmmAllocSwapXY()
functions.
|
|
BugID: 1599074
|
Video
corruption in game Hitman: Contracts
|
D3D
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM Chipset, Intel(r) 915G Chipset, Intel(R) 945G Express Chipset,
Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: The Z-Bias implementation was
causing visual anomalies.
Fix Description: Instead of modifying the
z-component of each vertex in the vertex buffer, implement the solution
based off of the hardware solution for the SLOPE SCALE DEPTH BIAS and DEPTH
BIAS found in Microsoft's DX9 API.
|
|
BugID: 1619751
|
Hardware
accelerated stretching not working properly (neither RGB BitBLT with
stretching, nor YUV BitBLT with stretching).
|
D3D, DD
|
Windows*
XP, Windows* 2000
|
Intel(r) 855GME Chipset, Intel(r) 865G
|
|
Resolution Description:
Root Cause: App uses mmx instructions
without clearing mmx state. This
causes floating point errors.
Fix Description: Clear mmx state in stretch blit
before we execute floating point operations.
|
|
BugID: 1612989
|
MIT/Aspen-U/XP/Rev1:
Brightness of panel will become dark after do S/R (Nx51v_E1.100)
|
softBIOS
|
Windows*
XP, Windows* 2000
|
Intel(r)
855GME Chipset, Intel(r) 915GM Chipset
|
|
Resolution Description:
Root Cause:For edid panels we used to get
BLC/BIA data from panel #16 even though OEM selected a different panel in
VBT.
Fix Description: use the panel selected by OEM in
VBT
|
|
BugID: 1622405
|
VCD
display always shows on CRT in display both mode by VGA driver 6.14.10.4277
on xpsp2 os.
|
miniPort,
RM
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM Chipset
|
|
Resolution Description:
Root Cause: The recursive function used to
reserve bandwidth for Video Overlay is failing when both Overlay and Sprite
are enabled. The first call is
destroying the Video parameter used by the second call.
Fix Description: Save/Restore the input video
parameter in the function that calculates the bandwidth needed for overlay.
|
|
BugID: 1603157, 1616930
|
Functional
Failure / CUI2 - BMP, fwhflash.exe, work_slider - The message telling the power
api is locked does not pop after opening the Power Settings page and
launching work_slider application.
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: setpowerlock api was not working
and second is incorrect coding to disable buttons when Apply Button Always
active in Advanced dialog issue.
Fix Description: Correction for Power API mutex
lock issue and for disable buttons.
|
|
BugID: 1625232
|
Bug fix
for 1625232
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM Chipset
|
|
Resolution Description:
Root Cause: However CUI should not populate
win_vga standard to the UI in the combos.
Fix Description: When driver returns the
available standards, CUI checks whether win_vga is present in the available
standards, if it is present it will remove win_vga from the standard list.
|
|
BugID: 1620716
|
Zoom app
starts malfunctioning after 7 minutes
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: The Zoom App does not destroy
the bitmap handles created when DrawCursor function is called. Also, the
ZoomRestore function does not release the DeviceContext it created, causing
GDI memory leak.
Fix Description: Destroyed the handles returned
by the call ‘GetIconInfo’. Also, released the DC created by GetDC.
|
|
BugID: 1626044
|
The word
contrast does not belong in the UI for TV settings -> color
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause:The string IDS_HUE was set as
‘Contrast’ in resources.
Fix Description: Added a new string IDS_TV_HUE
(2918) and set to ‘Hue’
which is used in TV page while the old is used in Overlay page. We need the
string IDS_HUE for overlay page.
|
|
BugID: 1623738
|
CUI2_HEB_ARA_
Change/Delete Scheme dialog is a mirror
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause:The MessageBox function was
called without the flag MB_RTLREADING for the languages HEB and ARA.
Fix Description: Added the flag MB_RTLREADING for
message boxes (3)
when the dialog layout is RTL_LAYOUT
|
|
BugID: 1623874
|
Composite
TV display quality is not acceptable
|
softBIOS
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM Chipset
|
|
Resolution Description:
Root Cause: In latest TV has inbuilt notch
filters by default and by enabling in driver is causing the low
quality of display
Fix Description: Enable in the notch filter which
are as part of the horizontal luma
filters for Composite
|
|
Issues Resolved in 4306
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1612989
|
MIT/Aspen-U/XP/Rev1:
Brightness of panel will become dark after do S/R (Nx51v_E1.100)
|
softBIOS
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM Chipset, Intel(r) 855GME Chipset
|
|
Resolution Description:
Root Cause: For edid panels we used to get BLC/BIA data from panel
#16 even though OEM selected a different panel in VBT.
Fix Description: Use the panel selected by OEM in VBT
|
|
BugID: 1620451
|
CRT does
not turn on when suspend and resume with DDC mode.
|
miniPort
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM Chipset, Intel(r) 855GME Chipset
|
|
Resolution Description:
Root Cause: On resume from standby in DDC, Miniport queries softBIOS
for boot preferences and LFP single is returned.
Fix Description: Query and apply the boot preferences only during
resuming from hibernation or during boot time. During resuming from standby
the user can not change the boot preferences so do not query for boot
preferences.
|
|
BugID: 1613881
|
DCT 5.3
DXVA Display Test BSOD with VLD enabled.
|
DxVA
|
Windows*
XP
|
Intel(R)
945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: GPF was caused by the invalid pass-in pointer.
Fix Description: Check if dwSliceDataLocation is valid before using
it for memory access. In Addition, add WA for invalid dwNumMacroblock when
playing .drv-ms files and use width and height to calculate buffer size.
|
|
BugID: 1619765
|
Intel
14.12_4288: Limestone UMA Video driver displays video corruption when in
windowed mode.
|
D3D
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: The texture aliasing code was causing an issue when GMM
happened to flush commands when trying to do a block allocation. The 'GetTextureAlias' function made a very
poor assumption that it can delete the previous GMM Block in the case where
there was no longer a tag pending.
Fix Description: Properly tag and delete the original GMM after the
the copy BLT has been performed.
|
|
BugID: 1620828
|
Functional
Failure / DVD - Going into standby with Media Center App running in full
screen mode will cause BSOD
|
D3D
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM, Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset,
Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: FinishBinning was being called by the destructor when destroying
the context immediately after restoring from S3 state. This has a poor side effect when the
binner has already been reset, which happens on resume from ACPI Power
Management events .
Fix Description: Prevent sending anything to the HWB ring if it has
already been reset because of a return from S3 or any other power
management event.
|
|
BugID: 1617441, 1619268, 1623388
|
The
incorrect playback is displayed on LFP while playing the DVD movie on DDC mode.[RTA][64-bit]
|
DD
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM, Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: GMM_IMC_2_4_SURFACE was being reset by call to
GmmGetSurfaceFlags() instead of using the value passed into the
function.This is causing the right side of the overlay on the DFP to be
corrupt. Fix Description: The flag is set and maintained
when passing in parameters to the
AllocateIntermediateBuffer() function, instead of overriding the
flags with a call to GmmGetSurfaceFlags()
|
|
BugID: 1623245
|
Aspen
UMA PV1 M4, 14.13.0.4280: BSOD when close lid and open lid during MOH.
|
OGL
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM, Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause:Flip instructions are queued on a disabled plane causing
the ring buffer to hang.
Fix Description: In __glSrvCanFlip we disallow flipping when there
are no active planes.
|
|
Issues Resolved in 4302
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1593592, 1620563
|
The
corruption is displayed while playing the Far Cry game several minutes with
OpenGL mode.
|
OGL
|
Windows*
XP, Windows* XP-64,Windows* 2000
|
Intel(r)
915G Chipset, Intel(r) 855GME Chipset, Intel(R) 945G Express Chipset
|
|
Resolution Description:
Root Cause: The Shader Cache Index does not take into account the
shader constant, and the shader had been cached with th eincorrect
constant.
Fix Description: Remove the constants from the cache and change the
shader cache flow to accomodate this: gfxInitTexEnvConstants and
gfxGenTexenvShader were modified slightly.
The __GFXhnapaTexEnvSignatureState structure was placed inside the
__GFXhnapaTexEnvShaderCache structure and envColorUseMask and
oneMinusEnvColorUseMask are now cached.
The Store and Restore signature state is now done via pointer
reference rather than a structure copy.
To fix
1620563 the max FP Program ENV (__GL_MAX_PROGRAM_ENV_PARAMETERS_FP) limit
was raised from 24 to 32
|
|
BugID: 1584364
|
Second
Life exits to desktop
|
OGL
|
Windows*
XP
|
Intel(r)
915G Chipset
|
|
Resolution Description:
Root Cause: Missing allocation for a variable for the slice addresses
when mipmap generation was enabled.
Fix Description: Allocate appropriate space for slice addresses
that get used in the software path for mipmapgen also.
|
|
BugID: 1601077
|
The screen
shows abnormal blue color while running Glaze3D.
|
OGL
|
Windows*
XP, Windows* 2000
|
Intel(R)
945G Express Chipset
|
|
Resolution Description:
Root Cause: 16 bit Color Masks sent to GDI were wrong, should be RGB
instead of BGR.
Fix Description: Fixed Color Masks
|
|
BugID: 1622202
|
Check in
the file igfxcfg\Publics.h which was not checked in under the DCN 427406
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM, Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: Could not build MCCS on Settings page, because using old
MCCS_GET variable.
Fix Description: Update Variable in header file to match new
algorithm.
|
|
BugID: 1617109
|
F1 help dialog
box displayed outside of application window.
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G, Intel(r) 915GM, Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: The help text was
displayed at the mouse cursor position.
Fix Description: The help
text is displayed at the control position
|
|
BugID: 1621731
|
The Dot Crawl
cannot be reset to default setting after clicking the Restore
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G Chipset, Intel(r) 855GME
|
|
Resolution Description:
Root Cause: Code for Dotcrawl not restored on Restore defaults.
Fix Description: Include missing code.
|
|
BugID: 1619726
|
Unable to change Graphic Options from desktop
|
CUI2
|
Windows*
2000
|
Intel(r)
865G, Intel(r) 915GM, Chipset, Intel(r) 915G Chipset, Intel(r) 855GME
Chipset, Intel(R) 945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: The menu handles were destroyed before displaying on
Windows 2000.
Fix Description: The
destruction of menu handle is postponed till the menu is closed and then
destroyed.
|
|
BugID: 1621793
|
Video Standard
NTSC-443 showing incorrect ACP signal
|
AIM
|
Windows*
XP, Windows* 2000
|
Intel(r)
865G
|
|
Resolution Description:
Root Cause: the issue was due to incorrect programming of the Macrovision
register for NTSC-433 standard.
Fix Description: Corrected the programming for removing the color
stripes in NTSC-433 format when MV is enabled
|
|
BugID: 1618303
|
Unable to play macrovision protected DVD using component
connection on HDTV
|
AIM
|
Windows*
XP, Windows* 2000
|
Intel(r)
915G Chipset, Intel(R) 945G Express Chipset
|
|
Resolution Description:
Root Cause: On the player query of the MS function, driver masked
out the incorrect HDTV standard bits, making the current format
unavailable.
Fix Description: Do not mask out current formats, instead mask out
the unavailable SDVO HDTV out per the SDVO capabilities.
|
|
BugID: 1595690
|
Enabling
Additional Supported Input Resolutions for SDVO HDTV Device
|
AIM
|
Windows*
XP, Windows* 2000
|
Intel(r)
915G Chipset, Intel(R) 945G Express Chipset
|
|
Resolution Description:
Root Cause: Not all under scan modes for CH HDTV were supported.
Fix Description: Added support in the code for the new under scan
modes. The Chrontel CH7021 TV out part supports scaling many different
resolutions to various HDTV formats.
The CUI will not automatically change resolution or TV standard
based on making a change in the other field. Multiple combinations are selectable.
Root
Cause:
Fix
Description:
|
|
BugID: 1611327
|
Creating
failure paths for memory allocation failures
|
RM
|
Windows*
XP, Windows* 2000
|
Intel(r)
915G Chipset
|
|
Resolution Description:
Root Cause: Not handling failed memory allocations in RM was causing
BSOD with Driver Verifier.
Fix Description: Added graceful handling of memory allocations
|
|
BugID: 1556152, 1436715
|
Display
corruption or blank screen after lid close and open in extended desktop
with CRT+TV
|
miniPort
|
Windows*
XP
|
Intel(r)
915GM Chipset, Intel(r) 855GME Chipset
|
|
Resolution Description:
Root Cause: Miniport function MD_Lid_Validation allowed display switches
on all lid close events, even when the system was in Extended Desktop with
two non-LFP displays. This would
result in the existing configuration being disturbed and possible loss of
primary display.
Fix Description: Closing the lid will no longer force a display
switch when the system is in Extended Desktop using two non-LFP displays
|
|
BugID: 1621376
|
System
will Blue Screen when it returns from S4.
|
softBIOS
|
Windows*
XP
|
Intel(r)
915GM Chipset, Intel(r) 855GME Chipset
|
|
Resolution Description:
Root Cause: When setting timings for DVO/SDVO device, we access Null
Pointer in Mode_CenterImage function.
Fix Description: Return unsuccessful for null pointers in the
Centering function.
|
|
BugID: 1621796
|
Video
Standard NTSC-443/PAL-M showing incorrect ACP signal
|
softBIOS
|
Windows*
XP
|
Intel(r)
915GM Chipset
|
|
Resolution Description:
Root Cause: for NTSC 433 and PAL M if mode 2 and 3 are requested by
app, driver was setting separate register for them and it was showing
incorrect ACP signal.
Fix Description: For NTSC 433 and PAL_M even though if mode 2 and 3
are requested, driver has to set only the mode 1 registers. Includes fix
for ETM issue to avoid the standard TV change while CP is playing and and
remove showing of HDTV modes when SDTV connected.
|
|
BugID: 1597958, 1615711
|
Display is not centred and corruption is seen on 1080
HDTV standard
|
softBIOS
|
Windows*
XP
|
Intel(r)
915GM Chipset
|
|
Resolution Description:
Root Cause: unsupported mode was applied on TV Std and was causing
display blank.
Fix Description: Dynamically change the EDID for every TV standard,
to maintain consistence between OS and CUI and prevent unsupported modes to
be applied.
|
|
BugID: 1590887
|
Not able to apply 1080i and 780p on HDTV
|
softBIOS
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(R)
945G Express Chipset
|
|
Resolution Description:
Root Cause:The CE extension timings were not proper in m_CeShortVideoModes table for 1080i and 720p.
Fix Description: Updated the timing information for “1080i and
720p” in m_CeShortVideoModes table and include code to give priority to DTD
timings if the timings are present in both DTD and in CE extension.
|
|
Issues Resolved in 4293
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 426452
|
Star
Wars Knights of Old Republic - The Star Wars game exits to the desktop
automatically and the tool bar disappears after switching it back to full
screen mode.
|
OGL
|
Windows*
XP
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset, Intel(r) 865G Chipset, Intel(R) 945G
Chipset, Intel(R) 945GM Chipset
|
|
Resolution Description:
Root Cause: The OpenGL driver was freeing
resources incorrectly because not all items were being tagged correctly.
Fix Description: Fix was to make our
tagging mechaninsim more robust so that we could track all of our textures
resources.
|
|
BugID: 426965
|
Functional
Failure / PM – A blank screen is displayed on the TV after resuming from S1
standby.
|
AIM
|
Windows*
2000, Windows* XP
|
Intel(r)
915G Chipset, Intel(r) 945G Chipset
|
|
Resolution Description:
Root Cause: External TV-Out component was changing a command issued
to go to a lower D state, which happens on Standby request when two
displays are enabled.
Fix Description: Driver will work
around the issue by not programming the encoder twice, if already in that
state.
|
|
BugID: 427288
|
Constrain
User-Toggle of WaFlipStatus to Lakeport A0-A1/Calistoga A0
|
Miniport/KCH
|
Windows*
2000, Windows* XP
|
Intel(R)
945G Chipset, Intel(R) 945GM Chipset
|
|
Resolution Description:
Constrain the use of the WaFlipStatus registry key to toggle the
workaround of the same name to Lakeport A0/A1 and Calistoga A0 only. On
Lakeport production, the key is a No Op.
Root Cause:
Fix Description:
|
|
BugID: 427100
|
After
resuming from S3 No Display on TV
|
SoftBIOS
|
Windows*
XP
|
Intel(r) 915GM Chipset, Intel(R) 945GM
Chipset
|
|
Resolution Description:
Root Cause: The Powerstate variable for
Integrated TV out is initialized after the function that sets the HW state
is called, resulting in setting the previous power State, instead of
current requets.
Fix Description: Initialize the
Powerstate variable as part of the function IntTV_UpdateHardware, in the
correct sequence. Include some changes related to Pamimobj->Dwenabled
state and update the Power state
when ever there is a change in the DAC state.
|
|
BugID: 426933
|
Don't Disable unused pipes in Intel(R) 945G Chipset
|
SoftBIOS
|
Windows*
2000, Windows* XP
|
Intel(R)
945G Chipset
|
|
Resolution Description:
Currently we are disabling Unused pipes for Lakeport and above.
VBIOS wants this feature only for Mobile
platforms since they dont have space to incorporate this feature for
Desktops. Driver will not disable unused pipes for Lakeport to be in sync
with VBIOS
Root Cause:
Fix Description:
|
|
BugID: 427406
|
It takes
over 20 seconds to display the IGT page with MCCS Monitor.
|
CUI2
|
Windows*
2000, Windows* XP
|
Intel(r)
915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset, Intel(r) 865G Chipset, Intel(R) 945G Chipset,
Intel(R) 945GM Chipset
|
|
Resolution Description:
Root Cause: While launching, CUI looks
for MCCS capability which is slowing down the launch.
Fix Description: Query for MCCS capability
ony when Settings page is clicked. This will make CUI launch faster. Also,
when MCCS monitor attached.
|
|
BugID: 427404
|
Change
SetStdCtrl to SetVideoParam
|
CUI2
|
Windows*
2000, Windows* XP
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset, Intel(r) 865G Chipset, Intel(R) 945G
Chipset, Intel(R) 945GM Chipset
|
|
Resolution Description:
Change SetStdCtrl method to SetVideoParam method so that we
are in sync with the OS
Root Cause:
Fix Description:
|
|
BugID: 427140
|
Functional
Failure / Multiple_DTD_Panel_SD – CUI2 – CUI Information Page is missing
modes for LFP that support 50Hz
|
CUI2
|
Windows*
2000, Windows* XP
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset, Intel(r) 865G Chipset, Intel(R) 945G
Chipset, Intel(R) 945GM Chipset
|
|
Resolution Description:
Root Cause: Validation for Config data
when the hotplug/unplug occurs was incorrectly handling Clone devices,
leading to blank combos with 50Hz panels
and incorrect hotplug strings displayed.
Fix Description: Added code to handle
the Clone in the service layer.
|
|
BugID: 427098
|
Porting
fix to main: Display does not switch from TV to DFP when you are in 720p
standard on TV
|
CUI2
|
Windows*
2000, Windows* XP
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset, Intel(r) 865G Chipset, Intel(R) 945G
Chipset, Intel(R) 945GM Chipset
|
|
Resolution Description:
Root Cause: The display doesn't switch
from TV to DFP because the driver does not get the Dislplay switch call
from CUI for this format.
Fix Description: Confirm the successful
display configuration for this resolution/format to ensure we get the IOCTL
change message.
|
|
BugID: 427291
|
The
display cannot be switched to EFP from the DTCM.
|
CUI2
|
Windows*
2000, Windows* XP
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset, Intel(r) 865G Chipset, Intel(R) 945G
Chipset, Intel(R) 945GM Chipset
|
|
Resolution Description:
Root Cause: Incorrect array order for
tray/cm was preventing the display switch to occur from tray.
Fix Description: Changed the order in
commonutils.h file
|
|
BugID: 427251
|
Fix for customization
bugs and code changes for DCN 427289
|
CUI2
|
Windows*
2000, Windows* XP
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset, Intel(r) 865G Chipset, Intel(R) 945G
Chipset, Intel(R) 945GM Chipset
|
|
Resolution Description:
Root Cause: When the registry key is
used to get the data on whether the tray icon is enabled or disabled, it
uses KEY_WRITE access also. In this case CUI doesn't get the value from registry
and hence the tray icon is visible to the user.
Fix Description: Added code to check
for reading and doesn’t use the KEY_WRITE access; use this flag only for
writing it into the registry. Also for tray icon/context menu the tray icon
enable/disable option should not be available for limited user, which is an
issue where enable/disable from context menu won't work for limited user,
this is same as hotkeys enable/disable
|
|
BugID: 427413
|
Banner of
desktop moves when resuming from S3. Migrating fix to main baseline.
|
CUI2
|
Windows*
2000, Windows* XP
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset, Intel(r) 865G Chipset, Intel(R) 945G
Chipset, Intel(R) 945GM Chipset
|
|
Resolution Description:
Root Cause: Persistence module is
setting wparam & lparam to 0 when WM_DISPLAYCHANGE call is made to
broadcast messages as part of persistence algorithm (allowed by MS documentation).
This 3rd party app (softNavi) is incorrectly using these parameters for
processing data during power events, in which case this app is crashing.
Fix Description: Replace 0 with the
bits/pixel, width/height values when sending the WM_DISPLAYCHANGE.
|
|
BugID: 427286
|
CUI2
Localization merges to CUI2 main for linguistic fixes.
|
CUI2
|
Windows*
2000, Windows* XP
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset, Intel(r) 865G Chipset, Intel(R) 945G
Chipset, Intel(R) 945GM Chipset
|
|
Resolution Description: Merge various
linguistic fixes
Root Cause:
Fix Description:
|
|
Issues Resolved in 4281
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1612566
|
ADD2
card: Does Not detect TV.
|
AIM
|
Windows*
2000, Windows* XP
|
Intel(r)
915G Chipset, Intel(R) 945G Express Chipset
|
|
Resolution Description:
Root Cause: Incorrect wiring info was causing the
tv detection failure.
Fix Description: When setting the
input-out SDVO map for the existing configuration in the hardware, include
logic for the multifunction device and existence of two devices of the same
type (Cx).
|
|
BugID: 1603736
|
Anomalies
on game menu show up in online game
|
D3D
|
Windows*
XP
|
Intel(r)
915G Chipset, Intel(R) 945G Express Chipset
|
|
Resolution Description:
Root Cause: The application was clearing the Color
and Depth Buffers the first frame only, then clearing only the Color Buffer
afterwards. The hardware was not
evicting the Z-data for the whole Z-buffer on the first frame.
Fix Description: Changed the
default value to force Z-Evictions for the first frame
|
|
BugID: 1590946
|
Display on LFP is blank after resuming from Hibernation
by disconnecting CRT in EDS.
|
GDI, softBIOS
|
Windows*
2000, Windows* XP
|
Intel(r)
915GM Chipset
|
|
Resolution Description:
Root Cause: : RM/Softbios will not understand
DISPLAY_DONT_CARE as it is not a proper display device. Hence GDI used to
recieve junk value from RM which used to result in system blackout at the
end.
Fix Description: Avoid calling
RM/Softbios whenever a call is coming in for setting/resetting mode, GDI
will return true to OS.
|
|
BugID: 1596324
|
Optimize the physical page allocation and mapping for IMC
format video surface
|
GMM
|
Windows* XP
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset, Intel(R)
945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
When allocating physical pages for IMC (video) surface format,
there are some wastage to the righ of Y, U and V planes. There is even more
wastage when we can not find the heap with matching pitch and have
toallocate from a double pitch heap. This optimization will help our
"out of physical memory" issues to some degree when running
video.
Root Cause:
Fix Description:
|
|
BugID: 1610988
|
64-bit driver files do not follow 8.3 naming convention
|
OGL
|
Windows*
2000, Windows* XP, Windows*XP-64
|
Intel(r)
915G Chipset, Intel(R) 945G Express Chipset
|
|
Resolution Description:
Root Cause: DOS Naming convention not met.
Fix Description: Rename both 32
bit and 64 bit OGL Dll files and the code involved the names
|
|
BugID: 1440243
|
W-Macrovision BSOD-0xEA playing DVD and PM resume
|
softBIOS
|
Windows*
XP
|
Intel(r)
855GME Chipset, Intel(r) 865G Chipset
|
|
Resolution Description:
Root Cause: On Monitor Off, driver gets called to
turn off TV, which involves turning off Plane/pipe/DPLL. Turning off P/P/P
is done disregarding checking for flip pending hasn't been completed and on
the next driver call system has BSOD.
Fix Description: Separate the S
and D power state change sequences. This will allow us not to program the
Gfx Chipset PLLs in case of display power state change sequence
|
|
BugID: 1602960
|
Dynamic hotkey changes & Migrating Schemes changes to
Main with resource changes as requested by
loc team
|
CUI2
|
Windows*
2000, Windows* XP
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset, Intel(R)
945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description:
Root Cause: Dynamic hot keys cannot be Localized.
Fix Description: : Add method to
allow translation of hot keys, including call to GetKeyNameText function
and check keyboard layout. Include a
fix for panel hot key issue found during testing.
|
|
BugID: 1610664
|
CUI2: Schemes does not inform the user when overwriting
the existing scheme
|
CUI2
|
Windows*
2000, Windows* XP
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset, Intel(R)
945G Express Chipset, Intel(R) 945GM Express Chipset
|
|
Resolution Description: Localization changes as requested by loc
team.
Root Cause: Message box code missing in schemes.
Fix Description: Added the
confirmation message box
|
|
BugID: 1611243
|
Some controls in the TV-Settings are Greyed out for ADD2N
card for HDTV standards.
|
CUI2
|
Windows*
2000, Windows* XP
|
Intel(r)
915G Chipset, Intel(R) 945G Chipset
|
|
Resolution Description:
Root Cause: The values for adaptive flicker filter
to show/hide are not correct, and the flags used in Extended TV Param are also
incorrect.
Fix Description: Fix adaptive
flicker filter to show/hide and Extended TV Param flags.
|
|
BugID: 1578892
|
CUI Advanced Settings under Display Settings are not
functional
|
softBIOS
|
Windows*
2000, Windows* XP
|
Intel(R)
945G Chipset, Intel(R) 945GM Chipset
|
|
Resolution Description:
Root Cause: GMBUS transactions fail on some MCCS
monitors
Fix Description: Use BitBashing for
MCCS transactions on LPT and CST.
|
|
Issues Resolved in 4271
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1584561
|
Visual
anomaly / OGL - Hitman - some scenes are washed out with OGL enabled
|
OGL
|
Windows*
XP
|
Intel(R)
945G Chipset, Intel(R) 945GM Chipset
|
|
Resolution Description:
Root Cause: The shader cache key was being programmed with the
wrong texture unit data causing an erroneous cache hit. As a result the wrong shader was being
used which produced the "washed out" scenes with the OpenGL
renderer.
Fix Description: The fix is to program the shader cache key with
the correct texture unit data. The shader cache correctly registers a
"miss" and the correct shader is generated.
|
|
BugID: 1603409
|
System Blue
Screens while resuming from standby.
|
miniPort
|
Windows*
2000, Windows* XP
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset, Intel(R)
945G Chipset, Intel(R) 945GM Chipset
|
|
Resolution Description:
Root Cause: Register Save/Restore is not correct, but on
resume the state is corrected by driver mode set. Also FB access is done
before PTEs are enabled, causing invalid access.
Fix Description: Change the sequence of save/restore calls to softBIOs
and KCH for ACPI Power Management calls.
|
|
BugID: 1596947
|
when
overlay is running on single display you can see the clone settings are
visible
|
GDI
|
Windows*
2000, Windows* XP
|
Intel(r)
915GM Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset Intel(R) 945G
Chipset, Intel(R) 945GM Chipset
|
|
Resolution Description:
Root Cause: VBT bit check for block display switch when
overlay running was not ported to new DDC redesign.
Fix Description: Provide the 2 functions. IsCurrentMode() that
checks whether the request is for the current configuration and IsOverlayEnabled() checks whether overlay
is enabled or not if VBT bit is set to Block Display Switch When Overlay is
ON.
|
|
BugID: 1584456
|
When
HDTV connected using RGB Cable (HDTV 480i59). Corrupted fickling Screen is
displayed when Size and Position sliders moved to Zero in Televsion
Settings page.
|
softBIOS
|
Windows*
XP
|
Intel(r)
915GM Chipset
|
|
Resolution Description:
Root Cause: The position of the window size will go behind the
expected size before reaching the max under percentage.
Fix Description: hactive and vactive were wrong and these values
are re-calculated for P and I.
|
|
BugID: 1439345
|
FSDOS
comes in NTSC format when Display is set in PAL
|
softBIOS
|
Windows*
2000, Windows* XP
|
Intel(r)
915GM Chipset
|
|
Resolution Description:
Root Cause: Scratch pad implementation was not done to inform
the TV Std to VBIOS.
Fix Description: Scratch pad status flag register implementation is
done as part of the #RCR 289449 and Fix is required from VBIOS also.
|
|
BugID: 1579594
|
Port
Issue 1579594 to SourceCUI2
|
CUI2
|
Windows*
XP
|
Intel(r)
915GM Chipset, Calistoga
|
|
Resolution Description:
Remove
the workaround SyncOSAndDriver from CUI . Call SyncOSAndDriverThread when
config apply fail for LID Event in case of Persistence enabled
Root Cause:
Fix Description:
|
|
BugID: 1571077, 1596941
|
It
cannot set from D1(HDTV_480i59) to another mode on Device Settings
property.
|
CUI2
|
Windows*
2000, Windows* XP
|
Intel(r)
915G Chipset, Intel(R) 945G Chipset
|
|
Resolution Description:
Root Cause: This is new feature for SDVO TV out, Integrated TV
Out doesn’t cover this.
Fix Description: Add flip logic for both primary / secondary
devices as HDTV. Also fixed Hidemodes Unchecked issue due to HDTV work
around, Fixed 15 Sec message box not appearing issue.
|
|
BugID: 1603879
|
We get BSOD when change from NTSC- PAL-G standard in
SDTV
|
AIM
|
Windows*
2000, Windows* XP
|
Intel(r)
915G Chipset, Intel(R) 945G Chipset
|
|
Resolution Description:
Root Cause: The issue was that 720x480 resolution was not
supported for PAL modes.
Fix Description: Removed 720x540 mode and added 720x480 mode as a
common mode for all SDTV formats
|
|
BugID: 1602036
|
[Virtual
Desktop]Overlay does not follow scrolling of a desktop.
|
DD
|
Windows*
XP
|
Intel(r)
915GM Chipset, Intel(r) 915G Chipset, Intel(r), Intel(R) 945G Chipset,
Intel(R) 945GM Chipset
|
|
Resolution Description:
Root Cause: The function PanOverlays only called
HalUpdateOverlay2 if the ENABLE_SPRITE macro returned TRUE. The ENABLE_SPRITE macro should only be
called if we are in DDC. It was
being called in the single display case when the VBT bit to disable the 2nd
sprite was on. This caused the
ENABLE_SPRITE macro to return false and subsequently not call
HalUpdateOverlay2 leading to the 2nd overlay not to be updated on a screen
movement.
Fix Description: Change the ENABLE_SPRITE macro to only return FALSE
and check the VBT bit if we are in DDC.
Also moved the calls to ENABLE_SPRITE to inside HalUpdateOverlay2
and HalFlipOverlay2 so ENABLE_SPRITE doesn't have to wrap ever call.
|
|
BugID: 1601041
|
Overlay that
starts outside viewport in panning mode does not show up until you move the
mouse.
|
DD
|
Windows*
XP
|
Intel(r)
915GM Chipset, Intel(r) 915G Chipset, Intel(r), Intel(R) 945G Chipset,
Intel(R) 945GM Chipset
|
|
Resolution Description:
Root Cause: In HalUpdateOverlay and HalUpdateOverlay2, if
VIDEO_SHOW is requested but the overlay falls outside the viewport, overlay
is hidden
Fix Description: If the app is showing the sprite but it's out of the
viewport, make sure the OverlayOn SW flag is set.
|
|
BugID: 1592248
|
Intermittent
D3D failures reported when running 3DMark 2005
|
GMM
|
Windows*
XP
|
Intel(R)
945G Chipset, Intel(R) 945GM Chipset
|
|
Resolution Description:
Root Cause: The surface Flags passed in 3DMark2005 to D3D has
the Lakeport surface layout (below, right) flag bits 24 set. D3D was
passing this bit back to GMM causing a "right" layout surface to
be mapped as "Below" layout causing double mapping GTT entries.
Fix Description: : Mask the bits that should be internal use only
in GMM preventing clients from accessing them. Also in this fix we changed
the BPL alignment in for Lakeport from 4K to 16K per the spec.
|
|
Issues Resolved in 4264
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1583007
|
945G-Screen
does not refreshed correctly when enable Shadow Mapping on ShadowMap.
|
OGL
|
Windows*
XP
|
Intel(R)
945G Chipset
|
|
Resolution Description:
Root Cause: Clears were
not affecting the entire screen. In some situations the clip rectangle
dirty bit was not being set.
Fix Description: Added code
to set the clip rectangle dirty bit.
|
|
BugID:1598378
|
Movie
file is not played correctly on @EIZOKAN v.1.9.
|
DD
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(R) 945G Chipset
|
|
Resolution Description:
Root Cause: Incorrect code added that disallows
color-converting blits involving system memory XRGB8888 to ARGB8888.
Fix Description: Re-enable "no alpha" to "has
alpha" system-to-video blits (for XRGB to ARGB).
|
|
BugID:1581098
|
[WinXP-TE] Shape of eraser of
Windows Journal corrupt when using 90 degree or 270 degree with some
settings.
|
Rotation
|
TabletPC,
Windows* XP, Windows* 2000
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset, Intel(R)
945G Chipset
|
|
Resolution Description:
Root Cause: Incorrect creation and update of rotated cursor
bitmap.
Fix Description: Create bitmap for Rotated cursor and include code
in update cursor function to handle non-byte aligned widths case.
|
|
BugID:1594241
|
Stretch
mode does not effect between BIOS setting and CUI
|
miniPort
|
Windows*
2000
|
Intel(r) 915GM, Intel(r) 855GME Chipset
|
|
Resolution Description:
Root Cause: For SMI1.3,
driver does not update panel fitting status as setup by CMOS settings and
restores fitting status as saved in Registry.
Fix Description: For SMI revision 1.3 and above and the override
flag set on call, enable CMOS settings over the registry values
|
|
BugID:1574532
|
[Win2000
only]CRT does not turn ON, after Suspend LCD/CRT connecting CRT and Resume
disconnecting CRT.
|
miniPort
|
Windows*
2000
|
Intel(r) 915GM, Intel(r) 855GME Chipset
|
|
Resolution Description:
Root Cause: miniPort Driver does not correctly update the
internal structures power state of the Active device when
connect/disconnect CRT per SMI080Dh, and as a result softBIOS does not
receive the correct device UID for the W2000 re-enumeration of Power
events.
Fix Description: Update the miniport internal data structure with
required info before the DPC is queued for the OS enumeration.
|
|
BugID:1592409
|
Overlay
clipped 1 pixel off right and bottom in panning mode
|
GDI
|
Windows*
2000
|
Intel(r)
915GM
|
|
Resolution Description:
Root Cause: The overlay
pan window is not correctly initialized/updated (one less line
right/bottom).
Fix Description: Update overlay pan window for initialization of
shared info, cursor movement, get viewport.
|
|
BugID:1382048, 1382005
|
Full
sized DOS prompt may corrupted in DDC mode.
|
softBIOS
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(R) 945G Chipset
|
|
Resolution Description:
Root Cause: VGA Plane Control Register and DAC are incorrect.
Fix Description: After loading the VGA regs for mode 3, driver needs
to program the mode 3 timings in Full Screen DOS
|
|
BugID:1599092
|
Functional
Failure / CUI2 - The Driver Version is displayed Not Available in the IGT
page. [64-bit][REG]
|
CUI2
|
XP64
|
Intel(r)
915G Chipset
|
|
Resolution Description:
Root Cause: Filesystem redirector call sequence issue.
Fix Description: Disabled the redirector and re-enable it
after the getting the version.
|
|
BugID:1580590
|
On Lid
switch ON/OFF, CRT hotplug ,DFP hotplug does not persist the resolution
when Persistence bit is set in VBT
|
CUI2
|
Windows*
XP
|
Intel(r) 915GM, Intel(r) 855GME Chipset
|
|
Resolution Description:
Root Cause: When driver algorithm is set in VBT for
persistence when Lid switch ON/OFF, DFP hotplug was doing display switch to
correct device, butt it was not
setting the correct resolution.
Fix Description: In CUI and COM, add SetCurrentConfig code for DDC
/MDS modes to set correct resolution in these configurations.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Issues Resolved in
4259
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1598973
|
It will
become BSOD when movie is played on "Virtual Desktop".
|
DD
|
Windows*
XP
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(R) 945G Chipset, Intel(R)
945GM Chipset
|
|
Resolution Description:
Root Cause: The ClipRectsOutsideViewport function did not
return FALSE (to hide the overlay) when the overlay height or width was
zero after it was clipped to the viewport.
When function to show overlay was called and the scaling factor was
calculated, driver executed division with 0(height, width).
Fix Description:
The
ClipRectsOutsideViewport function checks the overlay width and height after
clipping, and if either one is zero, it should return FALSE to notify the
caller that the rectangles are not visible, and thus not take the code path
to hide the overlay.
|
|
BugID: 1590205
|
Monitor
signal out of range when loading GMCH driver
|
miniPort,
softBIOS
|
Windows*
XP
|
Intel(r)
915GM, Intel(r) 855GME Chipset, Intel(r) 915G Chipset, Intel(R) 945G
Chipset, Intel(R) 945GM Chipset
|
|
Resolution Description:
Root Cause: For Legacy
Monitor, we report the capabilities based on the static mode table build in
the driver/softBIOS, which includes all modes up to 20x15 and this highest
mode is used as Native mode.
Fix Description:
For No EDID
monitors, use the mode defined in VBT as Native mode, and for the VBT as 0,
default to 8x6 as the Native mode.
|
|
BugID: 1575182
|
Hang /
OGL - Homeworld 2 - The blue screen is displayed and the system hangs while
changing the Homeworld 2 game resolution.[REG]
|
OGL
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM
|
|
Resolution Description:
Root Cause: srvDrawable pointer was invalid, when referenced
it caused the blue screen.
Fix Description:
Created function
to search for the srvDrawable object.
If the object is not on the current srvDrawable list than the call
is ignored. This is valid because
the window event to destroy the wndobj structure can come after the
srvDrawable has been deleted.
|
|
BugID: 1594718
|
Can not
get scaling status correctly by API ICUIExternal2::GetConfiguation
|
CUI2
|
Windows*
XP, Windows* 2000,
|
Intel(r)
915GM
|
|
Resolution Description:
Root Cause: The scaling information was hard coded to false in
GetDeviceParameters function.
Fix Description:
Set the scaling information:
for Full Screen, Scaling is enabled, for Centering and MAS, Scaling is not
enabled
|
|
BugID: 1597477
|
Functional
Failure / CUI2 - Mosquito - The Restore Default values of Video Overlay do not
change to supposed value after reinstalling the graphics driver with the
modified INF file.[RTA]
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset, Intel(r) 865G Chipset, Intel(R) 945G
Chipset, Intel(R) 945GM Chipset
|
|
Resolution Description:
Root Cause:
Fix Description:
When
registry entries present CUI will be using the Driver IOCTL call to get default
values, else it will hard code these values. The default values can't be
all 0's.
|
|
Issues Resolved in
4251
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1594429
|
Checked
in code for co installer change for Calistoga
|
Co-Installer
|
Windows*
XP, Windows* 2000,
|
Intel(R)
945GM Chipset
|
|
Resolution Description: Add support for Calistoga DEVIDs for uninstall.
Root Cause:
Fix Description:
|
|
BugID: 1592361, 1593310
|
The line
control signals of D-connector is not working
|
AIM
|
Windows*
XP, Windows* 2000,
|
Intel(r)
915G
|
|
Resolution Description:
Root Cause: D-connector sense lines were not working; also
including additional fixes with D-Connector SDVO TV-out.
Fix Description:
Added workaround
in SetTiming function for activating the sense lines signals (using Get/Set
Ancillary op code); add fix for D-Connector switching to S-Video. For HDTV,
removed 640x400 and added 800x600 and 1024x768. Fixed format switching when MV protected
content is playing.
|
|
BugID: 1591429
|
W-DCT5.3
- LPG - DirectDraw\Overlay - Failed
|
DD
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(R) 945G Chipset, Intel(R)
945GM Chipset
|
|
Resolution Description:
Root Cause: Driver does
not initialize complete front and
back buffer, for the YUY2 overlay.
Fix Description:
Modify the clear
surface to ensure ample data is cleared for the overlay buffer. Do this by using the pitch instead of the
actual width, thus ensure the whole overlay is cleared to black at
creation.
|
|
BugID: 1591672
|
Hang /
PM - The blank screen is display and the system hangs while suspending the system
to S4 via pressing the lid switch with only LFP connected.[RTA]
|
softBIOS
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM Chipset, Intel(r) 855GME Chipset
|
|
Resolution Description:
Root Cause: Cannot disable pipe in LFP only mode with lid
switch close event based on attached ports; there won't be any ports
attached to the pipe since
GDI/Miniport has already disabled the display during hibernation.
Fix Description:
Disable pipe
only if the power state value is not a HIBERNATE call.
|
|
BugID: 1594864
|
Reboot/system
error occurs when changing window of Windows Media Player* in playing MPEG1
file.
|
DD
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset
|
|
Resolution Description:
Root Cause: RB hang due to driver incorrect overshrink
condition (Division by 2 was allowing a shrink ratio greater than 2:1 to be
written to GDG HW). caused by comparison
in new code that fails to downscale.
Fix Description:
When calculating the downscale, instead of dividing
source/destination, double destination and compare with source.
|
|
BugID: 1583982
|
Desktop screen
becomes 640x480/4bit after installing the ZoomText8.11.
|
GDI
|
Windows*
XP
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(R) 945G Chipset, Intel(R)
945GM Chipset
|
|
Resolution Description:
Root Cause: For apps calling
DrvEnableDriver more than once, the second call will incorrectly fail
without checking semaphore status.
Fix Description: Fail the second call only if the
semaphore was not already created.
|
|
BugID: 1401473
|
CH7017/CH7304
on MGM. when card is connected driver installation hangs or crashes
|
softBIOS
|
Windows*
XP, Windows* 2000,
|
Intel(r)
855GME Chipset
|
|
Resolution Description:
Root Cause: For MGM
the variable bInterna_LVDS is always true, and driver assumes that internal
lvds is always active.
Fix Description:
Add condition to
check for external LVDS when initializing devices, when setting timings and
power calls.
|
|
BugID: 1584037, 1591784, 1592118
|
Power
Settings shown in CRT when LFP is selected in Display Devices and clicked
on Display Settings. Bug fixes for 1584037, 1591784, 1592118
|
CUI2
|
Windows*
XP
|
Intel(R)
945G Chipset
|
|
Resolution Description:
Root Cause: N/A
Fix Description:
When display change happens CUI should close the
Power/MCCS/TV Settings dialog. Also, Include Dev ID for Lakeport for these
fixes.
|
|
BugID: 1592112
|
Code Checkin
for HDTV changes in CUI2
|
CUI2
|
Windows*
XP, Windows* 2000,
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(R) 945G Chipset, Intel(R)
945GM Chipset
|
|
Resolution Description:
Support for HDTV
in CUI2
Root Cause:
Fix Description:
|
|
BugID: 1594111
|
bug
fixes for 1594111 and other fixes found during testing
|
CUI2
|
Windows*
XP, Windows* 2000,
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(R) 945G Chipset, Intel(R)
945GM Chipset
|
|
Resolution Description:
Root Cause: Incorrect Refresh rate used for TV monitors caused
the mode set to fail.
Fix Description:
Pass to
requested Refresh Rate to CUI COM for TV modes. Includes Bug fix for paint
issues.
|
|
BugID: 1565504
|
TV
settings: Full screen check box not available
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
855GM Chipset, Intel(r) 915GM, Chipset, Intel(r) 915G Chipset, Intel(R)
945G Chipset, Intel(R) 945GM Chipset
|
|
Resolution Description:
Root Cause: TV Settings Fullscreeen check box resource not
called.
Fix Description:
Added resource
for check box and called the appropriate method.
|
|
BugID: 1584483
|
2048x1536x16/32@60Hz
cannot be set. But 75Hz can be set with panning. (EIZO T961)
|
softBIOS
|
Windows*
XP, Windows* 2000,
|
Intel(r)
915GM, Chipset
|
|
Resolution Description:
Root Cause: For LVDS,
we mark GTF modes encoder specific we get from AIM as EDID. These modes
will get disabled from the mode tables.
Fix Description:
Modify the
CheckGTF function to handle these special modes.
|
|
BugID: 1562365
|
Garbage display
on CRT after enable "Same
display configuration driven on both displays" on 2KSP4 OS.
|
softBIOS
|
Windows*
XP, Windows* 2000,
|
Intel(r)
915GM, Chipset
|
|
Resolution Description:
Root Cause: The garbage display or the flicker visible on CRT
in a CRT+LFP system, when hotkey ctrl+alt+F11 is pressed. The flicker was
observed on CRT because the DAC port was turned off in
util_set_controller_timing although the timing was to be set for pipe B which
is attached to the LVDS port.
Fix Description:
Check the DAC
port register value to obtain the pipe it is connected to. And if the
current pipe for which timing calculations are to be done is different from
the pipe DAC port is attached to then don’t turn off the DAC port.
|
|
Issues Resolved in
4245
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1562002
|
PC14.8
Visual Anomaly/LiveTV – WinXP MCE RTM – Texture flickering in Quake 3 Arena
while LiveTV is playing in the background.
|
OGL
|
Windows*
XP, Windows* 2000,
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(R) 945G Chipset, Intel(R)
945GM Chipset
|
|
Resolution Description:
Root Cause: When Quake3 is running in full screen and media
player is running with a higher resolution and is "off screen"
relative to Quake3's .d the code in __glSrvFlushScene() prevents GHAL_SyncFlushOp()
from performing a mid-zone Sync Flush via MMIO register INSTPM.
Fix Description:
Use syncTag
write to signify the end of OpenGL zone rendering.
|
|
BugID: 1557011
|
Fix for
MCE HDTV not running smoothly
|
D3D, GMM,
GDI, GDI
|
Windows*
XP, Windows* 2000,
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(R) 945G Chipset, Intel(R)
945GM Chipset
|
|
Resolution Description:
Root Cause: Driver cannot
map 256MB of aperture, because OS/DX is causing failure, which is apparent
when running HDTV live.
Fix Description:
Break the
aperture in Lockable/unlockable. Additional enhancement fix from PC14.10
solution is allowing the memory allocation from "User Lockable"
to exceed the 128MB of pre-assigned space. The driver will allocate from
the "Not User Lockable" space but it will map the block to get
"user map address" which can be used when DDraw wants to lock the
surface. This gives more dynamic flexibility in allocating Lockable space
and with this fix the HD content does not run into memory limitation.
|
|
BugID: 1582027
|
[Clone Mode]When "Fn+F5"
is pushed, a screen flickers violently or nothing is displayed.
|
softBIOS
|
Windows*
XP, Windows* 2000,
|
Intel(r)
915GM,
|
|
Resolution Description:
Root Cause: Flicker is caused because panel fitting is applied
incorrectly by VBIOS, when SoftBIOS does not update the correct mode in the
scratch registers the VBIOS functions use.
Fix Description:
softBIOS update
scratch registers with correct mode in Twin.
|
|
BugID: 1579404, 1579404
|
Fix for
Display switching issues in CUI2 - CT
|
CUI2
|
Windows*
XP, Windows* 2000,
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(R) 945G Chipset, Intel(R)
945GM Chipset
|
|
Resolution Description:
Root Cause: Registry access
from driver (the
VideoGUID location) and CUI are out-of-sync in 64bit OS.
Fix Description:
If CUICOM is
running in a WOW64 context, go to the driver for registry accesses.
|
|
Issues Resolved in
4240
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1574637
|
The
desktop screen does not scroll on SVGA and XGA.
|
AIM,
softBIOS
|
Windows*
XP, Windows* 2000, Tablet PC
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(R) 945G Chipset, Intel(R) 945GM
Chipset
|
|
Resolution Description:
Root Cause: SDVO encoder does not support scaling for HDTV
modes, as a result start button on most of the HD format get clipped.
Fix Description:
Enumerate specific
HD, underscanned modes using fake EDID and then center to fix TV
appearance. Only driver changes, no CUI2 changes.
|
|
BugID: 1591680, 1592355
|
Functional
Failure / DD - GenOV2003 - The background of the overlay is changed from
red to blue while running the GenOV2003 with setting Stretch function to
45% at the YVU9 pixel format.[RTA]
|
DD
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(R) 945G Chipset, Intel(R)
945GM Chipset
|
|
Resolution Description:
Root Cause: UV buffers are swapped at the wrong time causing
image corruption in DownscaleSourceBy3D.
Fix Description:
Downscale3D uses
a modified FourCC code when it should use original.
|
|
BugID: 1581428
|
DD
doesn't handle locks on non-lockable surfaces
|
DD
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(R) 945G Chipset, Intel(R)
945GM Chipset
|
|
Resolution Description:
Root Cause: DDraw was not handling requests to lock&draw non-lockable
surfaces used for off-screen bits, thus causing screen corruptions (cursor
missing or other blt corruptions).
Fix Description:
At the time of
the lock, copy the surface memory to newly allocated OS user space,
remembering where the original non-lockable memory was. To unlock, copy the surface data back
from user space to its original location and complete by deferring AGP free
calls.
|
|
BugID: 1578318
|
String
shaken by vertical scroll on Outlook Express
|
GDI
|
Windows*
XP, Windows* 2000
|
Intel(r) 915GM, Intel(r) 855GME Chipset
|
|
Resolution Description:
Root Cause: Ring buffer crossing Upper Watermark is causing
the problem. The ring buffer needs to be flushed, which takes time and hence
causes a shake.
Fix Description:
Calling “Idle Hardware” intermittently fixes the
issue, but there can be an impact on WinBench* benchmark numbers. To minimize the impact, “Idle Hardware”
is called only if “DrvCopyBit” is not being punted.
|
|
BugID: 1562366
|
Visual
Anomaly / CUI2 – Radio Buttons in CUI2 move or shift upwards when you click
on them.
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM, Intel(r) 855GME Chipset, Intel(r) 915G Chipset, Intel(R) 945G Chipset,
Intel(R) 945GM Chipset
|
|
Resolution Description:
Root Cause:
Fix Description:
Recalculate the
rect for drawing the radio button.
|
|
BugID: 1584278
|
Visual Anomaly
/ CUI2 - Gamma slider displays incorrect range
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM, Intel(r) 855GME Chipset, Intel(r) 915G Chipset, Intel(R) 945G
Chipset, Intel(R) 945GM Chipset
|
|
Resolution Description:
Root Cause: The numbers needed to be scaled from 10-50 to 1.0
to 5.0
Fix Description:
Scaled down the
numbers.
|
|
BugID: 1580393
|
Supported
Modes is "Not Available" in CUI page.
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM, Intel(r) 855GME Chipset, Intel(r) 915G Chipset, Intel(R) 945G
Chipset, Intel(R) 945GM Chipset
|
|
Resolution Description:
Root Cause: CUI should not display the page when EDID is not
supported.
Fix Description:
Added code to which will not display
the device page when EDID is not supported.
|
|
Issues Resolved in
4017
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1584835
|
INF file
change for changing the Resource ID of igfxdia.exe from 560 to 2945
|
INF
|
Windows*
XP
|
Intel(r)
915GM, Intel(r) 855GME Chipset, Intel(r) 915G Chipset, Intel(R) 945G
Chipset, Intel(R) 945GM Chipset
|
|
Resolution Description:
Root Cause: 560 is now a 270 degree Rotation ID while 2945 is
diag exe ID.
Fix Description:
Change the
Resource ID for igfxdiag from 560 to 2945
|
|
BugID: 1580053
|
W -
DCT5.2 - Intel(r) 915GM D5 BSOD OpenGL Scenario
|
OGL
|
Windows*
XP
|
Intel(r)
915GM
|
|
Resolution Description:
Root Cause: The rotated buffer was being placed on the buffer
list as well as attached to the render target, and was being freed at both
locations. This double free is causing the bluescreen.
Fix Description:
Free only the
rotation buffer from the attachment to the render target. Additional fixes:
"Default" return values from escapes that fail are corrected;
fixed blit throttling in rotation code; fixed access violation in ialmgdev
when a context creation fails.
|
|
BugID: 1565907
|
Corruption
/ D3D – Max Payne – Pressing ALT + TAB will create massive of display
corruption in the game
|
D3D
|
Windows*
XP
|
Intel(r)
915G Chipset, Intel(R) 945G Chipset
|
|
Resolution Description:
Root Cause: Persistent z-evict was not being turned back on
after Alt Tab was pressed, because it expected a primitive count of 0. Since the clear which was done after Alt
Tab counted as 1 primitive, the test didn't pass.
Fix Description:
Removed the
requirement of 0 primitives, which wasn't necessary.
|
|
BugID: 1582233
|
DVD
overlay is not shown on CRT in DDC mode.
|
DD
|
Windows*
XP
|
Intel(r)
855GME Chipset
|
|
Resolution Description:
Root Cause: Enabling/disabling Sprite in DDC was changed to be
based on VBT setting only (with some exception for older bios). For mobile platforms already in the
market with VBT bit 0, the customer
can no longer see overlay on both displays in DDC. On Desktop, there is no option for them to turn
enable sprite in DDC mode.
Fix Description:
Use the registry entry instead of VBT bit, and if the
registry does not exist, then the VBT bit setting will be used.
|
|
BugID: 1583156
|
Corruption
/ DD - GenOV2003 - The corruption is displayed while running the GenOV2003
with setting Mirror U/D function at the YVU9 pixel format. [RTA]
|
DD
|
Windows*
XP
|
Intel(r)
915G Chipset
|
|
Resolution Description:
Root Cause: Occurs because YUV Stride is not recalculated in
each flip for mirroring.
Fix Description:
Adjust Overlay
Regs YUVBuffers for mirroring.
|
|
BugID: 1584849
|
Execute plane
A, B, C, or overlay flips, go into PM event, return, and then system hangs
|
miniPort/KCH
|
Windows*
XP
|
Intel(R)
945G Chipset
|
|
Resolution Description:
Root Cause: On resume, IMR was restored with flip-status bits
unmasked, but the IIR was reset to 0, since it can't be written (so this
looks like flips-pending to any code that checks).
Fix Description:
Re-mask
flip-status in the IMR on resume.
|
|
BugID: 1582222
|
LID
disable extended view (CRT&DVI) and enable dual view (CRT&LCD) or
CRT only once it is release.
|
miniPort
|
Windows*
XP
|
Intel(r) 915GM, Intel(r) 855GME
|
|
Resolution Description:
Root Cause: Driver verifies that the UIDs passed by GDI with
mode set call are actually present
in the display UID list maintained by softbios. On a Lid Close in Extended
Desktop, the driver loops indefinitely when the UID on secondary is Don't Care is not found in the driver list.
Fix Description:
Checks whether any of the UIDs passed by GDI with mode
set call is Don't care and skips
searching for these UID.
|
|
BugID: 1579683
|
1280x720
resolution no longer available for LFP when hide mode unchecked
|
softBIOS
|
Windows*
XP
|
Intel(r)
915GM
|
|
Resolution Description:
Root Cause: Driver only adds those GTF modes which don’t have corresponding
EDID mode in Modetable. In case of LVDS, GTF Encoder specific modes are
marked Edid modes and modes are disabled from our modetable.
Fix Description:
Modify the
CheckGTF function to handle these modes as GTF modes.
|
|
BugID: 1572061
|
Half or
no display on wide screen panel
|
softBIOS
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM
|
|
Resolution Description:
Root Cause: We see corruption/bloom since we set Dual Channel Bit
for a Single Channel Panel.
Fix Description:
We should not
assume all panels with Xres >= 1280 are Dual Channel and need to read
LPF data from we determine whether a panel is Dual Channel or Single
Channel.
|
|
Issues Resolved in 4005
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1573625
|
On CUI Phase
2, the message in display device tab is erased when the window is moved to
bottom.
|
CUI2
|
Windows*
XP
|
Intel(R)
945G Chipset
|
|
Resolution Description:
Root Cause: Paint was not happening.
Fix Description:
When erase, invalidate
the page controls.
|
|
BugID: 1580635
|
Functional
Failure / CUI2 - CUI will hang after clicking "What's This?"
|
CUI2
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM, Intel(r) 855GME Chipset, Intel(r) 915G Chipset, Intel(R) 945G
Chipset, Intel(R) 945G Chipset
|
|
Resolution Description:
Root Cause:. Resource Issue.
Fix Description:
Fixed it by changing the property of
dialog
|
|
BugID: 1579422
|
Unable
to Apply new color Setting with CRT and DFP in Twin Mode
|
CUI2
|
Windows*
XP
|
Intel(R)
945G Chipset
|
|
Resolution Description:
Root Cause: For twin configuration, the primary/Secondary
devices concept is not used, so we should not treat the devices connected
as two separate devices as they are connected to a single pipe.
Fix Description:
Update the Builder structure only for one device.
|
|
BugID: 1558588
|
Unable
to check fullscreen check box for TV (if checked).
|
CUI2
|
Windows*
XP
|
Intel(r)
915GM
|
|
Resolution Description:
Root Cause: As the driver sends the fullscreen resolution as 1000
and Y= “don’t care”, we can check for X and Return the overscan value.
Fix Description:
|
|
BugID: 1581137
|
Display
on EFP flickers during mouse right click
|
UAIM
|
Windows*
XP
|
Intel(r)
915G Chipset
|
|
Resolution Description:
Root Cause: Right click
on DVI or TV causes both DVI detection and TV detection. During TV
detection, there is a flicker observed on DVI caused by turning SDVOB on
and then off.
Fix Description:
Remove SDVOB
On/Off.
|
|
BugID: 1582221
|
W-DCT5.2
- DirectDraw - Overlay Test &
VMR - "Overlay Scale" - Tests Fail
|
DD
|
Windows*
XP
|
Intel(r)
915G Chipset
|
|
Resolution Description:
Root Cause: Clipping function was not taking Pixel Doubling
into account when performing calculations.
Fix Description:
Added
adjustments in the Clipping function for handling pixel doubling.
|
|
BugID: 1575277
|
DMX 3.0
has black screen or video driver unloads when click on play.
|
DD
|
Windows*
XP
|
Intel(r)
865G
|
|
Resolution Description:
Root Cause: Bug occurs because YUV flipping chain has
different pitches.
Fix Description:
We must update
the YUV stride registers at every flip and we also must deny using surfaces
that have larger pitch than is supported by the overlay.
|
|
BugID: 1579367
|
865G
DCT5.2 XMemory/Texture Management failed
|
GMM
|
Windows*
XP
|
Intel(r)
865G
|
|
Resolution Description:
Root Cause: The issue shows up since we switched the
allocation algorithm in GMM to allocate from double pitched heap before
allocating linear heap (lockable aperture fix).
Fix Description:
Go back to the
old (pre-lock/unlock aperture) allocation algorithm.
|
|
BugID: 1578073
|
System
locks up/blue screen after hot undock with DVI
|
softBIOS
|
Windows*
XP
|
Intel(r)
915GM
|
|
Resolution Description:
Root Cause: On restore state from Hot-Dock linked to Suspend,
softBIOS was using incorrect saved mode returned from SDVO DVI functions.
Fix Description:
Modified
function to return current mode following AC/DC, display switch and docking
events.
|
|
BugID: 1566131
|
W-DCT5.2
- Power Management (PMTE) S3&S4
|
softBIOS
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM
|
|
Resolution Description:
Root Cause: When saving VGA Attribute Controller Registers,
the read requires a wait for Vertical Sync, which was timing out when VGA
plane/PipeA are disabled.
Fix Description:
Remove the
Util_Wait_Vsync.
|
|
BugID: 1551485
|
Integrated
TV display detection code does not allow PLLs to be stabilized before
device detection
|
softBIOS
|
Windows*
XP, Windows* 2000
|
Intel(r)
915GM
|
|
Resolution Description:
Root Cause: After setting
the dot clock frequency the PLL should get stabilized,
Fix Description:
After setting
the dot clock frequency, wait for 50micro seconds to stabilize PLL. Also
the Dot Clock frequency set was set in Hz but need to be set in KHz
|
|
BugID: 1574369
|
EFP is
not detected in PCI primary multimonitor settings. Screen goes to analog
display.
|
softBIOS
|
Windows*
XP
|
Intel(r)
915G Chipset
|
|
Resolution Description:
Root Cause: If UAIM SPD detection fails, it returns failure for desktop platforms.
Fix Description:
Some desktops support VBT based AIM devices, if SPD
detection fails, use VBT detection.
|
|
BugID: 1560489, 1579435
|
Functional
failure / CUI2 – PriRotateCUINormal – Hotkeys and DTCM transfers rotation
settings from secondary display (from Extended Desktop mode) to primary
display (Single display mode).
|
CUI2
|
Windows*
2000
|
Intel(r) 915GM, Intel(r)
855GME
|
|
Resolution Description:
Root Cause: When a new device is selected, CUI is trying to
get the new device’s rotation angle, which is not correct.
Fix Description: CUI
applies Current rotation for the newly selected device.
|
|
BugID: 1565825
|
System
hangs with black screen while system goes Standby (ACPI S3).
|
softBIOS
|
Windows*
XP
|
Intel(r) 855GME Chipset
|
|
Resolution Description:
Root Cause: Before entering into S3 need to
disable port and pipe but not PLL.
Fix Description:
Disabled the port and pipe (A, B) while entering into S3
|
|
Issues Resolved in
3997
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1561509
|
Screen
is interrupted/corrupted/Noise appear at Windows Logo screen while PowerON/restart WindowsXP.
|
AIM
|
Windows*
XP
|
Intel(r)
865G
|
|
Resolution Description:
Root Cause: Screen corruption is caused due to the change in
operating settings within a short span during post boot.
Fix Description:
Already
initialized by VBIOS, so there is no need to "reinitialize/change
operating settings". Change the operating setting whenever required.
|
|
BugID: 1580905
|
LID
disable extended view (CRT&DVI) and enable dual view (CRT&LCD) or
CRT only once it is release.
|
miniPort
|
Windows*
XP
|
Intel(r)
915GM Express Chipset
|
|
Resolution Description: Blue screen after changing
resolution while completing install by INF.
Root Cause: After installing without restarting the system, if we
try to change display resolution via OS display property sheet, a pointer
is dereferenced and this causes BSOD.
Fix Description:
The logic was
modified to make it generic so that it is not dependent on the order.
|
|
BugID: 1577503
|
W-DCT5.3
- 32BIT- LPG - Blt-Stretch&ColorKey - Vid to Primary - Failed
|
Rotation
|
Windows*
XP
|
Intel(R)
945G Chipset
|
|
Resolution Description:
Root Cause: We were using dual ring buffers.
Fix Description:
Disable dual ring
buffers.
|
|
BugID: 1579167
|
Unable
to Enable/Disable rotation in Extended desktop mode with CRT and DFP connected
|
CUI2
|
Windows*
XP
|
Intel(R)
945G Chipset
|
|
Resolution Description:
Root Cause: When Pri/Sec settings are Enabled/Disabled, need to
update both the devices in MDS Config.
Fix Description:
In MDS Config,
update rotation enable flag for both the devices.
|
|
BugID: 1579422
|
In
Aspect Scaling customization, when the Full Screen, MAS, and Centering are
removed, xxxFullScreen Buttonxxx
checkbox is shown.
|
CUI2
|
Windows*
XP
|
Intel(R)
945G Chipset
|
|
Resolution Description:
Root
Cause: For twin configuration, the
Primary/Secondary device concept is not used, so we should not treat the
devices connected as two separate devices as they are connected to a single
pipe.
Fix Description:
Treat the
devices as a single device.
|
|
BugID: 1439558
|
Rotation
Tab appears when change vBIOS even though Rotation is disabled by vBIOS.
|
Rotation
|
Windows*
XP
|
Intel(r) 915GM, Intel(r) 855GME Chipset
|
|
Resolution Description:
Root Cause: Was checking for the presence of a registry entry
Display1_EnableRotation, created by CUI based on VBIOS. On a VBIOS update,
Display1_EnableRotation was incorrectly created, even when rotation was not
enabled in the first place.
Fix Description:
Define new value
to Display1_EnableRotation, which will be set when Display1_EnableRotation
is not present in registry. Rotation will be received directly from VBIOS,
when Display1_EnableRotation is present
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Issues Resolved in
3998
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1561509
|
Screen
is interrupted/corrupted/Noise appear at Windows Logo screen while PowerON/restart WindowsXP.
|
AIM
|
Windows*
XP
|
Intel(r)
865G
|
|
Resolution Description:
Root Cause: Screen corruption is caused due to the change in
operating settings within a short span during post boot.
Fix Description:
Already
initialized by VBIOS, so there is no need to "reinitialize/change
operating settings". Change the operating setting whenever required.
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1580905
|
LID
disable extended view (CRT&DVI) and enable dual view (CRT&LCD) or
CRT only once it is release.
|
miniPort
|
Windows*
XP
|
Intel(r)
915GM Express Chipset
|
|
Resolution Description:
Blue Screen
after changing resolution while completing install by INF.
Root Cause: After installing without restarting the system, if we
try to change display resolution via OS display property sheet, a pointer
is dereferenced and this causes a BSOD.
Fix Description:
: The logic was
modified to make it generic so that it was not dependent on the order of
installation.
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1577503
|
W-DCT5.3
- 32BIT- LPG - Blt-Stretch&ColorKey - Vid to Primary - Failed
|
Rotation
|
Windows*
XP
|
Intel(R)
945G Chipset
|
|
Resolution Description:
Root Cause: We were using dual ring buffers.
Fix Description:
Disable dual
ring buffers.
|
|
Reference
No.
|
Description
|
Affected
Component(s)
|
Affected
OS(s)
|
Affected
Project(s)
|
|
BugID: 1579167
|
Unable
to Enable/Disable rotation in Extended desktop mode with CRT and DFP connected
|
CUI2
|
Windows*
XP
|
Intel(R)
945G Chipset
|
|
Resolution Description:
Root Cause: When Pri/Sec settings are Enabled/Disabled, need to
update both the devices in MDS Config.
Fix Description:
In MDS Config,
update rotation enable flag for both the devices.
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1579422
|
In
Aspect Scaling customization, when the Full Screen, MAS, and Centering are
removed, xxxFullScreen Buttonxxx
checkbox is shown.
|
CUI2
|
Windows*
XP
|
Intel(R)
945G Chipset
|
|
Resolution Description:
Root
Cause: For twin configuration, the
Primary/Secondary devices concept is not used, so we should not treat the
devices connected as two separate devices as they are connected to a single
pipe.
Fix Description:
Fixed the issue by treating the devices as a single device.
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1439558
|
Rotation
Tab appears when change vBIOS even though Rotation is disabled by vBIOS.
|
Rotation
|
Windows*
XP
|
Intel(r) 915GM, Intel(r) 855GME Chipset
|
|
Resolution Description:
Root Cause: Checked for the presence of a registry
entry created by CUI based on VBIOS. On a VBIOS update, registry key was
incorrectly created, even when rotation was not enabled in the first place.
Fix Description: Defined new value for registry
key, which will be set when key is not present in registry. Rotation will
be received directly from VBIOS when new value is set.
|
|
Issues Resolved in 3993
|
|
Reference No.
|
Description
|
Affected Component(s)
|
Affected OS(s)
|
Affected Project(s)
|
|
BugID: 1569890
|
Corrupted
video screen on 1280x720 when fullscreen WinDVD PB
|
DD
|
Windows*
XP
|
Intel(r)
915GM Express Chipset
|
|
Resolution Description:
Root Cause: Same
driver code used both when panel is in aspect ratio mode and in full
screen
Fix Description:
Code now only applies to full screen mode cases
|
|
BugID: 1566741
|
Fix for
slow scroll and DPST backlight save and restore
|
miniPort
|
Windows*
2000, Windows* XP
|
Intel(r)
915GM
|
|
Resolution
Description:
Root Cause: Gfx controller
register bit 11 was not properly being restored after resuming from S3 or
S4
Fix
Description: Register
is masked, so those mask bits needed to be accounted for. Includes fix for
issue where DPST is preventing the correct backlight settings based on
AC/DC, because of incorrect saving and restoring of the LBPC register.
|
|
BugID: 13955182
|
GHAL
support for DD_ColorKey Blt
|
DD
|
Windows*
2000, Windows* XP
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset, Intel(r) 865G Chipset, Intel(R) 945G
Chipset, Intel(R) 945GM Chipset
|
|
Resolution
Description:
Root Cause: We
were missing support of destination color key in combination with
mirroring, shrink/stretching and source color key
Fix
Description: Added
support for all combinations of RGB->RGB source/destination color key
and mirroring
|
|
BugID: 1433665
|
System
can not wake up from standby (ACPI:S3) when monitor power is working together
with PC. Both DDC1/DDC2B supported monitor is needed to replicate this
issue.
|
miniPort
|
Windows*
XP
|
Intel(r)
915GM, Chipset, Intel(r) 915G Chipset, Intel(r) 855GME Chipset, Intel(r) 865G Chipset, Intel(R) 945G
Chipset, Intel(R) 945GM Chipset
|
|
Resolution
Description:
Root Cause: On
monitors that support both DDC1 and DDC2, on resume, the monitor first
comes to DDC1 mode and after a random amount of time switches to DDC2B. As
a result, the driver does not detect the monitor saved.
Fix
Description: Search
the EDID and then apply mode set
|
|
BugID: 1558587
|
Right
move button for TV controls does not get disabled once limit reached
|
softBIOS
|
Windows*
2000, Windows* XP
|
Intel(r)
915GM
|
|
Resolution
Description:
Root Cause: We
were missing support of destination color key in combination with
mirroring, shrink/stretching and source color key
Fix
Description: Added
support for all combinations of RGB->RGB source/destination color key
and mirroring
|
|
BugID: 1572993
|
Problem
on DXVA host IDCT decode function
|
DxVA
|
Windows*
XP
|
Intel(r)
915GM, Intel(r) 855GME Chipset, Intel(r) 915G Chipset, Intel(R) 945G
Express Chipset (Lakeport), Calistoga
|
|
Resolution
Description:
Root Cause: DVD content
size is unexpected (mepg2 content is 1440x1080)
Fix
Description: Use
an equation to calculate the size of batch buffer dynamically
|
|
BugID: 1551154, 1551838
|
On lid
close in full screen DOS, display does not turn off
|
GDI,
miniport
|
Windows*
XP
|
Intel(r) 915GM, Intel(r) 855GME Chipset
|
|
Resolution
Description:
Root Cause: VBT
bits are set that do not allow display switch when in full screen DOS or
while overlay is on
Fix
Description: When
display switching is not allowed, turn off LVDS port if LFP was active
during lid close and subsequently turn it on lid open.
Clear
Hotkey and Lid flags in case of full screen DOS when display switching is
allowed.
|
|
BugID: 1579096
|
Multiple
CUI2 issues resolved: assigning hot
keys in CUI is not intuitive, color page applies wrong colors, Rotation hot
keys has an enable/disable issue
|
CUI2
|
Windows*
2000, Windows* XP
|
Intel(r)
915GM, Intel(r) 855GME Chipset, Intel(r) 915G Chipset, Intel(R) 945GM
Express Chipset (Lakeport), Calistoga
|
|
Resolution Description:
Root Cause:
Fix
Description: When values are different then all radio boxes
will be grayed out. When rotation is disabled, corresponding hotkey will get
disabled. When rotation is disabled, rotation menu will be hidden.
|
|
BugID: 1577778
|
With CUI
open, select any CRT panel fit option from DTCM, the same does not get
refreshed in CUI page
|
CUI2
|
Windows*
2000, Windows* XP
|
Intel(r)
915GM, Intel(r) 855GME Chipset, Intel(r) 915G Chipset, Intel(R) 945GM
Express Chipset (Lakeport), Calistoga
|
|
Resolution Description:
Root Cause: When
switching from DTCM, igfxcfg doesn't get display change notification
Fix
Description: Broadcast display change
|
|
BugID: 1578581
|
IGT tab
in Display Properties shows "Intel(R) Extreme Graphics 2" instead
of "Intel(R) Graphics Media Accelerator"
|
CUI2
|
Windows*
2000, Windows* XP
|
Intel(r)
915GM, Intel(r) 855GME Chipset, Intel(r) 915G Chipset, Intel(R) 945GM
Express Chipset (Lakeport), Calistoga
|
|
Resolution Description:
Root Cause: Incorrect
String table strings
Fix
Description: Updated string table strings
|
|
BugID: 1577795
|
CUI page
does not gets refreshed during ACPI hotkey switching
|
CUI2
|
Windows*
2000, Windows* XP
|
Intel(r)
915GM, Intel(r) 855GME Chipset, Intel(r) 915G Chipset, Intel(R) 945GM
Express Chipset (Lakeport), Calistoga
|
|
Resolution Description:
Root Cause:
Broadcast
display change functions missing
Fix
Description: Added broadcast display change in functions.
|
|
BugID: 1567368
|
Functional
Failure: cube tears into lines
|
DD,
softBIOS
|
Windows*
2000, Windows* XP
|
Intel(R)
945GM Express Chipset (Lakeport), Calistoga
|
|
Resolution Description:
Root Cause:
Fix
Description: Driver will check the IIR for flip-status
instead of polling the ISR
|
|
BugID: 1559757
|
Assume
CRT device for Secondary Extended device
|
GDI
|
Windows*
2000, Windows* XP
|
Intel(r)
915GM
|
|
Resolution Description:
Root Cause:
Incorrect behavior in
Extended Desktop with VBT bit. Display must be attached for switching with
hot-key.
Fix
Description: We now distinguish between the BIOS hotkeys and
OS property pages calls and enable or disable hotkey detection for BIOS
hot-keys or OS property pages respectively
|
|
BugID: 1573078
|
[Excel2003]
When "Smart Loupe" is installed, the character of the font of 20
points or more disappears.
|
GDI
|
Windows*
XP
|
Intel(r)
915GM, Intel(r) 855GME Chipset, Intel(r) 915G Chipset, Intel(R) 945GM
Express Chipset (Lakeport), Calistoga
|
|
Resolution Description:
Root Cause:
Anti aliasing not supported
Fix
Description: Punt the calls for antialiased text
|
|
|
|
|
|
|
|
|
|
|
|
|