Issues
Resolved in Build 4820
|
Reference
No.
|
Description
|
Affected
Component(s)
|
Affected
OS(s)
|
Affected
Project(s)
|
BugID:
2333313, 2334906, 2436382, 2439343
|
Extended Second Display using the integrated video card
shows blank after wakeup from sleep.
|
Resource Manager
|
Windows* XP, Windows* XP-64
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root
Cause: In DPDD S3/Resume, Secondary Bandwidth is getting added
up twice leading to bandwidth leakage. Bandwidth exhaust after couple of
S3/Resum calls. This is due to change in ED call sequence for Windows
Vista* in case of S3/Resum. In this case, Secondary SPSD call comes first
unlike normal sequence where Primary SPSD comes. This is leading to
overwriting GRM structures leading to wrong bandwidth calculations.
Fix Description: To avoid previous bandwidth issues also ported fix
changes done for Issue # 2292924 which was checked into main but not 15.2.
This fix removes the dependency of sequence calls. Cache the Value of
AttchedDisplayUID in DPDD primary case and use this to decide upon
bandwidth allocation for both primary and secondary. This fix takes care of
using right values for calculating bandwidth.
|
BugID:
2332069
|
The display blank out for DDC CRT+DFP applied through
context menu.
|
CUI2
|
Windows* XP, Windows* XP-64
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root
Cause: The issue was happening because with CRT+LFP, CRT
does not support 58 hz.
Fix Description: In validateConfig(), we check whether the RR is in
modelist or not. If not, we pass the defualt RR 60Hz to driver.
|
Issues
Resolved in Build 4793
|
Reference
No.
|
Description
|
Affected
Component(s)
|
Affected
OS(s)
|
Affected
Project(s)
|
BugID:
2256822
|
Image corrupted when Present operation is performed to a
Primary in LINEAR format.
|
Miniport, SoftBIOS
|
Windows* XP, Windows* 2000 Windows* XP-64, Windows
Vista*, Windows Vista*-64
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: There were 2 issues
here with respect to handling dynamic change in surface format in Windows
Vista* without a mode change. 1) In Miniport where only the parameter would
get updated if it was Tiled. So, Linear-->Tiled was taken care of, but
Tiled-->Linear changes were missing.
Fix Description: Fix was to handle this for cases which
would result in subsequent update of the Plane registers viz. 2) In SB
where it assumed that surface format would not change without an
EnablePlane. Corrected the logic to always set the tileX bit (Bit10) in
Plane Control Register appropriately depending on Tiled or Linear format.
Miniport functions which change Plane start addr -
Topology_SetSourceAddress and Topology_SetFlip. Also fixed another issue of
updating gamma in 8bpp modes via CUI path- ComSetGamma.
|
BugID:
2332452
|
LFP Polarity of H-Sync and V-Sync are incorrect.
|
SoftBIOS
|
Windows* XP, Windows Vista*
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root
Cause: In GetLVDSTimingInfo(), we were not getting the
flags from the DTD timing and we were saving the restoring the values as
were before. Because of the above, even if we specify the Negative polarity
in the DTD timing, we were not actually populating the same in the flags.
So the register 61180 was not getting updated with proper polarity. It is
always set to default values.
Fix Description: Changes are done along with the timing. Also populate
the Flags taken from DTD timing and removed the code of save and restore.
|
BugID:
2077788, 2329497
|
Adding
new indirect state model for handling CURBE data and surface
states. This model changes default model from a caching model to a
streaming model, however both models still exist in the code. Tests ran on
3Dmark03, 3DMark05, 3DMark06, FEAR, Sims2, Quake3, Doom3 on XP, and
3DMark03 on Vista.
|
D3D, GHAL3D, OGL
|
Windows* XP, Windows* XP-64, Windows Vista*, Windows Vista*-64
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root
Cause: GHAL3D CURBE and Surface State
Streaming Optimization.
Fix Description:
|
Issues
Resolved in Build 4786
|
Reference
No.
|
Description
|
Affected
Component(s)
|
Affected
OS(s)
|
Affected
Project(s)
|
BugID:
2329076, 2327752, 2329076
|
ACPI Hot Keys will not work after hot plug for the first
time.
|
Miniport
|
Windows* 2000, Windows* XP, Tablet Pc, Windows* 20003,
Windows* XP-64, Windows* 20003-64
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root
Cause: Event Manager Display List was not getting updated
during enumeration which is the list used while validating the
configuration to be set after hotkeys in the function EM_ValidateConfig.
Fix Description: Since this function was only being used in another place
apart from above where we do Display Switch in FSDOS, we removed the device
list connectivity check which will fix the issue. So basically we are not
updating the Event Manager Display List during enumeration as mentioned in
previous description. The reason was that to allow event manager display
list to be populated as need based on the events and flags rather during
enumeration.
|
BugID:
2077788, 2329497
|
GHAL3D CURBE and Surface State
Streaming Optimization.
|
Direct3D*, GHAL3D, OpenGL*
|
Windows* XP, Windows* XP-64, Windows Vista*, Windows Vista*-64
|
IntelŪ G965 Express Chipset
|
Resolution Description:
Root Cause: Adding new
indirect state model for handling CURBE data and surface states. This model changes default model from a
caching model to a streaming model, however both models still exist in the
code. Tests ran on 3Dmark03, 3DMark05, 3DMark06, FEAR, Sims2, Quake3, Doom3
on Windows* XP, and 3DMark03 on Windows Vista*.
|
BugID:
2326223
|
Video fades in and out when resuming from hibernation.
(Porting into PC14.27 branch)
|
SoftBIOS
|
Windows Vista*, Windows Vista*-64
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: Optimize the
persitence code which was causing too many flickers.
|
BugID:
2288162
|
Fix failures in DTM StretchRect test in BW-G Windows* XP.
(TIBET
#2288162)
|
Direct3D*
|
Windows* XP
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: The first
failing test #12316 has a resource that is a flip chain (front buffer +
back buffer). We called SurfaceBlt() with dstSubResourceIndex = 1, but did
not pass this index to D3DBltSrc(). In D3DBltSrc(), we recalculated the
index = 0. So we were copying to the dst surface with index 0 instead of
the surface with index 1. That was why the square sky image was missing. To
fix the tests, we need to pass in the dstSubResourceIndex (and
srcSubResourceIndex) to D3DBltSrc(). This re-test is only to test Windows*
XP drivers. The changes do not touch Windows Vista*. StretchRect, 3DMark05,
has been tested.
|
BugID:
2255316
|
Adding support for Continue and While Labels.
|
Direct3D*, GHAL3D, OpenGL*
|
Windows* XP, Windows* XP-64, Windows Vista*, Windows Vista*-64
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: Add support
for "Continue" and "While" Labels to more accurately
target jump destinations for these instructions. Also lay the groundwork
for function calls.
|
BugID:
2255316
|
DX10 Framework Changes.
|
Direct3D*, GHAL3D, OpenGL*
|
Windows* XP, Windows* XP-64, Windows Vista*, Windows Vista*-64
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: The DX10
framework includes the files necessary to compile the D3D10 dlls. For the
most part these changes will have a minimal impact on D3D9. There are some
changes to D3D9 and GHAL2D to support the new GHAL3D interface changes in
this drop which require the ETM. There are also some small changes to
GHAL3D internal implementation to support DX10. Inf changes were also
necessary to properly install the D3D10 dll if it was built prior to
makefile execution.
|
Issues
Resolved in Build 4775
|
Reference
No.
|
Description
|
Affected
Component(s)
|
Affected
OS(s)
|
Affected
Project(s)
|
BugID:
2077576, 2326924
|
GHAL3D VS - constant buffer optimization for indirectly
addressed constants.
|
Direct3D*, GHAL3D, OpenGL*
|
Windows* XP, Windows* XP-64, Windows Vista*, Windows Vista*-64
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: Optimization for
HW vertex processing. This optimization gets applied for workloads
containing constants with indirect registers. This optimization will
attempt to batch up 'send' instructions to minimize the number of sends
that we make.
|
BugID:
2326537
|
Incorrect Adapter Chip Type String for G965.
|
Miniport
|
Windows* 2000, Windows* XP, Tablet Pc, Windows* 20003,
Windows Media* Center, Windows*
XP-64, Windows* 20003-64, Windows Vista*, Windows Vista*-64
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: To change the
Chip Type string for G965 Broadater chipset to "Intel(R) GMA
X3000" instead of "Intel(R) GMA3000".
|
BugID:
2325767
|
DXVA encryption memory leak.
|
DVD
|
Windows* XP, Windows Media* Center, Windows Vista*
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: An unexpected
usage model of DXVA encryption in which encryption is toggled On and Off
with the same video session caused a memory leak because the scratch memory
for encryption was reallocated over and overwithout being freed. The
solution is simply only to allocate the scratch surface once and then check
it before allocating again.
|
BugID:
2291781
|
When HDMI is connected devicename in SystemPage, Device
tab is not in sync.
|
CUI2
|
Windows Vista*, Windows*
XP
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: While loading
the strings for Digital display in information page, the check for HDMI TV
was missing.
|
BugID:
2255351
|
Panel Fitting changes for Windows* XP and Windows Vista*.
|
Miniport, Power Conservation
|
Windows* 2000, Windows* XP, Windows* XP-64, Windows
Vista*, Windows Vista*-64
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: Purpose: This
changes were intended to support Panel Fitting via BIOS hotkeys. Details:
The support was required for both Windows* XP and Windows Vista*. To
Support panel fitting it required some changes in handling ASLE event.
|
BugID:
2255316
|
Removal of compiler and linker warnings in Direct3D*,
DIRECTDRAW*, and GHAL2D components.
|
Direct3D*, DirectDraw*, OpenGL*
|
Windows* XP, Windows* XP-64, Windows Vista*, Windows Vista*-64
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: This check-in
addresses the many compiler and linker warnings seen in the Direct3D*,
DIRECTDRAW*, GHAL2D, and some of the warnings in GHAL. Also the "Error
on warning flag" has been turned on for all components and configurations
except GHAL. (The GMM team has some work to do to address the rest of the
warnings in GHAL.)
|
BugID:
2255316
|
Fix using namespace GHAL3D in GHAL2D.
|
Direct3D*, DirectDraw*, DVD, GDI
|
Windows* XP, Windows Vista*
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: This DCN
removes the "using namespace GHAL3D" from the GHAL2D files.
Instead we will be prepending GHAL3D:: to all GHAL3D types to avoid naming
conflicts.
|
BugID:
2255316
|
Dynamic URB Allocation.
|
Direct3D*, GHAL3D, OpenGL*
|
Windows* XP, Windows Vista*
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: Change
allocation of URB space from a static model based on enabled fixed function
units to a dynamic model based on vertex size, per fixed function URB space
requirements, and an algorithm for optimal distribution of space.
|
Issues
Resolved in Build 4770
|
Reference
No.
|
Description
|
Affected
Component(s)
|
Affected
OS(s)
|
Affected
Project(s)
|
BugID:
2255620, 2257115
|
Display mode cannot be switched on CRT only mode via
pressing the ACPI hot key.
|
Miniport
|
Windows* 2000,
Windows* XP,
Windows* 20003,
Windows Media*
Center,
Windows* XP-64,
Windows* 2003-64,
Tablet Pc
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: There are two
issues targetted with this code fix. One is 2255620 and another is 2257115.
For details on issues, please refer to the peer review document in the
attachment. These issues are resolved in the latest version of Os and we
could not reproduce them later. So, passing the EMT as passed with issues
as we have found 14 production sightings which will be attached to the DCN.
|
BugID:
2194906, 2221920, 2287914,
|
Fix for DCN516790's Breaking Pre-IntelŪ Q33/Q35/G33
Express Chipsets OpenGL*.
|
OpenGL*
|
Windows* 2000,
Windows* XP,
Windows* 2003,
Windows Media* Center,
Windows* XP-64,
Windows* 2003-64,
Tablet Pc
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset,
|
Resolution Description:
Root Cause: DCN516790 had
OpenGL* pass the GMM a __GMM_NO_GPU_ONLY flag to let the GMM know OpenGL*
is not interested in receiving GPU_ONLY memory. The flag passing was meant
to be atemporary fix (until OpenGL* adds GPU_ONLY support), and since the
GMM was out of usable flag bits for this purpose, the __GMM_NO_GPU_ONLY is
actually the GMM_ GPU_ONLY flag, being used for a second purpose. GMM
support was only implemented to have OpenGL* to pass this flag in on >=
Bearlake, but the OpenGL* implementation passes the flag in on all
platforms. This breaks things when the GMM gets the unexpected flag on
pre-Bearlake platforms. This fix is simply to have OpenGL* only pass-in the
__GMM_NO_GPU_ONLY flag on platforms >= Bearlake (BLB + >= Gen4).
|
BugID:
2194906, 2221920, 2287914
|
Adding SW-Tiling Copy Engine to support 384MB DVMT GPU
per surface tiling.
|
Direct3D*,
DirectDraw*,
DVD,
GMM,
OpenGL*
|
Windows* 2000,
Windows* XP,
Windows* 2003,
Windows Media* Center,
Windows* XP-64,
Windows* 2003-64,
Tablet Pc
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: We are
tacking-on to the DCN (post-ETM) a registry setting with which we can
control the GPU_ONLY allocation system--So tiled GPU_ONLY or all of
GPU_ONLY can be disabled if necessary.
|
BugID:
2102022, 2255316
|
Gen5+ GHAL3D updates and fixes.
|
Direct3D*,
GHAL3D,
OpenGL*
|
Windows* XP,
Windows Vista*
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: 1) Change
line antialias alpha value from U0.4 to U0.8 when performing line
antialiasing in the pixel shader. 2) Fix compute into mrf assert with
Gen5+. 3) Switch threads in system thread when writing certainarchitecture
registers. 4) Add new SetDepthBufferCoordinateOffset client interface.
|
BugID:
2102022, 2255316
|
Implement AutoGenMipMap for IntelŪ 945G Express Chipset
and IntelŪ G965 Express Chipset Windows Vista*.
|
Direct3D*
|
Windows* XP,
Windows Vista*
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: Implement
AutoGenMipMap support for IntelŪ 945G Express Chipset and IntelŪ G965
Express Chipset Windows Vista*. Fix StretchRect failures seen in BW-G
Windows Vista* from first ETM run and DTM GetRenderTargetData failures on
BW-G Windows Vista* from second ETM.
|
BugID:
2292574
|
Hotplug detection does not work.
|
AIM
|
Windows* XP,
Windows Vista*
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root
Cause: When hot plug is enabled in CH7315, it generates an
interrupt. But driver does not receive this interrupt as interrupt is not
enabled in the GMCH by this time. As a result interrupt state is not
cleared in the encoder and consequently it blocks all future hot plug
interrupts. Code Changes: Since the issue is seen only with CH7315, fixed
the issue by putting a workaround in UAIMEH. Added a new function
UAIM_HotPlugInterruptWA() in UAIMEHDFP.c, which gets called after UAIM
initialization. In this function, hotplug is enabled for the output
corresponding to the pAIMObj. If there is an interrupt event at that
output, interrupt state is cleared. However, hotplug enable status is
restored before leaving this function.
|
BugID:
2290681
|
W - DCT5.3_QFE - IntelŪ G965 Express Chipset - PMTE
S3&S4 "Display settings cannot be restored" Error and
Corruption.
|
SoftBIOS
|
Windows* XP
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: Fix memory
leak in CSurfaceManager::DestroySurface2D().
|
BugID:
2290646
|
Verify_PanelPowerCycleDelay_and_ReferenceRegister - Panel
values inconsistent returning from Standby (S3).
|
SoftBIOS
|
Windows* XP
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: Issue: While
loading LVDS registers, SB uses current values of the registers with mask. But
while doing OR operation between calculated and current values, whole of
the current value is used instead of using approriate/masked bits only. For
PannelPowerCycleDelay and Reference Register (0x61210), current values are
different while rebooting and resuming from standby, so there are different
values at the end in these scenarios. Fix: Now ensuring that before loading
LVDS registers, only appropriate bit-values are taken from current value of
the register. Also, added code for having masks depending upon platform
type. Also, added changes to remove couple of warnings in SB.
|
BugID:
2290642
|
Playback of WMV Clip shows corruption.
|
DVD
|
Windows* XP,
Windows Media* Center,
Windows* XP-64
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: In VC1
advance frofile, use the exact cropped picture size as boundary instead of
MB aligned size passed to driver as uncompressed size.
|
BugID:
2255639
|
W - DTM RTM - Blt - (32-bit) -
Failing Log.
|
DirectDraw*
|
Windows* XP
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: Fix some unfinished
code. For DirectDraw* 16/32-bit depth fill blits, adjust the destination
format so it will work with GHAL2D colorfill (same as LH DepthFill API).
|
BugID:
2255316
|
For performance optimization in Windows* XP and Windows
Vista*: avoid looping for blockdesc (ie., in IsLost()) by adding a flag
HasMultipleAllocations to D3DResource; set flag to true if all surface
views share a single gmmblockdescriptor.
|
Direct3D*
|
Windows* XP,
Windows Vista*
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: This code
implementation is for performance gain. We see a degradation in Windows* XP::IsLost()
where we loop through every surface view to check the block descriptor.
This loop can be avoided by adding a flag in D3DResource that gets set if
all the surfaces share a single block descriptor. Fixed subtle bug after
modifying IsLost() to be better optimized. Further optimized IsLost() in
Windows* XP and fixed BSOD on FarCry seen on IntelŪ G965 Express Chipset
32. Also ran FEAR, 3dMark03, and 3dMark05 using the new ETM driver with no
issues.
|
BugID:
2255316
|
Change outdated filter type enums in SetSamplerMinFilter
and SetSamplerMagFilter.
|
Direct3D*
|
Windows* XP,
Windows Vista*
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: This is a side
work when I work on the bug 2172184 (i.e. it is not the cause of the
bug).In D3d\ibdw\SamplerStateManager.cpp, in two functions,
SetSamplerMagFilter and SetSamplerMinFilter, outdated
MAPFILTER_PYRAMIDALQUAD and MAPFILTER_NEAREST are used, which should be
D3DDDITEXF_PYRAMIDALQUAD and D3DDDITEXF_POINT. We are using the following
lookup table to convert from Direct3D* filter to GHAL3D filter. Fortunately
MAPFILTER_PYRAMIDALQUAD (4) will be mapping to
GHAL3D::SAMPLER_MAPFILTER_POINT, exactly same as what we want. Also
MAPFILTER_POINT (0) will be mapping to GHAL3D::SAMPLER_MAPFILTER_POINT. So
hardware was NOT programmed incorrectly due to these outdated enums for
filters.
|
BugID:
2223147
|
When system boots, noize appears after progress bar of
Windows Vista*.
|
SoftBIOS
|
Windows* XP,
Windows Vista*
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: Issue: We are
seeing some corruption during setmode in few customer systems. During Post,
VBIOS uses High Res Plane to display OEM Logos. This data in the frame
buffer is not cleared and hence the corruption. SB do not program Start
Address as part of setmode. There are cases that Start Address is not
available during setmode in Windows Vista*. Windows Vista* OS will call
Driver to BlankVideo before setmode and enables back the video after
setmode through SetSourceVisibilty DDI. But SB internally disables and
enables back the video as part of setmode. This causes the corruption. This
was required for Windows* XP to synchronize with render. The same is not
required in Windows Vista*.
Fix Description: I have added a new flag bVideoBlankingReqd in setmode interface.
Clients have to set/reset this flag based on their need. Windows* XP
setmode path has to set this flag and Windows Vista* setmode path has to
reset this flag. We need to set this flag in all cases within SB where we
use SetMode Interface.
|
BugID:
2211051
|
Choppy Playback with Region-2 DVD on Sonic Player.
|
DVD
|
Windows* XP,
Windows* XP-64,
Windows Media* Center
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: During
deinterlacing, when the formats of current and reference samples donot match,
instead of switching from Adaptive DI to BOB, the DXVA driver was
incorrectly switching to progressive mode and causing shaking of frames.
This DCN also modifies the way content discontinuity is detected when
advanced DI(Modeast or Adaptive) is requested. Instead of checking for
renderTarget time-stamp to be within a band, it checks if the reference
sample is correct one (true prev sample).
|
BugID:
2197021
|
Flicker is displayed on HDTV after switching the desktop resolution
to 1766x1000 with HDTV_1080i59 video standard.
|
TVout
|
Windows* XP,
Windows Media* Center,
Windows Vista*,
Windows Vista*-64
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root
Cause: Because 3 Tap Filter was not enabled, hence the
text readability issue.
Fix Description: 1) If the vertical filter is being used the pipe
interlace bit should be off and the TV interlace bit should be on. 2) If
the vertical filter is not being used the pipe interlace bit should be on
and the TV interlace bit should be off. 3) If Vertical filters are enabled
then we have to program the Field_swap mode (BIT15 of 68000) to Normal mode. 4) Load 3 Tap Filter Coefficients as
given by Todd. Currently only one set of coefficients is given and its
doesnt have the granularity as found in 5 Tap filters where we have
atleast 11 sets for each of the sharpness value. 5) 3 Tap Filter needs to
be enabled only when the displays resolution is greater than 1024.
|
Issues
Resolved in Build 4762
|
Reference
No.
|
Description
|
Affected
Component(s)
|
Affected
OS(s)
|
Affected
Project(s)
|
BugID:
2194906, 2221920, 2287914
|
Adding SW-Tiling Copy Engine to support 384MB DVMT GPU
per surface tiling.
|
Direct3D*, DirectDraw*, DVD, GMM, OpenGL*
|
Windows* 2000, Windows* XP, Windows* 20003, Windows
Media* Center, Windows* XP-64, Windows* 20003-64, Tablet Pc
|
IntelŪ GM965 Express Chipset
IntelŪ GME965 Express Chipset
|
Resolution Description:
Root Cause: a setting
with which we can control the allocation system--So tiled GPU_ONLY or all
of GPU_ONLY can be disabled if necessary. Also removing the 224MB gfx mem
capping for IntelŪ Q33/Q35/G33 Express Chipsets.
|
Issues Resolved
in Build 4755
|
Reference
No.
|
Description
|
Affected
Component(s)
|
Affected
OS(s)
|
Affected
Project(s)
|
ID: 628767
|
Combined
scissor and VB optimizations.
|
OGL
|
Windows* XP, Windows* XP-64, Windows* 2000-64, Windows
Vista*, Windows Vista*-64
|
All Platforms
|
Resolution Description:
A combination of
optimizations involving scissor rectangle and vertex buffers that increase
the driver performance as measured by some benchmarks.
Root Cause:
|
Issues
Resolved in Build 4752
|
Reference
No.
|
Description
|
Affected
Component(s)
|
Affected
OS(s)
|
Affected
Project(s)
|
BugID:
2222127
|
BSpec Violation - Multiple events occur during a
MI_WAIT_FOR_EVENT instruction.
|
DirectDraw*
|
Windows* XP, Windows* XP-64, Windows* 2000, Windows*
2000, Windows* 2000-64, Windows Media* Center.
|
All Platforms
|
Resolution Description:
Root Cause: This fixes
DCN #516827 which ORed in a GFX_WAIT_FLIPPENDING_B or GFX_WAIT_FLIPPENDING
to a "GFX_MI_WAIT_FOR_EVENT | GFX_WAIT_FLIPPENDING_OV". A wait
for event can only have one event/condition specified. This was causing a
hang on IntelŪ GM965 Express Chipset and IntelŪ Q33/Q35/G33 Express
Chipsets-B. Will fix by separating into two separate wait for events.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|