201

(63 replies, posted in DriverPack Mass Storage)

I wonder, to resolve last doubt:

did anyone tried default XP sources and added iastor 7.8 at F6 floppy?
Do you get a BSOD at ICH8-M AHCI still?

202

(63 replies, posted in DriverPack Mass Storage)

I tried a ICH7-M at ACHI mode: PCI\VEN_8086&DEV_27C5&CC_0106

Base8040RC, MassStorage 804 and XP SP3RC2.

Windows installation did work, no BSOD.

Dirvers from \D\M\I3\IAAHCI.inf and \D\M\I3\Iastor.sys are used.
A ICH8-M should use this drivers too. And Lenovo distribute iastor.sys v7.8 for ICH8-M too.
Sorry no idea.

203

(63 replies, posted in DriverPack Mass Storage)

@adam_w
Which driver do you use?
Do you use a ThinkPad driver?
http://www-307.ibm.com/pc/support/site. … MIGR-62909

@abdou
You may change to AHCI, if you prepared running windows.

Another question:
Did you enable 'DriverPack MassStorage text mode at DriverPacks BASE ?

ricktendo64 wrote:

I have a old PCI Geforce 5 card and it only works good with the driver v6.14.10.9371

NVidia offers driver from 2002
http://www.nvidia.com/object/winxp-2k_archive.html
http://www.nvidia.com/object/winxp_2k_93.71_2.html

@Hectorfx

Be aware, I've no appropiate hardware to test.
I'm guessing. This may work or be nonsense.

Idea was

[V2]
ms_1_deviceName="VIA ATA/ATAPI HOST CONTROLLER"
...
ms_1_exc_disableIfOS="w2k,wxp,w2k3"

[V3]
ms_1_deviceName="VIA SATA/PATA IDE"
...
ms_1_exc_disableIfOS="w2k,wxp,w2k3"

[V4]
ms_1_deviceName="VIA HYPERION SATA IDE"
...
ms_1_exc_disableIfOS="w2k,wxp,w2k3"

PCI\VEN_1106&DEV_5372&SUBSYS_1B471019&REV_00\3&267A616A&0&78: VIA Serial ATA Controller - 5372

M\V4\viprt.inf contains
%PCI\VEN_1106&DEV_5372&CC_0101.DeviceDesc%=StorageBus_Inst_x32, PCI\VEN_1106&DEV_5372&CC_0101
That's a SATA controller at IDE emulation mode.

VIA Hyperion contains a txtsetup.oem for VIA RAID mode.
Contrary there is no txtsetup.oem for VIA IDE mode.

Remember a driver floppy is the generic solution to include massstorage drivers.
A textsup.oem is required for F6 floppy drivers.
And txtsetup.sif contains a fallback: PCI\CC_0101 = "pciide"

vminiide.inf:
; INF File for MiniIDE and Hot-Plug Driver

I wonder:
Why dosn't include VIA a IDE txtsetup.oem?
Does VIA IDE drivers support textmode setup?
Is a Hot-Plug Driver required at textmode setup?
How to install XP?

VIA's opinion could be:
a end user install XP at textmode PCI\CC_0101 = "pciide"

To summarize, what about VIA IDE:
-do not include any drivers for textmode setup
-include VIA IDE drivers for PNP setup

catdog8 wrote:

I'm using bartpe only.  I do have other mass storage drivers installed. Not LSI but several others.

For testing purposes, use driverpacks massstorage drivers only.
Do NOT add own massstorage drivers. They may interfere, results maybe confusing.

PCI\VEN_8086&DEV_27C3 remids to a ICH7R.

http://download.intel.com/support/mothe … ps_1_1.pdf
There is a  Intel  82801GR (ICH7R) mentioned, no addional massstorage controller.

This describes a 'Embedded SATA Software RAID for ICH7R'
http://manuals.fujitsu-siemens.com/serv … a-raid.pdf

Therefore a PCI\VEN_8086&DEV_27C3&SUBSYS_348F8086 use ICH7R hardware and LSI drivers.

Does a iastor.sys conflict exist?

DP_MassStorage_wnt5_x86-32_8022nightly_bartPEtest.7z :
"txtsetup.sif","HardWareIdsDatabase","PCI\VEN_8086&DEV_27c3&SUBSYS_348F8086", """MegaINTL"""
sil 3132:
"txtsetup.sif","HardWareIdsDatabase","PCI\VEN_1095&DEV_3132&SUBSYS_71321095", """3132R5C"""

Already integrated driver should work.
Try current test release. Again, do NOT add own massstorage drivers.

Do you use XP source files? Try 2003 source files too.

False question. Do not mix driver packs.

Which massstorage driver pack version do you use?

I uderstand, you use http://www.911cd.net/forums//index.php? … p;p=142156

PCI\VEN_1095&DEV_3132&SUBSYS_71321095&REV_01\4&2D8B019B&0&00E4
Name: Silicon Image SiI 3132 SoftRaid 5 Controller

PCI\VEN_8086&DEV_27C3&SUBSYS_348F8086&REV_01\3&61AAA01&0&FA
Name: Intel Embedded server RAID Technology

PCI\VEN_8086&DEV_27DF&SUBSYS_348F8086&REV_01\3&61AAA01&0&F9
Name: Intel® 82801G (ICH7 Family) Ultra ATA Storage Controllers - 27DF

Silicon Image SiI 3132 SoftRaid 5 Controller dosn't work.

Do you use Silicon Image firmware 7.4.05 ?
http://www.siliconimage.com/support/sup … sid=4&
There are driver version 1.5.18.0.

SC?\SI3132 contain different SoftRaid 5 Controller already.
How to select a working one?

Adding massstorage drivers dosn't change NIC drivers.
Maybe another change gives different behaviour.

Acutally a SiI3132 example, but there is a full Hardware ID list.
As known Silicon Image offers different firmware files.
http://www.siliconimage.com/support/ind … cid=3&

catdog6 flashed different firmware and deliverd full HardwareList
http://www.911cd.net/forums//index.php? … p;p=142234

I flashed the firmware with the b7405 and the r7405. Here are the results.

B7405

PCI\VEN_1095&DEV_3132&SUBSYS_31321095&REV_01\4&2D8B019B&0&00E4
Name: Mass Storage Controller
Hardware ID's:
PCI\VEN_1095&DEV_3132&SUBSYS_31321095&REV_01
PCI\VEN_1095&DEV_3132&SUBSYS_31321095
PCI\VEN_1095&DEV_3132&CC_018000
PCI\VEN_1095&DEV_3132&CC_0180
Compatible ID's:
PCI\VEN_1095&DEV_3132&REV_01
PCI\VEN_1095&DEV_3132
PCI\VEN_1095&CC_018000
PCI\VEN_1095&CC_0180
PCI\VEN_1095
PCI\CC_018000
PCI\CC_0180
1 matching device(s) found.

Here's the R7405

PCI\VEN_1095&DEV_3132&SUBSYS_71321095&REV_01\4&2D8B019B&0&00E4
Name: RAID Controller
Hardware ID's:
PCI\VEN_1095&DEV_3132&SUBSYS_71321095&REV_01
PCI\VEN_1095&DEV_3132&SUBSYS_71321095
PCI\VEN_1095&DEV_3132&CC_010400
PCI\VEN_1095&DEV_3132&CC_0104
Compatible ID's:
PCI\VEN_1095&DEV_3132&REV_01
PCI\VEN_1095&DEV_3132
PCI\VEN_1095&CC_010400
PCI\VEN_1095&CC_0104
PCI\VEN_1095
PCI\CC_010400
PCI\CC_0104
1 matching device(s) found.

This HardwareID classify different firmware at this example:
PCI\VEN_1095&DEV_3132&CC_0180 : B7405
PCI\VEN_1095&DEV_3132&CC_0104 : R7405

Can this be used at PCI\VEN_1095&DEV_3112&CC_0104 and PCI\VEN_1095&DEV_3132&CC_0180 too?

What about enhancing 'HWID Tool'? Store 'devcon.exe hwids "PCI\CC_01*"' too?
Some driver inf files use CC_01*.
Knowing full Hardware ID list may get clearer results.
In addition there maybe new solutions possible in future. e.g. use REV_*

Jaak wrote:

25 INF have &CC_010

Yes, that's true.

However I count zero SUBSYS_*&CC_01*.

I don't own a Si3112r, I used VMware instead for testing purposes.
devcon list

PCI\VEN_8086&DEV_7111&SUBSYS_197615AD&REV_01\3&61AAA01&0&39
    Name: IDE Controller
    Hardware ID's:
        PCI\VEN_8086&DEV_7111&SUBSYS_197615AD&REV_01
        PCI\VEN_8086&DEV_7111&SUBSYS_197615AD
        PCI\VEN_8086&DEV_7111&CC_01018A
        PCI\VEN_8086&DEV_7111&CC_0101
    Compatible ID's:
        PCI\VEN_8086&DEV_7111&REV_01
        PCI\VEN_8086&DEV_7111
        PCI\VEN_8086&CC_01018A
        PCI\VEN_8086&CC_0101
        PCI\VEN_8086
        PCI\CC_01018A
        PCI\CC_0101
1 matching device(s) found.

There is DEV_*&CC_0101.
Contrary there is no SUBSYS_*&CC_01*.


Next I set intelide at
Control\CriticalDeviceDatabase\PCI#VEN_8086&DEV_7111&SUBSYS_197615AD&CC_0101
and removed intelide from
Control\Class\{4D36E96A-E325-11CE-BFC1-08002BE10318}

Reboot: BSOD 0x7b


Next I set intelide
Control\CriticalDeviceDatabase\PCI#VEN_8086&DEV_7111&SUBSYS_197615AD

Reboot: Windows does boot.


I'm convinced: SUBSYS_*&CC_01* is not a valid HardwareID.

Jaak wrote:

There are a few drivers which do use those tags in PnP driver.

I've seen PCI\VEN_*&DEV_*&CC_01* at driver inf files.

Which driver file use PCI\VEN_*&DEV_*&SUBSYS_*&CC_01* ?

Skeptik wrote:

Hardware IDs for this notebook:
PCI\VEN_1106&DEV_0571&SUBSYS_06641558&REV_07\3&61AAA01&0&79 : VIA Bus Master IDE Controller - 0571
PCI\VEN_1106&DEV_0591&SUBSYS_06641558&REV_80\3&61AAA01&0&78 : VIA Serial ATA Controller - 0591

ViPrt.inf contains "VIA Serial ATA Controller - 0591"
Therefore BIOS uses IDE mode. This should be a PCI\VEN_1106&DEV_0591&CC_0101
To confirm please run 'devcon.exe hwids PCI\CC_01*'. Post output.

VIA HyperionPro contains a txtsetup.oem for VRAIDDrv version.
viamraid.sys should work at textmode.

But there is no txtsetup.oem for SATAIDE.
ViPrt.sys may not be tested at textmode. ViPrt.sys may result a BSOD 0x7b at textmode.

Can you disable [V4] section and try again?

Jaak wrote:

fixed; no more CC_0101 in S4 (I listened)

What about CC_0104 in S5?

I doubt that's a valid PNP ID: PCI\VEN_1095&DEV_3112&SUBSYS_*&CC_0104

Well, quoting myself:

This RIS Driver is NOT recommended for normal windows.

Yes RIS driver and normal driver use the same HardwareID. Mixing both will get funny results.

Massstorage section use a *.ini file.  Base read this ini file.
NET section use a *.ini file too. Does base support excluding NET drivers?

Idea:
Include both: RIS and normal driver. Mark them at ini file.
At base select either RIS or normal driver.

Generally Net drivers does work at PE.
Of course there are exeptions. Some Net drivers behave different.

I know so far:

Broadcom NetXtreme II
http://www.broadcom.com/support/etherne … remeii.php
This driver use a Virtual Bus Device architecture.
D\L\B4: There are two inf and sys files for one nic device.
This driver dosn't work at RIS.
Maybe this could work at PE, a special plugin would be required.

Broadcom recommends the RIS driver for a Windows Preinstallation Environment.
There is one inf and sys files for one nic device.
PE community use this RIS driver.


Another major pain is nforce. There are a lot of driver versions.
And there are two inf and sys files for one nic device.

Nvidia dosn't offer RIS drivers to a end customer.
Some OEM offers nforce RIS drivers.
There is a pdf included: Nvidia recommend a RIS Driver for WinPE.
This RIS Driver is NOT recommended for normal windows.

http://support.fujitsu-siemens.com/Down … 5DCE8A753F
nForce Family Network Driver  Date: 05.10.2006   Version:    65.3.4 RIS_PE

There are drivers 67.5.3 from 2007:
ftp://ftp.asus.com.tw/pub/ASUS/misc/uti … S_6753.zip



Some other exceptions:
Attensic use a strange subfolder: system32\Attansic\L1\atcInst.dll
http://www.911cd.net/forums//index.php? … p;p=122098

D-Link DGE-530T is similar: system32\wbem\mof\gewmi.mof
http://www.911cd.net/forums//index.php? … mp;p=87509


Unfortunately I don't have acess to above mentioned hardware.
I've to guess from user experiences.

Freece at monitor installation:
this reminds to
http://www.msfn.org/board/Unattend-Inst … 39479.html
http://support.microsoft.com/kb/883667/

As for sysprep http://support.microsoft.com/kb/910678/

Thanks for nvidia version compare.
It's not strange, it's nvidia wink

Sorry I'm not convinced yet.
A search *.inf and *.ini dosn't match PCI\VEN_10DE&DEV_044C&CC_0104 and not PCI\VEN_10DE&DEV_044D&CC_0106.
Does DP_MassStorage_wnt5_x86-32_7108X.7z support this two devices?

About Silicon image SUBSYS_*CC_010* tagging:
I'm guessing only, because I don't have this hardware.
Based on devcon output, I would remove CC_010* tagging.

Jaak wrote:

For Nvidia, I looked at the duplicate HWIDs.
The generic TXTmode driver we use has pretty much all HWIDs and is most current.

DP_MassStorage_wnt5_x86-32_7108X.7z

What about:
\12.15_nforce_520_winxp32_english_whql\IDE\WinXP\sataraid\nvrd32.inf
%NVSTOR_DESC%=Crush11_Inst,PCI\VEN_10DE&DEV_044C&CC_0104

\12.15_nforce_520_winxp32_english_whql\IDE\WinXP\sata_ide\nvgts.inf
%NVSTOR_DESC%=Crush11_Inst,PCI\VEN_10DE&DEV_044D&CC_0106


Added:
I wonder about, haven't tested this:
[S-5]
ms_1_hwids="PCI\VEN_1095&DEV_3112&SUBSYS_61121095&CC_0104

Devcon list:

devcon hwids PCI\CC_01* 

PCI\VEN_104B&DEV_1040&SUBSYS_1040104B&REV_01\3&61AAA01&0&80
Name: SCSI Controller
Hardware ID's:
PCI\VEN_104B&DEV_1040&SUBSYS_1040104B&REV_01
PCI\VEN_104B&DEV_1040&SUBSYS_1040104B
PCI\VEN_104B&DEV_1040&CC_010000
PCI\VEN_104B&DEV_1040&CC_0100
Compatible ID's:
PCI\VEN_104B&DEV_1040&REV_01
PCI\VEN_104B&DEV_1040
PCI\VEN_104B&CC_010000
PCI\VEN_104B&CC_0100
PCI\VEN_104B
PCI\CC_010000
PCI\CC_0100

Devcon dosn't list VEN_*&DEV_*&SUBSYS_*&CC_*.
Is VEN_*&DEV_*&SUBSYS_*&CC_* a valid HardwareID ?

Does different Silicon Image controller BIOS use different REV_* numbers?
If, then VEN_*&DEV_*&SUBSYS_*&REV_* maybe a unique HardwareID.

When I used the driverINfinfo spotlight tool and sort formula, all sorts of "things" came to light.
Because it looks like this is going to be our solution to many problems we had with silicon, promise, and other drivers which caused clashes between raid/non-raid variants, work will continue on the 'Xperimental'.

Thanks, driverINfinfo spotlight tool is great. This come in handy.

Unfortunalely I have limited different hardware.

L\4  updated to 01/27/2006,1.21.12.00

Nvidia, Looked at the duplicate HWIDs. Looks like it can be left as is.
The TXTmode driver we use has pretty much all HWIDs and is most current.

What about: http://www.nvidia.com/object/nforce_520 … 12.15.html
nForce 520/560 
Version:    12.15
Release Date:    July 24, 2007

@capacity

Expanding %ComSpec% fails.
Does %ComSpec% link to cmd.exe still?
Do you use some kind of 4NT?

Dichani wrote:

The Hardware ID is as follows,
PCI\VEN_1023&DEV_9850&SUBSYS_98501023&REV_F3\3&225B1D41&0&0008

Strange, XP should support this graphic card out of the box.

Does your XP CD contain files sgiu.inf, sgiulnt5.sys and sgiul50.dll?
Based on HardwareID, these three files should support your graphic card.

NT4 drivers:
http://www.downloadsource.net/869/Tride … indows-NT/

Good luck.

Added:
Which HardwareId does a Trident 3D Image 9850 use?

Does default XP sgiu.inf match?

[trident.Mfg]
Trident Video Accelerator 3D Image985     = sgiul50, PCI\VEN_1023&DEV_9850
chud wrote:

I124 Doing copy-only install of "PCI\VEN_8086&DEV_2572&SUBSYS_01511028&REV_02\3&172E68DD&0&10".
#-011 Installing section [iegd_alm] from "c:\windows\inf\oem21.inf".
#E358 An unsigned or incorrectly signed file "c:\windows\inf\oem21.inf" for driver "Intel Corporation 865 Embedded Graphics Controller" blocked (server install). Error 1168: Element not found.

Cat file is missing in Intel Embedded drivers. Intel didn't include Embedded cat file.

R79733.EXE does not contain embedded drivers.

Try last test package, but remove Intel Embedded drivers.
Does graphics drivers load at Dell GX260,GX270 and GX280?


Added:

#I022 Found "PCI\VEN_8086&DEV_2562&SUBSYS_01261028" in c:\D\G\I\12\ialmnt5.inf; Device: "Intel(R) 82845G/GL/GE/PE/GV Graphics Controller"; Driver: "Intel(R) 82845G/GL/GE/PE/GV Graphics Controller"; Provider: "Intel Corporation"; Mfg: "Intel Corporation"; Section name: "i845G".
#I087 Driver node not trusted, rank changed from 0x00000001 to 0x0000c001.

Idea:
Dell edited file ialmnt5.inf, SUBSYS was added: PCI\VEN_8086&DEV_*&SUBSYS_*
However there is a old Intel default *.cat tile included. This cat file is invalid to the edited Dell ialmnt5.inf.
Therefore rank 0xc001.
However according to *.cat file date, Dell may have added a Dell *.cat file too.
How to recognice a *.cat distributer.

Actually I'm wondering: why are Intel embedded drives included?

As known there are two driver sets:
-Intel desktop and laptop graphics drivers
-Intel embedded graphics drivers
Both driver set use the same HardwareID.

Which driver does most user use?

What about deleting embedded graphics drivers?

Or break signed, give PNP a low priority

[Version] 
Signature="$WINDOWS NT$" 
Class=Display
ClassGUID={4D36E968-E325-11CE-BFC1-08002BE10318} 
Provider=%Intel% 
;CatalogFile=iegd.cat
;DriverVer = 06/04/2007,7.0.0.1004

225

(11 replies, posted in DriverPack Mass Storage)

Promise FastTrak TX2200/TX2300 use different driver version for a BIOS version too.
http://www.promise.com/support/download … &go=GO

I don't own any previous mentioned hardware and can't test myself.

The big question:
Can a false driver version destroy data? Destroy a existing RAID?
Can anybody try this condition at real hardware?

What about: don't include any above drivers to DriverPack by default?
A customer has to include appropiate driver always.