Hi all,
I’ve recently installed SoftICE (Driver studio), then Syser on a Windows 7 machine.
In both cases, the result was that I got a blue screen on every Windows startup and had to resort to system restore.
Has anyone managed to run any kernel debugger (aside from WinDbg/KD) on Windows 7?
Thanks,
–Jonathan
SoftICE has been unsupported for a long time before Windows 7, and for some
of us was broken even on versions that worked. Why do you want another
debugger that WinDBG?
–
Don Burn (MVP, Windows DKD)
Windows Filesystem and Driver Consulting
Website: http://www.windrvr.com
Blog: http://msmvps.com/blogs/WinDrvr
wrote in message news:xxxxx@ntdev…
> Hi all,
>
> I’ve recently installed SoftICE (Driver studio), then Syser on a Windows 7
> machine.
> In both cases, the result was that I got a blue screen on every Windows
> startup and had to resort to system restore.
>
> Has anyone managed to run any kernel debugger (aside from WinDbg/KD) on
> Windows 7?
>
> Thanks,
> --Jonathan
>
>
> Information from ESET NOD32 Antivirus, version of virus
> signature database 4409 (20090909)
>
> The message was checked by ESET NOD32 Antivirus.
>
> http://www.eset.com
>
>
>
Information from ESET NOD32 Antivirus, version of virus signature database 4409 (20090909)
The message was checked by ESET NOD32 Antivirus.
http://www.eset.com
Softice: dead product doesn’t work
Syser: I have no idea, perhaps you should contact them, they seem to
be selling it.
Windbg: free supported works, requires two systems for breakpoint debugging.
Mark Roddy
On Wed, Sep 9, 2009 at 6:27 AM, wrote:
> Hi all,
>
> I’ve recently installed SoftICE (Driver studio), then Syser on a Windows 7 machine.
> In both cases, the result was that I got a blue screen on every Windows startup and had to resort to system restore.
>
> Has anyone managed to run any kernel debugger (aside from WinDbg/KD) on Windows 7?
>
> Thanks,
> --Jonathan
>
> —
> NTDEV is sponsored by OSR
>
> For our schedule of WDF, WDM, debugging and other seminars visit:
> http://www.osr.com/seminars
>
> To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer
>
The reason I don’t want to use WinDbg is because I don’t want to have to use 2 systems.
Having the debugger on the tested machine is easier to setup and use.
On Wed, Sep 9, 2009 at 7:50 AM, wrote:
> The reason I don’t want to use WinDbg is because I don’t want to have to use 2 systems.
> Having the debugger on the tested machine is easier to setup and use.
>
Good luck getting this easier to set up and use debugging
configuration to work. I’d try support at syser myself, of course you
probably have to first pay them for their product.
Mark Roddy
Doing all your work on one system is just asking for disaster. One
screw up in a driver could cause disk corruption. That is why when
we teach at OSR we ALWAYS tell people to use 2 systems.
–Mark Cariddi
OSR, Open Systems Resource, Inc.
-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of
xxxxx@n-trig.com
Sent: Wednesday, September 09, 2009 7:51 AM
To: Windows System Software Devs Interest List
Subject: RE:[ntdev] SoftICE/Syser/Other on Windows 7?
The reason I don’t want to use WinDbg is because I don’t want to have to
use 2 systems.
Having the debugger on the tested machine is easier to setup and use.
NTDEV is sponsored by OSR
For our schedule of WDF, WDM, debugging and other seminars visit:
http://www.osr.com/seminars
To unsubscribe, visit the List Server section of OSR Online at
http://www.osronline.com/page.cfm?name=ListServer
Take the advice of a non expert (but I have written drivers successfuly): I
too was annoyed about needing two systems. When I finally tried it it turns
out to be easy to set up, not that hard to learn to use, and you also get a
feeling of power at being able to “control” windows on the debug machine.
Then you realise that since you’re debugging kernel code, having a
development machine that doesn’t crash even though the driver under
development crashes is great, it is a fast loop to fix bugs and test. Windbg
is quite a steep learning curve, but so must be every such tool. Since
everyone else is using it this forum is full of helpful people and advice. A
second machine doesn’t cost much. A cable is pretty cheap. I have used COM
ports with no problem, and also firewire which is faster (but not so
practical if you’re actually writing a firewire driver). It’s easy to swap
to different test machines or reboot one into different OSes. And it is real
hardware, unlike any virtual machine. Bite the bullet… M
The reason I don’t want to use WinDbg is because I don’t want to have to
use 2 systems.
Having the debugger on the tested machine is easier to setup and use. <
I don’t know about Syser, but the last version of Windows support by SI (as far as I know) was XP (maybe 2K3; not sure), so I seriously doubt that it will work on Win7.
I used to be a SI user, and when I switched to WinDbg, I really, really hated it at first; however, over time, I’ve found it to be a MUCH better debugger. Granted, the docs are atrocious in some places and it’s a lot more complicated all around, but it can so much more.
I agree with everything said, but if you want to do the one machine thing and you’re not developing something for actual hardware, you might want to consider using VMWare Workstation.
If you do go the WinDbg route, I would use 1394 instead of serial as it’s much faster.
Good luck,
mm
Really? Are you sure? How much experience do you have with Windows driver
development? I’ve been developing low level code using in circuit emulation,
logic analyzers, single system debuggers, and target/host combinations for
multiple decades. I’ve even used SoftIce a time or two, and by far prefer
WinDbg.
Sure, at first glance SI is easier, but it isn’t. SI is stuck in Windows XP,
SP1, if you are lucky, and will not work on Vista or Win7 as you have found
out. And even if you can cobble it up to work, you either will not have
symbols or will have hell of a lot of work to do to get symbols. The first
time you trash your boot disk because the driver under debug decides to
crash during a critical phase of disk IO, you will understand. If you were
smart, and during the young and arrogant phase of development we generally
are not smart, you have a handy dandy image available that you can use to
re-image your target system. If you don’t, and remember the rule of young
and arrogant, you most likely will spend a significant part of a morning or
afternoon rebuilding the target. To only do it all over again when the boot
disk gets hosed, again, during another crash. Now, imagine that is your
development system crashing on you. Count at least half a day or more for
reinstalling not only the OS but all the software you use on a daily basis.
Maybe even multiple days if you have to go to your IT department to get the
disk re-imaged.
Using WinDbg, mapped hard drives, and .kdfiles is possibly the simplest,
most efficient, and most productive, debug environment I have found.
The personal opinion of
Gary G. Little
-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of xxxxx@n-trig.com
Sent: Wednesday, September 09, 2009 6:51 AM
To: Windows System Software Devs Interest List
Subject: RE:[ntdev] SoftICE/Syser/Other on Windows 7?
The reason I don’t want to use WinDbg is because I don’t want to have to use
2 systems.
Having the debugger on the tested machine is easier to setup and use.
NTDEV is sponsored by OSR
For our schedule of WDF, WDM, debugging and other seminars visit:
http://www.osr.com/seminars
To unsubscribe, visit the List Server section of OSR Online at
http://www.osronline.com/page.cfm?name=ListServer
__________ Information from ESET Smart Security, version of virus signature
database 4409 (20090909) __________
The message was checked by ESET Smart Security.
__________ Information from ESET Smart Security, version of virus signature
database 4410 (20090909) __________
The message was checked by ESET Smart Security.
> -----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of Mark Cariddi
Sent: Wednesday, September 09, 2009 2:09 PM
To: Windows System Software Devs Interest List
Subject: RE: [ntdev] SoftICE/Syser/Other on Windows 7?Doing all your work on one system is just asking for disaster. One
screw up in a driver could cause disk corruption. That is why when
we teach at OSR we ALWAYS tell people to use 2 systems.
Aaaaah, that’s a myth. Unless you’re writing FS and disk related
drivers. If one is careful enough, the danger is small. I’m doing it for
13 years now (at NT, did it always in DOS and WinDOS) and the worst
problem I had (once) was few zeroed files when I forgot to flush disk
caches before reloading driver which caused BSOD. Careful enough means:
Well, when I was writting FS filter, I used Ghost on daily base to
backup my drive to another one. And tested the filter on dedicated
partition only. Still never needed this backup.
I concede one machine debugging isn’t for everybody but for experienced
(not necessarily in NT drivers) and careful developers.
I even believe one machine debugging leads to better drivers because if
one runs driver under development on main machine, he has very good
motivation to write quality code and to fix every problem soon 
Best regards,
Michal Vodicka
UPEK, Inc.
[xxxxx@upek.com, http://www.upek.com]
There are a lot more cases than your simple two Michal. I have encountered
corruptions on many efforts besides this. Besides which the other reason
for two is that it reduces the heisen-bugs, i.e. the presence of the
debugger disturbs the bug. This is always present to some degree, but with
SI it was huge.
–
Don Burn (MVP, Windows DKD)
Windows Filesystem and Driver Consulting
Website: http://www.windrvr.com
Blog: http://msmvps.com/blogs/WinDrvr
“Michal Vodicka” wrote in message
news:xxxxx@ntdev…
> -----Original Message-----
> From: xxxxx@lists.osr.com
> [mailto:xxxxx@lists.osr.com] On Behalf Of Mark Cariddi
> Sent: Wednesday, September 09, 2009 2:09 PM
> To: Windows System Software Devs Interest List
> Subject: RE: [ntdev] SoftICE/Syser/Other on Windows 7?
>
> Doing all your work on one system is just asking for disaster. One
> screw up in a driver could cause disk corruption. That is why when
> we teach at OSR we ALWAYS tell people to use 2 systems.
Aaaaah, that’s a myth. Unless you’re writing FS and disk related
drivers. If one is careful enough, the danger is small. I’m doing it for
13 years now (at NT, did it always in DOS and WinDOS) and the worst
problem I had (once) was few zeroed files when I forgot to flush disk
caches before reloading driver which caused BSOD. Careful enough means:
- use NTFS which is more resistent against FS corruption
- flush disk caches always before reloading driver under development
- use PC-lint or something similar to catch stupid bugs before execution
- backup often
Well, when I was writting FS filter, I used Ghost on daily base to
backup my drive to another one. And tested the filter on dedicated
partition only. Still never needed this backup.
I concede one machine debugging isn’t for everybody but for experienced
(not necessarily in NT drivers) and careful developers.
I even believe one machine debugging leads to better drivers because if
one runs driver under development on main machine, he has very good
motivation to write quality code and to fix every problem soon
Best regards,
Michal Vodicka
UPEK, Inc.
[xxxxx@upek.com, http://www.upek.com]
Information from ESET NOD32 Antivirus, version of virus signature
database 4410 (20090909)
The message was checked by ESET NOD32 Antivirus.
http://www.eset.com
Information from ESET NOD32 Antivirus, version of virus signature database 4410 (20090909)
The message was checked by ESET NOD32 Antivirus.
http://www.eset.com
Well, I’d say generally it depends on the kind of driver you write. For
virtual, network, simple WDM and WDF driver I don’t see big danger. I
can imagine corruptions there but I can also imagine meteorite hitting
my head when I go on the street It is matter of probability and
carefullness.
Debugger changes environment and timing for sure. It can cover some
problems but can also uncover some missed on clean system. I saw such
cases. Don’t know about you but we test drivers on many computers
without debugger present so bugs masked by debugger are found. Saying
this I have to mention I don’t use any debugger since SI died but traces
are another kind of one machine debugging 
We also shouldn’t forget about UMDF which is like an application.
Anyway, it can still cause BSOD because of bugs in OS code (and does).
Best regards,
Michal Vodicka
UPEK, Inc.
[xxxxx@upek.com, http://www.upek.com]
-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of Don Burn
Sent: Wednesday, September 09, 2009 4:54 PM
To: Windows System Software Devs Interest List
Subject: Re:[ntdev] SoftICE/Syser/Other on Windows 7?There are a lot more cases than your simple two Michal. I
have encountered
corruptions on many efforts besides this. Besides which the
other reason
for two is that it reduces the heisen-bugs, i.e. the presence of the
debugger disturbs the bug. This is always present to some
degree, but with
SI it was huge.–
Don Burn (MVP, Windows DKD)
Windows Filesystem and Driver Consulting
Website: http://www.windrvr.com
Blog: http://msmvps.com/blogs/WinDrvr“Michal Vodicka” wrote in message
> news:xxxxx@ntdev…
> > -----Original Message-----
> > From: xxxxx@lists.osr.com
> > [mailto:xxxxx@lists.osr.com] On Behalf Of Mark Cariddi
> > Sent: Wednesday, September 09, 2009 2:09 PM
> > To: Windows System Software Devs Interest List
> > Subject: RE: [ntdev] SoftICE/Syser/Other on Windows 7?
> >
> > Doing all your work on one system is just asking for disaster. One
> > screw up in a driver could cause disk corruption. That
> is why when
> > we teach at OSR we ALWAYS tell people to use 2 systems.
>
> Aaaaah, that’s a myth. Unless you’re writing FS and disk related
> drivers. If one is careful enough, the danger is small. I’m
> doing it for
> 13 years now (at NT, did it always in DOS and WinDOS) and the worst
> problem I had (once) was few zeroed files when I forgot to flush disk
> caches before reloading driver which caused BSOD. Careful
> enough means:
>
> - use NTFS which is more resistent against FS corruption
> - flush disk caches always before reloading driver under development
> - use PC-lint or something similar to catch stupid bugs
> before execution
> - backup often
>
> Well, when I was writting FS filter, I used Ghost on daily base to
> backup my drive to another one. And tested the filter on dedicated
> partition only. Still never needed this backup.
>
> I concede one machine debugging isn’t for everybody but for
> experienced
> (not necessarily in NT drivers) and careful developers.
>
> I even believe one machine debugging leads to better drivers
> because if
> one runs driver under development on main machine, he has very good
> motivation to write quality code and to fix every problem soon
>
> Best regards,
>
> Michal Vodicka
> UPEK, Inc.
> [xxxxx@upek.com, http://www.upek.com]
>
>
>
> Information from ESET NOD32 Antivirus, version of
> virus signature
> database 4410 (20090909)
>
> The message was checked by ESET NOD32 Antivirus.
>
> http://www.eset.com
>
>
>
>
>
> Information from ESET NOD32 Antivirus, version of
> virus signature database 4410 (20090909)
>
> The message was checked by ESET NOD32 Antivirus.
>
> http://www.eset.com
>
>
>
>
>
> —
> NTDEV is sponsored by OSR
>
> For our schedule of WDF, WDM, debugging and other seminars visit:
> http://www.osr.com/seminars
>
> To unsubscribe, visit the List Server section of OSR Online
> at http://www.osronline.com/page.cfm?name=ListServer
>
Call me a careless dev (been called worse!), but I’m not running my
encryption filter on my workstation until I’m *really* sure that it’s
working properly
(And am I the only whose workstation doubles as a music player all day?
Having to start Slacker again every time I reboot would be the worst part.)
-scott
–
Scott Noone
Consulting Associate
OSR Open Systems Resources, Inc.
http://www.osronline.com
“Don Burn” wrote in message news:xxxxx@ntdev…
> There are a lot more cases than your simple two Michal. I have
> encountered corruptions on many efforts besides this. Besides which the
> other reason for two is that it reduces the heisen-bugs, i.e. the presence
> of the debugger disturbs the bug. This is always present to some degree,
> but with SI it was huge.
>
>
> –
> Don Burn (MVP, Windows DKD)
> Windows Filesystem and Driver Consulting
> Website: http://www.windrvr.com
> Blog: http://msmvps.com/blogs/WinDrvr
>
>
> “Michal Vodicka” wrote in message
> news:xxxxx@ntdev…
>> -----Original Message-----
>> From: xxxxx@lists.osr.com
>> [mailto:xxxxx@lists.osr.com] On Behalf Of Mark Cariddi
>> Sent: Wednesday, September 09, 2009 2:09 PM
>> To: Windows System Software Devs Interest List
>> Subject: RE: [ntdev] SoftICE/Syser/Other on Windows 7?
>>
>> Doing all your work on one system is just asking for disaster. One
>> screw up in a driver could cause disk corruption. That is why when
>> we teach at OSR we ALWAYS tell people to use 2 systems.
>
> Aaaaah, that’s a myth. Unless you’re writing FS and disk related
> drivers. If one is careful enough, the danger is small. I’m doing it for
> 13 years now (at NT, did it always in DOS and WinDOS) and the worst
> problem I had (once) was few zeroed files when I forgot to flush disk
> caches before reloading driver which caused BSOD. Careful enough means:
>
> - use NTFS which is more resistent against FS corruption
> - flush disk caches always before reloading driver under development
> - use PC-lint or something similar to catch stupid bugs before execution
> - backup often
>
> Well, when I was writting FS filter, I used Ghost on daily base to
> backup my drive to another one. And tested the filter on dedicated
> partition only. Still never needed this backup.
>
> I concede one machine debugging isn’t for everybody but for experienced
> (not necessarily in NT drivers) and careful developers.
>
> I even believe one machine debugging leads to better drivers because if
> one runs driver under development on main machine, he has very good
> motivation to write quality code and to fix every problem soon
>
> Best regards,
>
> Michal Vodicka
> UPEK, Inc.
> [xxxxx@upek.com, http://www.upek.com]
>
>
>
> Information from ESET NOD32 Antivirus, version of virus
> signature database 4410 (20090909)
>
> The message was checked by ESET NOD32 Antivirus.
>
> http://www.eset.com
>
>
>
>
>
> Information from ESET NOD32 Antivirus, version of virus
> signature database 4410 (20090909)
>
> The message was checked by ESET NOD32 Antivirus.
>
> http://www.eset.com
>
>
>
>
>
Most drivers don’t need reboot to reload. And encryption filter is a
good candidate for debugging inside VM
Best regards,
Michal Vodicka
UPEK, Inc.
[xxxxx@upek.com, http://www.upek.com]
-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of Scott Noone
Sent: Wednesday, September 09, 2009 5:14 PM
To: Windows System Software Devs Interest List
Subject: Re:[ntdev] SoftICE/Syser/Other on Windows 7?Call me a careless dev (been called worse!), but I’m not running my
encryption filter on my workstation until I’m *really* sure that it’s
working properly(And am I the only whose workstation doubles as a music
player all day?
Having to start Slacker again every time I reboot would be
the worst part.)-scott
–
Scott Noone
Consulting Associate
OSR Open Systems Resources, Inc.
http://www.osronline.com“Don Burn” wrote in message news:xxxxx@ntdev…
> > There are a lot more cases than your simple two Michal. I have
> > encountered corruptions on many efforts besides this.
> Besides which the
> > other reason for two is that it reduces the heisen-bugs,
> i.e. the presence
> > of the debugger disturbs the bug. This is always present
> to some degree,
> > but with SI it was huge.
> >
> >
> > –
> > Don Burn (MVP, Windows DKD)
> > Windows Filesystem and Driver Consulting
> > Website: http://www.windrvr.com
> > Blog: http://msmvps.com/blogs/WinDrvr
> >
> >
> > “Michal Vodicka” wrote in message
> > news:xxxxx@ntdev…
> >> -----Original Message-----
> >> From: xxxxx@lists.osr.com
> >> [mailto:xxxxx@lists.osr.com] On Behalf Of
> Mark Cariddi
> >> Sent: Wednesday, September 09, 2009 2:09 PM
> >> To: Windows System Software Devs Interest List
> >> Subject: RE: [ntdev] SoftICE/Syser/Other on Windows 7?
> >>
> >> Doing all your work on one system is just asking for
> disaster. One
> >> screw up in a driver could cause disk corruption. That
> is why when
> >> we teach at OSR we ALWAYS tell people to use 2 systems.
> >
> > Aaaaah, that’s a myth. Unless you’re writing FS and disk related
> > drivers. If one is careful enough, the danger is small. I’m
> doing it for
> > 13 years now (at NT, did it always in DOS and WinDOS) and the worst
> > problem I had (once) was few zeroed files when I forgot to
> flush disk
> > caches before reloading driver which caused BSOD. Careful
> enough means:
> >
> > - use NTFS which is more resistent against FS corruption
> > - flush disk caches always before reloading driver under development
> > - use PC-lint or something similar to catch stupid bugs
> before execution
> > - backup often
> >
> > Well, when I was writting FS filter, I used Ghost on daily base to
> > backup my drive to another one. And tested the filter on dedicated
> > partition only. Still never needed this backup.
> >
> > I concede one machine debugging isn’t for everybody but for
> experienced
> > (not necessarily in NT drivers) and careful developers.
> >
> > I even believe one machine debugging leads to better
> drivers because if
> > one runs driver under development on main machine, he has very good
> > motivation to write quality code and to fix every problem soon
> >
> > Best regards,
> >
> > Michal Vodicka
> > UPEK, Inc.
> > [xxxxx@upek.com, http://www.upek.com]
> >
> >
> >
> > Information from ESET NOD32 Antivirus, version of virus
> > signature database 4410 (20090909)
> >
> > The message was checked by ESET NOD32 Antivirus.
> >
> > http://www.eset.com
> >
> >
> >
> >
> >
> > Information from ESET NOD32 Antivirus, version of virus
> > signature database 4410 (20090909)
> >
> > The message was checked by ESET NOD32 Antivirus.
> >
> > http://www.eset.com
> >
> >
> >
> >
> >
>
>
>
> —
> NTDEV is sponsored by OSR
>
> For our schedule of WDF, WDM, debugging and other seminars visit:
> http://www.osr.com/seminars
>
> To unsubscribe, visit the List Server section of OSR Online
> at http://www.osronline.com/page.cfm?name=ListServer
>
My SI days are definitely over, and I much prefer windbg at this point, but like Michal, I used it for years without a single instance of corruption, though in my case, it was almost 100% for software only stuff.
I don’t think that Michal was suggesting that everyone use it for everything. He clearly specified the conditions under which he used it successfully, and the thing is that this guy apparently wants to use it.
What did you encounter?
mm
Mike Kemp wrote:
A second machine doesn’t cost much.
I have a Pentium III laptop here with Windows 2000 that makes a
perfectly usable windbg host. I keep it around because it still has a
serial port…
–
Tim Roberts, xxxxx@probo.com
Providenza & Boekelheide, Inc.
You can get a low end system to act as the debugger host for on the order of $200 or less nowadays, I’d go that route and save yourself the grief.
-----Original Message-----
From: xxxxx@n-trig.com
Sent: Wednesday, September 09, 2009 04:50
To: Windows System Software Devs Interest List
Subject: RE:[ntdev] SoftICE/Syser/Other on Windows 7?
The reason I don’t want to use WinDbg is because I don’t want to have to use 2 systems.
Having the debugger on the tested machine is easier to setup and use.
—
NTDEV is sponsored by OSR
For our schedule of WDF, WDM, debugging and other seminars visit:
http://www.osr.com/seminars
To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer
> The reason I don’t want to use WinDbg is because I don’t want to have to use 2 systems.
MS Virtual PC is another free solution, or, for Windows Server - Hyper-V.
–
Maxim S. Shatskih
Windows DDK MVP
xxxxx@storagecraft.com
http://www.storagecraft.com
>We also shouldn’t forget about UMDF which is like an application.
Anyway, it can still cause BSOD because of bugs in OS code (and does).
Can you share some details on this with us?
This can be really interesting.
–
Maxim S. Shatskih
Windows DDK MVP
xxxxx@storagecraft.com
http://www.storagecraft.com
> -----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of Maxim
S. Shatskih
Sent: Wednesday, September 09, 2009 11:27 PM
To: Windows System Software Devs Interest List
Subject: Re:[ntdev] SoftICE/Syser/Other on Windows 7?>We also shouldn’t forget about UMDF which is like an application.
>Anyway, it can still cause BSOD because of bugs in OS code
(and does).Can you share some details on this with us?
This can be really interesting.
Well, hotfix should be “under construction” now so I’d wait with details
until available (hopefully several weeks).
It is like any other OS bug which can be invoked from user mode. It
can’t cause BSOD by definition and if it does, it is kernel code bug.
One of the best advantages of UMDF drivers – all BSODs are MS’s
There are also other ways. If you configure UMDF process host timeout to
be long enough and block power IRP processing for more than 10 minutes,
timeout expires in kernel and invokes BSOD (0x9F). Maybe also other
IRPs. This timeout and BSOD is one of the greatest improvements
introduced in Vista. I believe any user is completely happy if detaches
an unimportant USB device and after 10 minutes gets sudden BSOD which
destroys all his unsaved work
Best regards,
Michal Vodicka
UPEK, Inc.
[xxxxx@upek.com, http://www.upek.com]