Two and a half weeks has past, the problem about 1394/sbp2port i encountered has been solved today. About the problem i encountered , you can see the following link:
http://www.osronline.com/showthread.cfm?link=112158
http://www.osronline.com/showthread.cfm?link=112751
It was solved at the moment i was trying deassemblying the lower driver(microsoft supplied)to find what on earth happened, and the most important thing is that the problem was solved by our customer who don’t know much about driver developing in windows platform, again the way to solve the problem is only installing the update in the following link:
http://support.microsoft.com/kb/929762/en-us
I can’t relax myself for that thing, for the process of solving the problem. In fact, i have searched the update in the following link earlier, but i mass it. The reason is simple: the update has nothing to do with my problem. My problem is the irp i allocated sometimes never be completed by the lower driver, while the update solve this kinds of problem:You receive a Stop 0x9F error when you wake a Windows Vista-based computer that is connected to an IEEE 1394-based device. I didn’t encounter the stop eeror and i never operate like that----wake vista based computer, what i did was sometimes plug the storage device when PC running, and my driver block the system, for the completion routine for the irp passed to the lower driver in my driver never got invoked. Of course, DDK will never tell you that why your driver gets STATUS_PENDING from iocalldriver() while your completion routine for the irp never get invoked, it assume that it shouldn’t and never happen. Unfortunately, the problem was encountered by me, and the lower driver who return STATUS_PENDING is microsoft supplied. Did i crash the memory of the irp which then cause the problem? I check the code, no result and the system never crash. Did i shouldn’t allocate the irp in the workitem routine? Of course, that seems ridiculous, but i tried not allocating the irp in the workitem, no effect, the problem is the same, then can i cancel the irp? I abandon the way at last, for the cancelroutine pointer in the irp never been filled by the lower. Can i report the situation to microsoft as a developer? In microsoft opinion, the best way to report a bug is to do it by internet when your application or driver has been dead abnormally, all other ways are not acceptable. Before the problem got solved, i almost sure it was not my problem, at least, according to WDK. Is it microsotf’s problem? I guess it is, but not sure. This problem can’t be report as a bug, for i couldn’t represent it every time, and of course i couldn’t send the device to microsoft to let them represent the problem. I tried, tried to find every possible forum or blog, espacially microsoft’s, no one seems to have experienced this. At last, i hate to tell my boss that i need to ask microsoft to solve the problem, you need to pay for it. All the things above frustrate me very much.
I was confused that in one aspect, microsoft let the driver developers all over the world free download the latest WDK to develop more device drivers for vista, and spent lots of money on advertisement all over the world for vista, while in another aspect, the real driver developer was so hard to send the feedback of the product to microsoft efficiently.
I forwarded this email to the 1394/sbp2port developers to see if they
have seen this issue before. Hopefully we can see if this is a known or
not.
d
-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of activism@163.com
Sent: Wednesday, July 11, 2007 7:08 AM
To: Windows System Software Devs Interest List
Subject: [ntdev] Cheers or be cool?
Two and a half weeks has past, the problem about 1394/sbp2port i
encountered has been solved today. About the problem i encountered , you
can see the following link:
http://www.osronline.com/showthread.cfm?link=112158
http://www.osronline.com/showthread.cfm?link=112751
It was solved at the moment i was trying deassemblying the lower
driver(microsoft supplied)to find what on earth happened, and the most
important thing is that the problem was solved by our customer who don’t
know much about driver developing in windows platform, again the way to
solve the problem is only installing the update in the following link:
http://support.microsoft.com/kb/929762/en-us
I can’t relax myself for that thing, for the process of solving the
problem. In fact, i have searched the update in the following link
earlier, but i mass it. The reason is simple: the update has nothing to
do with my problem. My problem is the irp i allocated sometimes never be
completed by the lower driver, while the update solve this kinds of
problem:You receive a Stop 0x9F error when you wake a Windows
Vista-based computer that is connected to an IEEE 1394-based device. I
didn’t encounter the stop eeror and i never operate like that----wake
vista based computer, what i did was sometimes plug the storage device
when PC running, and my driver block the system, for the completion
routine for the irp passed to the lower driver in my driver never got
invoked. Of course, DDK will never tell you that why your driver gets
STATUS_PENDING from iocalldriver() while your completion routine for the
irp never get invoked, it assume that it shouldn’t and never happen.
Unfortunately, the problem was encountered by me, and the lower driver
who return STATUS_PENDING is microsoft supplied. Did i crash the memory
of the irp which then cause the problem? I check the code, no result and
the system never crash. Did i shouldn’t allocate the irp in the workitem
routine? Of course, that seems ridiculous, but i tried not allocating
the irp in the workitem, no effect, the problem is the same, then can i
cancel the irp? I abandon the way at last, for the cancelroutine pointer
in the irp never been filled by the lower. Can i report the situation to
microsoft as a developer? In microsoft opinion, the best way to report a
bug is to do it by internet when your application or driver has been
dead abnormally, all other ways are not acceptable. Before the problem
got solved, i almost sure it was not my problem, at least, according to
WDK. Is it microsotf’s problem? I guess it is, but not sure. This
problem can’t be report as a bug, for i couldn’t represent it every
time, and of course i couldn’t send the device to microsoft to let them
represent the problem. I tried, tried to find every possible forum or
blog, espacially microsoft’s, no one seems to have experienced this. At
last, i hate to tell my boss that i need to ask microsoft to solve the
problem, you need to pay for it. All the things above frustrate me very
much.
I was confused that in one aspect, microsoft let the driver developers
all over the world free download the latest WDK to develop more device
drivers for vista, and spent lots of money on advertisement all over the
world for vista, while in another aspect, the real driver developer was
so hard to send the feedback of the product to microsoft efficiently.
Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256
To unsubscribe, visit the List Server section of OSR Online at
http://www.osronline.com/page.cfm?name=ListServer
Doron Holan:
->I forwarded this email to the 1394/sbp2port developers to see if they
->have seen this issue before. Hopefully we can see if this is a known or
->not.
Greatly appreciate. So far, i don’t know why the problem disappears after the update. I really want to know why the problem occured, and why it disappeared after the update(i forgot to tell that the CurrentStackLocation pointer in the stuck irp pointed to sbp2port stacklocation).
wrote in message news:xxxxx@ntdev…
> Doron Holan:
> ->I forwarded this email to the 1394/sbp2port developers to see if they
> ->have seen this issue before. Hopefully we can see if this is a known or
> ->not.
> Greatly appreciate. So far, i don’t know why the problem disappears after
> the update. I really want to know why the problem occured, and why it
> disappeared after the update(i forgot to tell that the
> CurrentStackLocation pointer in the stuck irp pointed to sbp2port
> stacklocation).
>
The reason the problem disappears after the update is because the update
provides a new driver for sbp2port.sys. There was a similar sounding
problem with earlier versions of Windows that was fixed in a May 17, 2007 in
KB311430. Its likely that the reason this driver malfunctioned on Vista is
the same reason as that update for earlier versions of Windows. That also
suggests that your driver may not work properly on earlier versions of
Windows if they are not updated either. Here’s a link:
http://support.microsoft.com/kb/311430
SYMPTOMS
A 1394 storage device with more than one Logical Unit Number (LUN) may stop
functioning if another 1394 device is either inserted or removed from the
computer.
CAUSE
When a 1394 bus reset occurs after a device is inserted or removed, the 1394
mass-storage driver (Sbp2port.sys) must reconnect to each LUN on a 1394
mass-storage device. 1394 mass-storage devices can only handle one reconnect
request at a time. The Sbp2port.sys driver often submits more than one
request a time, one for each LUN on the device. The mass-storage device
ignores the multiple reconnect requests, and only reconnects the first LUN.
Because of this, other LUNs on the device are not reconnected, and become
non-functional.
The key here seems to be the part about how the 1394 interface “can only
handle one request at a time”, but sometimes the faulty spb2port.sys tried
to send more than one request at the same time. Oops! Where it says that
the “mass-storage device ignores the multiple reconnect requests”, probably
what happens is spb2port.sys was pending those IRPs or SRBs and sending the
requests off to the 1394 hardware. The hardware only responded to one so it
got its request completed and it didn’t ever respond to the other so that
IRP just got stuck. I know nothing about 1394 reconnection requests, so I’m
not sure if those connection requests come through your driver and you might
have a possibility of forcing them to be performed synchronously to work
around this bug on machines that don’t have the update, or if those
connection requests originate at sbp2port.sys itself where there is no
solution except to install the update.
Matthew Carter:
Thanks a lot. There are three point i may need to classify when the problem was encountered:
1 Only use one storage device plug and unplug and when plug the device, the irp was stuck some times. In fact, the stuck irp was the second IRP that the device received from my driver, as i described in the following link:
http://www.osronline.com/showthread.cfm?link=112158
when the IRP_MN_START_DEVICE successfully return from the lower, my driver don’t do other thing when the lower returned successfully, just send a command to the device to tell the device that my driver will send other commands to it, then the irp for second command sometimes got stuck.
2 I set the timeout value in srb, and indeed the request will be completed when the time beyond in most cases(e.g, after a request was send to the device, i unpluged the device, beyond the timeout value, the request was completed). why sometimes the irp never been completed, thougth i set the timeout value(the timeout value never been overwrite, i promise every time it is right).
3 In the case that irp got stuck, the sbp2port send reset command two times, but sometimes through sbp2port send two times reset command(when plug in), the irp didn’t get stuck.
I have a similar problem with Microsoft.
I have a device that is using winusb.sys on Vista. On the Swedish version
of Vista, I get
an error when installing the device. I used an inf file that should
install the winusb.sys that already is installed on Vista for my device.
This inf works OK with the US version of Vista where it copies the driver
from the file repository, but on the Swedish version it instead tries to
copy the driver from c:/windows/inf directory, which fails, because the
file doesn’t exist.
I have not found any way to report bugs to Microsoft.
I think it is highly remarkable that Microsoft doesn’t have any decent way
to report bugs. If Microsoft is really interested in fixing bugs, why
don’t make everything possible to get bug reports? I also think that the
more information you have about a bug, the easier it is to fix it.
Microsoft should really take care of bug reports from programmers, that
mostly have very clear error descriptions.
In the present situation, where almost all computers are using Windows and
there are almost no alternative, Microsoft should really try everything to
fix the bugs.
Kurt Nyström
Two and a half weeks has past, the problem about 1394/sbp2port i
encountered has been solved today. About the problem i encountered , you
can see the following link:
http://www.osronline.com/showthread.cfm?link=112158
http://www.osronline.com/showthread.cfm?link=112751
It was solved at the moment i was trying deassemblying the lower
driver(microsoft supplied)to find what on earth happened, and the most
important thing is that the problem was solved by our customer who don’t
know much about driver developing in windows platform, again the way to
solve the problem is only installing the update in the following link:
http://support.microsoft.com/kb/929762/en-us
I can’t relax myself for that thing, for the process of solving the
problem. In fact, i have searched the update in the following link
earlier, but i mass it. The reason is simple: the update has nothing to
do with my problem. My problem is the irp i allocated sometimes never be
completed by the lower driver, while the update solve this kinds of
problem:You receive a Stop 0x9F error when you wake a Windows
Vista-based computer that is connected to an IEEE 1394-based device. I
didn’t encounter the stop eeror and i never operate like that----wake
vista based computer, what i did was sometimes plug the storage device
when PC running, and my driver block the system, for the completion
routine for the irp passed to the lower driver in my driver never got
invoked. Of course, DDK will never tell you that why your driver gets
STATUS_PENDING from iocalldriver() while your completion routine for the
irp never get invoked, it assume that it shouldn’t and never happen.
Unfortunately, the problem was encountered by me, and the lower driver
who return STATUS_PENDING is microsoft supplied. Did i crash the memory
of the irp which then cause the problem? I check the code, no result and
the system never crash. Did i shouldn’t allocate the irp in the workitem
routine? Of course, that seems ridiculous, but i tried not allocating
the irp in the workitem, no effect, the problem is the same, then can i
cancel the irp? I abandon the way at last, for the cancelroutine pointer
in the irp never been filled by the lower. Can i report the situation to
microsoft as a developer? In microsoft opinion, the best way to report a
bug is to do it by internet when your application or driver has been
dead abnormally, all other ways are not acceptable. Before the problem
got solved, i almost sure it was not my problem, at least, according to
WDK. Is it microsotf’s problem? I guess it is, but not sure. This
problem can’t be report as a bug, for i couldn’t represent it every
time, and of course i couldn’t send the device to microsoft to let them
represent the problem. I tried, tried to find every possible forum or
blog, espacially microsoft’s, no one seems to have experienced this. At
last, i hate to tell my boss that i need to ask microsoft to solve the
problem, you need to pay for it. All the things above frustrate me very
much.
I was confused that in one aspect, microsoft let the driver developers
all over the world free download the latest WDK to develop more device
drivers for vista, and spent lots of money on advertisement all over
the world for vista, while in another aspect, the real driver developer
was so hard to send the feedback of the product to microsoft
efficiently.
Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256To unsubscribe, visit the List Server section of OSR Online at
http://www.osronline.com/page.cfm?name=ListServer
Open an Microsoft support case, you can get a fix that way.
d
-----Original Message-----
From: xxxxx@lists.osr.com [mailto:xxxxx@lists.osr.com] On Behalf Of xxxxx@compevo.se
Sent: Thursday, July 12, 2007 2:42 PM
To: Windows System Software Devs Interest List
Subject: Re: [ntdev] Cheers or be cool?
I have a similar problem with Microsoft.
I have a device that is using winusb.sys on Vista. On the Swedish version
of Vista, I get
an error when installing the device. I used an inf file that should
install the winusb.sys that already is installed on Vista for my device.
This inf works OK with the US version of Vista where it copies the driver
from the file repository, but on the Swedish version it instead tries to
copy the driver from c:/windows/inf directory, which fails, because the
file doesn’t exist.
I have not found any way to report bugs to Microsoft.
I think it is highly remarkable that Microsoft doesn’t have any decent way
to report bugs. If Microsoft is really interested in fixing bugs, why
don’t make everything possible to get bug reports? I also think that the
more information you have about a bug, the easier it is to fix it.
Microsoft should really take care of bug reports from programmers, that
mostly have very clear error descriptions.
In the present situation, where almost all computers are using Windows and
there are almost no alternative, Microsoft should really try everything to
fix the bugs.
Kurt Nystr?m
Two and a half weeks has past, the problem about 1394/sbp2port i
encountered has been solved today. About the problem i encountered , you
can see the following link:
http://www.osronline.com/showthread.cfm?link=112158
http://www.osronline.com/showthread.cfm?link=112751
It was solved at the moment i was trying deassemblying the lower
driver(microsoft supplied)to find what on earth happened, and the most
important thing is that the problem was solved by our customer who don’t
know much about driver developing in windows platform, again the way to
solve the problem is only installing the update in the following link:
http://support.microsoft.com/kb/929762/en-us
I can’t relax myself for that thing, for the process of solving the
problem. In fact, i have searched the update in the following link
earlier, but i mass it. The reason is simple: the update has nothing to
do with my problem. My problem is the irp i allocated sometimes never be
completed by the lower driver, while the update solve this kinds of
problem:You receive a Stop 0x9F error when you wake a Windows
Vista-based computer that is connected to an IEEE 1394-based device. I
didn’t encounter the stop eeror and i never operate like that----wake
vista based computer, what i did was sometimes plug the storage device
when PC running, and my driver block the system, for the completion
routine for the irp passed to the lower driver in my driver never got
invoked. Of course, DDK will never tell you that why your driver gets
STATUS_PENDING from iocalldriver() while your completion routine for the
irp never get invoked, it assume that it shouldn’t and never happen.
Unfortunately, the problem was encountered by me, and the lower driver
who return STATUS_PENDING is microsoft supplied. Did i crash the memory
of the irp which then cause the problem? I check the code, no result and
the system never crash. Did i shouldn’t allocate the irp in the workitem
routine? Of course, that seems ridiculous, but i tried not allocating
the irp in the workitem, no effect, the problem is the same, then can i
cancel the irp? I abandon the way at last, for the cancelroutine pointer
in the irp never been filled by the lower. Can i report the situation to
microsoft as a developer? In microsoft opinion, the best way to report a
bug is to do it by internet when your application or driver has been
dead abnormally, all other ways are not acceptable. Before the problem
got solved, i almost sure it was not my problem, at least, according to
WDK. Is it microsotf’s problem? I guess it is, but not sure. This
problem can’t be report as a bug, for i couldn’t represent it every
time, and of course i couldn’t send the device to microsoft to let them
represent the problem. I tried, tried to find every possible forum or
blog, espacially microsoft’s, no one seems to have experienced this. At
last, i hate to tell my boss that i need to ask microsoft to solve the
problem, you need to pay for it. All the things above frustrate me very
much.
I was confused that in one aspect, microsoft let the driver developers
all over the world free download the latest WDK to develop more device
drivers for vista, and spent lots of money on advertisement all over
the world for vista, while in another aspect, the real driver developer
was so hard to send the feedback of the product to microsoft
efficiently.
Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256To unsubscribe, visit the List Server section of OSR Online at
http://www.osronline.com/page.cfm?name=ListServer
Questions? First check the Kernel Driver FAQ at http://www.osronline.com/article.cfm?id=256
To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer
Thanks Doron for answering,
What I have understood I will have to pay for a support case.
Why should I pay Microsoft for a fix of Microsoft’s software?
I have a work-around for the problem.
Kurt Nyström
Open an Microsoft support case, you can get a fix that way.
d
-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of
xxxxx@compevo.se Sent: Thursday, July 12, 2007 2:42 PM
To: Windows System Software Devs Interest List
Subject: Re: [ntdev] Cheers or be cool?I have a similar problem with Microsoft.
I have a device that is using winusb.sys on Vista. On the Swedish
version of Vista, I get
an error when installing the device. I used an inf file that should
install the winusb.sys that already is installed on Vista for my device.
This inf works OK with the US version of Vista where it copies the
driver from the file repository, but on the Swedish version it instead
tries to copy the driver from c:/windows/inf directory, which fails,
because the file doesn’t exist.I have not found any way to report bugs to Microsoft.
I think it is highly remarkable that Microsoft doesn’t have any decent
way to report bugs. If Microsoft is really interested in fixing bugs,
why don’t make everything possible to get bug reports? I also think that
the more information you have about a bug, the easier it is to fix it.
Microsoft should really take care of bug reports from programmers, that
mostly have very clear error descriptions.In the present situation, where almost all computers are using Windows
and there are almost no alternative, Microsoft should really try
everything to fix the bugs.Kurt Nyström
> Two and a half weeks has past, the problem about 1394/sbp2port i
> encountered has been solved today. About the problem i encountered ,
> you can see the following link:
> http://www.osronline.com/showthread.cfm?link=112158
> http://www.osronline.com/showthread.cfm?link=112751
> It was solved at the moment i was trying deassemblying the lower
> driver(microsoft supplied)to find what on earth happened, and the most
> important thing is that the problem was solved by our customer who
> don’t know much about driver developing in windows platform, again the
> way to solve the problem is only installing the update in the
> following link: http://support.microsoft.com/kb/929762/en-us
> I can’t relax myself for that thing, for the process of solving the
> problem. In fact, i have searched the update in the following link
> earlier, but i mass it. The reason is simple: the update has nothing
> to do with my problem. My problem is the irp i allocated sometimes
> never be completed by the lower driver, while the update solve this
> kinds of problem:You receive a Stop 0x9F error when you wake a Windows
> Vista-based computer that is connected to an IEEE 1394-based device. I
> didn’t encounter the stop eeror and i never operate like that----wake
> vista based computer, what i did was sometimes plug the storage device
> when PC running, and my driver block the system, for the completion
> routine for the irp passed to the lower driver in my driver never got
> invoked. Of course, DDK will never tell you that why your driver gets
> STATUS_PENDING from iocalldriver() while your completion routine for
> the irp never get invoked, it assume that it shouldn’t and never
> happen. Unfortunately, the problem was encountered by me, and the
> lower driver who return STATUS_PENDING is microsoft supplied. Did i
> crash the memory of the irp which then cause the problem? I check the
> code, no result and the system never crash. Did i shouldn’t allocate
> the irp in the workitem routine? Of course, that seems ridiculous, but
> i tried not allocating the irp in the workitem, no effect, the problem
> is the same, then can i cancel the irp? I abandon the way at last, for
> the cancelroutine pointer in the irp never been filled by the lower.
> Can i report the situation to microsoft as a developer? In microsoft
> opinion, the best way to report a bug is to do it by internet when
> your application or driver has been dead abnormally, all other ways
> are not acceptable. Before the problem got solved, i almost sure it
> was not my problem, at least, according to WDK. Is it microsotf’s
> problem? I guess it is, but not sure. This problem can’t be report as
> a bug, for i couldn’t represent it every time, and of course i
> couldn’t send the device to microsoft to let them represent the
> problem. I tried, tried to find every possible forum or blog,
> espacially microsoft’s, no one seems to have experienced this. At
> last, i hate to tell my boss that i need to ask microsoft to solve the
> problem, you need to pay for it. All the things above frustrate me
> very much.
> I was confused that in one aspect, microsoft let the driver
> developers
> all over the world free download the latest WDK to develop more device
> drivers for vista, and spent lots of money on advertisement all over
> the world for vista, while in another aspect, the real driver
> developer was so hard to send the feedback of the product to microsoft
> efficiently.
>
>
> —
> Questions? First check the Kernel Driver FAQ at
> http://www.osronline.com/article.cfm?id=256
>
> To unsubscribe, visit the List Server section of OSR Online at
> http://www.osronline.com/page.cfm?name=ListServer
Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256To unsubscribe, visit the List Server section of OSR Online at
http://www.osronline.com/page.cfm?name=ListServer
Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256To unsubscribe, visit the List Server section of OSR Online at
http://www.osronline.com/page.cfm?name=ListServer
If you find and report a legitimate bug, there is no cost AFAIK.
d
-----Original Message-----
From: xxxxx@lists.osr.com [mailto:xxxxx@lists.osr.com] On Behalf Of xxxxx@compevo.se
Sent: Thursday, July 12, 2007 3:54 PM
To: Windows System Software Devs Interest List
Subject: RE: [ntdev] Cheers or be cool?
Thanks Doron for answering,
What I have understood I will have to pay for a support case.
Why should I pay Microsoft for a fix of Microsoft’s software?
I have a work-around for the problem.
Kurt Nystr?m
Open an Microsoft support case, you can get a fix that way.
d
-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of
xxxxx@compevo.se Sent: Thursday, July 12, 2007 2:42 PM
To: Windows System Software Devs Interest List
Subject: Re: [ntdev] Cheers or be cool?I have a similar problem with Microsoft.
I have a device that is using winusb.sys on Vista. On the Swedish
version of Vista, I get
an error when installing the device. I used an inf file that should
install the winusb.sys that already is installed on Vista for my device.
This inf works OK with the US version of Vista where it copies the
driver from the file repository, but on the Swedish version it instead
tries to copy the driver from c:/windows/inf directory, which fails,
because the file doesn’t exist.I have not found any way to report bugs to Microsoft.
I think it is highly remarkable that Microsoft doesn’t have any decent
way to report bugs. If Microsoft is really interested in fixing bugs,
why don’t make everything possible to get bug reports? I also think that
the more information you have about a bug, the easier it is to fix it.
Microsoft should really take care of bug reports from programmers, that
mostly have very clear error descriptions.In the present situation, where almost all computers are using Windows
and there are almost no alternative, Microsoft should really try
everything to fix the bugs.Kurt Nystr?m
> Two and a half weeks has past, the problem about 1394/sbp2port i
> encountered has been solved today. About the problem i encountered ,
> you can see the following link:
> http://www.osronline.com/showthread.cfm?link=112158
> http://www.osronline.com/showthread.cfm?link=112751
> It was solved at the moment i was trying deassemblying the lower
> driver(microsoft supplied)to find what on earth happened, and the most
> important thing is that the problem was solved by our customer who
> don’t know much about driver developing in windows platform, again the
> way to solve the problem is only installing the update in the
> following link: http://support.microsoft.com/kb/929762/en-us
> I can’t relax myself for that thing, for the process of solving the
> problem. In fact, i have searched the update in the following link
> earlier, but i mass it. The reason is simple: the update has nothing
> to do with my problem. My problem is the irp i allocated sometimes
> never be completed by the lower driver, while the update solve this
> kinds of problem:You receive a Stop 0x9F error when you wake a Windows
> Vista-based computer that is connected to an IEEE 1394-based device. I
> didn’t encounter the stop eeror and i never operate like that----wake
> vista based computer, what i did was sometimes plug the storage device
> when PC running, and my driver block the system, for the completion
> routine for the irp passed to the lower driver in my driver never got
> invoked. Of course, DDK will never tell you that why your driver gets
> STATUS_PENDING from iocalldriver() while your completion routine for
> the irp never get invoked, it assume that it shouldn’t and never
> happen. Unfortunately, the problem was encountered by me, and the
> lower driver who return STATUS_PENDING is microsoft supplied. Did i
> crash the memory of the irp which then cause the problem? I check the
> code, no result and the system never crash. Did i shouldn’t allocate
> the irp in the workitem routine? Of course, that seems ridiculous, but
> i tried not allocating the irp in the workitem, no effect, the problem
> is the same, then can i cancel the irp? I abandon the way at last, for
> the cancelroutine pointer in the irp never been filled by the lower.
> Can i report the situation to microsoft as a developer? In microsoft
> opinion, the best way to report a bug is to do it by internet when
> your application or driver has been dead abnormally, all other ways
> are not acceptable. Before the problem got solved, i almost sure it
> was not my problem, at least, according to WDK. Is it microsotf’s
> problem? I guess it is, but not sure. This problem can’t be report as
> a bug, for i couldn’t represent it every time, and of course i
> couldn’t send the device to microsoft to let them represent the
> problem. I tried, tried to find every possible forum or blog,
> espacially microsoft’s, no one seems to have experienced this. At
> last, i hate to tell my boss that i need to ask microsoft to solve the
> problem, you need to pay for it. All the things above frustrate me
> very much.
> I was confused that in one aspect, microsoft let the driver
> developers
> all over the world free download the latest WDK to develop more device
> drivers for vista, and spent lots of money on advertisement all over
> the world for vista, while in another aspect, the real driver
> developer was so hard to send the feedback of the product to microsoft
> efficiently.
>
>
> —
> Questions? First check the Kernel Driver FAQ at
> http://www.osronline.com/article.cfm?id=256
>
> To unsubscribe, visit the List Server section of OSR Online at
> http://www.osronline.com/page.cfm?name=ListServer
Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256To unsubscribe, visit the List Server section of OSR Online at
http://www.osronline.com/page.cfm?name=ListServer
Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256To unsubscribe, visit the List Server section of OSR Online at
http://www.osronline.com/page.cfm?name=ListServer
Questions? First check the Kernel Driver FAQ at http://www.osronline.com/article.cfm?id=256
To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer
> ----------
From: xxxxx@lists.osr.com[SMTP:xxxxx@lists.osr.com] on behalf of xxxxx@compevo.se[SMTP:xxxxx@compevo.se]
Reply To: Windows System Software Devs Interest List
Sent: Friday, July 13, 2007 12:53 AM
To: Windows System Software Devs Interest List
Subject: RE: [ntdev] Cheers or be cool?What I have understood I will have to pay for a support case.
Why should I pay Microsoft for a fix of Microsoft’s software?
You won’t pay if MS admits a bug. However, it usually takes months and even if the bug is found, you’ll have to convince them to make a hotfix.
I have a work-around for the problem.
It is the best case. Use a workaround and in parallel open support case and problem can be fixed in the next SP.
Best regards,
Michal Vodicka
UPEK, Inc.
[xxxxx@upek.com, http://www.upek.com]
Thanks for answers,
I am the only developer in a small company. I can’t put a lot of time or
any money in anything that is really not my problem. Why should I start a
support case, pay money for it, put time in it and only get money back if
I can convince Microsoft that there is a bug?
I am just saying that there should be a way to just file bug reports.
It should really be in Microsoft’s interest to get information about bugs,
if they are interested in removing them.
I did make a new try to find any place at Microsoft’s site to report bugs
or even start a support case, but I can’t find anything suitable.
Kurt Nyström
> What I have understood I will have to pay for a support case.
> Why should I pay Microsoft for a fix of Microsoft’s software?
>
You won’t pay if MS admits a bug. However, it usually takes months and
even if the bug is found, you’ll have to convince them to make a hotfix.> I have a work-around for the problem.
>
It is the best case. Use a workaround and in parallel open support case
and problem can be fixed in the next SP.Best regards,
Michal Vodicka
UPEK, Inc.
[xxxxx@upek.com, http://www.upek.com]
xxxxx@compevo.se wrote:
I did make a new try to find any place at Microsoft’s site to report bugs
or even start a support case, but I can’t find anything suitable.
Sign up at connect.microsoft.com and use the “Feedback” function for the
“Participation” you are interested in. See my other post.
Can anyone from Microsoft comment on this, please? (Doron?)
-H
I thought someone already mentioned this, but if not you can leave bug
reports and feedback at the Microsoft Connect website for the latest version
of the WDK. You can read and vote on bug reports that other people left as
well. Now I wouldn’t necessarily expect any immediate action since people
who pay for support get precedence, but they do accept and record bug
reports. Go to the Windows Logo Kit (WLK), Windows Driver Kit (WDK) and
Windows Driver Framework (WDF) connection (sign up for it if you haven’t
already), and choose Feedback. You can leave bug reports for Windows Server
2008 Beta at Microsoft Connect as well, if you have an issue with
functionality that still exists in the latest version of Windows, but is not
driver related.
wrote in message news:xxxxx@ntdev…
> Thanks for answers,
>
> I am the only developer in a small company. I can’t put a lot of time or
> any money in anything that is really not my problem. Why should I start a
> support case, pay money for it, put time in it and only get money back if
> I can convince Microsoft that there is a bug?
>
> I am just saying that there should be a way to just file bug reports.
> It should really be in Microsoft’s interest to get information about bugs,
> if they are interested in removing them.
>
> I did make a new try to find any place at Microsoft’s site to report bugs
> or even start a support case, but I can’t find anything suitable.
> ----------
From: xxxxx@lists.osr.com[SMTP:xxxxx@lists.osr.com] on behalf of Hagen Patzke[SMTP:xxxxx@hotmail.com]
Reply To: Windows System Software Devs Interest List
Sent: Friday, July 13, 2007 12:58 PM
To: Windows System Software Devs Interest List
Subject: Re:[ntdev] Cheers or be cool?xxxxx@compevo.se wrote:
> I did make a new try to find any place at Microsoft’s site to report bugs
> or even start a support case, but I can’t find anything suitable.Sign up at connect.microsoft.com and use the “Feedback” function for the
“Participation” you are interested in. See my other post.
In my experience it is waste of time. I filed several cases during Vista beta testing and wasn’t safistied with any. Most of them were untouched for a long time and then suddenly, short before Vista release, were all closed for ridiculous reasons. Some of these bugs were confirmed and fixed later but feedback didn’t influence it; I had to use standard or not-quite-standard support channels.
I take this approach as discourteous and decided to not used Connect feedback anymore. When I spend time writing bugreport, and I believe I’m able to write good bugreports, I’d expect somebody would at least examine it. Other companies do.
Maybe others have different experience. Is there anybody who is safisfied with this way?
Best regards,
Michal Vodicka
UPEK, Inc.
[xxxxx@upek.com, http://www.upek.com]
> ----------
From: xxxxx@lists.osr.com[SMTP:xxxxx@lists.osr.com] on behalf of xxxxx@compevo.se[SMTP:xxxxx@compevo.se]
Reply To: Windows System Software Devs Interest List
Sent: Friday, July 13, 2007 11:16 AM
To: Windows System Software Devs Interest List
Subject: RE: [ntdev] Cheers or be cool?I am the only developer in a small company. I can’t put a lot of time or
any money in anything that is really not my problem.
Do you think it is different in bigger companies? 
Why should I start a
support case, pay money for it, put time in it and only get money back if
I can convince Microsoft that there is a bug?
Well, I never paid anything as I tend to report real bugs only I believe if you ever pay, you’d pay after case is closed.
I am just saying that there should be a way to just file bug reports.
It should really be in Microsoft’s interest to get information about bugs,
if they are interested in removing them.
I’d agree.
I did make a new try to find any place at Microsoft’s site to report bugs
or even start a support case, but I can’t find anything suitable.
One suitable way is to clearly describe problem and send it here. There is a good chance one of MS list participants will be interested and send it to proper people.
Best regards,
Michal Vodicka
UPEK, Inc.
[xxxxx@upek.com, http://www.upek.com]
I agree with Michal on the slow response on the non-pay support. I never bother going this route nor am I interested in having the OS bugs fixed unless the bug is severe enough to prevent my product from shipping.
I don’t usually file a paid support for OS bugs (although I knew it’s going to be free) unless my OEM customers ask me to (in order to get a case# for them to work with MSFT directly).
I found that having OEMs deal with MSFT works the best because they buy huge amount of Windows licenses every month hence MSFT’s better off listening to them. Money rocks as always.
Calvin Guan
Broadcom Corp.
Connecting Everything (r)
[/quote]
In my experience it is waste of time. I filed several cases during Vista beta
testing and wasn’t safistied with any. Most of them were untouched for a long
time and then suddenly, short before Vista release, were all closed for
ridiculous reasons. Some of these bugs were confirmed and fixed later but
feedback didn’t influence it; I had to use standard or not-quite-standard
support channels.
I take this approach as discourteous and decided to not used Connect feedback
anymore. When I spend time writing bugreport, and I believe I’m able to write
good bugreports, I’d expect somebody would at least examine it. Other companies
do.
Maybe others have different experience. Is there anybody who is safisfied with
this way?
Best regards,
Michal Vodicka
UPEK, Inc.
[xxxxx@upek.com, http://www.upek.com]
[quote]
> ----------
From: xxxxx@lists.osr.com[SMTP:xxxxx@lists.osr.com] on behalf of xxxxx@yahoo.ca[SMTP:xxxxx@yahoo.ca]
Reply To: Windows System Software Devs Interest List
Sent: Friday, July 13, 2007 6:40 PM
To: Windows System Software Devs Interest List
Subject: RE:[ntdev] Cheers or be cool?I found that having OEMs deal with MSFT works the best because they buy huge amount of Windows licenses every month hence MSFT’s better off listening to them. Money rocks as always.
I have the same experience. What normally takes weeks or months, they can have within days or even hours. Not all problems can be solved this way, though, but if possible, it is the best.
Best regards,
Michal Vodicka
UPEK, Inc.
[xxxxx@upek.com, http://www.upek.com]
Getting back to the original question, as far as Vista goes, http://support.microsoft.com/kb/936150/en-us contains the latest sbp2port.sys. Talking to the support folks, they think that this will address your issues of the hung request.
d
-----Original Message-----
From: xxxxx@lists.osr.com [mailto:xxxxx@lists.osr.com] On Behalf Of xxxxx@compevo.se
Sent: Friday, July 13, 2007 2:16 AM
To: Windows System Software Devs Interest List
Subject: RE: [ntdev] Cheers or be cool?
Thanks for answers,
I am the only developer in a small company. I can’t put a lot of time or
any money in anything that is really not my problem. Why should I start a
support case, pay money for it, put time in it and only get money back if
I can convince Microsoft that there is a bug?
I am just saying that there should be a way to just file bug reports.
It should really be in Microsoft’s interest to get information about bugs,
if they are interested in removing them.
I did make a new try to find any place at Microsoft’s site to report bugs
or even start a support case, but I can’t find anything suitable.
Kurt Nystr?m
> What I have understood I will have to pay for a support case.
> Why should I pay Microsoft for a fix of Microsoft’s software?
>
You won’t pay if MS admits a bug. However, it usually takes months and
even if the bug is found, you’ll have to convince them to make a hotfix.> I have a work-around for the problem.
>
It is the best case. Use a workaround and in parallel open support case
and problem can be fixed in the next SP.Best regards,
Michal Vodicka
UPEK, Inc.
[xxxxx@upek.com, http://www.upek.com]
Questions? First check the Kernel Driver FAQ at http://www.osronline.com/article.cfm?id=256
To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer
Doron Holan:
Thanks for your answer.
It a pity that i haven’t encountered the problem the update fixed both the two:
http://support.microsoft.com/kb/929762/en-us
http://support.microsoft.com/kb/936150/en-us
I have tested many times on vista platform, i never encountered bugchecks though i tried to wake up PC from sleep state or restart PC from hibernate, but I do encountered the irp was pending forever even I have installed the update KB929762( this update fix the problem that when plugging storage device into pc, sometimes the irp for the second command sent to device was pending forever) , this cause the system hung(Like the Plug into PC, the problem occurred sometimes wake up from sleep state, not every time) again.
The problem now like the following:
1 when plug our device into PC, the irp( used to send the second command to the device) sometimes was pending forever, the problem seems have been fixed by the update KB929762( we haven’t encountered the problem again after installed the update).
2 when wake up from sleep state, the irp again was pending forever sometimes, like the first problem, I don’t know why and when the irp will be pended forever, the update KB929762 not fix the problem.
Look forward your information.
Have you tried the binary from KB936150? It contains all of the fixes
from 929762 + more changes. While the articles do not describe your
particular problem, they are addressing some of the underlying issues
with logging into the device (which is the current theory of why your
i/o is stuck).
d
-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of activism@163.com
Sent: Tuesday, July 17, 2007 3:19 AM
To: Windows System Software Devs Interest List
Subject: RE:[ntdev] Cheers or be cool?
Doron Holan:
Thanks for your answer.
It a pity that i haven’t encountered the problem the update fixed both
the two:
http://support.microsoft.com/kb/929762/en-us
http://support.microsoft.com/kb/936150/en-us
I have tested many times on vista platform, i never encountered
bugchecks though i tried to wake up PC from sleep state or restart PC
from hibernate, but I do encountered the irp was pending forever even I
have installed the update KB929762( this update fix the problem that
when plugging storage device into pc, sometimes the irp for the second
command sent to device was pending forever) , this cause the system
hung(Like the Plug into PC, the problem occurred sometimes wake up from
sleep state, not every time) again.
The problem now like the following:
1 when plug our device into PC, the irp( used to send the second command
to the device) sometimes was pending forever, the problem seems have
been fixed by the update KB929762( we haven’t encountered the problem
again after installed the update).
2 when wake up from sleep state, the irp again was pending forever
sometimes, like the first problem, I don’t know why and when the irp
will be pended forever, the update KB929762 not fix the problem.
Look forward your information.
Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256
To unsubscribe, visit the List Server section of OSR Online at
http://www.osronline.com/page.cfm?name=ListServer