Re: 1394 Error code 0xC0120090

Tech-Archive recommends: Repair Windows Errors & Optimize Windows Performance

Actually, "if (status == STATUS_INVALID_GENERATION)" is perfectly valid.
The generation count can change asynchronously, so it is generally a good
idea to check for this failure status and update the generation count and
retry if this failure is received.

I noticed this generation status value changing several years ago and
emailed MSFT and asked them to please stop changing status values. I don't
know if they listened to me, or just realized they were screwing the world
here, but they finally stopped changing this value with each new DDK. I
don't have my 1394 code in front of me, but I believe they actually changed
it three times, the original value on 98, new value for Win2K and one more
for XP. Nice!

Bill M.

"Gary G. Little" <gary.g.little@xxxxxxxxxxx> wrote in message
news:43ba929c$0$15177$6d36acad@xxxxxxxxxxxxxxxxxxxxx
> Of course I must mention that "if (status == STATUS_INVALID_GENERATION)"
> is an improper form in the first place. You should be using
> NT_SUCCESS(status) or NT_ERROR(status). Is it beyond your kith and kin to
> mask off the offending bit; status &= 0xFFFF7FFF?
>
> The personal opinion of
> Gary G. Little
>
> "Hemant Mohan" <hmohan@xxxxxxxxx> wrote in message
> news:1136279545.977835.241140@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
>> Thanks for your response.
>> Here is some funny observation regarding this error code.
>> In the DDK 2600 the w2k/1394.h file defines this
>> STATUS_INVALID_GENERATION as 0xC0128090, while wxp/1394.h defines it as
>> 0xC0120090.
>>
>> While testing on a W2k the bus driver returned the number as 0xC0128090
>> and we modified the logic to solve the race condition issue with this
>> number. Now while testing it on XP the errno returned by lower bus
>> driver is 0xC0120090. So if we want a binary compatible driver then we
>> cannot check on STATUS_INVALID_GENERATION but need to explictly hard
>> code the numbers.
>>
>> Don't know what these folks at M$ are upto. They just can't keep upto
>> any standards.
>>
>
>


.

Relevant Pages

  • Re: 1394 Error code 0xC0120090
    ... > retry if this failure is received. ... > Bill M. ... >>> driver is 0xC0120090. ... >>> Don't know what these folks at M$ are upto. ...
    (microsoft.public.development.device.drivers)
  • Re: 1394 Error code 0xC0120090
    ... Here is some funny observation regarding this error code. ... In the DDK 2600 the w2k/1394.h file defines this ... While testing on a W2k the bus driver returned the number as 0xC0128090 ... Don't know what these folks at M$ are upto. ...
    (microsoft.public.development.device.drivers)
  • Re: Toyota acceleration Was Re: Snow Cover On Roof Provides Wind Protection?
    ... Not reacting properly when there apparently was quite a lot of time makes the participant an accomplice in the result of a failure not of their doing initially. ... The counter example cited is too dissimilar to be of any import -- in that case the remedy is to take some unusual precaution a priori; in the case under discussion it's the lack of an appropriate response to the event after it has occurred when there is ample opportunity to take corrective action. ... I'll disagree with the assertion that there's no culpability in severity of outcome independent of the driver in the incidents until and unless it's shown that the remedial actions were unavailable. ... wheel, covered her face and started screaming. ...
    (alt.home.repair)
  • Re: Max SAN time out
    ... It depends on if a hard error was raised by the IO driver or the OS, ... A Soft failure is more tolerable. ... If say a SAN Fiber Switch keeps the port up to the SQL Server, but the SAN drops totally on another port, and the switch does not return an error code, you then get into IO driver and OS defined timeouts. ... Soft failure followed by a hard failure. ...
    (microsoft.public.sqlserver.clustering)
  • Re: Max SAN time out
    ... A Soft failure is more tolerable. ... If a hard error comes up the driver stack to SQL Server, ... If say a SAN Fiber Switch keeps the port up to the SQL Server, ...
    (microsoft.public.sqlserver.clustering)