Hello jerome,
sorry, my mistake. those are for minifilters only… I thought
for some reason that it was a minifilter you were developing…
Perhaps one of the gurus here can answer your question.
–
Best regards,
Alex mailto:xxxxx@bitdefender.com
Wednesday, September 14, 2005, 2:53:46 AM, you wrote:
Hi Alex,
Thanks for your fast reply.
----- Original Message -----
From: “Alex Carp”
> Newsgroups: ntfsd
> To: “Windows File Systems Devs Interest List”
> Sent: Tuesday, September 13, 2005 7:22 AM
> Subject: Re: RegisterFileSystemFilterCallbacks
>> Hello jerome,
>>
>> not sure I got your question right,
> Sorry if the message is confused. The question is why I don’t get the
> …release_for_mod_write action ?
>>but when I need my PostXxx
>> routine called, i return FLT_PREOP_SUCCESS_WITH_CALLBACK in the
>> PreXxx operation. Otherwise, I return
>> FLT_PREOP_SUCCESS_NO_CALLBACK.
>>
>> Hope it helps…
> Seems very interesting but I don’t seem to find these values in my
> environment ?
> (my ddk build is 3790 and ifkit build number is 3790.1159)
> Thanks,
> Jerome.
>>
>>
>> –
>> Best regards,
>> Alex mailto:xxxxx@bitdefender.com
>>
>> Wednesday, September 14, 2005, 2:11:28 AM, you wrote:
>>
>> > Hello,
>>
>> > My filter driver is using
>> > FsRtlRegisterFileSystemFilterCallbacks() on
>> > windows XP and the 2003 Ifs kit.
>>
>> > These callbacks functions seems to work properly for
>> > FS_FILTER_ACQUIRE_FOR_CC_FLUSH,
>> FS_FILTER_RELEASE_FOR_CC_FLUSH meaning
> that
>> > I get called for pre/post processing for each action.
>>
>> > But the modified page writer thread is only calling my function at pre
>> > processing time and only for FS_FILTER_ACQUIRE_FOR_MOD_WRITE no post
>> > processing and no release action are called ? I don’t think it is the
>> > expected behavior ?
>>
>> > My understanding is that the modified page thread in the IO completion
>> > process is checking a variable (resource to release ?) against
> NULL -which
>> > is always true- and does not bother to send the release message to the
>> > filter after write pages to disk because it wrongly assumes no resource
> was
>> > locked at preprocessing time in the filter. (I think this is confirmed
> from
>> > a previous post from Molly Brown of Microsoft?)
>>
>> > If so, how do I set the resource to release argument ?
>>
>> > I tried to set it like: FS_FILTER_CALLBACK_DATA
>> Data->>Parameters.AcquireForModifiedPageWriter.ResourceToRelease =… or
>> > through the CompletionContext pointer without effect.
>>
>> > Note that always I’m using the same function callback with a switch
>> > statement to differentiate actions MOD_WRITE, CC_FLUSH…
>>
>> > Should I return another value than STATUS_SUCCESS ? (like
>> > STATUS_FSFILTER_OP_COMPLETED_SUCCESSFULLY 0x00000126L) or in this case
> does
>> > returning …INVALID_REQUEST like the old fastio method force the kernel
> to
>> > adopt the default behavior ?
>>
>> > Thank you.
>>
>> > Jerome.
>>
>>
>>
>> > —
>> > Questions? First check the IFS FAQ at
>> > https://www.osronline.com/article.cfm?id=17
>>
>> > You are currently subscribed to ntfsd as: xxxxx@bitdefender.com
>> > To unsubscribe send a blank email to
>> xxxxx@lists.osr.com
>>
> —
> Questions? First check the IFS FAQ at
> https://www.osronline.com/article.cfm?id=17
> You are currently subscribed to ntfsd as: xxxxx@bitdefender.com
> To unsubscribe send a blank email to xxxxx@lists.osr.com
–
This message was scanned for spam and viruses by BitDefender.
For more information please visit http://www.bitdefender.com/
Hi Alex,
“Alex Carp” wrote in message news:xxxxx@ntfsd…
> Hello jerome,
>
> sorry, my mistake. those are for minifilters only… I thought
> for some reason that it was a minifilter you were developing…
Sorry I should have mention it was a ‘full’ filter driver.
For what it is worth, it seems that returning STATUS_INVALID_DEVICE_REQUEST
in the ‘prefilter’ function for registerfileSystemFilterCallbacks() with
FS_FILTER_ACQUIRE_FOR_MOD_WRITE operation does trigger the
FS_FILTER_RELEASE_FOR_MOD_WRITE callback later and this whatever the OUT
arguments settings when acquiring the resource in the first callback.
Jerome.
>
> Perhaps one of the gurus here can answer your question.
>
>
> –
> Best regards,
> Alex mailto:xxxxx@bitdefender.com
>
> Wednesday, September 14, 2005, 2:53:46 AM, you wrote:
>
> > Hi Alex,
>
> > Thanks for your fast reply.
>
> > ----- Original Message -----
> > From: “Alex Carp”
> > Newsgroups: ntfsd
> > To: “Windows File Systems Devs Interest List”
> > Sent: Tuesday, September 13, 2005 7:22 AM
> > Subject: Re: RegisterFileSystemFilterCallbacks
>
>
> >> Hello jerome,
> >>
> >> not sure I got your question right,
>
> > Sorry if the message is confused. The question is why I don’t get the
> > …release_for_mod_write action ?
>
> >>but when I need my PostXxx
> >> routine called, i return FLT_PREOP_SUCCESS_WITH_CALLBACK in the
> >> PreXxx operation. Otherwise, I return
> >> FLT_PREOP_SUCCESS_NO_CALLBACK.
> >>
> >> Hope it helps…
>
> > Seems very interesting but I don’t seem to find these values in my
> > environment ?
> > (my ddk build is 3790 and ifkit build number is 3790.1159)
>
> > Thanks,
>
> > Jerome.
>
> >>
> >>
> >> –
> >> Best regards,
> >> Alex mailto:xxxxx@bitdefender.com
> >>
> >> Wednesday, September 14, 2005, 2:11:28 AM, you wrote:
> >>
> >> > Hello,
> >>
> >> > My filter driver is using
> >> > FsRtlRegisterFileSystemFilterCallbacks() on
> >> > windows XP and the 2003 Ifs kit.
> >>
> >> > These callbacks functions seems to work properly for
> >> > FS_FILTER_ACQUIRE_FOR_CC_FLUSH,
> >> FS_FILTER_RELEASE_FOR_CC_FLUSH meaning
> > that
> >> > I get called for pre/post processing for each action.
> >>
> >> > But the modified page writer thread is only calling my function at
pre
> >> > processing time and only for FS_FILTER_ACQUIRE_FOR_MOD_WRITE no post
> >> > processing and no release action are called ? I don’t think it is the
> >> > expected behavior ?
> >>
> >> > My understanding is that the modified page thread in the IO
completion
> >> > process is checking a variable (resource to release ?) against
> > NULL -which
> >> > is always true- and does not bother to send the release message to
the
> >> > filter after write pages to disk because it wrongly assumes no
resource
> > was
> >> > locked at preprocessing time in the filter. (I think this is
confirmed
> > from
> >> > a previous post from Molly Brown of Microsoft?)
> >>
> >> > If so, how do I set the resource to release argument ?
> >>
> >> > I tried to set it like: FS_FILTER_CALLBACK_DATA
> >> Data->>Parameters.AcquireForModifiedPageWriter.ResourceToRelease =… or
> >> > through the CompletionContext pointer without effect.
> >>
> >> > Note that always I’m using the same function callback with a switch
> >> > statement to differentiate actions MOD_WRITE, CC_FLUSH…
> >>
> >> > Should I return another value than STATUS_SUCCESS ? (like
> >> > STATUS_FSFILTER_OP_COMPLETED_SUCCESSFULLY 0x00000126L) or in this
case
> > does
> >> > returning …INVALID_REQUEST like the old fastio method force the
kernel
> > to
> >> > adopt the default behavior ?
> >>
> >> > Thank you.
> >>
> >> > Jerome.
> >>
> >>
> >>
> >> > —
> >> > Questions? First check the IFS FAQ at
> >> > https://www.osronline.com/article.cfm?id=17
> >>
> >> > You are currently subscribed to ntfsd as: xxxxx@bitdefender.com
> >> > To unsubscribe send a blank email to
> >> xxxxx@lists.osr.com
> >>
>
>
>
> > —
> > Questions? First check the IFS FAQ at
> > https://www.osronline.com/article.cfm?id=17
>
> > You are currently subscribed to ntfsd as: xxxxx@bitdefender.com
> > To unsubscribe send a blank email to xxxxx@lists.osr.com
>
>
>
>
>
> –
> This message was scanned for spam and viruses by BitDefender.
> For more information please visit http://www.bitdefender.com/
>
>