If I send an IRP with IRP_MJ_WRITE (non-cached,
but not paging IO), the byte count *must not* be rounded
up to the sector size, but the buffer *must* be of size at least
the nearest multiplier of sector size.
Is it true ?
L.
If I send an IRP with IRP_MJ_WRITE (non-cached,
but not paging IO), the byte count *must not* be rounded
up to the sector size, but the buffer *must* be of size at least
the nearest multiplier of sector size.
Is it true ?
L.
The byte count “MUST” be sector aligned and the buffer must be sector
aligned too. Since there is no cache manager in the picture, you can not to
byte-oriented reads.
Jamey
-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of Ladislav Zezula
Sent: Monday, October 25, 2004 4:59 AM
To: Windows File Systems Devs Interest List
Subject: [ntfsd] Number of bytes to write when IRP_NOCACHE
If I send an IRP with IRP_MJ_WRITE (non-cached,
but not paging IO), the byte count *must not* be rounded
up to the sector size, but the buffer *must* be of size at least
the nearest multiplier of sector size.
Is it true ?
L.
Questions? First check the IFS FAQ at
https://www.osronline.com/article.cfm?id=17
You are currently subscribed to ntfsd as: xxxxx@storagecraft.com
To unsubscribe send a blank email to xxxxx@lists.osr.com
Hi, Jamey,
Thank you for the response.
The byte count “MUST” be sector aligned and the buffer must be sector
aligned too. Since there is no cache manager in the picture, you can not
to
byte-oriented reads.
I thought so too. But recently, I investigated
what the redirector does when synchronizing Offline Folders.
When synchronization runs, the redirector sends down
a non-cached, not-paging-IO, write request(s) to locally
cached offline file(s). The offset seems to be always aligned to sector
size (I haven’t found a case where not), but the write length
is e.g. 156 bytes.
Why Microsoft violates its own rule ?
L.
File systems do not enforce sector aligned length if you are writing up to
the EOF of beyond EOF.
In this case actual write length is rounded up to the sector boundary but
EOF is not extended to the sector boundary.
Alexei.
“Ladislav Zezula” wrote in message news:xxxxx@ntfsd…
> Hi, Jamey,
>
> Thank you for the response.
>
> > The byte count “MUST” be sector aligned and the buffer must be sector
> > aligned too. Since there is no cache manager in the picture, you can not
> > to
> > byte-oriented reads.
>
> I thought so too. But recently, I investigated
> what the redirector does when synchronizing Offline Folders.
> When synchronization runs, the redirector sends down
> a non-cached, not-paging-IO, write request(s) to locally
> cached offline file(s). The offset seems to be always aligned to sector
> size (I haven’t found a case where not), but the write length
> is e.g. 156 bytes.
> Why Microsoft violates its own rule ?
>
> L.
>
>
>