On Thu, 7 Sep 2006, Siddharth Choudhuri wrote:
> Thanks. I put printk() in sd_init_command() to output
> block(rq->sector) and count(Scpnt->request_bufflen >>
> 9). Now, I can see the [read/write, block, count] as I
> start some file I/O activity on a mounted usb flash
> disk (moutned as vfat). I
>
> If you want to get this information, you're probably
> better off getting it
> at the point where it is sent to the device, not at
> the points where the
> conversions are made. For example, take a look at
> sd_init_command() in
> drivers/scsi/sd.c.
>
Thanks. I put printk() in sd_init_comman
On Wed, 6 Sep 2006 10:44:11 -0700 (PDT), Siddharth Choudhuri <[EMAIL
PROTECTED]> wrote:
> > 2800 0020 2000 --
> > SCSI command: 0x28=READ(10), Block=0x20,
> I got usbmon running on 2.6.16 kernel. I was wondering
> how did you get block number = 0x20 and length =
> 0x20*512 i
> For example:
>
> ca6a349c 3984599108 S Bo:004:02 -115 31 = 55534243
> 4900 0040 8a28 2020
> 00
> ...
> ...
> Data:
> 55534243 Magic
> 4900 0040
> 8000 Input data transfer
> 0a10 byte CDB
> 2800 0020 2000 --
>
On Mon, 4 Sep 2006, Siddharth Choudhuri wrote:
> > > I am trying to implement a kernel module that can
> > > track read/write requests to a USB-mounted flash
> > > device (say, mounted as FAT filesystem).
> >
> > When you say "track", exactly what do you mean?
> > Like Luiz said, usbmon
> > wi
On Mon, 4 Sep 2006, Pete Zaitcev wrote:
> On Mon, 4 Sep 2006 17:24:52 -0700 (PDT), Siddharth Choudhuri <[EMAIL
> PROTECTED]> wrote:
>
> > [...] What I need is [request_type (R/W), size,
> > logical address (sector number)] being issued to the
> > usb-flash storage device. My guess is block numbe
> > I am trying to implement a kernel module that can
> > track read/write requests to a USB-mounted flash
> > device (say, mounted as FAT filesystem).
>
> When you say "track", exactly what do you mean?
> Like Luiz said, usbmon
> will produce a log of all USB messages; is that what
> you want?
On Mon, 4 Sep 2006 17:24:52 -0700 (PDT), Siddharth Choudhuri <[EMAIL
PROTECTED]> wrote:
> [...] What I need is [request_type (R/W), size,
> logical address (sector number)] being issued to the
> usb-flash storage device. My guess is block number is
> being converted to logical address (sector nu
> > | I am unable to figure out how the read,write
> requests
> > | that are in the form of [sector, size] in the
> VFS
> > | layer are passed into urb structure before
> calling
> > | usb_submit_urb(). My understanding is at the
> lowest
> > | level a call to usb_submit_urb() is required to
> > |
On Sat, 2 Sep 2006, Siddharth Choudhuri wrote:
> Hi,
>
> I am trying to implement a kernel module that can
> track read/write requests to a USB-mounted flash
> device (say, mounted as FAT filesystem).
When you say "track", exactly what do you mean? Like Luiz said, usbmon
will produce a log of
On Sat, 2 Sep 2006 20:40:50 -0300, "Luiz Fernando N. Capitulino" <[EMAIL
PROTECTED]> wrote:
> | I am unable to figure out how the read,write requests
> | that are in the form of [sector, size] in the VFS
> | layer are passed into urb structure before calling
> | usb_submit_urb(). My understanding
Hi Siddharth,
On Sat, 2 Sep 2006 15:03:03 -0700 (PDT)
Siddharth Choudhuri <[EMAIL PROTECTED]> wrote:
| Hi,
|
| I am trying to implement a kernel module that can
| track read/write requests to a USB-mounted flash
| device (say, mounted as FAT filesystem).
|
| I am unable to figure out how the
Hi,
I am trying to implement a kernel module that can
track read/write requests to a USB-mounted flash
device (say, mounted as FAT filesystem).
I am unable to figure out how the read,write requests
that are in the form of [sector, size] in the VFS
layer are passed into urb structure before calli
13 matches
Mail list logo