mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
* Block size problem
@ 2002-09-10 14:44 Jan Kara
  2002-09-10 17:58 ` OGAWA Hirofumi
  0 siblings, 1 reply; 3+ messages in thread
From: Jan Kara @ 2002-09-10 14:44 UTC (permalink / raw)
  To: linux-kernel

  Hello,

  my friend has a following problem: He has FAT filesystem on MO disk
  and computer with SCSI drive reading MO disk. The problem is that
  smallest blocksize supported by the driver is larger than 512 bytes
  which FAT needs. What is the right solution? Copying the device and
  loopback mounting is always possible but it's not nice...  
  Should I fix the driver to support 512 byte sectors (I'm not sure if
  SCSI drivers are supposed to support that)? Another solution I though
  about is creating loopback directly to device but loopback device
  supports only blocksize same as underlying device... So do you think
  it would be nice/useful if loopback device supported any blocksize?

  								Honza
-- 
Jan Kara <jack@suse.cz>
SuSE CR Labs

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Block size problem
  2002-09-10 14:44 Block size problem Jan Kara
@ 2002-09-10 17:58 ` OGAWA Hirofumi
  2002-09-10 18:03   ` Jan Kara
  0 siblings, 1 reply; 3+ messages in thread
From: OGAWA Hirofumi @ 2002-09-10 17:58 UTC (permalink / raw)
  To: Jan Kara; +Cc: linux-kernel

Jan Kara <jack@suse.cz> writes:

>   Hello,
> 
>   my friend has a following problem: He has FAT filesystem on MO disk
>   and computer with SCSI drive reading MO disk. The problem is that
>   smallest blocksize supported by the driver is larger than 512 bytes
>   which FAT needs. What is the right solution?

AFAIK, since originally FAT driver also isn't supporting blocksize
smaller than device sector size, I think he should use blocksize
larger than device sector size.

for example,

    $ mkdosfs -S 2048 /dev/xxx

>   Another solution I though about is creating loopback directly to
>   device but loopback device supports only blocksize same as
>   underlying device... So do you think it would be nice/useful if
>   loopback device supported any blocksize?

Since it isn't the problem of only FAT, I think it would be useful.
-- 
OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Block size problem
  2002-09-10 17:58 ` OGAWA Hirofumi
@ 2002-09-10 18:03   ` Jan Kara
  0 siblings, 0 replies; 3+ messages in thread
From: Jan Kara @ 2002-09-10 18:03 UTC (permalink / raw)
  To: OGAWA Hirofumi; +Cc: linux-kernel

> Jan Kara <jack@suse.cz> writes:
> 
> >   Hello,
> > 
> >   my friend has a following problem: He has FAT filesystem on MO disk
> >   and computer with SCSI drive reading MO disk. The problem is that
> >   smallest blocksize supported by the driver is larger than 512 bytes
> >   which FAT needs. What is the right solution?
> 
> AFAIK, since originally FAT driver also isn't supporting blocksize
> smaller than device sector size, I think he should use blocksize
> larger than device sector size.
> 
> for example,
> 
>     $ mkdosfs -S 2048 /dev/xxx
  Sadly that isn't a choice because disks are created by EWSD phone
  exchange...
  
> >   Another solution I though about is creating loopback directly to
> >   device but loopback device supports only blocksize same as
> >   underlying device... So do you think it would be nice/useful if
> >   loopback device supported any blocksize?
> 
> Since it isn't the problem of only FAT, I think it would be useful.

								Honza
-- 
Jan Kara <jack@suse.cz>
SuSE CR Labs

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2002-09-10 17:58 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2002-09-10 14:44 Block size problem Jan Kara
2002-09-10 17:58 ` OGAWA Hirofumi
2002-09-10 18:03   ` Jan Kara

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®