Prev: Race in wb_do_writeback() ???
Next: Length and alignment in scsi_execute_async and blk_rq_map_user
From: David Howells on 29 Jun 2010 17:20
Bernd Schubert <bernd.schubert(a)fastmail.fm> wrote:

> Is there any chance we can use that chance and also add a field
>
> unsigned long long st_gen
>
> to struct_ xstat? Inode generation numbers really would be useful for
> userspace NFS servers and some fuse filesystems.

That would be st_inode_version (equivalent to i_generation internally).

David
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Bernd Schubert on 29 Jun 2010 17:30
On Tuesday, June 29, 2010, David Howells wrote:
> Bernd Schubert <bernd.schubert(a)fastmail.fm> wrote:
> > Is there any chance we can use that chance and also add a field
> >
> > unsigned long long st_gen
> >
> > to struct_ xstat? Inode generation numbers really would be useful for
> > userspace NFS servers and some fuse filesystems.
>
> That would be st_inode_version (equivalent to i_generation internally).

Ah, great, so already there :) I was looking for st_gen, as it is called that
way on BSD. And as BSD already has it for a long time, shouldn't linux use the
BSD identifier?


Thanks,
Bernd
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: David Howells on 29 Jun 2010 17:30
Bernd Schubert <bernd.schubert(a)fastmail.fm> wrote:

> Ah, great, so already there :) I was looking for st_gen, as it is called
> that way on BSD. And as BSD already has it for a long time, shouldn't linux
> use the BSD identifier?

Sure. I guess you'd also want it to be a u64?

David
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Bernd Schubert on 29 Jun 2010 18:00
On Tuesday, June 29, 2010, David Howells wrote:
> Bernd Schubert <bernd.schubert(a)fastmail.fm> wrote:
> > Ah, great, so already there :) I was looking for st_gen, as it is called
> > that way on BSD. And as BSD already has it for a long time, shouldn't
> > linux use the BSD identifier?
>
> Sure. I guess you'd also want it to be a u64?

Hmm, as far as I can see, BSD has u32. I only need it to verify for recycled
inodes and at least for me the probability of a recyled inode + 32 bit
generation number that overflew to exactly the same value as the previous
inode had would be sufficiently small.


Thanks a lot for your work on this,
Bernd
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Maciej W. Rozycki on 29 Jun 2010 19:10
On Tue, 29 Jun 2010, David Howells wrote:

> > Ah, great, so already there :) I was looking for st_gen, as it is called
> > that way on BSD. And as BSD already has it for a long time, shouldn't linux
> > use the BSD identifier?
>
> Sure. I guess you'd also want it to be a u64?

Note the Alpha port has had an st_gen member reserved in its struct stat
for many years now ;) -- which could have been DEC OSF/1 legacy. I'm glad
to see this member seriously considered after these many years and
previously rejected proposals.

Maciej
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
First  |  Prev  | 
Pages: 1 2
Prev: Race in wb_do_writeback() ???
Next: Length and alignment in scsi_execute_async and blk_rq_map_user