[prev in list] [next in list] [prev in thread] [next in thread] 

List:       linux-fsdevel
Subject:    Re: Q: sb->s_bdev same device as sb->s_dev?
From:       Alexander Viro <viro () math ! psu ! edu>
Date:       2001-03-12 7:59:28
[Download RAW message or body]



On Mon, 12 Mar 2001 Andries.Brouwer@cwi.nl wrote:

> static dev_t next_unnamed_device = 0x10000000000ULL;
> 
> kdev_t get_unnamed_dev(void) {
> 	return to_kdev_t(next_unnamed_device++);
> }
> 
> void put_unnamed_dev(kdev_t dev) {
> }
> 
>     We need to put something into ->st_dev upon stat(2). And with the current
>     icache we need something unique in ->i_dev.
> 
> Yes, no changes there.

For the first part - sure. For the second... I'd rather avoid long long
in __iget(). So we are down to 32 bits and we'd better make sure that
there will be no clashes. I really don't think that st_dev should have
anything to icache search. OK, anyway - I'll do the patch that switches
icache to i_sb as a key, put it on anonftp and announce here. I suspect
that actual code is a better way to explain what I mean ;-)
							Cheers,
								Al

-
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to majordomo@vger.kernel.org

[prev in list] [next in list] [prev in thread] [next in thread]