Re: New filesystem for Linux

[Adrian Bunk <bunk@stusta.de>]

On Fri, Nov 03, 2006 at 02:31:46PM +0100, Mikulas Patocka wrote:
> >>>I am fully aware the counters are effectively 48-bit.  If they were
> >>>just 32-bit, you would likely have hit the problem yourself already.
> >>
> >>Given the seek time 0.01s, 31-bit value would last for minimum time of 248
> >>days when doing only syncs and nothing else. 47-bit value will last for
> >>reasonably long.
> >
> >So you can at most do one transaction per drive seek?  That would
> >definitely solve the overflow case, but hardly sounds like a
> >high-performance filesystem. :)
> 
> Really it can batch any number of modifications into one transaction 
> (unless fsync or sync is called). Transaction is closed only on 
> fsync/sync, if 2 minutes pass (can be adjusted) or when the disk runs out 
> of space.

O_DIRECT ?
O_SYNC ?

> Mikulas

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/