On Fri, 17 Mar 2000, Alan Cox wrote:
> > That's for cache coherency between different machines. On NFS, it is
> > usual to expect that if you call close() on one machine and then open()
> > on another, the other machine will see the written data.
>
> And for close down errors. NFS does writebehind, this means the fact
> that the disk is full may only be reported to you when the close() is called
> and the close fails to write the rest of the data.
Isn't asynchronous write another case of (granted, short-timed)
write-behind ? What happens if an I/O error happens on an asynchronously
written sector ?
Well, indeed fsync() already solves this for critical stuff, as Jamie
said.
-- Cyrille
------------------------------------------------------------------------------
Grumpf.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Thu Mar 23 2000 - 21:00:23 EST