Re: [ Mind testing experimental one-liner? ]
Simon Kirby (sim@netnation.com)
Mon, 18 Jan 1999 11:12:29 -0800 (PST)
On Mon, 18 Jan 1999, Benjamin LaHaise wrote:
> On Mon, 18 Jan 1999, Simon Kirby wrote:
>
> > Hmm...What would this do? Allow flushing to be done asynchronously?
> > Is there anything that could be unsafe with this change?
>
> It should be as safe as anything else (it's the fallback for filesystems
> that don't implement fsync). file_fsync does a sync on the device the
> filesystem is mounted on, which covers everything except dirty shared
> mmap()ings of the file (which ext2_file_sync doesn't deal with anyways);
> it just generates more disk writes than nescessary, but no reads.
So how is it better than ext2_file_sync()? Would it sort the blocks
or write asynchronously or something? It seems to me that the biggest
problem for the machine currently is that it can't read when
flushing...would this change allow that?
Simon-
| Simon Kirby | Systems Administration |
| mailto:sim@netnation.com | NetNation Communications |
| http://www.netnation.com/ | Tech: (604) 684-6892 |
-
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/