Re: Should raw I/O be added to the kernel?

Chris Wedgwood (cw@ix.net.nz)
Sat, 23 Jan 1999 18:19:48 +1300


On Fri, Jan 22, 1999 at 05:31:53PM -0600, Jason Froebe wrote:

> NO! Not to improve performance but to improve recoverability of
> databases. One will get 2 - 3 times better performance on a
> filesystem but we can't guarantee that the write has been
> completed.

Yes you can -- either using O_SYNC of fsync. Sure, it still copied to
the buffer cache, so in theory could be faster if it were direct, but
either of the above will guarantee (as far as the kernel can) the
data has been written to persistent storage.

-cw

-
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/