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

Larry McVoy (lm@bitmover.com)
Mon, 21 Dec 1998 15:06:26 -0800


"Theodore Y. Ts'o" <tytso@MIT.EDU> sayeth:
: As far as I know, POSIX has always required that fsync() return when the
: cache was flushed to disk, and I didn't think there were any fsync()
: implementations that didn't follow this rule.

Part of the problem is that OS implementations have sometimes forgotten to
sync associated metadata as part of fsync(). Given that people may have
been bitten by this in the past, they don't always trust fsync() to do the
right thing.

In the raw I/O case, the application is managing the meta data so it
can do whatever it wants.

-
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/