Re: [PATCH] barrier patch set

From: Jeff Garzik
Date: Wed Mar 31 2004 - 16:36:43 EST

Chris Mason wrote:
On Wed, 2004-03-31 at 09:08, Stephen C. Tweedie wrote:


On Tue, 2004-03-30 at 23:21, Jeff Garzik wrote:

For IDE, O_DIRECT and O_SYNC can use special "FUA" commands, which don't return until the data is on the platter.

fsync() is still really nasty, because that can require that we wait on
IO that was submitted by the VM before we knew that there was a
synchronous IO wait coming.

Yes, it gets ugly in a hurry. Jeff, look at the whole thread about the
O_DIRECT read vs buffered write races. I don't think we can use FUA for

Yes, I'm aware of the thread...

fsync or O_SYNC without using it for every write.

Why not for O_SYNC? Is some crazy userspace application flipping this bit on and off rapidly?

We might be able to get away with using it on O_DIRECT.



To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at