Re: [RFC PATCH 2/2] mm, fs: daxfile, an interface for byte-addressable updates to pmem

From: Darrick J. Wong
Date: Thu Jun 22 2017 - 03:11:23 EST

On Tue, Jun 20, 2017 at 09:42:55AM -0600, Ross Zwisler wrote:
> On Mon, Jun 19, 2017 at 10:22:14PM -0700, Darrick J. Wong wrote:
> <>
> > Fourth, the VFS entry points for things like read, write, truncate,
> > utimes, fallocate, etc. all just bail out if S_IOMAP_FROZEN is set on a
> > file, so that the block map cannot be modified. mmap is still allowed,
> > as we've discussed. /Maybe/ we can allow fallocate to extend a file
> > with zeroed extents (it will be slow) as I've heard murmurs about
> > wanting to be able to extend a file, maybe not.
> Read and write should still be allowed, right?

<shrug> I had thought the usage model was pretty slanted towards mmap,
but it's not a big deal to turn read/writes into glorified memcpy,
provided we reject the io request if it goes past EOF.


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