On Tue, 18 Apr 2006 10:25:36 +1000, Daniel Carosone <dan@geek.com.au>
wrote:

> On Mon, Apr 17, 2006 at 02:13:09AM +1000, Simon Burge wrote:
> > tmpfs is still marked with the comment "experimental - Efficient memory
> > file-system" and not enabled anywhere (except for evbarm/HDL_G and
> > mvme68k/VME177).  Is there any reason to not remove this warning and
> > enable it everywhere?
> 
> I'd also like to see it enabled everywhere, pre 4.0 branching.
> 
> There are some comments in the notes about various issues with
> allocation or nfs sharing, but I'm not clear what these refer to.  Is
> there a PR or other suitable reference for further details?
> 
One issue I know of is kern/32034.  There was a fair amount of discussion
on it, but it boils down to misbehavior of a Linux library compared to the
semantics of lseek() on tmpfs.

		--Steven M. Bellovin, http://www.cs.columbia.edu/~smb