Subject: Re: cp(1) vs. UVM/UBC
To: Todd Vierling <email@example.com>
From: Ignatios Souvatzis <firstname.lastname@example.org>
Date: 07/20/2001 22:52:42
On Thu, Jul 19, 2001 at 10:29:50PM -0400, Todd Vierling wrote:
> On Wed, 18 Jul 2001, Jason R Thorpe wrote:
> : > would there be any point to looking at the output of fstatfs() after
> : > the open() to decide a code path?
> : Let's not start special-casing various file systems in tools :-)
> We could always have the file system itself fail the mmap in the interim, if
> need be, until it is in fact fixed. Not all filesystems necessarily *must*
> support mmap; just return -1 and an errno, and let smart code fall back. 8-)
how does it do it?