[parisc-linux] Re: Problems with kernel mmap (failing tst-mmap-eofsync in glibc on parisc)

[David S. Miller]

On Sat, 23 Aug 2003 08:22:19 +0100 (BST)
Hugh Dickins <hugh@veritas.com> wrote:

> On 22 Aug 2003, James Bottomley wrote:
> > 
> > I suppose if we had a way of telling if any of the i_mmap list members
> > were really MAP_SHARED semantics mappings, then we could alter our
> > flush_dcache_page() implementation to work.
> 
> Good idea.  It's VM_MAYSHARE you need to check for.

Nope, please see my other email for why all of these ideas
simply will not work.  If the first fault-in of a MAP_PRIVATE
page is a read, it's just like a MAP_SHARED read-only page until
the first write occurs.