From mboxrd@z Thu Jan 1 00:00:00 1970 From: James Bottomley Subject: Re: Patch for segfaults in minifail tests Date: Fri, 30 Apr 2010 16:13:44 -0500 Message-ID: <1272662024.2844.44.camel@mulgrave.site> References: <1272651697.2844.39.camel@mulgrave.site> <20100430211136.GA11644@hiauly1.hia.nrc.ca> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Cc: linux-parisc@vger.kernel.org, Helge Deller To: John David Anglin Return-path: In-Reply-To: <20100430211136.GA11644@hiauly1.hia.nrc.ca> List-ID: List-Id: linux-parisc.vger.kernel.org On Fri, 2010-04-30 at 17:11 -0400, John David Anglin wrote: > On Fri, 30 Apr 2010, James Bottomley wrote: > > > It's the essential patch we need to fix up our kmapping. Right at the > > moment kmap of a page with a dirty cache line in userspace sees stale > > data and kunmap of a page the kernel has modified will see likewise. > > Did you try using the version of copy_user_page_asm that uses > equivalent aliasing? There is an issue with needing cache flushing > noted in pgtable.h, but if this works it may be less of a hammer. Not yet. I want to isolate the root cause first before I start adding optimisations. Although the optimisation is a fairly safe one, since it's used by a lot of other architectures. James