From mboxrd@z Thu Jan 1 00:00:00 1970 From: Konrad Rzeszutek Wilk Subject: Re: Re: ATI radeon fails with "iommu=soft swiotlb=force" (seen on RV730/RV740 and RS780/RS800) Date: Fri, 2 Oct 2009 10:44:19 -0400 Message-ID: <20091002144419.GA4239@phenom.dumpdata.com> References: <20090930152149.GA29603@phenom.dumpdata.com> <20091001173558.GA1921@phenom.dumpdata.com> <4AC4FDEA.3070909@goop.org> <20091001192140.GB2367@phenom.dumpdata.com> <4AC613BE0200007800017B2C@vpn.id2.novell.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <4AC613BE0200007800017B2C@vpn.id2.novell.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Jan Beulich Cc: Jeremy Fitzhardinge , dri-devel@lists.sourceforge.net, xen-devel@lists.xensource.com List-Id: xen-devel@lists.xenproject.org On Fri, Oct 02, 2009 at 01:52:46PM +0100, Jan Beulich wrote: > >>> Konrad Rzeszutek Wilk 01.10.09 21:21 >>> > >The are other users of 'vmalloc_32' that look like they depend on this > >memory being under the 4GB mark. Most of them are do video capture through > >USB - so it probably is limited to only accessing up to 4GB. > > I just went through all of the users of vmalloc_32(), and more than half of > them seem bogus (like some legitimate use of it got cloned many times > without really needing all the restrictions that come with this). Of course > I can't verify that I'm right with all of those, so I'm not sure how to > proceed with trying to do some clean up here... I actually have some of those devices in the basement or I buy some of them at the MIT Flea. Will dig them up and make sure they work properly.