From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from gate.crashing.org ([63.228.1.57]:48344 "EHLO gate.crashing.org") by vger.kernel.org with ESMTP id S932135AbWB1Jcx (ORCPT ); Tue, 28 Feb 2006 04:32:53 -0500 Subject: Re: [Fwd: [PATCH] Add mm->task_size and fix powerpc vdso] From: Benjamin Herrenschmidt In-Reply-To: <10152.1141118520@warthog.cambridge.redhat.com> References: <1141115747.3767.53.camel@localhost.localdomain> <10152.1141118520@warthog.cambridge.redhat.com> Content-Type: text/plain Date: Tue, 28 Feb 2006 20:32:45 +1100 Message-Id: <1141119165.3767.55.camel@localhost.localdomain> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Sender: linux-arch-owner@vger.kernel.org To: David Howells Cc: Linux Arch list List-ID: On Tue, 2006-02-28 at 09:22 +0000, David Howells wrote: > Benjamin Herrenschmidt wrote: > > > Forgot this list on the CC ... Comments welcome, the idea is to finally > > track if an mm is 32 or 64 bits (knowing that I have an urgent need to > > get that fixed in 2.6.16 due to a problem with the powerpc vdso). > > Is it possible to end up with a 32-bit and a 64-bit process sharing an > mm_struct? I assume not as I imagine this is set up by execve() which'll > destroy any CLONE_VM sharing. Well, if it's possible, I don't see how it would be possible for those to have a vDSO at all ... But I don't think it's possible with any current code path. Ben.