From mboxrd@z Thu Jan 1 00:00:00 1970 From: Steven Rostedt Subject: Re: [PATCH] Fix >4G i386 PAE grant table interface Date: Fri, 03 Nov 2006 08:43:00 -0500 Message-ID: <454B4764.6080709@redhat.com> References: <454A2271.90000@redhat.com> <454B0F19.76E4.0078.0@novell.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <454B0F19.76E4.0078.0@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: xen-devel@lists.xensource.com List-Id: xen-devel@lists.xenproject.org Jan Beulich wrote: >>>> Steven Rostedt 02.11.06 17:53 >>> >> It has been discovered that i386 boxes with more than 4G of RAM would >> randomly crash. It was traced to the interface of blktap using >> gnttab_set_map_op. >> >> It would pass in the 64 bit pte entry, but the gnttab_set_map_op would >> only take a 32 bit (on i386) unsigned long as a parameter. So we lose >> the top 32bits. > > Could you use maddr_t here rather than uint64_t? For non-PAE i386 > Linux, especially when using CONFIG_REGPARM, adding a useless > argument slot seems wasteful... > Actually, it makes no difference to me. In fact uint64_t was my third incarnation, since I wasn't sure what the best would be. I started with unsigned long long, then switched to u64, and then noticed that since host_addr is uint64_t, that seemed the proper thing to use. So a maddr_t would work too. Do you want to do the patch, or would you like me to send another patch that would do this change? -- Steve