From: Tim Deegan <Tim.Deegan@citrix.com> To: Igor Mammedov <imammedo@redhat.com> Cc: <xen-devel@lists.xensource.com>, Keir Fraser <keir@xen.org>, Stefano Stabellini <stefano.stabellini@eu.citrix.com>, "containers@lists.linux-foundation.org" <containers@lists.linux-foundation.org>, Li Zefan <lizf@cn.fujitsu.com>, "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>, Michal Hocko <mhocko@suse.cz>, "linux-mm@kvack.org" <linux-mm@kvack.org>, Keir Fraser <keir.xen@gmail.com>, "akpm@linux-foundation.org" <akpm@linux-foundation.org>, "balbir@linux.vnet.ibm.com" <balbir@linux.vnet.ibm.com>, Paul Menage <menage@google.com>, KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>, Hiroyuki Kamezawa <kamezawa.hiroyuki@gmail.com> Subject: Re: [Xen-devel] Possible shadow bug Date: Fri, 10 Jun 2011 11:01:39 +0100 [thread overview] Message-ID: <20110610100139.GG5098@whitby.uk.xensource.com> (raw) In-Reply-To: <4DF0F90D.4010900@redhat.com> Hi, At 18:47 +0200 on 09 Jun (1307645229), Igor Mammedov wrote: > It's rhel5.6 xen. I've tried to test on SLES 11 that has 4.0.1 xen, however > wasn't able to reproduce problem. (I'm not sure if hap was turned > off in this case). More detailed info can be found at RHBZ#700565 The best way to be sure whether HAP is in use is to connect to the serial line, hit ^A^A^A to switch input to Xen, and hit 'q' to dump per-domain state. The printout for the guest domain should either say "paging assistance: shadow refcounts translate external" or "paging assistance: hap refcounts translate external". (If you don't have serial you can get the same info by running "xm debug-keys q" and then "xm dmesg" to read the output.) > >you're willing to try recompiling Xen with some small patches that > >disable the "cleverer" parts of the shadow pagetable code that might > >indicate something. (Of course, it might just change the timing to > >obscure a real linux bug too.) > > > Haven't got to this part yet. But looks like it's the only option left. Actually, looking at the disassembly you posted, it looks more like it might be an emulator bug in Xen; if Xen finds itself emulating the IMUL instruction and either gets the logic wrong or does the memory access wrong, it could cause that failure. And one reason that Xen emulates instructions is if the memory operand is on a pagetable that's shadowed (which might be a page that was recently a pagetable). ISTR that even though the RHEL xen reports a 3.0.x version it has quite a lot of backports in it. Does it have this patch? http://hg.uk.xensource.com/xen-3.1-testing.hg/rev/e8fca4c42d05 Cheers, Tim. -- Tim Deegan <Tim.Deegan@citrix.com> Principal Software Engineer, Xen Platform Team Citrix Systems UK Ltd. (Company #02937203, SL9 0BG)
WARNING: multiple messages have this Message-ID (diff)
From: Tim Deegan <Tim.Deegan@citrix.com> To: Igor Mammedov <imammedo@redhat.com> Cc: xen-devel@lists.xensource.com, Keir Fraser <keir@xen.org>, Stefano Stabellini <stefano.stabellini@eu.citrix.com>, "containers@lists.linux-foundation.org" <containers@lists.linux-foundation.org>, Li Zefan <lizf@cn.fujitsu.com>, "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>, Michal Hocko <mhocko@suse.cz>, "linux-mm@kvack.org" <linux-mm@kvack.org>, Keir Fraser <keir.xen@gmail.com>, "akpm@linux-foundation.org" <akpm@linux-foundation.org>, "balbir@linux.vnet.ibm.com" <balbir@linux.vnet.ibm.com>, Paul Menage <menage@google.com>, KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>, Hiroyuki Kamezawa <kamezawa.hiroyuki@gmail.com> Subject: Re: [Xen-devel] Possible shadow bug Date: Fri, 10 Jun 2011 11:01:39 +0100 [thread overview] Message-ID: <20110610100139.GG5098@whitby.uk.xensource.com> (raw) In-Reply-To: <4DF0F90D.4010900@redhat.com> Hi, At 18:47 +0200 on 09 Jun (1307645229), Igor Mammedov wrote: > It's rhel5.6 xen. I've tried to test on SLES 11 that has 4.0.1 xen, however > wasn't able to reproduce problem. (I'm not sure if hap was turned > off in this case). More detailed info can be found at RHBZ#700565 The best way to be sure whether HAP is in use is to connect to the serial line, hit ^A^A^A to switch input to Xen, and hit 'q' to dump per-domain state. The printout for the guest domain should either say "paging assistance: shadow refcounts translate external" or "paging assistance: hap refcounts translate external". (If you don't have serial you can get the same info by running "xm debug-keys q" and then "xm dmesg" to read the output.) > >you're willing to try recompiling Xen with some small patches that > >disable the "cleverer" parts of the shadow pagetable code that might > >indicate something. (Of course, it might just change the timing to > >obscure a real linux bug too.) > > > Haven't got to this part yet. But looks like it's the only option left. Actually, looking at the disassembly you posted, it looks more like it might be an emulator bug in Xen; if Xen finds itself emulating the IMUL instruction and either gets the logic wrong or does the memory access wrong, it could cause that failure. And one reason that Xen emulates instructions is if the memory operand is on a pagetable that's shadowed (which might be a page that was recently a pagetable). ISTR that even though the RHEL xen reports a 3.0.x version it has quite a lot of backports in it. Does it have this patch? http://hg.uk.xensource.com/xen-3.1-testing.hg/rev/e8fca4c42d05 Cheers, Tim. -- Tim Deegan <Tim.Deegan@citrix.com> Principal Software Engineer, Xen Platform Team Citrix Systems UK Ltd. (Company #02937203, SL9 0BG) -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2011-06-10 10:01 UTC|newest] Thread overview: 85+ messages / expand[flat|nested] mbox.gz Atom feed top 2011-06-01 10:44 [PATCH] memcg: do not expose uninitialized mem_cgroup_per_node to world Igor Mammedov 2011-06-01 12:39 ` Michal Hocko 2011-06-01 12:39 ` Michal Hocko 2011-06-01 13:07 ` Igor Mammedov 2011-06-01 13:07 ` Igor Mammedov [not found] ` <4DE6399C.8070802-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> 2011-06-01 13:41 ` Michal Hocko 2011-06-01 13:41 ` Michal Hocko 2011-06-01 13:41 ` Michal Hocko [not found] ` <20110601134149.GD4266-VqjxzfR4DlwKmadIfiO5sKVXKuFTiq87@public.gmane.org> 2011-06-01 14:39 ` Igor Mammedov 2011-06-01 14:39 ` Igor Mammedov 2011-06-01 14:39 ` Igor Mammedov 2011-06-01 15:20 ` Michal Hocko 2011-06-01 15:20 ` Michal Hocko [not found] ` <20110601152039.GG4266-VqjxzfR4DlwKmadIfiO5sKVXKuFTiq87@public.gmane.org> 2011-06-01 16:42 ` Igor Mammedov 2011-06-01 16:42 ` Igor Mammedov 2011-06-01 23:10 ` Hiroyuki Kamezawa 2011-06-01 23:10 ` Hiroyuki Kamezawa [not found] ` <BANLkTimbqHPeUdue=_Z31KVdPwcXtbLpeg-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org> 2011-06-03 12:35 ` Igor Mammedov 2011-06-03 12:35 ` Igor Mammedov 2011-06-03 12:35 ` Igor Mammedov 2011-06-03 13:00 ` Hiroyuki Kamezawa 2011-06-03 13:00 ` Hiroyuki Kamezawa 2011-06-07 13:25 ` Igor Mammedov 2011-06-07 13:25 ` Igor Mammedov [not found] ` <4DEE26E7.2060201-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> 2011-06-08 3:35 ` KAMEZAWA Hiroyuki 2011-06-08 3:35 ` KAMEZAWA Hiroyuki 2011-06-08 3:35 ` KAMEZAWA Hiroyuki 2011-06-08 21:09 ` Andrew Morton 2011-06-08 21:09 ` Andrew Morton 2011-06-08 23:44 ` KAMEZAWA Hiroyuki 2011-06-08 23:44 ` KAMEZAWA Hiroyuki [not found] ` <20110608140951.115ab1dd.akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org> 2011-06-08 23:44 ` KAMEZAWA Hiroyuki 2011-06-10 16:57 ` Igor Mammedov 2011-06-10 16:57 ` Igor Mammedov 2011-06-10 16:57 ` Igor Mammedov 2011-07-26 21:17 ` Andrew Morton 2011-07-26 21:17 ` Andrew Morton [not found] ` <20110726141754.c69b96c6.akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org> 2011-07-27 7:58 ` Michal Hocko 2011-07-27 7:58 ` Michal Hocko 2011-07-27 7:58 ` Michal Hocko 2011-07-27 9:30 ` Igor Mammedov 2011-07-27 9:30 ` Igor Mammedov 2011-07-27 9:57 ` Michal Hocko 2011-07-27 9:57 ` Michal Hocko [not found] ` <4E2FDAA0.5020702-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> 2011-07-27 9:57 ` Michal Hocko [not found] ` <20110727075845.GA4024-VqjxzfR4DlwKmadIfiO5sKVXKuFTiq87@public.gmane.org> 2011-07-27 9:30 ` Igor Mammedov [not found] ` <4DF24D04.1080802-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> 2011-07-26 21:17 ` Andrew Morton [not found] ` <20110608123527.479e6991.kamezawa.hiroyu-+CUm20s59erQFUHtdCDX3A@public.gmane.org> 2011-06-08 21:09 ` Andrew Morton 2011-06-09 8:11 ` Igor Mammedov 2011-06-09 8:11 ` Igor Mammedov 2011-06-09 8:11 ` Igor Mammedov 2011-06-09 12:40 ` Possible shadow bug (was: Re: [PATCH] memcg: do not expose uninitialized mem_cgroup_per_node to world) Stefano Stabellini 2011-06-09 12:40 ` Stefano Stabellini 2011-06-09 12:40 ` Stefano Stabellini 2011-06-09 15:01 ` [Xen-devel] " Tim Deegan 2011-06-09 15:01 ` Tim Deegan 2011-06-09 15:01 ` Tim Deegan 2011-06-09 16:47 ` [Xen-devel] Possible shadow bug Igor Mammedov 2011-06-09 16:47 ` Igor Mammedov 2011-06-10 10:01 ` Tim Deegan [this message] 2011-06-10 10:01 ` Tim Deegan 2011-06-10 10:10 ` Tim Deegan 2011-06-10 10:10 ` Tim Deegan [not found] ` <20110610101011.GH5098-uBdcGoUfBNNYtxbxJUhB2Dgeux46jI+i@public.gmane.org> 2011-06-10 11:48 ` Pasi Kärkkäinen 2011-06-10 13:55 ` Igor Mammedov 2011-06-10 11:48 ` Pasi Kärkkäinen 2011-06-10 11:48 ` Pasi Kärkkäinen 2011-06-10 12:40 ` Tim Deegan 2011-06-10 12:40 ` Tim Deegan 2011-06-10 15:38 ` Igor Mammedov 2011-06-10 15:38 ` Igor Mammedov [not found] ` <20110610124034.GI5098-uBdcGoUfBNNYtxbxJUhB2Dgeux46jI+i@public.gmane.org> 2011-06-10 15:38 ` Igor Mammedov [not found] ` <20110610114821.GB32595-GxtO3QLqHcLR7s880joybQ@public.gmane.org> 2011-06-10 12:40 ` Tim Deegan 2011-06-10 13:55 ` Igor Mammedov 2011-06-10 13:55 ` Igor Mammedov [not found] ` <20110610100139.GG5098-uBdcGoUfBNNYtxbxJUhB2Dgeux46jI+i@public.gmane.org> 2011-06-10 10:10 ` Tim Deegan [not found] ` <4DF0F90D.4010900-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> 2011-06-10 10:01 ` Tim Deegan [not found] ` <20110609150133.GF5098-uBdcGoUfBNNYtxbxJUhB2Dgeux46jI+i@public.gmane.org> 2011-06-09 16:47 ` Igor Mammedov [not found] ` <4DF0801F.9050908-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> 2011-06-09 12:40 ` Possible shadow bug (was: Re: [PATCH] memcg: do not expose uninitialized mem_cgroup_per_node to world) Stefano Stabellini [not found] ` <BANLkTinMamg_qesEffGxKu3QkT=zyQ2MRQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org> 2011-06-07 13:25 ` [PATCH] memcg: do not expose uninitialized mem_cgroup_per_node to world Igor Mammedov [not found] ` <4DE8D50F.1090406-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> 2011-06-03 13:00 ` Hiroyuki Kamezawa [not found] ` <4DE66BEB.7040502-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> 2011-06-01 23:10 ` Hiroyuki Kamezawa [not found] ` <4DE64F0C.3050203-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> 2011-06-01 15:20 ` Michal Hocko 2011-06-01 13:49 ` Igor Mammedov 2011-06-01 13:49 ` Igor Mammedov
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=20110610100139.GG5098@whitby.uk.xensource.com \ --to=tim.deegan@citrix.com \ --cc=akpm@linux-foundation.org \ --cc=balbir@linux.vnet.ibm.com \ --cc=containers@lists.linux-foundation.org \ --cc=imammedo@redhat.com \ --cc=kamezawa.hiroyu@jp.fujitsu.com \ --cc=kamezawa.hiroyuki@gmail.com \ --cc=keir.xen@gmail.com \ --cc=keir@xen.org \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-mm@kvack.org \ --cc=lizf@cn.fujitsu.com \ --cc=menage@google.com \ --cc=mhocko@suse.cz \ --cc=stefano.stabellini@eu.citrix.com \ --cc=xen-devel@lists.xensource.com \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.