Re[2]: vcpu-pin cause dom0 kernel panic

From: Andrew Xu <xu.an@cloudex.cn>
To: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: xen-devel@lists.xensource.com, yunhong.jiang@intel.com
Subject: Re[2]: vcpu-pin cause dom0 kernel panic
Date: Fri, 27 May 2011 17:38:23 +0800	[thread overview]
Message-ID: <20110527173823.68CA.3A8D29D5@cloudex.cn> (raw)
In-Reply-To: <20110524140150.GE10926@dumpdata.com>

I update update RAID firmware from 12.10.1.001 to 12.12.0.048.
pin-2 dom0 can find disk,
but, pin-1 dom0 still panic.
------------------------------------------------
(XEN) [VT-D]iommu.c:821: iommu_fault_status: Primary Pending Fault
(XEN) [VT-D]iommu.c:796: DMAR:[DMA Read] Request device [03:00.0] fault addr bf173000, iommu reg = ffff82c3fff57000
(XEN) DMAR:[fault reason 06h] PTE Read access is not set
(XEN) print_vtd_entries: iommu = ffff83103fdf7ea0 bdf = 3:0.0 gmfn = bf173
(XEN)     root_entry = ffff83103fd38000
(XEN)     root_entry[3] = 1011b5d001
(XEN)     context = ffff831011b5d000
(XEN)     context[0] = 2_1019da3005
(XEN)     l4 = ffff831019da3000
(XEN)     l4_index = 0
(XEN)     l4[0] = 1019da2003
(XEN)     l3 = ffff831019da2000
(XEN)     l3_index = 2
(XEN)     l3[2] = 101999f003
(XEN)     l2 = ffff83101999f000
(XEN)     l2_index = 1f8
(XEN)     l2[1f8] = 10197a6003
(XEN)     l1 = ffff8310197a6000
(XEN)     l1_index = 173
(XEN)     l1[173] = 0
(XEN)     l1[173] not present
-------------------------------------------------

When I add "loglvl=none" to xen bootoptions, pin-1 dom0 run OK.
It is so !!!strange!!!

On Tue, 24 May 2011 10:01:50 -0400
Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> wrote:

> On Thu, May 19, 2011 at 09:08:24PM +0800, Andrew Xu wrote:
> > I use xen 4.0(dom0 is suse11.sp1,2.6.32 x86_64) on 
> 
> Did you open a Bugzilla with SLES?
> 
> > Dell R710 with PERC H700 RAID adapter.
> > --Intel(R) Xeon(R) CPU E5620 @ 2.40GHz
> > --8 CPU cores.
> > --Memory 64G
> > --RAID5 4.5T
> > 
> > When I dedicated (pin) a CPU core only for dom0 use.
> > (I specify "dom0_max_vcpus=1 dom0_vcpus_pin" options for Xen)
> > I got dom0 kernel panic error(pin-1-5.30.bmp)
> > 
> > When I pin 2 core to dom0, the dom0 system can boot up,
> > but it can't find any disk devices.
> > 
> > But when I pin 4 core to dom0, it run OK.
> > 
> > Does anyone can help me, please.
> > 
> > I have read this mail, and updated my megaraid_sas to version 05.30.
> > http://xen.1045712.n5.nabble.com/HOWTO-Update-megaraid-sas-driver-in-Linux-2-6-32-x-to-make-it-boot-and-work-with-Dell-LSI-RAID-adapts-td3254332.html
> > Anyone can help me?
> > 
> > I download megaraid_sas_05.30 driver from LSI web:
> > http://www.lsi.com/storage_home/products_home/internal_raid/megaraid_sas/6gb_s_value_line/sas9260-8i/index.html
> 
> 
> > _______________________________________________
> > Xen-devel mailing list
> > Xen-devel@lists.xensource.com
> > http://lists.xensource.com/xen-devel
> 
> 
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel

**************************************************
徐安（Andrew Xu）
部门  ：云快线 - 运营支撑中心 - 研发中心
手机  : 18910391796
E-mail：xu.an@cloudex.cn
地址  ：北京市朝阳区酒仙桥东路1号M5楼
**************************************************