From mboxrd@z Thu Jan 1 00:00:00 1970 From: Keir Fraser Subject: Re: Fatal page fault on xen-unstable tip (vt-d related maybe?) Date: Thu, 12 Jun 2008 08:28:39 +0100 Message-ID: References: <1213241283.7094.24.camel@ip6-localhost> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Return-path: In-Reply-To: <1213241283.7094.24.camel@ip6-localhost> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: "Yang, Xiaowei" , dan.magenheimer@oracle.com Cc: xen-devel List-Id: xen-devel@lists.xenproject.org If this is a common issue (DMA outside RAM and RMRR ranges) then we may hav= e to turn off VT-d for PV guests, including dom0, by default. -- Keir On 12/6/08 04:28, "Yang, Xiaowei" wrote: > Some info got from you log: > Strange to see there is VTd write fault from =EF=BB=BFa high address f1bffb000. > It's not covered by e820. Could it be in one MMIO range? >=20 > I can't figure out why a page fault from =EF=BB=BF7b9c51e0 (not in Xen HV), > while on the top of the stack is =EF=BB=BFiommu_enable_translation(). It seems > output of print_vtd_entries() is not finished. And as we met Page fault > when executing it before, could it be the cause?? >=20 > P.S. you log is really messed up. To make life easier, =EF=BB=BFsync_console Xe= n > option is recommended:) >=20 >=20 > Thanks, > Xiaowei >=20 >=20 >> -----Original Message----- >> From: xen-devel-bounces@lists.xensource.com >> [mailto:xen-devel-bounces@lists.xensource.com]On Behalf Of Dan >> Magenheimer >> Sent: Wednesday, June 11, 2008 3:26 PM >> To: Xen-Devel (E-mail) >> Subject: [Xen-devel] Fatal page fault on xen-unstable tip >> (vt-d related >> maybe?) >>=20 >>=20 >> I just updated to a more recent unstable (17826) and got this >> fatal page fault on boot. I backed off to 17772 (which I'm told >> was good on another machine) and got the same. I suspect it >> is something peculiar to my hardware -- an Intel "Weybridge" >> box -- which I think has VT-d support though I am not currently >> using it. >>=20 >> I can't binary search the exact cset right now, but thought >> I'd report it. >>=20 >> Dan >>=20 >> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D >>=20 >>=20 >> \ \/ /___ _ __ |___ / |___ / _ _ _ __ ___| |_ __ _| >> |__ | | ___ >>=20 >> \ // _ \ '_ \ |_ \ |_ \ __| | | | '_ \/ __| __/ _` | >> '_ \| |/ _ \ >>=20 >> / \ __/ | | | ___) | ___) |__| |_| | | | \__ \ || (_| | >> |_) | | __/ >>=20 >> /_/\_\___|_| |_| |____(_)____/ \__,_|_| >> |_|___/\__\__,_|_.__/|_|\___| >>=20 >>=20 >> (XEN) Xen version 3.3-unstable (root@(none)) (gcc version >> 4.1.1 20070105 (Red Ha >>=20 >> t 4.1.1-52)) Wed Jun 11 14:02:26 MDT 2008 >>=20 >> (XEN) Latest ChangeSet: Mon Jun 02 16:05:07 2008 +0100 >> 17772:fe60bf79d96f >>=20 >> (XEN) Command line: com1=3D115200,8n1 console=3Dcom1,vga noreboot >>=20 >> (XEN) Video information: >> (XEN) VGA is text mode 80x25, font 8x16 >>=20 >> (XEN) VBE/DDC methods: V2; EDID transfer time: 1 seconds >>=20 >> (XEN) Disc information: >> (XEN) Found 1 MBR signatures >> (XEN) Found 1 EDD information structures >>=20 >> (XEN) Xen-e820 RAM map: >> (XEN) 0000000000000000 - 000000000009cc00 (usable) >>=20 >> (XEN) 000000000009cc00 - 00000000000a0000 (reserved) >>=20 >> (XEN) 00000000000e0000 - 0000000000100000 (reserved) >>=20 >> (XEN) 0000000000100000 - 000000007d5b0000 (usable) >>=20 >> (XEN) 000000007d5b0000 - 000000007d5be000 (ACPI data) >>=20 >> (XEN) 000000007d5be000 - 000000 >> (XEN) 000000007d5f0000 - 000000007d600000 >> (reserved)nput/input2smp_apic_timer_i >> (XEN) 00000000fed90000 - 00000000fed94000 (reserved) >> Sta >> pnp: >> ACPI: LAPIC (acpi_id[0x04] lapic_id[ >> (XEN) 00000000fee00000 - 00000000fee01000 (reserved) ] >>=20 >> (XEN) 00000000ffa6f000 - 00000000ffa70000 (reserved) >> ACPI: IOAPIC (id[0x02] address[0xfec00000] gsi_b >> (XEN) 00000000ffb00000 - 0000000100000000 (reserved) >> echo >> Q >> usbcore: registered new driver h >> (XEN) System RAM: 2005MB (2053424kB) >> IOAPIC[0]: a >> (XEN) Xen heap: 9MB (10012kB)0xfec00000, GSI 0-23 >> (XEN) Domain heap initialised: DMA width 32 bits >> uhci_hcd 0000:00:1 >> PCI: I >> (XEN) Processor #0 6:15 APIC version 20 >>=20 >> ACPI: >> (XEN) Processor #1 6:15 APIC version 202 dfl dfl)] on >> usb-0000:00:1PU: >> (XEN) IOAPIC[0]: apic_id 2, version 32, address 0xf >>=20 >> (XEN) Brought up 2 CPUs00 (virtual 0xf8800000) >> (XEN) [VT-D]iommu.c:1700: Queued Invalidation hardware not >> found >> (XEN) [VT-D]iommu.c:1700: Queued Invalidation hardware not >> found >> hpet0: 4 64-bit timers, 14318180 Hzs SCSI revision: 05tr >> (XEN) [VT-D]iommu.c:1700: Queued Invalidation hardware not >> foundly >>=20 >> Using HPET for base-timercated, 0from 1 choice >> (XEN) [VT-D]iommu.c:1700: Queued Invalidation hardware not >> foundg timer specific routine.. 5990.25 BogoMIPS (lpj=3D2995129)cache h >> (XEN) [VT-D]iommu.c:1708: Interrupt Remapping hardware not >> founding >> (XEN) [VT-D]iommu.c:1708: Interrupt Remapping hardware not >> found >> usb >> Security Framework v1.0.0 in >> (XEN) [VT-D]iommu.c:1708: Interrupt Remapping hardware not >> found >> SELinux: >> (XEN) [VT-D]iommu.c:1708: Interrupt R >>=20 >> 20060707]-0:1.0: 2 >> ACPI: >> Intel mac >> (XEN) [VT-D]iommu.c:720: iommu_fault:DMA Write: 0:2.0 addr >> f1bffb000 REASON 5 io >>=20 >> ACPI (excon >>=20 >> mmu->reg =3D fff77000reporting enabled o >> (XEN) print_vtd_entries: iommu =3D ff1e7c00 bdf =3D 0:2:0 gmfn =3D >> f1bffb >>=20 >> (XEN) root_entry =3D fee17000uction... OK.1 choicetm2 cx16 x >> (XEN) root_entry[0] =3D 7c400001 >> 20060707] >> (XEN) context =3D fee54000hing to UP code C2[C2])CI Ho >> (XEN) context[10] =3D 101_7c9c5001 >>=20 >> (XEN) EFLAGS: 00010286 CONTEXT: hypervisor >> ENABLING IO-APIC IRQsx3f8 (irq =3D 4) i >> (XEN) eax: 00000000 ebx: 7b9c51e0 ecx: 0000000a edx: >> 00000000 apic1=3D0 pin1=3D2 apic2=3D-1 pin2=3D-100:07: ttyS1 at I/O >> 0x2f8 (irq =3D 3) >> (XEN) esi: 00000003 edi: 0000003c ebp: fee54000 esp: >> ff1efce0 >> checking TSC >> (XEN) cr0: 8005003b cr4: 000026f0 cr3: 00100120 cr2: >> 7b9c51e0GSI 17 (level, low) - >>=20 >> (XEN) ds: e010 es: e010 fs: e010 gs: e010 ss: e010 >> cs: e008CPUss: mounted filesy >> migration_cos >> (XEN) fee54000 00000001 156db6e3 ff1e7c00 00000005 >> 00f1bffb 00000000 ff12160b >> ACPI: bus type pci registeredE driver Revis >>=20 >> (XEN) ff1e7c00 00000000 00000010 00f1bffb 00000000 >> 00000002 00000000 1bffb000S Bug: MCFG area at e0000000 is not >> E820-reservedde: Assuming 33MHz system bus s >>=20 >> (XEN) 0000000f 00000005 fff77000 ff12363a ff23af00 >> 00000001 00000010 ff1e7c00 >>=20 >> PCI: Not using MMCONF >>=20 >> (XEN) 00000000 80000005 00000001 00000000 00000010 >> 00000200 0000 >>=20 >>=20 >> (XEN) ff1efde4 510cb033 00000000 fff77000 00000000 >> ff1e7c00 00000282 00000000 >> ACPI: PCI Root Bridge [PCI0] (0000: >>=20 >> (XEN) 00a00000 ff120f38 0000e008 00000282 ff1e7c00 >> ff198892 00000298 fff77000evice co >> PCI: Transparent bridge - 0000:00:1e.0 >>=20 >> (XEN) 00000000 00000000 10000000 ff23bb80 000000e0 >> 00000008 00000100 ff123b62 >> ACPI: PCI Interrupt Link [LNKA] (IRQs 3 4 5 6 7 *10 12 14 15)orts dete >>=20 >> (XEN) 80ffff00 00000000 00000004 00000000 ffbec718 >> ff19ab83 00000000 ff1e4bb4 >> md: bitmap version 4.39 jbd.ko mod >>=20 >>=20 >> (XEN) 00000001 00000003 00000000 ff10437c ff1e4080 >> 00000000 00000000 ff16b9e7 >> Loading libata.ko mod >>=20 >>=20 >>=20 >>=20 >>=20 >>=20 >>=20 >> (XEN) 00000003 ff19acd4 00000000 ff1ca144 00000000 >> 00000001 00000510 ff1c0ac8 >> NetLabel: Initializingbus registered, assign >>=20 >>=20 >> (XEN) 00000000 00000000 00000000 00000000 00000000 >> 00000000 fffffffb ff1b3070 >>=20 >> NetLabel: pro >>=20 >> (XEN) ff1cb1f0 ff1b2670 00021a00 00000000 00000000 >> 00000000 ff02cc80 ff021a00 >>=20 >>=20 >>=20 >> (XEN) [] iommu_enable_translation+0xa8/0xc0 >> (XEN) [] intel_iommu_domain_init+0x202/0x290 >> (XEN) [] arch_domain_create+0x1af/0x2a >=20 >=20 >> (XEN) [] evtchn_init+0x27/0x40 >> (XEN) [] domain_create+0x11c/0x250 >> (XEN) [] __hvm_register_PIC_save_and_restore+0x37/0x40 >> (XEN) [] __start_xen+0xcd8/0x1380 >> (XEN) [] e820nr+0x0/0x4 >> (XEN) [] e820map+0x0/0xa00 >> (XEN) [] __high_start+0x58/0x5a >> (XEN) >> (XEN) Pagetable walk from 7b9c51e0: >> (XEN) L3[0x001] =3D 00000000001f5001 55555555 >> (XEN) L2[0x1dc] =3D 0000000000000000 ffffffff >> (XEN) >> (XEN) **************************************** >> (XEN) Panic on CPU 0: >> (XEN) FATAL PAGE FAULT >> (XEN) [error_code=3D0000] >> (XEN) Faulting linear address: 7b9c51e0 >> (XEN) **************************************** >> (XEN) >> (XEN) Manual reset required ('noreboot' specified) >>=20 >=20 > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xensource.com > http://lists.xensource.com/xen-devel >=20 >=20 >=20 >=20 > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xensource.com > http://lists.xensource.com/xen-devel