LinuxLists.cc - kexec + USB storage in 2.6.19

2007-01-12 12:25:07

Subject: kexec + USB storage in 2.6.19

Hello,

After upgrading from 2.6.18.3 to 2.6.19.2 on an x86_64 machine I noticed
that the EHCI USB host is unable to work properly after a kexec invocation.
This makes it impossible to mount the rootfs in the configuration I'm using.

According to the prints, the irq changes from 23 to 10.

NOTE: Since the device is already connected at boot, I've added a patch
that disables the scanning delay for the first detected device, in order
to shorten the time it takes for the boot process. It worked on 2.6.18.3,
so I wonder what has changed...

[ 92.426719] ehci_hcd 0000:00:1d.7: EHCI Host Controller
[ 92.432008] ehci_hcd 0000:00:1d.7: new USB bus registered, assigned bus number 1
[ 92.439422] ehci_hcd 0000:00:1d.7: debug port 1
[ 92.443961] ehci_hcd 0000:00:1d.7: irq 10, io mem 0xdc001000
[ 92.453472] ehci_hcd 0000:00:1d.7: USB 2.0 started, EHCI 1.00, driver 10 Dec 2004
[ 92.461034] usb usb1: configuration #1 chosen from 1 choice
[ 92.466625] hub 1-0:1.0: USB hub found
[ 92.470373] hub 1-0:1.0: 8 ports detected
[ 92.581134] Initializing USB Mass Storage driver...
[ 92.828571] usb 1-1: new high speed USB device using ehci_hcd and address 2
[ 93.834658] ehci_hcd 0000:00:1d.7: Unlink after no-IRQ? Controller is probably using the wrong IRQ.
[ 104.374665] usb 1-1: device not accepting address 2, error -110
[ 104.494441] usb 1-1: new high speed USB device using ehci_hcd and address 3
[ 116.024567] usb 1-1: device not accepting address 3, error -110
[ 116.144343] usb 1-1: new high speed USB device using ehci_hcd and address 4
[ 126.556589] usb 1-1: device not accepting address 4, error -110
[ 126.676365] usb 1-1: new high speed USB device using ehci_hcd and address 5
[ 137.088612] usb 1-1: device not accepting address 5, error -110
[ 137.336145] usb 1-2: new high speed USB device using ehci_hcd and address 6
[ 148.866271] usb 1-2: device not accepting address 6, error -110
[ 148.982055] usb 1-2: new high speed USB device using ehci_hcd and address 7
[ 160.512181] usb 1-2: device not accepting address 7, error -110
[ 160.627965] usb 1-2: new high speed USB device using ehci_hcd and address 8
[ 171.040211] usb 1-2: device not accepting address 8, error -110
[ 171.155995] usb 1-2: new high speed USB device using ehci_hcd and address 9
[ 181.568241] usb 1-2: device not accepting address 9, error -110
[ 181.574184] usbcore: registered new interface driver usb-storage
[ 181.580182] USB Mass Storage support registered.

A normal boot works properly:

[ 78.139976] ACPI: PCI Interrupt 0000:00:1d.7[D] -> GSI 23 (level, low) -> IRQ 23
[ 78.148121] ehci_hcd 0000:00:1d.7: EHCI Host Controller
[ 78.153411] ehci_hcd 0000:00:1d.7: new USB bus registered, assigned bus number 1
[ 78.160821] ehci_hcd 0000:00:1d.7: debug port 1
[ 78.165362] ehci_hcd 0000:00:1d.7: irq 23, io mem 0xdc001000
[ 78.174871] ehci_hcd 0000:00:1d.7: USB 2.0 started, EHCI 1.00, driver 10 Dec 2004
[ 78.182454] usb usb1: configuration #1 chosen from 1 choice
[ 78.188045] hub 1-0:1.0: USB hub found
[ 78.191791] hub 1-0:1.0: 8 ports detected
[ 78.303392] Initializing USB Mass Storage driver...
[ 78.546834] usb 1-1: new high speed USB device using ehci_hcd and address 2
[ 78.692172] usb 1-1: configuration #1 chosen from 1 choice
[ 78.942075] usb 1-2: new high speed USB device using ehci_hcd and address 3
[ 79.083324] usb 1-2: configuration #1 chosen from 1 choice
[ 79.089135] scsi0 : SCSI emulation for USB Mass Storage devices
[ 79.095172] scsi1 : SCSI emulation for USB Mass Storage devices
[ 79.101184] usbcore: registered new interface driver usb-storage

- Dan

2007-01-12 14:05:42

by Eric W. Biederman

[permalink] [raw]

Subject: Re: kexec + USB storage in 2.6.19

Dan Aloni <[email protected]> writes:

> Hello,
>
> After upgrading from 2.6.18.3 to 2.6.19.2 on an x86_64 machine I noticed
> that the EHCI USB host is unable to work properly after a kexec invocation.
> This makes it impossible to mount the rootfs in the configuration I'm using.
>
> According to the prints, the irq changes from 23 to 10.
>
> NOTE: Since the device is already connected at boot, I've added a patch
> that disables the scanning delay for the first detected device, in order
> to shorten the time it takes for the boot process. It worked on 2.6.18.3,
> so I wonder what has changed...

At first glance it looks like acpi didn't come on in the kexec'd kernel.
Do you see anything like the line below.
> [ 78.139976] ACPI: PCI Interrupt 0000:00:1d.7[D] -> GSI 23 (level, low) -> IRQ

Could your provide the full bootlogs instead of these partial ones?
There is enough context missing I don't think anyone can do more than
agree you are seeing a problem.

Eric

2007-01-12 14:56:40

by Dan Aloni

[permalink] [raw]

Subject: Re: kexec + USB storage in 2.6.19

On Fri, Jan 12, 2007 at 07:05:09AM -0700, Eric W. Biederman wrote:
> Dan Aloni <[email protected]> writes:
>
> > Hello,
> >
> > After upgrading from 2.6.18.3 to 2.6.19.2 on an x86_64 machine I noticed
> > that the EHCI USB host is unable to work properly after a kexec invocation.
> > This makes it impossible to mount the rootfs in the configuration I'm using.
> >
> > According to the prints, the irq changes from 23 to 10.
> >
> > NOTE: Since the device is already connected at boot, I've added a patch
> > that disables the scanning delay for the first detected device, in order
> > to shorten the time it takes for the boot process. It worked on 2.6.18.3,
> > so I wonder what has changed...
>
> At first glance it looks like acpi didn't come on in the kexec'd kernel.
> Do you see anything like the line below.
> > [ 78.139976] ACPI: PCI Interrupt 0000:00:1d.7[D] -> GSI 23 (level, low) -> IRQ
>
> Could your provide the full bootlogs instead of these partial ones?
> There is enough context missing I don't think anyone can do more than
> agree you are seeing a problem.

Looks like you were right about the ACPI being disabled during boot.

I'm attaching the full logs.

Note that there are a few additions to the kernel, some are known to
everyone (kdb), and the others are self-bred facilities for tracing,
logging across kexec-invocations, and specialized memory pools
(respectively, dandalf, permamem, prebb), but these are not related
to the problem that we see here.

Just for the record, I also used a modified version of gcc 4.1.1 that
adds a switch (see the 'Re: [PATCH] add -foverride-comp-dir' thread
on gcc-devel) - it should also be completely disconnected from this
problem as well.

This kernel was also patched for retrying to mount the root dev with
a timeout, also to reduce execution time.

Sorry that I can't test with the vanilla kernel - it's a specialized
environment. All changes are available under GPL on request.

[ 8655.114721] Starting new kernel
[ 0.000000] Linux version 2.6.19.2-xiv-34-x86_64-misc (dan@pro135) (gcc version 4.1.1 (xiv)) #4 SMP Fri Jan 12 14:03:34 IST 2007
[ 0.000000] Command line: ro root=/dev/sda1 rootdelay=10 init=/xiv/system/xinit console=ttyS0,115200
[ 0.000000] BIOS-provided physical RAM map:
[ 0.000000] BIOS-e820: 0000000000000100 - 000000000009b800 (usable)
[ 0.000000] BIOS-e820: 000000000009b800 - 00000000000a0000 (reserved)
[ 0.000000] BIOS-e820: 0000000000100000 - 00000000cff70000 (usable)
[ 0.000000] BIOS-e820: 00000000cff70000 - 00000000cff78000 (ACPI data)
[ 0.000000] BIOS-e820: 00000000cff78000 - 00000000cff80000 (ACPI NVS)
[ 0.000000] BIOS-e820: 00000000cff80000 - 00000000d0000000 (reserved)
[ 0.000000] BIOS-e820: 00000000e0000000 - 00000000f0000000 (reserved)
[ 0.000000] BIOS-e820: 00000000fec00000 - 00000000fec10000 (reserved)
[ 0.000000] BIOS-e820: 00000000fee00000 - 00000000fee01000 (reserved)
[ 0.000000] BIOS-e820: 00000000ff800000 - 00000000ffc00000 (reserved)
[ 0.000000] BIOS-e820: 00000000fffffc00 - 0000000100000000 (reserved)
[ 0.000000] BIOS-e820: 0000000100000000 - 0000000130000000 (usable)
[ 0.000000] end_pfn_map = 1245184
[ 0.000000] DMI present.
[ 0.000000] Zone PFN ranges:
[ 0.000000] DMA 1 -> 4096
[ 0.000000] DMA32 4096 -> 1048576
[ 0.000000] Normal 1048576 -> 1245184
[ 0.000000] early_node_map[3] active PFN ranges
[ 0.000000] 0: 1 -> 155
[ 0.000000] 0: 256 -> 851824
[ 0.000000] 0: 1048576 -> 1245184
[ 0.000000] ACPI: PM-Timer IO Port: 0x1008
[ 0.000000] ACPI: LAPIC (acpi_id[0x00] lapic_id[0x00] enabled)
[ 0.000000] Processor #0 (Bootup-CPU)
[ 0.000000] ACPI: LAPIC (acpi_id[0x01] lapic_id[0x06] enabled)
[ 0.000000] Processor #6
[ 0.000000] ACPI: LAPIC (acpi_id[0x02] lapic_id[0x01] enabled)
[ 0.000000] Processor #1
[ 0.000000] WARNING: NR_CPUS limit of 2 reached. Processor ignored.
[ 0.000000] ACPI: LAPIC (acpi_id[0x03] lapic_id[0x07] enabled)
[ 0.000000] Processor #7
[ 0.000000] WARNING: NR_CPUS limit of 2 reached. Processor ignored.
[ 0.000000] ACPI: LAPIC_NMI (acpi_id[0x00] high edge lint[0x1])
[ 0.000000] ACPI: LAPIC_NMI (acpi_id[0x01] high edge lint[0x1])
[ 0.000000] ACPI: LAPIC_NMI (acpi_id[0x02] high edge lint[0x1])
[ 0.000000] ACPI: LAPIC_NMI (acpi_id[0x03] high edge lint[0x1])
[ 0.000000] ACPI: IOAPIC (id[0x02] address[0xfec00000] gsi_base[0])
[ 0.000000] IOAPIC[0]: apic_id 2, address 0xfec00000, GSI 0-23
[ 0.000000] ACPI: IOAPIC (id[0x03] address[0xfec80000] gsi_base[24])
[ 0.000000] IOAPIC[1]: apic_id 3, address 0xfec80000, GSI 24-47
[ 0.000000] ACPI: IOAPIC (id[0x04] address[0xfec80400] gsi_base[48])
[ 0.000000] IOAPIC[2]: apic_id 4, address 0xfec80400, GSI 48-71
[ 0.000000] ACPI: IOAPIC (id[0x05] address[0xfec84000] gsi_base[72])
[ 0.000000] IOAPIC[3]: apic_id 5, address 0xfec84000, GSI 72-95
[ 0.000000] ACPI: IOAPIC (id[0x08] address[0xfec84400] gsi_base[96])
[ 0.000000] IOAPIC[4]: apic_id 8, address 0xfec84400, GSI 96-119
[ 0.000000] ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 high edge)
[ 0.000000] ACPI: INT_SRC_OVR (bus 0 bus_irq 9 global_irq 9 high level)
[ 0.000000] Setting APIC routing to flat
[ 0.000000] Using ACPI (MADT) for SMP configuration information
[ 0.000000] Nosave address range: 000000000009b000 - 000000000009c000
[ 0.000000] Nosave address range: 000000000009c000 - 00000000000a0000
[ 0.000000] Nosave address range: 00000000000a0000 - 0000000000100000
[ 0.000000] Nosave address range: 00000000cff70000 - 00000000cff78000
[ 0.000000] Nosave address range: 00000000cff78000 - 00000000cff80000
[ 0.000000] Nosave address range: 00000000cff80000 - 00000000d0000000
[ 0.000000] Nosave address range: 00000000d0000000 - 00000000e0000000
[ 0.000000] Nosave address range: 00000000e0000000 - 00000000f0000000
[ 0.000000] Nosave address range: 00000000f0000000 - 00000000fec00000
[ 0.000000] Nosave address range: 00000000fec00000 - 00000000fec10000
[ 0.000000] Nosave address range: 00000000fec10000 - 00000000fee00000
[ 0.000000] Nosave address range: 00000000fee00000 - 00000000fee01000
[ 0.000000] Nosave address range: 00000000fee01000 - 00000000ff800000
[ 0.000000] Nosave address range: 00000000ff800000 - 00000000ffc00000
[ 0.000000] Nosave address range: 00000000ffc00000 - 0000000100000000
[ 0.000000] Allocating PCI resources starting at d1000000 (gap: d0000000:10000000)
[ 0.000000] PERCPU: Allocating 34176 bytes of per cpu data
[ 0.000000] Built 1 zonelists. Total pages: 1029907
[ 0.000000] Kernel command line: ro root=/dev/sda1 rootdelay=10 init=/xiv/system/xinit console=ttyS0,115200
[ 0.000000] Initializing CPU#0
[ 0.000000] PID hash table entries: 4096 (order: 12, 32768 bytes)
[ 8655.473910] Console: colour VGA+ 80x25
[ 8655.911117] Dentry cache hash table entries: 524288 (order: 10, 4194304 bytes)
[ 8655.921834] Inode-cache hash table entries: 262144 (order: 9, 2097152 bytes)
[ 8655.929466] Checking aperture...
[ 8655.932837] PCI-DMA: Using software bounce buffering for IO (SWIOTLB)
[ 8655.978621] Placing software IO TLB between 0x5900000 - 0x9900000
[ 8655.984713] permamem: initializing
[ 8655.988113] permamem: allocating dandalf at 0000000020000000 with 48 MBs, virt ffff810020000000, page ffff810001700000
[ 8656.000350] prebb: allocation for objsize 1048576 (2 objects) succeeded
[ 8656.013001] prebb: allocation for objsize 8192 (1280 objects) succeeded
[ 8656.019631] prebb: activated with 2 prebb pools
[ 8656.062066] Memory: 3985264k/4980736k available (2867k kernel code, 207768k reserved, 1839k data, 264k init)
[ 8656.149100] Calibrating delay using timer specific routine.. 5604.62 BogoMIPS (lpj=11209254)
[ 8656.178702] kdb version 4.4 by Keith Owens, Scott Lurndal. Copyright SGI, All Rights Reserved
kdb_cmd[0]: defcmd archkdb "" "First line arch debugging"
kdb_cmd[7]: defcmd archkdbcpu "" "archkdb with only tasks on cpus"
kdb_cmd[14]: defcmd archkdbshort "" "archkdb with less detailed backtrace"
kdb_cmd[21]: defcmd archkdbcommon "" "Common arch debugging"
[ 8656.210293] dandalf: partition entries size 15442 KB
[ 8656.215252] dandalf: partition vardata size 3676 KB
[ 8656.220121] dandalf: partition logptrs size 2451 KB
[ 8656.224990] dandalf: partition modentries size 490 KB
[ 8656.230032] dandalf: partition modvardata size 2451 KB
[ 8656.235340] dandalf: initialization succeeded
[ 8656.239746] Mount-cache hash table entries: 256
[ 8656.244416] CPU: Trace cache: 12K uops, L1 D cache: 16K
[ 8656.249656] CPU: L2 cache: 2048K
[ 8656.252876] using mwait in idle threads.
[ 8656.256791] CPU: Physical Processor ID: 0
[ 8656.260794] CPU: Processor Core ID: 0
[ 8656.264477] Freeing SMP alternatives: 24k freed
[ 8656.269024] ACPI: Core revision 20060707
[ 8656.272980] ACPI Error (tbxfroot-0512): Could not map memory at 0000040E for length 2 [20060707]
[ 8656.281789] ACPI Exception (tbxfroot-0400): AE_NO_MEMORY, RSDP structure not found - Flags=8 [20060707]
[ 8656.291201] ACPI: System description tables not found
[ 8656.296239] ACPI Exception (tbxface-0076): AE_NOT_FOUND, Could not get the RSDP [20060707]
[ 8656.304521] ACPI Exception (tbxface-0120): AE_NOT_FOUND, Could not load tables [20060707]
[ 8656.312711] ACPI: Unable to load the System Description Tables
[ 8656.359049] Using local APIC timer interrupts.
[ 8656.399123] result 12500612
[ 8656.401912] Detected 12.500 MHz APIC timer.
[ 8656.408772] Booting processor 1/2 APIC 0x6
[ 8656.423423] Initializing CPU#1
[ 8656.500442] Calibrating delay using timer specific routine.. 5600.74 BogoMIPS (lpj=11201480)
[ 8656.500456] CPU: Trace cache: 12K uops, L1 D cache: 16K
[ 8656.500460] CPU: L2 cache: 2048K
[ 8656.500464] CPU: Physical Processor ID: 3
[ 8656.500466] CPU: Processor Core ID: 0
[ 8656.500478] CPU1: Thermal monitoring enabled (TM1)
[ 8656.500907] Intel(R) Xeon(TM) CPU 2.80GHz stepping 03
[ 8656.504479] Brought up 2 CPUs
[ 8656.546386] testing NMI watchdog ... OK.
[ 8656.590251] time.c: Using 3.579545 MHz WALL PM GTOD PIT/TSC timer.
[ 8656.596421] time.c: Detected 2800.161 MHz processor.
[ 8656.977181] migration_cost=1789
[ 8656.980875] NET: Registered protocol family 16
[ 8656.989756] PCI: Using MMCONFIG at e0000000
[ 8656.994212] PCI: No mmconfig possible on device 08:01
[ 8657.000006] ACPI: Interpreter disabled.
[ 8657.003847] Linux Plug and Play Support v0.97 (c) Adam Belay
[ 8657.009501] pnp: PnP ACPI: disabled
[ 8657.018700] SCSI subsystem initialized
[ 8657.022518] usbcore: registered new interface driver usbfs
[ 8657.028018] usbcore: registered new interface driver hub
[ 8657.033352] usbcore: registered new device driver usb
[ 8657.038431] PCI: Probing PCI hardware
[ 8657.042503] PCI quirk: region 1000-107f claimed by ICH4 ACPI/GPIO/TCO
[ 8657.048928] PCI quirk: region 1180-11bf claimed by ICH4 GPIO
[ 8657.054601] PCI: Ignoring BAR0-3 of IDE controller 0000:00:1f.2
[ 8657.060578] PCI: PXH quirk detected, disabling MSI for SHPC device
[ 8657.066784] PCI: PXH quirk detected, disabling MSI for SHPC device
[ 8657.073305] PCI: PXH quirk detected, disabling MSI for SHPC device
[ 8657.079518] PCI: PXH quirk detected, disabling MSI for SHPC device
[ 8657.086047] PCI: Transparent bridge - 0000:00:1e.0
[ 8657.091757] PCI: Using IRQ router PIIX/ICH [8086/24d0] at 0000:00:1f.0
[ 8657.098423] PCI-GART: No AMD northbridge found.
[ 8657.103100] PCI: Bridge: 0000:01:00.0
[ 8657.106754] IO window: 2000-2fff
[ 8657.110155] MEM window: dc200000-dc2fffff
[ 8657.114330] PREFETCH window: d1800000-d18fffff
[ 8657.118941] PCI: Bridge: 0000:01:00.2
[ 8657.122597] IO window: 3000-3fff
[ 8657.126000] MEM window: dc300000-dc3fffff
[ 8657.130175] PREFETCH window: disabled.
[ 8657.134095] PCI: Bridge: 0000:00:02.0
[ 8657.137752] IO window: 2000-3fff
[ 8657.141156] MEM window: dc100000-dc3fffff
[ 8657.145331] PREFETCH window: d1800000-d18fffff
[ 8657.149943] PCI: Bridge: 0000:00:04.0
[ 8657.153597] IO window: disabled.
[ 8657.156992] MEM window: disabled.
[ 8657.160476] PREFETCH window: disabled.
[ 8657.164398] PCI: Bridge: 0000:05:00.0
[ 8657.168052] IO window: 4000-4fff
[ 8657.171454] MEM window: dc500000-dcbfffff
[ 8657.175629] PREFETCH window: d1000000-d13fffff
[ 8657.180240] PCI: Bridge: 0000:05:00.2
[ 8657.183897] IO window: 5000-5fff
[ 8657.187299] MEM window: dcc00000-dd0fffff
[ 8657.191477] PREFETCH window: d1400000-d17fffff
[ 8657.196086] PCI: Bridge: 0000:00:06.0
[ 8657.199743] IO window: 4000-5fff
[ 8657.203145] MEM window: dc400000-dd0fffff
[ 8657.207320] PREFETCH window: d1000000-d17fffff
[ 8657.211935] PCI: Bridge: 0000:00:1e.0
[ 8657.215589] IO window: 6000-6fff
[ 8657.218988] MEM window: dd100000-deffffff
[ 8657.223169] PREFETCH window: d1900000-d19fffff
[ 8657.227903] NET: Registered protocol family 2
[ 8657.267053] IP route cache hash table entries: 131072 (order: 8, 1048576 bytes)
[ 8657.274895] TCP established hash table entries: 131072 (order: 10, 4194304 bytes)
[ 8657.284596] TCP bind hash table entries: 65536 (order: 9, 2097152 bytes)
[ 8657.292489] TCP: Hash tables configured (established 131072 bind 65536)
[ 8657.299102] TCP reno registered
[ 8657.302360] Simple Boot Flag at 0x39 set to 0x80
[ 8657.307636] JFFS2 version 2.2. (NAND) (C) 2001-2006 Red Hat, Inc.
[ 8657.313794] SGI XFS with large block/inode numbers, no debug enabled
[ 8657.320360] io scheduler noop registered
[ 8657.324297] io scheduler anticipatory registered (default)
[ 8657.329806] io scheduler deadline registered
[ 8657.334104] io scheduler cfq registered
[ 8657.354521] Real Time Clock Driver v1.12ac
[ 8657.358616] Linux agpgart interface v0.101 (c) Dave Jones
[ 8657.364003] Serial: 8250/16550 driver $Revision: 1.90 $ 4 ports, IRQ sharing disabled
[ 8657.371908] serial8250: ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A
[ 8657.378011] serial8250: ttyS1 at I/O 0x2f8 (irq = 3) is a 16550A
[ 8657.384307] Floppy drive(s): fd0 is 1.44M
[ 8657.408129] FDC 0 is a National Semiconductor PC87306
[ 8657.414492] RAMDISK driver initialized: 16 RAM disks of 80000K size 1024 blocksize
[ 8657.422224] loop: loaded (max 8 devices)
[ 8657.426252] Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2
[ 8657.432595] ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
[ 8657.440617] ICH5-SATA: IDE controller at PCI slot 0000:00:1f.2
[ 8657.446452] ICH5-SATA: chipset revision 2
[ 8657.450454] ICH5-SATA: not 100% native mode: will probe irqs later
[ 8657.456625] ide0: BM-DMA at 0x14b0-0x14b7, BIOS settings: hda:DMA, hdb:pio
[ 8657.463883] ide1: BM-DMA at 0x14b8-0x14bf, BIOS settings: hdc:pio, hdd:pio
[ 8657.762190] hda: HDS725050KLA360, ATA DISK drive
[ 8658.440928] ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
[ 8659.586567] hda: max request size: 1024KiB
[ 8659.596977] hda: 976773168 sectors (500107 MB) w/15607KiB Cache, CHS=60801/255/63, UDMA(33)
[ 8659.605577] hda: cache flushes supported
[ 8659.609553] hda: hda1
[ 8659.617253] block2mtd: version $Revision: 1.30 $
[ 8659.621859] usbmon: debugfs is not available
[ 8659.626919] ehci_hcd 0000:00:1d.7: EHCI Host Controller
[ 8659.632208] ehci_hcd 0000:00:1d.7: new USB bus registered, assigned bus number 1
[ 8659.639615] ehci_hcd 0000:00:1d.7: debug port 1
[ 8659.644155] ehci_hcd 0000:00:1d.7: irq 10, io mem 0xdc001000
[ 8659.653678] ehci_hcd 0000:00:1d.7: USB 2.0 started, EHCI 1.00, driver 10 Dec 2004
[ 8659.661254] usb usb1: configuration #1 chosen from 1 choice
[ 8659.666838] hub 1-0:1.0: USB hub found
[ 8659.670587] hub 1-0:1.0: 8 ports detected
[ 8659.782260] Initializing USB Mass Storage driver...
[ 8659.787138] usbcore: registered new interface driver usb-storage
[ 8659.793137] USB Mass Storage support registered.
[ 8659.797792] PNP: No PS/2 controller found. Probing ports directly.
[ 8659.806486] serio: i8042 KBD port at 0x60,0x64 irq 1
[ 8659.811448] serio: i8042 AUX port at 0x60,0x64 irq 12
[ 8659.816506] mice: PS/2 mouse device common for all mice
[ 8659.821727] EDAC MC: Ver: 2.0.1 Jan 12 2007
[ 8659.826054] EDAC e752x: tolm = d0000, remapbase = 100000, remaplimit = 12c000
[ 8659.833227] EDAC MC0: Giving out device to e752x_edac E7520: DEV 0000:00:00.0
[ 8659.840385] EDAC e752x: Non-Fatal Error PCI Express B
[ 8659.845431] EDAC e752x: Non-Fatal Error PCI Express B
[ 8659.845443] TCP cubic registered
[ 8659.853688] NET: Registered protocol family 1
[ 8659.858043] NET: Registered protocol family 17
[ 8659.862480] 802.1Q VLAN Support v1.8 Ben Greear <[email protected]>
[ 8659.869253] All bugs added by David S. Miller <[email protected]>
[ 8659.981933] VFS: retrying root mount
[ 8660.033685] usb 1-1: new high speed USB device using ehci_hcd and address 2
[ 8660.484933] VFS: retrying root mount
[ 8660.987963] VFS: retrying root mount
[ 8661.039752] ehci_hcd 0000:00:1d.7: Unlink after no-IRQ? Controller is probably using the wrong IRQ.
[ 8661.487010] VFS: retrying root mount
[ 8661.990045] VFS: retrying root mount
[ 8662.493080] VFS: retrying root mount
[ 8662.996115] VFS: retrying root mount
[ 8663.499152] VFS: retrying root mount
[ 8664.002186] VFS: retrying root mount
[ 8664.505221] VFS: retrying root mount
[ 8665.008258] VFS: retrying root mount
[ 8665.511294] VFS: retrying root mount
[ 8666.014330] VFS: retrying root mount
[ 8666.517366] VFS: retrying root mount
[ 8667.020401] VFS: retrying root mount
[ 8667.523439] VFS: retrying root mount
[ 8668.026473] VFS: retrying root mount
[ 8668.529508] VFS: retrying root mount
[ 8669.032544] VFS: retrying root mount
[ 8669.535580] VFS: retrying root mount
[ 8670.038617] VFS: Cannot open root device "sda1" or unknown-block(0,0)
[ 8670.045045] Please append a correct "root=" boot option
[ 8670.050259] Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(0,0)
[ 8670.058502]
Entering kdb (current=0xffff81012fcca7a0, pid 1) on processor 1 due to KDB_ENTER()
[1]kdb>

----------------

A normal boot log:

[ 0.000000] Linux version 2.6.19.2-xiv-34-x86_64-misc (dan@pro135) (gcc version 4.1.1 (xiv)) #4 SMP Fri Jan 12 14:03:34 IST 2007
[ 0.000000] Command line: ro root=/dev/sda1 rootdelay=10 init=/xiv/system/xinit console=ttyS0,115200
[ 0.000000] BIOS-provided physical RAM map:
[ 0.000000] BIOS-e820: 0000000000000000 - 000000000009b800 (usable)
[ 0.000000] BIOS-e820: 000000000009b800 - 00000000000a0000 (reserved)
[ 0.000000] BIOS-e820: 00000000000e4000 - 0000000000100000 (reserved)
[ 0.000000] BIOS-e820: 0000000000100000 - 00000000cff70000 (usable)
[ 0.000000] BIOS-e820: 00000000cff70000 - 00000000cff78000 (ACPI data)
[ 0.000000] BIOS-e820: 00000000cff78000 - 00000000cff80000 (ACPI NVS)
[ 0.000000] BIOS-e820: 00000000cff80000 - 00000000d0000000 (reserved)
[ 0.000000] BIOS-e820: 00000000e0000000 - 00000000f0000000 (reserved)
[ 0.000000] BIOS-e820: 00000000fec00000 - 00000000fec10000 (reserved)
[ 0.000000] BIOS-e820: 00000000fee00000 - 00000000fee01000 (reserved)
[ 0.000000] BIOS-e820: 00000000ff800000 - 00000000ffc00000 (reserved)
[ 0.000000] BIOS-e820: 00000000fffffc00 - 0000000100000000 (reserved)
[ 0.000000] BIOS-e820: 0000000100000000 - 0000000130000000 (usable)
[ 0.000000] end_pfn_map = 1245184
[ 0.000000] DMI present.
[ 0.000000] Zone PFN ranges:
[ 0.000000] DMA 0 -> 4096
[ 0.000000] DMA32 4096 -> 1048576
[ 0.000000] Normal 1048576 -> 1245184
[ 0.000000] early_node_map[3] active PFN ranges
[ 0.000000] 0: 0 -> 155
[ 0.000000] 0: 256 -> 851824
[ 0.000000] 0: 1048576 -> 1245184
[ 0.000000] ACPI: PM-Timer IO Port: 0x1008
[ 0.000000] ACPI: LAPIC (acpi_id[0x00] lapic_id[0x00] enabled)
[ 0.000000] Processor #0 (Bootup-CPU)
[ 0.000000] ACPI: LAPIC (acpi_id[0x01] lapic_id[0x06] enabled)
[ 0.000000] Processor #6
[ 0.000000] ACPI: LAPIC (acpi_id[0x02] lapic_id[0x01] enabled)
[ 0.000000] Processor #1
[ 0.000000] WARNING: NR_CPUS limit of 2 reached. Processor ignored.
[ 0.000000] ACPI: LAPIC (acpi_id[0x03] lapic_id[0x07] enabled)
[ 0.000000] Processor #7
[ 0.000000] WARNING: NR_CPUS limit of 2 reached. Processor ignored.
[ 0.000000] ACPI: LAPIC_NMI (acpi_id[0x00] high edge lint[0x1])
[ 0.000000] ACPI: LAPIC_NMI (acpi_id[0x01] high edge lint[0x1])
[ 0.000000] ACPI: LAPIC_NMI (acpi_id[0x02] high edge lint[0x1])
[ 0.000000] ACPI: LAPIC_NMI (acpi_id[0x03] high edge lint[0x1])
[ 0.000000] ACPI: IOAPIC (id[0x02] address[0xfec00000] gsi_base[0])
[ 0.000000] IOAPIC[0]: apic_id 2, address 0xfec00000, GSI 0-23
[ 0.000000] ACPI: IOAPIC (id[0x03] address[0xfec80000] gsi_base[24])
[ 0.000000] IOAPIC[1]: apic_id 3, address 0xfec80000, GSI 24-47
[ 0.000000] ACPI: IOAPIC (id[0x04] address[0xfec80400] gsi_base[48])
[ 0.000000] IOAPIC[2]: apic_id 4, address 0xfec80400, GSI 48-71
[ 0.000000] ACPI: IOAPIC (id[0x05] address[0xfec84000] gsi_base[72])
[ 0.000000] IOAPIC[3]: apic_id 5, address 0xfec84000, GSI 72-95
[ 0.000000] ACPI: IOAPIC (id[0x08] address[0xfec84400] gsi_base[96])
[ 0.000000] IOAPIC[4]: apic_id 8, address 0xfec84400, GSI 96-119
[ 0.000000] ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 high edge)
[ 0.000000] ACPI: INT_SRC_OVR (bus 0 bus_irq 9 global_irq 9 high level)
[ 0.000000] Setting APIC routing to flat
[ 0.000000] Using ACPI (MADT) for SMP configuration information
[ 0.000000] Nosave address range: 000000000009b000 - 000000000009c000
[ 0.000000] Nosave address range: 000000000009c000 - 00000000000a0000
[ 0.000000] Nosave address range: 00000000000a0000 - 00000000000e4000
[ 0.000000] Nosave address range: 00000000000e4000 - 0000000000100000
[ 0.000000] Nosave address range: 00000000cff70000 - 00000000cff78000
[ 0.000000] Nosave address range: 00000000cff78000 - 00000000cff80000
[ 0.000000] Nosave address range: 00000000cff80000 - 00000000d0000000
[ 0.000000] Nosave address range: 00000000d0000000 - 00000000e0000000
[ 0.000000] Nosave address range: 00000000e0000000 - 00000000f0000000
[ 0.000000] Nosave address range: 00000000f0000000 - 00000000fec00000
[ 0.000000] Nosave address range: 00000000fec00000 - 00000000fec10000
[ 0.000000] Nosave address range: 00000000fec10000 - 00000000fee00000
[ 0.000000] Nosave address range: 00000000fee00000 - 00000000fee01000
[ 0.000000] Nosave address range: 00000000fee01000 - 00000000ff800000
[ 0.000000] Nosave address range: 00000000ff800000 - 00000000ffc00000
[ 0.000000] Nosave address range: 00000000ffc00000 - 0000000100000000
[ 0.000000] Allocating PCI resources starting at d1000000 (gap: d0000000:10000000)
[ 0.000000] PERCPU: Allocating 34176 bytes of per cpu data
[ 0.000000] Built 1 zonelists. Total pages: 1029907
[ 0.000000] Kernel command line: ro root=/dev/sda1 rootdelay=10 init=/xiv/system/xinit console=ttyS0,115200
[ 0.000000] Initializing CPU#0
[ 0.000000] PID hash table entries: 4096 (order: 12, 32768 bytes)
[ 87.927578] Console: colour VGA+ 80x25
[ 88.378396] Dentry cache hash table entries: 524288 (order: 10, 4194304 bytes)
[ 88.389160] Inode-cache hash table entries: 262144 (order: 9, 2097152 bytes)
[ 88.396798] Checking aperture...
[ 88.400165] PCI-DMA: Using software bounce buffering for IO (SWIOTLB)
[ 88.446650] Placing software IO TLB between 0x5900000 - 0x9900000
[ 88.452744] permamem: initializing
[ 88.456142] permamem: allocating dandalf at 0000000020000000 with 48 MBs, virt ffff810020000000, page ffff810001700000
[ 88.468414] prebb: allocation for objsize 1048576 (2 objects) succeeded
[ 88.481074] prebb: allocation for objsize 8192 (1280 objects) succeeded
[ 88.487706] prebb: activated with 2 prebb pools
[ 88.530267] Memory: 3985264k/4980736k available (2867k kernel code, 207772k reserved, 1839k data, 264k init)
[ 88.618737] Calibrating delay using timer specific routine.. 5604.65 BogoMIPS (lpj=11209307)
[ 88.649777] kdb version 4.4 by Keith Owens, Scott Lurndal. Copyright SGI, All Rights Reserved
kdb_cmd[0]: defcmd archkdb "" "First line arch debugging"
kdb_cmd[7]: defcmd archkdbcpu "" "archkdb with only tasks on cpus"
kdb_cmd[14]: defcmd archkdbshort "" "archkdb with less detailed backtrace"
kdb_cmd[21]: defcmd archkdbcommon "" "Common arch debugging"
[ 88.681922] dandalf: partition entries size 15442 KB
[ 88.686882] dandalf: partition vardata size 3676 KB
[ 88.692494] dandalf: partition logptrs size 2451 KB
[ 88.697363] dandalf: partition modentries size 490 KB
[ 88.702405] dandalf: partition modvardata size 2451 KB
[ 88.707715] dandalf: initialization succeeded
[ 88.712123] Mount-cache hash table entries: 256
[ 88.716798] CPU: Trace cache: 12K uops, L1 D cache: 16K
[ 88.722032] CPU: L2 cache: 2048K
[ 88.725256] using mwait in idle threads.
[ 88.729172] CPU: Physical Processor ID: 0
[ 88.733175] CPU: Processor Core ID: 0
[ 88.736844] CPU0: Thermal monitoring enabled (TM1)
[ 88.741637] Freeing SMP alternatives: 24k freed
[ 88.746175] ACPI: Core revision 20060707
[ 88.794449] Using local APIC timer interrupts.
[ 88.834526] result 12500652
[ 88.837310] Detected 12.500 MHz APIC timer.
[ 88.842473] Booting processor 1/2 APIC 0x6
[ 88.857137] Initializing CPU#1
[ 88.934146] Calibrating delay using timer specific routine.. 5600.65 BogoMIPS (lpj=11201313)
[ 88.934160] CPU: Trace cache: 12K uops, L1 D cache: 16K
[ 88.934164] CPU: L2 cache: 2048K
[ 88.934168] CPU: Physical Processor ID: 3
[ 88.934170] CPU: Processor Core ID: 0
[ 88.934182] CPU1: Thermal monitoring enabled (TM1)
[ 88.934593] Intel(R) Xeon(TM) CPU 2.80GHz stepping 03
[ 88.938180] Brought up 2 CPUs
[ 88.980735] testing NMI watchdog ... OK.
[ 89.024608] time.c: Using 3.579545 MHz WALL PM GTOD PIT/TSC timer.
[ 89.030776] time.c: Detected 2800.166 MHz processor.
[ 89.376365] migration_cost=1743
[ 89.380050] NET: Registered protocol family 16
[ 89.384546] ACPI: bus type pci registered
[ 89.392937] PCI: Using MMCONFIG at e0000000
[ 89.397396] PCI: No mmconfig possible on device 08:01
[ 89.406174] ACPI: Interpreter enabled
[ 89.409838] ACPI: Using IOAPIC for interrupt routing
[ 89.415085] ACPI: PCI Root Bridge [PCI0] (0000:00)
[ 89.422404] PCI quirk: region 1000-107f claimed by ICH4 ACPI/GPIO/TCO
[ 89.428831] PCI quirk: region 1180-11bf claimed by ICH4 GPIO
[ 89.434507] PCI: Ignoring BAR0-3 of IDE controller 0000:00:1f.2
[ 89.440483] PCI: PXH quirk detected, disabling MSI for SHPC device
[ 89.446688] PCI: PXH quirk detected, disabling MSI for SHPC device
[ 89.453210] PCI: PXH quirk detected, disabling MSI for SHPC device
[ 89.459417] PCI: PXH quirk detected, disabling MSI for SHPC device
[ 89.466711] PCI: Transparent bridge - 0000:00:1e.0
[ 89.479098] ACPI: PCI Interrupt Link [LNKA] (IRQs 3 4 *5 6 7 10 11 14 15)
[ 89.486286] ACPI: PCI Interrupt Link [LNKB] (IRQs 3 4 5 6 7 *10 11 14 15)
[ 89.493461] ACPI: PCI Interrupt Link [LNKC] (IRQs 3 4 5 6 *7 10 11 14 15)
[ 89.500635] ACPI: PCI Interrupt Link [LNKD] (IRQs 3 4 5 6 7 10 *11 14 15)
[ 89.507809] ACPI: PCI Interrupt Link [LNKE] (IRQs 3 4 5 6 7 10 11 14 15) *0, disabled.
[ 89.516137] ACPI: PCI Interrupt Link [LNKF] (IRQs 4 5 6 7 10 11 14 15) *0, disabled.
[ 89.524276] ACPI: PCI Interrupt Link [LNKG] (IRQs 3 4 5 6 7 10 11 14 15) *0, disabled.
[ 89.532604] ACPI: PCI Interrupt Link [LNKH] (IRQs 4 5 6 7 *10 11 14 15)
[ 89.539559] ACPI: Device [PRT] status [0000000c]: functional but not present; setting present
[ 89.549542] Linux Plug and Play Support v0.97 (c) Adam Belay
[ 89.555203] pnp: PnP ACPI init
[ 89.561717] pnp: PnP ACPI: found 12 devices
[ 89.571619] SCSI subsystem initialized
[ 89.575446] usbcore: registered new interface driver usbfs
[ 89.580948] usbcore: registered new interface driver hub
[ 89.586279] usbcore: registered new device driver usb
[ 89.591366] PCI: Using ACPI for IRQ routing
[ 89.595544] PCI: If a device doesn't work, try "pci=routeirq". If it helps, post a report
[ 89.603921] PCI-GART: No AMD northbridge found.
[ 89.608786] PCI: Bridge: 0000:01:00.0
[ 89.612443] IO window: 2000-2fff
[ 89.615843] MEM window: dc200000-dc2fffff
[ 89.620018] PREFETCH window: d1800000-d18fffff
[ 89.624630] PCI: Bridge: 0000:01:00.2
[ 89.628285] IO window: 3000-3fff
[ 89.631686] MEM window: dc300000-dc3fffff
[ 89.635862] PREFETCH window: disabled.
[ 89.639780] PCI: Bridge: 0000:00:02.0
[ 89.643436] IO window: 2000-3fff
[ 89.646838] MEM window: dc100000-dc3fffff
[ 89.651014] PREFETCH window: d1800000-d18fffff
[ 89.655623] PCI: Bridge: 0000:00:04.0
[ 89.659280] IO window: disabled.
[ 89.662681] MEM window: disabled.
[ 89.666164] PREFETCH window: disabled.
[ 89.670088] PCI: Bridge: 0000:05:00.0
[ 89.673741] IO window: 4000-4fff
[ 89.677143] MEM window: dc500000-dcbfffff
[ 89.681319] PREFETCH window: d1000000-d13fffff
[ 89.685930] PCI: Bridge: 0000:05:00.2
[ 89.689585] IO window: 5000-5fff
[ 89.692986] MEM window: dcc00000-dd0fffff
[ 89.697161] PREFETCH window: d1400000-d17fffff
[ 89.701772] PCI: Bridge: 0000:00:06.0
[ 89.705429] IO window: 4000-5fff
[ 89.708830] MEM window: dc400000-dd0fffff
[ 89.713005] PREFETCH window: d1000000-d17fffff
[ 89.717617] PCI: Bridge: 0000:00:1e.0
[ 89.721272] IO window: 6000-6fff
[ 89.725392] MEM window: dd100000-deffffff
[ 89.729567] PREFETCH window: d1900000-d19fffff
[ 89.734192] ACPI: PCI Interrupt 0000:00:02.0[A] -> GSI 16 (level, low) -> IRQ 16
[ 89.741630] ACPI: PCI Interrupt 0000:00:04.0[A] -> GSI 16 (level, low) -> IRQ 16
[ 89.749044] ACPI: PCI Interrupt 0000:00:06.0[A] -> GSI 16 (level, low) -> IRQ 16
[ 89.756514] NET: Registered protocol family 2
[ 89.796569] IP route cache hash table entries: 131072 (order: 8, 1048576 bytes)
[ 89.804412] TCP established hash table entries: 131072 (order: 10, 4194304 bytes)
[ 89.814087] TCP bind hash table entries: 65536 (order: 9, 2097152 bytes)
[ 89.821987] TCP: Hash tables configured (established 131072 bind 65536)
[ 89.828596] TCP reno registered
[ 89.831856] Simple Boot Flag at 0x39 set to 0x80
[ 89.837134] JFFS2 version 2.2. (NAND) (C) 2001-2006 Red Hat, Inc.
[ 89.843285] SGI XFS with large block/inode numbers, no debug enabled
[ 89.849848] io scheduler noop registered
[ 89.853786] io scheduler anticipatory registered (default)
[ 89.859287] io scheduler deadline registered
[ 89.863583] io scheduler cfq registered
[ 89.869471] ACPI: Power Button (FF) [PWRF]
[ 89.873599] ACPI: Power Button (CM) [PWRB]
[ 89.877792] Using specific hotkey driver
[ 89.882065] ACPI: Getting cpuindex for acpiid 0x2
[ 89.886771] ACPI: Getting cpuindex for acpiid 0x3
[ 89.906793] Real Time Clock Driver v1.12ac
[ 89.910887] Linux agpgart interface v0.101 (c) Dave Jones
[ 89.916274] Serial: 8250/16550 driver $Revision: 1.90 $ 4 ports, IRQ sharing disabled
[ 89.924184] serial8250: ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A
[ 89.930290] serial8250: ttyS1 at I/O 0x2f8 (irq = 3) is a 16550A
[ 89.936703] 00:09: ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A
[ 89.942418] 00:0a: ttyS1 at I/O 0x2f8 (irq = 3) is a 16550A
[ 89.948155] Floppy drive(s): fd0 is 1.44M
[ 89.971375] FDC 0 is a National Semiconductor PC87306
[ 89.977728] RAMDISK driver initialized: 16 RAM disks of 80000K size 1024 blocksize
[ 89.985465] loop: loaded (max 8 devices)
[ 89.989489] Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2
[ 89.995829] ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
[ 90.003849] ICH5-SATA: IDE controller at PCI slot 0000:00:1f.2
[ 90.009684] ACPI: PCI Interrupt 0000:00:1f.2[A] -> GSI 18 (level, low) -> IRQ 18
[ 90.017090] ICH5-SATA: chipset revision 2
[ 90.021088] ICH5-SATA: not 100% native mode: will probe irqs later
[ 90.027262] ide0: BM-DMA at 0x14b0-0x14b7, BIOS settings: hda:pio, hdb:pio
[ 90.034522] ide1: BM-DMA at 0x14b8-0x14bf, BIOS settings: hdc:pio, hdd:pio
[ 90.331630] hda: HDS725050KLA360, ATA DISK drive
[ 91.010375] ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
[ 92.156005] hda: max request size: 1024KiB
[ 92.171581] hda: 976773168 sectors (500107 MB) w/15607KiB Cache, CHS=60801/255/63, UDMA(33)
[ 92.180178] hda: cache flushes supported
[ 92.184147] hda: hda1
[ 92.189483] block2mtd: version $Revision: 1.30 $
[ 92.194093] usbmon: debugfs is not available
[ 92.198399] ACPI: PCI Interrupt 0000:00:1d.7[D] -> GSI 23 (level, low) -> IRQ 23
[ 92.206572] ehci_hcd 0000:00:1d.7: EHCI Host Controller
[ 92.211860] ehci_hcd 0000:00:1d.7: new USB bus registered, assigned bus number 1
[ 92.219272] ehci_hcd 0000:00:1d.7: debug port 1
[ 92.223809] ehci_hcd 0000:00:1d.7: irq 23, io mem 0xdc001000
[ 92.233343] ehci_hcd 0000:00:1d.7: USB 2.0 started, EHCI 1.00, driver 10 Dec 2004
[ 92.240924] usb usb1: configuration #1 chosen from 1 choice
[ 92.246515] hub 1-0:1.0: USB hub found
[ 92.250264] hub 1-0:1.0: 8 ports detected
[ 92.359690] Initializing USB Mass Storage driver...
[ 92.599138] usb 1-1: new high speed USB device using ehci_hcd and address 2
[ 92.744533] usb 1-1: configuration #1 chosen from 1 choice
[ 92.994378] usb 1-2: new high speed USB device using ehci_hcd and address 3
[ 93.139812] usb 1-2: configuration #1 chosen from 1 choice
[ 93.145732] scsi0 : SCSI emulation for USB Mass Storage devices
[ 93.151756] scsi1 : SCSI emulation for USB Mass Storage devices
[ 93.157745] usbcore: registered new interface driver usb-storage
[ 93.162086] scsi 0:0:0:0: Direct-Access OTi Flash Disk 2.00 PQ: 0 ANSI: 2
[ 93.171825] USB Mass Storage support registered.
[ 93.176503] PNP: PS/2 Controller [PNP0303:KBC0,PNP0f13:MSE0] at 0x60,0x64 irq 1,12
[ 93.184381] SCSI device sda: 2037760 512-byte hdwr sectors (1043 MB)
[ 93.191394] sda: Write Protect is off
[ 93.191886] serio: i8042 KBD port at 0x60,0x64 irq 1
[ 93.191894] serio: i8042 AUX port at 0x60,0x64 irq 12
[ 93.191916] mice: PS/2 mouse device common for all mice
[ 93.191927] EDAC MC: Ver: 2.0.1 Jan 12 2007
[ 93.214415] sda: assuming drive cache: write through
[ 93.219475] EDAC e752x: tolm = d0000, remapbase = 100000, remaplimit = 12c000
[ 93.226645] EDAC MC0: Giving out device to e752x_edac E7520: DEV 0000:00:00.0
[ 93.233832] TCP cubic registered
[ 93.237066] NET: Registered protocol family 1
[ 93.241421] NET: Registered protocol family 17
[ 93.245858] 802.1Q VLAN Support v1.8 Ben Greear <[email protected]>
[ 93.252639] All bugs added by David S. Miller <[email protected]>
[ 93.260363] SCSI device sda: 2037760 512-byte hdwr sectors (1043 MB)
[ 93.267221] sda: Write Protect is off
[ 93.270885] sda: assuming drive cache: write through
[ 93.275846] sda: sda1 sda2 sda3
[ 93.279962] sd 0:0:0:0: Attached scsi removable disk sda
[ 93.371900] VFS: Mounted root (ext2 filesystem) readonly.
[ 93.377313] xiv: board Supermicro X6DH8-XG2
[ 93.381538] Freeing unused kernel memory: 264k freed

--
Dan Aloni
XIV LTD, http://www.xivstorage.com
da-x (at) monatomic.org, dan (at) xiv.co.il

2007-01-12 14:57:26

by Dan Aloni

[permalink] [raw]

Subject: Re: kexec + USB storage in 2.6.19

2007-01-12 15:26:34

by Eric W. Biederman

[permalink] [raw]

Subject: Re: kexec + USB storage in 2.6.19

Dan Aloni <[email protected]> writes:

> I'm attaching the full logs.

Thanks.

> [ 8656.272980] ACPI Error (tbxfroot-0512): Could not map memory at 0000040E for length 2 [20060707]

Ok. This looks like the first sign of trouble.
Normally I would suspect a memory map issue but your e820 memory map looks fine,
although a little different between the two kernels.

Is this enough of a hint for you to dig more deeply?

Eric

2007-01-12 16:03:07

by Dan Aloni

[permalink] [raw]

Subject: kexec + ACPI in 2.6.19 (was: Re: kexec + USB storage in 2.6.19)

On Fri, Jan 12, 2007 at 08:26:03AM -0700, Eric W. Biederman wrote:
> Dan Aloni <[email protected]> writes:
>
> > I'm attaching the full logs.
>
> Thanks.
>
> > [ 8656.272980] ACPI Error (tbxfroot-0512): Could not map memory at 0000040E for length 2 [20060707]
>
> Ok. This looks like the first sign of trouble.
> Normally I would suspect a memory map issue but your e820 memory map looks fine,
> although a little different between the two kernels.
>
> Is this enough of a hint for you to dig more deeply?

Reverting just the ACPI code (everything under drivers/acpi/*)
back to the version of 2.6.18.3 doesn't fix the problem, so it
must be something else.

I'll try to dig it up more deeply..

--
Dan Aloni
XIV LTD, http://www.xivstorage.com
da-x (at) monatomic.org, dan (at) xiv.co.il

2007-01-12 16:28:09

by Dan Aloni

[permalink] [raw]

Subject: Re: kexec + ACPI in 2.6.19 (was: Re: kexec + USB storage in 2.6.19)

On Fri, Jan 12, 2007 at 06:02:43PM +0200, Dan Aloni wrote:
> On Fri, Jan 12, 2007 at 08:26:03AM -0700, Eric W. Biederman wrote:
> > Dan Aloni <[email protected]> writes:
> >
> > > I'm attaching the full logs.
> >
> > Thanks.
> >
> > > [ 8656.272980] ACPI Error (tbxfroot-0512): Could not map memory at 0000040E for length 2 [20060707]
> >
> > Ok. This looks like the first sign of trouble.
> > Normally I would suspect a memory map issue but your e820 memory map looks fine,
> > although a little different between the two kernels.
> >
> > Is this enough of a hint for you to dig more deeply?
>
> Reverting just the ACPI code (everything under drivers/acpi/*)
> back to the version of 2.6.18.3 doesn't fix the problem, so it
> must be something else.

Just occured to me that I didn't revert the relevant code under
arch/x86_64 so it might still be related somehow..

--
Dan Aloni
XIV LTD, http://www.xivstorage.com
da-x (at) monatomic.org, dan (at) xiv.co.il

2007-01-12 16:43:50

by Dan Aloni

[permalink] [raw]

Subject: Re: kexec + ACPI in 2.6.19 (was: Re: kexec + USB storage in 2.6.19)

On Fri, Jan 12, 2007 at 06:28:00PM +0200, Dan Aloni wrote:
> On Fri, Jan 12, 2007 at 06:02:43PM +0200, Dan Aloni wrote:
> > On Fri, Jan 12, 2007 at 08:26:03AM -0700, Eric W. Biederman wrote:
> > > Dan Aloni <[email protected]> writes:
> > >
> > > > I'm attaching the full logs.
> > >
> > > Thanks.
> > >
> > > > [ 8656.272980] ACPI Error (tbxfroot-0512): Could not map memory at 0000040E for length 2 [20060707]
> > >
> > > Ok. This looks like the first sign of trouble.
> > > Normally I would suspect a memory map issue but your e820 memory map looks fine,
> > > although a little different between the two kernels.
> > >
> > > Is this enough of a hint for you to dig more deeply?
> >
> > Reverting just the ACPI code (everything under drivers/acpi/*)
> > back to the version of 2.6.18.3 doesn't fix the problem, so it
> > must be something else.
>
> Just occured to me that I didn't revert the relevant code under
> arch/x86_64 so it might still be related somehow..

After adding a few prints inside __ioremap() it appears the function
exits for phys_addr==0x40e because (!PageReserved(page)).

Isn't page 0 supposed to be reserved? I clearly see that it is
being reserved under setup_arch().

Odd, I must say...

--
Dan Aloni
XIV LTD, http://www.xivstorage.com
da-x (at) monatomic.org, dan (at) xiv.co.il

2007-01-12 16:56:14

by Dan Aloni

[permalink] [raw]

Subject: Re: kexec + ACPI in 2.6.19 (was: Re: kexec + USB storage in 2.6.19)

On Fri, Jan 12, 2007 at 06:43:40PM +0200, Dan Aloni wrote:
> On Fri, Jan 12, 2007 at 06:28:00PM +0200, Dan Aloni wrote:
> > On Fri, Jan 12, 2007 at 06:02:43PM +0200, Dan Aloni wrote:
> > > On Fri, Jan 12, 2007 at 08:26:03AM -0700, Eric W. Biederman wrote:
> > > > Dan Aloni <[email protected]> writes:
> > > >
> > > > > I'm attaching the full logs.
> > > >
> > > > Thanks.
> > > >
> > > > > [ 8656.272980] ACPI Error (tbxfroot-0512): Could not map memory at 0000040E for length 2 [20060707]
> > > >
> > > > Ok. This looks like the first sign of trouble.
> > > > Normally I would suspect a memory map issue but your e820 memory map looks fine,
> > > > although a little different between the two kernels.
> > > >
> > > > Is this enough of a hint for you to dig more deeply?
> > >
> > > Reverting just the ACPI code (everything under drivers/acpi/*)
> > > back to the version of 2.6.18.3 doesn't fix the problem, so it
> > > must be something else.
> >
> > Just occured to me that I didn't revert the relevant code under
> > arch/x86_64 so it might still be related somehow..
>
> After adding a few prints inside __ioremap() it appears the function
> exits for phys_addr==0x40e because (!PageReserved(page)).
>
> Isn't page 0 supposed to be reserved? I clearly see that it is
> being reserved under setup_arch().
>
> Odd, I must say...

In __ioremap() I added this under if(!PageReserved(page)) {...}:

if (phys_addr == 0x40e) {
printk("PAGE %p (pfn=%ld): flags=%lx, count=%d\n",
page,
page_to_pfn(page),
page->flags,
atomic_read(&page->_count));
}

And I get:

[ 1013.864201] PAGE ffff810001000000 (pfn=0): flags=0, count=0

So at least no one is using that page. Still it is not clear why it
doesn't have the reserve flag turned on.

--
Dan Aloni
XIV LTD, http://www.xivstorage.com
da-x (at) monatomic.org, dan (at) xiv.co.il

2007-01-12 17:33:46

by Eric W. Biederman

[permalink] [raw]

Subject: Re: kexec + ACPI in 2.6.19 (was: Re: kexec + USB storage in 2.6.19)

Dan Aloni <[email protected]> writes:

> On Fri, Jan 12, 2007 at 06:43:40PM +0200, Dan Aloni wrote:
>> On Fri, Jan 12, 2007 at 06:28:00PM +0200, Dan Aloni wrote:
>> > On Fri, Jan 12, 2007 at 06:02:43PM +0200, Dan Aloni wrote:
>> > > On Fri, Jan 12, 2007 at 08:26:03AM -0700, Eric W. Biederman wrote:
>> > > > Dan Aloni <[email protected]> writes:
>> > > >
>> > > > > I'm attaching the full logs.
>> > > >
>> > > > Thanks.
>> > > >
>> > > > > [ 8656.272980] ACPI Error (tbxfroot-0512): Could not map memory at
> 0000040E for length 2 [20060707]
>> > > >
>> > > > Ok. This looks like the first sign of trouble.
>> > > > Normally I would suspect a memory map issue but your e820 memory map
> looks fine,
>> > > > although a little different between the two kernels.
>> > > >
>> > > > Is this enough of a hint for you to dig more deeply?
>> > >
>> > > Reverting just the ACPI code (everything under drivers/acpi/*)
>> > > back to the version of 2.6.18.3 doesn't fix the problem, so it
>> > > must be something else.
>> >
>> > Just occured to me that I didn't revert the relevant code under
>> > arch/x86_64 so it might still be related somehow..
>>
>> After adding a few prints inside __ioremap() it appears the function
>> exits for phys_addr==0x40e because (!PageReserved(page)).
>>
>> Isn't page 0 supposed to be reserved? I clearly see that it is
>> being reserved under setup_arch().
>>
>> Odd, I must say...
>
> In __ioremap() I added this under if(!PageReserved(page)) {...}:
>
> if (phys_addr == 0x40e) {
> printk("PAGE %p (pfn=%ld): flags=%lx, count=%d\n",
> page,
> page_to_pfn(page),
> page->flags,
> atomic_read(&page->_count));
> }
>
> And I get:
>
> [ 1013.864201] PAGE ffff810001000000 (pfn=0): flags=0, count=0
>
> So at least no one is using that page. Still it is not clear why it
> doesn't have the reserve flag turned on.

My hunch is that it might have something to do with the difference
in the e820 map. The original map reports that whole page as
being usable while in the kexec case the memory from 0x100 is
reported as being usable. Perhaps someone has a rounding error?

Eric

2007-01-12 18:49:56

by Dan Aloni

[permalink] [raw]

Subject: Re: kexec + ACPI in 2.6.19 (was: Re: kexec + USB storage in 2.6.19)

On Fri, Jan 12, 2007 at 10:33:34AM -0700, Eric W. Biederman wrote:
> > And I get:
> >
> > [ 1013.864201] PAGE ffff810001000000 (pfn=0): flags=0, count=0
> >
> > So at least no one is using that page. Still it is not clear why it
> > doesn't have the reserve flag turned on.
>
> My hunch is that it might have something to do with the difference
> in the e820 map. The original map reports that whole page as
> being usable while in the kexec case the memory from 0x100 is
> reported as being usable. Perhaps someone has a rounding error?

I added this right at the beginning of sanitize_e820_map():

for (i=0; i < *pnr_map; i++) {
printk("index %d: addr=%llx, size=%llx, type=%d\n",
i,
(unsigned long long)biosmap[i].addr,
(unsigned long long)biosmap[i].size,
(unsigned int)biosmap[i].type);
}

.. and got (kexec boot):

[ 0.000000] index 0: addr=100, size=9b700, type=1
[ 0.000000] index 1: addr=9b800, size=4800, type=2
[ 0.000000] index 2: addr=100000, size=cfe70000, type=1
[ 0.000000] index 3: addr=cff70000, size=8000, type=3
[ 0.000000] index 4: addr=cff78000, size=8000, type=4
[ 0.000000] index 5: addr=cff80000, size=80000, type=2
[ 0.000000] index 6: addr=e0000000, size=10000000, type=2
[ 0.000000] index 7: addr=fec00000, size=10000, type=2
[ 0.000000] index 8: addr=fee00000, size=1000, type=2
[ 0.000000] index 9: addr=ff800000, size=400000, type=2
[ 0.000000] index 10: addr=fffffc00, size=400, type=2
[ 0.000000] index 11: addr=100000000, size=30000000, type=1

Compared with a normal boot:

[ 0.000000] index 0: addr=0, size=9b800, type=1
[ 0.000000] index 1: addr=9b800, size=4800, type=2
[ 0.000000] index 2: addr=e4000, size=1c000, type=2
[ 0.000000] index 3: addr=100000, size=cfe70000, type=1
[ 0.000000] index 4: addr=cff70000, size=8000, type=3
[ 0.000000] index 5: addr=cff78000, size=8000, type=4
[ 0.000000] index 6: addr=cff80000, size=80000, type=2
[ 0.000000] index 7: addr=e0000000, size=10000000, type=2
[ 0.000000] index 8: addr=fec00000, size=10000, type=2
[ 0.000000] index 9: addr=fee00000, size=1000, type=2
[ 0.000000] index 10: addr=ff800000, size=400000, type=2
[ 0.000000] index 11: addr=fffffc00, size=400, type=2
[ 0.000000] index 12: addr=100000000, size=30000000, type=1

So that is the actual boot data that is passed to us.

Anyway, if I add this:

if (*pnr_map >= 1) {
if (biosmap[0].addr == 0x100)
biosmap[0].addr = 0;
}

kexec works. I'll use this for now, but obviously we _must_ come up
with a cleaner fix...

--
Dan Aloni
XIV LTD, http://www.xivstorage.com
da-x (at) monatomic.org, dan (at) xiv.co.il