Olivier Cochard-Labbé
2010-Jan-02 18:54 UTC
Regression with Asus K8N7-E deluxe motherboard on 8-stable: ACPI or PCI drivers bug ?
Hi,
I've met 3 regressions on 8-stable on my Asus K8N7-E deluxe motherboard:
- PATA drivers (kern/139143)
- USB 2.0 drivers (usb/139142)
- Ethernet (nfe) drivers
But, I don't think that the USB/PATA/nfe drivers are the problem: The
problem seems came from ACPI or PCI bus drivers bug with this
motherboard.
There are lot's of differences between the dmesg on 7.2 and 8.0
regarding the hardware ressources informations.
Full dmesg under FreeBSD 7.2 here:
http://pastebin.com/f7bbdcb3a
Full dmesg under FreeBSD 8.0 here:
http://pastebin.com/f52df9c4c
Here, are some examples of the diff between the 2 dmesg:
--------------
About ACPI, there is this error on 8.0:
ACPI Error: Package List length (6) larger than NumElements count (3), truncated
20090521 dsobject-590
---------------
About PCI link18 (Initial Probe IRQ diff):
7.2:
pci_link18: Index IRQ Rtd Ref IRQs
Initial Probe 0 255 N 0 18
Validation 0 255 N 0 18
After Disable 0 255 N 0 18
8.0:
pci_link18: Index IRQ Rtd Ref IRQs
Initial Probe 0 5 N 0 18
Validation 0 255 N 0 18
After Disable 0 255 N 0 18
(Note the IRQ diff)
---------------
And lot's of devices have map size diff:
on 7.2:
found-> vendor=0x10de, dev=0x0052, revid=0xa2
map[10]: type I/O Port, range 32, base 0xdc00, size 5, enabled
map[20]: type I/O Port, range 32, base 0x4c00, size 6, enabled
map[24]: type I/O Port, range 32, base 0x4c40, size 6, enabled
found-> vendor=0x10de, dev=0x0053, revid=0xf2
map[20]: type I/O Port, range 32, base 0xf000, size 4, enabled
found-> vendor=0x10de, dev=0x0055, revid=0xf3
map[10]: type I/O Port, range 32, base 0x9e0, size 3, enabled
map[14]: type I/O Port, range 32, base 0xbe0, size 2, enabled
map[18]: type I/O Port, range 32, base 0x960, size 3, enabled
map[1c]: type I/O Port, range 32, base 0xb60, size 2, enabled
map[20]: type I/O Port, range 32, base 0xc400, size 4, enabled
on 8.0:
found-> vendor=0x10de, dev=0x0052, revid=0xa2
map[10]: type I/O Port, range 32, base 0xdc00, size 10, enabled
map[20]: type I/O Port, range 32, base 0x4c00, size 10, enabled
map[24]: type I/O Port, range 32, base 0x4c40, size 6, enabled
found-> vendor=0x10de, dev=0x0053, revid=0xf2
map[20]: type I/O Port, range 32, base 0xf000, size 12, enabled
found-> vendor=0x10de, dev=0x0055, revid=0xf3
map[10]: type I/O Port, range 32, base 0x9e0, size 5, enabled
map[14]: type I/O Port, range 32, base 0xbe0, size 5, enabled
map[18]: type I/O Port, range 32, base 0x960, size 5, enabled
map[1c]: type I/O Port, range 32, base 0xb60, size 5, enabled
map[20]: type I/O Port, range 32, base 0xc400, size 10, enabled
(Note the size diff)
---------------
An unknown and USB error on 8.0:
7.2:
pcib0: slot 2 INTA routed to irq 21 via \_SB_.PCI0.APCF
8.0:
pcib0: slot 2 INTA routed to irq 21 via \_SB_.PCI0.APCF
unknown: Lazy allocation of 0x1000 bytes rid 0x10 type 3 at 0x80000000
unknown: Reserved 0x1000 bytes for rid 0x10 type 3 at 0x80000000
ohci early: SMM active, request owner change
ohci early: SMM does not respond, resetting
---------------
The memory addresses of devices or reserved addresse bytes are different:
7.2:
ohci0: <OHCI (generic) USB controller> mem 0xd3003000-0xd3003fff irq
21 at device 2.0 on pci0
ohci0: Reserved 0x1000 bytes for rid 0x10 type 3 at 0xd3003000
ioapic0: routing intpin 21 (PCI IRQ 21) to vector 49
atapci0: <nVidia nForce CK804 UDMA133 controller> port
0x1f0-0x1f7,0x3f6,0x170-0x177,0x376,0xf000-0xf00f at device 6.0 on
pci0
atapci0: Reserved 0x10 bytes for rid 0x20 type 4 at 0xf000
atapci1: <nVidia nForce CK804 SATA300 controller> port
0x9f0-0x9f7,0xbf0-0xbf3,0x970-0x977,0xb70-0xb73,0xd800-0xd80f mem
0xd3002000-0xd3002fff irq 22 at device 7.0 on pci0
atapci1: Reserved 0x10 bytes for rid 0x20 type 4 at 0xd800
ioapic0: routing intpin 22 (PCI IRQ 22) to vector 52
nfe0: <NVIDIA nForce4 CK804 MCP8 Networking Adapter> port
0xb000-0xb007 mem 0xd3000000-0xd3000fff irq 21 at device 10.0 on pci0
nfe0: Reserved 0x1000 bytes for rid 0x10 type 3 at 0xd3000000
miibus0: <MII bus> on nfe0
e1000phy0: <Marvell 88E1111 Gigabit PHY> PHY 15 on miibus0
e1000phy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseTX-FDX, auto
nfe0: bpf attached
nfe0: Ethernet address: 00:13:d4:d6:3a:7b
8.0:
ohci0: <OHCI (generic) USB controller> mem 0x80000000-0x80000fff irq
21 at device 2.0 on pci0
ioapic0: routing intpin 21 (PCI IRQ 21) to lapic 0 vector 49
(Note the mem diff)
atapci0: <nVidia nForce CK804 UDMA133 controller> port
0x1f0-0x1f7,0x3f6,0x170-0x177,0x376 at device 6.0 on pci0
atapci0: Lazy allocation of 0x1000 bytes rid 0x20 type 4 at 0x1000
atapci0: Reserved 0x1000 bytes for rid 0x20 type 4 at 0x1000
(Note the reserved bytes diff)
atapci2: <nVidia nForce CK804 SATA300 controller> irq 23 at device 8.0 on
pci0
atapci2: Lazy allocation of 0x400 bytes rid 0x20 type 4 at 0x2c00
atapci2: Reserved 0x400 bytes for rid 0x20 type 4 at 0x2c00
ioapic0: routing intpin 23 (PCI IRQ 23) to lapic 0 vector 52
(Note the reserved bytes diff)
nfe0: <NVIDIA nForce4 CK804 MCP8 Networking Adapter> irq 21 at device
10.0 on pci0
nfe0: Lazy allocation of 0x1000000 bytes rid 0x10 type 3 at 0x81000000
nfe0: Reserved 0x1000000 bytes for rid 0x10 type 3 at 0x81000000
nfe0: NFE_STATUS_RUNNING set!
nfe0: MII without any phy!
device_attach: nfe0 attach returned 6
(Note the reserved mem diff)
What can I do for helping you to debug this problem ?
Regards,
Olivier
Hans Petter Selasky
2010-Jan-02 21:44 UTC
Regression with Asus K8N7-E deluxe motherboard on 8-stable: ACPI or PCI drivers bug ?
On Saturday 02 January 2010 19:54:11 Olivier Cochard-Labb? wrote:> ohci early: SMM active, request owner change > ohci early: SMM does not respond, resettingIf the I/O address range changed, then the above can be explained by that we are reading/writing to an invalid range. USB debug sysctls: /boot/loader.conf hw.usb.ohci.debug=15 hw.usb.ehci.debug=15 --HPS