USB problem on x86_64: nommu_map_single() issue?
Hi
On 2.6.27, I can't use any USB device (mass storage, mouse) anymore on my x86_64 system, the devices are not detected: lsusb returns nothing except the hub.
I have the following logs in kern.log when booting with my USB mouse plugged :
Oct 13 15:43:48 brew kernel: USB Universal Host Controller Interface driver v3.0 Oct 13 15:43:48 brew kernel: uhci_hcd 0000:00:1d.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16 Oct 13 15:43:48 brew kernel: uhci_hcd 0000:00:1d.0: setting latency timer to 64 Oct 13 15:43:48 brew kernel: uhci_hcd 0000:00:1d.0: UHCI Host Controller Oct 13 15:43:48 brew kernel: uhci_hcd 0000:00:1d.0: new USB bus registered, assigned bus number 1 Oct 13 15:43:48 brew kernel: uhci_hcd 0000:00:1d.0: irq 16, io base 0x00002000 Oct 13 15:43:48 brew kernel: usb usb1: configuration #1 chosen from 1 choice Oct 13 15:43:48 brew kernel: hub 1-0:1.0: USB hub found Oct 13 15:43:48 brew kernel: hub 1-0:1.0: 2 ports detected Oct 13 15:43:48 brew kernel: intel_rng: Firmware space is locked read-only. If you can't or Oct 13 15:43:48 brew kernel: intel_rng: don't want to disable this in firmware setup, and if Oct 13 15:43:48 brew kernel: intel_rng: you are certain that your system has a functional Oct 13 15:43:48 brew kernel: intel_rng: RNG, try using the 'no_fwh_detect' option. Oct 13 15:43:48 brew kernel: uhci_hcd 0000:00:1d.1: PCI INT B -> GSI 19 (level, low) -> IRQ 19 Oct 13 15:43:48 brew kernel: uhci_hcd 0000:00:1d.1: setting latency timer to 64 Oct 13 15:43:48 brew kernel: uhci_hcd 0000:00:1d.1: UHCI Host Controller Oct 13 15:43:48 brew kernel: uhci_hcd 0000:00:1d.1: new USB bus registered, assigned bus number 2 Oct 13 15:43:48 brew kernel: uhci_hcd 0000:00:1d.1: irq 19, io base 0x00002020 Oct 13 15:43:48 brew kernel: usb usb2: configuration #1 chosen from 1 choice Oct 13 15:43:48 brew kernel: hub 2-0:1.0: USB hub found Oct 13 15:43:48 brew kernel: hub 2-0:1.0: 2 ports detected Oct 13 15:43:48 brew kernel: uhci_hcd 0000:00:1d.2: PCI INT C -> GSI 18 (level, low) -> IRQ 18 Oct 13 15:43:48 brew kernel: uhci_hcd 0000:00:1d.2: setting latency timer to 64 Oct 13 15:43:48 brew kernel: uhci_hcd 0000:00:1d.2: UHCI Host Controller Oct 13 15:43:48 brew kernel: uhci_hcd 0000:00:1d.2: new USB bus registered, assigned bus number 3 Oct 13 15:43:48 brew kernel: uhci_hcd 0000:00:1d.2: irq 18, io base 0x00002040 Oct 13 15:43:48 brew kernel: usb usb3: configuration #1 chosen from 1 choice Oct 13 15:43:48 brew kernel: hub 3-0:1.0: USB hub found Oct 13 15:43:48 brew kernel: hub 3-0:1.0: 2 ports detected Oct 13 15:43:48 brew kernel: uhci_hcd 0000:00:1d.3: PCI INT D -> GSI 22 (level, low) -> IRQ 22 Oct 13 15:43:48 brew kernel: uhci_hcd 0000:00:1d.3: setting latency timer to 64 Oct 13 15:43:48 brew kernel: uhci_hcd 0000:00:1d.3: UHCI Host Controller Oct 13 15:43:48 brew kernel: uhci_hcd 0000:00:1d.3: new USB bus registered, assigned bus number 4 Oct 13 15:43:48 brew kernel: uhci_hcd 0000:00:1d.3: irq 22, io base 0x00002060 Oct 13 15:43:48 brew kernel: usb usb4: configuration #1 chosen from 1 choice Oct 13 15:43:48 brew kernel: hub 4-0:1.0: USB hub found Oct 13 15:43:48 brew kernel: hub 4-0:1.0: 2 ports detected Oct 13 15:43:48 brew kernel: usb 2-1: new low speed USB device using uhci_hcd and address 2 Oct 13 15:43:48 brew kernel: nommu_map_single: overflow 12e8b78e0+8 of device mask ffffffff Oct 13 15:43:48 brew kernel: nommu_map_single: overflow 11f0fd480+64 of device mask ffffffff Oct 13 15:43:48 brew kernel: nommu_map_single: overflow 12e8b78e0+8 of device mask ffffffff Oct 13 15:43:48 brew kernel: nommu_map_single: overflow 11f0fd480+64 of device mask ffffffff Oct 13 15:43:48 brew kernel: nommu_map_single: overflow 12e8b78e0+8 of device mask ffffffff Oct 13 15:43:48 brew kernel: nommu_map_single: overflow 11f0fd480+64 of device mask ffffffff Oct 13 15:43:48 brew kernel: HDA Intel 0000:00:1b.0: PCI INT A -> GSI 21 (level, low) -> IRQ 21 Oct 13 15:43:48 brew kernel: HDA Intel 0000:00:1b.0: setting latency timer to 64 Oct 13 15:43:48 brew kernel: usb 2-1: device descriptor read/64, error -32 Oct 13 15:43:48 brew kernel: ehci_hcd 0000:00:1d.7: PCI INT A -> GSI 16 (level, low) -> IRQ 16 Oct 13 15:43:48 brew kernel: ehci_hcd 0000:00:1d.7: setting latency timer to 64 Oct 13 15:43:48 brew kernel: ehci_hcd 0000:00:1d.7: EHCI Host Controller Oct 13 15:43:48 brew kernel: ehci_hcd 0000:00:1d.7: new USB bus registered, assigned bus number 5 Oct 13 15:43:48 brew kernel: ehci_hcd 0000:00:1d.7: debug port 1 Oct 13 15:43:48 brew kernel: ehci_hcd 0000:00:1d.7: cache line size of 32 is not supported Oct 13 15:43:48 brew kernel: ehci_hcd 0000:00:1d.7: irq 16, io mem 0xf0204000 Oct 13 15:43:48 brew kernel: ehci_hcd 0000:00:1d.7: USB 2.0 started, EHCI 1.00, driver 10 Dec 2004 Oct 13 15:43:48 brew kernel: usb usb5: configuration #1 chosen from 1 choice Oct 13 15:43:48 brew kernel: hub 5-0:1.0: USB hub found Oct 13 15:43:48 brew kernel: hub 5-0:1.0: 8 ports detected Oct 13 15:43:48 brew kernel: nommu_map_single: overflow 12e9de620+8 of device mask ffffffff Oct 13 15:43:48 brew kernel: nommu_map_single: overflow 11ed07d80+64 of device mask ffffffff Oct 13 15:43:48 brew kernel: nommu_map_single: overflow 12e9de620+8 of device mask ffffffff Oct 13 15:43:48 brew kernel: nommu_map_single: overflow 11ed07d80+64 of device mask ffffffff Oct 13 15:43:48 brew kernel: nommu_map_single: overflow 12e9de620+8 of device mask ffffffff Oct 13 15:43:48 brew kernel: nommu_map_single: overflow 11ed07d80+64 of device mask ffffffff Oct 13 15:43:48 brew kernel: hub 2-0:1.0: unable to enumerate USB device on port 1
...
The whole kern.log and my config are available here :
http://chdir.org/~nbareil/kern.log
http://chdir.or/~nbareil/config-2.6.27-brew
Thanks!
Am Dienstag, 14. Oktober 2008 09:20:45 schrieb Nicolas Bareil:
Quoted 4 lines
Hi
On 2.6.27, I can't use any USB device (mass storage, mouse) anymore on my
x86_64system, the devices are not detected: lsusb returns nothing except the hub.
Is this a regression? Did it work on earlier kernels?
Quoted 2 lines
I have the following logs in kern.log when booting with my USB mouse plugged :
For confirmation boot with mem=2G on the kernel command line?
Regards
Oliver
Hi Oliver,
Quoted 4 lines · Oliver Neukum
Quoted 2 lines
On 2.6.27, I can't use any USB device (mass storage, mouse) anymore on my
x86_64system, the devices are not detected: lsusb returns nothing except the hub.Is this a regression? Did it work on earlier kernels?
Yes this is a regression: USB works in 2.6.26.x with (almost) the same configuration.
Quoted 3 lines
Quoted 1 line
I have the following logs in kern.log when booting with my USB mouse plugged :
For confirmation boot with mem=2G on the kernel command line?
With this workaround, the USB works with the following logs (don't know if it's useful):
Oct 14 12:30:22 brew kernel: ehci_hcd 0000:00:1d.7: PCI INT A -> GSI 16 (level, low) -> IRQ 16 Oct 14 12:30:22 brew kernel: ehci_hcd 0000:00:1d.7: setting latency timer to 64 Oct 14 12:30:22 brew kernel: ehci_hcd 0000:00:1d.7: EHCI Host Controller Oct 14 12:30:22 brew kernel: ehci_hcd 0000:00:1d.7: new USB bus registered, assigned bus number 1 Oct 14 12:30:22 brew kernel: ehci_hcd 0000:00:1d.7: debug port 1 Oct 14 12:30:22 brew kernel: ehci_hcd 0000:00:1d.7: cache line size of 32 is not supported Oct 14 12:30:22 brew kernel: ehci_hcd 0000:00:1d.7: irq 16, io mem 0xf0204000 Oct 14 12:30:22 brew kernel: USB Universal Host Controller Interface driver v3.0 Oct 14 12:30:22 brew kernel: ehci_hcd 0000:00:1d.7: USB 2.0 started, EHCI 1.00, driver 10 Dec 2004 Oct 14 12:30:22 brew kernel: usb usb1: configuration #1 chosen from 1 choice Oct 14 12:30:22 brew kernel: hub 1-0:1.0: USB hub found Oct 14 12:30:22 brew kernel: hub 1-0:1.0: 8 ports detected Oct 14 12:30:22 brew kernel: intel_rng: Firmware space is locked read-only. If you can't or Oct 14 12:30:22 brew kernel: intel_rng: don't want to disable this in firmware setup, and if Oct 14 12:30:22 brew kernel: intel_rng: you are certain that your system has a functional Oct 14 12:30:22 brew kernel: intel_rng: RNG, try using the 'no_fwh_detect' option. Oct 14 12:30:22 brew kernel: HDA Intel 0000:00:1b.0: PCI INT A -> GSI 21 (level, low) -> IRQ 21 Oct 14 12:30:22 brew kernel: HDA Intel 0000:00:1b.0: setting latency timer to 64 Oct 14 12:30:22 brew kernel: uhci_hcd 0000:00:1d.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16 Oct 14 12:30:22 brew kernel: uhci_hcd 0000:00:1d.0: setting latency timer to 64 Oct 14 12:30:22 brew kernel: uhci_hcd 0000:00:1d.0: UHCI Host Controller Oct 14 12:30:22 brew kernel: uhci_hcd 0000:00:1d.0: new USB bus registered, assigned bus number 2 Oct 14 12:30:22 brew kernel: uhci_hcd 0000:00:1d.0: irq 16, io base 0x00002000 Oct 14 12:30:22 brew kernel: usb usb2: configuration #1 chosen from 1 choice Oct 14 12:30:22 brew kernel: hub 2-0:1.0: USB hub found Oct 14 12:30:22 brew kernel: hub 2-0:1.0: 2 ports detected Oct 14 12:30:22 brew kernel: uhci_hcd 0000:00:1d.1: PCI INT B -> GSI 19 (level, low) -> IRQ 19 Oct 14 12:30:22 brew kernel: uhci_hcd 0000:00:1d.1: setting latency timer to 64 Oct 14 12:30:22 brew kernel: uhci_hcd 0000:00:1d.1: UHCI Host Controller Oct 14 12:30:22 brew kernel: uhci_hcd 0000:00:1d.1: new USB bus registered, assigned bus number 3 Oct 14 12:30:22 brew kernel: uhci_hcd 0000:00:1d.1: irq 19, io base 0x00002020 Oct 14 12:30:22 brew kernel: usb usb3: configuration #1 chosen from 1 choice Oct 14 12:30:22 brew kernel: hub 3-0:1.0: USB hub found Oct 14 12:30:22 brew kernel: hub 3-0:1.0: 2 ports detected Oct 14 12:30:22 brew kernel: uhci_hcd 0000:00:1d.2: PCI INT C -> GSI 18 (level, low) -> IRQ 18 Oct 14 12:30:22 brew kernel: uhci_hcd 0000:00:1d.2: setting latency timer to 64 Oct 14 12:30:22 brew kernel: uhci_hcd 0000:00:1d.2: UHCI Host Controller Oct 14 12:30:22 brew kernel: uhci_hcd 0000:00:1d.2: new USB bus registered, assigned bus number 4 Oct 14 12:30:22 brew kernel: uhci_hcd 0000:00:1d.2: irq 18, io base 0x00002040 Oct 14 12:30:22 brew kernel: usb usb4: configuration #1 chosen from 1 choice Oct 14 12:30:22 brew kernel: hub 4-0:1.0: USB hub found Oct 14 12:30:22 brew kernel: hub 4-0:1.0: 2 ports detected Oct 14 12:30:22 brew kernel: uhci_hcd 0000:00:1d.3: PCI INT D -> GSI 22 (level, low) -> IRQ 22 Oct 14 12:30:22 brew kernel: uhci_hcd 0000:00:1d.3: setting latency timer to 64 Oct 14 12:30:22 brew kernel: uhci_hcd 0000:00:1d.3: UHCI Host Controller Oct 14 12:30:22 brew kernel: uhci_hcd 0000:00:1d.3: new USB bus registered, assigned bus number 5 Oct 14 12:30:22 brew kernel: uhci_hcd 0000:00:1d.3: irq 22, io base 0x00002060 Oct 14 12:30:22 brew kernel: usb usb5: configuration #1 chosen from 1 choice Oct 14 12:30:22 brew kernel: hub 5-0:1.0: USB hub found Oct 14 12:30:22 brew kernel: hub 5-0:1.0: 2 ports detected Oct 14 12:30:22 brew kernel: usb 3-1: new low speed USB device using uhci_hcd and address 2
Thank you for your time!
Quoted 9 lines · On Tue, 14 Oct 2008 12:45:00 +0200 Nicolas Bareil <[email protected]> wrote:
Hi Oliver,
Quoted 4 lines · Oliver Neukum
Quoted 2 lines
On 2.6.27, I can't use any USB device (mass storage, mouse) anymore on my
x86_64system, the devices are not detected: lsusb returns nothing except the hub.Is this a regression? Did it work on earlier kernels?
Yes this is a regression: USB works in 2.6.26.x with (almost) the same configuration.
With old kernels, you can find something like the following line in the boot log?
PCI-DMA: Using software bounce buffering for IO (SWIOTLB)
Quoted 8 lines · FUJITA Tomonori
Quoted 3 lines
Quoted 1 line
Is this a regression? Did it work on earlier kernels?
Yes this is a regression: USB works in 2.6.26.x with (almost) the same configuration.
With old kernels, you can find something like the following line in the boot log?
PCI-DMA: Using software bounce buffering for IO (SWIOTLB)
Indeed, I have this line:
Oct 14 12:46:29 brew kernel: [ 0.004000] Calgary: detecting Calgary via BIOS EBDA area Oct 14 12:46:29 brew kernel: [ 0.004000] Calgary: Unable to locate Rio Grande table in EBDA - bailing! Oct 14 12:46:29 brew kernel: [ 0.004000] PCI-DMA: Using software bounce buffering for IO (SWIOTLB) Oct 14 12:46:29 brew kernel: [ 0.004000] Placing software IO TLB between 0x4000000 - 0x8000000 Oct 14 12:46:29 brew kernel: [ 0.004000] Memory: 4055544k/4980736k available (2225k kernel code, 138092k reserved, 1079k data, 392k init) Oct 14 12:46:29 brew kernel: [ 0.004000] CPA: page pool initialized 1 of 1 pages preallocated
The full kern.log is available here: http://chdir.org/~nbareil/kern.log-2.6.26-1
Thanks!
Quoted 10 lines · On Tue, 14 Oct 2008 13:05:39 +0200 Nicolas Bareil <[email protected]> wrote:
Quoted 8 lines · FUJITA Tomonori
Quoted 3 lines
Quoted 1 line
Is this a regression? Did it work on earlier kernels?
Yes this is a regression: USB works in 2.6.26.x with (almost) the same configuration.
With old kernels, you can find something like the following line in the boot log?
PCI-DMA: Using software bounce buffering for IO (SWIOTLB)
Quoted 8 lines
Indeed, I have this line:
Oct 14 12:46:29 brew kernel: [ 0.004000] Calgary: detecting Calgary via BIOS EBDA area Oct 14 12:46:29 brew kernel: [ 0.004000] Calgary: Unable to locate Rio Grande table in EBDA - bailing! Oct 14 12:46:29 brew kernel: [ 0.004000] PCI-DMA: Using software bounce buffering for IO (SWIOTLB) Oct 14 12:46:29 brew kernel: [ 0.004000] Placing software IO TLB between 0x4000000 - 0x8000000 Oct 14 12:46:29 brew kernel: [ 0.004000] Memory: 4055544k/4980736k available (2225k kernel code, 138092k reserved, 1079k data, 392k init) Oct 14 12:46:29 brew kernel: [ 0.004000] CPA: page pool initialized 1 of 1 pages preallocated
Probably, the changes to the initial memory setup code breaks the following code:
void __init pci_swiotlb_init(void)
{
/* don't initialize swiotlb if iommu=off (no_iommu=1) */
if (!iommu_detected && !no_iommu && max_pfn > MAX_DMA32_PFN)
swiotlb = 1;SWIOTLB should be used for your box but somehow pci-nommu is used.
Quoted 2 lines
The whole kern.log and my config are available here :
http://chdir.org/~nbareil/kern.log
The following part in the 2.6.27 boot log looks suspicious:
Oct 13 15:43:48 brew kernel: last_pfn = 0x130000 max_arch_pfn = 0x3ffffffff Oct 13 15:43:48 brew kernel: x86 PAT enabled: cpu 0, old 0x7040600070406, new 0x7010600070106 Oct 13 15:43:48 brew kernel: last_pfn = 0xcffc2 max_arch_pfn = 0x3ffffffff
CC'ed Yinghai and Ingo, who might have clues.
Quoted 35 lines · FUJITA Tomonori
Quoted 10 lines · On Tue, 14 Oct 2008 13:05:39 +0200 Nicolas Bareil <[email protected]> wrote:
Quoted 8 lines · FUJITA Tomonori
Quoted 3 lines
Quoted 1 line
Is this a regression? Did it work on earlier kernels?
Yes this is a regression: USB works in 2.6.26.x with (almost) the same configuration.
With old kernels, you can find something like the following line in the boot log?
PCI-DMA: Using software bounce buffering for IO (SWIOTLB)
Quoted 8 lines
Indeed, I have this line:
Oct 14 12:46:29 brew kernel: [ 0.004000] Calgary: detecting Calgary via BIOS EBDA area Oct 14 12:46:29 brew kernel: [ 0.004000] Calgary: Unable to locate Rio Grande table in EBDA - bailing! Oct 14 12:46:29 brew kernel: [ 0.004000] PCI-DMA: Using software bounce buffering for IO (SWIOTLB) Oct 14 12:46:29 brew kernel: [ 0.004000] Placing software IO TLB between 0x4000000 - 0x8000000 Oct 14 12:46:29 brew kernel: [ 0.004000] Memory: 4055544k/4980736k available (2225k kernel code, 138092k reserved, 1079k data, 392k init) Oct 14 12:46:29 brew kernel: [ 0.004000] CPA: page pool initialized 1 of 1 pages preallocatedProbably, the changes to the initial memory setup code breaks the following code:
void
__initpci_swiotlb_init(void){ /* don't initialize swiotlb if iommu=off (no_iommu=1) */ if (!iommu_detected && !no_iommu && max_pfn > MAX_DMA32_PFN) swiotlb = 1;SWIOTLB should be used for your box but somehow pci-nommu is used.
Quoted 6 lines
Quoted 2 lines
The whole kern.log and my config are available here :
http://chdir.org/~nbareil/kern.logThe following part in the 2.6.27 boot log looks suspicious:
Oct 13 15:43:48 brew kernel: last_pfn = 0x130000 max_arch_pfn = 0x3ffffffff
that is max_pfn
Quoted 2 lines
Oct 13 15:43:48 brew kernel: x86 PAT enabled: cpu 0, old 0x7040600070406, new 0x7010600070106 Oct 13 15:43:48 brew kernel: last_pfn = 0xcffc2 max_arch_pfn = 0x3ffffffff
that is max_low_pfn
YH
Am Dienstag, 14. Oktober 2008 12:45:00 schrieb Nicolas Bareil:
Hello,
Quoted 9 lines
Hi Oliver,
Quoted 4 lines · Oliver Neukum
Quoted 2 lines
On 2.6.27, I can't use any USB device (mass storage, mouse) anymore on my
x86_64system, the devices are not detected: lsusb returns nothing except the hub.Is this a regression? Did it work on earlier kernels?
Yes this is a regression: USB works in 2.6.26.x with (almost) the same configuration.
This a an IOMMU problem, not really a USB problem. Can you bisect the problem?
Regards
Oliver
Quoted 3 lines · Oliver Neukum
Quoted 1 line
Yes this is a regression: USB works in 2.6.26.x with (almost) the same configuration.
This a an IOMMU problem, not really a USB problem. Can you bisect the problem?
Ok, after 24 reboots, I have the following commit:
commit 6f21d806f698a5c437dcae1f03e8dd73e3f5aab1
Merge: e59e14b... 9f72632...
Author: Linus Torvalds <[email protected]>
Date: Sun Sep 21 12:40:56 2008 -0700
Merge branch 'kvm-updates/2.6.27' of git://git.kernel.org/pub/scm/linux/kernel/git/avi/kvm
* 'kvm-updates/2.6.27' of git://git.kernel.org/pub/scm/linux/kernel/git/avi/kvm:
KVM: ia64: 'struct fdesc' build fix
Do you think I'm good to bisect that tree or do you have an idea?
Quoted 11 lines · Nicolas Bareil
commit
6f21d806f698a5c437dcae1f03e8dd73e3f5aab1
Merge: e59e14b... 9f72632...
Author: Linus Torvalds <[email protected]>
Date: Sun Sep 21 12:40:56 2008 -0700Merge branch 'kvm-updates/2.6.27' of
git://git.kernel.org/pub/scm/linux/kernel/git/avi/kvm* 'kvm-updates/2.6.27' of
git://git.kernel.org/pub/scm/linux/kernel/git/avi/kvm:
KVM: ia64: 'struct fdesc' build fixDo you think I'm good to bisect that tree or do you have an idea?
Strangely, if I disable MMC, there is no error at all. I'm currently booting on a 2.6.27 kernel and everything seems to be ok. My kernel log is at http://chdir.org/~nbareil/kern.log.2.6.27.without.mmc for information.
Thanks
Quoted 10 lines · Oliver Neukum
Am Dienstag, 14. Oktober 2008 12:45:00 schrieb Nicolas Bareil:
Quoted 7 lines · Oliver Neukum
Quoted 4 lines
Quoted 2 lines
On 2.6.27, I can't use any USB device (mass storage, mouse) anymore on my
x86_64system, the devices are not detected: lsusb returns nothing except the hub.Is this a regression? Did it work on earlier kernels?
Yes this is a regression: USB works in 2.6.26.x with (almost) the same configuration.
This a an IOMMU problem, not really a USB problem. Can you bisect the problem?
It looks like I'm seeing something similar with the current linux-2.6.git. However, in my case, 2.6.27 seemed to work and I started seeing this only after the 2.6.27 release (though, git bisect ended up in 2.6.27-rc3). In addition, I'm seeing way more issues than just USB dying (both wired and wireless networking were hosed; r8169 was returning random memory or all zeroes in the payload of the frames, etc.).
After a painful git bisect (two other bugs that prevented boot and some build issues in the tested commits) the first bad commit turned out to be cf169702ba6928cee9d4f4adf3e932b643b8db7a (see below; author cc'ed). This is only adding a new PCI device id, so the commit itself seems to just trigger the issue somewhere else.
I've verified that I can get rid of the nommu_map_single() messages by reverting this patch (i.e., made kernel not find the northbridge) with the current linux-2.6.git head. Obviously, that does not sound like a proper fix here. Would anyone have any idea on what is the real issue here or what could be done to find it?
I'm seeing this issue on a HP Pavilion dv5 laptop that has AMD Turion X2 Ultra Dual-Core Mobile ZM-80. lspci shows the PCI id that was added (1022:1303). mem=2G works around the problem. pci_swiotlb_init() did not set swiotlb to 1, but hardcoding it to 1 there did not help either.
$ git bisect bad cf169702ba6928cee9d4f4adf3e932b643b8db7a is first bad commit commit cf169702ba6928cee9d4f4adf3e932b643b8db7a Author: Joerg Roedel <[email protected]> Date: Tue Sep 2 13:13:40 2008 +0200
x86, gart: add detection of AMD family 0x11 northbridges
This patch adds the detection of the northbridges in the AMD family 0x11 processors. It also fixes the magic numbers there while changing this code.
:040000 040000 183c03461f96be8bbf96cb51bae9bb0fb93dae8958638c91f03393225f3c48cb12476466f9c5e2be M arch
----------------------------- arch/x86/kernel/k8.c ----------------------------- index 7377ccb..304d8ba 100644
@@ -16,8 +16,9 @@ EXPORT_SYMBOL(num_k8_northbridges);
static u32 *flush_words;
struct pci_device_id k8_nb_ids[] = {
- { PCI_DEVICE(PCI_VENDOR_ID_AMD, 0x1103) },
- { PCI_DEVICE(PCI_VENDOR_ID_AMD, 0x1203) },
+ { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_K8_NB_MISC) },
+ { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_10H_NB_MISC) },
+ { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_11H_NB_MISC) },
{}
};
EXPORT_SYMBOL(k8_nb_ids);
- Jouni
PATCH 1/1 x86, gart: fix gart detection for Fam11h CPUs
The AMD Fam11h CPUs have a K8 northbridge. This northbridge is different from other familys because it lacks GART support (as I just learned). But the kernel implicitly expects a GART if it finds an AMD northbridge. Fix this by removing the Fam11h northbridge id from the scan list of K8 northbridges. This patch also changes the message in the GART driver about missing K8 northbridges to tell that the GART is missing which is the correct information in this case.
--- arch/x86/kernel/k8.c | 1 - arch/x86/kernel/pci-gart_64.c | 2 +- 2 files changed, 1 insertions(+), 2 deletions(-)
arch/x86/kernel/k8.c +0 -1
@@ -18,7 +18,6 @@ static u32 *flush_words;struct pci_device_id k8_nb_ids[] = { {PCI_DEVICE (PCI_VENDOR_ID_AMD ,PCI_DEVICE_ID_AMD_K8_NB_MISC ) }, {PCI_DEVICE (PCI_VENDOR_ID_AMD ,PCI_DEVICE_ID_AMD_10H_NB_MISC ) }, - {PCI_DEVICE (PCI_VENDOR_ID_AMD ,PCI_DEVICE_ID_AMD_11H_NB_MISC ) }, {} };EXPORT_SYMBOL (k8_nb_ids);
arch/x86/kernel/pci-gart_64.c +1 -1
@@ -744,7 +744,7 @@ void __init gart_iommu_init(void)long i;if (cache_k8_northbridges () <0 || num_k8_northbridges ==0 ) { -printk (KERN_INFO "PCI-GART: No AMD northbridge found.\n" ); +printk (KERN_INFO "PCI-GART: No AMD GART found.\n" );return ; }
Signature · 1 line
1.5.6.4
Ingo Molnar
AppliedRe: [PATCH] x86, gart: fix gart detection for Fam11h CPUs
Quoted 9 lines · Joerg Roedel
The AMD Fam11h CPUs have a K8 northbridge. This northbridge is different from other familys because it lacks GART support (as I just learned). But the kernel implicitly expects a GART if it finds an AMD northbridge. Fix this by removing the Fam11h northbridge id from the scan list of K8 northbridges. This patch also changes the message in the GART driver about missing K8 northbridges to tell that the GART is missing which is the correct information in this case.
Quoted 5 lines
Signed-off-by Joerg Roedel <[email protected]>--- arch/x86/kernel/k8.c | 1 - arch/x86/kernel/pci-gart_64.c | 2 +- 2 files changed, 1 insertions(+), 2 deletions(-)
applied to tip/x86/urgent, thanks Joerg!
Ingo
Re: [PATCH] x86, gart: fix gart detection for Fam11h CPUs
In reply to Joerg RoedelQuoted 9 lines · Joerg Roedel
The AMD Fam11h CPUs have a K8 northbridge. This northbridge is different from other familys because it lacks GART support (as I just learned). But the kernel implicitly expects a GART if it finds an AMD northbridge. Fix this by removing the Fam11h northbridge id from the scan list of K8 northbridges. This patch also changes the message in the GART driver about missing K8 northbridges to tell that the GART is missing which is the correct information in this case.
Signed-off-by Joerg Roedel <[email protected]>
Looks like this needs to be added to the 2.6.27-stable tree also, right?
thanks,
greg k-h
Quoted 38 lines
--- arch/x86/kernel/k8.c | 1 - arch/x86/kernel/pci-gart_64.c | 2 +- 2 files changed, 1 insertions(+), 2 deletions(-)arch/x86/kernel/k8.c +0
-1@@ -18,7 +18,6 @@ static u32 *flush_words;struct pci_device_id k8_nb_ids[] = { {PCI_DEVICE (PCI_VENDOR_ID_AMD ,PCI_DEVICE_ID_AMD_K8_NB_MISC ) }, {PCI_DEVICE (PCI_VENDOR_ID_AMD ,PCI_DEVICE_ID_AMD_10H_NB_MISC ) }, - {PCI_DEVICE (PCI_VENDOR_ID_AMD ,PCI_DEVICE_ID_AMD_11H_NB_MISC ) }, {} };EXPORT_SYMBOL (k8_nb_ids);arch/x86/kernel/pci-gart_64.c +1
-1@@ -744,7 +744,7 @@ void __init gart_iommu_init(void)long i;if (cache_k8_northbridges () <0 || num_k8_northbridges ==0 ) { -printk (KERN_INFO "PCI-GART: No AMD northbridge found.\n" ); +printk (KERN_INFO "PCI-GART: No AMD GART found.\n" );return ; }Signature · 7 lines
1.5.6.4 -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [PATCH] x86, gart: fix gart detection for Fam11h CPUs
Quoted 9 lines · Greg KH
Quoted 8 lines · Joerg Roedel
The AMD Fam11h CPUs have a K8 northbridge. This northbridge is different from other familys because it lacks GART support (as I just learned). But the kernel implicitly expects a GART if it finds an AMD northbridge. Fix this by removing the Fam11h northbridge id from the scan list of K8 northbridges. This patch also changes the message in the GART driver about missing K8 northbridges to tell that the GART is missing which is the correct information in this case.
Quoted 3 lines
Quoted 1 line
Signed-off-by Joerg Roedel <[email protected]>Looks like this needs to be added to the 2.6.27-stable tree also, right?
No, this patch not. The PCI id was added during the .28 merge window.
Joerg
Signature · 6 lines
| AMD Saxony Limited Liability Company & Co. KG
Operating | Wilschdorfer Landstr. 101, 01109 Dresden, Germany
System | Register Court Dresden: HRA 4896
Research | General Partner authorized to represent:
Center | AMD Saxony LLC (Wilmington, Delaware, US)
| General Manager of AMD Saxony LLC: Dr. Hans-R. Deppe, Thomas McCoyRe: [PATCH] x86, gart: fix gart detection for Fam11h CPUs
Quoted 10 lines · Joerg Roedel
Quoted 9 lines · Greg KH
Quoted 8 lines · Joerg Roedel
The AMD Fam11h CPUs have a K8 northbridge. This northbridge is different from other familys because it lacks GART support (as I just learned). But the kernel implicitly expects a GART if it finds an AMD northbridge. Fix this by removing the Fam11h northbridge id from the scan list of K8 northbridges. This patch also changes the message in the GART driver about missing K8 northbridges to tell that the GART is missing which is the correct information in this case.
Quoted 5 lines
Quoted 3 lines
Quoted 1 line
Signed-off-by Joerg Roedel <[email protected]>Looks like this needs to be added to the 2.6.27-stable tree also, right?
No, this patch not. The PCI id was added during the .28 merge window.
Ah, you are right, sorry about that, was looking at the wrong tree here.
greg k-h