Re: ACPI disables USB ports on Lenovo X200

[Andriy Gapon]

on 30/05/2010 21:32 Damian Gerow said the following:
> 
> 
> Trying a few other things, I tried to boot a GENERIC kernel.  Everything works
> just fine.  Though the mouse (and USB keys) still turn off shortly after
> ACPI gets going, they turn back on when the USB stack initializes, so it's
> not really an issue.
> 
> Specifically, it seems that at some point in time, I accidentally changed
> 'device uhci' to 'device ohci'.
> 
> Whoops.
> 
> Sorry for the noise, folks.
> 
> 

Thanks for sharing this.
I will add it to my list of things to double-check.

-- 
Andriy Gapon
___
freebsd-acpi@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-acpi
To unsubscribe, send any mail to "freebsd-acpi-unsubscr...@freebsd.org"

Re: ACPI disables USB ports on Lenovo X200

[Damian Gerow]

Damian Gerow wrote:
: : > : Yup.  I couldn't remember if hw.usb.debug was a blanket statement for
: : > : hw.usb*debug, so I also set hw.usb.ehci.debug, hw.usb.ohci.debug,
: : > : hw.usb.dev.debug, and hw.usb.ums.debug to 1.
: : > 
: : > For kicks, I just booted a kernel without any USB support at all, and
: : > experience the same symptoms: the mouse is functional (well, appears to 
be,
: : > anyhow) when the boot loader is active, but goes dark around when ACPI
: : > starts to do its thing.
: : 
: : Please also try to experiment with hw.pci.usb_early_takeover.
: 
: No love on that one, either: the mouse turns off at the same point in the
: boot sequence.



Trying a few other things, I tried to boot a GENERIC kernel.  Everything works
just fine.  Though the mouse (and USB keys) still turn off shortly after
ACPI gets going, they turn back on when the USB stack initializes, so it's
not really an issue.

Specifically, it seems that at some point in time, I accidentally changed
'device uhci' to 'device ohci'.

Whoops.

Sorry for the noise, folks.


___
freebsd-acpi@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-acpi
To unsubscribe, send any mail to "freebsd-acpi-unsubscr...@freebsd.org"

Re: ACPI disables USB ports on Lenovo X200

[Damian Gerow]

Andriy Gapon wrote:
: > : Yup.  I couldn't remember if hw.usb.debug was a blanket statement for
: > : hw.usb*debug, so I also set hw.usb.ehci.debug, hw.usb.ohci.debug,
: > : hw.usb.dev.debug, and hw.usb.ums.debug to 1.
: > 
: > For kicks, I just booted a kernel without any USB support at all, and
: > experience the same symptoms: the mouse is functional (well, appears to be,
: > anyhow) when the boot loader is active, but goes dark around when ACPI
: > starts to do its thing.
: 
: Please also try to experiment with hw.pci.usb_early_takeover.

No love on that one, either: the mouse turns off at the same point in the
boot sequence.
___
freebsd-acpi@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-acpi
To unsubscribe, send any mail to "freebsd-acpi-unsubscr...@freebsd.org"

Re: ACPI disables USB ports on Lenovo X200

[Damian Gerow]

Brandon Gooch wrote:
: > For kicks, I just booted a kernel without any USB support at all, and
: > experience the same symptoms: the mouse is functional (well, appears to be,
: > anyhow) when the boot loader is active, but goes dark around when ACPI
: > starts to do its thing.
: 
: Damien, sorry if you've mentioned this already, but have you taken a
: look at the BIOS settings? In particular, check out "Legacy USB
: Support" options and experiment with changing the value(s)...

I've only got two USB-related options in my BIOS:

USB BIOS Support: Toggles the support for booting from USB devices
Always On USB: Toggles the power status of USB ports as they related to the
   power status of the laptop

So, no legacy USB support knob.  And yes, I've tried setting changing both
of those options, to no avail.
___
freebsd-acpi@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-acpi
To unsubscribe, send any mail to "freebsd-acpi-unsubscr...@freebsd.org"

Re: ACPI disables USB ports on Lenovo X200

[Andriy Gapon]

on 30/05/2010 08:53 Damian Gerow said the following:
> Damian Gerow wrote:
> : : Hi Damian,
> : : Have you also tried setting hw.usb.debug=1 and see what's reported
> : : by the new USB stack?
> : 
> : Yup.  I couldn't remember if hw.usb.debug was a blanket statement for
> : hw.usb*debug, so I also set hw.usb.ehci.debug, hw.usb.ohci.debug,
> : hw.usb.dev.debug, and hw.usb.ums.debug to 1.
> 
> For kicks, I just booted a kernel without any USB support at all, and
> experience the same symptoms: the mouse is functional (well, appears to be,
> anyhow) when the boot loader is active, but goes dark around when ACPI
> starts to do its thing.

Please also try to experiment with hw.pci.usb_early_takeover.

-- 
Andriy Gapon
___
freebsd-acpi@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-acpi
To unsubscribe, send any mail to "freebsd-acpi-unsubscr...@freebsd.org"

Re: ACPI disables USB ports on Lenovo X200

[Brandon Gooch]

On Sun, May 30, 2010 at 12:53 AM, Damian Gerow  wrote:
> Damian Gerow wrote:
> : : Hi Damian,
> : :     Have you also tried setting hw.usb.debug=1 and see what's reported
> : : by the new USB stack?
> :
> : Yup.  I couldn't remember if hw.usb.debug was a blanket statement for
> : hw.usb*debug, so I also set hw.usb.ehci.debug, hw.usb.ohci.debug,
> : hw.usb.dev.debug, and hw.usb.ums.debug to 1.
>
> For kicks, I just booted a kernel without any USB support at all, and
> experience the same symptoms: the mouse is functional (well, appears to be,
> anyhow) when the boot loader is active, but goes dark around when ACPI
> starts to do its thing.

Damien, sorry if you've mentioned this already, but have you taken a
look at the BIOS settings? In particular, check out "Legacy USB
Support" options and experiment with changing the value(s)...

-Brandon
___
freebsd-acpi@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-acpi
To unsubscribe, send any mail to "freebsd-acpi-unsubscr...@freebsd.org"

Re: ACPI disables USB ports on Lenovo X200

[Damian Gerow]

Damian Gerow wrote:
: : Hi Damian,
: : Have you also tried setting hw.usb.debug=1 and see what's reported
: : by the new USB stack?
: 
: Yup.  I couldn't remember if hw.usb.debug was a blanket statement for
: hw.usb*debug, so I also set hw.usb.ehci.debug, hw.usb.ohci.debug,
: hw.usb.dev.debug, and hw.usb.ums.debug to 1.

For kicks, I just booted a kernel without any USB support at all, and
experience the same symptoms: the mouse is functional (well, appears to be,
anyhow) when the boot loader is active, but goes dark around when ACPI
starts to do its thing.
___
freebsd-acpi@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-acpi
To unsubscribe, send any mail to "freebsd-acpi-unsubscr...@freebsd.org"

Re: ACPI disables USB ports on Lenovo X200

[Damian Gerow]

Additional note: completely forgot to include a uname -a:

FreeBSD plebeian.afflictions.org 8.1-PRERELEASE FreeBSD 8.1-PRERELEASE #10 
r208630: Sat May 29 14:36:32 EDT 2010 
r...@plebeian.afflictions.org:/usr/obj/usr/src/sys/PLEBEIAN  amd64

Garrett Cooper wrote:
: > The laptop is a Lenovo X200, 7454-CTO, running BIOS 3.16. ?Which, um, is
: > more recent than the latest release from Lenovo, but was current as of about
: > six months ago.
: 
: Hi Damian,
: Have you also tried setting hw.usb.debug=1 and see what's reported
: by the new USB stack?

Yup.  I couldn't remember if hw.usb.debug was a blanket statement for
hw.usb*debug, so I also set hw.usb.ehci.debug, hw.usb.ohci.debug,
hw.usb.dev.debug, and hw.usb.ums.debug to 1.

Here's what happens when I plug in that mouse:

-
ehci_set_hw_power:
uhub_reattach_port: reattaching port 2
uhub_reattach_port: Port 2 is in Host Mode
ehci_disown: index=2 lowspeed=1
usbd_transfer_power_ref: Adding type 2 to power state
usbd_transfer_power_ref: needs power
ehci_set_hw_power:
ehci_set_hw_power: Async is active
usb_needs_explore:
usb_bus_powerd: bus=0xff8000373c90
ehci_set_hw_power:
ehci_set_hw_power: Async is active
-

Should I be sending this over to stable@ instead?
___
freebsd-acpi@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-acpi
To unsubscribe, send any mail to "freebsd-acpi-unsubscr...@freebsd.org"

Re: ACPI disables USB ports on Lenovo X200

[Garrett Cooper]

On Sat, May 29, 2010 at 12:25 PM, Damian Gerow  wrote:
> Some time ago, the external USB ports on my Lenovo X200 stopped working.
> Specifically, if I plug in an optical mouse, I see it light up for a brief
> moment, then turn off.  Nothing ever shows up in the system logs, even with
> USB debugging enabled.  The laptop also has an internal SD card reader,
> connected via USB, that works just fine
>
> After playing around, I've discovered that the USB ports work while the
> boot loader is active, but seem to be disabled around the time
> \\_SB_.PCI0.EHC0.U7CS and \\_SB_.PCI0.EHC1.U8CS are registered (discovered?)
> during the boot process.  In the snippet below, the system pauses briefly
> after the "PCIe: Memory Mapped configuration" line is displayed, and the USB
> mouse is still active at that time.  As soon as things pick up again,
> though, the mouse turns off.
>
> -
> acpi0:  on motherboard
> PCIe: Memory Mapped configuration base @ 0xe000
> CPU0: local APIC error 0x40
> ioapic0: routing intpin 9 (ISA IRQ 9) to lapic 0 vector 48
> acpi0: [MPSAFE]
> acpi0: [ITHREAD]
> acpi_ec0:  port 0x62,0x66 on acpi0
> AcpiOsDerivePciId: \\_SB_.PCI0.MHCS -> bus 0 dev 0 func 0
> AcpiOsDerivePciId: \\_SB_.PCI0.EHC0.U7CS -> bus 0 dev 29 func 7
> AcpiOsDerivePciId: \\_SB_.PCI0.EHC1.U8CS -> bus 0 dev 26 func 7
> acpi0: Power Button (fixed)
> acpi0: wakeup code va 0xff80f000 pa 0x4000
> AcpiOsDerivePciId: \\_SB_.PCI0.LPC_.LPCS -> bus 0 dev 31 func 0
> acpi0: reservation of 0, a (3) failed
> acpi0: reservation of 10, bff0 (3) failed
> ACPI timer: 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 -> 10
> Timecounter "ACPI-fast" frequency 3579545 Hz quality 1000
> acpi_timer0: <24-bit timer at 3.579545MHz> port 0x1008-0x100b on acpi0
> -
>
> Full copies of a verbose dmesg (including USB debugging enabled) as well as
> a copy of my ASL can be found at  and
> , respectively.
>
> Though this used to work, I don't know exactly when it broke.  I've tried
> poking around at a few BIOS settings to no avail, and booting without ACPI
> isn't an option (the system doesn't boot).
>
> The laptop is a Lenovo X200, 7454-CTO, running BIOS 3.16.  Which, um, is
> more recent than the latest release from Lenovo, but was current as of about
> six months ago.

Hi Damian,
Have you also tried setting hw.usb.debug=1 and see what's reported
by the new USB stack?
HTH,
-Garrett
___
freebsd-acpi@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-acpi
To unsubscribe, send any mail to "freebsd-acpi-unsubscr...@freebsd.org"