ThinkPad support

John Baldwin jhb at freebsd.org
Mon Sep 22 19:22:05 UTC 2014


On Wednesday, September 17, 2014 08:09:17 PM Ian Smith wrote:
> On Tue, 16 Sep 2014 22:38:35 +0100, isdtor wrote:
> 
> [.. I can't comment on your devd issue/s ..]
> 
>  > Sep 16 21:48:23 host kernel: pci0: failed to set ACPI power state D2 on
>  > \_SB_.PCI0.EXP1: AE_BAD_PARAMETER
> 
> This appears to be common to all Thinkpads for the last 6 years or so,
> when booting with verbose messages. This message is emitted late on the
> suspend path, and seems to indicate some device that has advertised D2
> as an available state, where D2 isn't apparently really supported.
> 
> On the resume path, not much later in the messages sequence, you should
> see power being set back to D0 state, successfully.  Perhaps - as in the
> case of my X200 - twice in a row, for devices \_SB_.PCI0.EXP0 thru .EXP3
> 
> I don't know if this failure means that the device was left in D0 state
> (run) or in D3 state (off) when power was removed in S3 suspend state.

The kernel probably left them in D0, but I'm sure the hardware removed power
in S3.
 
> I never have been able to connect the dots between these .EXPn devices
> and particular PCI devices in dmesg.  So far they appear to be harmless.

devinfo -v shows you ACPI namespace names for new-bus devices:

% devinfo -v | grep EXP
        pcib1 pnpinfo vendor=0x8086 device=0x1c10 subvendor=0x17aa subdevice=0x21da class=0x060400 at slot=28 function=0 handle=\_SB_.PCI0.EXP1
        pcib2 pnpinfo vendor=0x8086 device=0x1c12 subvendor=0x17aa subdevice=0x21da class=0x060400 at slot=28 function=1 handle=\_SB_.PCI0.EXP2
        pcib3 pnpinfo vendor=0x8086 device=0x1c16 subvendor=0x17aa subdevice=0x21da class=0x060400 at slot=28 function=3 handle=\_SB_.PCI0.EXP4
        pcib4 pnpinfo vendor=0x8086 device=0x1c18 subvendor=0x17aa subdevice=0x21da class=0x060400 at slot=28 function=4 handle=\_SB_.PCI0.EXP5
            sdhci_pci0 pnpinfo vendor=0x1180 device=0xe823 subvendor=0x17aa subdevice=0x21da class=0x088001 at slot=0 function=0 handle=\_SB_.PCI0.EXP5.SLOT
 
Certainly these root-ports do not support D2:

pcib1 at pci0:0:28:0:      class=0x060400 card=0x21da17aa chip=0x1c108086 rev=0xb4 hdr=0x01
    cap 10[40] = PCI-Express 2 root port slot max data 128(128) link x0(x1)
                 speed 0.0(5.0) ASPM L0s/L1(L0s/L1)
    cap 05[80] = MSI supports 1 message 
    cap 0d[90] = PCI Bridge card=0x21da17aa
    cap 01[a0] = powerspec 2  supports D0 D3  current D0

-- 
John Baldwin


More information about the freebsd-mobile mailing list