2.6.6-mm-rc3-mm2 USB 2.0 after suspend issue

2.6.6-mm-rc3-mm2 USB 2.0 after suspend issue

[Ari Pollak]

That reminds me, I seem to be having problems with EHCI in 2.6.6-rc3-mm2 
after a resume. With this: "0000:00:1d.7 USB Controller: Intel Corp. 
82801DB (ICH4) USB2 EHCI Controller (rev 01)" if I suspend & resume with 
APM, then plug in a USB 2.0 device, it won't work at all until I "rmmod 
ehci_hcd; modprobe ehci_hcd". I'd imagine it has something to do with 
these messages:
uhci_hcd 0000:00:1d.1: suspend_hc
ehci_hcd 0000:00:1d.7: suspend D0 --> D3
ehci_hcd 0000:00:1d.7: No PM capability
uhci_hcd 0000:00:1d.2: suspend D4 --> D3
uhci_hcd 0000:00:1d.2: suspend_hc
uhci_hcd 0000:00:1d.1: suspend D4 --> D3
uhci_hcd 0000:00:1d.1: suspend_hc
uhci_hcd 0000:00:1d.0: suspend D4 --> D3
uhci_hcd 0000:00:1d.0: suspend_hc
uhci_hcd 0000:00:1d.0: resume from state D4
PCI: Setting latency timer of device 0000:00:1d.0 to 64
uhci_hcd 0000:00:1d.1: resume from state D4
PCI: Setting latency timer of device 0000:00:1d.1 to 64
uhci_hcd 0000:00:1d.2: resume from state D4
PCI: Setting latency timer of device 0000:00:1d.2 to 64
ehci_hcd 0000:00:1d.7: resume from state D0
ehci_hcd 0000:00:1d.7: can't resume, not suspended!

Granted, this isn't a huge problem because I can easily work around it 
by reloading the ehci_hcd module. But it's still a little annoying. It 
doesn't look like this has changed as of 2.6.6-mm2 either.

Prakash K. Cheemplavam wrote:
> Please look at the end of dmesg output.
> 
> There appear lines like
> 
> usb usb2: string descriptor 0 read error: -108
> 
> bug or feature? They weren't there with 2.6.6-mm1. I have no usb2.0 
> stuff to actually test. My usb1 stuff seems to work though.
> 
> Prakash

Re: 2.6.6-mm-rc3-mm2 USB 2.0 after suspend issue

[David Brownell]

You said  2.6.6-rc3-mm2 gives you these strange
messages, but how does 2.6.6-mm2 behave?

> ehci_hcd 0000:00:1d.7: suspend D0 --> D3
> ehci_hcd 0000:00:1d.7: No PM capability

2.6.6-mm2 won't give that message pair.  The EHCI
suspend/resume should work better though.

> uhci_hcd 0000:00:1d.2: suspend D4 --> D3
> uhci_hcd 0000:00:1d.2: suspend_hc
> uhci_hcd 0000:00:1d.1: suspend D4 --> D3
> uhci_hcd 0000:00:1d.1: suspend_hc
> uhci_hcd 0000:00:1d.0: suspend D4 --> D3
> uhci_hcd 0000:00:1d.0: suspend_hc

Those messages are a bit strange.  If it still
misbehaves in 2.6.6-mm2, please send "lspci -vv"
info for these USB controllers.  (Preferably
to linux-usb-devel, where it's easier to see.)

> ...
> ehci_hcd 0000:00:1d.7: resume from state D0
> ehci_hcd 0000:00:1d.7: can't resume, not suspended!

Again, 2.6.6-mm2 shouldn't do that.

- Dave

Re: 2.6.6-mm-rc3-mm2 USB 2.0 after suspend issue

[Ari Pollak]

2.6.6-mm2 does indeed fix the EHCI problem and those messages; I guess I 
didn't read the patches closely enough. Instead, though, I have a new 
problem related to suspend/resume in 2.6.6-mm2 where an attached USB 
mouse will stop responding after a resume, but unplugging & plugging the 
device back in will work fine.
I was going to send a message off to usb-devel about it, but Thunderbird 
ended up losing my entire e-mail and I was too lazy to re-type it.

David Brownell wrote:
> Those messages are a bit strange.  If it still
> misbehaves in 2.6.6-mm2, please send "lspci -vv"
> info for these USB controllers.  (Preferably
> to linux-usb-devel, where it's easier to see.)
> 
>> ...
>> ehci_hcd 0000:00:1d.7: resume from state D0
>> ehci_hcd 0000:00:1d.7: can't resume, not suspended!
> 
> 
> Again, 2.6.6-mm2 shouldn't do that.
> 
> - Dave
> 
>