2.6.27-rc2 USB suspend regression

DISCUSSION Jeremy Fitzhardinge 4 participants 10 messages Aug 20, 2008 – Aug 28, 2008

Subsystems (2)

USB Bluetooth drivers

lore

Share & tools

lore

Jeremy Fitzhardinge

author

2008-08-20 20:49 UTC #0 lore raw

Using the Fedora Rawhide kernel RPM "2.6.27-0.238.rc2.fc10.i686" (kernel-2.6.27-0.238.rc2.fc10.i686), I'm seeing suspend to ram failures. The suspend seems to go OK, but then I just get endless:

usb 5-1: uhci_result_common: failed with status 440000

messages, with occasional flickers of something else being printed (perhaps). The machine is otherwise unresponsive.

Rafael J. Wysocki

2008-08-20 20:59 UTC #1 lore raw

Quoted 8 lines · Jeremy Fitzhardinge

Using the Fedora Rawhide kernel RPM "2.6.27-0.238.rc2.fc10.i686" (kernel-2.6.27-0.238.rc2.fc10.i686), I'm seeing suspend to ram failures. The suspend seems to go OK, but then I just get endless:
usb 5-1: uhci_result_common: failed with status 440000
messages, with occasional flickers of something else being printed (perhaps). The machine is otherwise unresponsive.

I haven't seen anything like this so far with 2.6.27-rc. No idea what might have caused that to happen.

Thanks,
Rafael

Alan Stern

2008-08-20 21:28 UTC #2 lore raw

In reply to Jeremy Fitzhardinge

Quoted 8 lines · Jeremy Fitzhardinge

Using the Fedora Rawhide kernel RPM "2.6.27-0.238.rc2.fc10.i686" (kernel-2.6.27-0.238.rc2.fc10.i686), I'm seeing suspend to ram failures. The suspend seems to go OK, but then I just get endless:
usb 5-1: uhci_result_common: failed with status 440000
messages, with occasional flickers of something else being printed (perhaps). The machine is otherwise unresponsive.

You could comment out that error message in order to see what else is in the log. It is located in drivers/usb/host/uhci-q.c in the uhci_result_common() function.

You could try running -rc3 instead of -rc2, although that probably won't make much difference.

What USB devices do you have on the failing system? In particular, what is 5-1? If you unplug it before suspending, does the problem go away? (It certainly should!)

Alan Stern

Jeremy Fitzhardinge

author

2008-08-20 21:42 UTC #3 lore raw

Quoted 17 lines · Alan Stern

Quoted 9 lines · Jeremy Fitzhardinge
Using the Fedora Rawhide kernel RPM "2.6.27-0.238.rc2.fc10.i686" (kernel-2.6.27-0.238.rc2.fc10.i686), I'm seeing suspend to ram failures. The suspend seems to go OK, but then I just get endless:
usb 5-1: uhci_result_common: failed with status 440000
messages, with occasional flickers of something else being printed (perhaps). The machine is otherwise unresponsive.
You could comment out that error message in order to see what else is in the log. It is located in drivers/usb/host/uhci-q.c in the uhci_result_common() function.

Quoted 8 lines

You could try running -rc3 instead of -rc2, although that probably won't make much difference.
What USB devices do you have on the failing system? In particular, what is 5-1? If you unplug it before suspending, does the problem go away? (It certainly should!)

No external usb devices. There are three internal USB devices:
Bluetooth, a Sierra CDMA modem thingy and a fingerprint reader.

$ lsusb
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 005 Device 003: ID 0483:2016 SGS Thomson Microelectronics
Fingerprint Reader
Bus 005 Device 002: ID 0a5c:2110 Broadcom Corp. Bluetooth Controller
Bus 005 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 003 Device 003: ID 1199:0218 Sierra Wireless, Inc. MC5720 Wireless Modem
Bus 002 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub

The rfkill switch "unplugs" the two wireless devices, so I'll try that before suspending next time (if I remember).

Alan Stern

2008-08-20 21:56 UTC #4 lore raw

Quoted 41 lines · Jeremy Fitzhardinge

Quoted 17 lines · Alan Stern
Quoted 9 lines · Jeremy Fitzhardinge
Using the Fedora Rawhide kernel RPM "2.6.27-0.238.rc2.fc10.i686" (kernel-2.6.27-0.238.rc2.fc10.i686), I'm seeing suspend to ram failures. The suspend seems to go OK, but then I just get endless:
usb 5-1: uhci_result_common: failed with status 440000
messages, with occasional flickers of something else being printed (perhaps). The machine is otherwise unresponsive.
You could comment out that error message in order to see what else is in the log. It is located in drivers/usb/host/uhci-q.c in the uhci_result_common() function.
Quoted 8 lines
You could try running -rc3 instead of -rc2, although that probably won't make much difference.
What USB devices do you have on the failing system? In particular, what is 5-1? If you unplug it before suspending, does the problem go away? (It certainly should!)
No external usb devices. There are three internal USB devices:
Bluetooth, a Sierra CDMA modem thingy and a fingerprint reader.
$ lsusb
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 005 Device 003: ID 0483:2016 SGS Thomson Microelectronics
Fingerprint Reader
Bus 005 Device 002: ID 0a5c:2110 Broadcom Corp. Bluetooth Controller
Bus 005 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 003 Device 003: ID 1199:0218 Sierra Wireless, Inc. MC5720 Wireless Modem
Bus 002 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub

Quoted 2 lines

The rfkill switch "unplugs" the two wireless devices, so I'll try that before suspending next time (if I remember).

Obviously 5-1 is either the Bluetooth controller or the fingerprint reader. It's understandable that either one of them might have problems with a suspend/resume cycle, but the kernel ought to be able to recover by itself.

It would help to see the USB debugging log (without all those useless uhci_result_common messages).

Alan Stern

2008-08-27 21:30 UTC #5 lore raw

In reply to Jeremy Fitzhardinge

Quoted 41 lines · Jeremy Fitzhardinge

Quoted 17 lines · Alan Stern
Quoted 9 lines · Jeremy Fitzhardinge
Using the Fedora Rawhide kernel RPM "2.6.27-0.238.rc2.fc10.i686" (kernel-2.6.27-0.238.rc2.fc10.i686), I'm seeing suspend to ram failures. The suspend seems to go OK, but then I just get endless:
usb 5-1: uhci_result_common: failed with status 440000
messages, with occasional flickers of something else being printed (perhaps). The machine is otherwise unresponsive.
You could comment out that error message in order to see what else is in the log. It is located in drivers/usb/host/uhci-q.c in the uhci_result_common() function.
Quoted 8 lines
You could try running -rc3 instead of -rc2, although that probably won't make much difference.
What USB devices do you have on the failing system? In particular, what is 5-1? If you unplug it before suspending, does the problem go away? (It certainly should!)
No external usb devices. There are three internal USB devices:
Bluetooth, a Sierra CDMA modem thingy and a fingerprint reader.
$ lsusb
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 005 Device 003: ID 0483:2016 SGS Thomson Microelectronics
Fingerprint Reader
Bus 005 Device 002: ID 0a5c:2110 Broadcom Corp. Bluetooth Controller
Bus 005 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 003 Device 003: ID 1199:0218 Sierra Wireless, Inc. MC5720 Wireless Modem
Bus 002 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub

Quoted 2 lines

The rfkill switch "unplugs" the two wireless devices, so I'll try that before suspending next time (if I remember).

Is this still a problem?

Is this related to other Bluetooth suspend/resume regressions in 2.6.27?

Alan Stern

2008-08-27 21:31 UTC #6 lore raw

In reply to Jeremy Fitzhardinge

Quoted 41 lines · Jeremy Fitzhardinge

Quoted 17 lines · Alan Stern
Quoted 9 lines · Jeremy Fitzhardinge
Using the Fedora Rawhide kernel RPM "2.6.27-0.238.rc2.fc10.i686" (kernel-2.6.27-0.238.rc2.fc10.i686), I'm seeing suspend to ram failures. The suspend seems to go OK, but then I just get endless:
usb 5-1: uhci_result_common: failed with status 440000
messages, with occasional flickers of something else being printed (perhaps). The machine is otherwise unresponsive.
You could comment out that error message in order to see what else is in the log. It is located in drivers/usb/host/uhci-q.c in the uhci_result_common() function.
Quoted 8 lines
You could try running -rc3 instead of -rc2, although that probably won't make much difference.
What USB devices do you have on the failing system? In particular, what is 5-1? If you unplug it before suspending, does the problem go away? (It certainly should!)
No external usb devices. There are three internal USB devices:
Bluetooth, a Sierra CDMA modem thingy and a fingerprint reader.
$ lsusb
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 005 Device 003: ID 0483:2016 SGS Thomson Microelectronics
Fingerprint Reader
Bus 005 Device 002: ID 0a5c:2110 Broadcom Corp. Bluetooth Controller
Bus 005 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 003 Device 003: ID 1199:0218 Sierra Wireless, Inc. MC5720 Wireless Modem
Bus 002 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub

Quoted 2 lines

The rfkill switch "unplugs" the two wireless devices, so I'll try that before suspending next time (if I remember).

Is this still a problem?

Is this related to other Bluetooth suspend/resume regressions in 2.6.27?

Alan Stern

Jeremy Fitzhardinge

author

2008-08-27 21:36 UTC #7 lore raw

Quoted 46 lines · Alan Stern

Quoted 43 lines · Jeremy Fitzhardinge
Quoted 18 lines · Alan Stern
Quoted 10 lines · Jeremy Fitzhardinge
Using the Fedora Rawhide kernel RPM "2.6.27-0.238.rc2.fc10.i686" (kernel-2.6.27-0.238.rc2.fc10.i686), I'm seeing suspend to ram failures. The suspend seems to go OK, but then I just get endless:
usb 5-1: uhci_result_common: failed with status 440000
messages, with occasional flickers of something else being printed (perhaps). The machine is otherwise unresponsive.
You could comment out that error message in order to see what else is in the log. It is located in drivers/usb/host/uhci-q.c in the uhci_result_common() function.
Quoted 9 lines
You could try running -rc3 instead of -rc2, although that probably won't make much difference.
What USB devices do you have on the failing system? In particular, what is 5-1? If you unplug it before suspending, does the problem go away? (It certainly should!)
No external usb devices. There are three internal USB devices:
Bluetooth, a Sierra CDMA modem thingy and a fingerprint reader.
$ lsusb
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 005 Device 003: ID 0483:2016 SGS Thomson Microelectronics
Fingerprint Reader
Bus 005 Device 002: ID 0a5c:2110 Broadcom Corp. Bluetooth Controller
Bus 005 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 003 Device 003: ID 1199:0218 Sierra Wireless, Inc. MC5720 Wireless Modem
Bus 002 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub

Quoted 6 lines

Quoted 3 lines
The rfkill switch "unplugs" the two wireless devices, so I'll try that before suspending next time (if I remember).
Is this still a problem?

I haven't seen it lately, but it was fairly erratic anyway. It happened a few times in quick succession, and then it all worked fine. The rawhide kernels are getting fairly old (I guess RH is still trying to pull its infrastructure together), so I'm running a fairly recent self-compiled kernel. Will report anything I see.

Quoted 3 lines

Is this related to other Bluetooth suspend/resume regressions in 2.6.27?

I'm not sure if its BT or Sierra wireless related.

Jeremy Fitzhardinge

author

2008-08-27 21:36 UTC #8 lore raw

In reply to Alan Stern

Quoted 46 lines · Alan Stern

Quoted 43 lines · Jeremy Fitzhardinge
Quoted 18 lines · Alan Stern
Quoted 10 lines · Jeremy Fitzhardinge
Using the Fedora Rawhide kernel RPM "2.6.27-0.238.rc2.fc10.i686" (kernel-2.6.27-0.238.rc2.fc10.i686), I'm seeing suspend to ram failures. The suspend seems to go OK, but then I just get endless:
usb 5-1: uhci_result_common: failed with status 440000
messages, with occasional flickers of something else being printed (perhaps). The machine is otherwise unresponsive.
You could comment out that error message in order to see what else is in the log. It is located in drivers/usb/host/uhci-q.c in the uhci_result_common() function.
Quoted 9 lines
You could try running -rc3 instead of -rc2, although that probably won't make much difference.
What USB devices do you have on the failing system? In particular, what is 5-1? If you unplug it before suspending, does the problem go away? (It certainly should!)
No external usb devices. There are three internal USB devices:
Bluetooth, a Sierra CDMA modem thingy and a fingerprint reader.
$ lsusb
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 005 Device 003: ID 0483:2016 SGS Thomson Microelectronics
Fingerprint Reader
Bus 005 Device 002: ID 0a5c:2110 Broadcom Corp. Bluetooth Controller
Bus 005 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 003 Device 003: ID 1199:0218 Sierra Wireless, Inc. MC5720 Wireless Modem
Bus 002 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub

Quoted 6 lines

Quoted 3 lines
The rfkill switch "unplugs" the two wireless devices, so I'll try that before suspending next time (if I remember).
Is this still a problem?

Quoted 3 lines

Is this related to other Bluetooth suspend/resume regressions in 2.6.27?

I'm not sure if its BT or Sierra wireless related.

Oliver Neukum

2008-08-28 07:13 UTC #9 lore raw

Am Mittwoch 27 August 2008 23:36:30 schrieb Jeremy Fitzhardinge:

Quoted 9 lines

Quoted 2 lines
Is this still a problem?
I haven't seen it lately, but it was fairly erratic anyway. It happened a few times in quick succession, and then it all worked fine. The rawhide kernels are getting fairly old (I guess RH is still trying to pull its infrastructure together), so I'm running a fairly recent self-compiled kernel. Will report anything I see.

Quoted 5 lines

Quoted 3 lines
Is this related to other Bluetooth suspend/resume regressions in 2.6.27?
I'm not sure if its BT or Sierra wireless related.

The attached patch is reported to make the problem go away for a user of btusb. He might be able to use it for narrowing down causes.

Regards
Oliver

drivers/usb/core/urb.c +7 -2

@@ -601,15 +601,20 @@ EXPORT_SYMBOL_GPL(usb_kill_anchored_urbs
 void usb_unlink_anchored_urbs(struct usb_anchor *anchor)
 {
 	struct urb *victim;
+	unsigned long flags;
 
-	spin_lock_irq(&anchor->lock);
+	spin_lock_irqsave(&anchor->lock, flags);
 	while (!list_empty(&anchor->urb_list)) {
 		victim = list_entry(anchor->urb_list.prev, struct urb,
 				    anchor_list);
+		usb_get_urb(victim);
+		spin_unlock_irqrestore(&anchor->lock, flags);
 		/* this will unanchor the URB */
 		usb_unlink_urb(victim);
+		usb_put_urb(victim);
+		spin_lock_irqsave(&anchor->lock, flags);
 	}
-	spin_unlock_irq(&anchor->lock);
+	spin_unlock_irqrestore(&anchor->lock, flags);
 }
 EXPORT_SYMBOL_GPL(usb_unlink_anchored_urbs);

drivers/bluetooth/btusb.c renamed from drivers/bluetooth/btusb.c.alt +113 ~~-32~~

276 lines not on this page · raw message