Skip to contentSkip to conversation

WARNING: at drivers/pci/pci.c:1397 pci_disable_device

DISCUSSION Jiri Slaby 2 participants 4 messages Feb 19, 2013 – Feb 25, 2013 PCI

lore
Share & tools
lore

Jiri Slaby

author
#0 lore raw

Hi,

so I hit that one:
+ dev_WARN_ONCE(&dev->dev, atomic_read(&dev->enable_cnt) <= 0,
+ "disabling already-disabled device");

during suspend (to ram):

WARNING: at drivers/pci/pci.c:1397 pci_disable_device+0x90/0xa0()
Hardware name: To Be Filled By O.E.M.
Device e1000e
disabling already-disabled device
Modules linked in: dvb_usb_dib0700 dib0090 dib7000p dib7000m dib0070
dib8000 dib3000mc dibx000_common microcode
Pid: 31027, comm: kworker/u:35 Not tainted 3.8.0-rc7-next-20130218_64+ #1768
Call Trace:
 [<ffffffff81313b00>] ? do_pci_disable_device+0x30/0x60
 [<ffffffff8107f2ff>] warn_slowpath_common+0x7f/0xc0
 [<ffffffff8107f3f6>] warn_slowpath_fmt+0x46/0x50
 [<ffffffff81313bc0>] pci_disable_device+0x90/0xa0
 [<ffffffff814c8152>] __e1000_shutdown+0x262/0x8b0
 [<ffffffff814c8813>] e1000_suspend+0x23/0x50
 [<ffffffff816be581>] ? wait_for_completion+0x31/0x100
 [<ffffffff81316337>] pci_pm_suspend+0x77/0x140
 [<ffffffff8144aa6a>] ? __pm_runtime_barrier+0x1a/0x130
 [<ffffffff813162c0>] ? pci_pm_poweroff+0xf0/0xf0
 [<ffffffff81447838>] dpm_run_callback+0x58/0x90
 [<ffffffff8144795b>] __device_suspend+0xeb/0x280
 [<ffffffff814482af>] async_suspend+0x1f/0xa0
 [<ffffffff810a9c9b>] async_run_entry_fn+0x3b/0x140
 [<ffffffff8109b9a4>] process_one_work+0x174/0x410
 [<ffffffff8109d426>] worker_thread+0x116/0x400
 [<ffffffff8109d310>] ? busy_worker_rebind_fn+0xc0/0xc0
 [<ffffffff810a3340>] kthread+0xc0/0xd0
 [<ffffffff810a3280>] ? kthread_create_on_node+0x130/0x130
 [<ffffffff816c089c>] ret_from_fork+0x7c/0xb0
 [<ffffffff810a3280>] ? kthread_create_on_node+0x130/0x130
---[ end trace 6c5060f8b8fb9175 ]---

e1000e 0000:00:19.0: System wakeup enabled by ACPI

Signature · 2 lines
js
suse labs
maintainer
#1 lore raw
Quoted 37 lines · Jiri Slaby

Hi,

so I hit that one:

(quoted diff) +2 -0
+       dev_WARN_ONCE(&dev->dev, atomic_read(&dev->enable_cnt) <= 0,
+                     "disabling already-disabled device");

during suspend (to ram):

WARNING: at drivers/pci/pci.c:1397 pci_disable_device+0x90/0xa0()
Hardware name: To Be Filled By O.E.M.
Device e1000e
disabling already-disabled device
Modules linked in: dvb_usb_dib0700 dib0090 dib7000p dib7000m dib0070
dib8000 dib3000mc dibx000_common microcode
Pid: 31027, comm: kworker/u:35 Not tainted 3.8.0-rc7-next-20130218_64+ #1768
Call Trace:
 [<ffffffff81313b00>] ? do_pci_disable_device+0x30/0x60
 [<ffffffff8107f2ff>] warn_slowpath_common+0x7f/0xc0
 [<ffffffff8107f3f6>] warn_slowpath_fmt+0x46/0x50
 [<ffffffff81313bc0>] pci_disable_device+0x90/0xa0
 [<ffffffff814c8152>] __e1000_shutdown+0x262/0x8b0
 [<ffffffff814c8813>] e1000_suspend+0x23/0x50
 [<ffffffff816be581>] ? wait_for_completion+0x31/0x100
 [<ffffffff81316337>] pci_pm_suspend+0x77/0x140
 [<ffffffff8144aa6a>] ? __pm_runtime_barrier+0x1a/0x130
 [<ffffffff813162c0>] ? pci_pm_poweroff+0xf0/0xf0
 [<ffffffff81447838>] dpm_run_callback+0x58/0x90
 [<ffffffff8144795b>] __device_suspend+0xeb/0x280
 [<ffffffff814482af>] async_suspend+0x1f/0xa0
 [<ffffffff810a9c9b>] async_run_entry_fn+0x3b/0x140
 [<ffffffff8109b9a4>] process_one_work+0x174/0x410
 [<ffffffff8109d426>] worker_thread+0x116/0x400
 [<ffffffff8109d310>] ? busy_worker_rebind_fn+0xc0/0xc0
 [<ffffffff810a3340>] kthread+0xc0/0xd0
 [<ffffffff810a3280>] ? kthread_create_on_node+0x130/0x130
 [<ffffffff816c089c>] ret_from_fork+0x7c/0xb0
 [<ffffffff810a3280>] ? kthread_create_on_node+0x130/0x130
---[ end trace 6c5060f8b8fb9175 ]---
Quoted 1 line

e1000e 0000:00:19.0: System wakeup enabled by ACPI

I think Konstantin posted an e1000e patch that would fix this, but it's going via the e1000e folks. Hopefully both my tree and the e1000e patch will be merged soon.

Bjorn

author
#2 lore raw
Quoted 40 lines · Bjorn Helgaas
Quoted 37 lines · Jiri Slaby

Hi,

so I hit that one:

(quoted diff) +2 -0
+       dev_WARN_ONCE(&dev->dev, atomic_read(&dev->enable_cnt) <= 0,
+                     "disabling already-disabled device");

during suspend (to ram):

WARNING: at drivers/pci/pci.c:1397 pci_disable_device+0x90/0xa0()
Hardware name: To Be Filled By O.E.M.
Device e1000e
disabling already-disabled device
Modules linked in: dvb_usb_dib0700 dib0090 dib7000p dib7000m dib0070
dib8000 dib3000mc dibx000_common microcode
Pid: 31027, comm: kworker/u:35 Not tainted 3.8.0-rc7-next-20130218_64+ #1768
Call Trace:
 [<ffffffff81313b00>] ? do_pci_disable_device+0x30/0x60
 [<ffffffff8107f2ff>] warn_slowpath_common+0x7f/0xc0
 [<ffffffff8107f3f6>] warn_slowpath_fmt+0x46/0x50
 [<ffffffff81313bc0>] pci_disable_device+0x90/0xa0
 [<ffffffff814c8152>] __e1000_shutdown+0x262/0x8b0
 [<ffffffff814c8813>] e1000_suspend+0x23/0x50
 [<ffffffff816be581>] ? wait_for_completion+0x31/0x100
 [<ffffffff81316337>] pci_pm_suspend+0x77/0x140
 [<ffffffff8144aa6a>] ? __pm_runtime_barrier+0x1a/0x130
 [<ffffffff813162c0>] ? pci_pm_poweroff+0xf0/0xf0
 [<ffffffff81447838>] dpm_run_callback+0x58/0x90
 [<ffffffff8144795b>] __device_suspend+0xeb/0x280
 [<ffffffff814482af>] async_suspend+0x1f/0xa0
 [<ffffffff810a9c9b>] async_run_entry_fn+0x3b/0x140
 [<ffffffff8109b9a4>] process_one_work+0x174/0x410
 [<ffffffff8109d426>] worker_thread+0x116/0x400
 [<ffffffff8109d310>] ? busy_worker_rebind_fn+0xc0/0xc0
 [<ffffffff810a3340>] kthread+0xc0/0xd0
 [<ffffffff810a3280>] ? kthread_create_on_node+0x130/0x130
 [<ffffffff816c089c>] ret_from_fork+0x7c/0xb0
 [<ffffffff810a3280>] ? kthread_create_on_node+0x130/0x130
---[ end trace 6c5060f8b8fb9175 ]---
Quoted 1 line

e1000e 0000:00:19.0: System wakeup enabled by ACPI

Quoted 3 lines

I think Konstantin posted an e1000e patch that would fix this, but it's going via the e1000e folks. Hopefully both my tree and the e1000e patch will be merged soon.

I'm using -next anyway, so is there any chance it is fixed there some time soon?

thanks,

Signature · 2 lines
js
suse labs
author
#3 lore raw
Quoted 41 lines · Jiri Slaby
Quoted 40 lines · Bjorn Helgaas
Quoted 38 lines · Jiri Slaby

Hi,

so I hit that one:

(quoted diff) +2 -0
+       dev_WARN_ONCE(&dev->dev, atomic_read(&dev->enable_cnt) <= 0,
+                     "disabling already-disabled device");

during suspend (to ram):

WARNING: at drivers/pci/pci.c:1397 pci_disable_device+0x90/0xa0()
Hardware name: To Be Filled By O.E.M.
Device e1000e
disabling already-disabled device
Modules linked in: dvb_usb_dib0700 dib0090 dib7000p dib7000m dib0070
dib8000 dib3000mc dibx000_common microcode
Pid: 31027, comm: kworker/u:35 Not tainted 3.8.0-rc7-next-20130218_64+ #1768
Call Trace:
 [<ffffffff81313b00>] ? do_pci_disable_device+0x30/0x60
 [<ffffffff8107f2ff>] warn_slowpath_common+0x7f/0xc0
 [<ffffffff8107f3f6>] warn_slowpath_fmt+0x46/0x50
 [<ffffffff81313bc0>] pci_disable_device+0x90/0xa0
 [<ffffffff814c8152>] __e1000_shutdown+0x262/0x8b0
 [<ffffffff814c8813>] e1000_suspend+0x23/0x50
 [<ffffffff816be581>] ? wait_for_completion+0x31/0x100
 [<ffffffff81316337>] pci_pm_suspend+0x77/0x140
 [<ffffffff8144aa6a>] ? __pm_runtime_barrier+0x1a/0x130
 [<ffffffff813162c0>] ? pci_pm_poweroff+0xf0/0xf0
 [<ffffffff81447838>] dpm_run_callback+0x58/0x90
 [<ffffffff8144795b>] __device_suspend+0xeb/0x280
 [<ffffffff814482af>] async_suspend+0x1f/0xa0
 [<ffffffff810a9c9b>] async_run_entry_fn+0x3b/0x140
 [<ffffffff8109b9a4>] process_one_work+0x174/0x410
 [<ffffffff8109d426>] worker_thread+0x116/0x400
 [<ffffffff8109d310>] ? busy_worker_rebind_fn+0xc0/0xc0
 [<ffffffff810a3340>] kthread+0xc0/0xd0
 [<ffffffff810a3280>] ? kthread_create_on_node+0x130/0x130
 [<ffffffff816c089c>] ret_from_fork+0x7c/0xb0
 [<ffffffff810a3280>] ? kthread_create_on_node+0x130/0x130
---[ end trace 6c5060f8b8fb9175 ]---

e1000e 0000:00:19.0: System wakeup enabled by ACPI

Quoted 6 lines
Quoted 3 lines

I think Konstantin posted an e1000e patch that would fix this, but it's going via the e1000e folks. Hopefully both my tree and the e1000e patch will be merged soon.

I'm using -next anyway, so is there any chance it is fixed there some time soon?

Nevermind, I've just applied "e1000e: fix pci-device enable-counter balance" which should fix that, I suppose.

Quoted 1 line

thanks,

Signature · 2 lines
js
suse labs