Re: [PATCH] rtl8192cu: Fix WARNING on suspend/resume

From: Daniel Halperin
Date: Wed Jan 04 2012 - 12:44:03 EST


On Wed, Jan 4, 2012 at 9:14 AM, Larry Finger <Larry.Finger@xxxxxxxxxxxx> wrote:
> On 01/04/2012 10:16 AM, Daniel Halperin wrote:
>>
>> Sorry, Larry, I didn't see this til it was included in John's mail this
>> morning.
>>
>> On Tue, Jan 3, 2012 at 10:39 AM, Larry Finger<Larry.Finger@xxxxxxxxxxxx>
>>  wrote:
>>>
>>>
>>> +#ifdef CONFIG_PM
>>> +static int rtl8192cu_usb_suspend(struct usb_interface *pusb_intf,
>>> +                                pm_message_t message)
>>> +{
>>> +       /* Increase usage_count to Save loaded fw across suspend/resume
>>> */
>>> +       atomic_inc(&usage_count);
>>> +       return 0;
>>> +}
>>> +
>>> +static int rtl8192cu_usb_resume(struct usb_interface *pusb_intf)
>>> +{
>>> +       atomic_dec(&usage_count);       /* after resume, decrease usage
>>> count */
>>> +       return 0;
>>> +}
>>> +#endif
>>
>>
>> Based on the warning-fix I submitted yesterday, this probably produces
>> a warning if CONFIG_PM is set but not CONFIG_PM_SLEEP. My patch was
>> for an ethernet driver, but iwlwifi and ath5k appear to follow the
>> same pattern using CONFIG_PM_SLEEP instead of the above.
>>
>> See: http://www.spinics.net/lists/netdev/msg184717.html for the patch
>> and http://www.spinics.net/lists/netdev/msg184715.html for the
>> discussion.
>
>
> Thanks for the note. I will prepare a revised patch.
>
> What architecture gave the warning? Of x86_64, I was unable to get
> CONFIG_PM_SLEEP without CONFIG_PM.
>

x86-64. I realized that I actually explained this problem rather
clearly when I fixed it for iwlwifi 5 months ago. :)
https://github.com/mirrors/linux/commit/f090fba305658fe6e464e2fbd25fad81957ece26

"""
CC [M] drivers/net/wireless/iwlwifi/iwl-pci.o
drivers/net/wireless/iwlwifi/iwl-pci.c:506: warning:
‘iwl_pci_suspend’ defined but not used
drivers/net/wireless/iwlwifi/iwl-pci.c:519: warning:
‘iwl_pci_resume’ defined but not used

These are only used if CONFIG_PM_SLEEP is enabled. CONFIG_PM depends
(CONFIG_PM_SLEEP || CONFIG_PM_RUNTIME), so it can be set without
CONFIG_PM_SLEEP selected.
"""

Dan
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/