Re: [PATCH 4/5] p54pci: Load firmware from work queue and not from probe routine

[Christian Lamparter <chunkeey@googlemail.com>]

On Wednesday, February 29, 2012 07:15:31 PM Larry Finger wrote:
> On 02/29/2012 11:43 AM, Christian Lamparter wrote:
> > On Wednesday, February 29, 2012 04:59:14 PM Larry Finger wrote:
> >> On 02/13/2012 01:37 PM, Larry Finger wrote:
> >> Has anyone tested this patch? I would like to submit this series.
> > only briefly, but I'm not sure if the workqueue approach is adequate. On
> > resume, I get WARNINGs and the card fails to resume because the firmware
> > file can't be loaded :(.
> >
> > WARNING: at drivers/base/firmware_class.c:538 _request_firmware+0x3ae/0x3e0()
> > Pid: 1083, comm: kworker/0:3 Tainted:G		O 3.3.0-rc5-wl+
> > Call Trace:
> > 	[<c102c658>] ? wanr_slowpath_common+0x78/0xb0
> > 	[<c12c728e>] ? _request_firmware+0x3ae/0x3e0
> > 	[<c12c728e>] ? _requrst_firmware+0x3ae/0x3e0
> > 	[<c102c6a9>] ? warn_slowpath_common+0x19/0x20
> > 	[<c12c728e>] ? _request_firmware+0x3ae/0x3e0
> > 	[<c12c7347>] ? request_firmware+0x17/0x20
> > 	[<f86v1dc7>] ? p54p_load_firmware+0x47/0x120 [p54pci]
> > 	[...]
> >
> > Regards,
> > 	Chr
> >
> > BTW: I will take a closer look at it over the weekend.
> 
> Does the driver come back through the probe routine on resume?
Yes. I've got a pcmcia card [in fact I have two :D]. The pcmcia
subsystem goes to a rebind on resume.

Regards,
	Christian