Re: [RFC 00/10] make iwlwifi RT friendly

[Henrik Austad <henrik@austad.us>]

[-- Attachment #1: Type: text/plain, Size: 1410 bytes --]

On Tue, Dec 17, 2013 at 09:32:14AM +0000, Grumbach, Emmanuel wrote:
> > 
> > On Sun, Dec 15, 2013 at 07:50:23PM +0200, Emmanuel Grumbach wrote:
> > >
> > >
> > > On 12/15/2013 07:46 PM, Sebastian Andrzej Siewior wrote:
> > > > On 12/15/2013 06:45 PM, Emmanuel Grumbach wrote:
> > > >> No we don't. The primary interrupt just tells the hardware to stop
> > > >> firing interrupts and wakes up the thread as written above.
> > > >
> > > > Sounds great then.
> > > >
> > >
> > > Ok - I guess I'll submit then internally then before we send them upstream.
> > 
> > I'm trying to mash this on top of 3.12.5 for the latest rt-release, and especially
> > drivers/net/wireless/iwlwifi/pcie/rx.c is not particularly compatible with your
> > patches, especially iwl_pcie_int_cause_non_ict(), which as been moved,
> > renamed and given new return values.
> 
> iwl_pcie_int_cause_non_ict didn't exist before my patch. I am not sure I follow here...

bah, git is messing with my head, iwl_pcie_is_non_ict() used a lot of the 
same code as iwl_pcie_isr(), so I mistakenly saw it as a rename (which got 
me a bit confused). Looking closer and applying enough braincells to form a 
synapse, I see that it has just moved further down.

> I am not sure I'll have the time to fulfill your request, but I'll try.

Fair enough, I'll give the rt-port of the patches another go :)

-- 
Henrik Austad

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 198 bytes --]