Re: [PATCH] PCI: dwc: fix enumeration end when reaching root subordinate

[Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>]

On Fri, Jan 12, 2018 at 04:56:33PM +0100, Koen Vandeputte wrote:
> Hi Lorenzo,
> 
> Apologies for the late reply, It's been really busy over here.
> 
> 
> On 2018-01-09 16:25, Lorenzo Pieralisi wrote:
> ><snip>
> >>Fix this by initializing the RC to a subordinate value of 0xff, meaning
> >>that all busses [0x00-0xff] are reachable through this RC.
> >This is not a correct description of the problem. AFAICS all busses
> >are reachable through this RC _regardless_ of whatever subordinate
> >bus number value you programme into it.
> Noted.
> 
> This was written meaning: "as seen be the probing functions below"
> I'll try harder in V2 to actually include the message that it's not
> a HW related problem or influencing HW in any way, but really
> "tricking" the probing functions below

Ok, add it to the log please.

> >You should extend the CC list to all dwc host submaintainers so
> >that you can actually get it tested.
> How to figure out who to include? (& please provide who to include)
> Reading online manuals on "how to send patches" only demo's
> "get_maintainer.pl" script

Well, you can do it manually, by checking all pci/dwc submaintainers
in MAINTAINERS.

There are other ways of course but it is simple enough.

> >><snip>
> >>---
> >>
> >>Will send separate patches to stable as this file got moved/renamed
> >Fixes: commit appeared at v4.15-rc1 (and v4.15 has not been released
> >yet) - there is no separate patch to be sent.
> This is something typical which is hard to learn/understand by just
> reading "how to send patches" docs available everywhere. (like I did
> before sending this one over ;-) )
> 
> 2 questions basically:
> - As the commit causing it was included in 4.15-*RC1*,  do I need to
> add a "Fixes; bla bla" at all?

Yes.

> - As the commit causing it was backported to 4.9, (how) should I
> send a separate patch in order to get it fixed there? [0]

So submitting a stable tag does make sense, sorry I missed that.

Or we can send to stable kernel specific backports.

> Probably annoying questions triggering a 'sigh' .. but I'm lacking
> experience here ..

No way, thank you for fixing it !

> Thanks for your time & patience so far,
> Highly appreciated,

Same here, thank you.

Lorenzo