On Mon, 7 May 2012 09:11:55 -0700, Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> wrote:
> On Mon, May 07, 2012 at 10:58:36AM +0300, Alexander Shishkin wrote:
> > On Mon, 7 May 2012 10:43:13 +0300, Felipe Balbi <balbi@xxxxxx> wrote:
> > > Hi,
> > >
> > > On Fri, May 04, 2012 at 04:47:13PM +0300, Alexander Shishkin wrote:
> > > > This is the 6th version of the patchset aiming to rework ci13xxx
> > > > driver. With this patchset, the driver also supports an older version
> > > > of chipidea found in Marvell's Feroceon SoC, that is, it works with
> > > > sheevaplug (provided you can find or make a USB A-to-A cable).
> > > >
> > > > This patchset fixes several bugs in the udc driver and addresses
> > > > some of the style issues. This patchset also includes conversion to
> > > > platform_device, which in case if msm is only compile-tested. While
> > > > at it, I have also converted "choice" of UDC controllers into a menu,
> > > > since we can probably have more than one UDC in the system, at least
> > > > theoretically.
> > > >
> > > > It also converts ci13xxx_udc to the "new style", so that we can finally
> > > > get rid of the global _udc pointer and theoretically can now have as
> > > > many chipidea gadgets in the system as we please.
> > > >
> > > > Then, it proceeds to move the driver to drivers/usb/chipidea and turn
> > > > it into a dual-role controller driver, complete with role switching
> > > > (manual and ID-pin based) and both device and host roles. Tested on
> > > > both sheevaplug and penwell. The former requires modifications to the
> > > > platform code in arch/arm, which are not part of these series. The
> > > > latter just works.
> > > >
> > > > Finally, it removes langwell_udc driver, as it is now covered by the
> > > > reworked chipidea driver too and adds a new entry in MAINTAINERS.
> > > >
> > > > New since the previous version:
> > > > * moved platform data of the pci driver to driver_data in pci_id
> > > > tables as Felipe suggested,
> > > > * split "new style" conversion into two patches, first of which moves
> > > > endpoint initialization from start() to probe(), per Felipe's request,
> > > > * fixed deferred address setting per Felipe's comment,
> > > > * minor fixes here and there per Felipe's comments,
> > > > * removed langwell_udc,
> > > > * added a MAINTAINERS entry
> > >
> > > unfortunately it doesn't apply cleanly, so I have dropped the series
> > > from my v3.5 pull request. It will have to wait until v3.6. I'm sorry.
> >
> > Uhhoh, seems like Greg has decided to patch ci13xxx_udc without telling
> > anybody,
>
> What? I copy everyone who is "responsible" for the file when I patch
> them, so don't say I didn't tell "anybody", that's disingenuous.
I sort of expected that you also copy linux-usb, which doesn't seem to
be the case, because searching my mail for the typo in the subject line
of that commit ("ci13xx_udc" as opposed to "ci13xxx_udc") doesn't find
it. That's how I'd have known that there's a conflict to resolve.
> > and doing something that's taken care of by this patchset. That's
> > nice.
>
> Sarcasm is fun :)
>
> What did I do that is also covered by this patchset? I haven't looked
> at this patchset at all, sorry.
The err() macro in the abovementioned driver.
> Is there a problem having this wait for 3.6?
My plan was that the sooner we get this in, the sooner we can point it
to the maintainers of other chipidea drivers in the kernel (mv_udc,
fsl-usb2-udc, and a bunch of echi- and otg drivers), so that we end up
having just one. But of course, they've been there for so long that
another 4 months won't change anything.
Regards,
--
Alex
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
