[U-Boot] [PATCH 1/2] dm: Protect device_unbind() with CONFIG_DM_DEVICE_REMOVE

[Simon Glass]

Hi Marek,

On 19 February 2015 at 11:19, Marek Vasut <marex at denx.de> wrote:
> On Thursday, February 19, 2015 at 04:24:02 PM, Simon Glass wrote:
>> Hi Marek,
>
> Hi Simon,
>
>> On 19 February 2015 at 07:34, Marek Vasut <marex at denx.de> wrote:
>> > On Thursday, February 19, 2015 at 03:29:52 PM, Simon Glass wrote:
>> >> Hi Marek,
>> >
>> > Hi Simon,
>> >
>> >> On 19 February 2015 at 01:50, Marek Vasut <marex at denx.de> wrote:
>> >> > On Thursday, February 19, 2015 at 02:06:38 AM, Simon Glass wrote:
>> >> >> On 18 February 2015 at 14:36, Marek Vasut <marex at denx.de> wrote:
>> >> >> > Since device_unbind() is also defined in device-remove.c,
>> >> >> > which is compiled in only in case CONFIG_DM_DEVICE_REMOVE
>> >> >> > is defined, protect the device_unbind() prototype with the
>> >> >> > same CONFIG_DM_DEVICE_REMOVE check.
>> >> >> >
>> >> >> > Signed-off-by: Marek Vasut <marex at denx.de>
>> >> >> > Cc: Simon Glass <sjg at chromium.org>
>> >> >> > Cc: Stefan Roese <sr at denx.de>
>> >> >> > Cc: Tom Rini <trini at ti.com>
>> >> >> > ---
>> >> >> >
>> >> >> >  include/dm/device-internal.h | 4 ++++
>> >> >> >  1 file changed, 4 insertions(+)
>> >> >>
>> >> >> Acked-by: Simon Glass <sjg at chromium.org>
>> >> >
>> >> > Tom, can you please pick this one, I need it to repair the socfpga,
>> >> > so I'd like to have it in before I submit new PR for that.
>> >>
>> >> I'll bring it into u-boot-dm in any case.
>> >>
>> >> I didn't see a build error though? What is broken at present?
>> >
>> > Try building u-boot-socfpga/master . The patches there enable DT
>> > (CONFIG_OF_CONTROL) on SoCFPGA always. And in case this is enabled,
>> > the Cadence SPI and DW SPI drivers are compiled in. But those two
>> > drivers require CONFIG_DM and CONFIG_DM_SPI ... and spi-uclass.c
>> > contains calls to device_unbind() .
>>
>> Ah OK, so this a problem you have seen in new patches, not in mainline.
>
> It's only triggered by the new code, it is not introduced by this new code.
> In case you enable either of those drivers as well as CONFIG_OF_CONTROL,
> you will trigger this bug.
>
> In case I base those patches on v2015.04-rc1 , I don't see this error, but
> in case I base those patches on top of v2015.04-rc2, I do see it. Do you
> have any hint which changes between these two points can cause this breakage
> please ?

It is probably the Kconfig conversion - now there will be extra options enabled.

>
> Also, about the Kconfig, shall I introduce the entries for those two drivers
> or what is the plan here ? I would like to get these patches into the current
> release.

Yes please.

Regards,
Simon