BBB MMC / SD detection instability with U-Boot 2014.04 (CPU 1GHz)

[SAITOU Toshihide]

In message: <537B8DFA.3000603 at hot.ee>
            "Sulev-Madis Silber (ketas)" <madis555 at hot.ee> writes:
> On 2014-05-20 18:58, SAITOU Toshihide wrote:
>> In message: <20140520.234245.38709064.toshi at ruby.ocn.ne.jp>
>>             SAITOU Toshihide <toshi at ruby.ocn.ne.jp> writes:
>>>>> u-boot-2014.04.tar.bz2
>>>>>
>>>>>   patch
>>>>>
>>>>>     1. apply these patches:
>>>>>
>>>>>          http://lists.freebsd.org/pipermail/freebsd-arm/2014-April/007922.html
>>>>>          ~/crochet-freebsd/board/BeagleBone/files/uboot-2013.04_api_api_storage.c.patch
>>>>>          ~/crochet-freebsd/board/BeagleBone/files/uboot-2013.04_drivers_mmc_mmc.c.patch
>> 
>> I forgot to mention that I had changed uboot-2013.04_drivers_mmc_mmc.c.patch a bit.
>> 
>> Sorry for bother you
> 
> Hmm, that doesn't seem to fix it. I have different uboot env but that
> should not matter here. Is this CURRENT? I'm on CURRENT and that patch
> doesn't get me two SD devices in loader. It's quite mysterious code
> there too...
> 
> Could you explain where and how exactly did you manage to get both
> devices found in loader. I'm not exactly sure uboot needs patch. I might
> start hacking loader next.

I'm using current 265876 and did not change the source code except for
am335x.dtsi and beaglebone-black.dts.


ubldr is built to change the load address as follows:

  # cat ubldr_make.sh
  buildenv=`make -C /usr/src TARGET=arm TARGET_ARCH=armv6 buildenvvars`
  eval $buildenv make -C /usr/src/sys/boot obj
  eval $buildenv make -C /usr/src/sys/boot UBLDR_LOADADDR=0x88000000 all

  # sh ubldr_make.sh

  # cp /usr/obj/arm.armv6/usr/src/sys/boot/arm/uboot/ubldr \
      /mnt/ubldr

  (maybe need to cp /usr/src/share/mk/src.opts.mk /usr/share/mk/)


u-boot (MLO, u-boot.img) is u-boot-2014.04 and apply these patch:

  http://lists.freebsd.org/pipermail/freebsd-arm/2014-April/007922.html
  uboot-2013.04_api_api_storage.c.patch
  uboot-2014.04_drivers_mmc_mmc.c.patch

  According to the third patch, it seems to retry when
  mmc_send_op_cond is failed.  I don't know why it failed.

-- 
SAITOU Toshihide