Bug#749484: linux: Please backport Cubietruck (sunxi) MMC driver/dt
On Tue, 2014-05-27 at 09:02 +0000, Andreas Henriksson wrote:
Apologies for the confusion with the pending tag earlier -- I
accidentally ran tagpending from a tree which had my WIP stuff for this
applied and then untagged. I've now pushed the changes and tagged again.
> The actual driver is in Ulf Hanssons "mmc-next_for_chris_3.16_take3" branch:
I couldn't see this in the linux-next tree or the mmc maintainers
tree. I could see 19b7f796c375 which is the same (modulo some changes
to the comment message) in both, but linux-next contains a commit
(47d673b7fd55 in next-20140530) which reverts it due to a build failure.
However it builds for me so I suspect this was just a synchronisation
issue between some of the different trees involed. Given that I went
ahead and used 19b7f796c375 for the backport.
> (Not mmc related, but was part of same review thread and also
> something we want if we don't already have it...)
I included this, with it (and some fiddling + non-free firmware) wifi
works for me on CT.
> Maybe also needed:
This isn't needed because we don't have 559482d1f950 "ARM: sunxi: Split
the various SoCs support in Kconfig".
In addition to the commits you listed I also needed:
a97181a clk: sunxi: Fixup clk_sunxi_mmc_phase_control to take a clk rather then a hw_clk
9797eb8 ARM: dts: sun6i: Add pin-muxing info for the mmc controllers
6da50f1 ARM: dts: sun5i: Add pin-muxing info for the mmc controllers
otherwise I got build errors/warnings.
With all that I got things building and MMCs seem to work on my CT so
I've pushed the backport to SVN trunk.
Thanks for doing the initial legwork, it was very useful!