HI, On Sat, Jan 21, 2017 at 01:10:38AM +0100, Heiko Stuebner wrote: > Hi Arnd, Kevin, Olof, > > please find below some rockchip-specific soc-driver updates for 4.11. > It is based on 4.10-rc2 due to the bug in 4.10-rc1 that caused the whole > arm64 to not build. > > The grf-thingy has been on the lists in some iterations and as the tag > message says is really only ever meant for stuff that really fits nowhere > else but can also not be set in firmware in a sane way. > > So if there are no complaints, please pull. I replied to one of the patches -- I think it can be done better but this can go in meanwhile and it can be switched over to later. > > The following changes since commit 0c744ea4f77d72b3dcebb7a8f2684633ec79be88: > > Linux 4.10-rc2 (2017-01-01 14:31:53 -0800) > > are available in the git repository at: > > git://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip.git tags/v4.11-armsoc-drivers1 > > for you to fetch changes up to 19f0f3021e7532a1056886e3ddf524f9847e4f15: > > ARM: rockchip: drop rk3288 jtag/mmc switch handling (2017-01-06 10:29:16 +0100) > > ---------------------------------------------------------------- > Some extensions to the power-domain driver to support domains in > hiword registers (write-mask in upper 16bit) and domain-definitions > for the rk3328 soc. > > Secondly a "driver" that attaches to the already existing grf nodes > and is able to set static defaults for settings that cannot really > be attached to any specific subsystem. > Most GRF settings can already be set from drivers using them, but there > are some behavioural settings like the mmc/jtag switch that cannot. > > As the commit message states this is really meant as a last line > of defence for things that neither belong to a subsystem nor to the > > Having this here allows arm64 socs to have this as well and also > moves another bit of code out of the arm32 mach-rockchip. Third paragraph looks cut off. Still, merged. -Olof
