Hi Shawn, Am Donnerstag, 22. September 2016, 12:02:19 CEST schrieb Shawn Lin: > Please don't add these for vcc_sd, and mmc-core/driver will control > it. Otherwise, it will waste energy even without sdmmc in slot. > > Moreover, it will causes a bug: > If we insert/remove sd card, we could see > [9.337271] mmc0: new ultra high speed SDR25 SDHC card at address 0007 > [9.345144] mmcblk0: mmc0:0007 SD32G 29.3 GiB > > This is okay for normal sd insert/remove test, but when I debug some > issues for sdmmc, I did unbind/bind test. And there is a interesting > phenomenon when we bind the driver again: > [58.314069] mmc0: new high speed SDHC card at address 0007 > [58.320282] mmcblk0: mmc0:0007 SD32G 29.3 GiB > > So the sd card could just support high speed without power cycle > since the vcc_sd is always on, which makes the sd card fail to > reinit its internal ocr mask. > > Signed-off-by: Shawn Lin <shawn.lin at rock-chips.com> > arch/arm64/boot/dts/rockchip/rk3368-geekbox.dts | 2 -- > arch/arm64/boot/dts/rockchip/rk3368-orion-r68-meta.dts | 2 -- I've applied the geekbox + r68 parts of the patch as fix for 4.9 ... > arch/arm64/boot/dts/rockchip/rk3368-px5-evb.dts | 2 -- and merged the px5-evb-part with the core patch adding the px5-evb, so that we don't introduce known regressions unecessarily. Thanks for fixing this Heiko
