Anson Huang Best Regards!
> -----Original Message----- > From: Fabio Estevam [mailto:feste...@gmail.com] > Sent: Monday, July 2, 2018 7:33 AM > To: Shawn Guo <shawn...@kernel.org>; Anson Huang > <anson.hu...@nxp.com> > Cc: Robin Gong <yibin.g...@nxp.com>; Mark Rutland > <mark.rutl...@arm.com>; open list:OPEN FIRMWARE AND FLATTENED > DEVICE TREE BINDINGS <devicet...@vger.kernel.org>; linux-kernel > <linux-kernel@vger.kernel.org>; Rob Herring <robh...@kernel.org>; > dl-linux-imx <linux-...@nxp.com>; Sascha Hauer <ker...@pengutronix.de>; > Fabio Estevam <fabio.este...@nxp.com>; moderated list:ARM/FREESCALE > IMX / MXC ARM ARCHITECTURE <linux-arm-ker...@lists.infradead.org> > Subject: Re: [PATCH v1] ARM: dts: imx6sl-evk: keep sw4 always on > > On Sun, Jul 1, 2018 at 6:34 AM, Shawn Guo <shawn...@kernel.org> wrote: > > On Mon, Jun 25, 2018 at 08:34:11PM +0800, Robin Gong wrote: > >> SW4 is one power rail for LPDDR2 on i.mx6sl-evk, so it should be kept > >> always on. But it's disabled after switch disabled interface > >> implemented in pfuze driver 'commit 5fe156f1cab4 > >> ("regulator: pfuze100: add enable/disable for switch")'.Thus, it > >> breaks kernel bootup. Add 'regulator-always-on' for SW4. > >> > >> Signed-off-by: Robin Gong <yibin.g...@nxp.com> > > > > Does that mean boards with existing DTB installed will stop working > > with new kernel? That's bad, and the kernel commit should probably be > > reverted. > > Yes, this is a good point. > > Anson, > > Should 5fe156f1cab4 ("regulator: pfuze100: add enable/disable for > switch") be reverted to avoid such breakage? Yes, I think we can revert it to avoid breakage. Didn't notice that some i.MX platform do NOT have those critical switches always-on. Anson.
