Messages in this thread | | | From | Doug Anderson <> | Date | Tue, 23 Aug 2022 13:18:43 -0700 | Subject | Re: [PATCH] regulator: qcom-rpmh: Implement get_optimum_mode(), not set_load() |
| |
Hi,
On Tue, Aug 23, 2022 at 6:06 AM Mark Brown <broonie@kernel.org> wrote: > > On Mon, Aug 22, 2022 at 01:13:55PM -0700, Doug Anderson wrote: > > > I guess at this point I'll wait for Mark to give his suggestion for > > what to do. Options I'm aware of: > > > a) ${SUBJECT} patch was written as a cleanup as per Mark's request and > > we could just revert it. > > > b) We could make it so that failures to get the input/output voltages > > doesn't count as an error when going through the get_optimum_mode() > > path. > > We could push the checks for a valid voltage down into the drivers, a > lot of things aren't particularly sensitive to the output voltage here > and are much more sensitive to the current draw. Depending on people's > attitudes to DT stability for Qualcomm platforms we could also fix the > affected DTs as well, though that doesn't stop us handling this in the > core too.
OK, patch posted with this approach. ("regulator: core: Require regulator drivers to check uV for get_optimum_mode()") [1]
[1] https://lore.kernel.org/r/20220823131629.RFT.1.I137e6bef4f6d517be7b081be926059321102fd3d@changeid
| |