On 09/19/2017 08:36 AM, Rob Herring wrote: > On Mon, Sep 18, 2017 at 3:50 PM, Andrew F. Davis <[email protected]> wrote: >> On TI platforms OP-TEE must be called using a modified SMC call, >> allow the selection of this though DT. >> >> Signed-off-by: Andrew F. Davis <[email protected]> >> --- >> Documentation/devicetree/bindings/arm/firmware/linaro,optee-tz.txt | 2 ++ >> drivers/tee/optee/core.c | 2 ++ >> 2 files changed, 4 insertions(+) >> >> diff --git >> a/Documentation/devicetree/bindings/arm/firmware/linaro,optee-tz.txt >> b/Documentation/devicetree/bindings/arm/firmware/linaro,optee-tz.txt >> index d38834c67dff..a3275ecdf186 100644 >> --- a/Documentation/devicetree/bindings/arm/firmware/linaro,optee-tz.txt >> +++ b/Documentation/devicetree/bindings/arm/firmware/linaro,optee-tz.txt >> @@ -20,6 +20,8 @@ the reference implementation maintained by Linaro. >> "hvc" : HVC #0, with the register assignments specified >> in drivers/tee/optee/optee_smc.h >> >> + "ti-smc" : Similar to "smc" with TI specific register >> + adjustments > > Sigh, really? IMO, this should be determined from the compatible > string. Then the next TI (or any vendor) specific thing can be handled > without a DT change. >
Which compatible string, do you mean the OP-TEE driver check the top level platform compatible string? > Maybe some day we'll figure out that not just h/w needs to be > probe-able, but all the firmware pieces do too. > > Rob >
