On 08/04/22 10:47AM, Tom Rini wrote: > On Thu, Mar 31, 2022 at 03:35:26PM -0400, Tom Rini wrote: > > > This is a little tricky since SoCFPGA has code to determine this as > > runtime. Introduce a guard variable for platforms to select if they > > have a static value to use. Then for ARCH_SOCFPGA, call > > cm_get_qspi_controller_clk_hz() and otherwise continue the previous > > behavior. > > > > Cc: Jagan Teki <[email protected]> > > Signed-off-by: Tom Rini <[email protected]> > > OK, so there's a problem with using IS_ENABLED(..) which is that we get > a failure to build in the not-enabled case over CONFIG_CQSPI_REF_CLK > being unset. And then we can't use if (CONFIG_VAL(..)) because we then > run in to needing it to exist for SPL as well. I'm going to go back to > v1 of this particular patch.
Interesting! I remember trying the IS_ENABLED() thing a while back, and I remember it working fine when the functions don't exist if the config is not enabled. I figured that was because of the compiler eliminating dead code. But I think here it is the preprocessor raising the error and maybe it is not so smart. Dunno. Anyway, I am fine with reverting this to v1 for now. I'll try to look at it later if I ever get some time to spare. -- Regards, Pratyush Yadav Texas Instruments Inc.
