[Bug target/67260] [sh] Register spill bug for sibcall+complex+softfloat

[bugdal at aerifal dot cx]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67260

--- Comment #15 from Rich Felker <bugdal at aerifal dot cx> ---
Is it related to the fact that the relative address load is tied to a specific
call site and that the call can't be cloned without producing a new relative
address load to go with the clone?