Question about MODES_TIEABLE_P

John Lu jlu@lsil.com
Fri Oct 24 00:03:00 GMT 2003


Hi All,

I'm trying to understand how the macro MODES_TIEABLE_P(MODE1,MODE2) should
be defined.  The gccint.info documentation states:

     If `HARD_REGNO_MODE_OK (R, MODE1)' and `HARD_REGNO_MODE_OK (R,
     MODE2)' are always the same for any R, then `MODES_TIEABLE_P
     (MODE1, MODE2)' should be nonzero.
     
It also states that it is highly desirable for modes to be tieable.
Based on this, should every port have MODES_TIEABLE_P() return true
if MODE1==MODE2?  There are some ports, in the gcc-3.3 release, where
MODES_TIEABLE_P is set to 0 for all inputs (avr.h, c4x.h, pdp11.h), which
made me wonder if I'm overlooking something.

Thanks,
John Lu



More information about the Gcc mailing list