George Koehler ac2b0710c8 Add more rules for single-precision reg_float. ... The result of single-precision fadds, fsubs, and such can go into a register variable, like we already do with double precision. This avoids an extra fmr from a temporary register to the regvar.		2017-10-17 17:53:03 -04:00
..
as	Add RELOLIS for PowerPC lis with ha16 or hi16.	2017-02-08 11:46:31 -05:00
libem	Stop inlining code to convert integers to floats.	2017-10-17 17:00:28 -04:00
libend	Mostly add support for the experimental and largely broken linuxppc platform.	2016-09-15 23:12:03 +02:00
mcg	Use a better NOT; and after remembering that PowerPC bit numbers are all	2017-01-07 01:03:15 +01:00
ncg	Add more rules for single-precision reg_float.	2017-10-17 17:53:03 -04:00
top	In PowerPC top, don't delete addi r0, r0, 0	2017-01-26 12:44:32 -05:00