On Tue, Jul 26, 2016 at 08:04:32PM -0400, Michael Meissner wrote: > > dg-do compile? That's not testing much then, as an executable test! > > Good catch. Hopefully, third time is a charm. I verified that changing the > dg-do compile to dg-do run did run properly on a big endian power7 (both > 32-bit > and 64-bit) and a little endian power8. > > [gcc] > 2016-07-26 Michael Meissner <meiss...@linux.vnet.ibm.com> > > PR target/71869 > * config/rs6000/rs6000.c (rs6000_generate_compare): Rework > __float128 support when we don't have hardware support, so that > the IEEE built-in functions like isgreater, first call __unordkf3 > to make sure neither operand is a NaN, and if both operands are > ordered, do the normal comparison. > > [gcc/testsuite] > 2016-07-26 Michael Meissner <meiss...@linux.vnet.ibm.com> > > PR target/71869 > * gcc.target/powerpc/float128-cmp.c: New test to make sure that > IEEE built-in functions handle quiet and signalling NaNs > correctly.
Okay for trunk. Okay for 6 once there is independent testing (on gcc-testresults, say) for all affected targets and you aren't comfortable waiting any longer ;-) Segher
