[PATCH] aarch64 sim fp min/max number bug fix

[Jim Wilson]

The fp min/max number routines are using fpclassify to try to
distinguish between a nan and a number, but this is handling zero,
inf, and subnormals incorrectly.  We get the right result if we just
use isnan.

The testcase works with the patch, and fails without the patch.  The
patch also reduces GCC C testsuite failures from 2416 to 2407.

Jim
-------------- next part --------------
A non-text attachment was scrubbed...
Name: aarch64-sim-fminnm.patch
Type: text/x-patch
Size: 3281 bytes
Desc: not available
URL: <http://sourceware.org/pipermail/gdb-patches/attachments/20170101/f603ec97/attachment.bin>