Hi all,
This patch fixes the aarch64 bootstrap failure I've encountered.
Richi suggested on IRC that we should be using .ulow () on the wide_int insetad
of
accessing elt (0) as that doesn't play well with the uninit analysis.
Bootstrapped and tested on aarch64-none-linux-gnu.
Committing to trunk in the interest of fixing the build
2017-03-03 Kyrylo Tkachov <kyrylo.tkac...@arm.com>
* config/aarch64/aarch64.c (aarch64_float_const_representable_p):
Use wide_int::ulow () instead of .elt (0).
diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c
index ab1bdc0..714bb79 100644
--- a/gcc/config/aarch64/aarch64.c
+++ b/gcc/config/aarch64/aarch64.c
@@ -12496,7 +12496,7 @@ aarch64_float_const_representable_p (rtx x)
/* If the low part of the mantissa has bits set we cannot represent
the value. */
- if (w.elt (0) != 0)
+ if (w.ulow () != 0)
return false;
/* We have rejected the lower HOST_WIDE_INT, so update our
understanding of how many bits lie in the mantissa and
