Fix powerpc-nofpu fmaxmagl, fminmagl spurious "invalid" exception (bug 22691) [committed]

Message ID alpine.DEB.2.20.1801100038380.21425@digraph.polyomino.org.uk
State New
Headers show
Series
  • Fix powerpc-nofpu fmaxmagl, fminmagl spurious "invalid" exception (bug 22691) [committed]
Related show

Commit Message

Joseph Myers Jan. 10, 2018, 12:38 a.m.
For soft-float powerpc, fmaxmagl and fminmagl generate spurious
"invalid" exceptions for quiet NaN arguments.  This is another case of
the problems with fabsl inline expansion via comparisons, and so is
fixed by building those functions with -fno-builtin-fabsl.

Tested for powerpc (soft-float).  Committed.

2018-01-10  Joseph Myers  <joseph@codesourcery.com>

	[BZ #22691]
	* sysdeps/powerpc/nofpu/Makefile [$(subdir) = math]
	(CFLAGS-s_fmaxmagl.c): New variable.
	[$(subdir) = math] (CFLAGS-s_fminmagl.c: Likewise.

Patch

diff --git a/sysdeps/powerpc/nofpu/Makefile b/sysdeps/powerpc/nofpu/Makefile
index 75b116c..463338d 100644
--- a/sysdeps/powerpc/nofpu/Makefile
+++ b/sysdeps/powerpc/nofpu/Makefile
@@ -47,4 +47,6 @@  CFLAGS-e_lgammal_r.c += -fno-builtin-fabsl
 CFLAGS-e_log10l.c += -fno-builtin-fabsl
 CFLAGS-e_log2l.c += -fno-builtin-fabsl
 CFLAGS-e_remainderl.c += -fno-builtin-fabsl
+CFLAGS-s_fmaxmagl.c += -fno-builtin-fabsl
+CFLAGS-s_fminmagl.c += -fno-builtin-fabsl
 endif