aboutsummaryrefslogtreecommitdiff
path: root/sysdeps/powerpc/fpu
diff options
context:
space:
mode:
authorAdhemerval Zanella <adhemerval.zanella@linaro.org>2022-04-01 10:19:57 -0300
committerAdhemerval Zanella <adhemerval.zanella@linaro.org>2022-05-23 17:49:18 -0300
commitefeb2bd1ab49215325040aa53396a2c19dfd1402 (patch)
tree0dda3bc772c3ac1c2e2b2a8a55a6f75c28e1efe8 /sysdeps/powerpc/fpu
parent04b30fe4f85d6e4c0390581a5e90db066a0b2f1f (diff)
downloadglibc-efeb2bd1ab49215325040aa53396a2c19dfd1402.zip
glibc-efeb2bd1ab49215325040aa53396a2c19dfd1402.tar.gz
glibc-efeb2bd1ab49215325040aa53396a2c19dfd1402.tar.bz2
math: Add math-use-builtins-fabs (BZ#29027)
Both float, double, and _Float128 are assumed to be supported (float and double already only uses builtins). Only long double is parametrized due GCC bug 29253 which prevents its usage on powerpc. It allows to remove i686, ia64, x86_64, powerpc, and sparc arch specific implementation. On ia64 it also fixes the sNAN handling: math/test-float64x-fabs math/test-ldouble-fabs Checked on x86_64-linux-gnu, i686-linux-gnu, powerpc-linux-gnu, powerpc64-linux-gnu, sparc64-linux-gnu, and ia64-linux-gnu.
Diffstat (limited to 'sysdeps/powerpc/fpu')
-rw-r--r--sysdeps/powerpc/fpu/math-use-builtins-fabs.h8
1 files changed, 8 insertions, 0 deletions
diff --git a/sysdeps/powerpc/fpu/math-use-builtins-fabs.h b/sysdeps/powerpc/fpu/math-use-builtins-fabs.h
new file mode 100644
index 0000000..8976d0b
--- /dev/null
+++ b/sysdeps/powerpc/fpu/math-use-builtins-fabs.h
@@ -0,0 +1,8 @@
+/* Generic implementations already assume builtin support for
+ float, double, and _Float128 . */
+
+/* We can't enable it for long double due GCC Bugzilla Bug 29253
+ "expand_abs wrong default code for floating point" where it can
+ cause spurious "invalid" exceptions from comparisons with NaN in
+ the code sequence generated. */
+#define USE_FABSL_BUILTIN 0