Since the AVX2 version of mathvec functions uses FMA, it can only be
used when FMA is usable.
[BZ #21966]
* sysdeps/x86_64/fpu/multiarch/ifunc-mathvec-avx2.h
(IFUNC_SELECTOR): Don't use the AVX2 version if FMA isn't
usable.
+2017-08-18 H.J. Lu <hongjiu.lu@intel.com>
+
+ [BZ #21966]
+ * sysdeps/x86_64/fpu/multiarch/ifunc-mathvec-avx2.h
+ (IFUNC_SELECTOR): Don't use the AVX2 version if FMA isn't
+ usable.
+
2017-08-17 DJ Delorie <dj@redhat.com>
* bug17079.c: Update to new test harness.
{
const struct cpu_features* cpu_features = __get_cpu_features ();
- if (CPU_FEATURES_ARCH_P (cpu_features, AVX2_Usable))
+ if (CPU_FEATURES_ARCH_P (cpu_features, FMA_Usable)
+ && CPU_FEATURES_ARCH_P (cpu_features, AVX2_Usable))
return OPTIMIZE (avx2);
return OPTIMIZE (sse_wrapper);