Hi Richard ! On Thu, Jun 30, 2022 at 4:45 PM Richard Earnshaw <richard.earns...@arm.com> wrote: > > I think the problem is valgrind's Makefiles are passing -mcpu=cortex-a8 > to the compiler. Cortex-a8 has Neon and the compiler now makes use of that. > > On the subject of the configuration of GCC > > --with-arch=armv7-a+fp
Thanks for the quick response. > *is* the correct configuration for the baseline GCC; it adds a vfpv3 > with 16 double-precision registers. +vfpv3-fp16 would add support for > the float16 load/store operations, while +nosimd is a nop because no > other options have been added to enable SIMD. I managed to misread the gcc arm options page (fp16 != d16). Thanks for the explanation. I've suggested valgrind people the following patch: % sed -i -e 's/cortex-a8/generic-armv7-a+vfpv3-d16/g' Makefile.all.am Hopefully my understanding is correct this time. -M