Hi Maciej:
Thanks for detail analysis and performance number report, I am concern
about this patch might let compiler schedule the fsflags/frflags with
other floating point instructions, and the major issue is we didn't
model fflags right in GCC as you mentioned in the first email.
So I think we
On Thu, 9 Jun 2022, Maciej W. Rozycki wrote:
> I'm yet running some benchmarking to see if the use of UNSPEC_VOLATILEs
> makes any noticeable performance difference, but I suspect it does not as
> the compiler could not do much about the original multiple-instruction
> single RTL insns anyway.
We have unordered FP comparisons implemented as RTL insns that produce
multiple machine instructions. Such RTL insns are hard to match with a
processor pipeline description and additionally there is a redundant
SNEZ instruction produced on the result of these comparisons even though
the FLT.fm
