On 6/12/23 01:36, Manolis Tsamis wrote:
Even if late, one question for the dynamic instruction numbers.
Was this measured just with f-m-o or with the stack pointer fold patch
applied too?
I remember I was getting better improvements in the past, but most of
the cases had to do with the stack pointer so the second patch is
necessary.
It was just the main f-m-o patch, so there'll be additional benefits
with the ability to cprop the stack pointer. And even if we don't get
measurable wins for something like mcf due to its memory bound nature,
smaller, tighter code is always preferable.
Jeff
