On 2020-Nov-10, David Rowley wrote:
> On Mon, 9 Nov 2020 at 16:29, Andy Fan <zhihui.fan1213@gmail.com> wrote:
> > However I believe v9
> > should be no worse than v8 all the time, Is there any theory to explain
> > your result?
>
> Nothing jumps out at me from looking at profiles. The only thing I
> noticed was the tuple deforming is more costly with v9. I'm not sure
> why.
Are you taking into account the possibility that generated machine code
is a small percent slower out of mere bad luck? I remember someone
suggesting that they can make code 2% faster or so by inserting random
no-op instructions in the binary, or something like that. So if the
difference between v8 and v9 is that small, then it might be due to this
kind of effect.
I don't know what is a good technique to test this hypothesis.