On 7/18/25 7:06 AM, Li, Pan2 wrote:
OK. Curious if you've seen this show up in practice and is using the
vaaddu and similar instructions actually profitable?
I remember avg will be presented in somewhere else but not for DImode.
I tried to add vaadd related vx combine and notice we don't have DImode for
rvv, so make it available before that.
Thanks.
Point is we probably need to be careful here and we probably want a
something in the tuning structure to steer us away from vaaddu and
friends on some designs. I was tentatively planning to have Austin get
that entry into the tuning structure and conditionalize the appropriate
code.
I see, thanks Jeff.
FWIW, we tested the BPI last night. It doesn't seem sensitive to the
VXRM issues, so hopefully the issues are just in the early sifive vector
designs.
Jeff
