https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115795
--- Comment #8 from Jordi Sala ---
(In reply to JuzheZhong from comment #7)
> (In reply to Jordi Sala from comment #6)
> > Perfect, that's what I was looking for. I'm thinking of adding a way to tell
> > GCC to minimize, maximize or preserve SEW
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115795
--- Comment #6 from Jordi Sala ---
Perfect, that's what I was looking for. I'm thinking of adding a way to tell
GCC to minimize, maximize or preserve SEW on vsetvl expand. Like
-mrvv-vsetvl-sew={maximize,minimize,preserve}
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115795
--- Comment #4 from Jordi Sala ---
problem is this is not related to the vectorizer as far as I'm aware, so
setting -mrvv-max-lmul=m8 does not change the fact that vsetvl pass is going to
change the loads and stores from m8 to m2
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115795
--- Comment #2 from Jordi Sala ---
Thanks, is it possible to force a bigger LMUL currently? It's mainly for
performance reasons.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115795
Bug ID: 115795
Summary: RISC-V: vsetvl step causes wrong codegen after fusing
info
Product: gcc
Version: 15.0
Status: UNCONFIRMED
Severity: normal
Pr
