On Jan 31, 2026, Jeffrey Law <[email protected]> wrote:

> In general we haven't put much time (yet) into making dynamic-lmul
> real.   Though I think everyone involved in the RVV effort expects
> that over time good support for dynamic lmul is where we want to go.

> FWIW, a larger LMUL isn't always a win, even if it doesn't trigger
> spilling or anything like that.  But my hope is that with few
> exceptions that a larger LMUL doesn't hurt performance in the future
> (it certainly can right now on current designs).

Cool, thanks for the tips, they're very welcome.  This has been my very
first close encounter of the third kind :-) with RVV, and it's been
quite exciting, but I've still got a long way to go to be able to label
myself as familiar with it :-)

-- 
Alexandre Oliva, happy hacker            https://blog.lx.oliva.nom.br/
Free Software Activist     FSFLA co-founder     GNU Toolchain Engineer
More tolerance and less prejudice are key for inclusion and diversity.
Excluding neuro-others for not behaving ""normal"" is *not* inclusive!

Reply via email to