On Tue, May 10, 2022 at 2:31 AM Borislav Petkov <bp@xxxxxxxxx> wrote:
>
> clear_user_original:
> Amean: 9219.71 (Sum: 6340154910, samples: 687674)
>
> fsrm:
> Amean: 8030.63 (Sum: 5522277720, samples: 687652)
Well, that's pretty conclusive.
I'm obviously very happy with fsrm. I've been pushing for that thing
for probably over two decades by now, because I absolutely detest
uarch optimizations for memset/memcpy that can never be done well in
software anyway (because it depends not just on cache organization,
but on cache sizes and dynamic cache hit/miss behavior of the load).
And one of the things I always wanted to do was to just have
memcpy/memset entirely inlined.
In fact, if you go back to the 0.01 linux kernel sources, you'll see
that they only compile with my bastardized version of gcc-1.40,
because I made the compiler inline those things with 'rep movs/stos',
and there was no other implementation of memcpy/memset at all.
That was a bit optimistic at the time, but here we are, 30+ years
later and it is finally looking possible, at least on some uarchs.
Linus
