Lmst

"A recent patch to glibc6 to add #POWER9 VSX strncpy was a whopping 250 hand-crafted assembly instructions, where its equivalent using Cray Vector principles is around 14."
https://archive.fosdem.org/2021/schedule/event/the_libresoc_project_simple_v_vectorisation/
(haven't watched yet)

This reminds me of that #MillComputing talk about how they automatically parallelize strlen.
I think it was this one:
https://redirect.invidious.io/watch?v=JS5hCjueqQ0&list=PLFls3Q5bBInj_FfNLrV7gGdVtikeGoUc9

:blobowo:
"There’s a lot to do, but we are out of research ind into development, and money and talent we can use both of now."
https://millcomputing.com/topic/news/#post-3533

was supposed to be cleaning my room but got stuck on this talk [1]
it's really cool how they vectorize that strcpy.
https://youtu.be/DZ8HN9Cnjhc

"Programs have got no business mucking with state. The hardware should muck with state because the hardware knows what state is."

> Stanford Seminar -Drinking from the Firehose: How the Mill #CPU Decodes 30+ Instructions per Cycle
#MillComputing

"Mill vs. #Spectre: Performance and Security" by Ivan Godard
https://invidio.us/watch?v=8E4qs2irmpc

@alcinnz This made me wonder what's up with them, and what do you know, Ivan Goddard posted a status update 49 seconds before I opened their forum:
https://millcomputing.com/topic/news/#post-3450

I wonder how the hype around #riscv will affect #millcomputing 🤔

#MillComputing