News for hackers. Stuff that matters.

A comment on Making AMD GPUs competitive for LLM inference. Parent comment

Re:Making AMD GPUs competitive for LLM inference by jereees on Thursday August 10, 2023 @08:53AM (#37075531)
He shows the git patch fix delivered by AMD engineers. The root cause is not explained and the change itself is about 2-3 SLoC — his frustration, what you call meltdown, is not about the interaction with AMD rather how troublesome it was to get AMD’s demo scripts to run on AMD’s chips.
[ Parent | on HN ]
Re:Making AMD GPUs competitive for LLM inference by pokeypokes on Thursday August 10, 2023 @10:29AM (#37076606)
Feeling frustration about something not working is a normal part of software development. Reporting a bug, or choosing not to use that software are normal too. Making a public video trashing the work of 100s, using the language he did, is a meltdown.

I've worked on public software before, and it's left me with an extremely low opinion of this kind of behaviour, and the kind of people that show it.

[ Parent | on HN ]