feld on Nostr: so I fired up my local model (don't even need the big scary Claude Mythos super ...

so I fired up my local model (don't even need the big scary Claude Mythos super brain), told it to review this kernel commit as an expert code reviewer and to not trust anything and investigate the full impact of changes and it spotted the null pointer dereference

MEANWHILE the commit was signed-off by 3 kernel developers and "reviewed" by one other

Do we get to call this human slop? Is this acceptable? Is there any reason that we should believe that this code was actually reviewed or that the "Signed-off-by" has any real value?

quoting
note19mp…yrsj

What good is code review and "signed off by" for kernel commits if nobody ever actually reads the code? These are supposed be some of the best developers on the planet right?

https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=linux-6.6.y&id=5a1e865e51063d6c56f673ec8ad4b6604321b455