Am 25.04.26 um 09:44 schrieb Eric Botcazou:
I’d also not accept big infrastructure designed by LLMs. I still hope for
sensible use of LLMs for GCC, like finishing long neglected partial
transitions. I do not think LLMs are ready to implement new compiler
features.
Yes, that seems to be a safe and sensible middle ground, in other words to
limit changes to the local "scope" so that any mistakes made are also local.
Something (slightly) above mechanical, but certainly nowhere near global.
Another point:
I think that patches which contain LLM-generated code need to receive
special care from *reviewers*, as well (to be formalized how?)
Best regards
Thomas
