On Fri, May 30, 2025 at 10:48:47AM -0700, Kees Cook wrote:

> This needs docs too. I think this is collapsing 1 and 2 argument WARNs
> into the ud1, and anything larger is explicitly calling __warn_printk +
> __WARN_FLAGS? If only 1 and 2 args can be collapsed, why reserve 0xfe00
> through 0xfeff? I feel like I'm missing something here...

I did something really evil in that first version, I encoded the
registers into the bug type. I've fixed that and have decode_bug emit a
ud_regs thingy now.

Reply via email to