Re: [PATCH, RFC] x86: also CFI-annotate certain inline asm()s

[Linus Torvalds]

On Mon, Nov 10, 2014 at 11:42 PM, Jan Beulich <JBeulich@xxxxxxxx> wrote:
>
> Nothing crashes with the unwind information being wrong. It is
> solely you who was claiming (without proof) years ago that the
> unwinder repeatedly caused issues.

Umm. We had oopses showing it. Several times.

> Yes, we did find a bug or two over the years in it

.. and you and Andi repeatedly refused to make the code more robust
when I asked.

Which is why I don't work with Andi or you directly any more, and why
the code got entirely removed when I got fed up with the last time it
crashed in ways that it *wouldn't* have crashed if it had been made
more robust.

Every time there were just excuses. Like now. "All code has bugs".
Sure. But debugging code had better be better than "all code", and it
had better be written to be robust, and it had better not make bugs
more _likely_ either by failing, or by making the non-debug code
harder to read.

I'm done with your crazy unwinder games. We removed a lot of CFI stuff
entirely because it made the asm code hard to read. Much of it ended
up getting re-introduced when you made the infrastructure more
bearable.

But this patch I NAK'ed because the code is not readable, and the
infrastructure is not bearable.

Live with it.

                      Linus
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/