Re: [RFC PATCH 4/7] kconfig: support new special property shell=

[Kees Cook]

On Sun, Feb 11, 2018 at 11:53 AM, Linus Torvalds
<torvalds@xxxxxxxxxxxxxxxxxxxx> wrote:
> On Sun, Feb 11, 2018 at 11:39 AM, Kees Cook <keescook@xxxxxxxxxxxx> wrote:
>> This made akpm and Arnd very very grumpy as it regressed their builds.
>> That's why I had to deal with the condition very carefully for _AUTO.
>
> Well, Arnd build new cross-tools last week, probably because you
> really need new tools for other reasons anyway (ie all the spectre
> mitigation). So I think Arnd is set.
>
> And akpm being on some ancient stone age system has been an issue
> before. The only way to fix it is to break it.

I don't seem to be the one that can declare such things, which is why
I pulled _AUTO out of -mm before the 4.15 merge window. It was
breaking akpm and Arnd, which seemed like a serious problem,
especially since you've yelled about how it's very bad to break
builds, etc.

> What I would worry about primarily is not one of the odd developers
> who can upgrade, but random people in the wild. I don't want to lose
> the occasional odd tester that does things nobody else does.
>
> But with gcc-4.5 being 7+ years old, I can't imagine it's a huge issue.

You have no objection from me at all on this. I have been yelled at
repeatedly (unjustly in this thread even) for "forcing options" on
people. I worked very hard to NOT break people with this, so it didn't
seem like a good plan to knowingly break builds. (And the other
options, _REGULAR and _STRONG, _do_ actually break the build if you
have a broken compiler. But for _AUTO, that seemed like a very bad
idea. I could only imagine the flames.)

-Kees

-- 
Kees Cook
Pixel Security