Re: [PATCH 2/2] Enable 'make CONFIG_FOO=y oldconfig'

[Randy Dunlap <rdunlap@xenotime.net>]

On Sun, 31 Jul 2011 08:33:39 +0100 David Woodhouse wrote:

> On Sat, 2011-07-30 at 21:06 -0400, Arnaud Lacombe wrote:
> > The principle of least surprise is broken anyway as the proposed patch
> > has absolutely no dependency checking and verification. You can `make
> > CONFIG_SATA_MV=y allnoconfig', you will _not_ get it set.
> 
> That's always true in kconfig *anyway*. We've *never* really had an
> option for "do whatever you need to enable this option". We've even
> hard-coded this failure in our language, by introducing this horrible
> 'select' thing to work around it.
> 
> I'd no more expect that, than I would for it to write the code for me if
> I type 'make CONFIG_BTRFSv2=y oldconfig'.
> 
> So no, it doesn't violate the principle of least surprise.
> 
> > ok, the issue is that you will only be allowed to change visible
> > symbols. CONFIG_64BIT is conditionally visible (when ARCH=x86), so
> > right now, you can not do on x86-64:
> > 
> > % make ARCH=i386 defconfig
> > % make CONFIG_64BIT=n oldconfig # [0]
> 
> That works fine here. What was ARCH set to in your second test? If it's
> ARCH=x86_64 then that's expected. That's the whole point of my *other*
> patch to make 'ARCH=x86' be the default, so that the value of
> CONFIG_64BIT in your .config is *not* forcibly overridden to match the
> build host. That's a *separate* bug, which I also have a patch for.

Simple question:  what does "ARCH=x86" mean?

It doesn't mean anything to me without SUBARCH or nnBIT specified.


---
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***