linux-kernel - Re: [PATCH 2/2] Enable 'make CONFIG

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CACqU3MVyjzr47N9v23SGaSjiEOGoEaP1-TZ4Z_RyVNgVXgPEwg@mail.gmail.com>
Date:	Wed, 10 Aug 2011 23:29:46 -0400
From:	Arnaud Lacombe <lacombar@...il.com>
To:	David Woodhouse <dwmw2@...radead.org>
Cc:	Michal Marek <mmarek@...e.cz>, "H. Peter Anvin" <hpa@...or.com>,
	linux-kernel@...r.kernel.org, linux-kbuild@...r.kernel.org
Subject: Re: [PATCH 2/2] Enable 'make CONFIG_FOO=y oldconfig'

Hi,

On Wed, Aug 10, 2011 at 7:16 PM, David Woodhouse <dwmw2@...radead.org> wrote:
> On Wed, 2011-08-10 at 18:33 -0400, Arnaud Lacombe wrote:
>> > We're not enabling anything that we're later going to break. I can't see
>> > many people *depending* on the fact that 'make CONFIG_SATA_MV=y
>> > oldconfig' actually does *nothing* in some cases.
>> >
>> you are wrong, you ends up with half-baked compile-time dependency,
>> which break the build:
>
> s/*nothing*/nothing useful/, for crying out loud.
>
> The point remains that we're not enabling anything which we're later
> going to break. Nobody is going to come to *depend* on this behaviour.
>
> And anyway, this behaviour exists even *before* my patches, as Michal
> pointed out in a far more helpful and constructive fashion in
> <CACqU3MU4wJb3ij6skod-ZiM+Q0OMTXNdbJ+qWjJW8VZNEP+x1g@...l.gmail.com>
> earlier today.
>
> It's simple enough to fix, too:
>
> diff --git a/Makefile b/Makefile
> index 1fc5172..6cc7f7b 100644
> --- a/Makefile
> +++ b/Makefile
> @@ -474,6 +474,9 @@ ifeq ($(KBUILD_EXTMOD),)
>         endif
>  endif
>
> +# This could probably be simpler?
> +CONFIG_OVERRIDES := $(patsubst line:%,%,$(filter line:%,$(foreach var, $(filter CONFIG_%,$(.VARIABLES)), $(origin $(var)):$(var))))
> +
>  ifeq ($(mixed-targets),1)
>  # ===========================================================================
>  # We're called with mixed targets (*config and build targets).
> @@ -507,6 +510,10 @@ else
>  # Build targets only - this includes vmlinux, arch specific targets, clean
>  # targets and others. In general all targets except *config targets.
>
> +ifneq ($(CONFIG_OVERRIDES),)
> +$(error Cannot perform build targets with CONFIG symbols overridden)
> +endif
> +
>  ifeq ($(KBUILD_EXTMOD),)
>  # Additional helpers built in scripts/
>  # Carefully list dependencies so we do not try to build scripts twice
>
how is that supposed to work with your other patches ?

% vi mail.txt
[strip all lines until 'diff --git a/Makefile b/Makefile']
% git apply mail.txt
% make CONFIG_NET=y allyesconfig drivers/ata/
[...]
scripts/kconfig/conf --allyesconfig Kconfig
#
# CONFIG_NET set to y from environment
#
#
# configuration written to .config
#
Makefile:504: *** Cannot perform build targets with CONFIG symbols
overridden.  Stop.
make: *** [prepare] Error 2

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