Re: [PATCH][RESEND] fix compilation error with CROSS_COMPILE

From: Sam Ravnborg
Date: Sat Aug 21 2010 - 08:59:19 EST


>
> However I confirmed that the linux-2.6 tree is OK, the above patch that caused
> the problem is still sitting in -mm tree:
>
> commit 7f48a0a8c15c41f382f2265088bc96007836e608
> Author: Tiago Maluta <tiago.maluta@xxxxxxxxx>
> Date: Mon Jun 21 22:00:17 2010 +0200
>
> Linux has an recent option to define CROSS_COMPILE in menuconfig.
>
> Typically, one of the options when building for another target is
> ('arm' is just an example):
>
> # make ARCH=arm CROSS_COMPILE=arm-cc O=../build
>
> If I omit CROSS_COMPILE and use the new way in .config:
> CONFIG_CROSS_COMPILE="arm-cc"
>
> I got:
>
> cc1: error: unrecognized command line option "-mlittle-endian"
> cc1: error: unrecognized command line option "-mabi=aapcs-linux"
> cc1: error: unrecognized command line option "-mno-thumb-interwork"
> /usr/src/linux/kernel/bounds.c:1: error: bad value (armv5t) for -march= switch
> /usr/src/linux/kernel/bounds.c:1: error: bad value (armv5t) for -mtune= switch
> make[2]: *** [kernel/bounds.s] Error 1
> make[1]: *** [prepare0] Error 2
> make: *** [sub-make] Error 2
>
> Verbose output clearly says that I'm using the host compiler instead cross
> compiler. This patch fixed my problem.
>
> Cc: Michal Marek <mmarek@xxxxxxx>
> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
>
> diff --git a/Makefile b/Makefile
> index 7431c28..13126ae 100644
> --- a/Makefile
> +++ b/Makefile
> @@ -190,7 +190,7 @@ SUBARCH := $(shell uname -m | sed -e s/i.86/i386/ -e s/sun4u/sparc64/ \
> export KBUILD_BUILDHOST := $(SUBARCH)
> ARCH ?= $(SUBARCH)
> CROSS_COMPILE ?=
> -CROSS_COMPILE ?= $(CONFIG_CROSS_COMPILE:"%"=%)
> +CROSS_COMPILE := $(shell grep CONFIG_CROSS_COMPILE .config | cut -d'"' -f2)

So this patch does two things:
1) It grep for CONFIG_CROSS_COMPILE because we may not have incldued .config
2) It no longer pick up CROSS_COMPILE from the environment.

Traditionally we have picked up ARCH and CROSS_COMPILE from the environmnet,
hence the use of "?=".
"?=" says - perfrom this assignmnet unless lhs is already defined.
And lhs may be already defined is we have the ysmbol defined in the
environmnet.

I think the original submitter chenged from ?= to := without
realising this.
So the patch should be adjusted to use ?= again as you suggest
before it is applied.
I assume Michal / Andrew will take care.

Subject on patch could also be better.
Suggestion: kbuild: fix reading CROSS_COMPILE from .config

Sam
--
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/