Re: 2.6.22-rc4-mm2: kvm compile breakage with X86_CMPXCHG64=n

[Dave Jones <davej@redhat.com>]

On Tue, Jun 12, 2007 at 03:03:57PM -0700, Andrew Morton wrote:
 > On Mon, 11 Jun 2007 23:22:24 -0400
 > Dave Jones <davej@redhat.com> wrote:
 > 
 > > Add -Werror-implicit-function-declaration
 > > This makes builds fail sooner if something is implicitly defined instead
 > > of having to wait half an hour for it to fail at the linking stage.
 > > 
 > > Signed-off-by: Dave Jones <davej@redhat.com>
 > > 
 > > --- linux-2.6/Makefile~	2007-06-04 16:46:24.000000000 -0400
 > > +++ linux-2.6/Makefile	2007-06-04 16:46:53.000000000 -0400
 > > @@ -313,7 +313,8 @@ LINUXINCLUDE    := -Iinclude \
 > >  CPPFLAGS        := -D__KERNEL__ $(LINUXINCLUDE)
 > >  
 > >  CFLAGS          := -Wall -Wundef -Wstrict-prototypes -Wno-trigraphs \
 > > -                   -fno-strict-aliasing -fno-common
 > > +		   -fno-strict-aliasing -fno-common \
 > > +		   -Werror-implicit-function-declaration
 > >  AFLAGS          := -D__ASSEMBLY__
 > >  
 > >  # Read KERNELRELEASE from include/config/kernel.release (if it exists)
 > 
 > This causes the i386 allmodconfig build to fail:

Seems to be doing its job rather effectively.

 > include/linux/uaccess.h: In function 'pagefault_disable':
 > include/linux/uaccess.h:23: error: implicit declaration of function '__memory_barrier'
 > 
 > I didn't look to see why...

include/linux/compiler.h ..

/* Optimization barrier */
#ifndef barrier
# define barrier() __memory_barrier()
#endif

We shouldn't be hitting this, because barrier should be getting defined
in the compiler specific headers above..

#if __GNUC__ >= 4
# include <linux/compiler-gcc4.h>
#elif __GNUC__ == 3 && __GNUC_MINOR__ >= 2
# include <linux/compiler-gcc3.h>
#else
# error Sorry, your compiler is too old/not recognized.
#endif

both of those include linux/compiler-gcc.h, which defines barrier.
How strange.  What compiler version is this?

	Dave

-- 
http://www.codemonkey.org.uk