Re: [tip:x86/vdso] x86, vdso: Instead of dummy functions, include < linux/spinlock_up.h>

[Stefani Seibold]

Am Montag, den 17.02.2014, 01:27 -0800 schrieb H. Peter Anvin:
> On 02/16/2014 11:42 PM, Stefani Seibold wrote:
> > I think for the first time it will be okay to kick out the
> > _ASM_X86_SPINLOCK_H hack and accept the C=1 warnings.
> >
> > At next step it is necessary to make the whole BUILD_VDSO32 path in
> > vclock_gettime.c independent from the kernel headers, only uapi/ should
> > be included.
> >
> > The use of cycle_t must be replaced with u64.
> >
> > We need a own copy of __native_read_tsc(), __iter_div_u64_rem, smp_rmb()
> > and cpu_relax().
> 
> All of which are quite trivial.
> 
> > For the non BUILD_VDSO32 path we must only move the #includes inside
> > this #ifndef BUILD_VDSO32
> 
> Sorry, didn't quite follow that.
> 

The solution is quite simple: In case of a 32 bit VDSO for a 64 bit
kernel fake a 32 bit kernel configuration. Than everything is fine and
all kernel headers will compile without warnings or errors, also make
C=1 will give no complains.

The arch/x86/vdso/vdso32/vclock_gettime.c will now look like:

#define BUILD_VDSO32

#ifdef CONFIG_X86_64

/*
 * in case of a 32 bit VDSO for a 64 bit kernel fake a 32 bit kernel
 * configuration
 */
#undef CONFIG_64BIT
#undef CONFIG_X86_64
#undef CONFIG_ILLEGAL_POINTER_VALUE

#define CONFIG_X86_32 1
#define CONFIG_PAGE_OFFSET 0
#define CONFIG_ILLEGAL_POINTER_VALUE 0

#define BUILD_VDSO32_64

#endif

#include "../vclock_gettime.c"

and the following modifications for arch/x86/include/asm/vgtod.h:

#ifdef BUILD_VDSO32_64
typedef u64 gtod_long_t;
#else
typedef unsigned long gtod_long_t;
#endif

I tested it and i see no side effects. What do you think?

- Stefani


--
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/