On Tue, Jul 30, 2013 at 2:38 PM, Arpit Goel [off-list ref] wrote:

This patch-set moves USECS_PER_JIFFY to architecture timex.h from architecture
specific C files. Patch-set further uses this to make PPC macro
spin_event_timeout() architecture independent. This change enables drivers to
use spin_event_timeout() even in non-PowerPC based SoC's.

This patchset has been compiled for ARM, PowerPC and x86.

Patch[1/2] : Moves USECS_PER_JIFFY to architecture specific timex.h files from
             architecture specific C files.

 arch/arm/include/asm/timex.h      |  2 ++
 arch/arm/kernel/time.c            |  3 ---
 arch/m32r/include/asm/timex.h     |  2 ++
 arch/m32r/kernel/time.c           |  3 ---
 arch/m68k/hp300/time.c            |  4 +---
 arch/m68k/include/asm/timex.h     |  2 ++
 arch/s390/include/asm/timex.h     |  2 ++
 arch/s390/kernel/time.c           |  2 --
 arch/sparc/include/asm/timex_32.h |  2 ++
 arch/sparc/include/asm/timex_64.h |  2 ++
 arch/sparc/kernel/pcic.c          |  1 -

As the definition is the same on all architectures, why not move it to
<linux/delay.h>?

Patch[2/2] : Converts PPC macro spin_event_timeout() in architecture independent
             macro.

 include/linux/delay.h             | 40 +++++++++++++++++++++++++++++++++++++++

After your first patch, USECS_PER_JIFFY is still available only on
arm/m32r/m68k/s390/sparc. What about other architectures?
If anyone uses spin_event_timeout() on e.g. x86, it will fail to compile.

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert at linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds