[PATCH 0/2] perf: perf_arch_fetch_regs_caller API change

From: Frederic Weisbecker
Date: Fri May 21 2010 - 04:11:20 EST


Hi,

I did some updates in this function, we don't need anymore
to skip further the first caller now. So I think it's worth
the simplification to lower the overhead in the tracing fast
path.

I took Paul updates for PowerPc and I did the changes on x86.
But I'm clearly useless about the sparc part, I'm still confused
about these window things. So all I did in this arch is to
wrap the older API with the new one, always passing the skip to 1.
I suspect David can be more useful than me to do a real change there :)

Tell me what you think,

Thanks.

Frederic Weisbecker (2):
x86: Unify dumpstack.h and stacktrace.h
perf: Drop the skip argument from perf_arch_fetch_regs_caller

arch/powerpc/include/asm/perf_event.h | 12 +++++++
arch/powerpc/kernel/misc.S | 26 ---------------
arch/sparc/include/asm/perf_event.h | 4 ++
arch/sparc/kernel/helpers.S | 6 ++--
arch/x86/include/asm/perf_event.h | 9 +++++
arch/x86/include/asm/stacktrace.h | 49 ++++++++++++++++++++++++++++
arch/x86/kernel/cpu/perf_event.c | 14 --------
arch/x86/kernel/dumpstack.c | 1 -
arch/x86/kernel/dumpstack.h | 56 ---------------------------------
arch/x86/kernel/dumpstack_32.c | 2 -
arch/x86/kernel/dumpstack_64.c | 1 -
arch/x86/kernel/stacktrace.c | 7 ++--
include/linux/perf_event.h | 32 ++++--------------
include/trace/ftrace.h | 2 +-
kernel/perf_event.c | 5 ---
kernel/trace/trace_event_perf.c | 2 -
16 files changed, 89 insertions(+), 139 deletions(-)
delete mode 100644 arch/x86/kernel/dumpstack.h

--
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/