Re: [RFT/PATCH v2 6/6] x86-64: Turn off -pg and turn on-foptimize-sibling-calls for vDSO

[Ingo Molnar]

* Andy Lutomirski <luto@xxxxxxx> wrote:

> The vDSO isn't part of the kernel, so profiling and kernel
> backtraces don't really matter.
> 
> Signed-off-by: Andy Lutomirski <luto@xxxxxxx>
> ---
>  arch/x86/vdso/Makefile |   15 ++++++++++++++-
>  1 files changed, 14 insertions(+), 1 deletions(-)
> 
> diff --git a/arch/x86/vdso/Makefile b/arch/x86/vdso/Makefile
> index a651861..bef0bc9 100644
> --- a/arch/x86/vdso/Makefile
> +++ b/arch/x86/vdso/Makefile
> @@ -37,11 +37,24 @@ $(obj)/%.so: OBJCOPYFLAGS := -S
>  $(obj)/%.so: $(obj)/%.so.dbg FORCE
>  	$(call if_changed,objcopy)
>  
> +#
> +# Don't omit frame pointers for ease of userspace debugging, but do
> +# optimize sibling calls.
> +#
>  CFL := $(PROFILING) -mcmodel=small -fPIC -O2 -fasynchronous-unwind-tables -m64 \
> -       $(filter -g%,$(KBUILD_CFLAGS)) $(call cc-option, -fno-stack-protector)
> +       $(filter -g%,$(KBUILD_CFLAGS)) $(call cc-option, -fno-stack-protector) \
> +       -fno-omit-frame-pointer -foptimize-sibling-calls
>  
>  $(vobjs): KBUILD_CFLAGS += $(CFL)
>  
> +#
> +# vDSO code runs in userspace and -pg doesn't help with profiling anyway.
> +#
> +CFLAGS_REMOVE_vdso-note.o = -pg
> +CFLAGS_REMOVE_vclock_gettime.o = -pg
> +CFLAGS_REMOVE_vgetcpu.o = -pg
> +CFLAGS_REMOVE_vvar.o = -pg

The comment makes no sense in this form, -pg is not used for profiling but 
in-kernel tracing.

Also, please do not do multiple changes in a single patch, split out the -pg, 
the -fomit-frame-pointers and the -foptimize-sibling-calls change into two 
patches. This will also help bisection, should some GCC version do something 
silly with -foptimize-sibling-calls ...

Thanks,

	Ingo
--
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/