Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1283169 > unrolled thread

Re: [PATCH v4 4/9] ppc64 ftrace_with_regs configuration variables

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: [PATCH v4 4/9] ppc64 ftrace_with_regs configuration variables Petr Mladek <pmladek@suse.com> - 2015-12-03 17:30 +0100
    Re: [PATCH v4 4/9] ppc64 ftrace_with_regs configuration variables Torsten Duwe <duwe@lst.de> - 2015-12-04 10:10 +0100

#1283169 — Re: [PATCH v4 4/9] ppc64 ftrace_with_regs configuration variables

On Wed 2015-11-25 17:37:33, Torsten Duwe wrote:
>   * Makefile:
>     - globally use -mprofile-kernel in case it's configured.
>   * arch/powerpc/Kconfig / kernel/trace/Kconfig:
>     - declare that ppc64le HAVE_MPROFILE_KERNEL and
>       HAVE_DYNAMIC_FTRACE_WITH_REGS, and use it.
> 
> --- a/arch/powerpc/Makefile
> +++ b/arch/powerpc/Makefile
> @@ -133,6 +133,13 @@ else
>  CFLAGS-$(CONFIG_GENERIC_CPU) += -mcpu=powerpc64
>  endif
>  
> +ifeq ($(CONFIG_PPC64),y)
> +ifdef CONFIG_HAVE_MPROFILE_KERNEL
> +CC_FLAGS_FTRACE	:= -pg $(call cc-option,-mprofile-kernel)

Do we want to define -pg even when -mprofile-kernel is not supported
by the used gcc, please?

> +KBUILD_CPPFLAGS	+= -DCC_USING_MPROFILE_KERNEL

IMHO, we should not define CC_USING_MPROFILE_KERNEL if it is not
supported by the compiler.

I took inspiration from the CC_USING_FENTRY handling in
linux/Makefile. The following code worked for me:

CC_USING_MPROFILE_KERNEL := $(call cc-option, -pg -mprofile-kernel -DCC_USING_MPROFILE_KERNEL)
CC_FLAGS_FTRACE := $(CC_USING_MPROFILE_KERNEL)
KBUILD_CPPFLAGS += $(CC_USING_MPROFILE_KERNEL)

I just do not understand why we need to add the flags also
to KBUILD_CPPFLAGS. It seems that they are duplicated
when compiling kernel/livepatch/core.o. But livepatching
did not work without it. I wonder if you found the culprit.

Best Regards,
Petr

> +endif
> +endif
> +
>  CFLAGS-$(CONFIG_CELL_CPU) += $(call cc-option,-mcpu=cell)
>  CFLAGS-$(CONFIG_POWER4_CPU) += $(call cc-option,-mcpu=power4)
>  CFLAGS-$(CONFIG_POWER5_CPU) += $(call cc-option,-mcpu=power5)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [next] | [standalone]

#1283660

On Thu, Dec 03, 2015 at 05:20:08PM +0100, Petr Mladek wrote:
>
> IMHO, we should not define CC_USING_MPROFILE_KERNEL if it is not
> supported by the compiler.

Yes, true.

> I took inspiration from the CC_USING_FENTRY handling in
> linux/Makefile. The following code worked for me:
>
> CC_USING_MPROFILE_KERNEL := $(call cc-option, -pg -mprofile-kernel -DCC_USING_MPROFILE_KERNEL)
> CC_FLAGS_FTRACE := $(CC_USING_MPROFILE_KERNEL)
> KBUILD_CPPFLAGS += $(CC_USING_MPROFILE_KERNEL)

Excellent!

> I just do not understand why we need to add the flags also
> to KBUILD_CPPFLAGS. It seems that they are duplicated
> when compiling kernel/livepatch/core.o. But livepatching
> did not work without it. I wonder if you found the culprit.

Some assembler-with-cpp files also need to be notified?

My plan is to first get this working reliably and then fine tune.

	Torsten

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web