diff mbox series

[v3,14/16] kconfig: Add support for 'as-option'

Message ID 1576486038-9899-15-git-send-email-amit.kachhap@arm.com (mailing list archive)
State New, archived
Headers show
Series None | expand

Commit Message

Amit Daniel Kachhap Dec. 16, 2019, 8:47 a.m. UTC
From: Vincenzo Frascino <vincenzo.frascino@arm.com>

Currently kconfig does not have a feature that allows to detect if the
used assembler supports a specific compilation option.

Introduce 'as-option' to serve this purpose in the context of Kconfig:

	config X
		def_bool $(as-option,...)

Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
Cc: linux-kbuild@vger.kernel.org
Signed-off-by: Vincenzo Frascino <vincenzo.frascino@arm.com>
Signed-off-by: Amit Daniel Kachhap <amit.kachhap@arm.com>
---
Changes since last version:
* None.

 scripts/Kconfig.include | 4 ++++
 1 file changed, 4 insertions(+)

Comments

Masahiro Yamada Dec. 17, 2019, 11:43 a.m. UTC | #1
On Mon, Dec 16, 2019 at 5:48 PM Amit Daniel Kachhap
<amit.kachhap@arm.com> wrote:
>
> From: Vincenzo Frascino <vincenzo.frascino@arm.com>
>
> Currently kconfig does not have a feature that allows to detect if the
> used assembler supports a specific compilation option.
>
> Introduce 'as-option' to serve this purpose in the context of Kconfig:
>
>         config X
>                 def_bool $(as-option,...)
>
> Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
> Cc: linux-kbuild@vger.kernel.org
> Signed-off-by: Vincenzo Frascino <vincenzo.frascino@arm.com>
> Signed-off-by: Amit Daniel Kachhap <amit.kachhap@arm.com>
> ---

If you apply this to arm64 tree, just go ahead.
Acked-by: Masahiro Yamada <masahiroy@kernel.org>

But, please be aware of a possible merge conflict with:
https://patchwork.kernel.org/patch/11285929/
Catalin Marinas Jan. 17, 2020, 11:33 a.m. UTC | #2
On Mon, Dec 16, 2019 at 02:17:16PM +0530, Amit Daniel Kachhap wrote:
> diff --git a/scripts/Kconfig.include b/scripts/Kconfig.include
> index d4adfbe..cc465dd 100644
> --- a/scripts/Kconfig.include
> +++ b/scripts/Kconfig.include
> @@ -31,6 +31,10 @@ cc-option = $(success,$(CC) -Werror $(CLANG_FLAGS) $(1) -E -x c /dev/null -o /de
>  # Return y if the linker supports <flag>, n otherwise
>  ld-option = $(success,$(LD) -v $(1))
>  
> +# $(as-option,<flag>)
> +# Return y if the assembler supports <flag>, n otherwise
> +as-option = $(success, $(CC) $(CLANG_FLAGS) $(1) -E -x assembler /dev/null -o /dev/null)

I had different experiments with this for MTE and noticed that clang
does not honour the -Wa, option (which you use in a subsequent patch).
So not sure how useful as-option is.
Masahiro Yamada Jan. 17, 2020, 5:51 p.m. UTC | #3
On Fri, Jan 17, 2020 at 8:33 PM Catalin Marinas <catalin.marinas@arm.com> wrote:
>
> On Mon, Dec 16, 2019 at 02:17:16PM +0530, Amit Daniel Kachhap wrote:
> > diff --git a/scripts/Kconfig.include b/scripts/Kconfig.include
> > index d4adfbe..cc465dd 100644
> > --- a/scripts/Kconfig.include
> > +++ b/scripts/Kconfig.include
> > @@ -31,6 +31,10 @@ cc-option = $(success,$(CC) -Werror $(CLANG_FLAGS) $(1) -E -x c /dev/null -o /de
> >  # Return y if the linker supports <flag>, n otherwise
> >  ld-option = $(success,$(LD) -v $(1))
> >
> > +# $(as-option,<flag>)
> > +# Return y if the assembler supports <flag>, n otherwise
> > +as-option = $(success, $(CC) $(CLANG_FLAGS) $(1) -E -x assembler /dev/null -o /dev/null)
>
> I had different experiments with this for MTE and noticed that clang
> does not honour the -Wa, option (which you use in a subsequent patch).
> So not sure how useful as-option is.


I think this is because it uses '-E' option.

To invoke the assembler, -c is needed.


I replaced -E with -c, and tested it.
It seems working for both gcc and clang.



I noticed a similar case for cc-option:
https://patchwork.kernel.org/patch/11339567/
Vincenzo Frascino Jan. 20, 2020, 10:26 a.m. UTC | #4
Hi Catalin,

On 17/01/2020 11:33, Catalin Marinas wrote:
> On Mon, Dec 16, 2019 at 02:17:16PM +0530, Amit Daniel Kachhap wrote:
>> diff --git a/scripts/Kconfig.include b/scripts/Kconfig.include
>> index d4adfbe..cc465dd 100644
>> --- a/scripts/Kconfig.include
>> +++ b/scripts/Kconfig.include
>> @@ -31,6 +31,10 @@ cc-option = $(success,$(CC) -Werror $(CLANG_FLAGS) $(1) -E -x c /dev/null -o /de
>>  # Return y if the linker supports <flag>, n otherwise
>>  ld-option = $(success,$(LD) -v $(1))
>>  
>> +# $(as-option,<flag>)
>> +# Return y if the assembler supports <flag>, n otherwise
>> +as-option = $(success, $(CC) $(CLANG_FLAGS) $(1) -E -x assembler /dev/null -o /dev/null)
> 
> I had different experiments with this for MTE and noticed that clang
> does not honour the -Wa, option (which you use in a subsequent patch).
> So not sure how useful as-option is.
> 

Not sure of your experiments for MTE, the kernel is built with clang but uses
the GNU assembler. The -Wa, option is passed down to the assembler (GNU) which
honors it.

Said that, I agree with Masahiro Yamada about the "-c" option to be used in
place of "-E" and on that sense I provided a patch to Amit last December which
should be integrated in the next series (Amit can confirm, I hope ;) ).
diff mbox series

Patch

diff --git a/scripts/Kconfig.include b/scripts/Kconfig.include
index d4adfbe..cc465dd 100644
--- a/scripts/Kconfig.include
+++ b/scripts/Kconfig.include
@@ -31,6 +31,10 @@  cc-option = $(success,$(CC) -Werror $(CLANG_FLAGS) $(1) -E -x c /dev/null -o /de
 # Return y if the linker supports <flag>, n otherwise
 ld-option = $(success,$(LD) -v $(1))
 
+# $(as-option,<flag>)
+# Return y if the assembler supports <flag>, n otherwise
+as-option = $(success, $(CC) $(CLANG_FLAGS) $(1) -E -x assembler /dev/null -o /dev/null)
+
 # check if $(CC) and $(LD) exist
 $(error-if,$(failure,command -v $(CC)),compiler '$(CC)' not found)
 $(error-if,$(failure,command -v $(LD)),linker '$(LD)' not found)