[PATCH] Wipe out traditional opt from x86_64 Makefile

[PATCH] Wipe out traditional opt from x86_64 Makefile

[Glauber de Oliveira Costa]

Among other things, using -traditional as a gcc option stops us from
using macro token pasting, which is a feature we heavily rely on.

There was still a use of -traditional in arch/x86/kernel/Makefile_64,
which this patch removes.

I don't see any problems building kernels in my x86_64 box without
-traditional.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
---
 arch/x86/kernel/Makefile_64 |    1 -
 1 files changed, 0 insertions(+), 1 deletions(-)

diff --git a/arch/x86/kernel/Makefile_64 b/arch/x86/kernel/Makefile_64
index 82da906..03a01f6 100644
--- a/arch/x86/kernel/Makefile_64
+++ b/arch/x86/kernel/Makefile_64
@@ -3,7 +3,6 @@
 #
 
 extra-y 	:= head_64.o head64.o init_task.o vmlinux.lds
-#EXTRA_AFLAGS	:= -traditional
 obj-y	:= process_64.o signal_64.o entry_64.o traps_64.o irq_64.o \
 		ptrace_64.o time_64.o ioport_64.o ldt_64.o setup_64.o i8259_64.o sys_x86_64.o \
 		x8664_ksyms_64.o i387_64.o syscall_64.o vsyscall_64.o \
-- 
1.4.4.2

Re: [PATCH] Wipe out traditional opt from x86_64 Makefile

[Glauber de Oliveira Costa]

Ok, ok... this was stupid.

I sent the patch against an already-commented out version, which is
the one I was using for testing. Really stupid.

On 10/25/07, Glauber de Oliveira Costa <gcosta@redhat.com> wrote:
> Among other things, using -traditional as a gcc option stops us from
> using macro token pasting, which is a feature we heavily rely on.
>
> There was still a use of -traditional in arch/x86/kernel/Makefile_64,
> which this patch removes.
>
> I don't see any problems building kernels in my x86_64 box without
> -traditional.
>
> Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
> ---
>  arch/x86/kernel/Makefile_64 |    1 -
>  1 files changed, 0 insertions(+), 1 deletions(-)
>
> diff --git a/arch/x86/kernel/Makefile_64 b/arch/x86/kernel/Makefile_64
> index 82da906..03a01f6 100644
> --- a/arch/x86/kernel/Makefile_64
> +++ b/arch/x86/kernel/Makefile_64
> @@ -3,7 +3,6 @@
>  #
>
>  extra-y        := head_64.o head64.o init_task.o vmlinux.lds
> -#EXTRA_AFLAGS  := -traditional
>  obj-y  := process_64.o signal_64.o entry_64.o traps_64.o irq_64.o \
>                 ptrace_64.o time_64.o ioport_64.o ldt_64.o setup_64.o i8259_64.o sys_x86_64.o \
>                 x8664_ksyms_64.o i387_64.o syscall_64.o vsyscall_64.o \
> --
> 1.4.4.2
>
>


-- 
Glauber de Oliveira Costa.
"Free as in Freedom"
http://glommer.net

"The less confident you are, the more serious you have to act."

Re: [PATCH] Wipe out traditional opt from x86_64 Makefile

[H. Peter Anvin]

Glauber de Oliveira Costa wrote:
> Among other things, using -traditional as a gcc option stops us from
> using macro token pasting, which is a feature we heavily rely on.
> 
> There was still a use of -traditional in arch/x86/kernel/Makefile_64,
> which this patch removes.
> 
> I don't see any problems building kernels in my x86_64 box without
> -traditional.
> 
> Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
> ---
>  arch/x86/kernel/Makefile_64 |    1 -
>  1 files changed, 0 insertions(+), 1 deletions(-)
> 
> diff --git a/arch/x86/kernel/Makefile_64 b/arch/x86/kernel/Makefile_64
> index 82da906..03a01f6 100644
> --- a/arch/x86/kernel/Makefile_64
> +++ b/arch/x86/kernel/Makefile_64
> @@ -3,7 +3,6 @@
>  #
>  
>  extra-y 	:= head_64.o head64.o init_task.o vmlinux.lds
> -#EXTRA_AFLAGS	:= -traditional
>  obj-y	:= process_64.o signal_64.o entry_64.o traps_64.o irq_64.o \
>  		ptrace_64.o time_64.o ioport_64.o ldt_64.o setup_64.o i8259_64.o sys_x86_64.o \
>  		x8664_ksyms_64.o i387_64.o syscall_64.o vsyscall_64.o \

At least *at one point*, ld scripts were whitespace-sensitive in a way 
that called for using -traditional.  Doesn't mean that anything else 
needs it, though.

	-hpa

Re: [PATCH] Wipe out traditional opt from x86_64 Makefile

[Adrian Bunk]

On Thu, Oct 25, 2007 at 09:47:52AM -0300, Glauber de Oliveira Costa wrote:
> Among other things, using -traditional as a gcc option stops us from
> using macro token pasting, which is a feature we heavily rely on.
> 
> There was still a use of -traditional in arch/x86/kernel/Makefile_64,
> which this patch removes.
> 
> I don't see any problems building kernels in my x86_64 box without
> -traditional.
>...

Does it also work with gcc 3.2 ?

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: [PATCH] Wipe out traditional opt from x86_64 Makefile

[H. Peter Anvin]

Adrian Bunk wrote:
> On Thu, Oct 25, 2007 at 09:47:52AM -0300, Glauber de Oliveira Costa wrote:
>> Among other things, using -traditional as a gcc option stops us from
>> using macro token pasting, which is a feature we heavily rely on.
>>
>> There was still a use of -traditional in arch/x86/kernel/Makefile_64,
>> which this patch removes.
>>
>> I don't see any problems building kernels in my x86_64 box without
>> -traditional.
>> ...
> 
> Does it also work with gcc 3.2 ?
> 

It wouldn't be a gcc version issue, it would be a binutils (ld) version 
issue.

	-hpa

Re: [PATCH] Wipe out traditional opt from x86_64 Makefile

[Adrian Bunk]

On Thu, Oct 25, 2007 at 10:56:58AM -0700, H. Peter Anvin wrote:
> Adrian Bunk wrote:
>> On Thu, Oct 25, 2007 at 09:47:52AM -0300, Glauber de Oliveira Costa wrote:
>>> Among other things, using -traditional as a gcc option stops us from
>>> using macro token pasting, which is a feature we heavily rely on.
>>>
>>> There was still a use of -traditional in arch/x86/kernel/Makefile_64,
>>> which this patch removes.
>>>
>>> I don't see any problems building kernels in my x86_64 box without
>>> -traditional.
>>> ...
>> Does it also work with gcc 3.2 ?
>
> It wouldn't be a gcc version issue, it would be a binutils (ld) version 
> issue.

Sorry for the dumb question, but how does or did the gcc -traditional 
option affect ld?

> 	-hpa

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: [PATCH] Wipe out traditional opt from x86_64 Makefile

[Jeremy Fitzhardinge]

Adrian Bunk wrote:
> Sorry for the dumb question, but how does or did the gcc -traditional 
> option affect ld?
>   

vmlinux.lds is run through cpp (as a .S file).

    J

Re: [PATCH] Wipe out traditional opt from x86_64 Makefile

[Glauber de Oliveira Costa]

On 10/25/07, Jeremy Fitzhardinge <jeremy@goop.org> wrote:
> Adrian Bunk wrote:
> > Sorry for the dumb question, but how does or did the gcc -traditional
> > option affect ld?
> >
>
After a while, there is no issue I can find on x86_64 by taking out
the -traditional flag. Everything seems fine.

-- 
Glauber de Oliveira Costa.
"Free as in Freedom"
http://glommer.net

"The less confident you are, the more serious you have to act."