sched.c: why -fno-omit-frame-pointer?

sched.c: why -fno-omit-frame-pointer?

[Adrian Bunk]

kernel/sched.c gets compiled with -fno-omit-frame-pointer, and this was 
already done in kernel 1.0 (sic).

Later, it has been modified to be this way only on some architectures.

It might not be an earthshaking amount, but removing it saves some 
bytes, and there's no visible breakage when running the modified kernel 
on i386.

$ ls -la kernel/sched.o*
-rw-rw-r-- 1 bunk bunk 25600 2007-08-19 16:10 kernel/sched.o
-rw-rw-r-- 1 bunk bunk 26084 2007-08-19 16:03 kernel/sched.o.old
$ size kernel/sched.o*
   text    data     bss     dec     hex filename
  11835     184    2000   14019    36c3 kernel/sched.o
  12321     184    2000   14505    38a9 kernel/sched.o.old
$ 

Is the -fno-omit-frame-pointer still required for some reason, or is 
this a relict that could be removed?

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: sched.c: why -fno-omit-frame-pointer?

[Arjan van de Ven]

On Sun, 2007-08-19 at 16:17 +0200, Adrian Bunk wrote:
> kernel/sched.c gets compiled with -fno-omit-frame-pointer, and this was 
> already done in kernel 1.0 (sic).
> 
> Later, it has been modified to be this way only on some architectures.
> 
> It might not be an earthshaking amount, but removing it saves some 
> bytes, and there's no visible breakage when running the modified kernel 
> on i386.
> 

vague memory: this was needed for wchan to work properly

Re: sched.c: why -fno-omit-frame-pointer?

[Andi Kleen]

Adrian Bunk <bunk@kernel.org> writes:
> 
> Is the -fno-omit-frame-pointer still required for some reason, or is 
> this a relict that could be removed?

It's needed so that the WCHAN /proc display can backtrace one
level up out of schedule()

-Andi

Re: sched.c: why -fno-omit-frame-pointer?

[Ingo Molnar]

* Andi Kleen <andi@firstfloor.org> wrote:

> Adrian Bunk <bunk@kernel.org> writes:
> > 
> > Is the -fno-omit-frame-pointer still required for some reason, or is 
> > this a relict that could be removed?
> 
> It's needed so that the WCHAN /proc display can backtrace one level up 
> out of schedule()

it's not just one level - we'll back out until we are in scheduler 
functions (up to 16 levels). So it will back out of io_schedule() and 
other functions as well.

	Ingo