lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAK7LNARPq7Bj+zp7bYyH_uPOyJ9a1-hD-aSq6GDLb8YdqbA-+A@mail.gmail.com>
Date:   Fri, 1 Feb 2019 19:09:36 +0900
From:   Masahiro Yamada <yamada.masahiro@...ionext.com>
To:     Borislav Petkov <bp@...en8.de>
Cc:     Linux Kbuild mailing list <linux-kbuild@...r.kernel.org>,
        LKML <linux-kernel@...r.kernel.org>, X86 ML <x86@...nel.org>,
        Michal Marek <michal.lkml@...kovi.net>
Subject: Re: [PATCH] kbuild: Disable extra debugging info in .s output

On Fri, Feb 1, 2019 at 6:50 PM Borislav Petkov <bp@...en8.de> wrote:
>
> On Fri, Feb 01, 2019 at 12:06:13PM +0900, Masahiro Yamada wrote:
> > This would make the build speed slower
> > because the compiler is invoked to check those flags in every directory.
> >
> > I tested this patch with x86_64_defconfig and GCC 8.2
> > (on ubuntu 18.10 in docker)
> > but I saw no effect of this patch.
>
> Yes, because that's the .s target which is meant to be used to convert
> only a *single* .c file to assembly in order to stare at the resulting
> asm. For example:
>
> make arch/x86/kernel/cpu/common.s


I do know this, and did compare the output
from the *single* target "make arch/x86/kernel/cpu/common.s"



> It is a debugging aid - and a very basic and important one - and is not
> usually used during a normal kernel build. AFAIK.
>
> So we shouldn't worry about any slowdowns here.


Your patch does slowdown the build system extremely.


I provide the result of "perf stat" of the incremental build for you.



Without your patch:


 Performance counter stats for 'make -j8' (20 runs):

       2.721312445 seconds time elapsed
          ( +-  1.02% )



With you patch:


 Performance counter stats for 'make -j8' (20 runs):

       5.038521415 seconds time elapsed
          ( +-  0.56% )




Why should the normal build affected by the debugging aid?




One more thing, you did not answer my question.

If you are complaining about the DWARF info
enabled by CONFIG_DEBUG_INFO,
I recommend to turn off CONFIG_DEBUG_INFO.



> Thx.
>
> --
> Regards/Gruss,
>     Boris.
>
> Good mailing practices for 400: avoid top-posting and trim the reply.

--
Best Regards
Masahiro Yamada

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ