Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
References: <202004020117.6E434C035@keescook> <CAK7LNAQGTAgtADfY4H-k8X1J9nTMeOWvo8ZFfrUSHQUbhgcLKw@mail.gmail.com>
In-Reply-To: <CAK7LNAQGTAgtADfY4H-k8X1J9nTMeOWvo8ZFfrUSHQUbhgcLKw@mail.gmail.com>
From:   Nick Desaulniers <ndesaulniers@google.com>
Date:   Thu, 2 Apr 2020 11:22:02 -0700
Message-ID: <CAKwvOdnO_-jiQzCmy7pQNMGmZsgD9J=PYXVzOQ1NF1+u44MsOA@mail.gmail.com>
Subject: Re: [PATCH] kbuild: mkcompile_h: Include $LD version in /proc/version
To:     Masahiro Yamada <masahiroy@kernel.org>
Cc:     Kees Cook <keescook@chromium.org>,
        Michal Marek <michal.lkml@markovi.net>,
        Thomas Gleixner <tglx@linutronix.de>,
        Linux Kbuild mailing list <linux-kbuild@vger.kernel.org>,
        clang-built-linux <clang-built-linux@googlegroups.com>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Sandeep Patil <sspatil@google.com>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Thu, Apr 2, 2020 at 10:02 AM Masahiro Yamada <masahiroy@kernel.org> wrote:
>
> Hi Kees,
>
> On Thu, Apr 2, 2020 at 5:18 PM Kees Cook <keescook@chromium.org> wrote:
> >
> > When doing Clang builds of the kernel, it is possible to link with
> > either ld.bfd (binutils) or ld.lld (LLVM), but it is not possible to
> > discover this from a running kernel. Add the "$LD -v" output to
> > /proc/version.
> >
> > Signed-off-by: Kees Cook <keescook@chromium.org>
> > ---
> >  init/Makefile       | 2 +-
> >  scripts/mkcompile_h | 8 ++++++--
> >  2 files changed, 7 insertions(+), 3 deletions(-)
> >
> > diff --git a/init/Makefile b/init/Makefile
> > index 6246a06364d0..82c15bdb42d7 100644
> > --- a/init/Makefile
> > +++ b/init/Makefile
> > @@ -35,4 +35,4 @@ include/generated/compile.h: FORCE
> >         @$($(quiet)chk_compile.h)
> >         $(Q)$(CONFIG_SHELL) $(srctree)/scripts/mkcompile_h $@   \
> >         "$(UTS_MACHINE)" "$(CONFIG_SMP)" "$(CONFIG_PREEMPT)"    \
> > -       "$(CONFIG_PREEMPT_RT)" "$(CC) $(KBUILD_CFLAGS)"
> > +       "$(CONFIG_PREEMPT_RT)" "$(LD)" "$(CC) $(KBUILD_CFLAGS)"
> > diff --git a/scripts/mkcompile_h b/scripts/mkcompile_h
> > index 3a5a4b210c86..f98c07709370 100755
> > --- a/scripts/mkcompile_h
> > +++ b/scripts/mkcompile_h
> > @@ -6,7 +6,8 @@ ARCH=$2
> >  SMP=$3
> >  PREEMPT=$4
> >  PREEMPT_RT=$5
> > -CC=$6
> > +LD=$6
> > +CC=$7
>
>
> Just a nit.
>
> If you just append 'LD',
> you do not need to touch the 'CC=$6' line.
>
>
> If you do not mind, I will fold the following
> on top of your patch.

With Kees patch + Masahiro's diff applied:
$ cat /proc/version
Linux version 5.6.0-rc7-next-20200330+ (ndesaulniers@<hostname
scrubbed>) (clang version 11.0.0 (git@github.com:llvm/llvm-project.git
6d71daed03ced011cd25f4e4f4122a368fdc757d), LLD 11.0.0
(git@github.com:llvm/llvm-project.git
6d71daed03ced011cd25f4e4f4122a368fdc757d)) #167 SMP Thu Apr 2 11:17:36
PDT 2020

Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Tested-by: Nick Desaulniers <ndesaulniers@google.com>

We could actually use this in Android for a VTS test I've, uh, been
meaning to write.  Also, LOL at scrubbing the "compatible with" part,
reminds me of user agent strings in the browser, which are god awful.


> --- a/init/Makefile
> +++ b/init/Makefile
> @@ -35,4 +35,4 @@ include/generated/compile.h: FORCE
>         @$($(quiet)chk_compile.h)
>         $(Q)$(CONFIG_SHELL) $(srctree)/scripts/mkcompile_h $@   \
>         "$(UTS_MACHINE)" "$(CONFIG_SMP)" "$(CONFIG_PREEMPT)"    \
> -       "$(CONFIG_PREEMPT_RT)" "$(LD)" "$(CC) $(KBUILD_CFLAGS)"
> +       "$(CONFIG_PREEMPT_RT)" "$(CC) $(KBUILD_CFLAGS)" "$(LD)"
> diff --git a/scripts/mkcompile_h b/scripts/mkcompile_h
> index 8b38a96163e2..5b80a4699740 100755
> --- a/scripts/mkcompile_h
> +++ b/scripts/mkcompile_h
> @@ -6,8 +6,8 @@ ARCH=$2
>  SMP=$3
>  PREEMPT=$4
>  PREEMPT_RT=$5
> -LD=$6
> -CC=$7
> +CC=$6
> +LD=$7


-- 
Thanks,
~Nick Desaulniers