linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: James Hogan <james.hogan@imgtec.com>
To: Rusty Russell <rusty@ozlabs.org>
Cc: Michal Marek <mmarek@suse.cz>, Takashi Iwai <tiwai@suse.de>,
	David Howells <dhowells@redhat.com>,
	<linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 2/3] MODSIGN: Avoid using .incbin in C source
Date: Mon, 10 Dec 2012 10:12:49 +0000	[thread overview]
Message-ID: <50C5B5A1.3040801@imgtec.com> (raw)
In-Reply-To: <87zk1qeap3.fsf@rustcorp.com.au>

On 07/12/12 04:40, Rusty Russell wrote:
> Michal Marek <mmarek@suse.cz> writes:
>> It looks good, but looking at your patch, I just noticed that we have two
>> versions of the SYMBOL_PREFIX macro in the kernel now:
>>
>> scripts/Makefile.lib has had since some time
>>
>> ifdef CONFIG_SYMBOL_PREFIX
>> _sym_flags = -DSYMBOL_PREFIX=$(patsubst "%",%,$(CONFIG_SYMBOL_PREFIX))
>> _cpp_flags += $(_sym_flags)
>> _a_flags += $(_sym_flags)
>> endif
>>
>> while include/linux/kernel.h now has
>>
>> /* This helps us to avoid #ifdef CONFIG_SYMBOL_PREFIX */
>> #ifdef CONFIG_SYMBOL_PREFIX
>> #define SYMBOL_PREFIX CONFIG_SYMBOL_PREFIX
>> #else
>> #define SYMBOL_PREFIX ""
>> #endif
> 
> What a mess!  AFAICT there's one place which a non-string SYMBOL_PREFIX:
>         include/asm-generic/vmlinux.lds.h
> 
> So we do need it, and the only sane way seems to be via the cmdline and
> the shell usefully stripping quotes as we do.
> 
> So I just took Takashi's do-it-all-in-asm fix, and removed this
> re-definition (we have the same thing, BTW, called
> MODULE_SYMBOL_PREFIX).
> 
> I added a comment about the cmdline SYMBOL_PREFIX, too.
> 
> Thanks,
> Rusty.
> 
> From: Takashi Iwai <tiwai@suse.de>
> Subject: MODSIGN: Avoid using .incbin in C source
> 
> Using the asm .incbin statement in C sources breaks any gcc wrapper which
> assumes that preprocessed C source is self-contained. Use a separate .S
> file to include the siging key and certificate.
> 
> [ This means we no longer need SYMBOL_PREFIX which is defined in kernel.h
>   from cbdbf2abb7844548a7d7a6a2ae7af6b6fbcea401, so I removed it -- RR ]
> 
> Tested-by: Michal Marek <mmarek@suse.cz>
> Signed-off-by: Takashi Iwai <tiwai@suse.de>
> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>

Looks good to me.
Acked-by: James Hogan <james.hogan@imgtec.com>

Thanks
James

> diff --git a/include/linux/kernel.h b/include/linux/kernel.h
> index 7d8dfc7..a123b13 100644
> --- a/include/linux/kernel.h
> +++ b/include/linux/kernel.h
> @@ -701,13 +701,6 @@ static inline void ftrace_dump(enum ftrace_dump_mode oops_dump_mode) { }
>  #define COMPACTION_BUILD 0
>  #endif
>  
> -/* This helps us to avoid #ifdef CONFIG_SYMBOL_PREFIX */
> -#ifdef CONFIG_SYMBOL_PREFIX
> -#define SYMBOL_PREFIX CONFIG_SYMBOL_PREFIX
> -#else
> -#define SYMBOL_PREFIX ""
> -#endif
> -
>  /* Rebuild everything on CONFIG_FTRACE_MCOUNT_RECORD */
>  #ifdef CONFIG_FTRACE_MCOUNT_RECORD
>  # define REBUILD_DUE_TO_FTRACE_MCOUNT_RECORD
> diff --git a/kernel/Makefile b/kernel/Makefile
> index 86e3285..0bd9d43 100644
> --- a/kernel/Makefile
> +++ b/kernel/Makefile
> @@ -54,7 +54,7 @@ obj-$(CONFIG_DEBUG_SPINLOCK) += spinlock.o
>  obj-$(CONFIG_PROVE_LOCKING) += spinlock.o
>  obj-$(CONFIG_UID16) += uid16.o
>  obj-$(CONFIG_MODULES) += module.o
> -obj-$(CONFIG_MODULE_SIG) += module_signing.o modsign_pubkey.o
> +obj-$(CONFIG_MODULE_SIG) += module_signing.o modsign_pubkey.o modsign_certificate.o
>  obj-$(CONFIG_KALLSYMS) += kallsyms.o
>  obj-$(CONFIG_BSD_PROCESS_ACCT) += acct.o
>  obj-$(CONFIG_KEXEC) += kexec.o
> @@ -139,7 +139,7 @@ ifeq ($(CONFIG_MODULE_SIG),y)
>  extra_certificates:
>  	touch $@
>  
> -kernel/modsign_pubkey.o: signing_key.x509 extra_certificates
> +kernel/modsign_certificate.o: signing_key.x509 extra_certificates
>  
>  ###############################################################################
>  #
> diff --git a/kernel/modsign_certificate.S b/kernel/modsign_certificate.S
> new file mode 100644
> index 0000000..246b4c6
> --- /dev/null
> +++ b/kernel/modsign_certificate.S
> @@ -0,0 +1,19 @@
> +/* SYMBOL_PREFIX defined on commandline from CONFIG_SYMBOL_PREFIX */
> +#ifndef SYMBOL_PREFIX
> +#define ASM_SYMBOL(sym) sym
> +#else
> +#define PASTE2(x,y) x##y
> +#define PASTE(x,y) PASTE2(x,y)
> +#define ASM_SYMBOL(sym) PASTE(SYMBOL_PREFIX, sym)
> +#endif
> +
> +#define GLOBAL(name)	\
> +	.globl ASM_SYMBOL(name);	\
> +	ASM_SYMBOL(name):
> +
> +	.section ".init.data","aw"
> +
> +GLOBAL(modsign_certificate_list)
> +	.incbin "signing_key.x509"
> +	.incbin "extra_certificates"
> +GLOBAL(modsign_certificate_list_end)
> diff --git a/kernel/modsign_pubkey.c b/kernel/modsign_pubkey.c
> index 767e559..045504f 100644
> --- a/kernel/modsign_pubkey.c
> +++ b/kernel/modsign_pubkey.c
> @@ -20,12 +20,6 @@ struct key *modsign_keyring;
>  
>  extern __initdata const u8 modsign_certificate_list[];
>  extern __initdata const u8 modsign_certificate_list_end[];
> -asm(".section .init.data,\"aw\"\n"
> -    SYMBOL_PREFIX "modsign_certificate_list:\n"
> -    ".incbin \"signing_key.x509\"\n"
> -    ".incbin \"extra_certificates\"\n"
> -    SYMBOL_PREFIX "modsign_certificate_list_end:"
> -    );
>  
>  /*
>   * We need to make sure ccache doesn't cache the .o file as it doesn't notice
> 


  reply	other threads:[~2012-12-10 10:13 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-12-04 10:18 [PATCH 1/3] MODSIGN: Fix comparison erros in scripts/sign-file Michal Marek
2012-12-04 10:18 ` [PATCH 2/3] MODSIGN: Avoid using .incbin in C source Michal Marek
2012-12-04 10:18 ` [PATCH 3/3] MODSIGN: Drop ccache hack Michal Marek
2012-12-04 18:15 ` [PATCH 1/3] MODSIGN: Fix comparison erros in scripts/sign-file David Howells
2012-12-04 18:17 ` [PATCH 2/3] MODSIGN: Avoid using .incbin in C source David Howells
2012-12-04 23:58   ` Rusty Russell
2012-12-05  7:39     ` Takashi Iwai
2012-12-05  9:50       ` Michal Marek
2012-12-05 10:06         ` Takashi Iwai
2012-12-05 10:30         ` James Hogan
2012-12-05 11:05           ` Michal Marek
2012-12-05 11:16             ` James Hogan
2012-12-07  4:40         ` Rusty Russell
2012-12-10 10:12           ` James Hogan [this message]
2012-12-05 10:30     ` David Howells
2012-12-05 10:54       ` Michal Marek
2012-12-05 12:35     ` David Howells
2012-12-05  7:46   ` Takashi Iwai
2012-12-04 18:29 ` [PATCH 3/3] MODSIGN: Drop ccache hack David Howells
2012-12-04 23:44 ` [PATCH 1/3] MODSIGN: Fix comparison erros in scripts/sign-file Rusty Russell

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=50C5B5A1.3040801@imgtec.com \
    --to=james.hogan@imgtec.com \
    --cc=dhowells@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mmarek@suse.cz \
    --cc=rusty@ozlabs.org \
    --cc=tiwai@suse.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).