linux-kernel - Re: [PATCH 7/7] modpost: handle relocations mismatch in __ex

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: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Wed, 15 Apr 2015 12:57:37 +0930
From:	Rusty Russell <rusty@...tcorp.com.au>
To:	Quentin Casasnovas <quentin.casasnovas@...cle.com>,
	Thierry Reding <thierry.reding@...il.com>
Cc:	Quentin Casasnovas <quentin.casasnovas@...cle.com>,
	lkml <linux-kernel@...r.kernel.org>,
	Oleg Nesterov <oleg@...hat.com>,
	Borislav Petkov <bp@...en8.de>,
	Linus Torvalds <torvalds@...ux-foundation.org>
Subject: Re: [PATCH 7/7] modpost: handle relocations mismatch in __ex_table.

Quentin Casasnovas <quentin.casasnovas@...cle.com> writes:
> On Tue, Apr 14, 2015 at 02:14:14PM +0200, Thierry Reding wrote:
>> On Tue, Mar 17, 2015 at 01:40:02PM +0100, Quentin Casasnovas wrote:
>> > If one of these addresses point to a non-executable section, something is
>> > seriously wrong since it either means the kernel will never fault from
>> > there or it will not be able to jump to there.  As both cases are serious
>> > enough, we simply error out in these cases so the build fails and the
>> > developper has to fix the issue.
>> > 
>> > Signed-off-by: Quentin Casasnovas <quentin.casasnovas@...cle.com>
>> > ---
>> >  scripts/mod/modpost.c | 141 ++++++++++++++++++++++++++++++++++++++++++++++++++
>> >  1 file changed, 141 insertions(+)
>> 
>> This causes a bunch of mismatch warnings on 32-bit and 64-bit ARM
>> because there are two additional sections, .text.fixup and
>> .exception.text that store executable code. I've attached a patch
>> to fix those, but feel free to squash that into the original commit
>> if that's still possible.
>>
>
> Thanks Thierry!
>
> Your patch looks good to me, though I was wondering if we should just add
> .text.* in the TEXT_SECTIONS macro.  Some architectures define
> -ffunction-sections (parisc, score, metag and frv) so there are tons of
> useless warnings on these..  It also means the current modpost sanity
> checks don't run for those so it might even uncover some real mismatch ;)

Yes, but this adds ".exception.text" so a .text.* wildcard won't quite
cover it.

I've applied his patch, then the following:

modpost: handle -ffunction-sections

52dc0595d540 introduced OTHER_TEXT_SECTIONS for identifying what
sections could validly have __ex_table entries.  Unfortunately, it
wasn't tested with -ffunction-sections, which some architectures
use.

Reported-by: kbuild test robot <fengguang.wu@...el.com>
Cc: Quentin Casasnovas <quentin.casasnovas@...cle.com>
Signed-off-by: Rusty Russell <rusty@...tcorp.com.au>

diff --git a/scripts/mod/modpost.c b/scripts/mod/modpost.c
index cbd53e08769d..22dbc604cdb9 100644
--- a/scripts/mod/modpost.c
+++ b/scripts/mod/modpost.c
@@ -876,7 +876,7 @@ static void check_section(const char *modname, struct elf_info *elf,
 #define TEXT_SECTIONS ".text", ".text.unlikely", ".sched.text", \
 		".kprobes.text"
 #define OTHER_TEXT_SECTIONS ".ref.text", ".head.text", ".spinlock.text", \
-		".fixup", ".entry.text"
+		".fixup", ".entry.text", ".exception.text", ".text.*"
 
 #define INIT_SECTIONS      ".init.*"
 #define MEM_INIT_SECTIONS  ".meminit.*"
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/