linux-kernel - Re: [PATCH v4 4/8] recordmcount: Rewrite error/success handling

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, 9 Oct 2019 12:46:26 +0200
From:   Uwe Kleine-König 
        <u.kleine-koenig@...gutronix.de>
To:     Matt Helsley <mhelsley@...are.com>
Cc:     LKML <linux-kernel@...r.kernel.org>,
        Ingo Molnar <mingo@...nel.org>,
        Steven Rostedt <rostedt@...dmis.org>, kernel@...gutronix.de,
        linux-arm-kernel@...ts.infradead.org
Subject: Re: [PATCH v4 4/8] recordmcount: Rewrite error/success handling

Hello,

On Wed, Jul 31, 2019 at 11:24:12AM -0700, Matt Helsley wrote:
> Recordmcount uses setjmp/longjmp to manage control flow as
> it reads and then writes the ELF file. This unusual control
> flow is hard to follow and check in addition to being unlike
> kernel coding style.
> 
> So we rewrite these paths to use regular return values to
> indicate error/success. When an error or previously-completed object
> file is found we return an error code following kernel
> coding conventions -- negative error values and 0 for success when
> we're not returning a pointer. We return NULL for those that fail
> and return non-NULL pointers otherwise.
> 
> One oddity is already_has_rel_mcount -- there we use pointer comparison
> rather than string comparison to differentiate between
> previously-processed object files and returning the name of a text
> section.
> 
> Signed-off-by: Matt Helsley <mhelsley@...are.com>

This is 3f1df12019f333442b12c3b5d110b8fc43eb0b36 in mainline now.

I have the following problem:

	$ make ARCH=arm arch/arm/crypto/
	  Using /home/uwe/gsrc/linux as source for kernel
	  GEN     Makefile
	  CALL    /home/uwe/gsrc/linux/scripts/checksyscalls.sh
	  CALL    /home/uwe/gsrc/linux/scripts/atomic/check-atomics.sh
	  CC      arch/arm/crypto/aes-cipher-glue.o
	arch/arm/crypto/aes-cipher-glue.o: failed
	make[2]: *** [/home/uwe/gsrc/linux/scripts/Makefile.build:281: arch/arm/crypto/aes-cipher-glue.o] Error 1
	make[2]: *** Deleting file 'arch/arm/crypto/aes-cipher-glue.o'
	make[1]: *** [/home/uwe/gsrc/linux/Makefile:1792: arch/arm/crypto/] Error 2
	make: *** [/home/uwe/gsrc/linux/Makefile:179: sub-make] Error 2

and bisection points to 3f1df12019f333442b12c3b5d110b8fc43eb0b36.

This doesn't affect all files, most compile just fine. After calling the
compiler by hand to get aes-cipher-glue.o back I have:

uwe@...rus:~/arietta/kbuild$ ./scripts/recordmcount "arch/arm/crypto/aes-cipher-glue.o"
arch/arm/crypto/aes-cipher-glue.o: failed

I didn't debug this further, if you have problems reproducing or need more
infos tell me. The defconfig I'm using is attached.

Best regards
Uwe

-- 
Pengutronix e.K.                           | Uwe Kleine-König            |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |

View attachment "arietta_defconfig" of type "text/plain" (4860 bytes)