linux-kernel - Re: [PATCH v3 2/4] perf tests: make objdump disassemble zero blocks

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]

Date:	Thu, 03 Sep 2015 18:19:04 +0200
From:	Jan Stancek <jstancek@...hat.com>
To:	Arnaldo Carvalho de Melo <acme@...nel.org>,
	Adrian Hunter <adrian.hunter@...el.com>
CC:	linux-kernel@...r.kernel.org, jolsa@...nel.org, dsahern@...il.com,
	cjashfor@...ux.vnet.ibm.com, fweisbec@...il.com, mingo@...nel.org,
	namhyung@...nel.org, paulus@...ba.org, a.p.zijlstra@...llo.nl
Subject: Re: [PATCH v3 2/4] perf tests: make objdump disassemble zero blocks

On 09/03/2015 05:14 PM, Arnaldo Carvalho de Melo wrote:
> Em Thu, Sep 03, 2015 at 02:35:55PM +0300, Adrian Hunter escreveu:
>> On 03/09/15 14:23, Jan Stancek wrote:
>>> Add -z parameter to avoid skipping zero blocks:
>>>
>>>  ffffffff816704fe <sysret_check+0x4b>:
>>>  ffffffff816704fe:  7b 34         jnp ffffffff81670534 <sysret_signal+0x1c>
>>>        ...
>>>  ffffffff81670501 <sysret_careful>:
>>>  ffffffff81670501:  0f ba e2 03   bt  $0x3,%edx
>>>  ffffffff81670505:  73 11         jae ffffffff81670518 <sysret_signal>
>>
>> Acked-by: Adrian Hunter <adrian.hunter@...el.com>
> 
> Ok, I am applying this, but it would be nice to know in which systems,
> with which objdump/binutils versions which 'perf test' entry fails, with
> the output of such failure.
> 
> Jan, can you please provide this info?

Since my original report last year [1], I've seen it fail many times on
various HW (we run perf tests daily on RHEL). I haven't noticed any pattern
regarding HW it fails on.

Binutils versions go from binutils-2.23.52 to current latest from git.

It was always about "object code reading", from times it was perf test 21,
until now, when it's perf test 23.

I'm attaching examples of test output along with objdump output, that I
just ran on a random system using kernel 4.2.0.

HTH, regards,
Jan

[1] https://lkml.org/lkml/2014/12/11/222


View attachment "perf_failures_examples_with_4.2.0.txt" of type "text/plain" (8677 bytes)