linux-kernel - Re: [PATCH] scripts/gdb: fix lx-dmesg when CONFIG_PRINTK

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:   Thu, 10 Oct 2019 19:36:55 +0000
From:   Leonard Crestez <leonard.crestez@....com>
To:     Jan Kiszka <jan.kiszka@...mens.com>,
        Joel Colledge <joel.colledge@...bit.com>
CC:     Kieran Bingham <kbingham@...nel.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] scripts/gdb: fix lx-dmesg when CONFIG_PRINTK_CALLER is
 set

On 10.10.2019 18:14, Jan Kiszka wrote:
> On 25.09.19 17:03, Joel Colledge wrote:
>> When CONFIG_PRINTK_CALLER is set, struct printk_log contains an
>> additional member caller_id. As a result, the offset of the log text is
>> different.
>>
>> This fixes the following error:
>>
>>    (gdb) lx-dmesg
>>    Python Exception <class 'ValueError'> embedded null character:
>>    Error occurred in Python command: embedded null character
>>
>> Signed-off-by: Joel Colledge <joel.colledge@...bit.com>
>> ---
>>   scripts/gdb/linux/constants.py.in | 1 +
>>   scripts/gdb/linux/dmesg.py        | 4 +++-
>>   2 files changed, 4 insertions(+), 1 deletion(-)
>>
>> diff --git a/scripts/gdb/linux/constants.py.in b/scripts/gdb/linux/constants.py.in
>> index 2efbec6b6b8d..3c9794a0bf55 100644
>> --- a/scripts/gdb/linux/constants.py.in
>> +++ b/scripts/gdb/linux/constants.py.in
>> @@ -74,4 +74,5 @@ LX_CONFIG(CONFIG_GENERIC_CLOCKEVENTS_BROADCAST)
>>   LX_CONFIG(CONFIG_HIGH_RES_TIMERS)
>>   LX_CONFIG(CONFIG_NR_CPUS)
>>   LX_CONFIG(CONFIG_OF)
>> +LX_CONFIG(CONFIG_PRINTK_CALLER)
>>   LX_CONFIG(CONFIG_TICK_ONESHOT)
>> diff --git a/scripts/gdb/linux/dmesg.py b/scripts/gdb/linux/dmesg.py
>> index 6d2e09a2ad2f..1352680ef731 100644
>> --- a/scripts/gdb/linux/dmesg.py
>> +++ b/scripts/gdb/linux/dmesg.py
>> @@ -14,6 +14,7 @@
>>   import gdb
>>   import sys
>>   
>> +from linux import constants
>>   from linux import utils
>>   
>>   
>> @@ -53,7 +54,8 @@ class LxDmesg(gdb.Command):
>>                   continue
>>   
>>               text_len = utils.read_u16(log_buf[pos + 10:pos + 12])
>> -            text = log_buf[pos + 16:pos + 16 + text_len].decode(
>> +            text_start = pos + (20 if constants.LX_CONFIG_PRINTK_CALLER else 16)
>> +            text = log_buf[text_start:text_start + text_len].decode(
>>                   encoding='utf8', errors='replace')
>>               time_stamp = utils.read_u64(log_buf[pos:pos + 8])
> 
> Sorry for the delay:
> 
> Reviewed-by: Jan Kiszka <jan.kiszka@...mens.com>
> 
> I suspect we will see more in nearer future with upcoming printk rework...

The patch looks correct but I'm curious: is there a reason this code 
doesn't use struct printk_log?

GDB already knows about struct offsets so there should be no need to 
handle ifdefs with arithmetic.

Is it realistic to use gdb without struct layout info?

--
Regards,
Leonard