linux-kernel - Re: [PATCH] Revert "f2fs: use printk_ratelimited for f2fs

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] [day] [month] [year] [list]

Date:   Wed, 15 Aug 2018 11:42:27 +0800
From:   Chao Yu <yuchao0@...wei.com>
To:     Jaegeuk Kim <jaegeuk@...nel.org>
CC:     <linux-f2fs-devel@...ts.sourceforge.net>,
        <linux-kernel@...r.kernel.org>, <chao@...nel.org>
Subject: Re: [PATCH] Revert "f2fs: use printk_ratelimited for f2fs_msg"

On 2018/8/15 11:25, Jaegeuk Kim wrote:
> On 08/15, Chao Yu wrote:
>> On 2018/8/15 11:01, Jaegeuk Kim wrote:
>>> On 08/15, Chao Yu wrote:
>>>> On 2018/8/15 1:09, Jaegeuk Kim wrote:
>>>>> On 08/13, Chao Yu wrote:
>>>>>> Don't limit printing log, so that we will not miss any key messages.
>>>>>
>>>>> For example, this can avoid lots of messages during roll-forward recovery.
>>>>
>>>> That's really important evidence that can prove filesystem didn't lose last
>>>> fsynced data, when file/data is missing after SPO, otherwise, filesystem will
>>>> still be treated as suspect... :(
>>>
>>> It's too bad that we need to rely on kernel messages for that purpose. Anyway,
>>
>> Yes, it's very hard to make sure the problem is not caused by filesystem's bug
>> when trouble shooting, unless there is explicit kernel message.
>>
>> Also, without message, trouble shooting becomes harder.
>>
>>> if so, we may need to use printk() for essential messages only.
>>
>> Now, during our debugging, we have to change to use printk to avoid potential
>> message missing...
> 
> Let me understand what messages are really important for your debugging purpose.

The most important message is about filesystem/data consistence.

> I believe it'd not require all of them.

Yeah, IMO, error injection prints too many messages.

Thanks,

> 
>>
>> Thanks,
>>
>>>
>>>>
>>>> Thanks,
>>>>
>>>>>
>>>>>>
>>>>>> This reverts commit a36c106dffb616250117efb1cab271c19a8f94ff.
>>>>>>
>>>>>> Signed-off-by: Chao Yu <yuchao0@...wei.com>
>>>>>> ---
>>>>>>  fs/f2fs/super.c | 2 +-
>>>>>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>>>>>
>>>>>> diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
>>>>>> index bf4c919fe418..3d89d94191e7 100644
>>>>>> --- a/fs/f2fs/super.c
>>>>>> +++ b/fs/f2fs/super.c
>>>>>> @@ -209,7 +209,7 @@ void f2fs_msg(struct super_block *sb, const char *level, const char *fmt, ...)
>>>>>>  	va_start(args, fmt);
>>>>>>  	vaf.fmt = fmt;
>>>>>>  	vaf.va = &args;
>>>>>> -	printk_ratelimited("%sF2FS-fs (%s): %pV\n", level, sb->s_id, &vaf);
>>>>>> +	printk("%sF2FS-fs (%s): %pV\n", level, sb->s_id, &vaf);
>>>>>>  	va_end(args);
>>>>>>  }
>>>>>>  
>>>>>> -- 
>>>>>> 2.18.0.rc1
>>>>>
>>>>> .
>>>>>
>>>
>>> .
>>>
> 
> .
>