linux-kernel - Re: [PATCH] printk/nmi: Prevent deadlock when serializing NMI backtraces

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]

Message-ID: <20180620015834.GD650@jagdpanzerIV>
Date:   Wed, 20 Jun 2018 10:58:34 +0900
From:   Sergey Senozhatsky <sergey.senozhatsky.work@...il.com>
To:     Steven Rostedt <rostedt@...dmis.org>
Cc:     Sergey Senozhatsky <sergey.senozhatsky.work@...il.com>,
        Petr Mladek <pmladek@...e.com>,
        Sergey Senozhatsky <sergey.senozhatsky@...il.com>,
        Peter Zijlstra <peterz@...radead.org>,
        Tetsuo Handa <penguin-kernel@...ove.SAKURA.ne.jp>,
        linux-kernel@...r.kernel.org, "4 . 13+" <stable@...r.kernel.org>
Subject: Re: [PATCH] printk/nmi: Prevent deadlock when serializing NMI
 backtraces

On (06/19/18 09:23), Steven Rostedt wrote:
> > On (06/19/18 09:52), Petr Mladek wrote:
> > > On Mon 2018-06-18 19:07:18, Sergey Senozhatsky wrote:  
> > > > On (06/18/18 11:39), Petr Mladek wrote:
> > > > [..]
> > > > 
> > > > Hmm. Can't answer right now :)  
> > > 
> > > Please, let me know what name you would like ;-)  
> > 
> > :) Wow, it's hard. Maybe we can derive some bits from the ftrace_dump()
> > function name? Dunno... Does printk_dump_nmi_enter() sound terrible?
> > 
> > Maybe Steven has some opinions on this?
> > 
> 
> What exactly is the question?

Which one of these you'd prefer to see in ftrace_dump():

- printk_nmi_direct_enter() / printk_nmi_direct_exit()
- printk_chatty_nmi_enter() / printk_chatty_nmi_exit()
- printk_large_nmi_enter() / printk_large_nmi_exit()
- printk_dump_nmi_enter() / printk_dump_nmi_exit()

> Also, from a previous email in this thread, if you have to risk
> interleaved output to solve a deadlock, then just do that.

Yep.

	-ss