linux-kernel - Re: [PATCH printk 10/18] kgbd: Pretend that console list walk is safe

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:   Fri, 30 Sep 2022 19:27:04 +0200
From:   Petr Mladek <pmladek@...e.com>
To:     John Ogness <john.ogness@...utronix.de>
Cc:     Doug Anderson <dianders@...omium.org>,
        Sergey Senozhatsky <senozhatsky@...omium.org>,
        Steven Rostedt <rostedt@...dmis.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        LKML <linux-kernel@...r.kernel.org>,
        Jason Wessel <jason.wessel@...driver.com>,
        Daniel Thompson <daniel.thompson@...aro.org>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Jiri Slaby <jirislaby@...nel.org>,
        Aaron Tomlin <atomlin@...hat.com>,
        Luis Chamberlain <mcgrof@...nel.org>,
        kgdb-bugreport@...ts.sourceforge.net, linux-serial@...r.kernel.org
Subject: Re: [PATCH printk 10/18] kgbd: Pretend that console list walk is safe

On Fri 2022-09-30 15:50:56, John Ogness wrote:
> On 2022-09-30, Petr Mladek <pmladek@...e.com> wrote:
> > Anyway, what about using the later added SRCU walk here?
> > After all, this is exactly what RCU is for, isn't it?
> 
> So I think a lot of the problems with this series is that SRCU is
> introduced too late. We are debating things in patch 6 that are
> irrelevant by patch 12.

> I will rework the series so that the changes come in the following
> order:
> 
> 1. provide an atomic console_is_enabled()
>
> 2. convert the list to SRCU
> 
> 3. move all iterators from console_lock()/console_trylock() to SRCU
> 
> Step 3 may result in console_lock()/console_trylock() calls disappearing
> or relocating to where they are needed for non-list-synchronization
> purposes.

I agree that introding SRCU as early as possible would
help. The current patchset converts the same code several times...

Best Regards,
Petr