linux-kernel - Re: [PATCH v2 0/2] printk: Reduce unnecessary wait times by pr

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, 11 Oct 2023 14:10:39 +0200
From:   Petr Mladek <pmladek@...e.com>
To:     John Ogness <john.ogness@...utronix.de>
Cc:     Sergey Senozhatsky <senozhatsky@...omium.org>,
        Steven Rostedt <rostedt@...dmis.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Todd Brandt <todd.e.brandt@...el.com>,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2 0/2] printk: Reduce unnecessary wait times by
 pr_flush()

On Tue 2023-10-10 13:44:38, John Ogness wrote:
> On 2023-10-10, Petr Mladek <pmladek@...e.com> wrote:
> >> This is an updated version of the patch
> >> https://lore.kernel.org/r/20230929113233.863824-1-john.ogness@linutronix.de
> >> 
> >> Changes since v1:
> >> 
> >>   + Split reduction of the pooling time into separate patch.
> >>   + Use jiffies to count the real sleeping time.
> >> 
> >> Note that my original POC counted jiffies spent in the entire cycle.
> >> It would count also the time spent by console_lock() and by flushing
> >> the legacy consoles in console_unlock().
> >> 
> >> The current version counts only the time spent by msleep(1) aka
> >> the pooling interval.
> >> 
> >> 
> >> John Ogness (1):
> >>   printk: flush consoles before checking progress
> >> 
> >> Petr Mladek (1):
> >>   printk: Reduce pr_flush() pooling time
> >
> > JFYI, the patchset has been committed into printk/linux.git,
> > branch rework/misc-cleanups.
> >
> > I am going to push it in the next merge window for 6.7.
> > IMHO, the issue is not critical and we are already at 6.6-rc5.
> 
> It is a real regression. I think it is important that it goes into
> 6.6. At least the first patch is simple enough and fixes the regression.

Fair enough. I have crated pull request with the 1st patch for
6.6-rc6, see ZSaPGo13Fz9TLUc6@...ey.

Note that I rebased the branch rework/misc-cleanups in
printk/linux.git to be able to crate a merge with the 1st patch only.
I am not sure if it could have been done without the rebase.

Best Regards,
Petr