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:   Fri, 8 Dec 2017 01:03:53 +0900
From:   Sergey Senozhatsky <sergey.senozhatsky@...il.com>
To:     Petr Mladek <pmladek@...e.com>
Cc:     Sergey Senozhatsky <sergey.senozhatsky@...il.com>,
        Steven Rostedt <rostedt@...dmis.org>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        Fengguang Wu <fengguang.wu@...el.com>,
        Kevin Hilman <khilman@...libre.com>,
        Mark Brown <broonie@...nel.org>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        LKML <linux-kernel@...r.kernel.org>,
        Sergey Senozhatsky <sergey.senozhatsky.work@...il.com>
Subject: Re: [RFC][PATCH] printk: add console_msg_format command line option

On (12/07/17 16:45), Petr Mladek wrote:
[.,]
> > hmm... `dmesg -S --raw' doesn't show anything on my system. `dmesg --raw'
> > matches the console_msg_format=syslog.
> 
> Huh, I am confused now. I thought that I saw the extended, /dev/kmsg - like,
> output with "dmesg --raw" but I do not see it now.
> 
> Anyway, I can see a difference related to
> /sys/module/printk/parameters/time. The command "dmesg --raw" always
> shows the timestamp help because it reads the messages via /dev/kmsg.
> But "dmesg -S --raw" shows the timestamp only when the "time"
> parameter is enabled:

aha, I see your point. -S forces dmesg to use syslog() syscall. so it
doesn't access the kmsg. otherwise it attempts to read from kmsg first.
you are right.

> $> echo 0 >/sys/module/printk/parameters/time
> $> dmesg -S --raw | head -n 3
> <5>Linux version 4.15.0-rc2-5-default+ (pmladek@...hway) (gcc version 4.8.5 (SUSE Linux)) #4180 SMP Thu Dec 7 16:06:42 CET 2017
> <6>Command line: BOOT_IMAGE=/boot/vmlinuz-4.15.0-rc2-5-default+ root=UUID=9ba368bb-c323-4131-ad94-c29cf3add76f resume=/dev/sda2 earlyprintk=ttyS0,115200 console=tty0 console=ttyS0,115200 console_msg_format=syslog ignore_loglevel nosplash showopts crashkernel=256M,high crashkernel=128M,low
> <6>x86/fpu: x87 FPU will use FXSAVE
> $> echo 1 >/sys/module/printk/parameters/time
> $> dmesg -S --raw | head -n 3
> <5>[    0.000000] Linux version 4.15.0-rc2-5-default+ (pmladek@...hway) (gcc version 4.8.5 (SUSE Linux)) #4180 SMP Thu Dec 7 16:06:42 CET 2017
> <6>[    0.000000] Command line: BOOT_IMAGE=/boot/vmlinuz-4.15.0-rc2-5-default+ root=UUID=9ba368bb-c323-4131-ad94-c29cf3add76f resume=/dev/sda2 earlyprintk=ttyS0,115200 console=tty0 console=ttyS0,115200 console_msg_format=syslog ignore_loglevel nosplash showopts crashkernel=256M,high crashkernel=128M,low
> <6>[    0.000000] x86/fpu: x87 FPU will use FXSAVE

I see.

> To be honest, I am somehow confused by the terminology.
> We should make it clear that the "syslog" console format
> is the same as the format used by the syslog syscall.

yeah. that was exactly my point. syslog format means that

 : we do what reading from /proc/kmsg does -- do_syslog(). which "is"
 : the same for some dmesg modes.

I thought that `dmesg --raw' always uses do_syslog(), apparently this
is not the case, and that "is" was not correct.

	-ss

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ