linux-kernel - Re: [patch] Add a simple backtrace test module

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]

Message-ID: <20080110091319.GB23878@elte.hu>
Date:	Thu, 10 Jan 2008 10:13:19 +0100
From:	Ingo Molnar <mingo@...e.hu>
To:	Arjan van de Ven <arjan@...radead.org>
Cc:	linux-kernel@...r.kernel.org, Thomas Gleixner <tglx@...utronix.de>,
	"H. Peter Anvin" <hpa@...or.com>
Subject: Re: [patch]  Add a simple backtrace test module


* Arjan van de Ven <arjan@...radead.org> wrote:

> During the work on the x86 32 and 64 bit backtrace code I found it 
> useful to have a simple test module to test a process and irq context 
> backtrace. Since the existing backtrace code was buggy, I figure it 
> might be useful to have such a test module in the kernel so that maybe 
> we can even detect such bugs earlier..

cool patch, applied!

a few suggestions:

a fundamental one: could you do a save_stack_trace() and check that both 
the process context and the irq context functions are present in that 
trace? If not then flag it as a regression and emit a real WARN_ON() 
warning.

i.e. use save_stack_trace() to do a "silent" test - instead of emitting 
backtraces during bootup. (which are marked via 'this is not a bug' but 
which are visually active nevertheless.)

the locking selftests use similar techniques to never emit real 
warnings, just a readable table of test results:

 | Locking API testsuite:
 ----------------------------------------------------------------------------
                                  | spin |wlock |rlock |mutex | wsem | rsem |
   --------------------------------------------------------------------------
                      A-A deadlock:  ok  |  ok  |  ok  |  ok  |  ok  |  ok  |
                  A-B-B-A deadlock:  ok  |  ok  |  ok  |  ok  |  ok  |  ok  |

internally, while the test is running, lockdep is triggered for real but 
the debug output and the backtraces are supressed.

and a few small details:

> +	printk("====[ backtrace testing ]===========\n");
> +	printk("Testing a backtrace from process context.\n");
> +	printk("The following trace is a kernel self test and not a bug!\n");

the printks need a KERN_ attribute.

> +	dump_stack();
> +
> +	init_timer(&backtrace_timer);
> +	backtrace_timer.function = backtrace_test_timer;
> +	mod_timer(&backtrace_timer, jiffies + 10);
> +
> +	msleep(10);
> +	printk("====[ end of backtrace testing ]====\n");

would be nice to have a testcase for the NMI watchdog and the softlockup 
watchdog as well: do they properly detect lockups on all CPUs?

> +static void exitf(void)

s/exitf/exit_backtrace_test

> +	  This option provides a kernel module that can be used to test
> +	  the kernel stack backtrace code. This option is not useful
> +	  for distributions or general kernels, but only for kernel
> +	  developers working on architecture code.

s/but only/only

	Ingo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/