linux-kernel - Re: [RFC PATCH 2/7] printk: Simple implementation for NMI backtracing

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: <550B19EA.6000202@linaro.org>
Date:	Thu, 19 Mar 2015 18:48:10 +0000
From:	Daniel Thompson <daniel.thompson@...aro.org>
To:	Peter Zijlstra <peterz@...radead.org>,
	Steven Rostedt <rostedt@...dmis.org>
CC:	linux-arm-kernel@...ts.infradead.org,
	Catalin Marinas <catalin.marinas@....com>,
	Will Deacon <will.deacon@....com>,
	linux-kernel@...r.kernel.org, patches@...aro.org,
	linaro-kernel@...ts.linaro.org,
	John Stultz <john.stultz@...aro.org>,
	Sumit Semwal <sumit.semwal@...aro.org>,
	Marc Zyngier <marc.zyngier@....com>,
	Andrew Thoelke <andrew.thoelke@....com>,
	Dongdong Deng <dongdong.deng@...driver.com>
Subject: Re: [RFC PATCH 2/7] printk: Simple implementation for NMI backtracing

On 19/03/15 18:30, Peter Zijlstra wrote:
> On Thu, Mar 19, 2015 at 01:39:58PM -0400, Steven Rostedt wrote:
>>> +void printk_nmi_backtrace_complete(void)
>>> +{
>>> +	struct nmi_seq_buf *s;
>>> +	int len, cpu, i, last_i;
>>> +
>>> +	/*
>>> +	 * Now that all the NMIs have triggered, we can dump out their
>>> +	 * back traces safely to the console.
>>> +	 */
>>> +	for_each_possible_cpu(cpu) {
>>> +		s = &per_cpu(nmi_print_seq, cpu);
>>> +		last_i = 0;
>>> +
>>> +		len = seq_buf_used(&s->seq);
>>> +		if (!len)
>>> +			continue;
>>> +
>>> +		/* Print line by line. */
>>> +		for (i = 0; i < len; i++) {
>>> +			if (s->buffer[i] == '\n') {
>>> +				print_seq_line(s, last_i, i);
>>> +				last_i = i + 1;
>>> +			}
>>> +		}
>>> +		/* Check if there was a partial line. */
>>> +		if (last_i < len) {
>>> +			print_seq_line(s, last_i, len - 1);
>>> +			pr_cont("\n");
>>> +		}
>>> +
>>> +		/* Wipe out the buffer ready for the next time around. */
>>> +		seq_buf_clear(&s->seq);
>>> +	}
>>> +
>>> +	clear_bit(0, &nmi_print_flag);
>>> +	smp_mb__after_atomic();
>>
>> Is this really necessary. What is the mb synchronizing?
>>
>> [ Added Peter Zijlstra to confirm it's not needed ]
>
> It surely looks suspect; and it lacks a comment, which is a clear sign
> its buggy.
>
> Now it if tries to order the accesses to the seqbuf againt the clearing
> of the bit one would have expected a _before_ barrier, not an _after_.

It's nothing to do with the seqbuf since I added the seqbuf code myself 
but the barrier was already in the code that I copied from.

In the mainline code today it looks like this as part of the x86 code 
(note that call to put_cpu() in my patchset but it lives in the arch/ 
specific code rather than the generic code):

:                 /* Check if there was a partial line. */
:                 if (last_i < len) {
:                         print_seq_line(s, last_i, len - 1);
:                         pr_cont("\n");
:                 }
:         }
:
:         clear_bit(0, &backtrace_flag);
:         smp_mb__after_atomic();
:         put_cpu();
: }

The barrier was not intended to have anything to do with put_cpu() 
either though since the barrier was added before put_cpu() arrived:
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=554ec063982752e9a569ab9189eeffa3d96731b2

There's nothing in the commit comment explaining the barrier and I 
really can't see what it is for.


Daniel.
--
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/