linux-kernel - Re: [PATCH] ARM: 8351/1: perf: fix memory leak on return

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:	Mon, 18 May 2015 11:12:07 +0100
From:	Colin Ian King <colin.king@...onical.com>
To:	Ingo Molnar <mingo@...nel.org>
CC:	Will Deacon <will.deacon@....com>,
	Peter Zijlstra <a.p.zijlstra@...llo.nl>,
	Paul Mackerras <paulus@...ba.org>,
	Ingo Molnar <mingo@...hat.com>,
	Arnaldo Carvalho de Melo <acme@...nel.org>,
	Russell King <linux@....linux.org.uk>,
	linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] ARM: 8351/1: perf: fix memory leak on return

On 16/05/15 08:09, Ingo Molnar wrote:
> 
> * Colin King <colin.king@...onical.com> wrote:
> 
>> From: Colin Ian King <colin.king@...onical.com>
>>
>> Recent commit 3b8786ff7a1b31645ae2c26a2ec32dbd42ac1094
>> ("ARM: 8352/1: perf: Fix the pmu node name in warning message")
>> introduced a memory leak of irqs on the "Don't bother with PPIs"
>> return path. This was picked up by static analysis by cppcheck:
>>
>> [arch/arm/kernel/perf_event_cpu.c:315]: (error) Memory leak: irqs
>>
>> simpele fix is to free irqs when returning.
>>
>> Signed-off-by: Colin Ian King <colin.king@...onical.com>
>> ---
>>  arch/arm/kernel/perf_event_cpu.c | 4 +++-
>>  1 file changed, 3 insertions(+), 1 deletion(-)
>>
>> diff --git a/arch/arm/kernel/perf_event_cpu.c b/arch/arm/kernel/perf_event_cpu.c
>> index 213919b..9e5b2a5 100644
>> --- a/arch/arm/kernel/perf_event_cpu.c
>> +++ b/arch/arm/kernel/perf_event_cpu.c
>> @@ -311,8 +311,10 @@ static int of_pmu_irq_cfg(struct platform_device *pdev)
>>  
>>  	/* Don't bother with PPIs; they're already affine */
>>  	irq = platform_get_irq(pdev, 0);
>> -	if (irq >= 0 && irq_is_percpu(irq))
>> +	if (irq >= 0 && irq_is_percpu(irq)) {
>> +		kfree(irqs);
>>  		return 0;
>> +	}
>>  
>>  	for (i = 0; i < pdev->num_resources; ++i) {
>>  		struct device_node *dn;
> 
> So returning from the middle of a function isn't very clean.
> 
> Also, why do we return 0 in an error case?

I believe that's explained in commit
338d9dd3e2aee00a9198e8bf6e7d535d3feeaf32 ("ARM: 8351/1: perf: don't warn
about missing interrupt-affinity property for PPIs"):

"PPIs are affine by nature, so the interrupt-affinity property is not
 used and therefore we shouldn't print a warning in its absence."

> 
> Furthermore, this function already has a (partially hidden) error 
> cleanup path:
> 
>         if (i == pdev->num_resources)
>                 cpu_pmu->irq_affinity = irqs;
>         else
>                 kfree(irqs);
> 
> So this code should use proper goto driven cleanup. That's faster and 
> cleaner, and is less likely to result in bugs like the above.
> 
> Thanks,
> 
> 	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/