linux-kernel - Re: [PATCH 3/3] thermal: exynos: Defer probe if vtmu is present but not registered

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: <56C69921.4050201@samsung.com>
Date:	Fri, 19 Feb 2016 13:25:05 +0900
From:	Krzysztof Kozlowski <k.kozlowski@...sung.com>
To:	Andi Shyti <andi.shyti@...sung.com>,
	Javier Martinez Canillas <javier@....samsung.com>
Cc:	linux-kernel@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
	Kukjin Kim <kgene@...nel.org>, linux-pm@...r.kernel.org,
	linux-samsung-soc@...r.kernel.org,
	Lukasz Majewski <l.majewski@...sung.com>,
	Zhang Rui <rui.zhang@...el.com>,
	Eduardo Valentin <edubezval@...il.com>
Subject: Re: [PATCH 3/3] thermal: exynos: Defer probe if vtmu is present but
 not registered

On 19.02.2016 13:20, Andi Shyti wrote:
> Hi Javier,
> 
> Just a question...
> 
>> diff --git a/drivers/thermal/samsung/exynos_tmu.c b/drivers/thermal/samsung/exynos_tmu.c
>> index f4f36bba7be9..f3ce94ec73b5 100644
>> --- a/drivers/thermal/samsung/exynos_tmu.c
>> +++ b/drivers/thermal/samsung/exynos_tmu.c
>> @@ -1318,6 +1318,8 @@ static int exynos_tmu_probe(struct platform_device *pdev)
>>  			return ret;
>>  		}
>>  	} else {
>> +		if (PTR_ERR(data->regulator) == -EPROBE_DEFER)
>> +			return -EPROBE_DEFER;
> 
> shouldn't we return also in case of -ENOMEM? -ENOMEM is a Kernel
> failure, not depending on the regulator itself.

Usually not because that would make this error path quite complicated
and difficult to keep consistent. If you choose this way then you will
have to probably extend the black list (e.g. EINVAL, some other)... or
use a white list.

Additionally the error codes can come from deeper layers. These layer
now can return ESOMETHING but later it might change to EDIFFERENT... you
cannot predict that.

Whatever the reason was (except defer), just ignore the regulator.

Best regards,
Krzysztof