linux-kernel - Re: [PATCH v2 3/3] i2c: xilinx: Use devm

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: <9b5a6672-2fd9-4900-b018-f044dee60646@VA3EHSMHS021.ehs.local>
Date:	Fri, 4 Oct 2013 14:04:06 +0200
From:	Michal Simek <michal.simek@...inx.com>
To:	Wolfram Sang <wsa@...-dreams.de>
CC:	Michal Simek <monstr@...str.eu>,
	Michal Simek <michal.simek@...inx.com>,
	<linux-kernel@...r.kernel.org>,
	Kedareswara rao Appana <appana.durga.rao@...inx.com>,
	Kedareswara rao Appana <appanad@...inx.com>,
	Jean Delvare <khali@...ux-fr.org>,
	Peter Korsgaard <jacmet@...site.dk>,
	<linux-i2c@...r.kernel.org>
Subject: Re: [PATCH v2 3/3] i2c: xilinx: Use devm_* functions

On 10/04/2013 01:58 PM, Wolfram Sang wrote:
> On Fri, Oct 04, 2013 at 11:16:20AM +0200, Michal Simek wrote:
>> On 10/04/2013 07:33 AM, Wolfram Sang wrote:
>>>
>>>> +	i2c->base = devm_ioremap_resource(&pdev->dev, res);
>>>> +	if (IS_ERR(i2c->base)) {
>>>> +		dev_err(&pdev->dev, "Could not allocate iomem\n");
>>>
>>> devm_ioremap_resource already prints error messages.
>>
>> you are right.
>>
>>>
>>>> +	ret = devm_request_irq(&pdev->dev, irq, xiic_isr, 0, pdev->name, i2c);
>>>
>>> This is too early. Can you find out why?
>>
>> Why do you think that it is too early?
> 
> The ISR uses spinlocks which are not initialized by then.

And waitqueue too.  Ok I will keep that request irq in current location
and I will add xiic_reinit() below this code.

>> I am looking at origin code again and I think that the code
>> is also problematic because in xiic_reinit() interrupts are enabled
>> but they are requested later.
>> Shouldn't be there a logic that interrupts should be enabled when
>> interrupts are registered by the kernel?
> 
> First register the handler, then activate interrupts. You are right,
> this needs to be fixed, too.

Do you want me to create separate patch just about moving request irq in front of
xiic_reinit()? And then devm_ conversion?

Thanks,
Michal


Download attachment "signature.asc" of type "application/pgp-signature" (265 bytes)