linux-kernel - Re: [PATCH] genirq/devres: Simplify API devm_free

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: <d55840b8-4cc8-4bac-bb20-6aa155688fdb@icloud.com>
Date: Fri, 18 Oct 2024 19:08:29 +0800
From: Zijun Hu <zijun_hu@...oud.com>
To: Thomas Gleixner <tglx@...utronix.de>,
 quic_zijuhu <quic_zijuhu@...cinc.com>
Cc: linux-kernel@...r.kernel.org
Subject: Re: [PATCH] genirq/devres: Simplify API devm_free_irq()
 implementation

On 2024/10/18 18:58, Thomas Gleixner wrote:
> On Fri, Oct 18 2024 at 17:28, quic zijuhu wrote:
>> On 10/18/2024 4:57 PM, Thomas Gleixner wrote:
>> if irq to free was ever requested by devm_request_irq()
>> then both logic is exactly same.

[snip]

>> if (irq is not requested by devm_request_irq() {
>> 	warn;
>> 	return;
>> }
>>
> 
> Ah, you are right. I thought there is a return there.
> 
> So you want to explain it maybe this way:
> 
>    If devres_destroy() does not find a matching devres entry, then
>    devm_free_irq() emits a warning and tries to free the interrupt.
> 
>    That's wrong as devm_free_irq() should only undo what
>    devm_request_irq() set up.
> 
>    Replace devres_destroy() with a call to devres_release() which only
>    invokes the release function (free_irq()) in case that a matching
>    devres entry was found.
> 

thank you. good proposal, let me update commit base on it and send v2.

> Or something like that.
> 
> Thanks,
> 
>         tglx