linux-kernel - Re: [PATCH] usb: gadget: fsl: Fix a wrong judgment in fsl_udc

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:   Fri, 10 Apr 2020 16:05:06 +0800
From:   Tang Bin <tangbin@...s.chinamobile.com>
To:     Markus Elfring <Markus.Elfring@....de>
Cc:     Li Yang <leoyang.li@....com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Felipe Balbi <balbi@...nel.org>, linux-usb@...r.kernel.org,
        linuxppc-dev@...ts.ozlabs.org, linux-kernel@...r.kernel.org,
        kernel-janitors@...r.kernel.org
Subject: Re: [PATCH] usb: gadget: fsl: Fix a wrong judgment in fsl_udc_probe()

Hi Markus

On 2020/4/10 15:33, Markus Elfring wrote:
>> If the function "platform_get_irq()" failed, the negative value
>> returned will not be detected here, including "-EPROBE_DEFER",
> I suggest to adjust this change description.
>
> Wording alternative:
>    The negative return value (which could eventually be “-EPROBE_DEFER”)
>    will not be detected here from a failed call of the function “platform_get_irq”.
Hardware experiments show that the negative return value is not just 
"-EPROBE_DEFER".
>
>> which causes the application to fail to get the correct error message.
> Will another fine-tuning become relevant also for this wording?
Maybe that's not quite accurate.
>
>
>> Thus it must be fixed.
> Wording alternative:
>    Thus adjust the error detection and corresponding exception handling.
Got it.
>
>
>> Signed-off-by: Tang Bin <tangbin@...s.chinamobile.com>
>> Signed-off-by: Shengju Zhang <zhangshengju@...s.chinamobile.com>
> How do you think about to add the tags “Fixes”, “Link” and “Reported-by”?
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/process/submitting-patches.rst?id=c0cc271173b2e1c2d8d0ceaef14e4dfa79eefc0d#n584
>
> usb: gadget: fsl_udc_core: Checking for a failed platform_get_irq() call in fsl_udc_probe()
> https://lore.kernel.org/linux-usb/36341bb1-1e00-5eb1-d032-60dcc614ddaf@web.de/
> https://lkml.org/lkml/2020/4/8/442
>
> …
>> +++ b/drivers/usb/gadget/udc/fsl_udc_core.c
>> @@ -2441,8 +2441,8 @@ static int fsl_udc_probe(struct platform_device *pdev)
>>   	udc_controller->max_ep = (dccparams & DCCPARAMS_DEN_MASK) * 2;
>>
>>   	udc_controller->irq = platform_get_irq(pdev, 0);
>> -	if (!udc_controller->irq) {
>> -		ret = -ENODEV;
>> +	if (udc_controller->irq <= 0) {
> Will such a failure predicate need any more clarification?
>
> How does this check fit to the current software documentation?
Maybe my tags are not suitable.
>
>
>> +		ret = udc_controller->irq ? : -ENODEV;
> Will it be clearer to specify values for all cases in such a conditional operator
> (instead of leaving one case empty)?

I don't know what you mean of "instead of leaving one case empty". But 
by experiment, "ret = udc_controller->irq ? : -ENODEV" or "ret = 
udc_controller->irq < 0 ? udc_controller->irq : -ENODEV" should be 
suitable here.


Thank you for your guidance.

Tang Bin