Re: [PATCH 04/19] v4l: omap3isp: fix handling platform_get_irq result

From: Andrzej Hajda
Date: Tue Nov 10 2015 - 05:00:13 EST


On 11/10/2015 09:53 AM, Laurent Pinchart wrote:
> Hi Andrzej,
>
> On Tuesday 10 November 2015 07:48:54 Andrzej Hajda wrote:
>> On 11/09/2015 09:16 PM, Laurent Pinchart wrote:
>>> On Thursday 24 September 2015 16:00:12 Andrzej Hajda wrote:
>>>> The function can return negative value.
>>>>
>>>> The problem has been detected using proposed semantic patch
>>>> scripts/coccinelle/tests/assign_signed_to_unsigned.cocci [1].
>>>>
>>>> [1]: http://permalink.gmane.org/gmane.linux.kernel/2046107
>>>>
>>>> Signed-off-by: Andrzej Hajda <a.hajda@xxxxxxxxxxx>
>>>> ---
>>>> Hi,
>>>>
>>>> To avoid problems with too many mail recipients I have sent whole
>>>> patchset only to LKML. Anyway patches have no dependencies.
>>>>
>>>> drivers/media/platform/omap3isp/isp.c | 5 +++--
>>>> 1 file changed, 3 insertions(+), 2 deletions(-)
>>>>
>>>> diff --git a/drivers/media/platform/omap3isp/isp.c
>>>> b/drivers/media/platform/omap3isp/isp.c index 56e683b..df9d2c2 100644
>>>> --- a/drivers/media/platform/omap3isp/isp.c
>>>> +++ b/drivers/media/platform/omap3isp/isp.c
>>>> @@ -2442,12 +2442,13 @@ static int isp_probe(struct platform_device
>>>> *pdev)
>>>> }
>>>>
>>>> /* Interrupt */
>>>> - isp->irq_num = platform_get_irq(pdev, 0);
>>>> - if (isp->irq_num <= 0) {
>>>> + ret = platform_get_irq(pdev, 0);
>>>> + if (ret <= 0) {
>>> Looking at platform_get_irq() all error values are negative. You could
>>> just test for ret < 0 here, and remove the ret = -ENODEV assignment below
>>> to keep the error code returned by platform_get_irq().
>>>
>>> If you're fine with that modification there's no need to resubmit, just
>>> let me know and I'll fix it when applying it to my tree.
>> I left it as before, as it was not related to the patch. Additionally I have
>> lurked little bit inside platform_get_irq and it looks little bit scary to
>> me: platform_get_irq returns value of of_irq_get if ret >= 0,
>> of_irq_get calls of_irq_parse_one which can return 0,
>> in such case irq_create_of_mapping value is returned which is unsigned int
>> and evaluates to 0 in case of failures.
>> I am not sure if above scenario can ever occur, but the code looks so messy
>> to me, that I gave up :)
>>
>> Anyway if you are sure about your change I am OK with it also :)
> You're right, that's indeed an issue. It looks like a 0 irq is valid or
> invalid depending on who you ask. NO_IRQ is defined differently depending on
> the architecture :-/ I'll thus keep your version of the patch.
>
> Nonetheless, the core issue should be fixed. Do you feel adventurous ? :-)

Currently I am busy with other tasks, so I will be happy if somebody will
take care of it :), if not I will return to it if time permits.

Regards
Andrzej

>
>>>> dev_err(isp->dev, "No IRQ resource\n");
>>>> ret = -ENODEV;
>>>> goto error_iommu;
>>>>
>>>> }
>>>>
>>>> + isp->irq_num = ret;
>>>>
>>>> if (devm_request_irq(isp->dev, isp->irq_num, isp_isr, IRQF_SHARED,
>>>>
>>>> "OMAP3 ISP", isp)) {

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/