Re: [PATCH] USB: host: ehci: Add error handlinginehci_mxc_drv_probe()

[Tang Bin]

Hi Sergei:

On 2020/5/9 4:27, Sergei Shtylyov wrote:
> On 05/08/2020 06:03 PM, Tang Bin wrote:
>
> > > > > On Fri, May 08, 2020 at 07:44:53PM +0800, Tang Bin wrote:
> > > > > > The function ehci_mxc_drv_probe() does not perform sufficient error
> > > > > > checking after executing platform_get_irq(), thus fix it.
> > > > > >
> > > > > > Fixes: 7e8d5cd93fa ("USB: Add EHCI support for MX27 and MX31 based boards")
> > > > > > Signed-off-by: Zhang Shengju <zhangshengju@xxxxxxxxxxxxxxxxxxxx>
> > > > > > Signed-off-by: Tang Bin <tangbin@xxxxxxxxxxxxxxxxxxxx>
> > > > > > ---
> > > > > >     drivers/usb/host/ehci-mxc.c | 2 ++
> > > > > >     1 file changed, 2 insertions(+)
> > > > > >
> > > > > > diff --git a/drivers/usb/host/ehci-mxc.c b/drivers/usb/host/ehci-mxc.c
> > > > > > index a1eb5ee77..a0b42ba59 100644
> > > > > > --- a/drivers/usb/host/ehci-mxc.c
> > > > > > +++ b/drivers/usb/host/ehci-mxc.c
> > > > > > @@ -50,6 +50,8 @@ static int ehci_mxc_drv_probe(struct platform_device *pdev)
> > > > > >         }
> > > > > >         irq = platform_get_irq(pdev, 0);
> > > > > > +    if (irq < 0)
> > > > > > +        return irq;
> > > > > <= ?
> > > > In the file 'drivers/base/platform.c'ï the function platform_get_irq() is
> > > > explained and used as follows:
> > > >
> > > >        * Gets an IRQ for a platform device and prints an error message if
> > > > finding the
> > > >        * IRQ fails. Device drivers should check the return value for errors so
> > > > as to
> > > >        * not pass a negative integer value to the request_irq() APIs.
> > > >        *
> > > >        * Example:
> > > >        *        int irq = platform_get_irq(pdev, 0);
> > > >        *        if (irq < 0)
> > > >        *            return irq;
> > > >        *
> > > >        * Return: IRQ number on success, negative error number on failure.
> > > >
> > > > And in my hardware experiment, even if I set the irq failed deliberately in
> > > > the DTS, the returned value is negative instead of zero.
> > > Please read the thread at
> > >      https://lore.kernel.org/r/20200501224042.141366-1-helgaas%40kernel.org
> > > for more details about this.
> > Great, It looks beautiful, finally someone took a knife to the file 'platform.c'.
>     I thought I did that already couple years ago, when returned 0 from platform_get_irq() could mean both IRQ # and error... :-)
Can you tell me what platform can returned 0? I want to do this test in 
the hardware.
> > I have been studied this place for a long time, and don't know what platform can return 0, which made me curious.
> >
> > So the example should be:
> >
> >       *        int irq = platform_get_irq(pdev, 0);
> >       *        if (irq <= 0)
> >       *            return irq;
>     And you then return 0 (success) as if your probe() succeeded. Congratulations! :-P

Thanks,

Tang Bin