Re: [PATCH v8 4/5] usb: ehci: Add new EHCI driver for Broadcom STB SoC's

From: Andy Shevchenko <andy.shevchenko@gmail.com>
To: Alan Stern <stern@rowland.harvard.edu>
Cc: Al Cooper <alcooperx@gmail.com>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	bcm-kernel-feedback-list <bcm-kernel-feedback-list@broadcom.com>,
	devicetree <devicetree@vger.kernel.org>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Krzysztof Kozlowski <krzk@kernel.org>,
	USB <linux-usb@vger.kernel.org>,
	Mathias Nyman <mathias.nyman@intel.com>,
	Rob Herring <robh+dt@kernel.org>,
	Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>,
	Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Subject: Re: [PATCH v8 4/5] usb: ehci: Add new EHCI driver for Broadcom STB SoC's
Date: Sat, 9 May 2020 15:04:26 +0300	[thread overview]
Message-ID: <CAHp75VfV8SwKG5aX-POce5aHCoqCHdnfpxzq6C6buJ_TaSacsg@mail.gmail.com> (raw)
In-Reply-To: <Pine.LNX.4.44L0.2005082035460.2445-100000@netrider.rowland.org>

On Sat, May 9, 2020 at 3:47 AM Alan Stern <stern@rowland.harvard.edu> wrote:
> On Fri, 8 May 2020, Al Cooper wrote:

...

> > +     irq = platform_get_irq(pdev, 0);
> > +     if (irq <= 0)
> > +             return irq;
>
> I don't want to get involved in the question of whether or not 0 is a
> valid IRQ number.  The consensus has gone back and forth over the
> years, and it just doesn't seem important.
>
> However, as Sergei points out, if 0 is going to be regarded as an
> invalid value then we shouldn't return 0 from the probe function here.

The most of the drivers rely on the fact that on platforms where it's
the case (yes, we have you-know-which architecture has some legacy
stuff there) request_irq() will fail anyway.

> I'll leave the decision on how to handle this matter up to Greg.  :-)

Me too.

-- 
With Best Regards,
Andy Shevchenko