Re: [PATCH 7/7] usb: host: ohci-exynos: Convert to devm_of_phy_optional_get()

From: Rob Herring <robh@kernel.org>
To: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: "Madalin Bucur" <madalin.bucur@nxp.com>,
	"David S . Miller" <davem@davemloft.net>,
	"Eric Dumazet" <edumazet@google.com>,
	"Jakub Kicinski" <kuba@kernel.org>,
	"Paolo Abeni" <pabeni@redhat.com>,
	"Horatiu Vultur" <horatiu.vultur@microchip.com>,
	UNGLinuxDriver@microchip.com,
	"Thierry Reding" <thierry.reding@gmail.com>,
	"Lorenzo Pieralisi" <lpieralisi@kernel.org>,
	"Krzysztof Wilczyński" <kw@linux.com>,
	"Bjorn Helgaas" <bhelgaas@google.com>,
	"Jonathan Hunter" <jonathanh@nvidia.com>,
	"Vinod Koul" <vkoul@kernel.org>,
	"Kishon Vijay Abraham I" <kishon@kernel.org>,
	"Alan Stern" <stern@rowland.harvard.edu>,
	"Greg Kroah-Hartman" <gregkh@linuxfoundation.org>,
	"Krzysztof Kozlowski" <krzysztof.kozlowski@linaro.org>,
	"Alim Akhtar" <alim.akhtar@samsung.com>,
	"Siddharth Vadapalli" <s-vadapalli@ti.com>,
	"Russell King" <linux@armlinux.org.uk>,
	netdev@vger.kernel.org, linux-tegra@vger.kernel.org,
	linux-pci@vger.kernel.org, linux-phy@lists.infradead.org,
	linux-usb@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
	linux-samsung-soc@vger.kernel.org
Subject: Re: [PATCH 7/7] usb: host: ohci-exynos: Convert to devm_of_phy_optional_get()
Date: Wed, 18 Jan 2023 13:49:10 -0600	[thread overview]
Message-ID: <CAL_JsqJWEzb_hxi0_sSj-5F0q4A9UcJEhwcSArWT6eAffpeqHA@mail.gmail.com> (raw)
In-Reply-To: <CAMuHMdXGsmNjYy-ofmuHLkr8yaDEzy+SGnhtbmc_2ezbEKAMjw@mail.gmail.com>

On Wed, Jan 18, 2023 at 12:28 PM Geert Uytterhoeven
<geert@linux-m68k.org> wrote:
>
> Hi Rob,
>
> On Wed, Jan 18, 2023 at 6:30 PM Rob Herring <robh@kernel.org> wrote:
> > On Wed, Jan 18, 2023 at 4:15 AM Geert Uytterhoeven
> > <geert+renesas@glider.be> wrote:
> > > Use the new devm_of_phy_optional_get() helper instead of open-coding the
> > > same operation.
> > >
> > > This lets us drop several checks for IS_ERR(), as phy_power_{on,off}()
> > > handle NULL parameters fine.
> > >
> > > Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
> > > ---
> > >  drivers/usb/host/ohci-exynos.c | 24 +++++++-----------------
> > >  1 file changed, 7 insertions(+), 17 deletions(-)
> > >
> > > diff --git a/drivers/usb/host/ohci-exynos.c b/drivers/usb/host/ohci-exynos.c
> > > index 8d7977fd5d3bd502..8dd9c3b2411c383f 100644
> > > --- a/drivers/usb/host/ohci-exynos.c
> > > +++ b/drivers/usb/host/ohci-exynos.c
> > > @@ -69,19 +69,12 @@ static int exynos_ohci_get_phy(struct device *dev,
> > >                         return -EINVAL;
> > >                 }
> > >
> > > -               phy = devm_of_phy_get(dev, child, NULL);
> > > +               phy = devm_of_phy_optional_get(dev, child, NULL);
> > >                 exynos_ohci->phy[phy_number] = phy;
> > >                 if (IS_ERR(phy)) {
> > > -                       ret = PTR_ERR(phy);
> > > -                       if (ret == -EPROBE_DEFER) {
> > > -                               of_node_put(child);
> > > -                               return ret;
> > > -                       } else if (ret != -ENOSYS && ret != -ENODEV) {
> > > -                               dev_err(dev,
> > > -                                       "Error retrieving usb2 phy: %d\n", ret);
> > > -                               of_node_put(child);
> > > -                               return ret;
> > > -                       }
> > > +                       of_node_put(child);
> > > +                       return dev_err_probe(dev, PTR_ERR(phy),
> > > +                                            "Error retrieving usb2 phy\n");
> >
> > Optional is really the only reason for the caller to decide whether to
> > print an error message or not. If we have both flavors of 'get', then
> > really the 'get' functions should print an error message.
>
> In case of a real error, both should print an error message, right?
>
> Anyway, I understand that's a three step operation:
>   1. Introduce and convert to the _optional variant,
>   2. Add error printing to callees.
>   3. Remove error printing from callers.

I think you only need 2 out of 3 steps depending on the situation. In
this case, you can add error printing in the _optional variant when
you introduce it and then convert callers to it.

Where we already have an optional variant, then you need steps 2 and 3.

Rob