Messages in this thread | | | Date | Wed, 1 Oct 2014 20:02:59 +0100 | From | Mark Einon <> | Subject | Re: [PATCH net-next] et131x: Add PCIe gigabit ethernet driver et131x to drivers/net |
| |
On Wed, Oct 01, 2014 at 04:14:49PM +0200, Tobias Klauser wrote: > On 2014-10-01 at 15:43:47 +0200, Joe Perches <joe@perches.com> wrote: > > On Wed, 2014-10-01 at 14:45 +0200, Tobias Klauser wrote: > > > On 2014-09-30 at 23:29:46 +0200, Mark Einon <mark.einon@gmail.com> wrote: > > > > This adds the ethernet driver for Agere et131x devices to > > > > drivers/net/ethernet. > > [] > > > > diff --git a/drivers/net/ethernet/agere/et131x.c b/drivers/net/ethernet/agere/et131x.c > > [] > > > > + rc = pci_enable_device(pdev); > > > > + if (rc < 0) { > > > > + dev_err(&pdev->dev, "pci_enable_device() failed\n"); > > > > + goto out; > > > > > > Nit: Just return rc here. > > > > I don't think it matters at all. > > Combined with my second remark this change makes the `out' label > unnecessary. If Mark decides to keep the goto here, at least the > position of the label should be changed to the end of the function, to > remain predictable and avoid jumping back.
Hi Tobias, Thanks for the review. Yes, I think it is a bit nit-picking, and I'd prefer to have just one return in the function. I don't think replacing the return with a goto adds much and in the interests of keeping churn down, I won't be making the change unless it prevents the driver moving out of staging.
Cheers,
Mark
| |