Messages in this thread |  | | Date | Thu, 17 Jan 2013 10:41:36 +0200 | From | Felipe Balbi <> | Subject | Re: [PATCH 03/10] goldfish: tty driver |
| |
Hi,
On Wed, Jan 16, 2013 at 06:25:20PM +0000, Alan Cox wrote: > On Wed, 16 Jan 2013 19:01:22 +0200 > Felipe Balbi <balbi@ti.com> wrote: > > > Hi, > > > > On Wed, Jan 16, 2013 at 04:59:03PM +0000, Alan Cox wrote: > > > +static int goldfish_tty_probe(struct platform_device *pdev) > > > +{ > > > + struct goldfish_tty *qtty; > > > + int ret = -EINVAL; > > > + int i; > > > + struct resource *r; > > > + struct device *ttydev; > > > + void __iomem *base; > > > + u32 irq; > > > + > > > + r = platform_get_resource(pdev, IORESOURCE_MEM, 0); > > > + if(r == NULL) > > > + return -EINVAL; > > > + > > > + base = ioremap(r->start, 0x1000); > > > > missing request_mem_region(), in fact you could just use > > devm_request_and_ioremap(). > > Actually the reservation is already done by the bus driver. If it sees > the goldfish interface it reserves the entire chunk and the emulator then > tells it which is in each chunk of that range.
a bit unusual, but fair enough.
cheers
-- balbi [unhandled content-type:application/pgp-signature] |  |