On Mon, Oct 06, 2014 at 08:00:56PM +0300, Laurent Pinchart wrote:
> Hi Felipe,
>
> On Monday 06 October 2014 11:53:52 Felipe Balbi wrote:
> > On Mon, Oct 06, 2014 at 07:15:42PM +0300, Laurent Pinchart wrote:
> > > > > +int isp1760_register(struct resource *mem, int irq, unsigned long
> > > > > irqflags, + int rst_gpio, struct device *dev, unsigned int
> > > > > devflags)
> > > > > +{
> > > > > + struct isp1760_device *isp;
> > > > > + int ret;
> > > > > +
> > > > > + if (usb_disabled())
> > > > > + return -ENODEV;
> > > > > +
> > > > > + /* prevent usb-core allocating DMA pages */
> > > > > + dev->dma_mask = NULL;
> > > > > +
> > > > > + isp = kzalloc(sizeof(*isp), GFP_KERNEL);
> > > > > + if (!isp)
> > > > > + return -ENOMEM;
> > > > > +
> > > > > + isp->rst_gpio = rst_gpio;
> > > > > +
> > > > > + isp->mem_start = mem->start;
> > > > > + isp->mem_size = resource_size(mem);
> > > > > +
> > > > > + isp->regs = ioremap(isp->mem_start, isp->mem_size);
> > > > > + if (!isp->regs) {
> > > > > + ret = -EIO;
> > > > > + goto error;
> > > > > + }
> > > >
> > > > where did request_mem_region() go ?
> > >
> > > It's still in the glue code.
> >
> > isn't kinda weird that request_mem_region() is handled by glue, but
> > relase_mem_region() isn't ?
>
> It can be :-) release_mem_region() will disappear when moving to the devm_*
> API though, which I was planning to do as a follow-up. I'll try to rebase the
> series on top of that change, which should fix this issue.
alright :-)
--
balbi
Attachment:
signature.asc
Description: Digital signature
