Re: [PATCH 05/14] lib: Add I/O map cache implementation

[Thierry Reding <thierry.reding@avionic-design.de>]

[-- Attachment #1: Type: text/plain, Size: 2504 bytes --]

On Thu, Jan 10, 2013 at 12:24:17PM -0700, Jason Gunthorpe wrote:
> On Thu, Jan 10, 2013 at 08:03:27PM +0100, Thierry Reding wrote:
> 
> > > > You'd piece a mapping together, each bus requires 16 64k mappings, a
> > > > simple 2d array of busnr*16 of pointers would do the trick. A more
> > > > clever solution would be to allocate contiguous virtual memory and
> > > > split that up..
>  
> > > Oh, I see. I'm not very familiar with the internals of remapping, so
> > > I'll need to do some more reading. Thanks for the hints.
> > 
> > I forgot to ask. What's the advantage of having a contiguous virtual
> > memory area and splitting it up versus remapping each chunk separately?
> 
> Not alot, really, but it saves you from the pointer array and
> associated overhead. IIRC it is fairly easy to do in the kernel.
> 
> Arnd's version is good too, but you would be restricted to aligned
> powers of two for the bus number range in the DT, which is probably
> not that big a deal either?

I've been trying to make this work, but this implementation always
triggers a BUG_ON() in lib/ioremap.c, line 27:

	27 		BUG_ON(!pte_none(*pte));

which seems to indicate that the page is already mapped, right?

Below is the relevant code:

struct tegra_pcie_bus {
	struct vm_struct *area;
	struct list_head list;
	unsigned int nr;
};

static struct tegra_pcie_bus *tegra_pcie_bus_alloc(struct tegra_pcie *pcie,
						   unsigned int busnr)
{
	unsigned long flags = VM_READ | VM_WRITE | VM_IO | VM_PFNMAP |
			      VM_DONTEXPAND | VM_DONTDUMP;
	phys_addr_t cs = pcie->cs->start;
	struct tegra_pcie_bus *bus;
	struct vm_struct *vm;
	unsigned int i;
	int err;

	bus = devm_kzalloc(pcie->dev, sizeof(*bus), GFP_KERNEL);
	if (!bus)
		return ERR_PTR(-ENOMEM);

	INIT_LIST_HEAD(&bus->list);
	bus->nr = busnr;

	bus->area = get_vm_area(SZ_1M, VM_IOREMAP);
	if (!bus->area) {
		err = -ENOMEM;
		goto free;
	}

	for (i = 0; i < 16; i++) {
		unsigned long virt = (unsigned long)bus->area->addr +
				     i * SZ_64K;
		phys_addr_t phys = cs + busnr * SZ_64K + i * SZ_1M;

		err = ioremap_page_range(virt, virt + SZ_64K - 1, phys,
					 vm_get_page_prot(flags));
		if (err < 0) {
			dev_err(pcie->dev, "ioremap_page_range() failed: %d\n",
				err);
			goto unmap;
		}
	}

	return bus;

unmap:
	vunmap(bus->area->addr);
	free_vm_area(bus->area);
free:
	devm_kfree(pcie->dev, bus);
	return ERR_PTR(err);
}

Anybody see what's wrong with that?

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]