[PATCH 2/2] PCI: iproc: add bcma pcie driver

[Rafał Miłecki]

On 13 May 2015 at 17:56, Ray Jui <rjui at broadcom.com> wrote:
> On 5/12/2015 11:27 PM, Rafał Miłecki wrote:
>> On 12 May 2015 at 23:23, Hauke Mehrtens <hauke at hauke-m.de> wrote:
>>> This driver adds support for the PCIe 2.0 controller found on the bcma
>>> bus. This controller can be found on (mostly) all Broadcom BCM470X /
>>> BCM5301X ARM SoCs.
>>>
>>> The driver found in the Broadcom SDK does some more stuff, like setting
>>> up some DMA memory areas, chaining MPS and MRRS to 512 and also some
>>> PHY changes like "improving" the PCIe jitter and doing some special
>>> initializations for the 3rd PCIe port.
>>>
>>> This was tested on a bcm4708 board with 2 PCIe ports and wireless cards
>>> connected to them.
>>>
>>> PCI_DOMAINS is needed by this driver, because normally there is more
>>> than one PCIe controller and without PCI_DOMAINS only the first
>>> controller gets registered.
>>> This controller gets 6 IRQs, the last one is trigged by all IRQ events.
>>>
>>> Signed-off-by: Hauke Mehrtens <hauke at hauke-m.de>
>>
>> Acked-by: Rafał Miłecki <zajec5 at gmail.com>
>>
>>
>>> +static int iproc_pcie_bcma_probe(struct bcma_device *bdev)
>>> +{
>>> +       struct iproc_pcie *pcie;
>>> +       LIST_HEAD(res);
>>> +       struct resource res_mem;
>>> +       int ret;
>>> +
>>> +       pcie = devm_kzalloc(&bdev->dev, sizeof(*pcie), GFP_KERNEL);
>>> +       if (!pcie)
>>> +               return -ENOMEM;
>>> +
>>> +       pcie->dev = &bdev->dev;
>>> +       bcma_set_drvdata(bdev, pcie);
>>> +
>>> +       pcie->base = bdev->io_addr;
>>> +
>>> +       res_mem.start = bdev->addr_s[0];
>>> +       res_mem.end = bdev->addr_s[0] + SZ_128M - 1;
>>> +       res_mem.name = "PCIe MEM space";
>>> +       res_mem.flags = IORESOURCE_MEM;
>>> +       pci_add_resource(&res, &res_mem);
>>> +
>>> +       pcie->resources = &res;
>>> +
>>> +       pcie->map_irq = iproc_pcie_bcma_map_irq;
>>> +
>>> +       ret = iproc_pcie_setup(pcie);
>>
>> I think I don't like this part of iproc design. It lefts
>> pcie->resources pointing to some random memory after the setup/probe
>> are done. Guess it should be a separated parameter or sth.
>>
>> The patch is still OK, I just refer to generic iproc possible issue.
>>
> Sorry Rafal, but could you please be more specific on this?

iproc_pcie_pltfm_probe (and iproc_pcie_bcma_probe) have local "res"
variable (each its own). They do:
pcie->resources = &res;
and then they call
iproc_pcie_setup(pcie);

After iproc_pcie_pltfm_probe / iproc_pcie_bcma_probe returns, the
pointer pcie->resources is not valid anymore, yet pcie struct is still
in use. Of course pcie->resources isn't used anymore, but still, it's
some in-struct pointer (to the random memory since local variable is
not accessible anymore).

I think you should drop
struct list_head *resources;
from the struct iproc_pcie and use
iproc_pcie_setup(struct iproc_pcie *pcie, struct list_head *resources)

-- 
Rafał