On Wed, Aug 21, 2024 at 11:36:36AM +0200, Philipp Stanner wrote:

On Wed, 2024-08-21 at 11:14 +0300, Andy Shevchenko wrote:

quoted

On Wed, Aug 21, 2024 at 10:19 AM Philipp Stanner
[off-list ref] wrote:

...

quoted

quoted

-       for (i = 0; i < PCI_STD_NUM_BARS; i++) {
-               if (pci_resource_len(pdev, i) == 0)
-                       continue;
-               pcim_iounmap_regions(pdev, BIT(i));

Here is the BARx, which contradicts the probe :-)

I'm not sure what should be done about it. The only interesting
question is whether the other code with pcim_iomap_regions(... BIT(i)
does also only grap BAR 0.
In that case the driver wouldn't even be knowing what its own hardware
is / does, though.

I think your patch does the right thing already.

quoted

quoted

-               break;
-       }

-- 
With Best Regards,
Andy Shevchenko