Add check for disabled nodes. These nodes should not be probed. Can
result in logging for HW which is not present.
Signed-off-by: Aryan Srivastava <[email protected]>
---
drivers/i2c/busses/i2c-thunderx-pcidrv.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/drivers/i2c/busses/i2c-thunderx-pcidrv.c b/drivers/i2c/busses/i2c-thunderx-pcidrv.c
index a77cd86fe75e..7fcdb237076a 100644
--- a/drivers/i2c/busses/i2c-thunderx-pcidrv.c
+++ b/drivers/i2c/busses/i2c-thunderx-pcidrv.c
@@ -158,6 +158,10 @@ static int thunder_i2c_probe_pci(struct pci_dev *pdev,
struct octeon_i2c *i2c;
int ret;
+ /* Ignore disabled nodes. */
+ if (!of_device_is_available(pdev->dev.of_node))
+ return 0;
+
i2c = devm_kzalloc(dev, sizeof(*i2c), GFP_KERNEL);
if (!i2c)
return -ENOMEM;
--
2.41.0
On Tue, Jul 11, 2023 at 05:41:46PM +1200, Aryan Srivastava wrote:
> Add check for disabled nodes. These nodes should not be probed. Can
> result in logging for HW which is not present.
>
> Signed-off-by: Aryan Srivastava <[email protected]>
I am confused. This is a PCI driver, no? Why should we deal with DT
settings here?
On 30/10/23 06:11, Wolfram Sang wrote:
> On Tue, Jul 11, 2023 at 05:41:46PM +1200, Aryan Srivastava wrote:
>> Add check for disabled nodes. These nodes should not be probed. Can
>> result in logging for HW which is not present.
>>
>> Signed-off-by: Aryan Srivastava<[email protected]>
> I am confused. This is a PCI driver, no? Why should we deal with DT
> settings here?
It is a PCI driver but the IP block is included on SoCs like the CN9670
and not all hardware designs make use of all the I2C interfaces that the
IP block supports.