On 11/4/2024 20:04, Jarkko Nikula wrote:
> On 10/23/24 8:51 AM, Shyam Sundar S K wrote:
>> On AMD platforms, the IRQ lines are shared between two instances of
>> I3C.
>> Add IRQF_SHARED flag during the interrupt registration process.
>>
>> Co-developed-by: Sanket Goswami <Sanket.Goswami@amd.com>
>> Signed-off-by: Sanket Goswami <Sanket.Goswami@amd.com>
>> Signed-off-by: Shyam Sundar S K <Shyam-sundar.S-k@amd.com>
>> ---
>> drivers/i3c/master/dw-i3c-master.c | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/i3c/master/dw-i3c-master.c
>> b/drivers/i3c/master/dw-i3c-master.c
>> index 1a7c300b6d45..fd58a95ae1c3 100644
>> --- a/drivers/i3c/master/dw-i3c-master.c
>> +++ b/drivers/i3c/master/dw-i3c-master.c
>> @@ -1578,7 +1578,7 @@ int dw_i3c_common_probe(struct dw_i3c_master
>> *master,
>> writel(INTR_ALL, master->regs + INTR_STATUS);
>> irq = platform_get_irq(pdev, 0);
>> ret = devm_request_irq(&pdev->dev, irq,
>> - dw_i3c_master_irq_handler, 0,
>> + dw_i3c_master_irq_handler, IRQF_SHARED,
>> dev_name(&pdev->dev), master);
>
> dw_i3c_master_irq_handler() seems to be otherwise ready for shared
> interrupts but reminded me it might have a similar issue than
> drivers/i2c/busses/i2c-designware-master.c had [1] because both are
> runtime PM managed.
>
> To me it looks dw_i3c_master_irq_handler() may incorrectly process
> interrupt from other device if register reads return all bits one when
> device is suspended. Worth to check.
>
> 1. Commit cdbd2f169bf1 ("i2c: designware: Do not process interrupt
> when device is suspended")
Makes sense. I will remove this change in the next revision.
Thanks,
Shyam