Re: [PATCH v2 05/10] PCI: keystone: Add ks_pcie_free_msi_irq() helper for cleanup

[Siddharth Vadapalli <s-vadapalli@ti.com>]

On Sat, Sep 20, 2025 at 12:02:34AM +0530, Manivannan Sadhasivam wrote:
> On Fri, Sep 12, 2025 at 05:46:16PM +0530, Siddharth Vadapalli wrote:
> > Introduce the helper function ks_pcie_free_msi_irq() which will undo the
> > configuration performed by the ks_pcie_config_msi_irq() function. This will
> > be required for implementing a future helper function to undo the
> > configuration performed by the ks_pcie_host_init() function.
> > 
> > Signed-off-by: Siddharth Vadapalli <s-vadapalli@ti.com>
> > ---
> > 
> > v1: https://lore.kernel.org/r/20250903124505.365913-6-s-vadapalli@ti.com/
> > No changes since v1.
> > 
> >  drivers/pci/controller/dwc/pci-keystone.c | 25 +++++++++++++++++++++++
> >  1 file changed, 25 insertions(+)
> > 
> > diff --git a/drivers/pci/controller/dwc/pci-keystone.c b/drivers/pci/controller/dwc/pci-keystone.c
> > index d03e95bf7d54..81c3686688c0 100644
> > --- a/drivers/pci/controller/dwc/pci-keystone.c
> > +++ b/drivers/pci/controller/dwc/pci-keystone.c
> > @@ -666,6 +666,31 @@ static void ks_pcie_intx_irq_handler(struct irq_desc *desc)
> >  	chained_irq_exit(chip, desc);
> >  }
> >  
> > +static void ks_pcie_free_msi_irq(struct keystone_pcie *ks_pcie)
> > +{
> > +	struct device_node *np = ks_pcie->np;
> > +	struct device_node *intc_np;
> > +	int irq_count, irq, i;
> > +
> > +	if (!IS_ENABLED(CONFIG_PCI_MSI))
> 
> Isn't the CONFIG_PCI_KEYSTONE_HOST always depend on PCI_MSI?

The reason I added the check is that it exists in 'ks_pcie_config_msi_irq()'.
But I realize now that it should be removed from
'ks_pcie_config_msi_irq()' as well. Since I had written the above
function with the objective of undoing the changes done by
'ks_pcie_config_msi_irq()', the 'config check' was retained since the
changes should be undone only if they were executed by
'ks_pcie_config_msi_irq()'. I will drop the check in the v3 series and
will also post a separate patch to drop if from 'ks_pcie_config_msi_irq()'
if that is acceptable. Please let me know.

Regards,
Siddharth.