Re: [PATCH 4/9] PCI: imx6: Using "linux,pci-domain" as slot ID

[Frank Li]

On Wed, Dec 06, 2023 at 10:59:53AM -0600, Bjorn Helgaas wrote:
> On Wed, Dec 06, 2023 at 11:50:16AM -0500, Frank Li wrote:
> > On Wed, Dec 06, 2023 at 10:36:56AM -0600, Bjorn Helgaas wrote:
> > > In subject, maybe you mean "Use 'linux,pci-domain' as slot ID"?
> > > "Using" is the wrong verb form here.
> > > 
> > > On Wed, Dec 06, 2023 at 10:58:58AM -0500, Frank Li wrote:
> > > > Avoid use get slot id by compared with register physical address. If there
> > > > are more than 2 slots, compared logic will become complex.
> > > 
> > > But this doesn't say anything about "linux,pci-domain", and I don't
> > > see anything about a register physical address in the patch.
> > > 
> > > Maybe this commit log was meant for a different patch?  I'm confused.
> > > 
> > > > Signed-off-by: Frank Li <Frank.Li@xxxxxxx>
> > > > ---
> > > >  drivers/pci/controller/dwc/pci-imx6.c | 6 ++++++
> > > >  1 file changed, 6 insertions(+)
> > > > 
> > > > diff --git a/drivers/pci/controller/dwc/pci-imx6.c b/drivers/pci/controller/dwc/pci-imx6.c
> > > > index 62d77fabd82a..239ef439ba70 100644
> > > > --- a/drivers/pci/controller/dwc/pci-imx6.c
> > > > +++ b/drivers/pci/controller/dwc/pci-imx6.c
> > > > @@ -33,6 +33,7 @@
> > > >  #include <linux/pm_domain.h>
> > > >  #include <linux/pm_runtime.h>
> > > >  
> > > > +#include "../../pci.h"
> > > >  #include "pcie-designware.h"
> > > >  
> > > >  #define IMX8MQ_GPR_PCIE_REF_USE_PAD		BIT(9)
> > > > @@ -1333,6 +1334,11 @@ static int imx6_pcie_probe(struct platform_device *pdev)
> > > >  					     "Failed to get PCIEPHY reset control\n");
> > > >  	}
> > > >  
> > > > +	/* Using linux,pci-domain as PCI slot id */
> > > > +	imx6_pcie->controller_id = of_get_pci_domain_nr(node);
> > > > +	if (imx6_pcie->controller_id)
> > > > +		imx6_pcie->controller_id = 0;
> > > 
> > > I don't understand what this is doing.  It looks the same as just:
> > 
> > Good capture. It should be 
> > if (imx6_pcie->controller_id < 0)
> > 	imx6_pcie->controller_id = 0;
> > 
> > for only one PCI controller case. I just tested first one slot before send
> > patch, so not met problem.
> > 
> > Previously, we use below logic
> > 	if (dbi_base->start == IMX8MQ_PCIE2_BASE_ADDR)
> > 		imx6_pcie->controller_id = 1;
> > 
> > It is not good to depend on register's base address. If there are 3
> > controllers, check logic will becomoe ugly.
> 
> Makes sense.  If the previous code depended on the base address, this
> patch would make more sense if it contained both the addition of the
> of_get_pci_domain_nr() call and the removal of the base address code.

Remove base address code will block existed functions. My plan is
1. this patch upstreamed
2. related dts add linux,pci-domain.
3. removed base address compare code.

Frank

> 
> > > Maybe this is a typo?  As written, it doesn't look like there's any
> > > point in calling of_get_pci_domain_nr().
> > > 
> > > >  	switch (imx6_pcie->drvdata->variant) {
> > > >  	case IMX7D:
> > > >  		if (dbi_base->start == IMX8MQ_PCIE2_BASE_ADDR)
> > > > -- 
> > > > 2.34.1
> > > >