Re: [PATCH] PCI: Fix link speed calculation on retrain failure

["Ilpo Järvinen" <ilpo.jarvinen@linux.intel.com>]

[-- Attachment #1: Type: text/plain, Size: 1331 bytes --]

On Tue, 24 Jun 2025, Lukas Wunner wrote:

> On Tue, Jun 24, 2025 at 02:23:33PM +0300, Ilpo Järvinen wrote:
> > On Mon, 23 Jun 2025, Lukas Wunner wrote:
> > > --- a/drivers/pci/quirks.c
> > > +++ b/drivers/pci/quirks.c
> > > @@ -108,7 +108,7 @@ int pcie_failed_link_retrain(struct pci_dev *dev)
> > >  	pcie_capability_read_word(dev, PCI_EXP_LNKCTL2, &lnkctl2);
> > >  	pcie_capability_read_word(dev, PCI_EXP_LNKSTA, &lnksta);
> > >  	if (!(lnksta & PCI_EXP_LNKSTA_DLLLA) && pcie_lbms_seen(dev, lnksta)) {
> > > -		u16 oldlnkctl2 = lnkctl2;
> > > +		u16 oldlnkctl2 = lnkctl2 & PCI_EXP_LNKCTL2_TLS;
> > >  
> > >  		pci_info(dev, "broken device, retraining non-functional downstream link at 2.5GT/s\n");
> > 
> > IIRC, there was a patch from somebody else which fixed this a bit 
> > differently but never got applied (many months ago by now).
> 
> Must be this one, still marked "New" in patchwork:
> 
> https://patchwork.kernel.org/project/linux-pci/patch/20250123055155.22648-2-sjiwei@163.com/
> 
> I don't care which one gets applied, as long as the issue is fixed.

Yes, that's the one.

It seemed pointless to me to require callers to apply the mask to 
the register value when the conversion helper could do that itself, 
especially when the macro takes in a parameter called "lnkctl2".

-- 
 i.