[PPC] Boot problems after the pci-v6.18-changes

[Manivannan Sadhasivam]

On Wed, Oct 15, 2025 at 02:59:01PM +0200, Herve Codina wrote:
> On Wed, 15 Oct 2025 14:27:00 +0200
> Christian Zigotzky <chzigotzky at xenosoft.de> wrote:
> 
> > > On 15 October 2025 at 01:58 pm, Herve Codina <herve.codina at bootlin.com> wrote:
> > > 
> > > Hi Christian,
> > >   
> > >> On Wed, 15 Oct 2025 13:30:44 +0200
> > >> Christian Zigotzky <chzigotzky at xenosoft.de> wrote:
> > >> 
> > >> Hello Herve,
> > >>   
> > >>>> On 15 October 2025 at 10:39 am, Herve Codina <herve.codina at bootlin.com> wrote:  
> > >>> 
> > >>> Hi All,
> > >>> 
> > >>> I also observed issues with the commit f3ac2ff14834 ("PCI/ASPM: Enable all
> > >>> ClockPM and ASPM states for devicetree platforms")    
> > >> 
> > >> Thanks for reporting.
> > >>   
> > >>> 
> > >>> Also tried the quirk proposed in this discussion (quirk_disable_aspm_all)
> > >>> an the quirk also fixes the timing issue.    
> > >> 
> > >> Where have you added quirk_disable_aspm_all?  
> > > 
> > > --- 8< ---
> > > diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
> > > index 214ed060ca1b..a3808ab6e92e 100644
> > > --- a/drivers/pci/quirks.c
> > > +++ b/drivers/pci/quirks.c
> > > @@ -2525,6 +2525,17 @@ static void quirk_disable_aspm_l0s_l1(struct pci_dev *dev)
> > >  */
> > > DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ASMEDIA, 0x1080, quirk_disable_aspm_l0s_l1);
> > > 
> > > +static void quirk_disable_aspm_all(struct pci_dev *dev)
> > > +{
> > > +       pci_info(dev, "Disabling ASPM\n");
> > > +       pci_disable_link_state(dev, PCIE_LINK_STATE_ALL);
> > > +}
> > > +
> > > +/* LAN966x PCI board */
> > > +DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_EFAR, 0x9660, quirk_disable_aspm_all);
> > > +
> > > /*
> > >  * Some Pericom PCIe-to-PCI bridges in reverse mode need the PCIe Retrain
> > >  * Link bit cleared after starting the link retrain process to allow this
> > > --- 8< ---
> > > 
> > > Best regards,
> > > Hervé  
> > 
> > It is the same patch, I use for my AMD Radeon cards.
> > 
> > In my point of view we have to add a lot of other devices.
> 
> Yes, probably!
> 
> > 
> > But if the computer does not boot, will the average user know that there is a problem with the power management and their graphics card?
> > I am unsure whether I can deliver the kernel to average users later on.
> 
> Also when it boots, it is not easy to know about the problem root cause.
> 
> In my case, it was not obvious to make the relationship on my side between
> my ping timings behavior and ASPM.
> 

Interesting. So in your case, the issue is that ASPM adds up latency of the
network card? If so, it is intended. The added latency should correspond to the
L0s/L1 exit latencies.

If you want performance, then you should select CONFIG_PCIEASPM_PERFORMANCE
using Kconfig or pass 'pcie_aspm=off' in cmdline or do,

'echo performance > /sys/module/pcie_aspm/parameters/policy'

By default, enabling ASPM saves power, but it comes with a cost of reducing
performance.

- Mani

-- 
மணிவண்ணன் சதாசிவம்