Re: [PATCH v2 03/13] PCI/ASPM: Disable ASPM when driver requests it

[Ilpo Järvinen]

On Wed, 11 Oct 2023, Bjorn Helgaas wrote:

> On Mon, Sep 18, 2023 at 04:10:53PM +0300, Ilpo Järvinen wrote:
> > PCI core/ASPM service driver allows controlling ASPM state through
> > pci_disable_link_state() and pci_enable_link_state() API. It was
> > decided earlier (see the Link below), to not allow ASPM changes when OS
> > does not have control over it but only log a warning about the problem
> > (commit 2add0ec14c25 ("PCI/ASPM: Warn when driver asks to disable ASPM,
> > but we can't do it")). Similarly, if ASPM is not enabled through
> > config, ASPM cannot be disabled.
> > ...
> 
> > +#ifndef CONFIG_PCIEASPM
> > +/*
> > + * Always disable ASPM when requested, even when CONFIG_PCIEASPM is
> > + * not build to avoid drivers adding code to do it on their own
> > + * which caused issues when core does not know about the out-of-band
> > + * ASPM state changes.
> > + */
> > +int pci_disable_link_state_locked(struct pci_dev *pdev, int state)
> > +{
> > +	struct pci_dev *parent = pdev->bus->self;
> > +	struct pci_bus *linkbus = pdev->bus;
> > +	struct pci_dev *child;
> > +	u16 aspm_enabled, linkctl;
> > +	int ret;
> > +
> > +	if (!parent)
> > +		return -ENODEV;
> 
> P.S. I think this should look the same to the user (same dmesg log and
> same taint, if we do that) as the CONFIG_PCIEASPM=y case.

Okay.

> > +	ret = pcie_capability_read_word(parent, PCI_EXP_LNKCTL, &linkctl);
> > +	if (ret != PCIBIOS_SUCCESSFUL)
> > +		return pcibios_err_to_errno(ret);
> > +	aspm_enabled = linkctl & PCI_EXP_LNKCTL_ASPMC;
> > +
> > +	ret = pcie_capability_read_word(pdev, PCI_EXP_LNKCTL, &linkctl);
> > +	if (ret != PCIBIOS_SUCCESSFUL)
> > +		return pcibios_err_to_errno(ret);
> > +	aspm_enabled |= linkctl & PCI_EXP_LNKCTL_ASPMC;
> > +
> > +	/* If no states need to be disabled, don't touch LNKCTL */
> > +	if (state & aspm_enabled)
> > +		return 0;
> > +
> > +	ret = pcie_capability_clear_word(parent, PCI_EXP_LNKCTL, PCI_EXP_LNKCTL_ASPMC);
> > +	if (ret != PCIBIOS_SUCCESSFUL)
> > +		return pcibios_err_to_errno(ret);
> > +	list_for_each_entry(child, &linkbus->devices, bus_list)
> > +		pcie_capability_clear_word(child, PCI_EXP_LNKCTL, PCI_EXP_LNKCTL_ASPMC);
> 
> This disables *all* ASPM states, unlike the version when
> CONFIG_PCIEASPM is enabled.  I suppose there's a reason, and maybe a
> comment could elaborate on it?
>
> When CONFIG_PCIEASPM is not enabled, I don't think we actively
> *disable* ASPM in the hardware; we just leave it as-is, so firmware
> might have left it enabled.

This whole trickery is intended for drivers that do not want to have ASPM 
because the devices are broken with it. So leaving it as-is is not really 
an option (as demonstrated by the custom workarounds).

> > +
> > +	return 0;
> > +}
> 
> Conceptually it seems like the LNKCTL updates here should be the same
> whether CONFIG_PCIEASPM is enabled or not (subject to the question
> above).
> 
> When CONFIG_PCIEASPM is enabled, we might need to do more stuff, but
> it seems like the core should be the same.

So you think it's safer to partially disable ASPM (as per driver's 
request) rather than disable it completely? I got the impression that the 
latter might be safer from what Rafael said earlier but I suppose I might 
have misinterpreted him since he didn't exactly say that it might be safer 
to _completely_ disable it.

-- 
 i.