[+cc Tony]
On Wed, Oct 07, 2026 at 03:13:04PM +0200, Stian Halseth wrote:
> sparc64 and powerpc create PCI devices from the device tree in
> of_create_pci_dev(), which calls set_pcie_port_type() but not
> pci_setup_device(). Since d2bd39c0456b, supported_speeds is only set in
> pci_setup_device(), so it stays 0 for these devices.
> pcie_get_speed_cap() then returns PCI_SPEED_UNKNOWN, max_link_speed
> reads "Unknown", and the failed link retraining quirk passes
> PCI_SPEED_UNKNOWN to pcie_set_target_speed(), which warns on a SPARC
> T7-1:
>
> WARNING: drivers/pci/pcie/bwctrl.c:125 at pcie_set_target_speed+0x284/0x2e0
>
> Set supported_speeds in set_pcie_port_type(), which all three paths
> call.
>
> Fixes: d2bd39c0456b ("PCI: Store all PCIe Supported Link Speeds")
> Closes: https://github.com/sparclinux/issues/issues/109
> Signed-off-by: Stian Halseth <[email protected]>
Applied with Tony's Tested-by to pci/enumeration for v7.4, thanks!
I added a stable tag since d2bd39c0456b appeared in v6.13.
> ---
> Tested on a SPARC T7-1: before, max_link_speed reads "Unknown" and the
> warning above appears with CONFIG_PCIEPORTBUS; after, all 23 PCIe
> devices report their real speed and the quirk lifts the 2.5GT/s
> restriction without a warning. Not tested on powerpc.
>
> drivers/pci/probe.c | 5 ++---
> 1 file changed, 2 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
> index 27008e2ea5afc..52a79b033b069 100644
> --- a/drivers/pci/probe.c
> +++ b/drivers/pci/probe.c
> @@ -1651,6 +1651,8 @@ void set_pcie_port_type(struct pci_dev *pdev)
> pdev->aspm_l1_support = 1;
> #endif
>
> + pdev->supported_speeds = pcie_get_supported_speeds(pdev);
> +
> parent = pci_upstream_bridge(pdev);
> if (!parent)
> return;
> @@ -2074,9 +2076,6 @@ int pci_setup_device(struct pci_dev *dev)
>
> set_pcie_untrusted(dev);
>
> - if (pci_is_pcie(dev))
> - dev->supported_speeds = pcie_get_supported_speeds(dev);
> -
> /* "Unknown power state" */
> dev->current_state = PCI_UNKNOWN;
>
> --
> 2.43.0
>
