LS7A EHCI controller doesn't have extended capabilities, so the EECP (EHCI Extended Capabilities Pointer) field of HCCPARAMS register should be 0x0, but it reads as 0xa0 now. This is a hardware flaw and will be fixed in future, now just clear the EECP field to avoid error messages on boot:
...... [ 0.581675] pci 0000:00:04.1: EHCI: unrecognized capability ff [ 0.581699] pci 0000:00:04.1: EHCI: unrecognized capability ff [ 0.581716] pci 0000:00:04.1: EHCI: unrecognized capability ff [ 0.581851] pci 0000:00:04.1: EHCI: unrecognized capability ff ...... [ 0.581916] pci 0000:00:05.1: EHCI: unrecognized capability ff [ 0.581951] pci 0000:00:05.1: EHCI: unrecognized capability ff [ 0.582704] pci 0000:00:05.1: EHCI: unrecognized capability ff [ 0.582799] pci 0000:00:05.1: EHCI: unrecognized capability ff ......
Cc: stable@vger.kernel.org Signed-off-by: Baoqi Zhang zhangbaoqi@loongson.cn Signed-off-by: Huacai Chen chenhuacai@loongson.cn --- V2: Add a comment and don't touch pci_ids.h.
drivers/usb/host/pci-quirks.c | 9 +++++++++ 1 file changed, 9 insertions(+)
diff --git a/drivers/usb/host/pci-quirks.c b/drivers/usb/host/pci-quirks.c index 1f9c1b1435d8..0404489c2f6a 100644 --- a/drivers/usb/host/pci-quirks.c +++ b/drivers/usb/host/pci-quirks.c @@ -958,6 +958,15 @@ static void quirk_usb_disable_ehci(struct pci_dev *pdev) * booting from USB disk or using a usb keyboard */ hcc_params = readl(base + EHCI_HCC_PARAMS); + + /* LS7A EHCI controller doesn't have extended capabilities, the + * EECP (EHCI Extended Capabilities Pointer) field of HCCPARAMS + * register should be 0x0 but it reads as 0xa0. So clear it to + * avoid error messages on boot. + */ + if (pdev->vendor == PCI_VENDOR_ID_LOONGSON && pdev->device == 0x7a14) + hcc_params &= ~(0xffL << 8); + offset = (hcc_params >> 8) & 0xff; while (offset && --count) { pci_read_config_dword(pdev, offset, &cap);
On Sun, Feb 2, 2025 at 9:12 PM ruik r.marek@assembler.cz wrote:
Hi,
Dne 02. 02. 25 v 13:49 Huacai Chen napsal(a):
if (pdev->vendor == PCI_VENDOR_ID_LOONGSON && pdev->device == 0x7a14)
hcc_params &= ~(0xffL << 8);
If it would be fixed in the future, would it make sense to check for revision ID as well?
Next revision still has no extended capability, and the register reads as 0x0, so the result is the same. New revisions after the next revision are no plan now.
Huacai
Thanks, Rudolf
linux-stable-mirror@lists.linaro.org