2.6.36-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Brian J. Tarricone <[email protected]>

commit a85b4e7f4481c5a1ca89fa63c9c871151965075e upstream.

Tested on MacBookAir3,1.  Without this, we get EPROTO errors when
fetching device config descriptors.

Signed-off-by: Brian Tarricone <[email protected]>
Reported-by: Benoit Gschwind <[email protected]>
Tested-by: Edgar Hucek <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

---
 drivers/usb/host/ehci-pci.c |   12 ++++++++++++
 1 file changed, 12 insertions(+)

--- a/drivers/usb/host/ehci-pci.c
+++ b/drivers/usb/host/ehci-pci.c
@@ -148,6 +148,18 @@ static int ehci_pci_setup(struct usb_hcd
                        if (pdev->revision < 0xa4)
                                ehci->no_selective_suspend = 1;
                        break;
+
+               /* MCP89 chips on the MacBookAir3,1 give EPROTO when
+                * fetching device descriptors unless LPM is disabled.
+                * There are also intermittent problems enumerating
+                * devices with PPCD enabled.
+                */
+               case 0x0d9d:
+                       ehci_info(ehci, "disable lpm/ppcd for nvidia mcp89");
+                       ehci->has_lpm = 0;
+                       ehci->has_ppcd = 0;
+                       ehci->command &= ~CMD_PPCEE;
+                       break;
                }
                break;
        case PCI_VENDOR_ID_VIA:


_______________________________________________
stable mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/stable

Reply via email to