Skip to content

Commit

Permalink
[PATCH] USB: EHCI and NF2 quirk
Browse files Browse the repository at this point in the history
This teaches the EHCI driver about a quirk seen in older NForce2 chips,
adding a workaround to ignore selective suspend requests.  Bus-wide
(so-called "global") suspend still works, as does USB wakeup of a
root hub that's globally suspended.

There's still a hole in this support though.  Strictly speaking, this
should _fail_ selective suspend requests, rather than ignoring them,
since doing it this way means that devices which should be able to issue
remote wakeup are not going to be able to do that.  For now, we'll just
live with that problem ... since usbcore expects to do selective suspend
on the way towards a full bus suspend, and usbcore needs to be able to
do full bus suspend.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
  • Loading branch information
David Brownell authored and Greg Kroah-Hartman committed Mar 20, 2006
1 parent 4186ecf commit f8aeb3b
Show file tree
Hide file tree
Showing 3 changed files with 30 additions and 2 deletions.
4 changes: 4 additions & 0 deletions drivers/usb/host/ehci-hub.c
Original file line number Diff line number Diff line change
Expand Up @@ -359,6 +359,8 @@ static int ehci_hub_control (
case USB_PORT_FEAT_SUSPEND:
if (temp & PORT_RESET)
goto error;
if (ehci->no_selective_suspend)
break;
if (temp & PORT_SUSPEND) {
if ((temp & PORT_PE) == 0)
goto error;
Expand Down Expand Up @@ -514,6 +516,8 @@ static int ehci_hub_control (
temp &= ~PORT_RWC_BITS;
switch (wValue) {
case USB_PORT_FEAT_SUSPEND:
if (ehci->no_selective_suspend)
break;
if ((temp & PORT_PE) == 0
|| (temp & PORT_RESET) != 0)
goto error;
Expand Down
25 changes: 24 additions & 1 deletion drivers/usb/host/ehci-pci.c
Original file line number Diff line number Diff line change
Expand Up @@ -106,11 +106,11 @@ static int ehci_pci_setup(struct usb_hcd *hcd)
}
break;
case PCI_VENDOR_ID_NVIDIA:
switch (pdev->device) {
/* NVidia reports that certain chips don't handle
* QH, ITD, or SITD addresses above 2GB. (But TD,
* data buffer, and periodic schedule are normal.)
*/
switch (pdev->device) {
case 0x003c: /* MCP04 */
case 0x005b: /* CK804 */
case 0x00d8: /* CK8 */
Expand All @@ -120,6 +120,14 @@ static int ehci_pci_setup(struct usb_hcd *hcd)
ehci_warn(ehci, "can't enable NVidia "
"workaround for >2GB RAM\n");
break;
/* Some NForce2 chips have problems with selective suspend;
* fixed in newer silicon.
*/
case 0x0068:
pci_read_config_dword(pdev, PCI_REVISION_ID, &temp);
if ((temp & 0xff) < 0xa4)
ehci->no_selective_suspend = 1;
break;
}
break;
}
Expand Down Expand Up @@ -163,6 +171,21 @@ static int ehci_pci_setup(struct usb_hcd *hcd)
device_init_wakeup(&pdev->dev, 1);
}

#ifdef CONFIG_USB_SUSPEND
/* REVISIT: the controller works fine for wakeup iff the root hub
* itself is "globally" suspended, but usbcore currently doesn't
* understand such things.
*
* System suspend currently expects to be able to suspend the entire
* device tree, device-at-a-time. If we failed selective suspend
* reports, system suspend would fail; so the root hub code must claim
* success. That's lying to usbcore, and it matters for for runtime
* PM scenarios with selective suspend and remote wakeup...
*/
if (ehci->no_selective_suspend && device_can_wakeup(&pdev->dev))
ehci_warn(ehci, "selective suspend/wakeup unavailable\n");
#endif

retval = ehci_pci_reinit(ehci, pdev);
done:
return retval;
Expand Down
3 changes: 2 additions & 1 deletion drivers/usb/host/ehci.h
Original file line number Diff line number Diff line change
Expand Up @@ -89,6 +89,8 @@ struct ehci_hcd { /* one per controller */
u32 command;

unsigned is_tdi_rh_tt:1; /* TDI roothub with TT */
unsigned no_selective_suspend:1;
u8 sbrn; /* packed release number */

/* irq statistics */
#ifdef EHCI_STATS
Expand All @@ -97,7 +99,6 @@ struct ehci_hcd { /* one per controller */
#else
# define COUNT(x) do {} while (0)
#endif
u8 sbrn; /* packed release number */
};

/* convert between an HCD pointer and the corresponding EHCI_HCD */
Expand Down

0 comments on commit f8aeb3b

Please sign in to comment.