Bug#622649: linux-image-2.6.38-2-amd64: unable to enumerate USB device on port 6
Am Freitag, 15. April 2011 schrieb Alan Stern:
> On Fri, 15 Apr 2011, Ben Hutchings wrote:
> > On Wed, 2011-04-13 at 18:32 +0200, Hans-J. Ullrich wrote:
> > [...]
> >
> > > I get massive messages from the kernel "unable to enumerate USB device
> > > on port 6" which is making the pc slower and
> > > which is filling up my syslog. The problem is caused by the ohci-hcd
> > > kernel module. Unloading it, is stopping
> > > the messages, but then the mouse will not work (of course).
> > >
> > > This bug appeared the first time at the change from
> > > linux-image-2.6.32-5-amd64-30 --> linux-image-2.6.32-5-amd64-31 (and
> > > higher!)
> > > and it appeared only on 64-bit-systems.
> >
> > [...]
> >
> > The only change to ohci-hcd between these versions was:
> >
> > commit 5f528de0ef9b3e092e276d95930830b847b33dc4
> > Author: Alan Stern <stern@rowland.harvard.edu>
> > Date: Fri Sep 10 16:37:05 2010 -0400
> >
> > OHCI: work around for nVidia shutdown problem
> >
> > commit 3df7169e73fc1d71a39cffeacc969f6840cdf52b upstream.
> >
> > The reporter's system does indeed have an nVidia MCP67 chipset. Full
> > details (kernel log, lspci, etc.) are at <http://bugs.debian.org/620848>
> > and <http://bugs.debian.org/622649>.
>
> Hans, can you test whether the patch below fixes your problem?
>
> Alan Stern
>
>
>
> Index: usb-2.6/drivers/usb/host/ohci-pci.c
> ===================================================================
> --- usb-2.6.orig/drivers/usb/host/ohci-pci.c
> +++ usb-2.6/drivers/usb/host/ohci-pci.c
> @@ -207,10 +207,18 @@ static int ohci_quirk_amd700(struct usb_
> */
> static int ohci_quirk_nvidia_shutdown(struct usb_hcd *hcd)
> {
> + struct pci_dev *pdev = to_pci_dev(hcd->self.controller);
> struct ohci_hcd *ohci = hcd_to_ohci(hcd);
>
> - ohci->flags |= OHCI_QUIRK_SHUTDOWN;
> - ohci_dbg(ohci, "enabled nVidia shutdown quirk\n");
> + /* Evidently nVidia fixed their later hardware; this is a guess at
> + * the changeover point.
> + */
> +#define PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_USB 0x026d
> +
> + if (pdev->device < PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_USB) {
> + ohci->flags |= OHCI_QUIRK_SHUTDOWN;
> + ohci_dbg(ohci, "enabled nVidia shutdown quirk\n");
> + }
>
> return 0;
> }
Hello Alan,
it lasted a little bit, as I am not very experienced with patching of source-
code. However, I did as you suggested, and hope I did everything correct.
I compiled the source again and it worked. Please take e look at the source
code. I send it below.
If I did it correct, I suggest, to patch all versions of the linux kernel
since version 2.6.32-5-amd64-30 and higher.
Maybe a new and patched kernel version should be uploaded to the repository as
soon as possible, what do you think?
Thank you very much for your help!!!
Cheers
Hans
/*
* OHCI HCD (Host Controller Driver) for USB.
*
* (C) Copyright 1999 Roman Weissgaerber <weissg@vienna.at>
* (C) Copyright 2000-2002 David Brownell <dbrownell@users.sourceforge.net>
*
* [ Initialisation is based on Linus' ]
* [ uhci code and gregs ohci fragments ]
* [ (C) Copyright 1999 Linus Torvalds ]
* [ (C) Copyright 1999 Gregory P. Smith]
*
* PCI Bus Glue
*
* This file is licenced under the GPL.
*/
#ifndef CONFIG_PCI
#error "This file is PCI bus glue. CONFIG_PCI must be defined."
#endif
#include <linux/pci.h>
#include <linux/io.h>
/* constants used to work around PM-related transfer
* glitches in some AMD 700 series southbridges
*/
#define AB_REG_BAR 0xf0
#define AB_INDX(addr) ((addr) + 0x00)
#define AB_DATA(addr) ((addr) + 0x04)
#define AX_INDXC 0X30
#define AX_DATAC 0x34
#define NB_PCIE_INDX_ADDR 0xe0
#define NB_PCIE_INDX_DATA 0xe4
#define PCIE_P_CNTL 0x10040
#define BIF_NB 0x10002
static struct pci_dev *amd_smbus_dev;
static struct pci_dev *amd_hb_dev;
static int amd_ohci_iso_count;
/*-------------------------------------------------------------------------*/
static int broken_suspend(struct usb_hcd *hcd)
{
device_init_wakeup(&hcd->self.root_hub->dev, 0);
return 0;
}
/* AMD 756, for most chips (early revs), corrupts register
* values on read ... so enable the vendor workaround.
*/
static int ohci_quirk_amd756(struct usb_hcd *hcd)
{
struct ohci_hcd *ohci = hcd_to_ohci (hcd);
ohci->flags = OHCI_QUIRK_AMD756;
ohci_dbg (ohci, "AMD756 erratum 4 workaround\n");
/* also erratum 10 (suspend/resume issues) */
return broken_suspend(hcd);
}
/* Apple's OHCI driver has a lot of bizarre workarounds
* for this chip. Evidently control and bulk lists
* can get confused. (B&W G3 models, and ...)
*/
static int ohci_quirk_opti(struct usb_hcd *hcd)
{
struct ohci_hcd *ohci = hcd_to_ohci (hcd);
ohci_dbg (ohci, "WARNING: OPTi workarounds unavailable\n");
return 0;
}
/* Check for NSC87560. We have to look at the bridge (fn1) to
* identify the USB (fn2). This quirk might apply to more or
* even all NSC stuff.
*/
static int ohci_quirk_ns(struct usb_hcd *hcd)
{
struct pci_dev *pdev = to_pci_dev(hcd->self.controller);
struct pci_dev *b;
b = pci_get_slot (pdev->bus, PCI_DEVFN (PCI_SLOT (pdev->devfn), 1));
if (b && b->device == PCI_DEVICE_ID_NS_87560_LIO
&& b->vendor == PCI_VENDOR_ID_NS) {
struct ohci_hcd *ohci = hcd_to_ohci (hcd);
ohci->flags |= OHCI_QUIRK_SUPERIO;
ohci_dbg (ohci, "Using NSC SuperIO setup\n");
}
pci_dev_put(b);
return 0;
}
/* Check for Compaq's ZFMicro chipset, which needs short
* delays before control or bulk queues get re-activated
* in finish_unlinks()
*/
static int ohci_quirk_zfmicro(struct usb_hcd *hcd)
{
struct ohci_hcd *ohci = hcd_to_ohci (hcd);
ohci->flags |= OHCI_QUIRK_ZFMICRO;
ohci_dbg(ohci, "enabled Compaq ZFMicro chipset quirks\n");
return 0;
}
/* Check for Toshiba SCC OHCI which has big endian registers
* and little endian in memory data structures
*/
static int ohci_quirk_toshiba_scc(struct usb_hcd *hcd)
{
struct ohci_hcd *ohci = hcd_to_ohci (hcd);
/* That chip is only present in the southbridge of some
* cell based platforms which are supposed to select
* CONFIG_USB_OHCI_BIG_ENDIAN_MMIO. We verify here if
* that was the case though.
*/
#ifdef CONFIG_USB_OHCI_BIG_ENDIAN_MMIO
ohci->flags |= OHCI_QUIRK_BE_MMIO;
ohci_dbg (ohci, "enabled big endian Toshiba quirk\n");
return 0;
#else
ohci_err (ohci, "unsupported big endian Toshiba quirk\n");
return -ENXIO;
#endif
}
/* Check for NEC chip and apply quirk for allegedly lost interrupts.
*/
static void ohci_quirk_nec_worker(struct work_struct *work)
{
struct ohci_hcd *ohci = container_of(work, struct ohci_hcd, nec_work);
int status;
status = ohci_init(ohci);
if (status != 0) {
ohci_err(ohci, "Restarting NEC controller failed in %s, %d\n",
"ohci_init", status);
return;
}
status = ohci_restart(ohci);
if (status != 0)
ohci_err(ohci, "Restarting NEC controller failed in %s, %d\n",
"ohci_restart", status);
}
static int ohci_quirk_nec(struct usb_hcd *hcd)
{
struct ohci_hcd *ohci = hcd_to_ohci (hcd);
ohci->flags |= OHCI_QUIRK_NEC;
INIT_WORK(&ohci->nec_work, ohci_quirk_nec_worker);
ohci_dbg (ohci, "enabled NEC chipset lost interrupt quirk\n");
return 0;
}
static int ohci_quirk_amd700(struct usb_hcd *hcd)
{
struct ohci_hcd *ohci = hcd_to_ohci(hcd);
u8 rev = 0;
if (!amd_smbus_dev)
amd_smbus_dev = pci_get_device(PCI_VENDOR_ID_ATI,
PCI_DEVICE_ID_ATI_SBX00_SMBUS, NULL);
if (!amd_smbus_dev)
return 0;
pci_read_config_byte(amd_smbus_dev, PCI_REVISION_ID, &rev);
/* SB800 needs pre-fetch fix */
if ((rev >= 0x40) && (rev <= 0x4f)) {
ohci->flags |= OHCI_QUIRK_AMD_PREFETCH;
ohci_dbg(ohci, "enabled AMD prefetch quirk\n");
}
if ((rev > 0x3b) || (rev < 0x30)) {
pci_dev_put(amd_smbus_dev);
amd_smbus_dev = NULL;
return 0;
}
amd_ohci_iso_count++;
if (!amd_hb_dev)
amd_hb_dev = pci_get_device(PCI_VENDOR_ID_AMD, 0x9600, NULL);
ohci->flags |= OHCI_QUIRK_AMD_ISO;
ohci_dbg(ohci, "enabled AMD ISO transfers quirk\n");
return 0;
}
/* nVidia controllers continue to drive Reset signalling on the bus
* even after system shutdown, wasting power. This flag tells the
* shutdown routine to leave the controller OPERATIONAL instead of RESET.
*/
static int ohci_quirk_nvidia_shutdown(struct usb_hcd *hcd)
{
struct pci_dev *pdev = to_pci_dev(hcd->self.controller);
struct ohci_hcd *ohci = hcd_to_ohci(hcd);
/* Evidently nVidia fixed their later hardware; this is a guess at
* the changeover point.
*/
#define PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_USB 0x026d
if (pdev->device < PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_USB) {
ohci->flags |= OHCI_QUIRK_SHUTDOWN;
ohci_dbg(ohci, "enabled nVidia shutdown quirk\n");
}
return 0;
}
/*
* The hardware normally enables the A-link power management feature, which
* lets the system lower the power consumption in idle states.
*
* Assume the system is configured to have USB 1.1 ISO transfers going
* to or from a USB device. Without this quirk, that stream may stutter
* or have breaks occasionally. For transfers going to speakers, this
* makes a very audible mess...
*
* That audio playback corruption is due to the audio stream getting
* interrupted occasionally when the link goes in lower power state
* This USB quirk prevents the link going into that lower power state
* during audio playback or other ISO operations.
*/
static void quirk_amd_pll(int on)
{
u32 addr;
u32 val;
u32 bit = (on > 0) ? 1 : 0;
pci_read_config_dword(amd_smbus_dev, AB_REG_BAR, &addr);
/* BIT names/meanings are NDA-protected, sorry ... */
outl(AX_INDXC, AB_INDX(addr));
outl(0x40, AB_DATA(addr));
outl(AX_DATAC, AB_INDX(addr));
val = inl(AB_DATA(addr));
val &= ~((1 << 3) | (1 << 4) | (1 << 9));
val |= (bit << 3) | ((!bit) << 4) | ((!bit) << 9);
outl(val, AB_DATA(addr));
if (amd_hb_dev) {
addr = PCIE_P_CNTL;
pci_write_config_dword(amd_hb_dev, NB_PCIE_INDX_ADDR, addr);
pci_read_config_dword(amd_hb_dev, NB_PCIE_INDX_DATA, &val);
val &= ~(1 | (1 << 3) | (1 << 4) | (1 << 9) | (1 << 12));
val |= bit | (bit << 3) | (bit << 12);
val |= ((!bit) << 4) | ((!bit) << 9);
pci_write_config_dword(amd_hb_dev, NB_PCIE_INDX_DATA, val);
addr = BIF_NB;
pci_write_config_dword(amd_hb_dev, NB_PCIE_INDX_ADDR, addr);
pci_read_config_dword(amd_hb_dev, NB_PCIE_INDX_DATA, &val);
val &= ~(1 << 8);
val |= bit << 8;
pci_write_config_dword(amd_hb_dev, NB_PCIE_INDX_DATA, val);
}
}
static void amd_iso_dev_put(void)
{
amd_ohci_iso_count--;
if (amd_ohci_iso_count == 0) {
if (amd_smbus_dev) {
pci_dev_put(amd_smbus_dev);
amd_smbus_dev = NULL;
}
if (amd_hb_dev) {
pci_dev_put(amd_hb_dev);
amd_hb_dev = NULL;
}
}
}
static void sb800_prefetch(struct ohci_hcd *ohci, int on)
{
struct pci_dev *pdev;
u16 misc;
pdev = to_pci_dev(ohci_to_hcd(ohci)->self.controller);
pci_read_config_word(pdev, 0x50, &misc);
if (on == 0)
pci_write_config_word(pdev, 0x50, misc & 0xfcff);
else
pci_write_config_word(pdev, 0x50, misc | 0x0300);
}
/* List of quirks for OHCI */
static const struct pci_device_id ohci_pci_quirks[] = {
{
PCI_DEVICE(PCI_VENDOR_ID_AMD, 0x740c),
.driver_data = (unsigned long)ohci_quirk_amd756,
},
{
PCI_DEVICE(PCI_VENDOR_ID_OPTI, 0xc861),
.driver_data = (unsigned long)ohci_quirk_opti,
},
{
PCI_DEVICE(PCI_VENDOR_ID_NS, PCI_ANY_ID),
.driver_data = (unsigned long)ohci_quirk_ns,
},
{
PCI_DEVICE(PCI_VENDOR_ID_COMPAQ, 0xa0f8),
.driver_data = (unsigned long)ohci_quirk_zfmicro,
},
{
PCI_DEVICE(PCI_VENDOR_ID_TOSHIBA_2, 0x01b6),
.driver_data = (unsigned long)ohci_quirk_toshiba_scc,
},
{
PCI_DEVICE(PCI_VENDOR_ID_NEC, PCI_DEVICE_ID_NEC_USB),
.driver_data = (unsigned long)ohci_quirk_nec,
},
{
/* Toshiba portege 4000 */
.vendor = PCI_VENDOR_ID_AL,
.device = 0x5237,
.subvendor = PCI_VENDOR_ID_TOSHIBA,
.subdevice = 0x0004,
.driver_data = (unsigned long) broken_suspend,
},
{
PCI_DEVICE(PCI_VENDOR_ID_ITE, 0x8152),
.driver_data = (unsigned long) broken_suspend,
},
{
PCI_DEVICE(PCI_VENDOR_ID_ATI, 0x4397),
.driver_data = (unsigned long)ohci_quirk_amd700,
},
{
PCI_DEVICE(PCI_VENDOR_ID_ATI, 0x4398),
.driver_data = (unsigned long)ohci_quirk_amd700,
},
{
PCI_DEVICE(PCI_VENDOR_ID_ATI, 0x4399),
.driver_data = (unsigned long)ohci_quirk_amd700,
},
{
PCI_DEVICE(PCI_VENDOR_ID_NVIDIA, PCI_ANY_ID),
.driver_data = (unsigned long) ohci_quirk_nvidia_shutdown,
},
/* FIXME for some of the early AMD 760 southbridges, OHCI
* won't work at all. blacklist them.
*/
{},
};
static int ohci_pci_reset (struct usb_hcd *hcd)
{
struct ohci_hcd *ohci = hcd_to_ohci (hcd);
int ret = 0;
if (hcd->self.controller) {
struct pci_dev *pdev = to_pci_dev(hcd->self.controller);
const struct pci_device_id *quirk_id;
quirk_id = pci_match_id(ohci_pci_quirks, pdev);
if (quirk_id != NULL) {
int (*quirk)(struct usb_hcd *ohci);
quirk = (void *)quirk_id->driver_data;
ret = quirk(hcd);
}
}
if (ret == 0) {
ohci_hcd_init (ohci);
return ohci_init (ohci);
}
return ret;
}
static int __devinit ohci_pci_start (struct usb_hcd *hcd)
{
struct ohci_hcd *ohci = hcd_to_ohci (hcd);
int ret;
#ifdef CONFIG_PM /* avoid warnings about unused pdev */
if (hcd->self.controller) {
struct pci_dev *pdev = to_pci_dev(hcd->self.controller);
/* RWC may not be set for add-in PCI cards, since boot
* firmware probably ignored them. This transfers PCI
* PM wakeup capabilities.
*/
if (device_can_wakeup(&pdev->dev))
ohci->hc_control |= OHCI_CTRL_RWC;
}
#endif /* CONFIG_PM */
ret = ohci_run (ohci);
if (ret < 0) {
ohci_err (ohci, "can't start\n");
ohci_stop (hcd);
}
return ret;
}
#ifdef CONFIG_PM
static int ohci_pci_suspend(struct usb_hcd *hcd, bool do_wakeup)
{
struct ohci_hcd *ohci = hcd_to_ohci (hcd);
unsigned long flags;
int rc = 0;
/* Root hub was already suspended. Disable irq emission and
* mark HW unaccessible, bail out if RH has been resumed. Use
* the spinlock to properly synchronize with possible pending
* RH suspend or resume activity.
*
* This is still racy as hcd->state is manipulated outside of
* any locks =P But that will be a different fix.
*/
spin_lock_irqsave (&ohci->lock, flags);
if (hcd->state != HC_STATE_SUSPENDED) {
rc = -EINVAL;
goto bail;
}
ohci_writel(ohci, OHCI_INTR_MIE, &ohci->regs->intrdisable);
(void)ohci_readl(ohci, &ohci->regs->intrdisable);
clear_bit(HCD_FLAG_HW_ACCESSIBLE, &hcd->flags);
bail:
spin_unlock_irqrestore (&ohci->lock, flags);
return rc;
}
static int ohci_pci_resume(struct usb_hcd *hcd, bool hibernated)
{
set_bit(HCD_FLAG_HW_ACCESSIBLE, &hcd->flags);
/* Make sure resume from hibernation re-enumerates everything */
if (hibernated)
ohci_usb_reset(hcd_to_ohci(hcd));
ohci_finish_controller_resume(hcd);
return 0;
}
#endif /* CONFIG_PM */
/*-------------------------------------------------------------------------*/
static const struct hc_driver ohci_pci_hc_driver = {
.description = hcd_name,
.product_desc = "OHCI Host Controller",
.hcd_priv_size = sizeof(struct ohci_hcd),
/*
* generic hardware linkage
*/
.irq = ohci_irq,
.flags = HCD_MEMORY | HCD_USB11,
/*
* basic lifecycle operations
*/
.reset = ohci_pci_reset,
.start = ohci_pci_start,
.stop = ohci_stop,
.shutdown = ohci_shutdown,
#ifdef CONFIG_PM
.pci_suspend = ohci_pci_suspend,
.pci_resume = ohci_pci_resume,
#endif
/*
* managing i/o requests and associated device resources
*/
.urb_enqueue = ohci_urb_enqueue,
.urb_dequeue = ohci_urb_dequeue,
.endpoint_disable = ohci_endpoint_disable,
/*
* scheduling support
*/
.get_frame_number = ohci_get_frame,
/*
* root hub support
*/
.hub_status_data = ohci_hub_status_data,
.hub_control = ohci_hub_control,
#ifdef CONFIG_PM
.bus_suspend = ohci_bus_suspend,
.bus_resume = ohci_bus_resume,
#endif
.start_port_reset = ohci_start_port_reset,
};
/*-------------------------------------------------------------------------*/
static const struct pci_device_id pci_ids [] = { {
/* handle any USB OHCI controller */
PCI_DEVICE_CLASS(PCI_CLASS_SERIAL_USB_OHCI, ~0),
.driver_data = (unsigned long) &ohci_pci_hc_driver,
}, { /* end: all zeroes */ }
};
MODULE_DEVICE_TABLE (pci, pci_ids);
/* pci driver glue; this is a "new style" PCI driver module */
static struct pci_driver ohci_pci_driver = {
.name = (char *) hcd_name,
.id_table = pci_ids,
.probe = usb_hcd_pci_probe,
.remove = usb_hcd_pci_remove,
.shutdown = usb_hcd_pci_shutdown,
#ifdef CONFIG_PM_SLEEP
.driver = {
.pm = &usb_hcd_pci_pm_ops
},
#endif
};
Reply to: