Running io_watchdog_func() while ohci_urb_enqueue() is running can
cause a race condition where ohci->prev_frame_no is corrupted and the
watchdog can mis-detect following error:
ohci-platform 664a0800.usb: frame counter not updating; disabled
ohci-platform 664a0800.usb: HC died; cleaning up
Specifically, following scenario causes a race condition:
1. ohci_urb_enqueue() calls spin_lock_irqsave(&ohci->lock, flags)
and enters the critical section
2. ohci_urb_enqueue() calls timer_pending(&ohci->io_watchdog) and it
returns false
3. ohci_urb_enqueue() sets ohci->prev_frame_no to a frame number
read by ohci_frame_no(ohci)
4. ohci_urb_enqueue() schedules io_watchdog_func() with mod_timer()
5. ohci_urb_enqueue() calls spin_unlock_irqrestore(&ohci->lock,
flags) and exits the critical section
6. Later, ohci_urb_enqueue() is called
7. ohci_urb_enqueue() calls spin_lock_irqsave(&ohci->lock, flags)
and enters the critical section
8. The timer scheduled on step 4 expires and io_watchdog_func() runs
9. io_watchdog_func() calls spin_lock_irqsave(&ohci->lock, flags)
and waits on it because ohci_urb_enqueue() is already in the
critical section on step 7
10. ohci_urb_enqueue() calls timer_pending(&ohci->io_watchdog) and it
returns false
11. ohci_urb_enqueue() sets ohci->prev_frame_no to new frame number
read by ohci_frame_no(ohci) because the frame number proceeded
between step 3 and 6
12. ohci_urb_enqueue() schedules io_watchdog_func() with mod_timer()
13. ohci_urb_enqueue() calls spin_unlock_irqrestore(&ohci->lock,
flags) and exits the critical section, then wake up
io_watchdog_func() which is waiting on step 9
14. io_watchdog_func() enters the critical section
15. io_watchdog_func() calls ohci_frame_no(ohci) and set frame_no
variable to the frame number
16. io_watchdog_func() compares frame_no and ohci->prev_frame_no
On step 16, because this calling of io_watchdog_func() is scheduled on
step 4, the frame number set in ohci->prev_frame_no is expected to the
number set on step 3. However, ohci->prev_frame_no is overwritten on
step 11. Because step 16 is executed soon after step 11, the frame
number might not proceed, so ohci->prev_frame_no must equals to
frame_no.
To address above scenario, this patch introduces a special sentinel
value IO_WATCHDOG_OFF and set this value to ohci->prev_frame_no when
the watchdog is not pending or running. When ohci_urb_enqueue()
schedules the watchdog (step 4 and 12 above), it compares
ohci->prev_frame_no to IO_WATCHDOG_OFF so that ohci->prev_frame_no is
not overwritten while io_watchdog_func() is running.
Signed-off-by: Shigeru Yoshida <Shigeru.Yoshida@windriver.com>
Signed-off-by: Haiqing Bai <Haiqing.Bai@windriver.com>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
It's good to have SPDX identifiers in all files to make it easier to
audit the kernel tree for correct licenses.
Update the drivers/usb/ and include/linux/usb* files with the correct
SPDX license identifier based on the license text in the file itself.
The SPDX identifier is a legally binding shorthand, which can be used
instead of the full boiler plate text.
This work is based on a script and data from Thomas Gleixner, Philippe
Ombredanne, and Kate Stewart.
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Kate Stewart <kstewart@linuxfoundation.org>
Cc: Philippe Ombredanne <pombredanne@nexb.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Acked-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Acked-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The "dbg_port" macro uses the "outside" parameter (="temp") instead of
the parameters (="value") given in the macro. As the macro can look
outside its definition this causes no direct problem.
Signed-off-by: Jelle Martijn Kok <jmkok@youcom.nl>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fix using the bare number to set the 'bDescriptorType' field of the Hub
Descriptor while the value is #define'd in <linux/usb/ch11.h>.
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fix using the bare numbers to set the 'wHubCharacteristics' field of the Hub
Descriptor while the values are #define'd in <linux/usb/ch11.h>.
While at it, fix the indentation.
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
USB hub has started to use a workqueue instead of kthread. Let's update
the documentation and comments here and there.
This patch mostly just replaces "khubd" with "hub_wq". There are only few
exceptions where the whole sentence was updated. These more complicated
changes can be found in the following files:
Documentation/usb/hotplug.txt
drivers/net/usb/usbnet.c
drivers/usb/core/hcd.c
drivers/usb/host/ohci-hcd.c
drivers/usb/host/xhci.c
Signed-off-by: Petr Mladek <pmladek@suse.cz>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Some OHCI controllers have a bug: They fail to add completed TDs to
the done queue. Examining this queue is the only method ohci-hcd has
for telling when a transfer is complete; failure to add a TD can
result in an URB that never completes and cannot be unlinked.
This patch adds a watchdog routine to ohci-hcd. The routine
periodically scans the active ED and TD lists, looking for TDs which
are finished but not on the done queue. When one is found, and it is
certain that the controller hardware will never add the TD to the done
queue, the watchdog routine manually puts the TD on the done list so
that it can be handled normally.
The watchdog routine also checks for a condition indicating the
controller has died. If the done queue is non-empty but the
HccaDoneHead pointer hasn't been updated for a few hundred
milliseconds, we assume the controller will never update it and
therefore is dead.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
URBs for a particular endpoint should complete sequentially. That is,
we shouldn't call the completion handler for one URB until the handler
for the previous URB has returned.
When the OHCI watchdog routine is added, there will be two paths for
completing URBs: interrupt handler and watchdog routine. Their
activities have to be synchronized so that completions don't occur in
multiple threads concurrently.
For that purpose, this patch creates an ohci_work() routine which will
be responsible for calling process_done_list() and finish_unlinks(),
the two routines that detect when an URB is complete. Everything will
funnel through ohci_work(), and it will be careful not to run in more
than one thread at a time.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch changes the way ohci-hcd handles the TD done list. In
addition to relying on the TD pointers stored by the controller
hardware, we need to handle TDs that the hardware has forgotten about.
This means the list has to exist even while the dl_done_list() routine
isn't running. That function essentially gets split in two:
update_done_list() reads the TD pointers stored by the hardware and
adds the TDs to the done list, and process_done_list() scans through
the list to handle URB completions. When we detect a TD that the
hardware forgot about, we will be able to add it to the done list
manually and then process it normally.
Since the list is really a queue, and because there can be a lot of
TDs, keep the existing singly linked implementation. To insure that
URBs are given back in order of submission, whenever a TD is added to
the done list, all the preceding TDs for the same endpoint must be
added as well (going back to the first one that isn't already on the
done list).
The done list manipulations must all be protected by the private
lock. The scope of the lock is expanded in preparation for the
watchdog routine to be added in a later patch.
We have to be more careful about giving back unlinked URBs. Since TDs
may be added to the done list by the watchdog routine and not in
response to a controller interrupt, we have to check explicitly to
make sure all the URB's TDs that were added to the done list have been
processed before giving back the URB.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Some OHCI controllers from ATI/AMD seem to have difficulty with
"global" USB suspend, that is, suspending an entire USB bus without
setting the suspend feature for each port connected to a device. When
we try to resume the child devices, the controller gives timeout
errors on the unsuspended ports, requiring resets, and can even cause
ohci-hcd to hang; see
http://marc.info/?l=linux-usb&m=139514332820398&w=2
and the following messages.
This patch fixes the problem by adding a new quirk flag to ohci-hcd.
The flag causes the ohci_rh_suspend() routine to suspend each
unsuspended, enabled port before suspending the root hub. This
effectively converts the "global" suspend to an ordinary root-hub
suspend. There is no need to unsuspend these ports when the root hub
is resumed, because the child devices will be resumed anyway in the
course of a normal system resume ("global" suspend is never used for
runtime PM).
This patch should be applied to all stable kernels which include
commit 0aa2832dd0 (USB: use "global suspend" for system sleep on
USB-2 buses) or a backported version thereof.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Reported-by: Peter Münster <pmlists@free.fr>
Tested-by: Peter Münster <pmlists@free.fr>
CC: <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Platform drivers sometimes need to perform specific handling of hub
control requests and status data. Make this possible by exporting the
ohci_hub_control() and ohci_hub_status_data() functions which can then
be called from custom hub operations in the default case.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
With dynamic debugging the selection is done in user space
Signed-off-by: Oliver Neukum <oneukum@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch fixes an endian-related error in ohci-hcd (detected by
sparse) and clarifies a comment explaining a peculiar locking
arrangement that sparse warns about.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch prepares ohci-hcd for being split up into a core
library and separate platform driver modules. A generic
ohci_hc_driver structure is created, containing all the "standard"
values, and a new mechanism is added whereby a driver module can
specify a set of overrides to those values. In addition the
ohci_restart(),ohci_suspend() and ohci_resume() routines need
to be EXPORTed for use by the drivers.
Added ohci_setip(() and ohci_start() routine for to start the generic
controller rather than each having its own idiosyncratic approach.
This allow to clean duplicated code in most of SOC driver
In V2:
-ohci_hcd_init() ohci_run() and ohci_stop() are not made non-static.
-Adds the ohci_setup() and ohci_start() routine.
In V3:
-purpose of ohci_setup() and ohci_start() function description written in the patch
description.
-ohci_init() are not made non-static but now called beginning of the ohci_restart().
-ohci_run() signature change reverted back.
-unrelated changes removed.
-duplicate comment line removed.
-inline ohci_suspend() and ohci_resume() is not needed so removed from ohci.h file.
In V4:
-ohci-init() EXPORTed because it is called by all bus glue modules.
-ohci-setup() removed from 1/2 added into 2/2 patch.
In V5:
-Again ohci_setup() is added and EXPORTed because to replace the ohci_init() from
all bus glues.
-ohci_init() is not made non-static function.
In V6:
-ohci_init() call is removed from ohci_quirk_nec_worker(), because it is already called in ohci_restart().
In V8:
-ohci_hcd_init() is called by ohci_setup() to make generic ohci initialization in all ohci drivers.
Signed-off-by: Manjunath Goudar <manjunath.goudar@linaro.org>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch (as1675) removes the CONFIG_USB_SUSPEND option, essentially
replacing it everywhere with CONFIG_PM_RUNTIME (except for one place
in hub.c, where it is replaced with CONFIG_PM because the code needs
to be used in both runtime and system PM). The net result is code
shrinkage and simplification.
There's very little point in keeping CONFIG_USB_SUSPEND because almost
everybody enables it. The few that don't will find that the usbcore
module has gotten somewhat bigger and they will have to take active
measures if they want to prevent hubs from being runtime suspended.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
CC: Peter Chen <peter.chen@freescale.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Merge ohci_finish_controller_resume with ohci_resume as suggested by Alan
Stern. Since ohci_finish_controller_resume no longer exists, update the
various OHCI drivers to call ohci_resume() instead. Some drivers used to set
themselves the bit HCD_FLAG_HW_ACCESSIBLE, which is now handled by
ohci_resume().
Acked-by: Jingoo Han <jg1.han@samsung.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Signed-off-by: Florian Fainelli <florian@openwrt.org>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
ohci_finish_controller_resume() is intended to be used in platform specific
drivers ohci-*.c, included from ohci-hcd.c. Some of them don't actually use
ohci_finish_controller_resume(), so mark it as __maybe_unused.
Signed-off-by: Roland Stigge <stigge@antcom.de>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch (as1500) removes all uses of the objectionable hcd->state
variable from the ohci-hcd family of drivers. It is replaced by a
private ohci->rh_state field, just as in uhci-hcd and ehci-hcd.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
At least some OHCI hardware (such as the MCP89) fails to flag any change
in the host status register or the port status registers when receiving
a remote wakeup while in D3 state. This results in the controller being
resumed but no device state change being noticed, at which point the
controller is put back to sleep again. Since there doesn't seem to be any
reliable way to identify the state change, just unconditionally resume the
hub. It'll be put back to sleep in the near future anyway if there are no
active devices attached to it.
Signed-off-by: Matthew Garrett <mjg@redhat.com>
Cc: stable <stable@vger.kernel.org>
Cc: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Update the USB core to deal with USB 3.0 hubs. These hubs have a slightly
different hub descriptor than USB 2.0 hubs, with a fixed (rather than
variable length) size. Change the USB core's hub descriptor to have a
union for the last fields that differ. Change the host controller drivers
that access those last fields (DeviceRemovable and PortPowerCtrlMask) to
use the union.
Translate the new version of the hub port status field into the old
version that khubd understands. (Note: we need to fix it to translate the
roothub's port status once we stop converting it to USB 2.0 hub status
internally.)
Add new code to handle link state change status. Send out new control
messages that are needed for USB 3.0 hubs, like Set Hub Depth.
This patch is a modified version of the original patch submitted by John
Youn. It's updated to reflect the removal of the "bitmap" #define, and
change the hub descriptor accesses of a couple new host controller
drivers.
Signed-off-by: John Youn <johnyoun@synopsys.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Cc: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Cc: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
Cc: Tony Olech <tony.olech@elandigitalsystems.com>
Cc: "Robert P. J. Day" <rpjday@crashcourse.ca>
Cc: Max Vozeler <mvz@vozeler.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Cc: Rodolfo Giometti <giometti@linux.it>
Cc: Mike Frysinger <vapier@gentoo.org>
Cc: Anton Vorontsov <avorontsov@mvista.com>
Cc: Sebastian Siewior <bigeasy@linutronix.de>
Cc: Lothar Wassmann <LW@KARO-electronics.de>
Cc: Olav Kongas <ok@artecdesign.ee>
Cc: Martin Fuzzey <mfuzzey@gmail.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: David Brownell <dbrownell@users.sourceforge.net>
Using a #define to redefine a common variable name is a bad thing,
especially when the #define is in a header. include/linux/usb/hcd.h
redefined bitmap to DeviceRemovable to avoid typing a long field in the
hub descriptor. This has unintended side effects for files like
drivers/usb/core/devio.c that include that file, since another header
included after hcd.h has different variables named bitmap.
Remove the bitmap #define and replace instances of it in the host
controller code. Cleanup the spaces around function calls and square
brackets while we're at it.
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Cc: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Cc: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
Cc: Tony Olech <tony.olech@elandigitalsystems.com>
Cc: "Robert P. J. Day" <rpjday@crashcourse.ca>
Cc: Max Vozeler <mvz@vozeler.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Cc: Rodolfo Giometti <giometti@linux.it>
Cc: Mike Frysinger <vapier@gentoo.org>
Cc: Anton Vorontsov <avorontsov@mvista.com>
Cc: Sebastian Siewior <bigeasy@linutronix.de>
Cc: Lothar Wassmann <LW@KARO-electronics.de>
Cc: Olav Kongas <ok@artecdesign.ee>
Cc: Martin Fuzzey <mfuzzey@gmail.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: David Brownell <dbrownell@users.sourceforge.net>
This patch (as1394) adds code to ehci-hcd, ohci-hcd, and uhci-hcd for
automatically resuming the root hub when the controller is resumed, if
the root hub has a wakeup request pending on some port.
During resume from system sleep this doesn't matter, because the root
hubs will naturally be resumed along with every other device in the
system. However it _will_ matter for runtime PM: If the controller is
suspended and a remote wakeup request is received then the controller
will autoresume, but we need to ensure that the root hub also
autoresumes. Otherwise the wakeup request would be ignored, the
controller would go back to sleep, and the cycle would repeat a large
number of times (I saw this happen before the patch was written).
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch (as1393) converts several of the single-bit fields in
struct usb_hcd to atomic flags. This is for safety's sake; not all
CPUs can update bitfield values atomically, and these flags are used
in multiple contexts.
The flag fields that are set only during registration or removal can
remain as they are, since non-atomic accesses at those times will not
cause any problems.
(Strictly speaking, the authorized_default flag should become atomic
as well. I didn't bother with it because it gets changed only via
sysfs. It can be done later, if anyone wants.)
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch (as1371) fixes a small bug in ohci-hcd. The HCD already
knows how many ports the controller has; there's no need to go looking
at the root hub's usb_device structure to find out. Especially since
the root hub's maxchild value is set correctly only while the root hub
is bound to the hub driver.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch (as1149) fixes an obscure problem in OHCI polling. In the
current code, if the RHSC interrupt status flag turns on at a time
when RHSC interrupts are disabled, it will remain on forever:
The interrupt handler is the only place where RHSC status
gets turned back off;
The interrupt handler won't turn RHSC status off because it
doesn't turn off status flags if the corresponding interrupt
isn't enabled;
RHSC interrupts will never get enabled because
ohci_root_hub_state_changes() doesn't reenable RHSC if RHSC
status is on!
As a result we will continue polling indefinitely instead of reverting
to interrupt-driven operation, and the root hub will not autosuspend.
This particular sequence of events is not at all unusual; in fact
plugging a USB device into an OHCI controller will usually cause it to
occur.
Of course, this is a bug. The proper thing to do is to turn off RHSC
status just before reading the actual port status values. That way
either a port status change will be detected (if it occurs before the
status read) or it will turn RHSC back on. Possibly both, but that
won't hurt anything.
We can still check for systems in which RHSC is totally broken, by
re-reading RHSC after clearing it and before reading the port
statuses. (This re-read has to be done anyway, to post the earlier
write.) If RHSC is on but no port-change statuses are set, then we
know that RHSC is broken and we can avoid re-enabling it.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch (as1134) attempts to improve the way we handle OHCI
controllers with broken Root Hub Status Change interrupt support. In
these controllers the RHSC interrupt bit essentially never turns off,
making RHSC interrupts useless -- they have to remain permanently
disabled.
Such controllers should still be allowed to turn off their root hubs
when no devices are attached. Polling for new connections can
continue while the root hub is suspended. The patch implements this
feature. (It won't have much effect unless CONFIG_PM is enabled and
CONFIG_USB_SUSPEND is disabled, but since the overhead is very small
we may as well do it.)
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch (as1069c) changes the way OHCI root-hub status-change
interrupts are enabled. Currently a special HCD method,
hub_irq_enable(), is called when the hub driver is finished using a
root hub. This approach turns out to be subject to races, resulting
in unnecessary polling.
The patch does away with the method entirely. Instead, the driver
automatically enables the RHSC interrupt when no more status changes
are present. This scheme is safe with controllers using
level-triggered semantics for their interrupt flags.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch (as1114) fixes a problem that was revealed by an earlier
patch (as1069b). Some broken controllers seem never to turn off their
RHCS interrupt status bit, even when told to do so. As a result they
generate an interrupt storm and hang the system.
The patch avoids enabling RHSC interrupt requests when the RHCS status
bit is already set. This should have no adverse affects on normal
controllers, since they won't set the status bit until a root-hub
status change actually occurs, in which case we wouldn't enable RHSC
interrupt requests anyway -- we would wait until the status change had
been processed and cleared.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Tested by: Andrey Borzenkov <arvidjaar@mail.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This reverts commit e872154921.
Andrey Borzenkov reports that it resulted in a totally hung machine for
him when loading the OHCI driver. Extensive netconsole capture with
SysRq output shows that modprobe gets stuck in ohci_hub_status_data()
when probing and enabling the OHCI controller, see for example
http://lkml.org/lkml/2008/7/5/236
for an analysis.
The problem appears to be an interrupt flood triggered by the commit
that gets reverted, and Andrey confirmed that the revert makes things
work for him again.
Reported-and-tested-by: Andrey Borzenkov <arvidjaar@mail.ru>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Acked-by: David Brownell <david-b@pacbell.net>
Cc: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The patch (as1086) works around a bogus "uninitialized variable"
warning generated by some versions of GCC.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch (as1069b) changes the way OHCI root-hub status-change
interrupts are enabled. Currently a special HCD method,
hub_irq_enable(), is called when the hub driver is finished using a
root hub. This approach turns out to be subject to races, resulting
in unnecessary polling.
The patch does away with the method entirely. Instead, the driver
automatically enables the RHSC interrupt when no more status changes
are present. This scheme is safe with controllers using
level-triggered semantics for their interrupt flags.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch (as1068b) disables the RD interrupt flag when an OHCI root
hub is suspended with remote wakeup disabled. Although the spec
clearly states that this flag permits the controller to issue an
interrupt when a resume request from downstream is detected and not
when a local status change occurs, some controllers mistakenly use it
for both types of event.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
When a USB device is suspended, whether or not it is enabled for
remote wakeup depends on the device_may_wakeup() setting. The setting
is then saved in the do_remote_wakeup flag.
Later on, however, the device_may_wakeup() value can change because of
user activity. So when testing whether a suspended device is or
should be enabled for remote wakeup, we should always test
do_remote_wakeup instead of device_may_wakeup(). This patch (as1076)
makes that change for root hubs in several places.
The patch also adjusts uhci-hcd so that when an autostopped controller
is suspended, the remote wakeup setting agrees with the value recorded
in the root hub's do_remote_wakeup flag.
And the patch adjusts ehci-hcd so that wakeup events on selectively
suspended ports (i.e., the bus itself isn't suspended) don't turn on
the PME# wakeup signal.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Drivers in the ohci-hcd family should perform certain tasks whenever
their controller device is resumed. These include checking for loss
of power during suspend, turning on port power, and enabling interrupt
requests.
Until now these jobs have been carried out when the root hub is
resumed, not when the controller is. Many drivers work around the
resulting awkwardness by automatically resuming their root hub
whenever the controller is resumed. But this is wasteful and
unnecessary.
To simplify the situation, this patch (as1066) adds a new core
routine, ohci_finish_controller_resume(), which can be used by all the
OHCI-variant drivers. They can call the new routine instead of
resuming their root hubs. And ohci-pci.c can call it instead of using
its own special-purpose handler.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This limits how long the OHCI port reset loop waits for the hardware
to do its job, if the controller either (a) dies, or (b) can't finish
the reset. Such limits are always a good idea.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch fixes a silicon bug in some NEC OHCI chips. The bug appears
at random times and is very, very difficult to reproduce. Without the
following patch, Linux would shut the chip and its associated devices
down. In Apple PowerBooks this leads to an unusable keyboard and mouse
(SSH still working). The idea of restarting the chip is taken from
public Darwin code.
Signed-off-by: Michael Hanselmann <linux-kernel@hansmi.ch>
Cc: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch (as901) fixes an oversight in ohci-hcd. The
hub_status_data routine must not try to access the controller's
memory-mapped registers if the controller is in a low-power state;
such attempts will cause a crash on some architectures (such as PPC).
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
I have found a problem where the root_port_reset() goes into an infinite
loop and stalls the kernel.
This happens when a hardware fault inside the machine occurs during a small
timing window. In case of USB device connection, if a USB device responds to
hcd_submit_urb(), and later the controller fails before root_port_reset(),
root_port_reset() will loop infinitely because ohci_readl() will always
return "-1". Such a failure can include ejecting a CardBus OHCI controller.
The probability of this problem is low, but it will increase if PnP type
usage is frequent. The attached patch can solve this problem and I believe
that it is better to fix this problem.
Signed-off-by: Takamasa Ohtake <ohtake-txa@necst.nec.co.jp>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This is an OHCI cleanup patch ... it removes a lot of erroneous whitespace
(space before tab, at end of line) as well as the obsolete inline changelog.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Unlike UHCI, OHCI does not exert any DMA load on the system when no
devices are connected. Consequently there is no advantage to doing
an autostop other than the power savings, so we shouldn't compile the
necessary code unless CONFIG_PM is enabled.
This patch (as820) makes the root-hub suspend and resume routines
conditional on CONFIG_PM. It also prevents autostop from activating
if the device_may_wakeup flag isn't set; some people use this flag to
alert the driver about Resume-Detect bugs in the hardware.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
All the other root-hub suspend or resume log messages, in ohci-hcd or
any of the other host controller drivers, use the debug priority
level. This patch (as815) makes the one single exception behave like
all the rest.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch (as811) removes some stale testing code from the root-hub
resume routine in ohci-hcd. It also adds a spin_lock_irq() call that
inadvertently got left out of an error pathway.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch (as808b) moves the Root Hub Status Change interrupt-disable
code in ohci-hcd back into the interrupt handler proper, to avoid the
chance of adverse interactions with mediocre hardware implementations.
It also deletes the root-hub status timer from within the interrupt-enable
routine. There's no need to poll for status any more once interrupts are
re-enabled.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch (as806) fixes a compiler warning when ohci-hcd is built
with CONFIG_PM turned off.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
When a suspended OHCI controller sees a port's status change, it sets
both the Root-Hub-Status-Change and the Resume-Detect bits in the
Interrupt Status register. Processing both these bits, the driver
tries to resume the root hub twice!
This patch (as807) fixes the bug by ignoring RD if RHSC is set. It
also prints a slightly more informative log message when a
remote-wakeup event occurs.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch (as822) prevents the OHCI autostop mechanism from kicking in
if the root hub is not able or not allowed to issue wakeup requests.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Maintain a per-CPU global "struct pt_regs *" variable which can be used instead
of passing regs around manually through all ~1800 interrupt handlers in the
Linux kernel.
The regs pointer is used in few places, but it potentially costs both stack
space and code to pass it around. On the FRV arch, removing the regs parameter
from all the genirq function results in a 20% speed up of the IRQ exit path
(ie: from leaving timer_interrupt() to leaving do_IRQ()).
Where appropriate, an arch may override the generic storage facility and do
something different with the variable. On FRV, for instance, the address is
maintained in GR28 at all times inside the kernel as part of general exception
handling.
Having looked over the code, it appears that the parameter may be handed down
through up to twenty or so layers of functions. Consider a USB character
device attached to a USB hub, attached to a USB controller that posts its
interrupts through a cascaded auxiliary interrupt controller. A character
device driver may want to pass regs to the sysrq handler through the input
layer which adds another few layers of parameter passing.
I've build this code with allyesconfig for x86_64 and i386. I've runtested the
main part of the code on FRV and i386, though I can't test most of the drivers.
I've also done partial conversion for powerpc and MIPS - these at least compile
with minimal configurations.
This will affect all archs. Mostly the changes should be relatively easy.
Take do_IRQ(), store the regs pointer at the beginning, saving the old one:
struct pt_regs *old_regs = set_irq_regs(regs);
And put the old one back at the end:
set_irq_regs(old_regs);
Don't pass regs through to generic_handle_irq() or __do_IRQ().
In timer_interrupt(), this sort of change will be necessary:
- update_process_times(user_mode(regs));
- profile_tick(CPU_PROFILING, regs);
+ update_process_times(user_mode(get_irq_regs()));
+ profile_tick(CPU_PROFILING);
I'd like to move update_process_times()'s use of get_irq_regs() into itself,
except that i386, alone of the archs, uses something other than user_mode().
Some notes on the interrupt handling in the drivers:
(*) input_dev() is now gone entirely. The regs pointer is no longer stored in
the input_dev struct.
(*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does
something different depending on whether it's been supplied with a regs
pointer or not.
(*) Various IRQ handler function pointers have been moved to type
irq_handler_t.
Signed-Off-By: David Howells <dhowells@redhat.com>
(cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
Shigeru Yoshida