pandora-kernel.git
13 years agousb: octeon2-common.c: Configure ports for proper electrical characteristics.
David Daney [Wed, 27 Apr 2011 17:54:21 +0000 (10:54 -0700)]
usb: octeon2-common.c: Configure ports for proper electrical characteristics.

Additional PHY tuning is needed to obtain compliant 'eye' diagram
electrical characteristics.

Signed-off-by: David Daney <ddaney@caviumnetworks.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: octeon2-common: Don't reinitialize the clocks.
David Daney [Wed, 27 Apr 2011 17:54:20 +0000 (10:54 -0700)]
USB: octeon2-common: Don't reinitialize the clocks.

The UCTL clock initialization will cause the ehci and ohci blocks to
become inoperable if the clocks are reinitialized.

Check to see if the clocks have already been initialized.

Also use a mutex to protect the clock initialization code so that
there can be no attempt to use the clocks before they are fully
configured.

Signed-off-by: David Daney <ddaney@caviumnetworks.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoMerge branch 'for-usb-next' of git+ssh://master.kernel.org/pub/scm/linux/kernel/git...
Greg Kroah-Hartman [Tue, 3 May 2011 00:05:19 +0000 (17:05 -0700)]
Merge branch 'for-usb-next' of git+ssh:///linux/kernel/git/sarah/xhci into usb-next

* 'for-usb-next' of git+ssh://master.kernel.org/pub/scm/linux/kernel/git/sarah/xhci:
  xhci 1.0: Set transfer burst last packet count field.
  xhci 1.0: Set transfer burst count field.
  xhci 1.0: Update TD size field format.
  xhci 1.0: Only interrupt on short packet for IN EPs.
  xhci: Remove sparse warning about cmd_status.
  usbcore: warm reset USB3 port in SS.Inactive state
  usbcore: Refine USB3.0 device suspend and resume
  xHCI: report USB3.0 portstatus comply with USB3.0 specification
  xHCI: Set link state support
  xHCI: Clear link state change support
  xHCI: warm reset support
  usb/ch9: use proper endianess for wBytesPerInterval
  xhci: Remove recursive call to xhci_handle_event
  xhci: Add an assertion to check for virt_dev=0 bug.
  xhci: Add rmb() between reading event validity & event data access.
  xhci: Make xHCI driver endian-safe

13 years agousb/isp1760: Improve urb queueing, get rid of BUG():s in normal code paths
Arvid Brodin [Tue, 26 Apr 2011 19:48:30 +0000 (21:48 +0200)]
usb/isp1760: Improve urb queueing, get rid of BUG():s in normal code paths

This patch replaces the code that handles qtds. Intead of directly allocating
chip mem and chip slot, enqueue the transfer in a list of queue heads. Use
a centralized function enqueue_qtds() to prioritize and enqueue transfers.

This removes all of the interrupt context BUG() calls when out of chip
mem or transfer slots. It also makes it possible to efficiently use the
dual-port mem on the chip for double-buffered transfers, which improve
transfer times to/from/between usb sticks by about 40 % on my HW.

With this patch it should also be possible to handle qtd scheduling outside
of the interrupt handler, for significantly improved kernel latency. I have
not implemented this since there are some locking issues which I haven't
had time to look at.

Signed-off-by: Arvid Brodin <arvid.brodin@enea.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb/isp1760: Report correct urb status after unlink
Arvid Brodin [Tue, 26 Apr 2011 19:46:47 +0000 (21:46 +0200)]
usb/isp1760: Report correct urb status after unlink

This fixes a bug in my previous (2.6.38) patch series which caused
urb->status value to be wrong after unlink (broke usbtest 11, 12).

Signed-off-by: Arvid Brodin <arvid.brodin@enea.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb/isp1760: Remove unneeded OR map and HcBufferStatus code
Arvid Brodin [Tue, 26 Apr 2011 19:48:02 +0000 (21:48 +0200)]
usb/isp1760: Remove unneeded OR map and HcBufferStatus code

Since we always set the OR flag for each transfer, we can just as well set all
these bits to 1 at init and be done with it. Also, HcBufferStatus can be set
at init as per the ISP1761 datasheet page 47 with no loss of performance.

Signed-off-by: Arvid Brodin <arvid.brodin@enea.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb/isp1760: Clean up urb enqueueing
Arvid Brodin [Tue, 26 Apr 2011 19:47:37 +0000 (21:47 +0200)]
usb/isp1760: Clean up urb enqueueing

This collects urb enqueue code that was spread out all over the place
into a couple of more readable functions.

Signed-off-by: Arvid Brodin <arvid.brodin@enea.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb/isp1760: Remove false error printout
Arvid Brodin [Tue, 26 Apr 2011 19:47:12 +0000 (21:47 +0200)]
usb/isp1760: Remove false error printout

This removes the "qh is 0" printout. qh == NULL if the urb has
been unlinked, so this condition is normal.

Signed-off-by: Arvid Brodin <arvid.brodin@enea.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: s3c-hsotg: Add copyright string
Anton Tikhomirov [Thu, 21 Apr 2011 08:06:43 +0000 (17:06 +0900)]
USB: s3c-hsotg: Add copyright string

Signed-off-by: Anton Tikhomirov <av.tikhomirov@samsung.com>
Cc: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Kukjin Kim <kgene.kim@samsung.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: s3c-hsotg: Fix hang up after reset
Anton Tikhomirov [Thu, 21 Apr 2011 08:06:42 +0000 (17:06 +0900)]
USB: s3c-hsotg: Fix hang up after reset

When File Storage gadget receives SET CONFIGURATION request it tries
to cancel all pending transfers. If some request is in progress,
gadget waits for its completion. This commit allows gadget to dequeue
invalid requests in progress left after reset.

Signed-off-by: Anton Tikhomirov <av.tikhomirov@samsung.com>
Reviewed-by: Kyoungil Kim<ki0351.kim@samsung.com>
Cc: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Kukjin Kim <kgene.kim@samsung.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: s3c-hsotg: Fix stall condition processing
Anton Tikhomirov [Thu, 21 Apr 2011 08:06:41 +0000 (17:06 +0900)]
USB: s3c-hsotg: Fix stall condition processing

The following should be done for requests after endpoint stall
condition is cleared:
1) 'in progress' request (if any) should be completed since
   Tx FIFO was flushed;
2) next request from queue (if any) should be started.

This commit does that.

Additionally set/clear stall condition code is fixed.

Signed-off-by: Anton Tikhomirov <av.tikhomirov@samsung.com>
Reviewed-by: Kyoungil Kim<ki0351.kim@samsung.com>
Cc: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Kukjin Kim <kgene.kim@samsung.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: s3c-hsotg: Fix control request processing
Anton Tikhomirov [Thu, 21 Apr 2011 08:06:40 +0000 (17:06 +0900)]
USB: s3c-hsotg: Fix control request processing

UDC driver does not need to generate reply to host if request is
delivered to gadget. This is gadget's responsibility. This commit
fixes that.

Signed-off-by: Anton Tikhomirov <av.tikhomirov@samsung.com>
Reviewed-by: Kyoungil Kim<ki0351.kim@samsung.com>
Cc: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Kukjin Kim <kgene.kim@samsung.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: s3c-hsotg: Fix interrupt cleaning code
Anton Tikhomirov [Thu, 21 Apr 2011 08:06:39 +0000 (17:06 +0900)]
USB: s3c-hsotg: Fix interrupt cleaning code

This commit does the following:
1) clears all pending interrupts before unmasking;
2) clears interrupts as soon as possible to avoid missing
   next coming that may occur during handling;
3) removes ineffective interrupt cleaning code.

Signed-off-by: Anton Tikhomirov <av.tikhomirov@samsung.com>
Reviewed-by: Kyoungil Kim<ki0351.kim@samsung.com>
Cc: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Kukjin Kim <kgene.kim@samsung.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: s3c-hsotg: Fix core reset
Anton Tikhomirov [Thu, 21 Apr 2011 08:06:38 +0000 (17:06 +0900)]
USB: s3c-hsotg: Fix core reset

This patch fixes code responsible for core reset.

Signed-off-by: Anton Tikhomirov <av.tikhomirov@samsung.com>
Reviewed-by: Kyoungil Kim<ki0351.kim@samsung.com>
Cc: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Kukjin Kim <kgene.kim@samsung.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: extend ehci-fsl and fsl_udc_core driver for OTG operation
Anatolij Gustschin [Mon, 18 Apr 2011 20:02:00 +0000 (22:02 +0200)]
USB: extend ehci-fsl and fsl_udc_core driver for OTG operation

Signed-off-by: Anatolij Gustschin <agust@denx.de>
Cc: Li Yang <leoli@freescale.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: add Freescale USB OTG Transceiver driver
Li Yang [Mon, 18 Apr 2011 20:01:59 +0000 (22:01 +0200)]
USB: add Freescale USB OTG Transceiver driver

Slightly reworked and cleaned up driver from Freescale LTIB
for MPC5121E. The driver has been ported to the current kernel,
proc interface "/proc/driver/fsl_usb2_otg" has been replaced by
sysfs interface.

Signed-off-by: Li Yang <leoli@freescale.com>
Signed-off-by: Anatolij Gustschin <agust@denx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: fsl_udc_core: support device mode of MPC5121E DR USB Controller
Anatolij Gustschin [Mon, 18 Apr 2011 20:01:58 +0000 (22:01 +0200)]
USB: fsl_udc_core: support device mode of MPC5121E DR USB Controller

Extend the FSL UDC driver to support MPC5121E DR USB Controller
operation in device mode. Add MPC5121E specific init/uninit
at probe and remove and isolate system interface register accesses
when running on MPC5121E SoC, as these registers are not available
on this platform. This patch relies on previous patch for supporting
big endian registers and descriptors access in the FSL UDC driver.
Additionally support endpoint FIFO status operation by providing
appropriate callback in endpoint ops structure.

Also flush cache for the req buffer used for GetStatus reply.
Without this, the correct reply to an endpoint GetStatus
is written to 'req', but doesn't make it out to the USB bus
since the buffer hasn't been flushed. This would cause the
USBCV Halt Endpoint test to fail (according to changelog in
Freescale LTIB driver code).

Signed-off-by: Anatolij Gustschin <agust@denx.de>
Cc: Li Yang <leoli@freescale.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: fsl_udc_core: prepare for SoCs with BE registers and descriptors
Anatolij Gustschin [Mon, 18 Apr 2011 20:01:57 +0000 (22:01 +0200)]
USB: fsl_udc_core: prepare for SoCs with BE registers and descriptors

On some SoCs, the USB controller registers and descriptors can be big
or little endian, depending on the version of the chip. In order to be
able to run the same kernel binary on different versions of an SoC, the
BE/LE decision must be made at run time. Provide appropriate register
and descriptor accessors which are configurable at run time using the
configuration flags from fsl_usb2_platform_data data structure.

This is in preparation for adding support for MPC5121E DR USB2 Controller
to the FSL UDC driver.

Signed-off-by: Anatolij Gustschin <agust@denx.de>
Cc: Li Yang <leoli@freescale.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: fix build of FSL MPH DR OF platform driver
Anatolij Gustschin [Mon, 18 Apr 2011 20:01:56 +0000 (22:01 +0200)]
USB: fix build of FSL MPH DR OF platform driver

When only FSL USB gadget driver is selected in the kernel
configuration the MPH DR OF driver for creation of FSL
USB platform devices from device tree won't be built.
As a result no USB platform devices for MPH DR USB controller
will be created at run time and no probing will be done in
the fsl_udc_core driver.

Add an entry to the Makefile to build the MPH DR OF
platform driver if CONFIG_USB_FSL_MPH_DR_OF is defined.

Signed-off-by: Anatolij Gustschin <agust@denx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: ehci-fsl: add MPC5121E specific suspend and resume
Anatolij Gustschin [Mon, 18 Apr 2011 20:01:55 +0000 (22:01 +0200)]
USB: ehci-fsl: add MPC5121E specific suspend and resume

Signed-off-by: Anatolij Gustschin <agust@denx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoxhci 1.0: Set transfer burst last packet count field.
Sarah Sharp [Wed, 20 Apr 2011 00:43:33 +0000 (17:43 -0700)]
xhci 1.0: Set transfer burst last packet count field.

The xHCI 1.0 specification defines a new isochronous TRB field, called
transfer burst last packet count (TBLPC).  This field defines the number
of packets in the last "burst" of packets in a TD.  Only SuperSpeed
endpoints can handle more than one burst, so this is set to the number for
packets in a TD for all non-SuperSpeed devices (minus one, since the field
is zero based).

This patch should have no effect on host controllers that don't advertise
the xHCI 1.0 (0x100) version number in their hci_version field.

Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
13 years agoxhci 1.0: Set transfer burst count field.
Sarah Sharp [Fri, 8 Apr 2011 16:37:29 +0000 (09:37 -0700)]
xhci 1.0: Set transfer burst count field.

The xHCI 1.0 specification adds a new field to the fourth dword in an
isochronous TRB: the transfer burst count (TBC).  This field is only
non-zero for SuperSpeed devices.  Each SS endpoint sets the bMaxBurst
field in the SuperSpeed endpoint companion descriptor, which indicates how
many max-packet-sized "bursts" it can handle in one service interval.  The
device driver may choose to burst less max packet sized chunks each
service interval (which is defined by one TD).  The xHCI driver indicates
to the host controller how many bursts it needs to schedule through the
transfer burst count field.

This patch will only effect xHCI hosts that advertise 1.0 support (0x100)
in the HCI version field of their capabilities register.

Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
13 years agoxhci 1.0: Update TD size field format.
Sarah Sharp [Fri, 1 Apr 2011 21:01:30 +0000 (14:01 -0700)]
xhci 1.0: Update TD size field format.

The xHCI 1.0 specification changes the format of the TD size field in
Normal and Isochronous TRBs.  The field in control TRBs is still set to
reserved zero.  Instead of representing the number of bytes left to
transfer in the TD (including the current TRB's buffer), it now represents
the number of packets left to transfer (*not* including this TRB).

See section 4.11.2.4 of the xHCI 1.0 specification for details.  The math
is basically copied straight from there.

Create a new function, xhci_v1_0_td_remainder(), that should be called for
all xHCI 1.0 host controllers.  The field location and maximum value is
still the same, so reuse the old function, xhci_td_remainder(), to handle
the bit shifting.

Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
13 years agoxhci 1.0: Only interrupt on short packet for IN EPs.
Sarah Sharp [Wed, 23 Mar 2011 23:26:26 +0000 (16:26 -0700)]
xhci 1.0: Only interrupt on short packet for IN EPs.

It doesn't make sense to set the interrupt on short packet (TRB_ISP) flag
for TRBs queued to endpoints that only receive packets from the host
controller (i.e. OUT endpoints).  Packets can only be short when they are
sent from a USB device.  Plus, the xHCI 1.0 specification forbids setting
the flag for anything but IN endpoints.

While we're at it, remove some of my snide remarks about the inefficiency
of event data TRBs.

Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
13 years agoxhci: Remove sparse warning about cmd_status.
Sarah Sharp [Thu, 28 Apr 2011 19:23:23 +0000 (12:23 -0700)]
xhci: Remove sparse warning about cmd_status.

Sparse complains about the arguments to xhci_evaluate_context_result() and
xhci_configure_endpoint_result():

  CHECK   drivers/usb/host/xhci.c
drivers/usb/host/xhci.c:1647:53: warning: incorrect type in argument 3 (different signedness)
drivers/usb/host/xhci.c:1647:53:    expected int *cmd_status
drivers/usb/host/xhci.c:1647:53:    got unsigned int [usertype] *[assigned] cmd_status
drivers/usb/host/xhci.c:1648:50: warning: incorrect type in argument 3 (different signedness)
drivers/usb/host/xhci.c:1648:50:    expected int *cmd_status
drivers/usb/host/xhci.c:1648:50:    got unsigned int [usertype] *[assigned] cmd_status

The command status is taken from the command completion event TRB, and
will always be a positive number.  Change the signature of
xhci_evaluate_context_result() and xhci_configure_endpoint_result() to
take a u32 for cmd_status.

Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
13 years agousbcore: warm reset USB3 port in SS.Inactive state
Andiry Xu [Wed, 27 Apr 2011 10:07:54 +0000 (18:07 +0800)]
usbcore: warm reset USB3 port in SS.Inactive state

Some USB3.0 devices go to SS.Inactive state when hot plug to USB3 ports.
Warm reset the port to transition it to U0 state.

This patch fixes the issue that Kingston USB3.0 flash drive can not be
recognized when hot plug to USB3 port.

Signed-off-by: Andiry Xu <andiry.xu@amd.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
13 years agousbcore: Refine USB3.0 device suspend and resume
Andiry Xu [Wed, 27 Apr 2011 10:07:50 +0000 (18:07 +0800)]
usbcore: Refine USB3.0 device suspend and resume

In the past, we use USB2.0 request to suspend and resume a USB3.0 device.
Actually, USB3.0 hub does not support Set/Clear PORT_SUSPEND request,
instead, it uses Set PORT_LINK_STATE request. This patch makes USB3.0 device
suspend/resume comply with USB3.0 specification.

This patch fixes the issue that USB3.0 device can not be suspended when
connected to a USB3.0 external hub.

Signed-off-by: Andiry Xu <andiry.xu@amd.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
13 years agoxHCI: report USB3.0 portstatus comply with USB3.0 specification
Andiry Xu [Wed, 27 Apr 2011 10:07:43 +0000 (18:07 +0800)]
xHCI: report USB3.0 portstatus comply with USB3.0 specification

USB3.0 specification has different wPortStatus and wPortChange definitions
from USB2.0 specification. Since USB3 root hub and USB2 root hub are split
now and USB3 hub only has USB3 protocol ports, we should modify the
portstatus and portchange report of USB3 ports to comply with USB3.0
specification.

Signed-off-by: Andiry Xu <andiry.xu@amd.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
13 years agoxHCI: Set link state support
Andiry Xu [Wed, 27 Apr 2011 10:07:39 +0000 (18:07 +0800)]
xHCI: Set link state support

This patch adds support for Set Port Feature(PORT_LINK_STATE) request.

The most significant byte (bits 15..8) of the wIndex field specifies
the U state the host software wants to put the link connected to the
port into. This request is only valid when the PORT_ENABLE bit is set
and the PORT_LINK_STATE should not be above value '5' (Rx.Detect).

This request will be later used to replace the set/clear suspend USB3
protocol ports in hub driver.

Signed-off-by: Andiry Xu <andiry.xu@amd.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
13 years agoxHCI: Clear link state change support
Andiry Xu [Wed, 27 Apr 2011 10:07:35 +0000 (18:07 +0800)]
xHCI: Clear link state change support

This patch adds support for Clear Port Feature(C_PORT_LINK_STATE)
request from usbcore.

Signed-off-by: Andiry Xu <andiry.xu@amd.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
13 years agoxHCI: warm reset support
Andiry Xu [Wed, 27 Apr 2011 10:07:29 +0000 (18:07 +0800)]
xHCI: warm reset support

This patch adds warm reset support to xhci hub control.
It handles Set Port Feature(BH_PORT_RESET) and Clear Port Feature
(C_BH_PORT_RESET) request from usbcore.
Note warm reset is called BH reset some places in USB3.0 specification.

Signed-off-by: Andiry Xu <andiry.xu@amd.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
13 years agousb/ch9: use proper endianess for wBytesPerInterval
Sebastian Andrzej Siewior [Mon, 11 Apr 2011 18:19:12 +0000 (20:19 +0200)]
usb/ch9: use proper endianess for wBytesPerInterval

while going through Tatyana's changes for the gadget framework I noticed
that this type is not defined as __le16.

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
13 years agoxhci: Remove recursive call to xhci_handle_event
Matt Evans [Tue, 29 Mar 2011 02:41:02 +0000 (13:41 +1100)]
xhci: Remove recursive call to xhci_handle_event

Make the caller loop while there are events to handle, instead.

Signed-off-by: Matt Evans <matt@ozlabs.org>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
13 years agoxhci: Add an assertion to check for virt_dev=0 bug.
Matt Evans [Tue, 29 Mar 2011 02:40:56 +0000 (13:40 +1100)]
xhci: Add an assertion to check for virt_dev=0 bug.

During a "plug-unplug" stress test on an NEC xHCI card, a null pointer
dereference was observed.  xhci_address_device() dereferenced a null
virt_dev (possibly an erroneous udev->slot_id?); this patch adds a WARN_ON &
message to aid debug if it can be recreated.

Signed-off-by: Matt Evans <matt@ozlabs.org>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
13 years agoxhci: Add rmb() between reading event validity & event data access.
Matt Evans [Tue, 29 Mar 2011 02:40:51 +0000 (13:40 +1100)]
xhci: Add rmb() between reading event validity & event data access.

On weakly-ordered systems, the reading of an event's content must occur
after reading the event's validity.

Signed-off-by: Matt Evans <matt@ozlabs.org>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
13 years agoxhci: Make xHCI driver endian-safe
Matt Evans [Tue, 29 Mar 2011 02:40:46 +0000 (13:40 +1100)]
xhci: Make xHCI driver endian-safe

This patch changes the struct members defining access to xHCI device-visible
memory to use __le32/__le64 where appropriate, and then adds swaps where
required.  Checked with sparse that all accesses are correct.

MMIO accesses use readl/writel so already are performed LE, but prototypes
now reflect this with __le*.

There were a couple of (debug) instances of DMA pointers being truncated to
32bits which have been fixed too.

Signed-off-by: Matt Evans <matt@ozlabs.org>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
13 years agousb: factor out state_string() on otg drivers
Anatolij Gustschin [Fri, 15 Apr 2011 14:18:38 +0000 (16:18 +0200)]
usb: factor out state_string() on otg drivers

Provide common otg_state_string() and use
it in drivers.

Signed-off-by: Anatolij Gustschin <agust@denx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb/usbtest: fix test10 on superpseed
Sebastian Andrzej Siewior [Thu, 14 Apr 2011 14:17:21 +0000 (16:17 +0200)]
usb/usbtest: fix test10 on superpseed

dummy_hcd + g_zero:

|./testusb -a -s 1024 -v 1024 -t 10
| usbtest 2-1:3.0: subtest 14 error, status 0

with patch:
  ./testusb -a -t 10
|unknown speed   /proc/bus/usb/001/002
| dummy_udc dummy_udc: disabled ep-a
| dummy_udc dummy_udc: disabled ep-b
| dummy_udc dummy_udc: enabled ep-a (ep1in-bulk) maxpacket 1024
| dummy_udc dummy_udc: enabled ep-b (ep2out-bulk) maxpacket 1024
| zero gadget: source/sink enabled
| usbtest 1-1:3.0: TEST 10:  queue 32 control calls, 1000 times
| dummy_hcd dummy_hcd: timer fired with no URBs pending?
|/proc/bus/usb/001/002 test 10,    0.022370 secs

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb/usbtest: print super on super speed
Sebastian Andrzej Siewior [Thu, 14 Apr 2011 13:45:42 +0000 (15:45 +0200)]
usb/usbtest: print super on super speed

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Reviewed-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoxhci-hcd: Include <linux/slab.h> in xhci-pci.c
Ben Hutchings [Mon, 25 Apr 2011 15:54:28 +0000 (16:54 +0100)]
xhci-hcd: Include <linux/slab.h> in xhci-pci.c

Commit b02d0ed677acb3465e7600366f2353413bf24074 ('xhci: Change
hcd_priv into a pointer') added calls to kzalloc() and kfree() in
xhci-pci.c.  On most architectures <linux/slab.h> is indirectly
included, but on some it is not.

Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Cc: Sarah Sharp <sarah.a.sharp@linux.intel.com>,
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb: core: Change usb_create_sysfs_intf_files()' return type to void
Michal Nazarewicz [Thu, 14 Apr 2011 15:47:09 +0000 (17:47 +0200)]
usb: core: Change usb_create_sysfs_intf_files()' return type to void

The usb_create_sysfs_intf_files() function always returned zero even
if it failed to create sysfs fails.  Since this is a desired behaviour
there is no need to return return code at all.  This commit changes
function's return type (form int) to void.

Signed-off-by: Michal Nazarewicz <mina86@mina86.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: add queued-unlinks test case to usbtest driver
Alan Stern [Thu, 14 Apr 2011 15:21:04 +0000 (11:21 -0400)]
USB: add queued-unlinks test case to usbtest driver

This patch (as1452b) adds a new test case to the usbtest driver.  Test
24 exercises the unlink-from-queue pathways in the host.  It queues a
user-specified number of bulk-OUT URBs of user-specified size, unlinks
the fourth- and second-from-last URBs in the queue, and then waits to
see if all the URBs complete in the expected way (except of course
that the unlinked URBs might complete normally, if they weren't
unlinked soon enough).

This new test has confirmed the existence of a bug in the ehci-hcd
driver, to be fixed by a separate patch.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: ehci: tegra: fix USB1 port reset issue
Jim Lin [Sun, 17 Apr 2011 08:58:25 +0000 (11:58 +0300)]
USB: ehci: tegra: fix USB1 port reset issue

Tegra USB1 port needs to issue Port Reset twice internally, otherwise it
fails to enumerate devices attached to it

Signed-off-by: Jim Lin <jilin@nvidia.com>
Signed-off-by: Olof Johansson <olofj@chromium.org>
[ squash two patches into one and minor style cleanups ]
Signed-off-by: Mike Rapoport <mike@compulab.co.il>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: ohci-s3c2410: use resource_size()
Jingoo Han [Thu, 14 Apr 2011 12:09:16 +0000 (21:09 +0900)]
USB: ohci-s3c2410: use resource_size()

This patch uses the resource_size help function instead of
manually calculating the resource size. It can reduce the chance
of introducing off-by-one errors.

Signed-off-by: Jingoo Han <jg1.han@samsung.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: ohci-s3c2410: use __devinit and __devexit macros for probe and remove
Jingoo Han [Thu, 14 Apr 2011 12:09:37 +0000 (21:09 +0900)]
USB: ohci-s3c2410: use __devinit and __devexit macros for probe and remove

The __devinit and __devexit macros were added to probe and remove
functions. The macros move the probe and remove functions to the
devinit and devexit sections.

Signed-off-by: Jingoo Han <jg1.han@samsung.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb: gadget: storage_common: use kstrto*() [bug fix]
Michal Nazarewicz [Thu, 14 Apr 2011 09:55:43 +0000 (11:55 +0200)]
usb: gadget: storage_common: use kstrto*() [bug fix]

This commit fixes an embarrassing bug in the "storage_common:
use kstrto*()" patch which caused fsg_store_ro() to return
zero instead of the length of the consumed buffer.

Signed-off-by: Michal Nazarewicz <mina86@mina86.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb/dummy_hcd: don't probe for udc if hcd failed
Sebastian Andrzej Siewior [Fri, 15 Apr 2011 18:37:06 +0000 (20:37 +0200)]
usb/dummy_hcd: don't probe for udc if hcd failed

the_controller is allocated in dummy_hcd_probe() and is NULL if the
allocation failed. The probe function of the udc driver is dereferencing
this pointer and fault.
Alan Stern suggested to abort the dummy_hcd driver probing so the module
is not loaded. The is abort-on-error has been also added to the udc
driver.

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb/hcd: don't return 0 on error in usb_add_hcd()
Sebastian Andrzej Siewior [Thu, 14 Apr 2011 09:22:32 +0000 (11:22 +0200)]
usb/hcd: don't return 0 on error in usb_add_hcd()

If USB type detections fails, we run into default and return 0.

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb: renesas_usbhs: add autonomy mode
Kuninori Morimoto [Thu, 28 Apr 2011 07:41:20 +0000 (16:41 +0900)]
usb: renesas_usbhs: add autonomy mode

Current renesas_usbhs was designed to save power when USB is not connected.
And it assumed platform uses callback to notify connection/disconnection
by external interrupt.

But some SuperH / platform board doesn't have such feature.

This patch adds autonomy mode which detect USB connection/disconnection
by internal interrupt.
But power will be always ON when autonomy mode is selected.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb: renesas_usbhs: use delayed_work instead of work_struct
Kuninori Morimoto [Thu, 28 Apr 2011 07:41:14 +0000 (16:41 +0900)]
usb: renesas_usbhs: use delayed_work instead of work_struct

This delay is used to overjump debounce.

And, this patch also move usbhsc_drvcllbck_notify_hotplug to global,
because it will be called from other files.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb: renesas_usbhs: prevent NULL pointer crash
Kuninori Morimoto [Thu, 28 Apr 2011 07:41:07 +0000 (16:41 +0900)]
usb: renesas_usbhs: prevent NULL pointer crash

usbhs_status_get_each_irq/usbhs_irq_callback_update might be called
with mod == NULL

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb: renesas_usbhs: add power control function
Kuninori Morimoto [Thu, 28 Apr 2011 07:41:02 +0000 (16:41 +0900)]
usb: renesas_usbhs: add power control function

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb: renesas_usbhs: move pdev_to_priv to global
Kuninori Morimoto [Thu, 28 Apr 2011 07:40:54 +0000 (16:40 +0900)]
usb: renesas_usbhs: move pdev_to_priv to global

usbhs_pdev_to_priv function will be used in other files.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb: renesas_usbhs: add error reason for usbhs_pipe_malloc
Kuninori Morimoto [Thu, 21 Apr 2011 05:10:24 +0000 (14:10 +0900)]
usb: renesas_usbhs: add error reason for usbhs_pipe_malloc

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb: renesas_usbhs: remove callback when module removed.
Kuninori Morimoto [Thu, 21 Apr 2011 05:10:16 +0000 (14:10 +0900)]
usb: renesas_usbhs: remove callback when module removed.

The callback function which is called from platform must be removed
if module removed.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb: renesas_usbhs: return error if wrong recip request
Kuninori Morimoto [Thu, 21 Apr 2011 05:10:12 +0000 (14:10 +0900)]
usb: renesas_usbhs: return error if wrong recip request

There are some USB Host which doesn't notice disconnection at once.
And it might try some request after reconnection with old settings.
Current renesas_usbhs will crash in such case.
This patch prevent this issue.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb: renesas_usbhs: modify spinlock method
Kuninori Morimoto [Thu, 21 Apr 2011 05:10:08 +0000 (14:10 +0900)]
usb: renesas_usbhs: modify spinlock method

Current renesas_usbhs driver was using spin_trylock to avoid
dead lock / nest lock.
But acording to CONFIG_DEBUG_SPINLOCK, it is BUG under UP environment.
This patch add usbhsg_trylock to avoid this issue.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb: renesas_usbhs: don't re-allocation pipe buffer
Kuninori Morimoto [Tue, 26 Apr 2011 00:21:35 +0000 (09:21 +0900)]
usb: renesas_usbhs: don't re-allocation pipe buffer

Because pipe buffer allocation is very picky and difficult,
current renesas_usbhs driver is not caring pipe re-allocation.

In this situation, driver will create new pipe without caring old pipe
if "usbhsg_ep_enable" is called after "usbhsg_ep_disable" on current driver.
This mean the limited pipe and buffer will be used as waste.
But it is possible to re-use same buffer to same pipe.

By this patch, driver will initialize pipe when it detected new connection or
new gadget, and doesn't try re-allocation for same pipe in above case.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb: renesas_usbhs: modify fifo clear timing
Kuninori Morimoto [Thu, 21 Apr 2011 05:09:58 +0000 (14:09 +0900)]
usb: renesas_usbhs: modify fifo clear timing

Pipe buffer should be cleaned before using it,
but should NOT be cleaned in pipe "prepare" function.
Because the pipe might be working in such timing.
This patch fixup this issue.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: g_printer required set interface request
Jonas Andersson [Mon, 21 Mar 2011 13:54:09 +0000 (14:54 +0100)]
USB: g_printer required set interface request

g_printer reqiured "set interface" request from host. Not all hosts send
this request.  This patch enable the interface when it get "set
configuration" request from host.

Signed-off-by: Jonas Andersson <jonas@microbit.se>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: dbpg gadget: dont mask out direction bit
Sven Schnelle [Wed, 23 Mar 2011 14:24:56 +0000 (15:24 +0100)]
USB: dbpg gadget: dont mask out direction bit

Stripping the direction bit off will produce an
invalid descriptor.

Signed-off-by: Sven Schnelle <svens@stackframe.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: dbgp gadget: set MaxpacketSize0
Sven Schnelle [Wed, 23 Mar 2011 20:24:55 +0000 (21:24 +0100)]
USB: dbgp gadget: set MaxpacketSize0

The current code doesn't set it, so linux complains about
it when connected, and ignores the device:

[104611.068082] usb 1-5: new high speed USB device using ehci_hcd and address 127
[104611.088368] usb 1-5: Invalid ep0 maxpacket: 0

Signed-off-by: Sven Schnelle <svens@stackframe.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: dbgp gadget: fix return value of dbgp_setup
Sven Schnelle [Wed, 23 Mar 2011 20:36:17 +0000 (21:36 +0100)]
USB: dbgp gadget: fix return value of dbgp_setup

Current code returns 0 even if it can't handle the request.
This leads to timeouts when an unhandled request is sent:

Bus 001 Device 003: ID 0525:c0de Netchip Technology, Inc.
Device Descriptor:
[..]
can't get device qualifier: Connection timed out
[..]

change the code to return EOPNOTSUPP in such cases.

Signed-off-by: Sven Schnelle <svens@stackframe.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb: renesas_usbhs: Driver is depend on SuperH
Kuninori Morimoto [Fri, 15 Apr 2011 05:19:46 +0000 (14:19 +0900)]
usb: renesas_usbhs: Driver is depend on SuperH

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb: renesas_usbhs: fixup cast warning
Kuninori Morimoto [Thu, 14 Apr 2011 04:57:06 +0000 (13:57 +0900)]
usb: renesas_usbhs: fixup cast warning

This patch fixup cast warning on 64 bit compiler

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reported-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: documentation update for the pre_reset method
Alan Stern [Thu, 28 Apr 2011 14:45:24 +0000 (10:45 -0400)]
USB: documentation update for the pre_reset method

This patch (as1459) updates the documentation for the pre_reset method
in struct usb_driver.  When a driver is notified of an impending
reset, it must cancel all outstanding I/O and not start any new I/O
until it has been notified that the reset is complete.

As far as I know, most existing drivers that implement pre_reset do
this now.  The major exceptions appear to be the SpeedTouch and
CDC-WDM drivers.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoBind only modem AT command endpoint to option module.
Marius B. Kotsbak [Mon, 21 Mar 2011 22:27:21 +0000 (23:27 +0100)]
Bind only modem AT command endpoint to option module.

Network interface is handled by upcoming gt_b3730 module.

Removed "GT-B3710" from comment, it is another modem with another USB ID.

Signed-off-by: Marius B. Kotsbak <marius@kotsbak.com>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: cdc-wdm: reset handling according to new requirements
Oliver Neukum [Fri, 29 Apr 2011 12:12:21 +0000 (14:12 +0200)]
USB: cdc-wdm: reset handling according to new requirements

This patch
- ensures no IO takes place during resets
- reports resets to user space

Signed-off-by: Oliver Neukum <oneukum@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: gamin_gps: Fix for data transfer problems in native mode
Hermann Kneissel [Fri, 29 Apr 2011 06:58:43 +0000 (08:58 +0200)]
USB: gamin_gps: Fix for data transfer problems in native mode

This patch fixes a problem where data received from the gps is sometimes
transferred incompletely to the serial port. If used in native mode now
all data received via the bulk queue will be forwarded to the serial
port.

Signed-off-by: Hermann Kneissel <herkne@gmx.de>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: gadget: g_multi: fixed vendor and product ID in inf files
Michal Nazarewicz [Tue, 26 Apr 2011 17:08:36 +0000 (19:08 +0200)]
USB: gadget: g_multi: fixed vendor and product ID in inf files

Commit 1c6529e92b "USB: gadget: g_multi: fixed vendor and
product ID" replaced g_multi's vendor and product ID with
proper ID's from Linux Foundation.  This commit now updates
INF files in the Documentation/usb directory which were
omitted in the original commit.

Signed-off-by: Michal Nazarewicz <mina86@mina86.com>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: serial: ftdi_sio: adding support for TavIR STK500
Benedek László [Wed, 20 Apr 2011 01:22:21 +0000 (03:22 +0200)]
USB: serial: ftdi_sio: adding support for TavIR STK500

Adding support for the TavIR STK500 (id 0403:FA33)
Atmel AVR programmer device based on FTDI FT232RL.

Signed-off-by: Benedek László <benedekl@gmail.com>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: moto_modem: Add USB identifier for the Motorola VE240.
Elizabeth Jennifer Myers [Sat, 16 Apr 2011 18:49:51 +0000 (14:49 -0400)]
USB: moto_modem: Add USB identifier for the Motorola VE240.

Tested on my phone, the ttyUSB device is created and is fully
functional.

Signed-off-by: Elizabeth Jennifer Myers <elizabeth@sporksirc.net>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: CP210x Add 4 Device IDs for AC-Services Devices
Craig Shelley [Sun, 20 Mar 2011 13:51:13 +0000 (13:51 +0000)]
USB: CP210x Add 4 Device IDs for AC-Services Devices

This patch adds 4 device IDs for CP2102 based devices manufactured by
AC-Services. See http://www.ac-services.eu for further info.

Signed-off-by: Craig Shelley <craig@microtron.org.uk>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoMerge 2.6.39-rc4 into usb-next
Greg Kroah-Hartman [Tue, 19 Apr 2011 12:50:26 +0000 (05:50 -0700)]
Merge 2.6.39-rc4 into usb-next

This is needed to help resolve some xhci issues and other minor
differences.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoLinux 2.6.39-rc4 v2.6.39-rc4
Linus Torvalds [Tue, 19 Apr 2011 04:26:00 +0000 (21:26 -0700)]
Linux 2.6.39-rc4

13 years agoMerge branch 'for-39-rc4' of git://codeaurora.org/quic/kernel/davidb/linux-msm
Linus Torvalds [Mon, 18 Apr 2011 22:44:29 +0000 (15:44 -0700)]
Merge branch 'for-39-rc4' of git://codeaurora.org/quic/kernel/davidb/linux-msm

* 'for-39-rc4' of git://codeaurora.org/quic/kernel/davidb/linux-msm:
  msm: timer: fix missing return value
  msm: Remove extraneous ffa device check

13 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
Linus Torvalds [Mon, 18 Apr 2011 20:29:03 +0000 (13:29 -0700)]
Merge branch 'for-linus' of git://git./linux/kernel/git/dtor/input

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:
  Input: xen-kbdfront - fix mouse getting stuck after save/restore
  Input: estimate number of events per packet
  Input: evdev - indicate buffer overrun with SYN_DROPPED
  Input: document event types and codes and their intended use
  Input: add KEY_IMAGES specifically for AL Image Browser
  Input: twl4030_keypad - fix potential NULL dereference in twl4030_kp_probe()
  Input: h3600_ts - fix error handling at connect
  Input: twl4030_keypad - avoid potential NULL-pointer dereference

13 years agoMerge branch 'for-linus' of git://git.kernel.dk/linux-2.6-block
Linus Torvalds [Mon, 18 Apr 2011 20:21:18 +0000 (13:21 -0700)]
Merge branch 'for-linus' of git://git.kernel.dk/linux-2.6-block

* 'for-linus' of git://git.kernel.dk/linux-2.6-block:
  block: add blk_run_queue_async
  block: blk_delay_queue() should use kblockd workqueue
  md: fix up raid1/raid10 unplugging.
  md: incorporate new plugging into raid5.
  md: provide generic support for handling unplug callbacks.
  md - remove old plugging code.
  md/dm - remove remains of plug_fn callback.
  md: use new plugging interface for RAID IO.
  block: drop queue lock before calling __blk_run_queue() for kblockd punt
  Revert "block: add callback function for unplug notification"
  block: Enhance new plugging support to support general callbacks

13 years agoMerge branch 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc
Linus Torvalds [Mon, 18 Apr 2011 19:24:24 +0000 (12:24 -0700)]
Merge branch 'merge' of git://git./linux/kernel/git/benh/powerpc

* 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc:
  powerpc/powermac: Build fix with SMP and CPU hotplug
  powerpc/perf_event: Skip updating kernel counters if register value shrinks
  powerpc: Don't write protect kernel text with CONFIG_DYNAMIC_FTRACE enabled
  powerpc: Fix oops if scan_dispatch_log is called too early
  powerpc/pseries: Use a kmem cache for DTL buffers
  powerpc/kexec: Fix regression causing compile failure on UP
  powerpc/85xx: disable Suspend support if SMP enabled
  powerpc/e500mc: Remove CPU_FTR_MAYBE_CAN_NAP/CPU_FTR_MAYBE_CAN_DOZE
  powerpc/book3e: Fix CPU feature handling on 64-bit e5500
  powerpc: Check device status before adding serial device
  powerpc/85xx: Don't add disabled PCIe devices

13 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable
Linus Torvalds [Mon, 18 Apr 2011 19:24:05 +0000 (12:24 -0700)]
Merge git://git./linux/kernel/git/mason/btrfs-unstable

* git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable: (24 commits)
  Btrfs: fix free space cache leak
  Btrfs: avoid taking the chunk_mutex in do_chunk_alloc
  Btrfs end_bio_extent_readpage should look for locked bits
  Btrfs: don't force chunk allocation in find_free_extent
  Btrfs: Check validity before setting an acl
  Btrfs: Fix incorrect inode nlink in btrfs_link()
  Btrfs: Check if btrfs_next_leaf() returns error in btrfs_real_readdir()
  Btrfs: Check if btrfs_next_leaf() returns error in btrfs_listxattr()
  Btrfs: make uncache_state unconditional
  btrfs: using cached extent_state in set/unlock combinations
  Btrfs: avoid taking the trans_mutex in btrfs_end_transaction
  Btrfs: fix subvolume mount by name problem when default mount subvolume is set
  fix user annotation in ioctl.c
  Btrfs: check for duplicate iov_base's when doing dio reads
  btrfs: properly handle overlapping areas in memmove_extent_buffer
  Btrfs: fix memory leaks in btrfs_new_inode()
  Btrfs: check for duplicate iov_base's when doing dio reads
  Btrfs: reuse the extent_map we found when calling btrfs_get_extent
  Btrfs: do not use async submit for small DIO io's
  Btrfs: don't split dio bios if we don't have to
  ...

13 years agoproc: do proper range check on readdir offset
Linus Torvalds [Mon, 18 Apr 2011 17:36:54 +0000 (10:36 -0700)]
proc: do proper range check on readdir offset

Rather than pass in some random truncated offset to the pid-related
functions, check that the offset is in range up-front.

This is just cleanup, the previous commit fixed the real problem.

Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agonext_pidmap: fix overflow condition
Linus Torvalds [Mon, 18 Apr 2011 17:35:30 +0000 (10:35 -0700)]
next_pidmap: fix overflow condition

next_pidmap() just quietly accepted whatever 'last' pid that was passed
in, which is not all that safe when one of the users is /proc.

Admittedly the proc code should do some sanity checking on the range
(and that will be the next commit), but that doesn't mean that the
helper functions should just do that pidmap pointer arithmetic without
checking the range of its arguments.

So clamp 'last' to PID_MAX_LIMIT.  The fact that we then do "last+1"
doesn't really matter, the for-loop does check against the end of the
pidmap array properly (it's only the actual pointer arithmetic overflow
case we need to worry about, and going one bit beyond isn't going to
overflow).

[ Use PID_MAX_LIMIT rather than pid_max as per Eric Biederman ]

Reported-by: Tavis Ormandy <taviso@cmpxchg8b.com>
Analyzed-by: Robert Święcki <robert@swiecki.net>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agoInput: xen-kbdfront - fix mouse getting stuck after save/restore
Igor Mammedov [Mon, 18 Apr 2011 17:17:17 +0000 (10:17 -0700)]
Input: xen-kbdfront - fix mouse getting stuck after save/restore

Mouse gets "stuck" after restore of PV guest but buttons are in working
condition.

If driver has been configured for ABS coordinates at start it will get
XENKBD_TYPE_POS events and then suddenly after restore it'll start getting
XENKBD_TYPE_MOTION events, that will be dropped later and they won't get
into user-space.

Regression was introduced by hunk 5 and 6 of
5ea5254aa0ad269cfbd2875c973ef25ab5b5e9db
("Input: xen-kbdfront - advertise either absolute or relative
coordinates").

Driver on restore should ask xen for request-abs-pointer again if it is
available. So restore parts that did it before 5ea5254.

Acked-by: Olaf Hering <olaf@aepfle.de>
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
[v1: Expanded the commit description]
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
13 years agoInput: estimate number of events per packet
Jeff Brown [Mon, 18 Apr 2011 17:08:02 +0000 (10:08 -0700)]
Input: estimate number of events per packet

Calculate a default based on the number of ABS axes, REL axes,
and MT slots for the device during input device registration.

Signed-off-by: Jeff Brown <jeffbrown@android.com>
Reviewed-by: Henrik Rydberg <rydberg@euromail.se>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
13 years agoBtrfs: fix free space cache leak
Chris Mason [Mon, 18 Apr 2011 12:55:34 +0000 (08:55 -0400)]
Btrfs: fix free space cache leak

The free space caching code was recently reworked to
cache all the pages it needed instead of using find_get_page everywhere.

One loop was missed though, so it ended up leaking pages.  This fixes
it to use our page array instead of find_get_page.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
13 years agoblock: add blk_run_queue_async
Christoph Hellwig [Mon, 18 Apr 2011 09:41:33 +0000 (11:41 +0200)]
block: add blk_run_queue_async

Instead of overloading __blk_run_queue to force an offload to kblockd
add a new blk_run_queue_async helper to do it explicitly.  I've kept
the blk_queue_stopped check for now, but I suspect it's not needed
as the check we do when the workqueue items runs should be enough.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
13 years agoblock: blk_delay_queue() should use kblockd workqueue
Jens Axboe [Mon, 18 Apr 2011 09:36:39 +0000 (11:36 +0200)]
block: blk_delay_queue() should use kblockd workqueue

Reported-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
13 years agomd: fix up raid1/raid10 unplugging.
NeilBrown [Mon, 18 Apr 2011 08:25:43 +0000 (18:25 +1000)]
md: fix up raid1/raid10 unplugging.

We just need to make sure that an unplug event wakes up the md
thread, which is exactly what mddev_check_plugged does.

Also remove some plug-related code that is no longer needed.

Signed-off-by: NeilBrown <neilb@suse.de>
13 years agomd: incorporate new plugging into raid5.
NeilBrown [Mon, 18 Apr 2011 08:25:43 +0000 (18:25 +1000)]
md: incorporate new plugging into raid5.

In raid5 plugging is used for 2 things:
 1/ collecting writes that require a bitmap update
 2/ collecting writes in the hope that we can create full
    stripes - or at least more-full.

We now release these different sets of stripes when plug_cnt
is zero.

Also in make_request, we call mddev_check_plug to hopefully increase
plug_cnt, and wake up the thread at the end if plugging wasn't
achieved for some reason.

Signed-off-by: NeilBrown <neilb@suse.de>
13 years agomd: provide generic support for handling unplug callbacks.
NeilBrown [Mon, 18 Apr 2011 08:25:42 +0000 (18:25 +1000)]
md: provide generic support for handling unplug callbacks.

When an md device adds a request to a queue, it can call
mddev_check_plugged.
If this succeeds then we know that the md thread will be woken up
shortly, and ->plug_cnt will be non-zero until then, so some
processing can be delayed.

If it fails, then no unplug callback is expected and the make_request
function needs to do whatever is required to make the request happen.

Signed-off-by: NeilBrown <neilb@suse.de>
13 years agomd - remove old plugging code.
NeilBrown [Mon, 18 Apr 2011 08:25:42 +0000 (18:25 +1000)]
md - remove old plugging code.

md has some plugging infrastructure for RAID5 to use because the
normal plugging infrastructure required a 'request_queue', and when
called from dm, RAID5 doesn't have one of those available.

This relied on the ->unplug_fn callback which doesn't exist any more.

So remove all of that code, both in md and raid5.  Subsequent patches
with restore the plugging functionality.

Signed-off-by: NeilBrown <neilb@suse.de>
13 years agomd/dm - remove remains of plug_fn callback.
NeilBrown [Mon, 18 Apr 2011 08:25:41 +0000 (18:25 +1000)]
md/dm - remove remains of plug_fn callback.

Now that unplugging is done differently, the unplug_fn callback is
never called, so it can be completely discarded.

Signed-off-by: NeilBrown <neilb@suse.de>
13 years agomd: use new plugging interface for RAID IO.
NeilBrown [Mon, 18 Apr 2011 08:25:41 +0000 (18:25 +1000)]
md: use new plugging interface for RAID IO.

md/raid submits a lot of IO from the various raid threads.
So adding start/finish plug calls to those so that some
plugging happens.

Signed-off-by: NeilBrown <neilb@suse.de>
13 years agoblock: drop queue lock before calling __blk_run_queue() for kblockd punt
Jens Axboe [Mon, 18 Apr 2011 07:59:55 +0000 (09:59 +0200)]
block: drop queue lock before calling __blk_run_queue() for kblockd punt

If we know we are going to punt to kblockd, we can drop the queue
lock before calling into __blk_run_queue() since it only does a
safe bit test and a workqueue call. Since kblockd needs to grab
this very lock as one of the first things it does, it's a good
optimization to drop the lock before waking kblockd.

Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
13 years agoRevert "block: add callback function for unplug notification"
Jens Axboe [Mon, 18 Apr 2011 07:54:05 +0000 (09:54 +0200)]
Revert "block: add callback function for unplug notification"

MD can't use this since it really requires us to be able to
keep more than a single piece of state for the unplug. Commit
048c9374 added the required support for MD, so get rid of this
now unused code.

This reverts commit f75664570d8b75469cc468f23c2b27220984983b.

Conflicts:

block/blk-core.c

Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
13 years agoblock: Enhance new plugging support to support general callbacks
NeilBrown [Mon, 18 Apr 2011 07:52:22 +0000 (09:52 +0200)]
block: Enhance new plugging support to support general callbacks

md/raid requires an unplug callback, but as it does not uses
requests the current code cannot provide one.

So allow arbitrary callbacks to be attached to the blk_plug.

Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
13 years agopowerpc/powermac: Build fix with SMP and CPU hotplug
Benjamin Herrenschmidt [Mon, 18 Apr 2011 05:46:35 +0000 (15:46 +1000)]
powerpc/powermac: Build fix with SMP and CPU hotplug

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
13 years agopowerpc/perf_event: Skip updating kernel counters if register value shrinks
Eric B Munson [Fri, 15 Apr 2011 08:12:30 +0000 (08:12 +0000)]
powerpc/perf_event: Skip updating kernel counters if register value shrinks

Because of speculative event roll back, it is possible for some event coutners
to decrease between reads on POWER7.  This causes a problem with the way that
counters are updated.  Delta calues are calculated in a 64 bit value and the
top 32 bits are masked.  If the register value has decreased, this leaves us
with a very large positive value added to the kernel counters.  This patch
protects against this by skipping the update if the delta would be negative.
This can lead to a lack of precision in the coutner values, but from my testing
the value is typcially fewer than 10 samples at a time.

Signed-off-by: Eric B Munson <emunson@mgebm.net>
Cc: stable@kernel.org
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
13 years agopowerpc: Don't write protect kernel text with CONFIG_DYNAMIC_FTRACE enabled
Stefan Roese [Thu, 14 Apr 2011 23:49:53 +0000 (23:49 +0000)]
powerpc: Don't write protect kernel text with CONFIG_DYNAMIC_FTRACE enabled

This problem was noticed on an MPC855T platform. Ftrace did oops
when trying to write to the kernel text segment.

Many thanks to Joakim for finding the root cause of this problem.

Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Joakim Tjernlund <joakim.tjernlund@transmode.se>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
13 years agopowerpc: Fix oops if scan_dispatch_log is called too early
Anton Blanchard [Thu, 7 Apr 2011 21:44:21 +0000 (21:44 +0000)]
powerpc: Fix oops if scan_dispatch_log is called too early

We currently enable interrupts before the dispatch log for the boot
cpu is setup. If a timer interrupt comes in early enough we oops in
scan_dispatch_log:

Unable to handle kernel paging request for data at address 0x00000010

...

.scan_dispatch_log+0xb0/0x170
.account_system_vtime+0xa0/0x220
.irq_enter+0x88/0xc0
.do_IRQ+0x48/0x230

The patch below adds a check to scan_dispatch_log to ensure the
dispatch log has been allocated.

Signed-off-by: Anton Blanchard <anton@samba.org>
Cc: <stable@kernel.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>