| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Our event ring consists of only one segment, and we risk filling
the event ring in case we get isoc transfers with short intervals
such as webcams that fill a TD every microframe (125us)
With 64 TRB segment size one usb camera could fill the event ring in 8ms.
A setup with several cameras and other devices can fill up the
event ring as it is shared between all devices.
This has occurred when uvcvideo queues 5 * 32TD URBs which then
get cancelled when the video mode changes. The cancelled URBs are returned
in the xhci interrupt context and blocks the interrupt handler from
handling the new events.
A full event ring will block xhci from scheduling traffic and affect all
devices conneted to the xhci, will see errors such as Missed Service
Intervals for isoc devices, and and Split transaction errors for LS/FS
interrupt devices.
Increasing the TRB_PER_SEGMENT will also increase the default endpoint ring
size, which is welcome as for most isoc transfer we had to dynamically
expand the endpoint ring anyway to be able to queue the 5 * 32TDs uvcvideo
queues.
The default size used to be 64 TRBs per segment
Cc: <stable@vger.kernel.org>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Isoc TDs usually consist of one TRB, sometimes two. When all goes well we
receive only one success event for a TD, and move the dequeue pointer to
the next TD.
This fails if the TD consists of two TRBs and we get a transfer error
on the first TRB, we will then see two events for that TD.
Fix this by making sure the event we get is for the last TRB in that TD
before moving the dequeue pointer to the next TD. This will resolve some
of the uvc and dvb issues with the
"ERROR Transfer event TRB DMA ptr not part of current TD" error message
Cc: <stable@vger.kernel.org>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Peter writes:
Only a small fix for /sys entry
|
|
Felipe writes:
usb: fixes for v4.1-rc2
Here's the first pull request for v4.1-rc cycle,
it contains a few interesting fixes including a
fix to correct register offsets on dwc3, a fix
for Kconfig dependencies on isp1301 phy driver,
and a bug fix for our configfs gadget creation
interface.
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
Make sure only to copy any actual data rather than the whole buffer,
when releasing the temporary buffer used for unaligned non-isochronous
transfers.
Compile-tested only.
Signed-off-by: Johan Hovold <johan@kernel.org>
Tested-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Make sure only to copy any actual data rather than the whole buffer,
when releasing the temporary buffer used for unaligned non-isochronous
transfers.
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
USB 2.01+ full-speed devices can have extended descriptor as well
and can support LPM.
Signed-off-by: Rupesh Tatiya <rtatiya@codeaurora.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Single transform macros with hidden arguments are not
particularly useful. Just use seq_printf directly instead.
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The USB PCI quirks code gets built into the kernel whenever CONFIG_PCI
is enabled, even if CONFIG_USB is not set. This can cause unnecessary
messages to show up in the kernel log, such as "CONFIG_USB_XHCI_HCD is
turned off, defaulting to EHCI" (which makes no sense when the kernel
has been configured without host-side USB support).
This patch addresses the problem by building pci-quirks.o only when
CONFIG_PCI and CONFIG_USB are both enabled.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Reported-by: Toralf Förster <toralf.foerster@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The Microsoft document "Using ACPI to Configure USB Ports on a Computer"
makes it clear that the removable flag will be cleared on ports that are
marked as unused by the firmware. Handle this case to match.
Signed-off-by: Matthew Garrett <mjg59@coreos.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Windows appears to pay more attention to the ACPI values than any hub
configuration, so prefer the firmware's opinion on whether a port is
fixed or removable before falling back to the hub values.
Signed-off-by: Matthew Garrett <mjg59@coreos.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Add a debugfs interface - softconnect - for host mode to
connect/disconnect the devices without physically remove the
them.
This adds the capability to re-enumerate the devices which are
permanently mounted on the board with the MUSB controller
together.
Signed-off-by: Bin Liu <b-liu@ti.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
RNDIS function has a limitation on the number of allowed instances.
So far it has been RNDIS_MAX_CONFIGS, which happens to be one.
In order to eliminate this kind of arbitrary limitation we should not
preallocate a predefined (RNDIS_MAX_CONFIGS) array of struct rndis_params
instances but instead allow allocating them on demand.
This patch allocates struct rndis_params on demand in rndis_register().
Coversly, the structure is free()'d in rndis_deregister().
If CONFIG_USB_GADGET_DEBUG_FILES is set, the proc files are created which
is the same behaviour as before, but the moment of creation is delayed
until struct rndis_params is actually allocated.
rnids_init() and rndis_exit() have nothing to do, so they are eliminated.
Signed-off-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
Don't use a space between function name and parameter list opening bracket.
All other functions in this file comply wich checkpatch rules.
Signed-off-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
RNDIS function has a limitation on the number of allowed instances.
So far it has been RNDIS_MAX_CONFIGS, which happens to be one.
In order to eliminate this kind of arbitrary limitation we should not
preallocate a predefined (RNDIS_MAX_CONFIGS) array of struct rndis_params
instances but instead allow allocating them on demand.
This patch prepares the elimination of the said limit by converting all the
functions which accept rndis config number to accept a pointer to the
actual struct rndis_params. Consequently, rndis_register() returns
a pointer to a corresponding struct rndis_params instance. The pointer
is then always used by f_rndis.c instead of config number when it talks
to rndis.c API.
A nice side-effect of the changes is that many lines of code in rndis.c
become shorter and fit in 80 columns.
If a function prototype changes in rndis.h a style cleanup is made
at the same time, otherwise checkpatch complains that the patch
has style problems.
Signed-off-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
Using unnecessary static char buffers isn't good.
Use the %pV extension instead.
Miscellanea:
o the dprintk return value is unused, make it void
o add __printf format and argument verification
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
The GPIO subsystem provides dummy GPIO consumer functions if GPIOLIB is
not enabled. Hence drivers that depend on GPIOLIB, but use GPIO consumer
functionality only, can still be compiled if GPIOLIB is not enabled.
Relax the dependency on GPIOLIB if COMPILE_TEST is enabled, where
appropriate.
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: <linux-usb@vger.kernel.org>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
The platform_device_id is not modified by the driver and core uses it as
const.
Signed-off-by: Krzysztof Kozlowski <k.kozlowski.k@gmail.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
Remove ifdefs for musb_host_rx to get closer to building in
all the DMA drivers.
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
Remove ifdefs for musb_host_rx to get closer to building in
all the DMA drivers.
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
Remove ifdefs for musb_host_rx to get closer to building in
all the DMA drivers.
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
Remove ifdefs for musb_host_rx to get closer to building in
all the DMA drivers.
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
Remove ifdefs for musb_host_rx to get closer to building in
all the DMA drivers.
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
We can remove the ifdefs by setting up helper functions for
mentor DMA and cppi/tusb DMA.
Note that I've kept the existing formatting as otherwise this
patch becomes pretty much unreadable.
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
Let's get rid of the horrible ifdef in middle of the expression.
We can do it by adding a variable for short_packet and testing
it separately for DMA related code.
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
For musb_core.c we can now just drop the DMA related
ifdef and use the already existing runtime test for
!is_cppi_enabled(musb) instead.
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
Set up function pointers for DMA so get closer to
being able to build in all the DMA engines.
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
Pass struct musb to tusb_dma_omap() and is_cppi_enabled(),
and add macros for the other DMA controllers. Populate the
platform specific quirks with the DMA type and use it during
runtime.
Note that platform glue layers with no custom DMA code are
tagged with MUSB_DMA_INVENTRA which may have a chance of
working. Looks like the defconfigs for these use PIO_ONLY,
so this should not break existing configs.
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
A string written by the user may not be zero terminated.
sscanf may read memory beyond the buffer if no zero byte
is found.
For testing build with CONFIG_USB_CHIPIDEA=y, CONFIG_USB_CHIPIDEA_DEBUG=y.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Peter Chen <peter.chen@freescale.com>
|
|
dwc2_hc_nak_intr could be called with a NULL qtd.
Ensure qtd exists before dereferencing it to avoid kernel panic.
This happens when using usb to ethernet adapter.
Acked-by: John Youn <johnyoun@synopsys.com>
Signed-off-by: Gregory Herrero <gregory.herrero@intel.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
dwc2 may not be able to exit from hibernation if the hardware
does not provide a way to detect resume signalling in this state.
Thus, add the possibility to disable hibernation feature.
Acked-by: John Youn <johnyoun@synopsys.com>
Signed-off-by: Gregory Herrero <gregory.herrero@intel.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
As dwc2 pci module is now exporting dwc2 platform device, include
platform.o in dwc2-y and remove USB_DWC2_PLATFORM configuration
option. Driver will be built as two modules, dwc2.ko and dwc2_pci.ko.
dwc2.ko is the new platform driver.
Remove all EXPORT_SYMBOL_GPL as they are not needed any more.
Acked-by: John Youn <johnyoun@synopsys.com>
Signed-off-by: Mian Yousaf Kaukab <yousaf.kaukab@intel.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
Add support for SetPortFeature(PORT_TEST) for root port.
Acked-by: John Youn <johnyoun@synopsys.com>
Signed-off-by: Jingwu Lin <jingwu.lin@intel.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
Align buffer must be allocated using kmalloc since irqs are disabled.
Coherency is handled through dma_map_single which can be used with irqs
disabled.
Reviewed-by: Julius Werner <jwerner@chromium.org>
Acked-by: John Youn <johnyoun@synopsys.com>
Signed-off-by: Gregory Herrero <gregory.herrero@intel.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
During urb_enqueue, if the urb can't be queued to the endpoint,
the urb is freed without any spinlock protection.
This leads to memory corruption when concurrent urb_dequeue try to free
same urb->hcpriv.
Thus, ensure the whole urb_enqueue in spinlocked.
Acked-by: John Youn <johnyoun@synopsys.com>
Signed-off-by: Gregory Herrero <gregory.herrero@intel.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
Once hub is runtime suspended, dwc2 must resume it
on port connect event.
Else, roothub will stay in suspended state and will
not resume transfers.
Acked-by: John Youn <johnyoun@synopsys.com>
Signed-off-by: Gregory Herrero <gregory.herrero@intel.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
Update controller state to indicate suspend entry.
Acked-by: John Youn <johnyoun@synopsys.com>
Signed-off-by: Gregory Herrero <gregory.herrero@intel.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
If phy driver is present register hcd handle to it.
Acked-by: John Youn <johnyoun@synopsys.com>
Signed-off-by: Mian Yousaf Kaukab <yousaf.kaukab@intel.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
Force argument is not used anymore. Clean up leftovers from
https://lkml.org/lkml/2014/12/9/283
Acked-by: John Youn <johnyoun@synopsys.com>
Signed-off-by: Mian Yousaf Kaukab <yousaf.kaukab@intel.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
Inform that device is otg-capable in case of otg configuration.
Acked-by: John Youn <johnyoun@synopsys.com>
Signed-off-by: Gregory Herrero <gregory.herrero@intel.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
During vbus session, usb controller needs to exit hibernation if it was
previously in suspend state.
Since controller will be resetted and configured, there is no need
to restore registers.
Moreover, set lx_state to L0 on B session. vbus_session callback may
not be used by all platforms. Thus, controller software state needs
to be set to L0 if the controller detects a valid B session.
Otherwise, lx_state will remain L2 and prevent any request submission.
Acked-by: John Youn <johnyoun@synopsys.com>
Signed-off-by: Gregory Herrero <gregory.herrero@intel.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
If usb controller is in partial power down, any write to registers may
cause unpredictable behavior.
Thus, prevent any new request submission once controller is in partial
power down.
Acked-by: John Youn <johnyoun@synopsys.com>
Signed-off-by: Gregory Herrero <gregory.herrero@intel.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
Nothing to be done in pm suspend/resume when controller is in L2.
Don't disconnect or reset. State is already saved when putting
controller in hibernation and will be restored on USB bus resume.
Acked-by: John Youn <johnyoun@synopsys.com>
Signed-off-by: Gregory Herrero <gregory.herrero@intel.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
ResetDet interrupt is used to detect a reset of the bus
while the controller is suspended.
This may happens for example when using Command Verifier.
Acked-by: John Youn <johnyoun@synopsys.com>
Signed-off-by: Gregory Herrero <gregory.herrero@intel.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
So the parameters can be used in both host and gadget modes.
Also consolidate param functions in the core.h
Acked-by: John Youn <johnyoun@synopsys.com>
Signed-off-by: Mian Yousaf Kaukab <yousaf.kaukab@intel.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
This is required due to an Intel specific hardware issue. Where id-
pin setup causes glitches on the interrupt line when CONIDSTSCHG
interrupt is enabled.
Specify external_id_pin_ctl when an external driver (for example phy)
can handle id change, so that CONIDSTSCHG interrupt can be disabled
from the controller.
Acked-by: John Youn <johnyoun@synopsys.com>
Signed-off-by: Gregory Herrero <gregory.herrero@intel.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
During suspend, there could a race condition between ep_queue and
suspend interrupt if lx_state is updated after releasing spinlock in
call_gadget(hsotg, suspend).
Acked-by: John Youn <johnyoun@synopsys.com>
Signed-off-by: Gregory Herrero <gregory.herrero@intel.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
Allow controller to enter in hibernation during usb bus suspend and
inform both phy and gadget about the suspended state.
While in hibernation, the controller can't detect the resume condition.
An external mechanism must call usb_phy_set_suspend on resume.
Exit hibernation when controller gets the resume interrupt and inform
only gadget driver about it.
Acked-by: John Youn <johnyoun@synopsys.com>
Signed-off-by: Gregory Herrero <gregory.herrero@intel.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
When suspending usb bus, phy driver may disable controller power.
In this case, registers need to be saved on suspend and restored
on resume.
Acked-by: John Youn <johnyoun@synopsys.com>
Signed-off-by: Gregory Herrero <gregory.herrero@intel.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
|
Dump all registers to take a complete snapshot of dwc2 state.
Code is inspired by dwc3/debugfs.c
Acked-by: John Youn <johnyoun@synopsys.com>
Signed-off-by: Mian Yousaf Kaukab <yousaf.kaukab@intel.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
|
Mathias Nyman
Greg Kroah-Hartman
Johan Hovold
Rupesh Tatiya
Joe Perches
Alan Stern
Matthew Garrett
Bin Liu
Andrzej Pietrasiewicz
Geert Uytterhoeven
Krzysztof Kozlowski
Tony Lindgren
Heinrich Schuchardt
Gregory Herrero
Mian Yousaf Kaukab
Jingwu Lin