- 11 May, 2012 18 commits
-
-
Michael Grzeschik authored
Move spin_lock under the done label, so the lock will also be pulled in the error paths. Signed-off-by:
Michael Grzeschik <m.grzeschik@pengutronix.de> [rebased on top of the patchset] Signed-off-by:
Alexander Shishkin <alexander.shishkin@linux.intel.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Alexander Shishkin authored
Get rid of trailing comments in the structure definitions in favor of kernel-doc. Signed-off-by:
-
Alexander Shishkin authored
The old implementation used global hw_bank, the new implementation uses udc-local hw_bank. This field seems to be a leftover from previous coding experiments. Signed-off-by:
-
Alexander Shishkin authored
Add some generic code for roles and implement simple role switching based on ID pin state and/or a sysfs file. At this, we also rename the device to ci_hdrc, which is what it is. The "manual" switch is made into a sysfs file and not debugfs, because it might be useful even in non-debug context. For some boards, like sheevaplug, it seems to be the only way to switch roles without modifying the hardware, since the ID pin is always grounded. Signed-off-by:
-
Alexander Shishkin authored
Split the driver into the following parts: * core -- resources, register access, capabilities, etc; * udc -- device controller functionality; * debug -- logging events. Signed-off-by:
-
Alexander Shishkin authored
Since chipidea is a dual role controller, it makes sense to move it to its own directory, where we can also have host, otg and platform code related to this controller. It also makes sense to break out the driver into several compilation units like udc, host, debugging code, etc. Firstly, let's move the udc and platform code to drivers/usb/chipidea. Signed-off-by:
-
Alexander Shishkin authored
Report basic information about capabilities and register addresses on probe. Signed-off-by:
-
Alexander Shishkin authored
Newer versions of the chipidea controller support the "advance" setting of usb address, which means instead of setting it immediately, deferring it till the status completion. Unfortunately, older versions of the controller don't have this feature, so in order to support those too, we simply don't use it. It's about 4 lines of extra code, and isn't in any way critical to performance. While at it, change the return value of the hw_usb_set_address() to void, since it can't fail in any measurable way. With this patch, ci13xxx_udc driver works with the chipidea controller in kirkwood (feroceon SoC), as found in, for example, sheevaplug. Signed-off-by:
-
Alexander Shishkin authored
The regmap field is an array of register pointers, not the other way around. Signed-off-by:
-
Alexander Shishkin authored
There is no point in having tracing output in the kernel these days. Signed-off-by:
-
Alexander Shishkin authored
Don't try to initialize devices that don't have driver_data assigned to their pci ids. Reported-by:
-
Bjørn Mork authored
Prevents dereferencing an invalid struct usb_interface pointer. Always delete entry from device list whether or not the rest of the device state cleanup is postponed. The device list uses desc->intf as key, and wdm_open will dereference this key while searching for a matching device. A device should not appear in the list unless probe() has succeeded and disconnect() has not finished. Cc: Oliver Neukum <oliver@neukum.org> Cc: stable <stable@vger.kernel.org> Signed-off-by:
Bjørn Mork <bjorn@mork.no> Signed-off-by:
-
Bjørn Mork authored
We cannot dereference a removed USB interface for dev_printk. Use pr_debug instead where necessary. Flush errors are expected if device is unplugged and are therefore best ingored at this point. Move the kill_urbs() call in wdm_release with dev_dbg() for the non disconnect, as we know it has already been called if WDM_DISCONNECTING is set. This does not actually fix anything, but keeps the code more consistent. Cc: Oliver Neukum <oliver@neukum.org> Cc: stable <stable@vger.kernel.org> Signed-off-by:
-
Bjørn Mork authored
Else the poll will be restarted indefinitely in a tight loop, preventing final device cleanup. Cc: Oliver Neukum <oliver@neukum.org> Cc: stable <stable@vger.kernel.org> Signed-off-by:
-
joseph daniel authored
The warning shown up when ran with randconfig, warning: (USB_DWC3) selects USB_XHCI_PLATFORM which has unmet direct dependencies (USB_SUPPORT && USB_XHCI_HCD) Signed-off-by:
joseph daniel <josephdanielwalter@gmail.com> Signed-off-by:
-
Sasha Levin authored
Use WARN_ON() instead of __WARN, which also means we won't use any internal macros. Signed-off-by:
Sasha Levin <levinsasha928@gmail.com> Signed-off-by:
-
Steven J. Hill authored
Fix driver to work properly in big endian mode. Signed-off-by:
Steven J. Hill <sjhill@mips.com> Acked-by:
Alan Stern <stern@rowland.harvard.edu> Cc: stable <stable@vger.kernel.org> Signed-off-by:
-
Darren Hart authored
The Kontron M2M development board, also known as the Fish River Island II, has an optional daughter card providing access to the PCH_UART (EG20T) via a ti_usb_3410_5052 uart to usb chip. http://us.kontron.com/products/systems+and+platforms/m2m/m2m+smart+services+developer+kit.htmlSigned-off-by:
Darren Hart <dvhart@linux.intel.com> CC: Al Borchers <alborchers@steinerpoint.com> CC: Peter Berger <pberger@brimson.com> Cc: stable <stable@vger.kernel.org> Signed-off-by:
-
- 10 May, 2012 15 commits
-
-
Greg Kroah-Hartman authored
With a previous patch, the usb_driver suspend/resume callbacks got overridden and were never called if a usb_serial driver defined them. This patch fixes the opticon driver to move the suspend/resume callbacks into the usb_serial_driver structure where they will be properly called. It then removes the unused usb_driver structure. Cc: Johan Hovold <jhovold@gmail.com> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Cc: Julia Lawall <julia@diku.dk> Cc: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: -
Greg Kroah-Hartman authored
There are a number of device ids that the quatech2 driver supports that are duplicated in the serqt_usb2 driver, so remove them from the serqt_usb2 driver as they do not belong there. Reported-by:
Bill Pemberton <wfp5p@virginia.edu> Signed-off-by:
-
Greg Kroah-Hartman authored
Now that we have a "real" driver for the quatech devices, we can remove the drivers/staging/quatech_usb2/ driver as it is no longer needed. Thanks to Bill Pemberton for writing the new driver. Cc: Richard Ash <richard@audacityteam.org> Cc: Bill Pemberton <wfp5p@virginia.edu> Signed-off-by: -
Bill Pemberton authored
Add a copyright and license statement to the head of quatech.c source file. No code change here. Signed-off-by:
-
Bill Pemberton authored
This supports the Quatech USB 2 usb to serial adapters. Signed-off-by:
-
Benoit Goby authored
Add a variant of rndis_bind_config to let gadget drivers change rndis vendorID and manufacturer parameters. Signed-off-by:
Benoit Goby <benoit@android.com> [make rndis_bind_config a static inline function] Signed-off-by:
Andrzej Pietrasiewicz <andrzej.p@samsung.com> Signed-off-by:
Kyungmin Park <kyungmin.park@samsung.com> Acked-by:
Michal Nazarewicz <mina86@mina86.com> Signed-off-by:
-
Benoit Goby authored
f_rndis checks if rndis_string_defs[0].id is null to setup rndis and allocate string ids when it is bound to the first configuration: /* maybe allocate device-global string IDs */ if (rndis_string_defs[0].id == 0) { /* ... and setup RNDIS itself */ status = rndis_init(); if (status < 0) return status; rndis_string_defs[0].id must be reset to 0 on unbind for rndis to be correctly initialized on the next composite_bind. Signed-off-by: -
Mike Lockwood authored
This is needed to avoid name collisions on SoCs that have both usb gadget and usb host, where usb0 may be the rndis interface or a usb ethernet adapter. Signed-off-by:
Mike Lockwood <lockwood@android.com> Signed-off-by:
-
Benoit Goby authored
Remove geserial_setup from the init section. The android gadget driver calls it after probe, after userspace has configured the gadget driver. Signed-off-by:
-
Andrzej Pietrasiewicz authored
Add iSerialNumber to usb_composite_driver to allow setting a default value. This is useful when the module is compiled-in. Then the composite_bind is executed at kernel boot and string id for iSerialNumber can be overridden even if there is no iSerialNumber kernel commandline parameter. If the string id is not overridden, then get_string will never attempt to look for the alternative string contents using cdev->serial_override. Signed-off-by:
-
Andrzej Pietrasiewicz authored
Enable module parameters to be modified at runtime, especially if the module is compiled-in. Signed-off-by:
-
Benoit Goby authored
Reset config->interface in usb_add_config, as it may contain pointers to functions from a previous session if config is removed and re-added. Signed-off-by:
-
Benoit Goby authored
Add usb_remove_config to unbind a configuration and remove it from the configs list. This allows implementing composite gadget drivers that can disconnect themself from the bus and that will later be re-enumerated with a different configuration. Gadget drivers must call usb_gadget_disconnect before calling this function to disable the pullup, disconnect the device from the host, and prevent the host from enumerating the device while we are changing the gadget configuration. Signed-off-by:
-
Greg Kroah-Hartman authored
As pointed out by Dan Carpenter, the check for NULL for the id variable is no longer needed, especially as we just dereferenced it a few lines earlier, causing an oops if it really was NULL. This was caused by 62bb84ed: "usb: gadget: ci13xxx: convert to platform device". Reported-by:
Dan Carpenter <dan.carpenter@oracle.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Signed-off-by:
-
Greg Kroah-Hartman authored
As pointed out by Dan Carpenter, we should not be trying to call dev_err() on a structure that was previously determined to be NULL, that's just foolish and asking for trouble. So just delete the message, it's not going to do anyone any good to have it anyway. This problem was caused by 0f089094: "usb: gadget: ci13xxx: replace home-brewed logging with dev_{err,warn,info}" Reported-by:
-
- 09 May, 2012 7 commits
-
-
Lukasz Majewski authored
This commit adds a cosmetic change to the s3c-hsotg UDC driver. It moves s3c-hsotg.h to other linux/ related inclusions. Signed-off-by:
Lukasz Majewski <l.majewski@samsung.com> Signed-off-by:
-
Nicolas Ferre authored
A possible race condition appears because we are not initializing the ohci->regs before calling usb_hcd_request_irqs(). We move the call to ohci_init() in hcd->driver->reset() instead of hcd->driver->start() to fix this. This was experienced when we share the same IRQ line between OHCI and EHCI controllers. Signed-off-by:
Nicolas Ferre <nicolas.ferre@atmel.com> Tested-by:
Christian Eggers <christian.eggers@kathrein.de> Cc: stable <stable@vger.kernel.org> Acked-by:
-
Nicolas Ferre authored
Signed-off-by:
-
Igor Grinberg authored
When PHY reset pin is connected to a GPIO on external GPIO chip (e.g. I2C), we should not call the gpio_set_value() function, but gpio_set_value_cansleep(). Signed-off-by:
Igor Grinberg <grinberg@compulab.co.il> Acked-by:
Felipe Balbi <balbi@ti.com> Tested-by:
Govindraj.R <govindraj.raja@ti.com> Acked-by:
-
Alexander Shishkin authored
Finally, convert to the new style framework, using udc_start/udc_stop methods. Since there is no need in the global _udc pointer, remove it. Signed-off-by:
-
Alexander Shishkin authored
Currently, endpoints are initialized in gadget start/stop methods, however for the new style gadgets it is expected that bind() can be called before controller's start(), and we need endpoints already initialized at that point. So, move endpoint initialization to controller's probe before we switch to the "new style" gadget framework. Signed-off-by:
-
Alexander Shishkin authored
Logging output in the driver is mostly done using custom err/warn/info macros which rely on the existence of the global variable _udc, which is a global reference to the udc controller structure. This reference will have to go in order to allow us to have more than one chipidea udc in the system. Thus, replace custom macros with dev_{err,warn,info} using the platform device where possible. The trace() macro, which is a nop by default is left for tracing purposes. Signed-off-by:
-
