The IRQF_SHARED flag needs to be set regardless of the bus type. Don't
require glue code to set it manually.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

Move core device initialization to a central location in order to share
it with the device mode implementation.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

The isp1760-rhcd.h header contains PTD constants and structures only
useful for the HCD implementation. It also contains register definitions
that will be needed by common code when implementing support for the
ISP1761 device controller, but doesn't contain the isp1760_hcd structure
definition that will also be used by common code.

Move definitions to the right location and create an isp1760-regs.h to
store register definitions. No change other than moving definitions and
modifying indentation is performed.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

The slotinfo and memory_chunk structures are specific to the driver and
defined in a header file. Prefix them with isp1760_ to avoid namespace
clashes.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

Allocate the driver private data structure manually instead of using the
usb_hcd private space. This will allow skipping hcd registration for the
isp1761 when used in device mode only.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

Move the initialization code out of the PCI probe function to a new
function in order to simplify and fix error handling.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

This simplifies error and remove code paths.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

The function takes quite a few arguments, passing the resource pointer
instead of the start address and length simplifies it a bit.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

The argument is unused, remove it.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

The two functions are specific to the driver but have very generic
names, subject to collisions. Rename them.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

The glue code probe functions don't need to access the hcd structure
anymore. Modify isp1760_register to return an integer error code instead
of the hcd pointer.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

Don't duplicate *_set_drvdata calls in glue code.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

The removal cleanup code is duplicated between the different bus glues.
Move it to a central location.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

Both handle platform devices, merge them.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

The I/O registers are mapped in the HCD code but must be unmapped by
glue code. The OF and PCI glue code unmap them correctly but the
platform glue code is missing an iounmap() call. Fix it.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

The platform driver remove function needs to release the memory resource
requested at probe time. Instead of retrieving the resource from the
platform device, retrieve it from the usb_hcd. This mimics the behaviour
of the PCI and OF glues, and will make it easier to share code between
all three glue layers.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

The structure is allocated for the sole purpose of holding a pointer to
the HCD and being stored in platform device driver data. Store the HCD
pointer directly instead.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

Switching to the managed gpio descriptor API simplifies both error and
cleanup code paths (by removing the need to free the gpio) and runtime
code (by removing manual handling of the active low flag). It also
permits handling the reset gpio entirely from within the HCD code,
sharing it between the different glue layers.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

Fixed the following warnings (reported by cppcheck):
[drivers/usb/gadget/function/f_sourcesink.c:1217]: (warning) %d in format string (no. 1)
requires 'int' but the argument type is 'unsigned int'.
[drivers/usb/gadget/function/f_sourcesink.c:1261]: (warning) %d in format string (no. 1)
requires 'int' but the argument type is 'unsigned int'.
[drivers/usb/gadget/function/f_sourcesink.c:1305]: (warning) %d in format string (no. 1)
requires 'int' but the argument type is 'unsigned int'.
[drivers/usb/gadget/function/f_sourcesink.c:1349]: (warning) %d in format string (no. 1)
requires 'int' but the argument type is 'unsigned int'.
[drivers/usb/gadget/function/f_sourcesink.c:1393]: (warning) %d in format string (no. 1)
requires 'int' but the argument type is 'unsigned int'.
[drivers/usb/gadget/function/f_sourcesink.c:1437]: (warning) %d in format string (no. 1)
requires 'int' but the argument type is 'unsigned int'.
[drivers/usb/gadget/function/f_sourcesink.c:1476]: (warning) %d in format string (no. 1)
requires 'int' but the argument type is 'unsigned int'.
[drivers/usb/gadget/function/f_sourcesink.c:1520]: (warning) %d in format string (no. 1)
requires 'int' but the argument type is 'unsigned int'.
[drivers/usb/gadget/function/f_sourcesink.c:1564]: (warning) %d in format string (no. 1)
requires 'int' but the argument type is 'unsigned int'.
[drivers/usb/gadget/function/f_sourcesink.c:1608]: (warning) %d in format string (no. 1)
requires 'int' but the argument type is 'unsigned int'

Reviewed-by: Amit Virdi <amit.virdi@st.com>
Signed-off-by: Asaf Vertz <asaf.vertz@tandemg.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

Add eventfd which notifies userspace about ep0 events and AIO completion
events. It simplifies using of FunctionFS with event loop, because now
we need to poll on single file (instead of polling on ep0 and eventfd's
supplied to AIO layer).

FunctionFS eventfd is not triggered if another eventfd is supplied to
AIO layer (in AIO request). It can be useful, for example, when we want
to handle AIO transations for chosen endpoint in separate thread.

Signed-off-by: Robert Baldyga <r.baldyga@samsung.com>
Acked-by: Michal Nazarewicz <mina86@mina86.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

Use if-loop variable 'epfile' instead of start variable 'epfiles'. Now the
correct endpoint file name is stored.

Signed-off-by: Mario Schuknecht <mario.schuknecht@dresearch-fe.de>
Signed-off-by: Felipe Balbi <balbi@ti.com>

USB vendor type is encoded in field bmRequestType. Make test USB_TYPE_VENDOR
with bRequestType instead of bRequest.

Signed-off-by: Mario Schuknecht <mario.schuknecht@dresearch-fe.de>
Signed-off-by: Felipe Balbi <balbi@ti.com>

1. set AUTOREQ to NONE at the beginning of teardown;

2. add delay for dma pipeline to drain;

3. Do not set USB_TDOWN bit for RX teardown.

  The CPPI hw has an issue that when tearing down a RX channel, if
  another RX channel is receiving data, the CPPI will lockup.

  To workaround the issue, do not set the CPPI TD bit. The steps before
  this point ensures the CPPI channel will be torn down properly.

Signed-off-by: Bin Liu <b-liu@ti.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

The macro EP_MODE_AUTOREG_* should be called EP_MODE_AUTOREQ_*, as they
are used for register AUTOREQ.

Signed-off-by: Bin Liu <b-liu@ti.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

This patch adds dma_request_slave_channel_reason() calling to request
dma slave channels for multiplatform environment.

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

To add support for requesting DT DMA in the future, this patch adds
usbhsf_dma_init_pdev() function.

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

Fix  using the  bare number  to set the 'wHubCharacteristics' field of the Hub
Descriptor while the values are #define'd in <linux/usb/ch11.h>.

Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

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>.

Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

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>.

Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

Fix  using the  bare number  to set the 'wHubCharacteristics' field of the Hub
Descriptor while the values are #define'd in <linux/usb/ch11.h>.

Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

__uvcg_fill_strm() called from __uvcg_iter_stream_cls()
might have advanced the "data" even if __uvcg_iter_stream_cls()
returns an error, so use a backup copy as an argument to kfree().

Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

Update the MAINTAINERS entry for the dwc2 driver to show John Youn
as the new maintainer

Signed-off-by: Paul Zimmerman <paulz@synopsys.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

This reverts commit 0cf884e8.
Even after applying the follow-on patch at
https://patchwork.kernel.org/patch/5325111


there are still problems with device connect on the Altera SOCFPGA
platform at least. One possible fix would be to add a whitelist
to enable suspend/resume on platforms where it does work correctly.

Signed-off-by: Paul Zimmerman <paulz@synopsys.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

Instead of custom approach the patch converts code to use %pM specifier to
print MAC.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

The use of GPIOs should be optional for the generic phy, otherwise
the Altera SOCFPGA platform at least is broken.

Fixes breakage caused by a combination of e9f2cefb "usb: phy:
generic: migrate to gpio_desc" and 135b3c43

 "usb: dwc2: platform:
add generic PHY framework support".

Reviewed-by: Robert Jarzmik <robert.jarzmik@free.fr>
Signed-off-by: Paul Zimmerman <paulz@synopsys.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

1) Change "conv" an "vnoc" to "to_cpu_endian" to "to_little_endian".
2) No need to check the "limit" because that is already handled in
   kstrtoXX so delete that parameter along with the check.
3) By using a "bits" parameter, we can combine the "uxx" parameter and
   the "str2u" parameters.
4) The kstrtou##bits() conversion does not need to be done under the
   mutex so move it to the start of the function.
5) Change the name of "identity_conv" to "noop_conversion".

Acked-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

We need to add a kfree(h) on an error path.

Acked-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

"num" is a u32 so "(num > 0xFFFFFFFF)" is never true.  Also the range
is already checked in kstrtou32().

Acked-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>

We're basically saying ERR_CAST(NULL) and PTR_ERR(NULL) here, which is
nonsensical.

Acked-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>