Since commit 511f3c53 (usb: gadget: udc-core: fix a regression during
gadget driver unbinding) usb_gadget_remove_driver will pass NULL for
the driver argument.
Signed-off-by: NMaarten ter Huurne <maarten@treewalker.org>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Add the missing unlock before return from function cppi41_dma_callback()
in the error handling case.
Signed-off-by: NWei Yongjun <yongjun_wei@trendmicro.com.cn>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Since dma_map_single() may fail it is good to actually check the return
code to see if it succeeded.
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Since the musb-gadget code now calls the dma engine properly it is
possible to enable it for the TX path in device mode.
AM335x Advisory 1.0.13 says that we may lose the toggle bit on multiple
RX transfers. There is a workaround in host mode but none in device mode
and therefore RX transfers are disabled.
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

This patch makes use of the two function is_cppi_enabled() and
tusb_dma_omap() instead of the ifdef for the proper DMA implementation
setup code. It basically shifts the code right by one indention level
and adds a few line breaks once the chars are crossed.
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

musb_dsps.c utilizes a symbol which is only
available when CONFIG_OF_IRQ is set, so make
it depend on that.
Signed-off-by: NFelipe Balbi <balbi@ti.com>

This driver is currently used by musb' cppi41 couter part. I may merge
both dma engine user of musb at some point but not just yet.

The driver seems to work in RX/TX mode in host mode, tested on mass
storage. I increaed the size of the TX / RX transfers and waited for the
core code to cancel a transfers and it seems to recover.

v2..3:
- use mall transfers on RX side and check data toggle.
- use rndis mode on tx side so we haveon interrupt for 4096 transfers.
- remove custom "transferred" hack and use dmaengine_tx_status() to
  compute the total amount of data that has been transferred.
- cancel transfers and reclaim descriptors

v1..v2:
- RX path added
- dma mode 0 & 1 is working
- device tree nodes re-created.

Cc: Vinod Koul <vinod.koul@intel.com>
Cc: Dan Williams <djbw@fb.com>
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

This moves the two instances from the big node into two child nodes. The
glue layer ontop does almost nothing.

There is one devices containing the control module for USB (2) phy,
(2) usb and later the dma engine. The usb device is the "glue device"
which contains the musb device as a child. This is what we do ever since.

The new file musb_am335x is just here to prob the new bus and populate
child devices.

There are a lot of changes to the dsps file as a result of the changes:

- musb_core_offset
  This is gone. The device tree provides memory ressources information
  for the device there is no need to "fix" things

- instances
  This is gone as well. If we have two instances then we have have two
  child enabled nodes in the device tree. For instance the SoC in beagle
  bone has two USB instances but only one has been wired up so there is
  no need to load and init the second instance since it won't be used.

- dsps_glue is now per glue device
  In the past there was one of this structs but with an array of two and
  each instance accessed its variable depending on the platform device
  id.

- no unneeded copy of structs
  I do not know why struct dsps_musb_wrapper is copied but it is not
  necessary. The same goes for musb_hdrc_platform_data which allocated
  on demand and then again by platform_device_add_data(). One copy is
  enough.
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

dsps uses a nop driver which is added in dsps itself and does the PHY
on/off calls within dsps. Since those calls are now moved the nop driver
itself, we can now request the phy proper phy and remove those calls.
Currently only the first musb interface is used so we only add one phy
node for now.
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

The "nop" driver isn't a do-nothing-stub but supports a couple functions
like clock on/off or is able to use a voltage regulator. This patch
simply renames the driver to "generic" since it is easy possible to
extend it by a simple function istead of writing a complete driver.
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Use the wrapper function for retrieving the platform data instead of
accessing dev->platform_data directly.
Signed-off-by: NJingoo Han <jg1.han@samsung.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

The EOI register is not present in the AM335x memory space according to
the TRM and thus removed.
Should any platform using the EOI register get merged then it may be
used again if the register address is not zero.
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

The ti81xx platform is not fully supported right now. This patch renames
the date structure to a am33xx prefix which is actually used.
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Commit 8d2421e6 ("usb: musb: kill global and static for multi instance")
removed the global dma_mask copy and replaced by parent's DMA mask. The
problem here is that if the parent does not have a dma_mask then
musb_remove() goes kaboom.
Instead trying to fix this I was thinking we do we need to erase
dma_mask in the first place?
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

ti81xx does not have a baseport mainline i.e. it should not boot. The
amount of rework that is required makes it easier to simply remove that
platform (i.e. that possible platform device) and add it later once it
comes back with DT support.
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

If the init / shutdown function of the phy moves out of dsps into the
phy driver, then dsps needs to call the callbacks of the phy driver to
ensure that this happens.
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

The musb driver no longer uses procfs, so get rid of the proc_dir_entry
variable in struct musb.
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

If the descriptor is missing the reqeust is never unmapped. This patch
changes this and renames the cleanup label to unlock since there is no
cleanup done. The cleanup would revert the allocation of ressource (i.e.
this dma mapping) but it does not, it simply unlocks and returns.
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

The cleanup in the error is missing the dma controller. The structure is
allocated at runtime and ux500 allocates even a little more than just
this struct. So cleanup!
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

This patch removes is_dma_capable() and an ifdef in the init/exit path
around init/de-init of the dma_controller. Since we have the empty stubs
in the PIO code we can call it without gcc trouble. Earlier we had an
ifdef and the is_dma_capable() macro where gcc ignored the if (0) path
even that the function was not around :)
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Add a dma_controller_create() returning NULL so a few ifdefs can
dropped.
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

The core code creates a controller and immediately after that it calls
the ->start() callback. This one might drop an error but nobody cares.
The same thing happens in the destroy corner: First ->stop() called
followed by destroy callback. So why not merge those two into the same
function since there is no difference.
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

This check is hardly required and alas is wrong. 'c' might be NULL but
the chances are low that 'controller' after the container_of() becomes
NULL.

Since no other DMA implementation is doing that and musb-core does not
call it with a NULL pointer it can dropped.
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

The ifdef reads somehow better than an ifndef
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

None of these USB files need idr.h, so don't include it.

Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Reviewed-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

some MUSB incarnations, such as those governed by
omap2430.c and tusb6010.c, have three resources, not
two.

Fix the bug created by commit 09fc7d22 (usb: musb:
fix incorrect usage of resource pointer) where only
two of the three resources would be passed to musb_core.c

[ balbi@ti.com : add tusb6010.c to original patch ]
Tested-by: NAaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: NKishon Vijay Abraham I <kishon@ti.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

it does not compile since 09fc7d ("usb: musb: fix incorrect usage of
resource pointer"). What makes me wonder most is if source of the
Tested-by tag :)
Acked-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch will allow ux500-musb to be probed and configured solely from
configuration found in Device Tree.

Cc: Rob Herring <rob.herring@calxeda.com>
Cc: linux-usb@vger.kernel.org
Cc: devicetree-discuss@lists.ozlabs.org
Acked-by: NFelipe Balbi <balbi@ti.com>
Acked-by: NFabio Baltieri <fabio.baltieri@linaro.org>
Signed-off-by: NLee Jones <lee.jones@linaro.org>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>

If we can ever get to a state where we can solely search for DMA channels
by name, this will almost completely alleviate the requirement to pass
copious amounts of information though platform data. Here we take the
first step towards this. The next step will be to enable Device Tree
complete with name<->event_line mapping.

Cc: linux-usb@vger.kernel.org
Acked-by: NFelipe Balbi <balbi@ti.com>
Acked-by: NFabio Baltieri <fabio.baltieri@linaro.org>
Signed-off-by: NLee Jones <lee.jones@linaro.org>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>

In its current state, the ux500-musb driver uses platform data pointers
blindly with no prior checking. If no platform data pointer is passed
this will Oops the kernel. In this patch we ensure platform data and
board data are present prior to using them.

Cc: linux-usb@vger.kernel.org
Acked-by: NFelipe Balbi <balbi@ti.com>
Acked-by: NFabio Baltieri <fabio.baltieri@linaro.org>
Signed-off-by: NLee Jones <lee.jones@linaro.org>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>

The dma_mask will always be the same as the coherent_dma_mask, so let's
cut down on the platform_data burden and set it as such in the driver.
This also saves us from supporting it separately when we come to enable
this driver for Device Tree.

Cc: linux-usb@vger.kernel.org
Acked-by: NFelipe Balbi <balbi@ti.com>
Acked-by: NFabio Baltieri <fabio.baltieri@linaro.org>
Signed-off-by: NLee Jones <lee.jones@linaro.org>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>

The MUSB HDRC configuration never changes between each of the ux500
supported platforms, so there's little point passing it though platform
data. If we set it in the driver instead, we can make good use of it
when booting with either ATAGs or Device Tree.

Cc: linux-usb@vger.kernel.org
Acked-by: NFelipe Balbi <balbi@ti.com>
Acked-by: NFabio Baltieri <fabio.baltieri@linaro.org>
Signed-off-by: NLee Jones <lee.jones@linaro.org>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>

For all ux500 based platforms the maximum number of end-points are used.
Move this knowledge into the driver so we can relinquish the burden from
platform data. This also removes quite a bit of complexity from the driver
and will aid us when we come to enable the driver for Device Tree.

Cc: linux-usb@vger.kernel.org
Acked-by: NFelipe Balbi <balbi@ti.com>
Acked-by: NFabio Baltieri <fabio.baltieri@linaro.org>
Signed-off-by: NLee Jones <lee.jones@linaro.org>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>

This patch attempts to fix the isochonour API in the musb host
driver.  In particular, the urb->start_frame field should always be
set by the driver; it isn't an input parameter.

The simplest way to accomplish this is to treat all URBs as though the
URB_ISO_ASAP flag was set.  This won't give the right behavior when an
underrun occurs, but I don't know enough about the musb driver to
handle that case.

Unfortunately, I have no way to test this change.
Signed-off-by: NAlan Stern <stern@rowland.harvard.edu>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

very minor patch fixing the following build warning on
a debug message:

drivers/usb/musb/musb_host.c: In function ‘musb_host_rx’:
drivers/usb/musb/musb_host.c:1763:4: warning: format ‘%x’ \
	expects argument of type ‘unsigned int’, but argument \
	6 has type ‘dma_addr_t’ [-Wformat]
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Since highmem PIO URB handling was introduced in:

8e8a5516 usb: musb: host: Handle highmem in PIO mode

when a URB is being handled it may happen that the static use_sg flag
was set by a previous URB with buffer in highmem.  This leads to error
in handling the present URB.

Fix this by making the use_sg flag URB specific.

Cc: stable <stable@vger.kernel.org> # 3.7+
Acked-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NVirupax Sadashivpetimath <virupax.sadashivpetimath@stericsson.com>
Signed-off-by: NFabio Baltieri <fabio.baltieri@linaro.org>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Various non functional coding style fixes on ux500_dma.c and
phy-ab8500-usb.c drivers.
Acked-by: NLinus Walleij <linus.walleij@linaro.org>
Acked-by: NMaxime Coquelin <maxime.coquelin@st.com>
Signed-off-by: NFabio Baltieri <fabio.baltieri@linaro.org>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Host side dma support for ux500 is enabled by piggybacking on Inventra
dma support.
Acked-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NMian Yousaf Kaukab <mian.yousaf.kaukab@stericsson.com>
Signed-off-by: NFabio Baltieri <fabio.baltieri@linaro.org>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

With the hcd is now a direct member of struct musb, we can now simply
eliminate the musb_to_hcd() macro. There aren't that many users left
anyway, as some where already fixed up when parts were factored out to
musb_host.c
Signed-off-by: NDaniel Mack <zonque@gmail.com>
Acked-by: NPeter Korsgaard <jacmet@sunsite.dk>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Initialize the host and gadget subsystems of the musb driver only when
the appropriate mode is selected from platform data, or device-tree
information, respectively.
Signed-off-by: NDaniel Mack <zonque@gmail.com>
Acked-by: NPeter Korsgaard <jacmet@sunsite.dk>
Signed-off-by: NFelipe Balbi <balbi@ti.com>