that structure currently only holds a device
pointer to our own platform_device and musb's
platform_device, but soon it will hold pointers
to our clock structures and glue-specific bits
and pieces.
Signed-off-by: NFelipe Balbi <balbi@ti.com>

that structure currently only holds a device
pointer to our own platform_device and musb's
platform_device, but soon it will hold pointers
to our clock structures and glue-specific bits
and pieces.
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Just adding its own platform_driver, not really
using it yet.

Later patches will come to split power management
code from musb_core and move it completely to HW
glue layer.
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Just adding its own platform_driver, not really
using it yet.

When all HW glue layers are converted, more patches
will come to split power management code from musb_core
and move it completely to HW glue layer.
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Just adding its own platform_driver, not really
using it yet.

When all HW glue layers are converted, more patches
will come to split power management code from musb_core
and move it completely to HW glue layer.
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Just adding its own platform_driver, not really
using it yet.

When all HW glue layers are converted, more patches
will come to split power management code from musb_core
and move it completely to HW glue layer.
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Just adding its own platform_driver, not really
using it yet.

When all HW glue layers are converted, more patches
will come to split power management code from musb_core
and move it completely to HW glue layer.
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Just adding its own platform_driver, not really
using it yet.

When all HW glue layers are converted, more patches
will come to split power management code from musb_core
and move it completely to HW glue layer.
Signed-off-by: NFelipe Balbi <balbi@ti.com>

change all ocurrences of musb_hdrc to musb-hdrc.

We will call glue layer drivers musb-<glue layer>,
so in order to keep things somewhat standard, let's
change the underscore into a dash.
Signed-off-by: NFelipe Balbi <balbi@ti.com>

This will make things simpler when choosing which
glue layer to compile. It avoids a lot of magic
around the "default" Kconfig option and lets the
user choose what exactly s/he wants to compile.
Signed-off-by: NFelipe Balbi <balbi@ti.com>

when we start splitting HW glue layer, it's
gonna make it easier to re-use that structure.
Signed-off-by: NFelipe Balbi <balbi@ti.com>

preparing to a big refactor on musb code. We need
to be able to compile in all glue layers (or at
least all ARM-based ones) together and have a
working binary.

While preparing for that, we move every glue
layer to export only one symbol, which is
a struct musb_platform_ops, and make all
other functions static.

Later patches will come to allow for compiling
all glue layers together and have a working
binary.
Signed-off-by: NFelipe Balbi <balbi@ti.com>

This will be passed to musb_core by platform glue
layer in order to make it easier to compile support
for several HW glue layers.

Later patches will come using this structure and
also moving HW glue layers to its own platform
driver; the idea is to be able to handle platform
peculiarities in a manner which doesn't affect one
another.
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Case we can't allocate struct musb_request,
prevent a NULL pointer dereference by returning
early.
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Soon resource data will get automatically
populated from a set of autogenerated data
from TI's hardware database for the OMAP
platform.

Such database, might not have resources at
the expected order by the current drivers.

While we could hack in some exceptions to
that tool to generate resources in a specific
order, it seems less fragile to use the
resource name instead. That way, no matter
what order the resources are generated, the
driver still work.

Modified the OMAP, Blackfin and Davinci
architecture files to add the name of the IRQs
in the resource structures and musb driver to
use the platform_get_irq_byname() api to get
the device and dma irq numbers instead of using
the index.

Cc: Tony Lindgren <tony@atomide.com>
Acked-by: NKevin Hilman <khilman@deeprootsystems.com>
Acked-by: NMike Frysinger <vapier@gentoo.org>
Signed-off-by: NHema HK <hemahk@ti.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Removed the board_data parameter being
passed to musb_platform_init function
as board_data can be extracted from
device structure which is already member
of musb structure.
Acked-by: NKevin Hilman <khilman@deeprootsystems.com>
Cc: Tony Lindgren <tony@atomide.com>
Signed-off-by: NHema HK <hemahk@ti.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

AM35x supports only 32bit read operations so we need to have
workaround for 8bit and 16bit read operations.
Signed-off-by: NAjay Kumar Gupta <ajay.gupta@ti.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

AM35x has musb interface and uses CPPI4.1 DMA engine.
Current patch supports only PIO mode. DMA support can be
added later once basic CPPI4.1 DMA patch is accepted.

Also added USB_MUSB_AM35X which is required to differentiate musb ips
between OMAP3x and AM35x. This config would be used to for below
purposes,
        - Select am35x.c instead of omap2430.c for compilation
          at drivers/usb/musb directory. Please note there are
          significant differneces in these two files as musb ip
          in quite different on AM35x.

	  Please note that in multi omap configuration only omap2430.c
	  file will get compiled and we would require to select only
	  AM35x based board config to compile am35x.c

        - Select workaround codes applicable for AM35x musb issues.
          one such workaround is for bytewise read issue on AM35x.
Signed-off-by: NAjay Kumar Gupta <ajay.gupta@ti.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

For all modules, change <module>-objs to <module>-y; remove
if-statements and replace with lists using the kbuild idiom; move
flags to the top of the file; and fix alignment while trying to
maintain the original scheme in each file.

None of the dependencies are modified.
Signed-off-by: Nmatt mooney <mfm@muteddisk.com>
Acked-by: NSam Ravnborg <sam@ravnborg.org>
Acked-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Blackfin's musb_platform_init() needs to call gpio_free() for error cleanup iff
otg_get_transceiver() call returns NULL.
Signed-off-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Acked-by: NMike Frysinger <vapier@gentoo.org>
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Cc: stable <stable@kernel.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Blackfin's musb_platform_exit() forgets to call usb_nop_xceiv_unregister().
While fixing this, also remove the unneeded blank line there.
Signed-off-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Acked-by: NMike Frysinger <vapier@gentoo.org>
Cc: stable <stable@kernel.org>
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Since commit 461972d8 (musb_core: don't call
musb_platform_exit() twice), unloading the driver module results in a WARNING
"kobject: '(null)' (c73de788): is not initialized, yet kobject_put() is being
called." (or even kernel oops) on e.g. DaVincis, though only in the OTG mode.
There exists dubious and unbalanced put_device() call in musb_free() which
takes place only in the OTG mode.  As this commit caused musb_platform_exit()
to be called (and so unregister the NOP transceiver) before this put_device()
call, this function references already freed memory.

On the other hand, all the glue layers miss the otg_put_transceiver() call,
complementary to the otg_get_transceiver() call that they do.  So, I think
the solution is to get rid of the strange put_device() call, and instead
call otg_put_transceiver() in the glue layers...
Signed-off-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: stable <stable@kernel.org>
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Replace EXTRA_CFLAGS with ccflags-y.
Signed-off-by: Nmatt mooney <mfm@muteddisk.com>
Acked-by: NWANG Cong <xiyou.wangcong@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

commit 633ba7876b96ec339ef685357e2f7c60b5a8ce85 broke
g_file_storage functionality by enabling AUTOCLEAR on
all cases without caring for all gadget drivers.

This patch will only enable AUTOCLEAR if our endpoint's
FIFO was configured with double buffering support. Note
this is not a complete fix, double buffered case still
doesn't work always, but that hasn't been working for
quite some time. Other than reverting the entire commit
and breaking testusb with double buffered case again,
I decided it was better to fix the single buffered case
and spend more time fixing double buffered case properly.
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

This patch fixes the problem reported by Sergei:

>how come? we need to send ZLP before giving back the request.
>Well, look at the code ionce again. We need to send ZLP *after*
>request->actual == request->length, but as the check is inserted
>after the ZLP send, ZLP *may* be sent once the first DMA completes,
>not the last.

The patch also has been discussed on the link below:

	http://marc.info/?t=128454814900001&r=1&w=2Signed-off-by: NMing Lei <tom.leiming@gmail.com>
Reported-by: NSergei Shtylyov <sshtylyov@mvista.com>
Cc: David Brownell <dbrownell@users.sourceforge.net>
Cc: Anand Gadiyar <gadiyar@ti.com>
Cc: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

This will ignore any SESSREQ interrupt if musb is B state.
Charger detection may cause spurious SESSREQ interrupts.
Signed-off-by: NHeikki Krogerus <ext-heikki.krogerus@nokia.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

musb_g_rx() always returns if next_request() call yields NULL, so the DBG()
near the function's end can never be invoked. Remove it along with unneeded
'return'; also remove the duplicate 'request' check...
Signed-off-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Wrap flags with uninitialized_var() to suppress this:

drivers/usb/musb/cppi_dma.c:1158: warning: 'flags' may be used uninitialized
in this function
Signed-off-by: NJon Povey <jon.povey@racelogic.co.uk>
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Add support for MUSB Host DMA transfers greater than max
channel length, so that such transfers won't be truncated.
Signed-off-by: NAnil Shetty <anil@ti.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Since not all platforms are using the same offset 0x10 in
musb_save/restore_context() eg Blackfin the offset is 0x40,
Change the indexed address to direct.
Signed-off-by: NBob Liu <lliubbo@gmail.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

The USB stack maps the buffer for DMA if the controller supports DMA.
MUSB controller can perform DMA as well as PIO transfers.
The buffer needs to be unmapped before CPU can perform
PIO data transfers.

Export unmap_urb_for_dma() so that drivers can perform
the DMA unmapping in a sane way.
Signed-off-by: NMaulik Mankad <x0082077@ti.com>
Acked-by: NAlan Stern <stern@rowland.harvard.edu>
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

MUSB's DMA controller max channel length was set to 64k.
Transfer length greater than this max value is being truncated.
Signed-off-by: NAnil Shetty <anil@ti.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

This patch has been tested OK on beagle B5 board and
use usbtest #15 and #16 as testcase.
Signed-off-by: NMing Lei <tom.leiming@gmail.com>
Reviewed-by: NSergei Shtylyov <sshtylyov@mvista.com>
Cc: David Brownell <dbrownell@users.sourceforge.net>
Cc: Anand Gadiyar <gadiyar@ti.com>
Cc: Mike Frysinger <vapier@gentoo.org>
Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Texas Instruments DA8xx/OMAP-L1x glue layer for the
MUSBMHRDC driver.
Signed-off-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: NYadviga Grigorieva <yadviga@ru.mvista.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Enabling CONFIG_USB_MUSB_DEBUG option causes -DDEBUG to be added to gcc's
command line, however the DBG() macro doesn't depend on DEBUG, so that the
debugging messages get printed regardless of the option, and I don't think
that this was intended. Get rid of otherwise unused xprintk() macro and make
DBG() macro directly call pr_debug() which only results in the actual code
generated if DEBUG is defined.

This change makes musb_hdrc.o ~30% less in size with CONFIG_USB_MUSB_DEBUG
disabled (in host mode).
Signed-off-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

To accomplish this the function to register a gadget driver takes the bind
function as a second argument.  To make things clearer rename the function
to resemble platform_driver_probe.

This fixes many section mismatches like

	WARNING: drivers/usb/gadget/g_printer.o(.data+0xc): Section mismatch in
	reference from the variable printer_driver to the function
	.init.text:printer_bind()
	The variable printer_driver references
	the function __init printer_bind()

All callers are fixed.
Signed-off-by: NUwe Kleine-König <u.kleine-koenig@pengutronix.de>
[m.nazarewicz@samsung.com: added dbgp]
Signed-off-by: NMichał Nazarewicz <m.nazarewicz@samsung.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

This is a partial revert of 7f26b3a7 ("drivers/usb: Remove unnecessary
return's from void functions") as this hunk will vanish in USB tree completely.
Reported-by: NStephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Commit 46034dca (USB: musb_gadget_ep0: stop
abusing musb_gadget_set_halt()) forgot to restart a queued request after
clearing the endpoint halt feature. This results in a couple of USB resets
while enumerating the file-backed storage gadget due to CSW packet not being
sent for the MODE SENSE(10) command.
Signed-off-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: stable@kernel.org
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

This patch fixes the issue which was observed while transfering
a large file ( > 20MB) over USB (OMAP MUSB controller acts as USB host)
to an attached USB thumb drive.

It was found that CDB field of CBW packet was set to 0x0. This was
due to missing a barrier before DMA engine starts transfer.
This  buffer is  allocated using dma_alloc_coherent which gives
non-cacheble but bufferable memory and hence needed a write
memory barrier to flush the write buffer.

More info on this thread is here:
	http://www.spinics.net/lists/linux-omap/msg33987.htmlSigned-off-by: NSantosh Shilimkar <santosh.shilimkar@ti.com>
Signed-off-by: NMaulik Mankad <x0082077@ti.com>
Cc: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

DMA length should not go beyond the availabe space
of request buffer, so fix it.

Also set max_len of cppi dma channel as max size of
int type, so make musb dma handling happier.
Signed-off-by: NMing Lei <tom.leiming@gmail.com>
Cc: David Brownell <dbrownell@users.sourceforge.net>
Cc: Anand Gadiyar <gadiyar@ti.com>
Cc: Mike Frysinger <vapier@gentoo.org>
Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>