This is the first step in separating USB transceivers from
USB OTG utilities.

Includes fixes to IMX code from Sascha Hauer.
Signed-off-by: NHeikki Krogerus <heikki.krogerus@linux.intel.com>
Acked-by: NSascha Hauer <s.hauer@pengutronix.de>
Acked-by: NPavankumar Kondeti <pkondeti@codeaurora.org>
Acked-by: NLi Yang <leoli@freescale.com>
Acked-by: NAlan Stern <stern@rowland.harvard.edu>
Acked-by: NIgor Grinberg <grinberg@compulab.co.il>
Reviewed-by: NMarek Vasut <marek.vasut@gmail.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

1/ remove incorrect comment (it is a non-blocking notifier)
2/ Use correct symbolic return value for notifier
3/ Make sure otg_notifier_work is cancelled before module exit.
Signed-off-by: NNeilBrown <neilb@suse.de>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

If we shutdown without stopping the gadget first or removing the cable,
gadget manages to configure itself again:

root@pandora /root# poweroff
The system is going down NOW!
Requesting system poweroff
[   47.714385] musb-hm halted.
[   48.120697]  gadget: suspend
[   48.123748]  gadget: reset config
[   48.127227]  gadget: ecm deactivated
[   48.130981] usb0: gether_disconnect
[   48.281799]  gadget: high-speed config #1: CDC Ethernet (ECM)
[   48.287872]  gadget: init ecm
[   48.290985]  gadget: notify connect false
[   48.295288]  gadget: notify speed 425984000

This is not only unwanted, it's also happening on half-unitialized
state, after musb_shutdown() has returned, which sometimes causes
hardware to fail to work after reboot. Let's better properly stop
gadget on shutdown too.

This patch moves musb_gadget_cleanup out of musb_free(), which has 2
callsites: probe error path and musb_remove. On probe error path it was
superflous since musb_gadget_cleanup is called explicitly there, and
musb_remove() calls musb_shutdown(), so cleanup will get called as before.
Signed-off-by: NGrazvydas Ignotas <notasas@gmail.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Commit 0020afb3 (ARM: mach-davinci:
remove mach/memory.h) removed mach/memory.h for DaVinci which broke
DaVinci MUSB build.

mach/memory.h is not actually needed in davinci.c, so remove it.
While at it, also remove some more machine specific inclulde
files which are not needed for build.

Tested on DM644x EVM using USB card reader.

Cc: stable@vger.kernel.org # v3.2
Signed-off-by: NSekhar Nori <nsekhar@ti.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

module_param(bool) used to counter-intuitively take an int.  In
fddd5201 (mid-2009) we allowed bool or int/unsigned int using a messy
trick.

It's time to remove the int/unsigned int option.  For this version
it'll simply give a warning, but it'll break next kernel version.
Acked-by: NMauro Carvalho Chehab <mchehab@redhat.com>
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>

This change removes confusing extern qualifier, which doesn't have any practical
sense there.
Signed-off-by: NVladimir Zapolskiy <vladimir.zapolskiy@nokia.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

The whole thing depends on USB_MUSB_HDRC, just add an 'if'.
Signed-off-by: NFelipe Contreras <felipe.contreras@gmail.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Shuffle the code a bit so the description is at the top.
Signed-off-by: NFelipe Contreras <felipe.contreras@gmail.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

These are handled by drivers core, and in a way that doesn't wake up the
devices.
Signed-off-by: NFelipe Contreras <felipe.contreras@gmail.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

And use dev instead of musb->controller.
Signed-off-by: NFelipe Contreras <felipe.contreras@gmail.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

enabled && driver || !enabled can be simplified to !enabled || driver.
Signed-off-by: NFelipe Contreras <felipe.contreras@gmail.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

In musb_init_controller() there's a pm_runtime_put(), but there's no
pm_runtime_get(), which creates a mismatch that causes the driver to
sleep when it shouldn't.

This was introduced in 7acc6197[1], but it wasn't triggered in my setup
until 18a26892[2] was merged to Linus' branch at point df0914[3]. IOW;
when PM is working as it was supposed to.

However, it seems most of the time this is used in a way that keeps the
counter above 0, so nobody noticed. Also, it seems to depend on the
configuration used in versions before 3.1, but not later (or in it).

I found the problem by loading isp1704_charger before any usb gadgets:
http://article.gmane.org/gmane.linux.kernel/1226122

All versions after 2.6.39 are affected.

[1] usb: musb: Idle path retention and offmode support for OMAP3
[2] OMAP2+: musb: hwmod adaptation for musb registration
[3] Merge branch 'omap-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap-2.6

Cc: stable@vger.kernel.org
Cc: Hema HK <hemahk@ti.com>
Signed-off-by: NFelipe Contreras <felipe.contreras@gmail.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

TXMAXP register is not getting programmed correctly for a full speed device
as can_bulk_split() have been removed by
"06624818: usb: musb: disable double buffering when it's broken" patch.

Adding back the case for can_bulk_split() to fix the reset message seen with
a full speed stick.
Signed-off-by: NAjay Kumar Gupta <ajay.gupta@ti.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

fix the following compile warning:

drivers/usb/musb/omap2430.c: In function 'musb_otg_notifier_work':
drivers/usb/musb/omap2430.c:279:3: warning: 'return' with a value, in
	function returning void
drivers/usb/musb/omap2430.c:282:2: warning: 'return' with a value, in
	function returning void
Signed-off-by: NFelipe Balbi <balbi@ti.com>

musb pm_runtime_get_sync call happens in intrrupt context on cable attach case
That can result in re-enabling the interrupts and cause side affects.

So move the code to a work queue.

Following is the error path hit on cable attach:

BUG: sleeping function called from invalid context at drivers/base/power/runtime.c:802
in_atomic(): 0, irqs_disabled(): 0, pid: 18, name: irq/378-twl6030

Backtrace:
[<c00520f0>] (dump_backtrace+0x0/0x110) from [<c054f454>] (dump_stack+0x18/0x1c)
[<c054f43c>] (dump_stack+0x0/0x1c) from [<c007f59c>] (__might_sleep+0x130/0x134)
[<c007f46c>] (__might_sleep+0x0/0x134) from [<c02c2794>] (__pm_runtime_resume+0x94/0x98)
[<c02c2700>] (__pm_runtime_resume+0x0/0x98) from [<c033e7e4>] (musb_otg_notifications+0x9c/0x164)
[<c033e748>] (musb_otg_notifications+0x0/0x164) from [<c00b3df0>] (notifier_call_chain+0x4c/0x8c)
[<c00b3da4>] (notifier_call_chain+0x0/0x8c) from [<c00b44a8>] (__atomic_notifier_call_chain+0x40/0x54)
[<c00b4468>] (__atomic_notifier_call_chain+0x0/0x54) from [<c00b44dc>] (atomic_notifier_call_chain+0x20/0x28)
[<c00b44bc>] (atomic_notifier_call_chain+0x0/0x28) from [<c033f124>] (twl6030_usb_irq+0xc8/0xdc)
[<c033f05c>] (twl6030_usb_irq+0x0/0xdc) from [<c00d79f8>] (irq_thread_fn+0x24/0x40)
[<c00d79d4>] (irq_thread_fn+0x0/0x40) from [<c00d7b64>] (irq_thread+0x150/0x1d8)
[<c00d7a14>] (irq_thread+0x0/0x1d8) from [<c00adf70>] (kthread+0x94/0x98)
[<c00adedc>] (kthread+0x0/0x98) from [<c0094388>] (do_exit+0x0/0x720)

Tested with:
MUSB Device mode: Cold boot / Hot plug
MUSB Host mode: Cold boot / Hot plug
Signed-off-by: NVikram Pandita <vikram.pandita@ti.com>
Signed-off-by: NMoiz Sonasath <m-sonasath@ti.com>
Signed-off-by: NVikram Pandita <vikram.pandita@ti.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

musb core driver and tusb6010 glue layer don't
depend on anything which is ARCH-specific. It
builds fine on x86 and ARM. Dropping the dependency
so we can compile-test on linux-next.
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Remove a few unnecessary headers from a
few files.
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Commit 1376d92f "usb: musb: allow musb and glue layers to be modules"
made the USB_MUSB_TUSB6010 option modular, but actually building
the driver as a module does not work, so various randconfig builds
actually fail. This changes all code that depends on the
option to also check for modular builds, and exports the necessary
symbols.
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Acked-by: NTony Lindgren <tony@atomide.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

The logic to allow only one DMA driver in MUSB is currently
flawed, because it also allows picking no DMA driver at all
and also not selecting PIO mode.

Using a choice statement makes this foolproof for now and
also simplifies the Makefile.

Unfortunately, we will have to revisit this when we start
supporting multiple ARM platforms in a single kernel binary,
because at that point we will actually need to select
multiple DMA drivers and pick the right one at run-time.
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Cc: Felipe Balbi <balbi@ti.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

we need to save and restore OTG_INTERFSEL register
else we will be unable to function on resume after
OFF mode.
Reported-by: NDevaraj Rangasamy <dev@ti.com>
Signed-off-by: NHema HK <hemahk@ti.com>
Signed-off-by: NKishon Vijay Abraham I <kishon@ti.com>
Signed-off-by: NVikram Pandita <vikram.pandita@ti.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

RxFifoSz, TxFifoSz, RxFifoAddr, TxFifoAddr
are all indexed registers.

So before doing a context save or restore, INDEX register
should be set, then only one gets to the right register offset.
Signed-off-by: NVikram Pandita <vikram.pandita@ti.com>
Signed-off-by: NAnand Gadiyar <gadiyar@ti.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Skip the use of work queue and call musb_dma_completion() directly from
DMA callback context.

Here follows measurements on a Snowball board with ondemand governor active.

Performance using work queue:
(105 MB) copied, 6.23758 s, 16.8 MB/s
(105 MB) copied, 5.7151 s, 18.3 MB/s
(105 MB) copied, 5.83583 s, 18.0 MB/s
(105 MB) copied, 5.93611 s, 17.7 MB/s

Performance without work queue
(105 MB) copied, 5.62173 s, 18.7 MB/s
(105 MB) copied, 5.61811 s, 18.7 MB/s
(105 MB) copied, 5.57817 s, 18.8 MB/s
(105 MB) copied, 5.58549 s, 18.8 MB/s
Signed-off-by: NPer Forlin <per.forlin@linaro.org>
Acked-by: NMian Yousaf Kaukab <mian.yousaf.kaukab@stericsson.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

This commit renames the “speed” field of the usb_gadget_driver
structure to “max_speed”.  This is so that to make it more
apparent that the field represents the maximum speed gadget
driver can support.

This also make the field look more like fields with the same
name in usb_gadget and usb_composite_driver structures.  All
of those represent the *maximal* speed given entity supports.

After this commit, there are the following fields in various
structures:
* usb_gadget::speed - the current connection speed,
* usb_gadget::max_speed - maximal speed UDC supports,
* usb_gadget_driver::max_speed - maximal speed gadget driver
  supports, and
* usb_composite_driver::max_speed - maximal speed composite
  gadget supports.
Signed-off-by: NMichal Nazarewicz <mina86@mina86.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

This commit replaces usb_gadget's is_dualspeed field with
a max_speed field.

[ balbi@ti.com : Fixed DWC3 driver ]
Signed-off-by: NMichal Nazarewicz <mina86@mina86.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Signed-off-by: NPaul Bolle <pebolle@tiscali.nl>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Several UDC drivers had a gadget driver's speed sanity check of the
form of:

	driver->speed != USB_SPEED_HIGH

or:

	driver->speed != USB_SPEED_HIGH && driver->speed != USB_SPEED_FULL

As more and more gadget drivers support USB SuperSpeed, driver->speed
may be set to USB_SPEED_SUPER and UDC driver should handle such gadget
correctly.  The above checks however fail to recognise USB_SPEED_SUPER
as a valid speed.

This commit changes the two checks to:

	driver->speed < USB_SPEED_HIGH

or:

	driver->speed < USB_SPEED_FULL

respectively.
Signed-off-by: NMichal Nazarewicz <mina86@mina86.com>
Reported-by: NAlan Stern <stern@rowland.harvard.edu>
Tested-by: NHeiko Stuebner <heiko@sntech.de>
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Currently the driver tries to save context in the suspend path, but
will cause an abort if the device is already runtime suspended.  This
happens, for example, if MUSB loaded/compiled-in, in host mode, but no
USB devices are attached.  MUSB will be runtime suspended, but then
attempting a system suspend will crash due to the context save
being attempted while the device is disabled.

On OMAP, as of v3.1, the driver's ->runtime_suspend() callback will be
called late in the suspend path (by the PM domain layer) if the driver
is not already runtime suspended, ensuring a full shutdown.

Therefore, the context save is not needed in the ->suspend() method
since it will be called in the ->runtime_suspend() method anyways
(similarily for resume.)

NOTE: this leaves the suspend/resume methods basically empty (with
      some FIXMEs and comments, but I'll leave it to the maintainers
      to decide whether to remove them.

Cc: stable@vger.kernel.org
Signed-off-by: NKevin Hilman <khilman@ti.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

fixup usage of dma direction by introducing dma_transfer_direction,
this patch moves usb/musb driver to use new enum
Signed-off-by: NVinod Koul <vinod.koul@linux.intel.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@suse.de>
Acked-by: NFelipe Balbi <balbi@ti.com>
Acked-by: NMian Yousaf Kaukab <mian.yousaf.kaukab@stericsson.com>

The HCD_FLAG_SAW_IRQ flag was introduced in order to catch IRQ routing
errors: If an URB was unlinked and the host controller hadn't gotten
any IRQs, it seemed likely that the IRQs were directed to the wrong
vector.

This warning hasn't come up in many years, as far as I know; interrupt
routing now seems to be well under control.  Therefore there's no
reason to keep the flag around any more.  This patch (as1495) finally
removes it.
Signed-off-by: NAlan Stern <stern@rowland.harvard.edu>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

commit de477254 (include: replace linux/module.h
with "struct module" wherever possible) introduced
a compilation breaked when it removed <linux/module.h>
from <linux/device.h> which musb glue layers were
(mistakenly) relying on.

Include that header to fix the compile error.
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

USB_MUSB_HDRC depends on USB_GADGET_DUALSPEED.  If HDRC is selected
but DUALSPEED is not, the kernel oopses:

[    3.132781] Unable to handle kernel NULL pointer dereference at virtual address 00000000
[    3.141296] pgd = c0004000
[    3.144134] [00000000] *pgd=00000000
[    3.147918] Internal error: Oops: 5 [#1] SMP
[    3.152404] Modules linked in:
[    3.155609] CPU: 0    Not tainted  (3.1.0-rc9-wl+ #417)
[    3.161132] PC is at composite_setup+0x738/0xbb4
[    3.165985] LR is at vprintk+0x400/0x47c
[    3.170135] pc : [<c031df90>]    lr : [<c004c150>]    psr: 60000093
[    3.170135] sp : c065dd50  ip : dfb1f0fc  fp : c065ddbc
[    3.182220] r10: 00000000  r9 : df8fcae8  r8 : df8fcaa0
[    3.187713] r7 : 00000000  r6 : df8eaa20  r5 : dfae8ea0  r4 : 00000000
[    3.194580] r3 : df8fcae8  r2 : 00010002  r1 : c065dc40  r0 : 00000047
[    3.201446] Flags: nZCv  IRQs off  FIQs on  Mode SVC_32  ISA ARM  Segment kernel
[    3.209228] Control: 10c53c7d  Table: 8000404a  DAC: 00000015
[    3.215270] Process swapper (pid: 0, stack limit = 0xc065c2f8)
[    3.221405] Stack: (0xc065dd50 to 0xc065e000)
[...]
[    3.415405] [<c031df90>] (composite_setup+0x738/0xbb4) from [<c0310d18>] (musb_g_ep0_irq+0x9d0/0xaf8)
[    3.425109] [<c0310d18>] (musb_g_ep0_irq+0x9d0/0xaf8) from [<c030fb04>] (musb_interrupt+0xb48/0xc74)
[    3.434722] [<c030fb04>] (musb_interrupt+0xb48/0xc74) from [<c030fc98>] (generic_interrupt+0x68/0x80)
[    3.444458] [<c030fc98>] (generic_interrupt+0x68/0x80) from [<c0095204>] (handle_irq_event_percpu+0x9c/0x234)
[    3.454925] [<c0095204>] (handle_irq_event_percpu+0x9c/0x234) from [<c00953e8>] (handle_irq_event+0x4c/0x6c)
[    3.465270] [<c00953e8>] (handle_irq_event+0x4c/0x6c) from [<c0097e10>] (handle_fasteoi_irq+0xd8/0x110)
[    3.475158] [<c0097e10>] (handle_fasteoi_irq+0xd8/0x110) from [<c0094d2c>] (generic_handle_irq+0x34/0x3c)
[    3.485260] [<c0094d2c>] (generic_handle_irq+0x34/0x3c) from [<c0014ae4>] (handle_IRQ+0x88/0xc8)
[    3.494537] [<c0014ae4>] (handle_IRQ+0x88/0xc8) from [<c00085b0>] (asm_do_IRQ+0x18/0x1c)
[    3.503051] [<c00085b0>] (asm_do_IRQ+0x18/0x1c) from [<c0429e78>] (__irq_svc+0x38/0xc0)

This patch changes Kconfig so that USB_GADGET_DUALSPEED is selected
automatically by USB_MUSB_HDRC.
Signed-off-by: NLuciano Coelho <coelho@ti.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

ARCH_U8500 covers both MACH_U8500 and MACH_U5500
Reported-by: NPaul Bolle <pebolle@tiscali.nl>
Signed-off-by: NMian Yousaf Kaukab <mian.yousaf.kaukab@stericsson.com>
Acked-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

that has already being done by udc-core.c. It's
unnecessary and might cause issues with some gadget
drivers.

Tested: Ajay Kumar Gupta <ajay.gupta@ti.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Nothing actually requires that <mach/hardware.h>, <mach/memory.h> nor
<asm/mach-types.h> be included here.
Signed-off-by: NNicolas Pitre <nicolas.pitre@linaro.org>
Acked-by: NFelipe Balbi <balbi@ti.com>

This flag is a NOOP and can be removed now.
Signed-off-by: NYong Zhang <yong.zhang0@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Signed-off-by: NJustin P. Mattock <justinmattock@gmail.com>
Signed-off-by: NVitaliy Ivanov <vitalivanov@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

There is a multi-year old bug in the MUSB hardware which is not documented.
It causes spurious interrupts and have various symptoms, like endless
"SetupEnd came in a wrong ep0stage" messages. The fix is taken from the
FreeBSD's musb driver.

How to reproduce:
For example issue clear-stall on a couple of endpoints very fast,
like one request per 125us. After a while the bug triggers and the
musb-chip becomes unusable until next re-enumeration.
Signed-off-by: NHans Petter Selasky <hps@bitfrost.no>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

This patch enables DMA mode1 for the RX path when we know
there won't be any short packets. We check that by looking
into the short_no_ok flag, if it's true we enable mode1, otherwise
we use mode0 to transfer the data.

This will result in a throughput performance gain of around
40% for USB mass-storage/mtp use cases.

[ balbi@ti.com : updated commit log and code comments slightly ]
Signed-off-by: NAnand Gadiyar <gadiyar@ti.com>
Signed-off-by: NMoiz Sonasath <m-sonasath@ti.com>
Signed-off-by: NVikram Pandita <vikram.pandita@ti.com>
Tested-by: NVikram Pandita <vikram.pandita@ti.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Now ${LINUX}/drivers/usb/* can use usb_endpoint_maxp(desc) to get maximum packet size
instead of le16_to_cpu(desc->wMaxPacketSize).
This patch fix it up

Cc: Armin Fuerst <fuerst@in.tum.de>
Cc: Pavel Machek <pavel@ucw.cz>
Cc: Johannes Erdfelt <johannes@erdfelt.com>
Cc: Vojtech Pavlik <vojtech@suse.cz>
Cc: Oliver Neukum <oliver@neukum.name>
Cc: David Kubicek <dave@awk.cz>
Cc: Johan Hovold <jhovold@gmail.com>
Cc: Brad Hards <bhards@bigpond.net.au>
Acked-by: NFelipe Balbi <balbi@ti.com>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Thomas Dahlmann <dahlmann.thomas@arcor.de>
Cc: David Brownell <david-b@pacbell.net>
Cc: David Lopo <dlopo@chipidea.mips.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Michal Nazarewicz <m.nazarewicz@samsung.com>
Cc: Xie Xiaobo <X.Xie@freescale.com>
Cc: Li Yang <leoli@freescale.com>
Cc: Jiang Bo <tanya.jiang@freescale.com>
Cc: Yuan-hsin Chen <yhchen@faraday-tech.com>
Cc: Darius Augulis <augulis.darius@gmail.com>
Cc: Xiaochen Shen <xiaochen.shen@intel.com>
Cc: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Cc: OKI SEMICONDUCTOR, <toshiharu-linux@dsn.okisemi.com>
Cc: Robert Jarzmik <robert.jarzmik@free.fr>
Cc: Ben Dooks <ben@simtec.co.uk>
Cc: Thomas Abraham <thomas.ab@samsung.com>
Cc: Herbert Pötzl <herbert@13thfloor.at>
Cc: Arnaud Patard <arnaud.patard@rtp-net.org>
Cc: Roman Weissgaerber <weissg@vienna.at>
Acked-by: NSarah Sharp <sarah.a.sharp@linux.intel.com>
Cc: Tony Olech <tony.olech@elandigitalsystems.com>
Cc: Florian Floe Echtler <echtler@fs.tum.de>
Cc: Christian Lucht <lucht@codemercs.com>
Cc: Juergen Stuber <starblue@sourceforge.net>
Cc: Georges Toth <g.toth@e-biz.lu>
Cc: Bill Ryder <bryder@sgi.com>
Cc: Kuba Ober <kuba@mareimbrium.org>
Cc: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
Signed-off-by: NKuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

In case one "forgot" to load the receiver i.e. doing
|modprobe omap2430
|modprobe musb_hdrc

he ends up with:

|musb-hdrc: version 6.0, ?dma?, otg (peripheral+host)
|HS USB OTG: no transceiver configured
|musb-hdrc musb-hdrc: musb_init_controller failed with status -19
|(NULL device *): gadget not registered.
|Unable to handle kernel NULL pointer dereference at virtual address 0000001c
|Internal error: Oops: 17 [#1] SMP
|[<c011383c>] (sysfs_find_dirent+0x4/0x60) from [<c01138c0>] (sysfs_get_dirent+0x28/0x78)
|[<c01138c0>] (sysfs_get_dirent+0x28/0x78) from [<c0115b78>] (sysfs_unmerge_group+0x1c/0x90)
|[<c0115b78>] (sysfs_unmerge_group+0x1c/0x90) from [<c0179ba4>] (dpm_sysfs_remove+0x14/0x3c)
|[<c0179ba4>] (dpm_sysfs_remove+0x14/0x3c) from [<c01742f8>] (device_del+0x40/0x1b4)
|[<c01742f8>] (device_del+0x40/0x1b4) from [<c0174478>] (device_unregister+0xc/0x18)
|[<c0174478>] (device_unregister+0xc/0x18) from [<bf0489b4>] (musb_free+0x24/0x88 [musb_hdrc])
|[<bf0489b4>] (musb_free+0x24/0x88 [musb_hdrc]) from [<bf057d18>] (musb_probe+0xb50/0xe3c [musb_hdrc])
|[<bf057d18>] (musb_probe+0xb50/0xe3c [musb_hdrc]) from [<c01779c4>] (platform_drv_probe+0x1c/0x24)

The problem is that musb_free() tries to figure out what was
initializued and what wasn't and clean up only the initialized part.
This works well for usb_del_gadget_udc() but device_unregister() can't
deal with it. Therefore we rely on the fact the we always have a parent
device and only then remove the device.
I broke this in 0f91349b ("usb: gadget: convert all users to the new udc
infrastructure")
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NFelipe Balbi <balbi@ti.com>