This patch will aggregate the probing of gadget/hcd driver into platform.c.
The gadget probe funtion is converted into gadget_init that is now only
responsible for gadget only initialization. All the gadget resources are now
handled by platform.c

Since the host workqueue will not get initialized if the driver is configured
for peripheral mode only. Thus we need to check for wq_otg before calling
queue_work().

Also, we move spin_lock_init to common location for both host and gadget that
is either in platform.c or pci.c.

We also move suspend/resume code to common platform code.

Lastly, move the "samsung,s3c6400-hsotg" binding into dwc2_of_match_table.
Signed-off-by: NDinh Nguyen <dinguyen@opensource.altera.com>
Acked-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

Hcd controller needs bus_suspend/resume, dwc2 controller make
root hub generate suspend/resume signal with hprt0 register
when work in host mode.
After the root hub enter suspend, we can make controller enter
low power state with PCGCTL register.

We also update the lx_state for hsotg state.

This patch has tested on rk3288 with suspend/resume.
Signed-off-by: NKever Yang <kever.yang@rock-chips.com>
Acked-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NFelipe Balbi <balbi@ti.com>

The driver's handling of DMA buffers for non-aligned transfers
was kind of nuts. For IN transfers, it left the URB DMA buffer
mapped until the transfer completed, then synced it, copied the
data from the bounce buffer, then synced it again.

Instead of that, just call usb_hcd_unmap_urb_for_dma() to unmap
the buffer before starting the transfer. Then no syncing is
required when doing the copy. This should also allow handling of
other types of mappings besides just dma_map_single() ones.

Also reduce the size of the bounce buffer allocation for Isoc
endpoints to 3K, since that's the largest possible transfer size.

Tested on Raspberry Pi and Altera SOCFPGA.
Signed-off-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The move from the staging tree to the main tree exposed a
longstanding memory corruption bug in the dwc2 driver. The
reordering of the driver initialization caused the dwc2 driver
to corrupt the initialization data of the sdhci driver on the
Raspberry Pi platform, which made the bug show up.

The error is in calling to_usb_device(hsotg->dev), since ->dev
is not a member of struct usb_device. The easiest fix is to
just remove the offending code, since it is not really needed.

Thanks to Stephen Warren for tracking down the cause of this.
Reported-by: NAndre Heider <a.heider@gmail.com>
Tested-by: NStephen Warren <swarren@wwwdotorg.org>
Signed-off-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The DWC2 driver should now be in good enough shape to move out of
staging. I have stress tested it overnight on RPI running mass
storage and Ethernet transfers in parallel, and for several days
on our proprietary PCI-based platform.
Signed-off-by: NPaul Zimmerman <paulz@synopsys.com>
Cc: Felipe Balbi <balbi@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Individual controller driver has different requirement for wakeup
setting, so move it from core to itself. In order to align with
current etting the default wakeup setting is enabled (except for
chipidea host).

Pass compile test with below commands:
	make O=outout/all allmodconfig
	make -j$CPU_NUM O=outout/all drivers/usb
Signed-off-by: NPeter Chen <peter.chen@freescale.com>
Acked-by: NAlan Stern <stern@rowland.harvard.edu>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

I fell over the problem reported in
https://github.com/raspberrypi/linux/pull/390:

"Issuing low-speed packets when the root port is in full-speed mode
 causes the root port to stop responding. Explicitly fail when
 enqueuing URBs to a LS endpoint on a FS bus."

with my dwc2 testing in NetBSD, so I adapted the change to dwc2.
Signed-off-by: NNick Hudson <skrll@netbsd.org>
[paulz: fixed up the patch to compile under Linux, and tested it]
Signed-off-by: NPaul Zimmerman <paulz@synopsys.com>
Tested-by: NStephen Warren <swarren@wwwdotorg.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Fixed a coding style issue, specifically:

Line 1798: Removed parentheses since return is not a function.
Signed-off-by: NAldo Iljazi <mail@aldo.io>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Fix some functions called by dwc2_hcd_qtd_add() to return either
a proper error code or 0, instead of somewhat random values.
Then change the caller of dwc2_hcd_qtd_add() to just check the
return value for 0.
Signed-off-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch fixes the following smatch warning in hcd.c:
drivers/staging/dwc2/hcd.c:787 dwc2_assign_and_init_hc() warn: unsigned 'urb->actual_length' is never less than zero.
Signed-off-by: NRashika Kheria <rashika.kheria@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The transfer scheduler in the dwc2 driver is pretty basic, not to
mention buggy. It works fairly well with just a couple of devices
plugged in, but if you add, say, multiple devices with periodic
endpoints, the scheduler breaks down and can't even enumerate all
the devices.

To improve this, import the "microframe scheduler" patch from the
driver in the downstream Raspberry Pi kernel, which is based on
the Synopsys vendor driver. The original patch came from Denx
(http://git.denx.de/?p=linux-denx.git) and was commited to the
raspberrypi.org git tree by "popcornmix" (Dom Cobley).

I have added a driver parameter for this, enabled by default, in
case anyone has problems with it and needs to disable it. I don't
think we should add a DT binding for that, though, since I plan
to remove the option once any bugs are fixed.

[raspberrypi.org patch from Dom Cobley]
Signed-off-by: NDom Cobley <popcornmix@gmail.com>
[adapted to dwc2 driver by Paul Zimmerman]
Signed-off-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

In dwc2_assign_and_init_hc(), validate urb->actual_length for OUT
endpoints before using the value. This fix is from the Synopsys
vendor driver.
Signed-off-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Before, the hwcfg registers were read at device init time, but
interpreted at various parts in the code. This commit unpacks the hwcfg
register values into a struct with properly labeled variables at init
time, which makes all the other code using these values more consise and
easier to read. Some values that were previously stored in the hsotg
struct are now moved into this new struct as well.

In addition to the hwcfg registers, the contents of some fifo size
registers are also unpacked. The hwcfg registers are read-only, so they
can be safely stored. The fifo size registers are read-write registers,
but their power-on values are significant: they give the maximum depth
of the fifo they describe.

This commit mostly moves code, but also attempts to simplify some
expressions from (val >> shift) & (mask >> shift) to
(val & mask) >> shift.

Finally, all of the parameters read from the hardware are debug printed
after unpacking them, so a bunch of debug prints can be removed from
other places.
Signed-off-by: NMatthijs Kooijman <matthijs@stdin.nl>
Acked-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This line extracted the available queue space without properly shifting
it. Since the code only cared wether it was zero or not, it worked as
expected without the shift, but adding shift makes the code cleaner.

While we're here, store the result in a helper variable that was already
declared to increase readability a bit more.
Signed-off-by: NMatthijs Kooijman <matthijs@stdin.nl>
Acked-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This commit changes expressions from (val >> shift) & (mask >> shift) to
(val & mask) >> shift.
Signed-off-by: NMatthijs Kooijman <matthijs@stdin.nl>
Acked-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This (read-only) register was read twice, storing it for later use the
second time. Now it is only read once, storing it right away.
Signed-off-by: NMatthijs Kooijman <matthijs@stdin.nl>
Acked-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Various register fields wider than one bit have constants defined for
their value. Previously, these registers would define the values as they
appear in the register, so shifted to the right to the position the
value appears in the register.

This commit changes those constants to their natural values (e.g, 0, 1,
2, etc.), as they are after shifting the register value to the right.
This also changes all relevant code to shift the values before comparing
them with constants.

This has the advantage that the values can be stored in smaller
variables (now they always require a u32) and makes the handling of
these values more consistent with other register fields that represent
natural numbers instead of enumerations (e.g., number of host channels).
Signed-off-by: NMatthijs Kooijman <matthijs@stdin.nl>
Acked-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

There was some code that cleared the dma_mask when dma was disabled in
the driver. Given that clearing the mask doesn't actually tell the usb
core we're not using dma, and a previous commit explicitely sets the
hcd->self.uses_dma value, it seems these values are unneeded and can
only potentially cause problems (when reloading a module, for example).
Signed-off-by: NMatthijs Kooijman <matthijs@stdin.nl>
Acked-by: NPaul Zimmerman <Paul.Zimmerman@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

When dma is disabled inside dwc2 (because the hardware does not support
it, or the code was changed to disable it for testing), let the usb core
know about this by clearing hcd->self.uses_dma.

By default, the usb core assumes that dma is used when a dma_mask is
set, but this might not always match the dma_enable value in dwc2. To
prevent problems resulting from a mismatch, better to explicitely
disable dma in this case (though everything seemed to work with the
wrong value of uses_dma as well, probably only resulted in some unneeded
work).
Signed-off-by: NMatthijs Kooijman <matthijs@stdin.nl>
Acked-by: NPaul Zimmerman <Paul.Zimmerman@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

If the platform or bus driver failed to setup a dma_mask, but the
hardware advertises support for DMA, before DMA would be enabled in
dwc2, but disabled in the usb core, making all connectivity break.

With this commit, the dwc2 driver will emit a warning and fall back to
slave mode in this case.

Note that since commit 642f2ecc (staging: dwc2: Fix dma-enabled platform
devices using a default dma_mask) the platform bindings make sure a DMA
mask is always present, but having this check here anyway is probably a
good from a defensive programming standpoint (in case of changes to
platform.c or addition of new glue layers).
Signed-off-by: NMatthijs Kooijman <matthijs@stdin.nl>
Acked-by: NPaul Zimmerman <Paul.Zimmerman@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This fix uses 'value' parameter as it should be instead
of hardcoded -1.
Signed-off-by: NJulien Delacou <julien.delacou@stericsson.com>
Acked-by: NPaul Zimmerman <paulz@synopys.com>
Reviewed-by: NMatthijs Kooijman <matthijs@stdin.nl>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Remove the use of bus_to_virt() and just fail the transfer if an
unaligned buffer with no virtual address is found. AFAIK that
can't happen anyway.

Also change setting of coherent DMA mask to the normal 32 bits.
31 bits was only needed when calling bus_to_virt() AFAICR.
Signed-off-by: NPaul Zimmerman <paulz@synopsys.com>
Tested-by: NStephen Warren <swarren@wwwdotorg.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Change the non-aligned buffer debug messages to dev_vdbg(). Also
remove some duplicated debug output when the driver is loaded.
Signed-off-by: NPaul Zimmerman <paulz@synopsys.com>
Tested-by: NStephen Warren <swarren@wwwdotorg.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The driver was lacking calls to usb_hcd_link_urb_to_ep(),
usb_hcd_unlink_urb_from_ep(), and usb_hcd_check_unlink_urb(). Add
those now.
Signed-off-by: NPaul Zimmerman <paulz@synopsys.com>
Tested-by: NStephen Warren <swarren@wwwdotorg.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The parameters to dwc2_host_complete() didn't make much sense.
The 'context' parameter always came from the ->priv member of the
'dwc2_urb' parameter, and both of those always came from a struct
dwc2_qtd. So just pass in the struct dwc2_qtd instead.

This also allows us to null out the dwc2_qtd->urb member after it
is freed, which the calling code forgot to do in several places,
causing random driver crashes from dereferencing the freed pointer.

This also requires the calls to dwc2_hc_handle_tt_clear() to be
moved before the calls to dwc2_host_complete(), otherwise that
routine would do nothing because dwc2_qtd->urb has already been
freed.
Signed-off-by: NPaul Zimmerman <paulz@synopsys.com>
Tested-by: NStephen Warren <swarren@wwwdotorg.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Passing the value DMA_BIT_MASK(31) to dma_set_mask() causes the
dwc2-pci driver to sometimes fail (cannot enumerate the connected
device). Change it to DMA_BIT_MASK(32) instead, which is a more
sensible value anyway.
Signed-off-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Change some dev_dbg() messages in dwc2_hcd_hub_control() to
dev_vdbg(), to prevent massive spew to the dmesg log when a device
is disconnected.
Signed-off-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Now the functions use proper const annotations, the global variable with
default params can be marked const, which prevents these values from
being changed for a specific device (in theory there could be multiple
controllers with different settings, for example).
Signed-off-by: NStephen Warren <swarren@wwwdotorg.org>
[matthijs@stdin.nl: Split patch from bigger patch, marked
		    dwc2_module_params in pci.c as const and added
		    commit message]
Signed-off-by: NMatthijs Kooijman <matthijs@stdin.nl>
Acked-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This makes the function name more clear and consistent with
dwc2_handle_common_intr().
Signed-off-by: NMatthijs Kooijman <matthijs@stdin.nl>
Acked-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The top-level hcd interrupt handlers already used irq_return_t, but the
functions to which it delegates the actual work and the common irq
handler returned plain ints. In addition, they used the IRQ_RETVAL in
the wrong way (but because of the values of the various constants, this
didn't result in wrong behaviour).
Signed-off-by: NStephen Warren <swarren@wwwdotorg.org>
[matthijs@stdin.nl: Split patch from bigger patch and added commit message]
Signed-off-by: NMatthijs Kooijman <matthijs@stdin.nl>
Acked-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This flag is a deprecated NOOP, interrupt handlers are always run
with interupts disabled.

See commit 6932bf37 (genirq: Remove IRQF_DISABLED from core code), and
include/linux/interrupt.h:

	 * IRQF_DISABLED - keep irqs disabled when calling the action handler.
	 *                 DEPRECATED. This flag is a NOOP and scheduled to be removed
Signed-off-by: NMatthijs Kooijman <matthijs@stdin.nl>
Acked-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This adds a dwc_platform.ko module that can be loaded by using
compatible = "snps,dwc2" in a device tree.
Signed-off-by: NMatthijs Kooijman <matthijs@stdin.nl>
Signed-off-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

A few debug messages were missing newlines, add them
Signed-off-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Fix to return a negative error code from the error handling
case instead of 0, as returned elsewhere in this function.
Signed-off-by: NWei Yongjun <yongjun_wei@trendmicro.com.cn>
Reviewed-by: NMatthijs Kooijman <matthijs@stdin.nl>
Acked-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Before, it was a function that would set all members of a given struct
containing only int members to -1. Now, it is renamed to
dwc_set_all_params and it works only on the dwc2_core_params struct.
This makes sure that all of the slightly dubious casting and assumptions
happen inside the function instead of by the caller.
Signed-off-by: NMatthijs Kooijman <matthijs@stdin.nl>
Acked-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Previously, it was "dwc_otg", but this does not correspond to the
directory name and might cause confusion with the old out-of-tree
dwc_otg driver of which many versions circulate.
Signed-off-by: NMatthijs Kooijman <matthijs@stdin.nl>
Acked-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Before, this was initialized in pci.c, after the dwc2_hcd_init was
called and the interrupts were enabled. This opened up a small time
window where common interrupts could be triggered, but there was no
handler for them, causing them to keep triggering infinitely and locking
up the machine.

On my RT3052 board this bug could be easily reproduced by hardcoding
the console log level to 8, so that a bunch of debug output from the dwc2
driver was generated inside this time window. This caused the interrupt
lockup to occur almost every time.

By requesting the irq inside dwc2_core_init and by disabling interrupts
before calling dwc2_core_init instead of after, we can be sure the
handler is registered before the interrupts are enabled, which should
close this window.
Reported-by: NStephen Warren <swarren@wwwdotorg.org>
Signed-off-by: NMatthijs Kooijman <matthijs@stdin.nl>
Acked-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This adds a config option USB_DWC2_DEBUG_PERIODIC that allows debugging
output be suppressed for periodic transfers. This helps when debugging
non-periodic transfers while there are also periodic transfers going on
(both to make the debug output less polluted and to prevent all CPU time
going to debug messages).

In addition, a debug message from dwc2_hcd_is_status_changed is removed
entirely, since it often floods the log regardless of periodic
transfers.
Signed-off-by: NMatthijs Kooijman <matthijs@stdin.nl>
Cc: Paul Zimmerman <Paul.Zimmerman@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

We were setting the DMA masks in dwc2_driver_probe(), but that is
before the driver parameters have been set to their default values.
That meant the DMA masks could be set wrong. Fix it by moving the
DMA mask setting into dwc2_hcd_init(), after the driver parameters
have been set.
Reported-by: NMatthijs Kooijman <matthijs@stdin.nl>
Signed-off-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Remove the unneeded struct device *dev argument from dwc2_hcd_init()
and dwc2_hcd_remove(), and pass in the value through the hsotg->dev
member instead
Signed-off-by: NPaul Zimmerman <paulz@synopsys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>