Driver currently notifies only USB2 PHY on USB mode change.
Extend this to USB3 PHY so that PHY drivers based on the
mode can release system resources - clocks, regulators etc.
Additionally Qualcomm QMP and QUSB2 PHY drivers need to
override VBUS signal in PHY wrapper in device mode as USB
VBUS line is not connected to PHYs. Also, remove NULL checks
for PHY when calling phy_set_mode as PHY ops already check this.
Signed-off-by: NManu Gautam <mgautam@codeaurora.org>
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

Driver currently resumes and increments pm usage_count
of its child device (dwc3 main) from its runtime_resume
handler. This requires dwc3 runtime_resume to perform
pm_runtime_put to decrement the pm usage_count. However
runtime_put from dwc3 happens for non pci drivers
(e.g. dwc3-if-simple.c) as well which results in dwc3
pm usage_count becoming negative after couple of
runtime suspend resume iterations. Fix this by
performing runtime_get/put from dwc3-pci driver only
using workqueue.
Signed-off-by: NManu Gautam <mgautam@codeaurora.org>
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

Driver powers-off PHYs and reinitializes DWC3 core and gadget on
resume. While this works fine for gadget mode but in host
mode there is not re-initialization of host stack. Also, resetting
bus as part of bus_suspend/resume is not correct which could affect
(or disconnect) connected devices.
Fix this by not reinitializing core on suspend/resume in host mode
for HOST only and OTG/drd configurations.
Signed-off-by: NManu Gautam <mgautam@codeaurora.org>
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

Add support to get a list of resets available for the device.
These resets must be kept de-asserted until the device is
in use.
Signed-off-by: NVivek Gautam <vivek.gautam@codeaurora.org>
[p.zabel@pengutronix.de: switch to hidden reset control array]
Signed-off-by: NPhilipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

Move clock handling after of_platform_depopulate to achieve
a sequence that is reverse of the probe sequence.

Cc: Felipe Balbi <balbi@kernel.org>
Signed-off-by: NVivek Gautam <vivek.gautam@codeaurora.org>
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

Many variables in ep0 are set but never used, so should be removed.
Signed-off-by: NChristos Gkekas <chris.gekas@gmail.com>
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

Now that all the information we need sits in struct dwc3_ep, we can
start taking only a pointer to struct dwc3_ep as an argument. This
allows us to clean the code up a bit.
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

This is in preparation to simplifying prototype of
__dwc3_gadget_kick_transfer().
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

There is more possibility for sharing code if we just realise that now
__dwc3_gadget_kic_transfer() knows to break out early if there are no
TRBs left.
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

This will let us call __dwc3_gadget_kick_transfer()
unconditionally. No functional changes, cleanup only.
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

Add compatible string to use this generic glue layer to support
Spreadtrum SC9860 platform's dwc3 controller.
Signed-off-by: NBaolin Wang <baolin.wang@spreadtrum.com>
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

If we don't assign a TRB to ep0 requests, we won't be able to unmap
the request later on resulting in starvation of DMA resources.

Fixes: 4a71fcb8 ("usb: dwc3: gadget: only unmap requests from DMA if mapped")
Reported-by: NThinh Nguyen <Thinh.Nguyen@synopsys.com>
Tested-by: NThinh Nguyen <Thinh.Nguyen@synopsys.com>
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

The header clk-provider.h is there for clock drivers (providers) to
include, not client drivers (consumers).  That said,
of_clk_get_parent_count() is a helper function for clock providers, not
a clk API for consumers.

Let's replace of_clk_get_parent_count() with of_count_phandle_with_args()
call, so that we can remove the include of clk-provider.h.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

 dev_pm_ops is not supposed to change at runtime. Marking it
 constant.
Signed-off-by: NDoug Wilson <doug.wilson@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

For 66AK2Gx there is a requirement to use PM Runtime to properly manage
clocks and the power domains. Therefore, add PM runtime support. Remove
legacy clock api's calls since other users of this driver worked without
these clock apis calls.
Signed-off-by: NFranklin S Cooper Jr <fcooper@ti.com>
Signed-off-by: NSekhar Nori <nsekhar@ti.com>
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

platform_get_irq() returns an error code, but the dwc3-omap driver
ignores it and always returns -EINVAL. This is not correct and,
prevents -EPROBE_DEFER from being propagated properly.

Notice that platform_get_irq() no longer returns 0 on error:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=e330b9a6bb35dc7097a4f02cb1ae7b6f96df92af

Print and propagate the return value of platform_get_irq on failure.

This issue was detected with the help of Coccinelle.
Signed-off-by: NGustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

dev_pm_ops is not supposed to change at runtime. Marking it constant.
Signed-off-by: NDoug Wilson <doug.wilson@gmail.com>
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

The PIDs for Isochronous data transfers are incorrect
for high bandwidth IN endpoints when the request length
is less than EP wMaxPacketSize.

As per spec correct PIDs for ISOC data transfers are:

1) For request length <= maxpacket
	- DATA0,

2) For maxpacket < length <= (2 * maxpacket)
	- DATA1, DATA0

3) For (2 * maxpacket) <  length <= (3 * maxpacket)
	- DATA2, DATA1, DATA0.

But driver always sets PCM fields based on wMaxPacketSize
due to which DATA2 happens even for small requests.

Fix this by setting the PCM field of trb->size depending
on request length rather than fixing it to the value
depending on wMaxPacketSize.

Ideally it shouldn't give any issues as dwc3 will send
0-length packet for next IN token if host sends (even
after receiving a short packet). Windows seems to ignore
this but with MacOS frame loss observed when using f_uvc.
Signed-off-by: NManu Gautam <mgautam@codeaurora.org>
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

A recent optimization was made so that a request put on the
pending_list wouldn't get mapped for DMA until just before
preparing a TRB for it. However, this poses a problem in case
the request is dequeued or the endpoint is disabled before the
mapping is done as that would lead to dwc3_gadget_giveback()
unconditionally calling usb_gadget_unmap_request_for_dev() with
an invalid request->dma handle. Depending on the platform's DMA
implementation the unmap operation could result in a panic.

Since we know a successful mapping is a prerequisite for getting
a TRB, the unmap can be conditionally called only when req->trb
is non-NULL.

Fixes: cdb55b39 ("usb: dwc3: gadget: lazily map requests for DMA")
Signed-off-by: NJack Pham <jackp@codeaurora.org>
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

IRQ_NOAUTOEN cannot be used with shared IRQs, since commit 04c848d3
("genirq: Warn when IRQ_NOAUTOEN is used with shared interrupts") and
kernel now throws a warn dump. But OMAP DWC3 driver uses this flag. As
per commit 12a7f17f ("usb: dwc3: omap: fix race of pm runtime with
irq handler in probe") that introduced this flag, PM runtime can race
with IRQ handler when deferred probing happens due to extcon,
therefore IRQ_NOAUTOEN needs to be set so that irq is not enabled until
extcon is registered.

Remove setting of IRQ_NOAUTOEN and move the registration of
shared irq to a point after dwc3_omap_extcon_register() and
of_platform_populate(). This avoids possibility of probe deferring and
above said race condition.
Reviewed-by: NGrygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: NVignesh R <vigneshr@ti.com>
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

commit f54edb53 ("usb: dwc3: core: initialize ULPI before trying to
get the PHY") moved call to dwc3_core_get_phy() from dwc3_probe() to
dwc3_core_init() after dwc3_core_soft_reset(). But
dwc3_core_soft_reset() calls phy_init(), therefore dwc3_core_get_phy()
needs to be called before dwc3_core_soft_reset().

Fix this by moving call to dwc3_core_get_phy() before
dwc3_core_soft_reset().

This fixes the following abort seen on DRA7xx platforms
[   24.769118] usb usb2: SerialNumber: xhci-hcd.1.auto
[   24.781144] hub 2-0:1.0: USB hub found
[   24.787836] hub 2-0:1.0: 1 port detected
[   24.809939] Unhandled fault: imprecise external abort (0x1406) at 0x00000000
Reported-by: NCarlos Hernandez <ceh@ti.com>
Signed-off-by: NVignesh R <vigneshr@ti.com>
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

The run_wake flag in struct dev_pm_info is used to indicate whether
or not the device is capable of generating remote wakeup signals at
run time (or in the system working state), but the distinction
between runtime remote wakeup and system wakeup signaling has always
been rather artificial.  The only practical reason for it to exist
at the core level was that ACPI and PCI treated those two cases
differently, but that's not the case any more after recent changes.

For this reason, get rid of the run_wake flag and, when applicable,
use device_set_wakeup_capable() and device_can_wakeup() instead of
device_set_run_wake() and device_run_wake(), respectively.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Acked-by: NBjorn Helgaas <bhelgaas@google.com>

clk_prepare_enable() can fail here and we must check its return value.
Reviewed-by: NKrzysztof Kozlowski <krzk@kernel.org>
Signed-off-by: NArvind Yadav <arvind.yadav.cs@gmail.com>
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

The commit 47d3946e

	usb: dwc3: refactor gadget endpoint count calculation

refactored dwc3_gadget_init_endpoints() and in particular changed in or
out endpoint numbering to be through. It's not always convenient and
makes code a slightly harder to read.

Introduce a new temporary variable to make it easier to understand what
is going on inside the function.

While doing that, rename local variables as follows:
	u8 num	-> u8 total
	int num -> int kbytes

Replace implicit direction check via epnum with explicit use of
direction variable.

While here, replace %d to %u when compounding endpoint name since we are
using unsigned type.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

If PHY is entering Host mode, we need to enable VBUS.
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

If don't reorder initialization like this, we will never be able to
get a reference to ULPI PHYs.
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

If PHY is suspended by the time we want to issue ULPI transfers, we
will observe timeouts on the ULPI interface. In order to avoid such
issue, let's make sure PHY is resumed before issuing a ULPI transfer.
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

We don't (yet) support PTM_STATUS messages so let's not reply to them
erroneously.
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

Use this method to make sure we don't try to connect on speeds not
supported by the gadget driver.
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

acpi_evaluate_dsm() and friends take a pointer to a raw buffer of 16
bytes. Instead we convert them to use guid_t type. At the same time we
convert current users.

acpi_str_to_uuid() becomes useless after the conversion and it's safe to
get rid of it.
Acked-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Acked-by: NDan Williams <dan.j.williams@intel.com>
Cc: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: NJarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Reviewed-by: NJani Nikula <jani.nikula@intel.com>
Acked-by: NJani Nikula <jani.nikula@intel.com>
Cc: Ben Skeggs <bskeggs@redhat.com>
Acked-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Acked-by: NJoerg Roedel <jroedel@suse.de>
Acked-by: NAdrian Hunter <adrian.hunter@intel.com>
Cc: Yisen Zhuang <yisen.zhuang@huawei.com>
Acked-by: NBjorn Helgaas <bhelgaas@google.com>
Acked-by: NFelipe Balbi <felipe.balbi@linux.intel.com>
Acked-by: NMathias Nyman <mathias.nyman@linux.intel.com>
Reviewed-by: NHeikki Krogerus <heikki.krogerus@linux.intel.com>
Acked-by: NMark Brown <broonie@kernel.org>
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>

Instead of *always* dumping raw ctrl bytes, let's decode standard
requests which will make the lives of those debugging DWC3 quite a bit
easier.

Output will now look like so:

irq/34-dwc3-1594  [000] d..1   107.573081: dwc3_ctrl_req: Get Device Descriptor(Index = 0, Length = 18)
irq/34-dwc3-1594  [000] d..1   107.573694: dwc3_ctrl_req: Set Address(Addr = 01)
irq/34-dwc3-1594  [000] d..1   107.588319: dwc3_ctrl_req: Get Device Descriptor(Index = 0, Length = 18)
irq/34-dwc3-1594  [000] d..1   107.588816: dwc3_ctrl_req: Get Configuration Descriptor(Index = 0, Length = 9)
irq/34-dwc3-1594  [000] d..1   107.589191: dwc3_ctrl_req: Set Configuration(Config = 3)
irq/34-dwc3-1594  [000] d..1   107.589846: dwc3_ctrl_req: Get BOS Descriptor(Index = 0, Length = 5)
irq/34-dwc3-1594  [000] d..1   107.590146: dwc3_ctrl_req: Get BOS Descriptor(Index = 0, Length = 22)
irq/34-dwc3-1594  [000] d..1   107.590546: dwc3_ctrl_req: Get Configuration Descriptor(Index = 0, Length = 9)
irq/34-dwc3-1594  [000] d..1   107.590840: dwc3_ctrl_req: Get Configuration Descriptor(Index = 0, Length = 69)
irq/34-dwc3-1594  [000] d..1   107.591138: dwc3_ctrl_req: Get Configuration Descriptor(Index = 1, Length = 9)
irq/34-dwc3-1594  [000] d..1   107.591541: dwc3_ctrl_req: Get Configuration Descriptor(Index = 1, Length = 32)
irq/34-dwc3-1594  [000] d..1   107.591834: dwc3_ctrl_req: Get Device Descriptor(Index = 0, Length = 18)
irq/34-dwc3-1594  [000] d..1   114.701005: dwc3_ctrl_req: Get Device Descriptor(Index = 0, Length = 18)
irq/34-dwc3-1594  [000] d..1   114.721080: dwc3_ctrl_req: Get Device Descriptor(Index = 0, Length = 18)
irq/34-dwc3-1594  [000] d..1   114.722709: dwc3_ctrl_req: Get Device Descriptor(Index = 0, Length = 18)
irq/34-dwc3-1594  [000] d..1   114.728979: dwc3_ctrl_req: Get Device Descriptor(Index = 0, Length = 18)
irq/34-dwc3-1594  [000] d..1   114.730544: dwc3_ctrl_req: Get Device Qualifier Descriptor(Index = 0, Length = 10)
irq/34-dwc3-1594  [000] d..1   115.776018: dwc3_ctrl_req: Get Configuration Descriptor(Index = 0, Length = 9)
irq/34-dwc3-1594  [000] d..1   115.776760: dwc3_ctrl_req: Set Configuration(Config = 0)
irq/34-dwc3-1594  [000] d..1   115.777676: dwc3_ctrl_req: Get Configuration(Length = 1)
irq/34-dwc3-1594  [000] d..1   115.924797: dwc3_ctrl_req: Get Device Descriptor(Index = 0, Length = 18)
irq/34-dwc3-1594  [000] d..1   115.929025: dwc3_ctrl_req: Get String Descriptor(Index = 0, Length = 500)
irq/34-dwc3-1594  [000] d..1   115.929566: dwc3_ctrl_req: Get String Descriptor(Index = 1, Length = 500)
irq/34-dwc3-1594  [000] d..1   115.930911: dwc3_ctrl_req: Get String Descriptor(Index = 0, Length = 500)
irq/34-dwc3-1594  [000] d..1   115.931528: dwc3_ctrl_req: Get String Descriptor(Index = 2, Length = 500)
irq/34-dwc3-1594  [000] d..1   115.932950: dwc3_ctrl_req: Get String Descriptor(Index = 0, Length = 500)
irq/34-dwc3-1594  [000] d..1   115.933533: dwc3_ctrl_req: Get String Descriptor(Index = 3, Length = 500)

Note that Class and Vendor requests won't be decoded for obvious
reasons. Those will be printed as a raw sequence of bytes.

This patch has been tested against a normal host (both Linux and
Windows) and USB30CV Chapter 9 tests.
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

Instead, we can require caller to pass a buffer for the function to
use. This cleans things quite a bit.
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

Instead of going for a 512 byte buffer and using snprintf(), let's
rely on helps __string() and __assign_str() where possible.
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

No functional changes, just a slight readability improvement.
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

Instead of printing out enqueue and dequeue pointer value as a header
to the output, let's mark the TRBs in question with 'E' and 'D'. The
output looks slightly easier to read.
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

No functional changes, just making sure we can use these for ReST docs
later.
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

%p will leak kernel pointers, so let's not expose the information on
dmesg and instead use %pK. %pK will only show the actual addresses if
explicitly enabled under /proc/sys/kernel/kptr_restrict.

Cc: <stable@vger.kernel.org>
Acked-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

This patch adds a quirk to disable USB 2.0 MAC linestate check
during HS transmit. Refer the dwc3 databook, we can use it for
some special platforms if the linestate not reflect the expected
line state(J) during transmission.

When use this quirk, the controller implements a fixed 40-bit
TxEndDelay after the packet is given on UTMI and ignores the
linestate during the transmit of a token (during token-to-token
and token-to-data IPGAP).

On some rockchip platforms (e.g. rk3399), it requires to disable
the u2mac linestate check to decrease the SSPLIT token to SETUP
token inter-packet delay from 566ns to 466ns, and fix the issue
that FS/LS devices not recognized if inserted through USB 3.0 HUB.
Acked-by: NRob Herring <robh@kernel.org>
Reviewed-by: NGuenter Roeck <groeck@chromium.org>
Signed-off-by: NWilliam Wu <william.wu@rock-chips.com>
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

Some functions might want to have very, very long request queues. We
can't make any assumptions about how many requests we *are* able to
map, so instead of mapping requests early, let's map them late. This
way, functions can queue as many requests as they'd like but we won't
take DMA resources until they are needed.

Also, we can now stop processing requests when we run out of DMA
resources but still keep requests in the queue for late processing.
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>

We don't need a big fat warning with stack dump at all. Running out of
TRBs is a normal condition and we will have more TRBs available as
soon as some transfers complete.
Signed-off-by: NFelipe Balbi <felipe.balbi@linux.intel.com>