usb: dwc3: gadget: Handle DEV_TXF_FLUSH_BYPASS capability

DWC_usb32 IP introduces a new behavior when handling NoStream event for
IN endpoints. If the controller is capable of DEV_TXF_FLUSH_BYPASS, then
the driver does not need to force to restart stream for IN endpoints.
The controller will generate ERDY and restart the stream periodically.

Acked-by: Felipe Balbi <balbi@kernel.org>
Signed-off-by: Thinh Nguyen <Thinh.Nguyen@synopsys.com>
Link: https://lore.kernel.org/r/65c3070c666cd6b8beeee62d7f8e3e704ebf2d32.1619134559.git.Thinh.Nguyen@synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc3: Capture new capability register GHWPARAMS9

DWC_usb32 introduces a new HW capability register GHWPARAMS9. Capture
this in the dwc->hwparams.hwparams9 field.

Acked-by: Felipe Balbi <balbi@kernel.org>
Signed-off-by: Thinh Nguyen <Thinh.Nguyen@synopsys.com>
Link: https://lore.kernel.org/r/f76cc4a9c8c4ab325f5babe03c57b039166360b0.1619134559.git.Thinh.Nguyen@synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: gadget: prevent a ternary sign expansion bug

The problem is that "req->actual" is a u32, "req->status" is an int, and
iocb->ki_complete() takes a long.  We would expect that a negative error
code in "req->status" would translate to a negative long value.

But what actually happens is that because "req->actual" is a u32, the
error codes is type promoted to a high positive value and then remains
a positive value when it is cast to long.  (No sign expansion).

We can fix this by casting "req->status" to long.

Acked-by: Felipe Balbi <balbi@kernel.org>
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Link: https://lore.kernel.org/r/YIE7RrBPLWc3XtMg@mwanda
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc3: core: Do core softreset when switch mode

From: John Stultz <john.stultz@linaro.org>

According to the programming guide, to switch mode for DRD controller,
the driver needs to do the following.

To switch from device to host:
1. Reset controller with GCTL.CoreSoftReset
2. Set GCTL.PrtCapDir(host mode)
3. Reset the host with USBCMD.HCRESET
4. Then follow up with the initializing host registers sequence

To switch from host to device:
1. Reset controller with GCTL.CoreSoftReset
2. Set GCTL.PrtCapDir(device mode)
3. Reset the device with DCTL.CSftRst
4. Then follow up with the initializing registers sequence

Currently we're missing step 1) to do GCTL.CoreSoftReset and step 3) of
switching from host to device. John Stult reported a lockup issue seen
with HiKey960 platform without these steps[1]. Similar issue is observed
with Ferry's testing platform[2].

So, apply the required steps along with some fixes to Yu Chen's and John
Stultz's version. The main fixes to their versions are the missing wait
for clocks synchronization before clearing GCTL.CoreSoftReset and only
apply DCTL.CSftRst when switching from host to device.

[1] https://lore.kernel.org/linux-usb/20210108015115.27920-1-john.stultz@linaro.org/
[2] https://lore.kernel.org/linux-usb/0ba7a6ba-e6a7-9cd4-0695-64fc927e01f1@gmail.com/

Fixes: 41ce1456e1db ("usb: dwc3: core: make dwc3_set_mode() work properly")
Cc: Andy Shevchenko <andy.shevchenko@gmail.com>
Cc: Ferry Toth <fntoth@gmail.com>
Cc: Wesley Cheng <wcheng@codeaurora.org>
Cc: <stable@vger.kernel.org>
Tested-by: John Stultz <john.stultz@linaro.org>
Tested-by: Wesley Cheng <wcheng@codeaurora.org>
Signed-off-by: Yu Chen <chenyu56@huawei.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Thinh Nguyen <Thinh.Nguyen@synopsys.com>
Link: https://lore.kernel.org/r/374440f8dcd4f06c02c2caf4b1efde86774e02d9.1618521663.git.Thinh.Nguyen@synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Get rid of useless error checks in suspend interrupt

Squashed from Douglas Anderson's suggested commit
"usb: dwc2: Get rid of useless error checks for
hibernation/partial power down"

 - After this commit there should never be any
case where dwc2_enter_partial_power_down() and
dwc2_enter_hibernation() are called when
'params.power_down' is not correct.  Get rid of
the pile of error checking.

- As part of this cleanup some of the error messages
not to have __func__ in them.  That's not needed
for dev_err() calls since they already have the
device name as part of the message.

Acked-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20210416124843.9EDCDA005D@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Update dwc2_handle_usb_suspend_intr function.

To avoid working in two modes (partial power down
and hibernation) changed conditions for entering
partial power down or hibernation.

Instead of checking hw_params.power_optimized and
hw_params.hibernation now checking power_down
param which already set to one of the options
(Hibernation or Partial Power Down) based on
OTG_EN_PWROPT.

Acked-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Signed-off-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Link: https://lore.kernel.org/r/20210416124835.9F257A005D@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Add exit hibernation mode before removing drive

When dwc2 core is in hibernation mode loading
driver again causes driver fail. Because in
that mode registers are not accessible.

In order to exit from hibernation checking
dwc2 core power saving state in "dwc2_driver_remove()"
function. If core is in hibernation, then checking the
operational mode of the driver. To check whether dwc2 core
is operating in host mode or device mode there is one way
which is retrieving the backup value of "gotgctl" and compare
the "CurMod" value. If previously core entered hibernation
in host mode then the exit is performed for host if not then
exit is performed for device mode. The introduced checking
is because in hibernation state all registers are not
accessible.

Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210416124827.9BB59A005D@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Add hibernation exiting flow by system resume

Adds a new flow of exiting hibernation when PC is resumed
from suspend state.

Acked-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210416124819.A3B26A005C@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Add hibernation entering flow by system suspend

Adds a new flow of entering hibernation when PC is
hibernated or suspended.

Acked-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210416124811.BBFDBA005C@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Allow exit hibernation in urb enqueue

When core is in hibernation state and an external
hub is connected, upper layer sends URB enqueue request,
which results in port reset issue.

- Added exit from hibernation state to avoid port
reset issue and process upper layer request properly.

Acked-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210416124803.D1C1FA005F@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Move exit hibernation to dwc2_port_resume() function

This move is done to call hibernation exit handler in
"dwc2_port_resume()" function when core receives port resume.
Otherwise it could be confusing to exit hibernation in
"dwc2_hcd_hub_control()" function but other power saving modes
in "dwc2_port_resume()" function.

Acked-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210416124755.E47F3A005D@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Move enter hibernation to dwc2_port_suspend() function

This move is done to call enter hibernation handler in
"dwc2_port_suspend()" function when core receives port suspend.
Otherwise it could be confusing to enter to hibernation in
"dwc2_hcd_hub_control()" function but other power saving modes
in "dwc2_port_suspend()" function.

Acked-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210416124747.EE79EA005C@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Clear GINTSTS_RESTOREDONE bit after restore is generated.

When hibernation exit is performed the dwc2_hib_restore_common()
function is called. In that function we wait until GINTSTS_RESTOREDONE
bit is set. However, after the setting of that bit we get a lot of
(dwc2_hsotg_irq:) interrupts which indicates that (GINTSTS.RstrDoneInt)
restore done interrupt is asserted.

To avoid restore done interrupt storm after restore is generated
clear GINTSTS_RESTOREDONE bit in GINTSTS register.

Acked-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210416124739.D6269A005D@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Clear fifo_map when resetting core.

Switching from device mode to host mode by disconnecting
device cable core enters and exits form hibernation.
However, the fifo map remains not cleared. It results
to a WARNING (WARNING: CPU: 5 PID: 0 at drivers/usb/dwc2/
gadget.c:307 dwc2_hsotg_init_fifo+0x12/0x152 [dwc2])
if in host mode we disconnect the micro a to b host
cable. Because core reset occurs.

To avoid the WARNING, fifo_map should be cleared
in dwc2_core_reset() function by taking into account configs.
fifo_map must be cleared only if driver is configured in
"CONFIG_USB_DWC2_PERIPHERAL" or "CONFIG_USB_DWC2_DUAL_ROLE"
mode.

- Added "static inline void dwc2_clear_fifo_map()" helper
function to clear fifo_map with peripheral or dual role mode.

- Added a dummy version of "dwc2_clear_fifo_map()" helper
for host-only mode.

Acked-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210416124731.C500AA005D@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Allow exiting hibernation from gpwrdn rst detect

When device cable is disconnected core receives suspend
interrupt and enters hibernation. After entering
into hibernation GPWRDN_RST_DET and GPWRDN_STS_CHGINT
interrupts are asserted.

Allowed exit from gadget hibernation from
GPWRDN_RST_DET by checking only linestate.

Changed the return type of "dwc2_handle_gpwrdn_intr()"
function from void to int because exit from hibernation
functions have a return value.

Acked-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210416124723.B6F17A005C@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Fix hibernation between host and device modes.

When core is in hibernation in host mode and a device cable
was connected then driver exited from device hibernation.
However, registers saved for host mode and when exited from
device hibernation register restore would be done for device
register which was wrong because there was no device registers
stored to restore.

- Added dwc_handle_gpwrdn_disc_det() function which handles
  gpwrdn disconnect detect flow and exits hibernation
  without restoring the registers.
- Updated exiting from hibernation in GPWRDN_STS_CHGINT with
  calling dwc_handle_gpwrdn_disc_det() function. Here no register
  is restored which is the solution described above.

Fixes: 65c9c4c6b01f ("usb: dwc2: Add dwc2_handle_gpwrdn_intr() handler")
Acked-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Signed-off-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Link: https://lore.kernel.org/r/20210416124715.75355A005D@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Fix host mode hibernation exit with remote wakeup flow.

Added setting "port_connect_status_change" flag to "1" in order
to re-enumerate, because after exit from hibernation port
connection status is not detected.

Fixes: c5c403dc4336 ("usb: dwc2: Add host/device hibernation functions")
Acked-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210416124707.5EEC2A005D@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Reset DEVADDR after exiting gadget hibernation.

Initially resetting device address was done in dwc2_hsotg_irq()
interrupt handler. However, when core is hibernated USB RESET
is not handled in dwc2_hsotg_irq() handler, instead USB RESET
interrupt is handled in dwc2_handle_gpwrdn_intr() handler.

- Added reset device address to zero when core exits from gadget
  hibernation.

Acked-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Signed-off-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Link: https://lore.kernel.org/r/20210416124659.652CFA005C@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Update exit hibernation when port reset is asserted

No need to check for "DWC2_POWER_DOWN_PARAM_HIBERNATION" param
as "hsotg->hibernated" flag is already enough for exiting from
hibernation mode.

- Removes checking of "DWC2_POWER_DOWN_PARAM_HIBERNATION" param.

- For code readability Hibernation exit code moved after
debug message print.

- Added "dwc2_exit_hibernation()" function error checking.

Acked-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210416124651.51C8DA005C@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Merge tag 'usb-serial-5.13-rc1' of https://git./linux/kernel/git/johan/usb-serial into usb-next

Johan writes:

USB-serial updates for 5.13-rc1

Here are the USB-serial updates for 5.13-rc1, including:

 - better type detection for pl2303
 - support for more line speeds for pl2303 (TA/TB)
 - fixed CSIZE handling for the new xr driver
 - core support for multi-interface functions
 - TIOCGSERIAL and TIOCSSERIAL fixes
 - generic TIOCSSERIAL support (e.g. for closing_wait)
 - fixed return value for unsupported ioctls
 - support for gpio valid masks in cp210x
 - drain-delay fixes and improvements
 - support for multi-port devices for xr
 - generalisation of the xr driver to support three new device classes
   (XR21B142X, XR21B1411 and XR2280X)

Included are also various clean ups.

All have been in linux-next with no reported issues.

* tag 'usb-serial-5.13-rc1' of https://git.kernel.org/pub/scm/linux/kernel/git/johan/usb-serial: (72 commits)
  USB: cdc-acm: add more Maxlinear/Exar models to ignore list
  USB: serial: xr: add copyright notice
  USB: serial: xr: reset FIFOs on open
  USB: serial: xr: add support for XR22801, XR22802, XR22804
  USB: serial: xr: add support for XR21B1411
  USB: serial: xr: add support for XR21B1421, XR21B1422 and XR21B1424
  USB: serial: xr: add type abstraction
  USB: serial: xr: drop type prefix from shared defines
  USB: serial: xr: move pin configuration to probe
  USB: serial: xr: rename GPIO-pin defines
  USB: serial: xr: rename GPIO-mode defines
  USB: serial: xr: add support for XR21V1412 and XR21V1414
  USB: serial: ti_usb_3410_5052: clean up termios CSIZE handling
  USB: serial: ti_usb_3410_5052: use kernel types consistently
  USB: serial: ti_usb_3410_5052: add port-command helpers
  USB: serial: ti_usb_3410_5052: clean up vendor-request helpers
  USB: serial: ti_usb_3410_5052: drop unnecessary packed attributes
  USB: serial: io_ti: drop unnecessary packed attributes
  USB: serial: io_ti: use kernel types consistently
  USB: serial: io_ti: add read-port-command helper
  ...

usb: dwc3: gadget: Remove FS bInterval_m1 limitation

The programming guide incorrectly stated that the DCFG.bInterval_m1 must
be set to 0 when operating in fullspeed. There's no such limitation for
all IPs. See DWC_usb3x programming guide section 3.2.2.1.

Fixes: a1679af85b2a ("usb: dwc3: gadget: Fix setting of DEPCFG.bInterval_m1")
Cc: <stable@vger.kernel.org>
Acked-by: Felipe Balbi <balbi@kernel.org>
Signed-off-by: Thinh Nguyen <Thinh.Nguyen@synopsys.com>
Link: https://lore.kernel.org/r/5d4139ae89d810eb0a2d8577fb096fc88e87bfab.1618472454.git.Thinh.Nguyen@synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: xhci-mtk: remove bus status check

PM will take care of the status of child device, so no need
check each port anymore.

Suggested-by: Ikjoon Jang <ikjn@chromium.org>
Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com>
Link: https://lore.kernel.org/r/1618555706-6810-1-git-send-email-chunfeng.yun@mediatek.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

dt-bindings: connector: Add slow-charger-loop definition

Allows PMIC charger loops which are slow(i.e. cannot meet the
15ms deadline) to still comply to pSnkStby i.e Maximum power
that can be consumed by sink while in Sink Standby state as defined
in 7.4.2 Sink Electrical Parameters of USB Power Delivery Specification
Revision 3.0, Version 1.2.

This patch introduces slow-charger-loop which when set makes
the port request PD_P_SNK_STDBY_MW(2.5W i.e 500mA@5V) upon entering
SNK_DISCOVERY (instead of 3A or the 1.5A during SNK_DISCOVERY) and the
actual currrent limit after RX of PD_CTRL_PSRDY for PD link or during
SNK_READY for non-pd link.

Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Badhri Jagan Sridharan <badhri@google.com>
Link: https://lore.kernel.org/r/20210414142656.63749-3-badhri@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Enable RPi in ACPI mode

The dwc2 driver has everything we need to run
in ACPI mode except for the ACPI module device table
boilerplate. With that added and identified as "BCM2848",
an id in use by other OSs for this device, the dw2
controller on the BCM2711 will work.

Signed-off-by: Jeremy Linton <jeremy.linton@arm.com>
Link: https://lore.kernel.org/r/20210413215834.3126447-2-jeremy.linton@arm.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

dt-bindings: usb: dwc3: Add disabling LPM for gadget

Add a new DT option to disable LPM for gadget and update the description
for usb2-lpm-disable related to host for clarity.

Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Thinh Nguyen <Thinh.Nguyen@synopsys.com>
Link: https://lore.kernel.org/r/f31348ba744318c83b3a9ab1eab75c61122b15ae.1618366071.git.Thinh.Nguyen@synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

USB: CDC-ACM: fix poison/unpoison imbalance

suspend() does its poisoning conditionally, resume() does it
unconditionally. On a device with combined interfaces this
will balance, on a device with two interfaces the counter will
go negative and resubmission will fail.

Both actions need to be done conditionally.

Fixes: 6069e3e927c8f ("USB: cdc-acm: untangle a circular dependency between callback and softint")
Signed-off-by: Oliver Neukum <oneukum@suse.com>
Cc: stable <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20210421074513.4327-1-oneukum@suse.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

USB: Add reset-resume quirk for WD19's Realtek Hub

Realtek Hub (0bda:5487) in Dell Dock WD19 sometimes fails to work
after the system resumes from suspend with remote wakeup enabled
device connected:
[ 1947.640907] hub 5-2.3:1.0: hub_ext_port_status failed (err = -71)
[ 1947.641208] usb 5-2.3-port5: cannot disable (err = -71)
[ 1947.641401] hub 5-2.3:1.0: hub_ext_port_status failed (err = -71)
[ 1947.641450] usb 5-2.3-port4: cannot reset (err = -71)

Information of this hub:
T:  Bus=01 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#= 10 Spd=480  MxCh= 5
D:  Ver= 2.10 Cls=09(hub  ) Sub=00 Prot=02 MxPS=64 #Cfgs=  1
P:  Vendor=0bda ProdID=5487 Rev= 1.47
S:  Manufacturer=Dell Inc.
S:  Product=Dell dock
C:* #Ifs= 1 Cfg#= 1 Atr=e0 MxPwr=  0mA
I:  If#= 0 Alt= 0 #EPs= 1 Cls=09(hub  ) Sub=00 Prot=01 Driver=hub
E:  Ad=81(I) Atr=03(Int.) MxPS=   1 Ivl=256ms
I:* If#= 0 Alt= 1 #EPs= 1 Cls=09(hub  ) Sub=00 Prot=02 Driver=hub
E:  Ad=81(I) Atr=03(Int.) MxPS=   1 Ivl=256ms

The failure results from the ETIMEDOUT by chance when turning on
the suspend feature for the specified port of the hub. The port
seems to be in an unknown state so the hub_activate during resume
fails the hub_port_status, then the hub will fail to work.

The quirky hub needs the reset-resume quirk to function correctly.

Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Chris Chiu <chris.chiu@canonical.com>
Cc: stable <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20210420174651.6202-1-chris.chiu@canonical.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: gadget: net2272: remove redundant initialization of status

The variable status is being initialized with a value that is
never read and it is being updated later with a new value.  The
initialization is redundant and can be removed and move the
declaration of status to the scope where it is used.

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Addresses-Coverity: ("Unused value")
Link: https://lore.kernel.org/r/20210420110622.377339-1-colin.king@canonical.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: storage: datafab: remove redundant assignment of variable result

The variable result is being assigned with a value that is
never read, the assignment is redundant and can be removed.

Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Addresses-Coverity: ("Unused value")
Link: https://lore.kernel.org/r/20210420113818.378478-1-colin.king@canonical.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: misc: adutux: fix whitespace coding style issue

Correct missing space error ceckpatch.pl is complaining about.

Signed-off-by: Malte Deiseroth <mdeiseroth88@gmail.com>
Link: https://lore.kernel.org/r/20210416080843.GA137657@utop
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: musb: musb_core: Add space after that ','

Fix Error reported by checkpatch.pl

ERROR: space required after that ',' (ctx:VxV)
+#define can_bulk_split(musb,type) \
                            ^

ERROR: space required after that ',' (ctx:VxV)
+#define can_bulk_combine(musb,type) \
                              ^

Signed-off-by: Saurav Girepunje <saurav.girepunje@google.com>
Link: https://lore.kernel.org/r/20210418174720.GA59520@user
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: gadget: dummy_hcd: fix gpf in gadget_setup

Fix a general protection fault reported by syzbot due to a race between
gadget_setup() and gadget_unbind() in raw_gadget.

The gadget core is supposed to guarantee that there won't be any more
callbacks to the gadget driver once the driver's unbind routine is
called. That guarantee is enforced in usb_gadget_remove_driver as
follows:

        usb_gadget_disconnect(udc->gadget);
        if (udc->gadget->irq)
                synchronize_irq(udc->gadget->irq);
        udc->driver->unbind(udc->gadget);
        usb_gadget_udc_stop(udc);

usb_gadget_disconnect turns off the pullup resistor, telling the host
that the gadget is no longer connected and preventing the transmission
of any more USB packets. Any packets that have already been received
are sure to processed by the UDC driver's interrupt handler by the time
synchronize_irq returns.

But this doesn't work with dummy_hcd, because dummy_hcd doesn't use
interrupts; it uses a timer instead. It does have code to emulate the
effect of synchronize_irq, but that code doesn't get invoked at the
right time -- it currently runs in usb_gadget_udc_stop, after the unbind
callback instead of before. Indeed, there's no way for
usb_gadget_remove_driver to invoke this code before the unbind callback.

To fix this, move the synchronize_irq() emulation code to dummy_pullup
so that it runs before unbind. Also, add a comment explaining why it is
necessary to have it there.

Reported-by: syzbot+eb4674092e6cc8d9e0bd@syzkaller.appspotmail.com
Suggested-by: Alan Stern <stern@rowland.harvard.edu>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Anirudh Rayabharam <mail@anirudhrb.com>
Link: https://lore.kernel.org/r/20210419033713.3021-1-mail@anirudhrb.com
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc3: gadget: Fix START_TRANSFER link state check

The START_TRANSFER command needs to be executed while in ON/U0 link
state (with an exception during register initialization). Don't use
dwc->link_state to check this since the driver only tracks the link
state when the link state change interrupt is enabled. Check the link
state from DSTS register instead.

Note that often the host already brings the device out of low power
before it sends/requests the next transfer. So, the user won't see any
issue when the device starts transfer then. This issue is more
noticeable in cases when the device delays starting transfer, which can
happen during delayed control status after the host put the device in
low power.

Fixes: 799e9dc82968 ("usb: dwc3: gadget: conditionally disable Link State change events")
Cc: <stable@vger.kernel.org>
Acked-by: Felipe Balbi <balbi@kernel.org>
Signed-off-by: Thinh Nguyen <Thinh.Nguyen@synopsys.com>
Link: https://lore.kernel.org/r/bcefaa9ecbc3e1936858c0baa14de6612960e909.1618884221.git.Thinh.Nguyen@synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: gadget: Fix double free of device descriptor pointers

Upon driver unbind usb_free_all_descriptors() function frees all
speed descriptor pointers without setting them to NULL. In case
gadget speed changes (i.e from super speed plus to super speed)
after driver unbind only upto super speed descriptor pointers get
populated. Super speed plus desc still holds the stale (already
freed) pointer. Fix this issue by setting all descriptor pointers
to NULL after freeing them in usb_free_all_descriptors().

Fixes: f5c61225cf29 ("usb: gadget: Update function for SuperSpeedPlus")
cc: stable@vger.kernel.org
Reviewed-by: Peter Chen <peter.chen@kernel.org>
Signed-off-by: Hemant Kumar <hemantk@codeaurora.org>
Signed-off-by: Wesley Cheng <wcheng@codeaurora.org>
Link: https://lore.kernel.org/r/1619034452-17334-1-git-send-email-wcheng@codeaurora.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: typec: tcpm: Fix error while calculating PPS out values

"usb: typec: tcpm: Address incorrect values of tcpm psy for pps supply"
introduced a regression for req_out_volt and req_op_curr calculation.

req_out_volt should consider the newly calculated max voltage instead
of previously accepted max voltage by the port partner. Likewise,
req_op_curr should consider the newly calculated max current instead
of previously accepted max current by the port partner.

Fixes: e3a072022487 ("usb: typec: tcpm: Address incorrect values of tcpm psy for pps supply")
Reviewed-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Badhri Jagan Sridharan <badhri@google.com>
Link: https://lore.kernel.org/r/20210415050121.1928298-1-badhri@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: xhci-mtk: remove unused members

Now some members about phys and wakeup are not used anymore,
remove them.

Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com>
Link: https://lore.kernel.org/r/1618031406-15347-5-git-send-email-chunfeng.yun@mediatek.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: xhci-mtk: use clock bulk to get clocks

Use clock bulk helpers to get/enable/disable clocks, meanwhile
make sys_ck optional, then will be easier to handle clocks.

Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com>
Link: https://lore.kernel.org/r/1618031406-15347-4-git-send-email-chunfeng.yun@mediatek.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: xhci-mtk: add support runtime PM

A dedicated wakeup irq will be used to handle runtime suspend/resume,
we use dev_pm_set_dedicated_wake_irq API to take care of requesting
and attaching wakeup irq, then the suspend/resume framework will help
to enable/disable wakeup irq.

The runtime PM is default off since some platforms may not support it.
users can enable it via power/control (set "auto") in sysfs.

Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com>
Link: https://lore.kernel.org/r/1618031406-15347-3-git-send-email-chunfeng.yun@mediatek.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

dt-bindings: usb: mtk-xhci: add wakeup interrupt

Add an interrupt which is EINT usually to support runtime PM,
meanwhile add "interrupt-names" property, for backward
compatibility, it's optional and used when wakeup interrupt
exists

Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com>
Link: https://lore.kernel.org/r/1618031406-15347-2-git-send-email-chunfeng.yun@mediatek.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: xhci-mtk: check return value in suspend/resume hooks

Return error number if encounter errors during suspend and resume.

Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com>
Link: https://lore.kernel.org/r/1618031406-15347-1-git-send-email-chunfeng.yun@mediatek.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: typec: tcpm: Allow slow charging loops to comply to pSnkStby

When a PD charger advertising Rp-3.0 is connected to a sink port, the
sink port current limit would 3A, during SNK_DISCOVERY, till power
negotiation starts. Once the negotiation starts the power limit needs
to drop down to pSnkStby(500mA @ 5V) and to negotiated current limit
once the explicit contract is in place. Not all charging loops can ramp
up to 3A and drop down to 500mA within tSnkStdby which is 15ms. The port
partner might hard reset if tSnkStdby is not met.

To solve this problem, this patch introduces slow-charger-loop which
when set makes the port request PD_P_SNK_STDBY_MW upon entering
SNK_DISCOVERY(instead of 3A or the 1.5A during SNK_DISCOVERY) and the
actual currrent limit after RX of PD_CTRL_PSRDY for PD link or during
SNK_READY for non-pd link.

Reviewed-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Signed-off-by: Badhri Jagan Sridharan <badhri@google.com>
Link: https://lore.kernel.org/r/20210414024000.4175263-2-badhri@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: typec: tcpm: Honour pSnkStdby requirement during negotiation

>From PD Spec:
The Sink Shall transition to Sink Standby before a positive or
negative voltage transition of VBUS. During Sink Standby
the Sink Shall reduce its power draw to pSnkStdby. This allows
the Source to manage the voltage transition as well as
supply sufficient operating current to the Sink to maintain PD
operation during the transition. The Sink Shall
complete this transition to Sink Standby within tSnkStdby
after evaluating the Accept Message from the Source. The
transition when returning to Sink operation from Sink Standby
Shall be completed within tSnkNewPower. The
pSnkStdby requirement Shall only apply if the Sink power draw
is higher than this level.

The above requirement needs to be met to prevent hard resets
from port partner.

Without the patch: (5V/3A during SNK_DISCOVERY all the way through
explicit contract)
[   95.711984] CC1: 0 -> 0, CC2: 0 -> 5 [state TOGGLING, polarity 0, connected]
[   95.712007] state change TOGGLING -> SNK_ATTACH_WAIT [rev3 NONE_AMS]
[   95.712017] pending state change SNK_ATTACH_WAIT -> SNK_DEBOUNCED @ 170 ms [rev3 NONE_AMS]
[   95.837190] VBUS on
[   95.882075] state change SNK_ATTACH_WAIT -> SNK_DEBOUNCED [delayed 170 ms]
[   95.882082] state change SNK_DEBOUNCED -> SNK_ATTACHED [rev3 NONE_AMS]
[   95.882086] polarity 1
[   95.883151] set_auto_vbus_discharge_threshold mode:0 pps_active:n vbus:5000 ret:0
[   95.883441] enable vbus discharge ret:0
[   95.883445] Requesting mux state 1, usb-role 2, orientation 2
[   95.883776] state change SNK_ATTACHED -> SNK_STARTUP [rev3 NONE_AMS]
[   95.883879] pending state change SNK_STARTUP -> SNK_DISCOVERY @ 500 ms [rev3 NONE_AMS]
[   96.038960] VBUS on
[   96.383939] state change SNK_STARTUP -> SNK_DISCOVERY [delayed 500 ms]
[   96.383946] Setting voltage/current limit 5000 mV 3000 mA
[   96.383961] vbus=0 charge:=1
[   96.386044] state change SNK_DISCOVERY -> SNK_WAIT_CAPABILITIES [rev3 NONE_AMS]
[   96.386309] pending state change SNK_WAIT_CAPABILITIES -> HARD_RESET_SEND @ 450 ms [rev3 NONE_AMS]
[   96.394404] PD RX, header: 0x2161 [1]
[   96.394408]  PDO 0: type 0, 5000 mV, 3000 mA [E]
[   96.394410]  PDO 1: type 0, 9000 mV, 2000 mA []
[   96.394412] state change SNK_WAIT_CAPABILITIES -> SNK_NEGOTIATE_CAPABILITIES [rev2 POWER_NEGOTIATION]
[   96.394416] Setting usb_comm capable false
[   96.395083] cc=0 cc1=0 cc2=5 vbus=0 vconn=sink polarity=1
[   96.395089] Requesting PDO 1: 9000 mV, 2000 mA
[   96.395093] PD TX, header: 0x1042
[   96.397404] PD TX complete, status: 0
[   96.397424] pending state change SNK_NEGOTIATE_CAPABILITIES -> HARD_RESET_SEND @ 60 ms [rev2 POWER_NEGOTIATION]
[   96.400826] PD RX, header: 0x363 [1]
[   96.400829] state change SNK_NEGOTIATE_CAPABILITIES -> SNK_TRANSITION_SINK [rev2 POWER_NEGOTIATION]
[   96.400832] pending state change SNK_TRANSITION_SINK -> HARD_RESET_SEND @ 500 ms [rev2 POWER_NEGOTIATION]
[   96.577315] PD RX, header: 0x566 [1]
[   96.577321] Setting voltage/current limit 9000 mV 2000 mA
[   96.578363] set_auto_vbus_discharge_threshold mode:3 pps_active:n vbus:9000 ret:0
[   96.578370] state change SNK_TRANSITION_SINK -> SNK_READY [rev2 POWER_NEGOTIATION]

With the patch:
[  168.398573] CC1: 0 -> 0, CC2: 0 -> 5 [state TOGGLING, polarity 0, connected]
[  168.398605] state change TOGGLING -> SNK_ATTACH_WAIT [rev3 NONE_AMS]
[  168.398619] pending state change SNK_ATTACH_WAIT -> SNK_DEBOUNCED @ 170 ms [rev3 NONE_AMS]
[  168.522348] VBUS on
[  168.568676] state change SNK_ATTACH_WAIT -> SNK_DEBOUNCED [delayed 170 ms]
[  168.568684] state change SNK_DEBOUNCED -> SNK_ATTACHED [rev3 NONE_AMS]
[  168.568688] polarity 1
[  168.569867] set_auto_vbus_discharge_threshold mode:0 pps_active:n vbus:5000 ret:0
[  168.570158] enable vbus discharge ret:0
[  168.570161] Requesting mux state 1, usb-role 2, orientation 2
[  168.570504] state change SNK_ATTACHED -> SNK_STARTUP [rev3 NONE_AMS]
[  168.570634] pending state change SNK_STARTUP -> SNK_DISCOVERY @ 500 ms [rev3 NONE_AMS]
[  169.070689] state change SNK_STARTUP -> SNK_DISCOVERY [delayed 500 ms]
[  169.070695] Setting voltage/current limit 5000 mV 3000 mA
[  169.070702] vbus=0 charge:=1
[  169.072719] state change SNK_DISCOVERY -> SNK_WAIT_CAPABILITIES [rev3 NONE_AMS]
[  169.073145] pending state change SNK_WAIT_CAPABILITIES -> HARD_RESET_SEND @ 450 ms [rev3 NONE_AMS]
[  169.077162] PD RX, header: 0x2161 [1]
[  169.077172]  PDO 0: type 0, 5000 mV, 3000 mA [E]
[  169.077178]  PDO 1: type 0, 9000 mV, 2000 mA []
[  169.077183] state change SNK_WAIT_CAPABILITIES -> SNK_NEGOTIATE_CAPABILITIES [rev2 POWER_NEGOTIATION]
[  169.077191] Setting usb_comm capable false
[  169.077753] cc=0 cc1=0 cc2=5 vbus=0 vconn=sink polarity=1
[  169.077759] Requesting PDO 1: 9000 mV, 2000 mA
[  169.077762] PD TX, header: 0x1042
[  169.079990] PD TX complete, status: 0
[  169.080013] pending state change SNK_NEGOTIATE_CAPABILITIES -> HARD_RESET_SEND @ 60 ms [rev2 POWER_NEGOTIATION]
[  169.083183] VBUS on
[  169.084195] PD RX, header: 0x363 [1]
[  169.084200] state change SNK_NEGOTIATE_CAPABILITIES -> SNK_TRANSITION_SINK [rev2 POWER_NEGOTIATION]
[  169.084206] Setting standby current 5000 mV @ 500 mA
[  169.084209] Setting voltage/current limit 5000 mV 500 mA
[  169.084220] pending state change SNK_TRANSITION_SINK -> HARD_RESET_SEND @ 500 ms [rev2 POWER_NEGOTIATION]
[  169.260222] PD RX, header: 0x566 [1]
[  169.260227] Setting voltage/current limit 9000 mV 2000 mA
[  169.261315] set_auto_vbus_discharge_threshold mode:3 pps_active:n vbus:9000 ret:0
[  169.261321] state change SNK_TRANSITION_SINK -> SNK_READY [rev2 POWER_NEGOTIATION]
[  169.261570] AMS POWER_NEGOTIATION finished

Fixes: f0690a25a140b ("staging: typec: USB Type-C Port Manager (tcpm)")
Reviewed-by: Guenter Roeck <linux@roeck-us.net>
Reviewed-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Signed-off-by: Badhri Jagan Sridharan <badhri@google.com>
Link: https://lore.kernel.org/r/20210414024000.4175263-1-badhri@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc3: gadget: Check for disabled LPM quirk

If the device doesn't support LPM, make sure to disable the LPM
capability and don't advertise to the host that it supports it.

Acked-by: Felipe Balbi <balbi@kernel.org>
Signed-off-by: Thinh Nguyen <Thinh.Nguyen@synopsys.com>
Link: https://lore.kernel.org/r/9e68527ff932b1646f92a7593d4092a903754666.1618366071.git.Thinh.Nguyen@synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc3: core: Add shutdown callback for dwc3

This patch adds a shutdown callback to USB DWC core driver to ensure that
it is properly shutdown in reboot/shutdown path. This is required
where SMMU address translation is enabled like on SC7180
SoC and few others. If the hardware is still accessing memory after
SMMU translation is disabled as part of SMMU shutdown callback in
system reboot or shutdown path, then IOVAs(I/O virtual address)
which it was using will go on the bus as the physical addresses which
might result in unknown crashes (NoC/interconnect errors).

Acked-by: Felipe Balbi <balbi@kernel.org>
Signed-off-by: Sandeep Maheswaram <sanm@codeaurora.org>
Link: https://lore.kernel.org/r/1618380209-20114-1-git-send-email-sanm@codeaurora.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc3: gadget: Ignore Packet Pending bit

Currently the controller handles single stream only. So, Ignore Packet
Pending bit for stream selection and don't search for another stream if
the host sends Data Packet with PP=0 (for OUT direction) or ACK with
NumP=0 and PP=0 (for IN direction). This slightly improves the stream
performance.

Acked-by: Felipe Balbi <balbi@kernel.org>
Signed-off-by: Thinh Nguyen <Thinh.Nguyen@synopsys.com>
Link: https://lore.kernel.org/r/097ba9e104c143f7ba0195ebff29390ec3043692.1618282705.git.Thinh.Nguyen@synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

USB: Add LPM quirk for Lenovo ThinkPad USB-C Dock Gen2 Ethernet

This is another branded 8153 device that doesn't work well with LPM
enabled:
[ 400.597506] r8152 5-1.1:1.0 enx482ae3a2a6f0: Tx status -71

So disable LPM to resolve the issue.

Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
BugLink: https://bugs.launchpad.net/bugs/1922651
Link: https://lore.kernel.org/r/20210412135455.791971-1-kai.heng.feng@canonical.com
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: typec: silence a static checker warning

Smatch complains about a potential missing error code:

    drivers/usb/typec/port-mapper.c:168 typec_link_port()
    warn: missing error code 'ret'

This is a false positive and returning zero is intentional.  Let's
re-arrange the code to silence the warning and make the intent more
clear.

Reviewed-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Link: https://lore.kernel.org/r/YHadaACH8Mq/10F7@mwanda
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

USB: cdc-acm: add more Maxlinear/Exar models to ignore list

Now that the xr_serial got support for other models, add their USB IDs
as well.

The Maxlinear/Exar USB UARTs can be used in either ACM mode using the
cdc-acm driver or in "custom driver" mode in which further features such
as hardware and software flow control, GPIO control and in-band
line-status reporting are available.

In ACM mode the device always enables RTS/CTS flow control, something
which could prevent transmission in case the CTS input isn't wired up
correctly.

Ensure that cdc_acm will not bind to these devices if the custom
USB-serial driver is enabled.

Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/5155887a764cbc11f8da0217fe08a24a77d120b4.1616571453.git.mchehab+huawei@kernel.org
[ johan: rewrite commit message, clean up entries ]
Cc: Oliver Neukum <oneukum@suse.com>
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: xr: add copyright notice

Add another copyright notice for the work done in 2021.

Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: xr: reset FIFOs on open

Reset the transmit and receive FIFOs before enabling the UARTs as part
of open() in order to flush any stale data.

Note that the XR21V141X needs a type-specific implementation due to its
UART Manager registers.

Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: xr: add support for XR22801, XR22802, XR22804

The XR22801, XR22802 and XR22804 are compound devices with an embedded
hub and up to seven downstream USB devices including one, two or four
UARTs respectively.

The UART function is similar to XR21B142X but most registers are offset
by 0x40, the register requests are different and are directed at the
device rather than interface, and 5 and 6-bit words are not supported.

Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: xr: add support for XR21B1411

The single-port XR21B1411 is similar to the XR21B142X type but uses
12-bit registers and 16-bit register addresses, the register requests
are different and are directed at the device rather than interface, and
5 and 6-bit words are not supported.

The register layout is very similar to XR21B142X except that most
registers are offset by 0xc00 (corresponding to a channel index of 12 in
the MSB of wIndex). As the device is single-port so that the derived
channel index is 0, the current register accessors can be reused after
simply changing the address width.

Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: xr: add support for XR21B1421, XR21B1422 and XR21B1424

The XR21B1421, XR21B1422 and XR21B1424 are the one-, two- and four-port
models of a second XR21B142X type of the Maxlinear/Exar USB UARTs.

The XR21B142X type differs from XR21V141X in several ways, including:

- register layout
- register width (16-bit instead of 8-bit)
- vendor register requests
- UART enable/disable sequence
- custom-driver mode flag
- three additional GPIOs (9 instead of 6)

As for XR21V141X, the XR21B142X vendor requests encode the channel index
in the MSB of wIndex, but it lacks the UART Manager registers which
have been replaced by regular UART registers. The new type also uses the
interface number of the control interface (0, 2, 4, 6) as channel index
instead of the channel number (0, 1, 2, 3).

The XR21B142X lacks the divisor and format registers used by XR21V141X
and instead uses the CDC SET_LINE_CONTROL request to configure the line
settings.

Note that the currently supported XR21V141X type lacks the custom-driver
mode flag that prevents the device from entering CDC-ACM mode when a CDC
requests is received. This specifically means that the SET_LINE_CONTROL
request cannot be used with XR21V141X even though it is otherwise
supported.

The UART enable sequence for XR21B142X does not involve explicitly
enabling the FIFOs, but according to datasheet the UART must be disabled
when writing any register but GPIO_SET, GPIO_CLEAR, TX_BREAK and
ERROR_STATUS.

Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: xr: add type abstraction

There are at least four types of Maxlinear/Exar USB UARTs which differ
in various ways such as in their register layouts:

XR21V141X
XR21B142X
XR21B1411
XR22804

It is not clear whether the device type can be inferred from the
descriptors so encode it in the device-id table for now.

Add a type structure that can be used to abstract the register layout
and other features, and use it when accessing the XR21V141X UART
registers that are shared by all types.

Note that the currently supported XR21V141X type is the only type that
has a set of UART Manager registers and that these will need to be
handled specifically.

Similarly, XR21V141X is the only type which has the divisor registers
and that needs to use the format register when configuring the line
settings.

Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: xr: drop type prefix from shared defines

In preparation for adding support for further types, drop the type
prefix from defines that are not specific to XR21V141X.

Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: xr: move pin configuration to probe

There's no need to configure the pins on every open and judging from the
vendor driver and datasheet it can be done before enabling the UART.

Move pin configuration from open() to port probe and make sure to
deassert DTR and RTS after configuring all pins as GPIO.

Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: xr: rename GPIO-pin defines

Rename the GPIO-pin defines so that they reflect how they are used.

Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: xr: rename GPIO-mode defines

Rename the GPIO mode defines so that they reflect the datasheet and how
they are used.

Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: xr: add support for XR21V1412 and XR21V1414

Add support for the two- and four-port variants of XR21V1410.

Use the interface number of each control interface (e.g. 0, 2, 4, 6) to
derive the zero-based channel index:

XR21V1410 0
XR21V1412 0, 1
XR21V1414 0, 1, 2, 3

Note that the UART registers reside in separate blocks per channel,
while the UART Manager functionality is implemented using per-channel
registers.

Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: ti_usb_3410_5052: clean up termios CSIZE handling

Remove the random white space from the CSIZE switch.

Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: ti_usb_3410_5052: use kernel types consistently

Replace the remaining uses of user-space __XX types.

Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: ti_usb_3410_5052: add port-command helpers

Add two port-command helpers to handle the UART module-id parameter
instead of open coding.

Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: ti_usb_3410_5052: clean up vendor-request helpers

Make the vendor-request helpers data parameters be void pointers and
drop the caller casts.

Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: ti_usb_3410_5052: drop unnecessary packed attributes

Drop unnecessary packed attributes from structures that don't need it.

Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: io_ti: drop unnecessary packed attributes

Drop unnecessary packed attributes from structures that don't need it
and use the __packed macro consistently.

Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: io_ti: use kernel types consistently

Use kernel types consistently by replacing the remaining __uXX types.

Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: io_ti: add read-port-command helper

Add a read-port-command helper analogous to the send-port-command
helper to take care of the UART module id instead of open coding.

Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: io_ti: add send-port-command helper

Add a send-port-command helper which takes care of determining the UART
module id when sending commands instead of doing so at every call site.

Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: io_ti: clean up vendor-request helpers

Clean up the vendor-request helpers by using kernel-types consistently
and using void pointers for the data arguments, which allows removing
a cast from one caller.

Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: ti_usb_3410_5052: drop drain delay for 3410

Unlike the TUSB5052, the TUSB3410 has an LSR TEMT bit to tell if both
the transmitter data and shift registers are empty.

Make sure to check also the shift register on TUSB3410 when waiting for
the transmit buffer to drain during close and drop the time-based
one-char delay which is otherwise needed (e.g. 90 ms at 110 bps).

Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: ti_usb_3410_5052: reduce drain delay to one char

The three-character drain delay was added by commit f1175daa5312 ("USB:
ti_usb_3410_5052: kill custom closing_wait") when removing the custom
closing-wait implementation, which used a fixed 20 ms poll period and
drain delay.

This was likely a bit too conservative as a one-character timeout (e.g.
33 ms at 300 bps) should be enough to compensate for the lack of a
transmitter empty bit in the TUSB5052 line-status register.

Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: io_ti: document reason for drain delay

Document that the device line-status register doesn't tell when the
transmitter shift register has emptied and that this is why the
one-character drain delay is needed.

Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: f81232: drop time-based drain delay

The f81232 driver now waits for the transmit FIFO to drain during close
so there is no need to keep the time-based drain delay, which would add
up to two seconds on every close for low line speeds.

Fixes: 98405f81036d ("USB: serial: f81232: add tx_empty function")
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>

usb: dwc2: Add exit clock gating before removing driver

When dwc2 core is in clock gating mode loading driver
again causes driver fail. Because in that mode
registers are not accessible.

Added a flow of exiting clock gating mode
to avoid the driver reload failure.

Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210413073731.3C81BA022E@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Add clock gating exiting flow by system resume

If not hibernation nor partial power down are supported,
port resume is done using the clock gating programming flow.

Adds a new flow of exiting clock gating when PC is
resumed.

Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210413073723.BA0FEA022E@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Add clock gating entering flow by system suspend

If not hibernation nor partial power down are supported,
clock gating is used to save power.

Adds a new flow of entering clock gating when PC is
suspended.

Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210413073716.30C13A0094@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Allow exit clock gating in urb enqueue

When core is in clock gating state and an external
hub is connected, upper layer sends URB enqueue request,
which results in port reset issue.

Added exit from clock gating state to avoid port
reset issue and process upper layer request properly.

Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210413073708.ADFC6A0094@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Update exit clock gating when port is resumed

Updates the implementation of exiting clock gating mode
when core receives port resume.
Instead of setting the required bit fields of the registers
inline, called the "dwc2_host_exit_clock_gating()" function.

Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210413073701.367E0A022E@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Update enter clock gating when port is suspended

Updates the implementation of entering clock gating mode
when core receives port suspend.
Instead of setting the required bit fields of the registers
inline, called the "dwc2_host_enter_clock_gating()" function.

Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210413073653.9F493A0094@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Add exit clock gating when port reset is asserted

Adds clock gating exit flow when set port feature
reset is received in suspended state.

Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210413073646.27217A0094@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Add exit clock gating from session request interrupt

Added clock gating exit flow from session
request interrupt handler according programming guide.

Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210413073638.921E8A0099@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Add exit clock gating from wakeup interrupt

Added exit from clock gating mode when wakeup interrupt
is detected. To exit from the clock gating
in device mode "dwc2_gadget_exit_clock_gating()"
function is used with rem_wakeup parameter 0. To exit
clock gating in host mode "dwc2_host_exit_clock_gating()"
with rem_wakeup parameter 1.

Acked-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210413073630.EF2CEA0094@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Allow entering clock gating from USB_SUSPEND interrupt

If core doesn't support hibernation or partial power
down power saving options, power can still be saved
using clock gating on all the clocks.

- Added entering clock gating state from USB_SUSPEND
  interrupt.

Acked-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210413073623.65355A022E@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Add host clock gating support functions

Added host clock gating support functions according
programming guide.

Added function names:
dwc2_host_enter_clock_gating()
dwc2_host_exit_clock_gating()

Acked-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210413073615.B3E84A022E@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc2: Add device clock gating support functions

Added device clock gating support functions according
programming guide.

Moved "bus_suspended" flag to "dwc2_hsotg" struct because
we need to set that flag while entering to clock gating
in case when the driver is built in peripheral mode.

Added function names:
dwc2_gadget_enter_clock_gating()
dwc2_gadget_exit_clock_gating()

Acked-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Signed-off-by: Artur Petrosyan <Arthur.Petrosyan@synopsys.com>
Link: https://lore.kernel.org/r/20210413073607.F41E8A0094@mailhost.synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Merge tag 'thunderbolt-for-v5.13-rc1' of git://git./linux/kernel/git/westeri/thunderbolt into usb-next

Mika writes:

thunderbolt: Changes for v5.13 merge window

This includes following Thunderbolt/USB4 changes for v5.13 merge window:

  * Debugfs improvements

  * Align the inter-domain (peer-to-peer) support with the USB4
    inter-domain spec for better interoperability

  * Add support for USB4 DROM and the new product descriptor

  * More KUnit tests

  * Detailed uevent for routers

  * Few miscellaneous improvements

All these have been in linux-next without reported issues.

* tag 'thunderbolt-for-v5.13-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/westeri/thunderbolt: (24 commits)
  thunderbolt: Hide authorized attribute if router does not support PCIe tunnels
  thunderbolt: Add details to router uevent
  thunderbolt: Unlock on error path in tb_domain_add()
  thunderbolt: Add support for USB4 DROM
  thunderbolt: Check quirks in tb_switch_add()
  thunderbolt: Add KUnit tests for DMA tunnels
  thunderbolt: Add KUnit tests for XDomain properties
  net: thunderbolt: Align the driver to the USB4 networking spec
  thunderbolt: Allow multiple DMA tunnels over a single XDomain connection
  thunderbolt: Drop unused tb_port_set_initial_credits()
  thunderbolt: Use dedicated flow control for DMA tunnels
  thunderbolt: Add support for maxhopid XDomain property
  thunderbolt: Add tb_property_copy_dir()
  thunderbolt: Align XDomain protocol timeouts with the spec
  thunderbolt: Use pseudo-random number as initial property block generation
  thunderbolt: Do not re-establish XDomain DMA paths automatically
  thunderbolt: Add more logging to XDomain connections
  Documentation / thunderbolt: Drop speed/lanes entries for XDomain
  thunderbolt: Decrease control channel timeout for software connection manager
  thunderbolt: Do not pass timeout for tb_cfg_reset()
  ...

Merge tag 'usb-v5.13-rc1' of git://git./linux/kernel/git/peter.chen/usb into usb-next

Peter writes:

Several Cadence3 improvements are introduced in v5.13-rc1:
- Add recovery during resume if the controller was lost power at system suspend
- Reduce DMA memory footprint
- Other small improvements

* tag 'usb-v5.13-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/peter.chen/usb:
  usb: cdnsp: Fixes issue with Configure Endpoint command
  usb: cdnsp: remove redundant initialization of variable ret
  usb: cdns3: delete repeated clear operations
  usb: cdns3: Optimize DMA request buffer allocation
  usb: cdns3: Use dma_pool_* api to alloc trb pool
  usb: cdns3: fix static checker warning.
  usb: cdns3: imx: mark cdns_imx_system_resume as __maybe_unused
  usb: cdns3: trace: delete the trace parameter for request->trb
  usb: cdns3: imx: add power lost support for system resume
  usb: cdns3: add power lost support for system resume

Revert "usb: Link the ports to the connectors they are attached to"

This reverts commit 63cd78617350dae99cc5fbd8f643b83ee819fe33 as it
causes a build error:
depmod: ERROR: Cycle detected: usbcore -> typec -> usbcore
depmod: ERROR: Found 2 modules in dependency cycles!

Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Link: https://lore.kernel.org/r/20210412213655.3776e15e@canb.auug.org.au
Cc: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: cdnsp: Fixes issue with Configure Endpoint command

Patch adds flag EP_UNCONFIGURED to detect whether endpoint was
unconfigured. This flag is set in cdnsp_reset_device after Reset Device
command. Among others this command disables all non control endpoints.
Flag is used in cdnsp_gadget_ep_disable to protect controller against
invoking Configure Endpoint command on disabled endpoint. Lack of this
protection in some cases caused that Configure Endpoint command completed
with Context State Error code completion.

Fixes: 3d82904559f4 ("usb: cdnsp: cdns3 Add main part of Cadence USBSSP DRD Driver")
Signed-off-by: Pawel Laszczak <pawell@cadence.com>
Signed-off-by: Peter Chen <peter.chen@kernel.org>

usb: cdnsp: remove redundant initialization of variable ret

The variable ret is being initialized with a value that is
never read and it is being updated later with a new value.  The
initialization is redundant and can be removed.

Addresses-Coverity: ("Unused value")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Pawel Laszczak <pawell@cadence.com>
Signed-off-by: Peter Chen <peter.chen@kernel.org>

usb: cdns3: delete repeated clear operations

dma_alloc_coherent already zeroes out memory, so memset is not needed.

Signed-off-by: Wang Qing <wangqing@vivo.com>
Reviewed-by: Pawel Laszczak <pawell@cadence.com>
Acked-by: Pawel Laszczak <pawell@cadence.com>
Signed-off-by: Peter Chen <peter.chen@kernel.org>

usb: cdns3: Optimize DMA request buffer allocation

dma_alloc_coherent() might fail on the platform with a small
DMA region.

To avoid such failure in cdns3_prepare_aligned_request_buf(),
dma_alloc_coherent() is replaced with dma_alloc_noncoherent()
to allocate aligned request buffer of dynamic length.

Reported-by: Aswath Govindraju <a-govindraju@ti.com>
Signed-off-by: Sanket Parmar <sparmar@cadence.com>
Signed-off-by: Peter Chen <peter.chen@kernel.org>

usb: cdns3: Use dma_pool_* api to alloc trb pool

Allocation of DMA coherent memory in atomic context using
dma_alloc_coherent() might fail on platforms with smaller
DMA region.

To fix it, dma_alloc_coherent() is replaced with dma_pool
API to allocate a smaller chunk of DMA coherent memory for
TRB rings.

Reported-by: Aswath Govindraju <a-govindraju@ti.com>
Signed-off-by: Sanket Parmar <sparmar@cadence.com>
Signed-off-by: Peter Chen <peter.chen@kernel.org>

usb: cdns3: fix static checker warning.

The patch c450e48eb570: "usb: cdns3: add power lost support for
system resume" from Feb 18, 2021, leads to the following static
checker warning:

        drivers/usb/cdns3/core.c:551 cdns_resume()
        error: uninitialized symbol 'ret'.

drivers/usb/cdns3/core.c
544
545     if (!role_changed) {
546            if (cdns->role == USB_ROLE_HOST)
547                  ret = cdns_drd_host_on(cdns);
548     else if (cdns->role == USB_ROLE_DEVICE)
549            ret = cdns_drd_gadget_on(cdns);

"ret" is uninitialized at else branch.

Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Frank Li <frank.li@nxp.com>
Signed-off-by: Peter Chen <peter.chen@kernel.org>

usb: cdns3: imx: mark cdns_imx_system_resume as __maybe_unused

The function cdns_imx_system_resume() may have no callers depending
on configuration, so it must be marked __maybe_unused to avoid
harmless warning:

drivers/usb/cdns3/cdns3-imx.c:378:12: warning:
 'cdns_imx_system_resume' defined but not used [-Wunused-function]
  378 | static int cdns_imx_system_resume(struct device *dev)
      |            ^~~~~~~~~~~~~~~~~~~~~~

Fixes: 67982dfa59de ("usb: cdns3: imx: add power lost support for system resume")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: Peter Chen <peter.chen@kernel.org>

usb: cdns3: trace: delete the trace parameter for request->trb

It is not correct using %pa to print virtual address of request->trb, and
it is hard to print its physical address due to the virtual address is
zero before using. It could use index (start_trb/end_trb) to know the
current trb position, so no matter virtual address or physical address
for request-trb is not so meaningful.

Reported-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Peter Chen <peter.chen@kernel.org>

usb: cdns3: imx: add power lost support for system resume

imx need special handle when controller lost power

Signed-off-by: Peter Chen <peter.chen@nxp.com>
Signed-off-by: Frank Li <frank.li@nxp.com>
Signed-off-by: Peter Chen <peter.chen@kernel.org>

usb: cdns3: add power lost support for system resume

If the controller lost its power during the system suspend, we need
to do all initialiation operations.

Signed-off-by: Peter Chen <peter.chen@nxp.com>
Signed-off-by: Frank Li <frank.li@nxp.com>
Signed-off-by: Peter Chen <peter.chen@kernel.org>

USB: serial: cp210x: add gpio-configuration debug printk

Add a debug printk to dump the GPIO configuration stored in EEPROM
during probe.

Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: Johan Hovold <johan@kernel.org>

USB: serial: cp210x: provide gpio valid mask

Use the new GPIO valid-mask feature to inform gpiolib which pins are
available for use instead of handling that in a request callback.

This also allows user space to figure out which pins are available
through the chardev interface without having to request each pin in
turn.

Note that the return value when requesting an unavailable pin will now
be -EINVAL instead of -ENODEV.

Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: Johan Hovold <johan@kernel.org>