[ Upstream commit 33a06f1300a79cfd461cea0268f05e969d4f34ec ]
When gadget registration fails, one should not call usb_del_gadget_udc().
Ensure this by setting gadget->udc to NULL. Also in case of a failure
there is no need to disable low-level hardware, so return immiedetly
instead of jumping to error_init label.
This fixes the following kernel NULL ptr dereference on gadget failure
(can be easily triggered with g_mass_storage without any module
parameters):
dwc2 12480000.hsotg: dwc2_check_params: Invalid parameter besl=1
dwc2 12480000.hsotg: dwc2_check_params: Invalid parameter g_np_tx_fifo_size=1024
dwc2 12480000.hsotg: EPs: 16, dedicated fifos, 7808 entries in SPRAM
Mass Storage Function, version: 2009/09/11
LUN: removable file: (no medium)
no file given for LUN0
g_mass_storage 12480000.hsotg: failed to start g_mass_storage: -22
8<--- cut here ---
Unable to handle kernel NULL pointer dereference at virtual address 00000104
pgd = (ptrval)
[00000104] *pgd=00000000
Internal error: Oops: 805 [#1] PREEMPT SMP ARM
Modules linked in:
CPU: 0 PID: 12 Comm: kworker/0:1 Not tainted 5.8.0-rc5 #3133
Hardware name: Samsung Exynos (Flattened Device Tree)
Workqueue: events deferred_probe_work_func
PC is at usb_del_gadget_udc+0x38/0xc4
LR is at __mutex_lock+0x31c/0xb18
...
Process kworker/0:1 (pid: 12, stack limit = 0x(ptrval))
Stack: (0xef121db0 to 0xef122000)
...
[<c076bf3c>] (usb_del_gadget_udc) from [<c0726bec>] (dwc2_hsotg_remove+0x10/0x20)
[<c0726bec>] (dwc2_hsotg_remove) from [<c0711208>] (dwc2_driver_probe+0x57c/0x69c)
[<c0711208>] (dwc2_driver_probe) from [<c06247c0>] (platform_drv_probe+0x6c/0xa4)
[<c06247c0>] (platform_drv_probe) from [<c0621df4>] (really_probe+0x200/0x48c)
[<c0621df4>] (really_probe) from [<c06221e8>] (driver_probe_device+0x78/0x1fc)
[<c06221e8>] (driver_probe_device) from [<c061fcd4>] (bus_for_each_drv+0x74/0xb8)
[<c061fcd4>] (bus_for_each_drv) from [<c0621b54>] (__device_attach+0xd4/0x16c)
[<c0621b54>] (__device_attach) from [<c0620c98>] (bus_probe_device+0x88/0x90)
[<c0620c98>] (bus_probe_device) from [<c06211b0>] (deferred_probe_work_func+0x3c/0xd0)
[<c06211b0>] (deferred_probe_work_func) from [<c0149280>] (process_one_work+0x234/0x7dc)
[<c0149280>] (process_one_work) from [<c014986c>] (worker_thread+0x44/0x51c)
[<c014986c>] (worker_thread) from [<c0150b1c>] (kthread+0x158/0x1a0)
[<c0150b1c>] (kthread) from [<c0100114>] (ret_from_fork+0x14/0x20)
Exception stack(0xef121fb0 to 0xef121ff8)
...
---[ end trace 9724c2fc7cc9c982 ]---
While fixing this also fix the double call to dwc2_lowlevel_hw_disable()
if dr_mode is set to USB_DR_MODE_PERIPHERAL. In such case low-level
hardware is already disabled before calling usb_add_gadget_udc(). That
function correctly preserves low-level hardware state, there is no need
for the second unconditional dwc2_lowlevel_hw_disable() call.
Fixes: 207324a321a8 ("usb: dwc2: Postponed gadget registration to the udc class driver")
Acked-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Felipe Balbi <balbi@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 4fdf228cdf6925af45a2066d403821e0977bfddb upstream.
To avoid lot of interrupts from dwc2 core, which can be asserted in
specific conditions need to disable interrupts on HW level instead of
disable IRQs on Kernel level, because of IRQ can be shared between
drivers.
Cc: stable@vger.kernel.org
Fixes: a40a00318c ("usb: dwc2: add shutdown callback to platform variant")
Tested-by: Frank Mori Hess <fmh6jj@gmail.com>
Reviewed-by: Alan Stern <stern@rowland.harvard.edu>
Reviewed-by: Doug Anderson <dianders@chromium.org>
Reviewed-by: Frank Mori Hess <fmh6jj@gmail.com>
Signed-off-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Felipe Balbi <balbi@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 207324a321a866401b098cadf19e4a2dd6584622 upstream.
During dwc2 driver probe, after gadget registration to the udc class
driver, if exist any builtin function driver it immediately bound to
dwc2 and after init host side (dwc2_hcd_init()) stucked in host mode.
Patch postpone gadget registration after host side initialization done.
Fixes: 117777b2c3 ("usb: dwc2: Move gadget probe function into platform code")
Reported-by: kbuild test robot <lkp@intel.com>
Tested-by: Marek Vasut <marex@denx.de>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Minas Harutyunyan <hminas@synopsys.com>
Link: https://lore.kernel.org/r/f21cb38fecc72a230b86155d94c7e60c9cb66f58.1591690938.git.hminas@synopsys.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 8c935deacebb8fac8f41378701eb79d12f3c2e2d ]
When the remote wakeup interrupt is triggered, lx_state is resumed from L2
to L0 state. But when the gadget resume is called, lx_state is still L2.
This prevents the resume callback to queue any request. Any attempt
to queue a request from resume callback will result in:
- "submit request only in active state" debug message to be issued
- dwc2_hsotg_ep_queue() returns -EAGAIN
Call the gadget resume routine after the core is in L0 state.
Fixes: f81f46e1f5 ("usb: dwc2: implement hibernation during bus suspend/resume")
Acked-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Fabrice Gasnier <fabrice.gasnier@st.com>
Signed-off-by: Felipe Balbi <balbi@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 9a0d6f7c0a83844baae1d6d85482863d2bf3b7a7 upstream.
SET/CLEAR_FEATURE for Remote Wakeup allowance not handled correctly.
GET_STATUS handling provided not correct data on DATA Stage.
Issue seen when gadget's dr_mode set to "otg" mode and connected
to MacOS.
Both are fixed and tested using USBCV Ch.9 tests.
Signed-off-by: Minas Harutyunyan <hminas@synopsys.com>
Fixes: fa389a6d77 ("usb: dwc2: gadget: Add remote_wakeup_allowed flag")
Tested-by: Jack Mitchell <ml@embed.me.uk>
Cc: stable@vger.kernel.org
Signed-off-by: Felipe Balbi <balbi@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 5acb4b970184d189d901192d075997c933b82260 ]
Fix calculation of transfer size on completion in function
dwc2_gadget_get_xfersize_ddma().
Added increment of descriptor pointer to move to next descriptor in
the loop.
Fixes: aa3e8bc813 ("usb: dwc2: gadget: DDMA transfer start and complete")
Signed-off-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 6689f0f4bb14e50917ba42eb9b41c25e0184970c upstream.
Testing on different generations of Lantiq MIPS SoC based boards, showed
that it takes up to 1500 us until the core reset bit is cleared.
The driver from the vendor SDK (ifxhcd) uses a 1 second timeout. Use the
same timeout to fix wrong hang detections and make the driver work for
Lantiq MIPS SoCs.
At least till kernel 4.14 the hanging reset only caused a warning but
the driver was probed successful. With kernel 4.19 errors out with
EBUSY.
Cc: linux-stable <stable@vger.kernel.org> # 4.19+
Signed-off-by: Mathias Kresin <dev@kresin.me>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit c216765d3a1defda5e7e2dabd878f99f0cd2ebf2 ]
The bug would let the usb controller enter partial power down,
which was formally known as hibernate, upon boot if nothing was plugged
in to the port. Partial power down couldn't be exited properly, so any
usb devices plugged in after boot would not be usable.
Before the name change, params.hibernation was false by default, so
_dwc2_hcd_suspend() would skip entering hibernation. With the
rename, _dwc2_hcd_suspend() was changed to use params.power_down
to decide whether or not to enter partial power down.
Since params.power_down is non-zero by default, it needs to be set
to 0 for rockchip devices to restore functionality.
This bug was reported in the linux-usb thread:
REGRESSION: usb: dwc2: USB device not seen after boot
The commit that caused this regression is:
6d23ee9caa
Signed-off-by: SolidHal <hal@halemmerich.com>
Acked-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit cd7cd0e6cedfda8da6668a4af6748f96bbb6fed4 ]
When using external vbus supply regulator, it should be enabled
synchronously with PWR bit in HPRT register. This also fixes
unbalanced use of this optional regulator (This can be reproduced
easily when unbinding the driver).
Fixes: 531ef5ebea ("usb: dwc2: add support for host mode external
vbus supply")
Tested-by: Artur Petrosyan <arturp@synopsys.com>
Acked-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Fabrice Gasnier <fabrice.gasnier@st.com>
Signed-off-by: Amelie Delaunay <amelie.delaunay@st.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 4fe4f9fecc36956fd53c8edf96dd0c691ef98ff9 upstream.
Disabling all EP's allow to reset EP's to initial state.
Introduced new function dwc2_hsotg_ep_disable_lock() which
before calling dwc2_hsotg_ep_disable() function acquire
hsotg->lock and release on exiting.
From dwc2_hsotg_ep_disable() function removed acquiring
hsotg->lock.
In dwc2_hsotg_core_init_disconnected() function when USB
reset interrupt asserted disabling all ep’s by
dwc2_hsotg_ep_disable() function.
This updates eliminating sparse imbalance warnings.
Reverted changes in dwc2_hostg_disconnect() function.
Introduced new function dwc2_hsotg_ep_disable_lock().
Changed dwc2_hsotg_ep_ops. Now disable point to
dwc2_hsotg_ep_disable_lock() function.
In functions dwc2_hsotg_udc_stop() and dwc2_hsotg_suspend()
dwc2_hsotg_ep_disable() function replaced by
dwc2_hsotg_ep_disable_lock() function.
In dwc2_hsotg_ep_disable() function removed acquiring
of hsotg->lock.
Fixes: dccf1bad4be7 ("usb: dwc2: Disable all EP's on disconnect")
Signed-off-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Signed-off-by: Amit Pundir <amit.pundir@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit dccf1bad4be7eaa096c1f3697bd37883f9a08ecb upstream.
Disabling all EP's allow to reset EP's to initial state.
On disconnect disable all EP's instead of just killing
all requests. Because of some platform didn't catch
disconnect event, same stuff added to
dwc2_hsotg_core_init_disconnected() function when USB
reset detected on the bus.
Changed from version 1:
Changed lock acquire flow in dwc2_hsotg_ep_disable()
function.
Signed-off-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Signed-off-by: Amit Pundir <amit.pundir@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit dfc4fdebc5d62ac4e2fe5428e59b273675515fb2 upstream.
Use a 10000us AHB idle timeout in dwc2_core_reset() and make it
consistent with the other "wait for AHB master IDLE state" ocurrences.
This fixes a problem for me where dwc2 would not want to initialize when
updating to 4.19 on a MIPS Lantiq VRX200 SoC. dwc2 worked fine with
4.14.
Testing on my board shows that it takes 180us until AHB master IDLE
state is signalled. The very old vendor driver for this SoC (ifxhcd)
used a 1 second timeout.
Use the same timeout that is used everywhere when polling for
GRSTCTL_AHBIDLE instead of using a timeout that "works for one board"
(180us in my case) to have consistent behavior across the dwc2 driver.
Cc: linux-stable <stable@vger.kernel.org> # 4.19+
Acked-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit babd183915e91a64e976b9e8ab682bb56624df76 upstream.
In commit abb621844f ("usb: ch9: make usb_endpoint_maxp() return
only packet size") the API to usb_endpoint_maxp() changed. It used to
just return wMaxPacketSize but after that commit it returned
wMaxPacketSize with the high bits (the multiplier) masked off. If you
wanted to get the multiplier it was now up to your code to call the
new usb_endpoint_maxp_mult() which was introduced in
commit 541b6fe630 ("usb: add helper to extract bits 12:11 of
wMaxPacketSize").
Prior to the API change most host drivers were updated, but no update
was made to dwc2. Presumably it was assumed that dwc2 was too
simplistic to use the multiplier and thus just didn't support a
certain class of USB devices. However, it turns out that dwc2 did use
the multiplier and many devices using it were working quite nicely.
That means that many USB devices have been broken since the API
change. One such device is a Logitech HD Pro Webcam C920.
Specifically, though dwc2 didn't directly call usb_endpoint_maxp(), it
did call usb_maxpacket() which in turn called usb_endpoint_maxp().
Let's update dwc2 to work properly with the new API.
Fixes: abb621844f ("usb: ch9: make usb_endpoint_maxp() return only packet size")
Cc: stable@vger.kernel.org
Acked-by: Minas Harutyunyan <hminas@synopsys.com>
Reviewed-by: Matthias Kaehlcke <mka@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 4a4863bf2e7932e584a3a462d3c6daf891142ddc upstream.
Insert a padding between data and the stored_xfer_buffer pointer to
ensure they are not on the same cache line.
Otherwise, the stored_xfer_buffer gets corrupted for IN URBs on
non-cache-coherent systems. (In my case: Lantiq xRX200 MIPS)
Fixes: 3bc04e28a0 ("usb: dwc2: host: Get aligned DMA in a more supported way")
Fixes: 56406e017a ("usb: dwc2: Fix DMA alignment to start at allocated boundary")
Cc: <stable@vger.kernel.org>
Tested-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Acked-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Martin Schiller <ms@dev.tdt.de>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 54f37f56631747075f1f9a2f0edf6ba405e3e66c ]
Some function drivers queueing more than 128 ISOC requests at a time.
To avoid "descriptor chain full" cases, increasing descriptors count
from MAX_DMA_DESC_NUM_GENERIC to MAX_DMA_DESC_NUM_HS_ISOC for ISOC's
only.
Signed-off-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 35a6054132286a4ab92b536595093b82e6bdfcbc ]
Power down feature of DWC2 module integrated in Samsung SoCs doesn't work
properly or needs some additional handling in PHY or SoC glue layer, so
disable it for now. Without disabling power down, DWC2 causes random memory
trashes and fails enumeration if there is no USB link to host on driver
probe.
Fixes: 03ea6d6e9e ("usb: dwc2: Enable power down")
Acked-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit cc10ce0c51b13d1566d0ec1dcb472fb86330b391 upstream.
Disable power_down by setting the parameter to
DWC2_POWER_DOWN_PARAM_NONE. This fixes a problem on various Amlogic
Meson SoCs where USB devices are only recognized when plugged in before
booting Linux. A hot-plugged USB device was not detected even though the
device got power (my USB thumb drive for example has an LED which lit
up).
A similar fix was implemented for Rockchip SoCs in commit c216765d3a1def
("usb: dwc2: disable power_down on rockchip devices"). That commit
suggests that a change in the dwc2 driver is the cause because the
default value for the "hibernate" parameter (which then got renamed to
"power_down" to support other modes) was changed in the v4.17 merge
window with:
commit 6d23ee9caa ("Merge tag 'usb-for-v4.17' of git://git.kernel.org/pub/scm/linux/kernel/git/balbi/usb into usb-testing").
Cc: <stable@vger.kernel.org> # 4.19
Acked-by: Minas Harutyunyan <hminas@synopsys.com>
Suggested-by: Christian Hewitt <christianshewitt@gmail.com>
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 6ed30a7d8ec29d3aba46e47aa8b4a44f077dda4e upstream.
Modify the wait delay utilize the high resolution timer API to allow for
more precisely scheduled callbacks.
A previous commit added a 1ms retry delay after multiple consecutive
NAKed transactions using jiffies. On systems with a low timer interrupt
frequency, this delay may be significantly longer than specified,
resulting in misbehavior with some USB devices.
This scenario was reached on a Raspberry Pi 3B with a Macally FDD-USB
floppy drive (identified as 0424:0fdc Standard Microsystems Corp.
Floppy, based on the USB97CFDC USB FDC). With the relay delay, the drive
would be unable to mount a disk, replying with NAKs until the device was
reset.
Using ktime, the delta between starting the timer (in dwc2_hcd_qh_add)
and the callback function can be determined. With the original delay
implementation, this value was consistently approximately 12ms. (output
in us).
<idle>-0 [000] ..s. 1600.559974: dwc2_wait_timer_fn: wait_timer delta: 11976
<idle>-0 [000] ..s. 1600.571974: dwc2_wait_timer_fn: wait_timer delta: 11977
<idle>-0 [000] ..s. 1600.583974: dwc2_wait_timer_fn: wait_timer delta: 11976
<idle>-0 [000] ..s. 1600.595974: dwc2_wait_timer_fn: wait_timer delta: 11977
After converting the relay delay to using a higher resolution timer, the
delay was much closer to 1ms.
<idle>-0 [000] d.h. 1956.553017: dwc2_wait_timer_fn: wait_timer delta: 1002
<idle>-0 [000] d.h. 1956.554114: dwc2_wait_timer_fn: wait_timer delta: 1002
<idle>-0 [000] d.h. 1957.542660: dwc2_wait_timer_fn: wait_timer delta: 1004
<idle>-0 [000] d.h. 1957.543701: dwc2_wait_timer_fn: wait_timer delta: 1002
The floppy drive operates properly with delays up to approximately 5ms,
and sends NAKs for any delays that are longer.
Fixes: 38d2b5fb75 ("usb: dwc2: host: Don't retry NAKed transactions right away")
Cc: <stable@vger.kernel.org>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Acked-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Terin Stock <terin@terinstock.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 3c135e8900199e3b9375c1eff808cceba2ee37de ]
We added some error handling to this function but forgot to set the
error code on this path.
Fixes: ecd29dabb2 ("usb: dwc2: pci: Handle error cleanup in probe")
Acked-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 5aa678c7fd5371769efde30763fb43a43a118cd0 ]
dwc2_vbus_supply_exit() may call regulator_disable(). It shouldn't be
called with interrupts disabled as it might sleep.
This is seen with DEBUG_ATOMIC_SLEEP=y.
Fixes: 531ef5ebea ("usb: dwc2: add support for host mode external
vbus supply")
Tested-by: Artur Petrosyan <arturp@synopsys.com>
Acked-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Fabrice Gasnier <fabrice.gasnier@st.com>
Signed-off-by: Amelie Delaunay <amelie.delaunay@st.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Some SoC/IP as STM32F469, the snpsid can only be read after clock is
enabled, otherwise it will read as 0, and the dwc2_check_core_endianness
will assume the core and AHB have opposite endianness, leading to the
following error:
[ 1.976339] dwc2 50000000.usb: 50000000.usb supply vusb_d not found, using dummy regulator
[ 1.986124] dwc2 50000000.usb: Linked as a consumer to regulator.0
[ 1.992711] dwc2 50000000.usb: 50000000.usb supply vusb_a not found, using dummy regulator
[ 2.003672] dwc2 50000000.usb: dwc2_core_reset: HANG! AHB Idle timeout GRSTCTL GRSTCTL_AHBIDLE
[ 2.015176] dwc2: probe of 50000000.usb failed with error -16
The proposed patch changes the location where dwc2_check_core_endianness
is called, allowing the clock peripheral to be enabled first.
Acked-by: Minas Harutyunyan <hminas@synopsys.com>
Tested-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Bruno Meirelles Herrera <bmh@certi.org.br>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Not a big pull request with only 37 non-merge commits, most of which
are touching dwc2 (74% of the changes).
The most important changes are dwc2's support for uframe scheduling
and its endian-agnostic readl/writel wrappers.
From dwc3 side we have a special new glue layer for Synopsys HAPS
which will help Synopsys running FPGA validation using our upstream
driver. We also have the beginnings of dual-role support for Intel
Merrifield platform.
Apart from these, just a series of non-critical changes.
-----BEGIN PGP SIGNATURE-----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=HiNz
-----END PGP SIGNATURE-----
Merge tag 'usb-for-v4.19' of git://git.kernel.org/pub/scm/linux/kernel/git/balbi/usb into usb-next
Felipe writes:
usb: changes for v4.19
Not a big pull request with only 37 non-merge commits, most of which
are touching dwc2 (74% of the changes).
The most important changes are dwc2's support for uframe scheduling
and its endian-agnostic readl/writel wrappers.
From dwc3 side we have a special new glue layer for Synopsys HAPS
which will help Synopsys running FPGA validation using our upstream
driver. We also have the beginnings of dual-role support for Intel
Merrifield platform.
Apart from these, just a series of non-critical changes.
There's no reason to have the uframe scheduler off on dwc2. Running
with uframe_sched = False is equivalent to saying "I don't want to run
the correct code, I want to run the old and incorrect code".
The uframe scheduler has been off on stm32f4x9_fsotg since commit
e35b135055 ("usb: dwc2: Add support for STM32F429/439/469 USB OTG
HS/FS in FS mode (internal PHY)"). That commit is pretty recent, so
it's unclear to me why the uframe scheduler was left off. Hopefully
it's because someone copied it from other parameters and didn't think
to try it?
Presumably if everyone is good w/ the uframe_sched turned back on we
can kill all the old and crufty non-uframe sched code.
Reviewed-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
There's no reason to have the uframe scheduler off on dwc2. Running
with uframe_sched = False is equivalent to saying "I don't want to run
the correct code, I want to run the old and incorrect code".
The uframe scheduler has been off on Amlogic since commit f94310ac07
("usb: dwc2: add support for Meson8b and GXBB SoCs"). While this was
after most of the recent improvements, notably the commit 9f9f09b048
("usb: dwc2: host: Totally redo the microframe scheduler"), presumably
the parameters were copied from another platform and the uframe
scheduler wasn't tried.
Presumably if everyone is good w/ the uframe_sched turned back on we
can kill all the old and crufty non-uframe sched code.
Reviewed-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
There's no reason to have the uframe scheduler off on dwc2. Running
with uframe_sched = False is equivalent to saying "I don't want to run
the correct code, I want to run the old and incorrect code".
The uframe scheduler has been off on HiSilicon since commit
37dd9d65cc ("usb: dwc2: add support of hi6220"). Since then there
have been many many improvements, notably the commit 9f9f09b048
("usb: dwc2: host: Totally redo the microframe scheduler")
Presumably if everyone is good w/ the uframe_sched turned back on we
can kill all the old and crufty non-uframe sched code.
Acked-by: Minas Harutyunyan <hminas@synopsys.com>
Reviewed-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Tested-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
There's no reason to have the uframe scheduler off on dwc2. Running
with uframe_sched = False is equivalent to saying "I don't want to run
the correct code, I want to run the old and incorrect code".
The uframe scheduler has been off on Broadcom since commit
58b179dcf2 ("staging: dwc2: disable uframe_sched on the bcm2835").
Since then there have been many many improvements, notably the commit
9f9f09b048 ("usb: dwc2: host: Totally redo the microframe
scheduler")
Presumably if everyone is good w/ the uframe_sched turned back on we
can kill all the old and crufty non-uframe sched code.
Reviewed-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Tested-by: Stefan Wahren <stefan.wahren@i2se.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
To start ISOC transfers in handlers dwc2_gadget_handle_nak() and
dwc2_gadget_handle_out_token_ep_disabled() driver reads current frame
number, based on which, set target frame number to start first ISOC
transfer.
In case if system's high IRQ latency and multiple EP's asserted
interrupt in same frame, there are high probability that when reading
current frame number in EP's handlers, actual frame number can be
increased. As result for bInterval > 1, starting target frame
will be set wrongly and all ISOC packets will be dropped.
In patch "usb: dwc2: Change reading of current frame number flow"
reading of current frame number done ASAP in common interrupt handler.
This frame number stored in frame_number variable which used as
starting frame number for ISOC EP's in above mentioned handlers.
Signed-off-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Declared dwc2_check_core_endianness() function for dynamicly check
core endianness.
Added needs_byte_swap flag to hsotg structure, and depending on
flag swap value inside dwc2_readl/writel functions.
Signed-off-by: Gevorg Sahakyan <sahakyan@synopsys.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
dwc2_readl_rep/dwc2_writel_rep functions using readl/writel in a
loop.
Signed-off-by: Gevorg Sahakyan <sahakyan@synopsys.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Added hsotg argument to dwc2_readl/writel function prototype,
and also instead of address pass offset of register.
hsotg will contain flag field for endianness.
Also customized dwc2_set_bit and dwc2_clear_bit function for
dwc2_readl/writel functions.
Signed-off-by: Gevorg Sahakyan <sahakyan@synopsys.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Moved dwc2_readl/writel functions after hsotg declaration for
adding hsotg structure to dwc2_readl/writel function prototypes.
Acked-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Gevorg Sahakyan <sahakyan@synopsys.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Make sure only to copy any actual data rather than the whole buffer,
when releasing the temporary buffer used for unaligned non-isochronous
transfers.
Taken directly from commit 0efd937e27 ("USB: ehci-tegra: fix inefficient
copy of unaligned buffers")
Tested with Lantiq xRX200 (MIPS) and RPi Model B Rev 2 (ARM)
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Antti Seppälä <a.seppala@gmail.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
The commit 3bc04e28a0 ("usb: dwc2: host: Get aligned DMA in a more
supported way") introduced a common way to align DMA allocations.
The code in the commit aligns the struct dma_aligned_buffer but the
actual DMA address pointed by data[0] gets aligned to an offset from
the allocated boundary by the kmalloc_ptr and the old_xfer_buffer
pointers.
This is against the recommendation in Documentation/DMA-API.txt which
states:
Therefore, it is recommended that driver writers who don't take
special care to determine the cache line size at run time only map
virtual regions that begin and end on page boundaries (which are
guaranteed also to be cache line boundaries).
The effect of this is that architectures with non-coherent DMA caches
may run into memory corruption or kernel crashes with Unhandled
kernel unaligned accesses exceptions.
Fix the alignment by positioning the DMA area in front of the allocation
and use memory at the end of the area for storing the orginal
transfer_buffer pointer. This may have the added benefit of increased
performance as the DMA area is now fully aligned on all architectures.
Tested with Lantiq xRX200 (MIPS) and RPi Model B Rev 2 (ARM).
Fixes: 3bc04e28a0 ("usb: dwc2: host: Get aligned DMA in a more supported way")
Cc: <stable@vger.kernel.org>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Antti Seppälä <a.seppala@gmail.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
For unidirectional endpoints, the endpoint pointer will be NULL for the
unused direction. Check that the endpoint is active before
dereferencing this pointer.
Fixes: 1b4977c793 ("usb: dwc2: Update dwc2_handle_incomplete_isoc_in() function")
Fixes: 689efb2619 ("usb: dwc2: Update dwc2_handle_incomplete_isoc_out() function")
Fixes: d84845522d ("usb: dwc2: Update GINTSTS_GOUTNAKEFF interrupt handling")
Signed-off-by: John Keeping <john@metanate.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
When handling split transactions we will try to delay retry after
getting a NAK from the device. This works well for BULK transfers that
can be polled for essentially forever. Unfortunately, on slower systems
at boot time, when the kernel is busy enumerating all the devices (USB
or not), we issue a bunch of control requests (reading device
descriptors, etc). If we get a NAK for the IN part of the control
request and delay retry for too long (because the system is busy), we
may confuse the device when we finally get to reissue SSPLIT/CSPLIT IN
and the device will respond with STALL. As a result we end up with
failure to get device descriptor and will fail to enumerate the device:
[ 3.428801] usb 2-1.2.1: new full-speed USB device number 9 using dwc2
[ 3.508576] usb 2-1.2.1: device descriptor read/8, error -32
[ 3.699150] usb 2-1.2.1: device descriptor read/8, error -32
[ 3.891653] usb 2-1.2.1: new full-speed USB device number 10 using dwc2
[ 3.968859] usb 2-1.2.1: device descriptor read/8, error -32
...
Let's not delay retries of split CONTROL IN transfers, as this allows us
to reliably enumerate devices at boot time.
Fixes: 38d2b5fb75 ("usb: dwc2: host: Don't retry NAKed transactions right away")
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
PID bitfield in descriptor should be set based on particular request
length, not based on EP's mc value. PID value can't be set to 0 even
request length is 0.
Signed-off-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Freed allocated request for ep0 to prevent memory leak in case when
dwc2_driver_probe() failed.
Cc: Stefan Wahren <stefan.wahren@i2se.com>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Tested-by: Stefan Wahren <stefan.wahren@i2se.com>
Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>
Acked-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Grigor Tovmasyan <tovmasya@synopsys.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
If isoc split in transfer with no data (the length of DATA0
packet is zero), we can't simply return immediately. Because
the DATA0 can be the first transaction or the second transaction
for the isoc split in transaction. If the DATA0 packet with no
data is in the first transaction, we can return immediately.
But if the DATA0 packet with no data is in the second transaction
of isoc split in transaction sequence, we need to increase the
qtd->isoc_frame_index and giveback urb to device driver if needed,
otherwise, the MDATA packet will be lost.
A typical test case is that connect the dwc2 controller with an
usb hs Hub (GL852G-12), and plug an usb fs audio device (Plantronics
headset) into the downstream port of Hub. Then use the usb mic
to record, we can find noise when playback.
In the case, the isoc split in transaction sequence like this:
- SSPLIT IN transaction
- CSPLIT IN transaction
- MDATA packet (176 bytes)
- CSPLIT IN transaction
- DATA0 packet (0 byte)
This patch use both the length of DATA0 and qtd->isoc_split_offset
to check if the DATA0 is in the second transaction.
Tested-by: Gevorg Sahakyan <sahakyan@synopsys.com>
Tested-by: Heiko Stuebner <heiko@sntech.de>
Acked-by: Minas Harutyunyan hminas@synopsys.com>
Signed-off-by: William Wu <william.wu@rock-chips.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
The commit 3bc04e28a0 ("usb: dwc2: host: Get aligned DMA in
a more supported way") rips out a lot of code to simply the
allocation of aligned DMA. However, it also introduces a new
issue when use isoc split in transfer.
In my test case, I connect the dwc2 controller with an usb hs
Hub (GL852G-12), and plug an usb fs audio device (Plantronics
headset) into the downstream port of Hub. Then use the usb mic
to record, we can find noise when playback.
It's because that the usb Hub uses an MDATA for the first
transaction and a DATA0 for the second transaction for the isoc
split in transaction. An typical isoc split in transaction sequence
like this:
- SSPLIT IN transaction
- CSPLIT IN transaction
- MDATA packet
- CSPLIT IN transaction
- DATA0 packet
The DMA address of MDATA (urb->dma) is always DWORD-aligned, but
the DMA address of DATA0 (urb->dma + qtd->isoc_split_offset) may
not be DWORD-aligned, it depends on the qtd->isoc_split_offset (the
length of MDATA). In my test case, the length of MDATA is usually
unaligned, this cause DATA0 packet transmission error.
This patch use kmem_cache to allocate aligned DMA buf for isoc
split in transaction. Note that according to usb 2.0 spec, the
maximum data payload size is 1023 bytes for each fs isoc ep,
and the maximum allowable interrupt data payload size is 64 bytes
or less for fs interrupt ep. So we set the size of object to be
1024 bytes in the kmem cache.
Tested-by: Gevorg Sahakyan <sahakyan@synopsys.com>
Tested-by: Heiko Stuebner <heiko@sntech.de>
Acked-by: Minas Harutyunyan hminas@synopsys.com>
Signed-off-by: William Wu <william.wu@rock-chips.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
The dwc2_get_ls_map() use ttport to reference into the
bitmap if we're on a multi_tt hub. But the bitmaps index
from 0 to (hub->maxchild - 1), while the ttport index from
1 to hub->maxchild. This will cause invalid memory access
when the number of ttport is hub->maxchild.
Without this patch, I can easily meet a Kernel panic issue
if connect a low-speed USB mouse with the max port of FE2.1
multi-tt hub (1a40:0201) on rk3288 platform.
Fixes: 9f9f09b048 ("usb: dwc2: host: Totally redo the microframe scheduler")
Cc: <stable@vger.kernel.org>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Acked-by: Minas Harutyunyan hminas@synopsys.com>
Signed-off-by: William Wu <william.wu@rock-chips.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
In case when a hub is connected to DWC2 host
auto suspend occurs and host goes to
hibernation. When any device connected to hub
host hibernation exiting incorrectly.
- Added dwc2_hcd_rem_wakeup() function call to
exit from suspend state by remote wakeup.
- Increase timeout value for port suspend bit to be set.
Acked-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Artur Petrosyan <arturp@synopsys.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
In ISOC OUT transfer, when the OUT token received while EP disabled,
we shouldn't complete a usb request. The current flow completed one
usb request, this will lead to a packet drop to function driver.
Signed-off-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
In ISOC transfer, when the NAK interrupt happens, we shouldn't complete
a usb request, the current flow will complete one usb request with no
hardware transfer, this will lead to a packet drop on the usb bus.
Acked-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Zeng Tao <prime.zeng@hisilicon.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
In case of requests queue is empty reset EP target_frame to
initial value.
This allow restarting ISOC traffic in case when function
driver queued requests with interruptions.
Tested-by: Zeng Tao <prime.zeng@hisilicon.com>
Signed-off-by: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
When calling debugfs functions, there is no need to ever check the
return value. The function can work or not, but the code logic should
never do something different based on this.
Cc: Minas Harutyunyan <hminas@synopsys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When _param is unsigned and the minimum value of range is 0, it gives
the following warning:
COVERITY NO_EFFECT: This less-than-zero comparison of an unsigned value
is never true.
Converting ._param to int to avoid this warning.
Signed-off-by: Grigor Tovmasyan <tovmasya@synopsys.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
