Introduce tb_switch_downstream_port() helper function that returns the
downstream port of a parent switch that is connected to the upstream
port of specified switch. From now on, we use it all across the driver
where applicable.

While there fix a whitespace in comment and rename 'downstream' to
'down' to be consistent with the rest of the driver.
Signed-off-by: Gil Fine <gil.fine@intel.com>
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

Add support for Maple Ridge discrete USB4 host controller from Intel
which has a single USB4 port (versus the already supported dual port
Maple Ridge USB4 host controller).

Cc: stable@vger.kernel.org
Signed-off-by: Gil Fine <gil.fine@intel.com>
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

Intel Meteor Lake has the same integrated Thunderbolt/USB4 controller as
Intel Alder Lake. Add the Intel Meteor Lake PCI IDs to the driver list
of supported devices.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

Intel Raptor Lake has the same integrated Thunderbolt/USB4 controller as
Intel Alder Lake. By default it is still using firmware based connection
manager so we can use most of the Alder Lake flows.
Signed-off-by: George D Sworo <george.d.sworo@intel.com>
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

kmemdup() may return NULL if there is not enough memory available. Check
this and bail out early in this case. While there move INIT_WORK() to
happen after we have allocated all the memory needed for the event
handling to avoid doing unnecessary work.
Signed-off-by: Xiaoke Wang <xkernel.wang@foxmail.com>
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

Alder Lake has the same integrated Thunderbolt/USB4 controller as
Intel Tiger Lake. By default it is still using firmware based connection
manager so we can use most of the Tiger Lake flows.

Add the Alder Lake PCI IDs to the driver list of supported devices.
Signed-off-by: Azhar Shaikh <azhar.shaikh@intel.com>
Reviewed-by: Yehezkel Bernat <YehezkelShB@gmail.com>
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

In Intel Tiger Lake and beyond it takes some time after the force power
is set until the firmware connection manager is ready. So instead of
reading it once we poll it for 10ms before giving up.
Signed-off-by: Gil Fine <gil.fine@intel.com>
Reviewed-by: Yehezkel Bernat <YehezkelShB@gmail.com>
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

Currently we have had an artificial limitation of a single DMA tunnel
per XDomain connection. However, hardware wise there is no such limit
and software based connection manager can take advantage of all the DMA
rings available on the host to establish tunnels.

For this reason make the tb_xdomain_[enable|disable]_paths() to take the
DMA ring and HopID as parameter instead of storing them in the struct
tb_xdomain. We also add API functions to allocate input and output
HopIDs of the XDomain connection that the service drivers can use
instead of hard-coding.

Also convert the two existing service drivers over to this API.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

When the firmware connection manager is not proxying between the
software and the hardware we can decrease the timeout for control
packets significantly. The USB4 spec recommends 10 ms +- 1 ms but we use
slightly larger value (100 ms) which is recommendation from Intel
Thunderbolt firmware folks. When firmware connection manager is running
then we keep using the existing 5000 ms.

To implement this we move the control channel allocation to
tb_domain_alloc(), and pass the timeout from that function to the
tb_ctl_alloc(). Then make both connection manager implementations pass
the timeout when they alloc the domain structure.

While there update kernel-doc of struct tb_ctl to match the reality.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

This allows disabling XDomain protocol completely if the user does not
plan to use the USB4/Thunderbolt peer-to-peer functionality, or for
security reasons.

XDomain protocol is enabled by default but with this commit it is
possible to disable it by passing "xdomain=0" as module parameter (or
through the kernel command line).
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Acked-by: Yehezkel Bernat <YehezkelShB@gmail.com>

Fixes the following W=1 kernel build warning(s):

 drivers/thunderbolt/icm.c:122: warning: Function parameter or member 'xdomain_connected' not described in 'icm'
 drivers/thunderbolt/icm.c:122: warning: Function parameter or member 'xdomain_disconnected' not described in 'icm'

Cc: Andreas Noever <andreas.noever@gmail.com>
Cc: Michael Jamet <michael.jamet@intel.com>
Cc: Yehezkel Bernat <YehezkelShB@gmail.com>
Cc: linux-usb@vger.kernel.org
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

The tb_dbg() call is using %#x that already adds the 0x prefix so don't
duplicate it.

Fixes: 9039387e ("thunderbolt: Add USB4 router operation proxy for firmware connection manager")
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: Yehezkel Bernat <YehezkelShB@gmail.com>

Maple Ridge is first discrete USB4 host controller from Intel. It comes
with firmware based connection manager and the flows are similar as used
in Intel Titan Ridge.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

Intel Maple Ridge and Tiger Lake connection manager firmware implements
a USB4 router operation proxy that should be used instead of direct
register access to avoid races with the firmware. This is supported in
all firmwares where the protocol version field returned in the driver
ready response is 3 (or higher).

This adds the USB4 router proxy operations support to the driver so that
we first check the protocol version and if it is 3 (or higher) the USB4
router operation is run through the firmware provided proxy. Otherwise
the native version is used.

Most USB4 router proxy operations are pretty straightforward except
NVM_AUTH where the firmware only responds once the router is restarted
but before it sends device connected notification. To support this we
split the operation so that the reply is received asynchronously and
stored to struct icm. This last reply is then returned in
icm_usb4_switch_nvm_authenticate_status() if available.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

When doing device firmware upgrade the device will disconnect for a
while and then reconnect back. Keep the parent device (and the whole
domain) powered for a while so we don't need to runtime resume
immediately when the device is connected back after the device upgrade
completes.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

This makes it easier to figure out whether the driver is using firmware
or software based connection manager implementation.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

This makes the kernel-doc to match the ordering and also this is better
place for it, not between upstream_port and vnd_cap that are used
together.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

Paulian reported a crash that happens when a dock is unplugged during
hibernation:

[78436.228217] thunderbolt 0-1: device disconnected
[78436.228365] BUG: kernel NULL pointer dereference, address: 00000000000001e0
...
[78436.228397] RIP: 0010:icm_free_unplugged_children+0x109/0x1a0
...
[78436.228432] Call Trace:
[78436.228439]  icm_rescan_work+0x24/0x30
[78436.228444]  process_one_work+0x1a3/0x3a0
[78436.228449]  worker_thread+0x30/0x370
[78436.228454]  ? process_one_work+0x3a0/0x3a0
[78436.228457]  kthread+0x13d/0x160
[78436.228461]  ? kthread_park+0x90/0x90
[78436.228465]  ret_from_fork+0x1f/0x30

This happens because remove_unplugged_switch() calls tb_switch_remove()
that releases the memory pointed by sw so the following lines reference
to a memory that might be released already.

Fix this by saving pointer to the parent device before calling
tb_switch_remove().
Reported-by: Paulian Bogdan Marinca <paulian@marinca.net>
Fixes: 4f7c2e0d ("thunderbolt: Make sure device runtime resume completes before taking domain lock")
Cc: stable@vger.kernel.org
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Intel Tiger Lake-H has the same Thunderbolt/USB4 controller as Tiger
Lake-LP. Add the Tiger Lake-H PCI IDs to the driver list of supported
devices.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

In Tiger Lake the Firmware CM is always enabled (so bit 0 is always set)
but it may be in "pass through" mode which means it requires Software CM
instead. This can be determined by checking bit 31 instead.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

Tiger Lake integrated Thunderbolt/USB4 controller is quite close to
Intel Ice Lake. By default it is still using firmware based connection
manager so we can use most of the Ice Lake flows in Tiger Lake as well.
We check if the firmware connection manager is running and in that case
use it, otherwise use the software based connection manager.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Acked-by: Yehezkel Bernat <yehezkelshb@gmail.com>

The current codebase makes use of the zero-length array language
extension to the C90 standard, but the preferred mechanism to declare
variable-length types such as these ones is a flexible array member[1][2],
introduced in C99:

struct foo {
        int stuff;
        struct boo array[];
};

By making use of the mechanism above, we will get a compiler warning
in case the flexible array does not occur last in the structure, which
will help us prevent some kind of undefined behavior bugs from being
inadvertenly introduced[3] to the codebase from now on.

This issue was found with the help of Coccinelle.

[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit 76497732 ("cxgb3/l2t: Fix undefined behaviour")
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

Since now we can do pretty much the same thing in the software
connection manager than the firmware would do, there is no point
starting it by default. Instead we can just continue using the software
connection manager.

Make it possible for user to switch between the two by adding a module
pararameter (start_icm) which is by default false. Having this ability
to enable the firmware may be useful at least when debugging possible
issues with the software connection manager implementation.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

Lane bonding allows aggregating two 10/20 Gb/s (depending on the
generation) lanes into a single 20/40 Gb/s bonded link. This allows
sharing the full bandwidth more efficiently. In order to establish lane
bonding we need to check that lane bonding is possible through link
controller and that both ends of the link actually supports 2x widths.
This also means that all the paths should be established through the
primary port so update tb_path_alloc() to handle this as well.

Lane bonding is supported starting from Falcon Ridge (2nd generation)
controllers.

We also expose the current speed and number of lanes under each device
except the host router following similar attribute naming than USB bus.
Expose speed and number of lanes for both directions to allow possibility
of asymmetric link in the future.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

Currently add_switch() takes a huge amount of parameters that makes it
hard to maintain. Instead of passing all those parameters we can split
the function into two parts (alloc and add) and fill the additional
switch fields directly in the functions calling those.

While there remove redundant error logging in case kmemdup() fails.

No functional changes.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

There are quite many places in the driver where we iterate over each
port in the switch. To make it bit more convenient, add a macro that can
be used to iterate over each port and convert existing call sites to use it.

This is based on code by Lukas Wunner.

No functional changes.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

The Thunderbolt controller is integrated into the Ice Lake CPU itself
and requires special flows to power it on and off using force power bit
in NHI VSEC registers. Runtime PM (RTD3) and Sx flows also differ from
the discrete solutions. Now the firmware notifies the driver whether
RTD3 entry or exit are possible. The driver is responsible of sending
Go2Sx command through link controller mailbox when system enters Sx
states (suspend-to-mem/disk). Rest of the ICM firwmare flows follow
Titan Ridge.
Signed-off-by: Raanan Avargil <raanan.avargil@intel.com>
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: Yehezkel Bernat <YehezkelShB@gmail.com>
Tested-by: Mario Limonciello <mario.limonciello@dell.com>

This is depends on the controller and on the platform/CPU we are
running. Move it to struct icm so we can set it per controller.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: Yehezkel Bernat <YehezkelShB@gmail.com>
Tested-by: Mario Limonciello <mario.limonciello@dell.com>

When starting ICM firmware on Apple systems we need to perform CIO reset
as part of the flow. However, it turns out that the reset register has
changed to another location in Titan Ridge.

Fix this by introducing ->cio_reset() callback with corresponding
implementations for Alpine and Titan Ridge.

Fixes: c4630d6a ("thunderbolt: Start firmware on Titan Ridge Apple systems")
Reported-by: Peter Bowen <pzb@amazon.com>
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

When a device is authorized from userspace by writing to authorized
attribute we first take the domain lock and then runtime resume the
device in question. There are two issues with this.

First is that the device connected notifications are blocked during this
time which means we get them only after the authorization operation is
complete. Because of this the authorization needed flag from the
firmware notification is not reflecting the real authorization status
anymore. So what happens is that the "authorized" keeps returning 0 even
if the device was already authorized properly.

Second issue is that each time the controller is runtime resumed the
connection_id field of device connected notification may be different
than in the previous resume. We need to use the latest connection_id
otherwise the firmware rejects the authorization command.

Fix these by moving runtime resume operations to happen before the
domain lock is taken, and waiting for the updated device connected
notification from the firmware before we allow runtime resume of a
device to complete.

While there add missing locking to tb_switch_nvm_read().

Fixes: 09f11b6c ("thunderbolt: Take domain lock in switch sysfs attribute callbacks")
Reported-by: Pengfei Xu <pengfei.xu@intel.com>
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

Titan Ridge flow to start the firmware is the same as Alpine Ridge so we
can do the same on Titan Ridge based Apple systems.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

Now that the driver can handle every possible tunnel types there is no
point to log everything as info level so turn these to happen at debug
level instead.

While at it remove duplicated tunnel activation log message
(tb_tunnel_activate() calls tb_tunnel_restart() which print the same
message) and add one missing '\n' termination.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

In order to detect possible connections to other domains we need to be
able to find out why tb_switch_alloc() fails so make it return ERR_PTR()
instead. This allows the caller to differentiate between errors such as
-ENOMEM which comes from the kernel and for instance -EIO which comes
from the hardware when trying to access the possible switch.

Convert all the current call sites to handle this properly.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

Currently the driver only assigns remote port for the primary port if in
case of dual link. This makes things such as walking from one port to
another more complex than necessary because the code needs to change
from secondary to primary port if the path that is established is
created using secondary links.

In order to always assign both remote pointers we need to prevent the
scanning code from following the secondary link. Failing to do that
might cause problems as the same switch may be enumerated twice (or
removed in case of unplug). Handle that properly by introducing a new
function tb_port_has_remote() that returns true only for the primary
port. We also update tb_is_upstream_port() to support both dual link
ports, make it take const port pointer and move it below
tb_upstream_port() to keep similar functions close.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

Maximum depth in Thunderbolt topology is 6 so make sure it is not
possible to allocate switches that exceed the depth limit.

While at it update tb_switch_alloc() to use upper/lower_32_bits()
following tb_switch_alloc_safe_mode().
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

tb_switch_find_by_route() does the same already so use it instead and
remove duplicated get_switch_at_route().
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: Lukas Wunner <lukas@wunner.de>

uuid in add_switch is allocted via kmemdup which can fail. The patch
logs the error and cleans up the allocated memory for switch.
Signed-off-by: Aditya Pakki <pakki001@umn.edu>
Reviewed-by: Mukesh Ojha <mojha@codeaurora.org>
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

This gets rid of the licence boilerplate duplicated in each file. While
there fix doubled space in domain.c author line.

No functional changes intended.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Acked-by: Yehezkel Bernat <yehezkelshb@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

If there is a long chain of devices connected when the driver is loaded
ICM sends device connected event for each and those are put to tb->wq
for later processing. Now if the driver gets unloaded in the middle, so
that the work queue is not yet empty it gets flushed by tb_domain_stop().
However, by that time the root switch is already removed so the driver
crashes when it tries to dereference it in ICM event handling callbacks.

Fix this by checking whether the root switch is already removed. If it
is we know that the domain is stopped and we should merely skip handling
the event.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

When Thunderbolt host controller is set to RTD3 mode (Runtime D3) it is
present all the time. Because of this it is important to runtime suspend
the controller whenever possible. In case of ICM we have following rules
which all needs to be true before the host controller can be put to D3:

  - The controller firmware reports to support RTD3
  - All the connected devices announce support for RTD3
  - There is no active XDomain connection

Implement this using standard Linux runtime PM APIs so that when all the
children devices are runtime suspended, the Thunderbolt host controller
PCI device is runtime suspended as well. The ICM firmware then starts
powering down power domains towards RTD3 but it can prevent this if it
detects that there is an active Display Port stream (this is not visible
to the software, though).

The Thunderbolt host controller will be runtime resumed either when
there is a remote wake event (device is connected or disconnected), or
when there is access from userspace that requires hardware access.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>