The raw message frame length is unaligned and explicitly marked as
little endian. It should not be accessed without the appropriate
accessor functions. Fix this.
Note that payload.len already contains the correct length after parsing
via sshp_parse_frame(), so we can simply use that instead.
Reported-by: kernel-test-robot <lkp@intel.com>
Fixes: c167b9c7e3 ("platform/surface: Add Surface Aggregator subsystem")
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Acked-by: Mark Gross <mgross@linux.intel.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20210211124149.2439007-1-luzmaximilian@gmail.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Some Surface Book 2 and 3 models have a discrete GPU (dGPU) that is
hot-pluggable. On those devices, the dGPU is contained in the base,
which can be separated from the tablet part (containing CPU and
touchscreen) while the device is running.
It (in general) is presented as/behaves like a standard PCIe hot-plug
capable device, however, this device can also be put into D3cold. In
D3cold, the device itself is turned off and can thus not submit any
standard PCIe hot-plug events. To properly detect hot-(un)plugging while
the dGPU is in D3cold, out-of-band signaling is required. Without this,
the device state will only get updated during the next bus-check, eg.
via a manually issued lspci call.
This commit adds a driver to handle out-of-band PCIe hot-(un)plug events
on Microsoft Surface devices. On those devices, said events can be
detected via GPIO interrupts, which are then forwarded to the
corresponding ACPI DSM calls by this driver. The DSM then takes care of
issuing the appropriate bus-/device-check, causing the PCI core to
properly pick up the device change.
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Link: https://lore.kernel.org/r/20210205012657.1951753-1-luzmaximilian@gmail.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Explicitly check the status rather then relying on output.pointer staying
NULL on an error. This silences the following compiler warning:
drivers/platform/surface/surface3-wmi.c:60:14: warning: variable 'status' set but not used [-Wunused-but-set-variable]
Reported-by: kernel test robot <lkp@intel.com>
Reviewed-by: Maximilian Luz <luzmaximilian@gmail.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20210204113848.105994-1-hdegoede@redhat.com
The braces of the unlikely() macro inside the if condition only cover
the subtraction part, not the whole statement. This causes the result of
the subtraction to be converted to zero or one. While that still works
in this context, it causes static analysis tools to complain (and is
just plain wrong).
Fix the bracket placement and, while at it, simplify the if-condition.
Also add a comment to the if-condition explaining what we expect the
result to be and what happens on the failure path, as it seems to have
caused a bit of confusion.
This commit should not cause any difference in behavior or generated
code.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Fixes: c167b9c7e3 ("platform/surface: Add Surface Aggregator subsystem")
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Link: https://lore.kernel.org/r/20210126172202.1428367-1-luzmaximilian@gmail.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Both, ssh_rtl_rx_start() and ssh_rtl_tx_start() functions, do not exist
and have been consolidated into ssh_rtl_start(). Nevertheless,
kernel-doc references the former functions. Replace those references
with references to ssh_rtl_start().
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Link: https://lore.kernel.org/r/20210114150826.19109-1-luzmaximilian@gmail.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
A function has a different name between their prototype
and its kernel-doc markup:
../drivers/platform/surface/aggregator/ssh_request_layer.c:1065: warning: expecting prototype for ssh_rtl_tx_start(). Prototype was for ssh_rtl_start() instead
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Reviewed-by: Maximilian Luz <luzmaximilian@gmail.com>
Link: https://lore.kernel.org/r/4a6bf33cfbd06654d78294127f2b6d354d073089.1610610937.git.mchehab+huawei@kernel.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
CI static analysis complains about the allocation size in payload and
response buffers being unchecked. In general, these allocations should
be safe as the user-input is u16 and thus limited to U16_MAX, which is
only slightly larger than the theoretical maximum imposed by the
underlying SSH protocol.
All bounds on these values required by the underlying protocol are
enforced in ssam_request_sync() (or rather the functions called by it),
thus bounds here are only relevant for allocation.
Add comments explaining that this should be safe.
Reported-by: Colin Ian King <colin.king@canonical.com>
Fixes: 178f6ab77e ("platform/surface: Add Surface Aggregator user-space interface")
Addresses-Coverity: ("Untrusted allocation size")
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Link: https://lore.kernel.org/r/20210111154851.325404-3-luzmaximilian@gmail.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
When copy_struct_from_user() in ssam_cdev_request() fails, we directly
jump to the 'out' label. In this case, however 'spec' and 'rsp' are not
initialized, but we still access fields of those variables. Fix this by
initializing them at the time of their declaration.
Reported-by: Colin Ian King <colin.king@canonical.com>
Fixes: 178f6ab77e ("platform/surface: Add Surface Aggregator user-space interface")
Addresses-Coverity: ("Uninitialized pointer read")
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Link: https://lore.kernel.org/r/20210111154851.325404-2-luzmaximilian@gmail.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
The left shift of int 32 bit integer constant 1 is evaluated using 32 bit
arithmetic and then passed as a 64 bit function argument. In the case where
func is 32 or more this can lead to an oveflow. Avoid this by shifting
using the BIT_ULL macro instead.
Addresses-Coverity: ("Unintentional integer overflow")
Fixes: fc00bc8ac1 ("platform/surface: Add Surface ACPI Notify driver")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Maximilian Luz <luzmaximilian@gmail.com>
Link: https://lore.kernel.org/r/20210111144648.20498-1-colin.king@canonical.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
The Surface ACPI Notify (SAN) device provides an ACPI interface to the
Surface Aggregator EC, specifically the Surface Serial Hub interface.
This interface allows EC requests to be made from ACPI code and can
convert a subset of EC events back to ACPI notifications.
Specifically, this interface provides a GenericSerialBus operation
region ACPI code can execute a request by writing the request command
data and payload to this operation region and reading back the
corresponding response via a write-then-read operation. Furthermore,
this interface provides a _DSM method to be called when certain events
from the EC have been received, essentially turning them into ACPI
notifications.
The driver provided in this commit essentially takes care of translating
the request data written to the operation region, executing the request,
waiting for it to finish, and finally writing and translating back the
response (if the request has one). Furthermore, this driver takes care
of enabling the events handled via ACPI _DSM calls. Lastly, this driver
also exposes an interface providing discrete GPU (dGPU) power-on
notifications on the Surface Book 2, which are also received via the
operation region interface (but not handled by the SAN driver directly),
making them accessible to other drivers (such as a dGPU hot-plug driver
that may be added later on).
On 5th and 6th generation Surface devices (Surface Pro 5/2017, Pro 6,
Book 2, Laptop 1 and 2), the SAN interface provides full battery and
thermal subsystem access, as well as other EC based functionality. On
those models, battery and thermal sensor devices are implemented as
standard ACPI devices of that type, however, forward ACPI calls to the
corresponding Surface Aggregator EC request via the SAN interface and
receive corresponding notifications (e.g. battery information change)
from it. This interface is therefore required to provide said
functionality on those devices.
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20201221183959.1186143-10-luzmaximilian@gmail.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Add a misc-device providing user-space access to the Surface Aggregator
EC, mainly intended for debugging, testing, and reverse-engineering.
This interface gives user-space applications the ability to send
requests to the EC and receive the corresponding responses.
The device-file is managed by a pseudo platform-device and corresponding
driver to avoid dependence on the dedicated bus, allowing it to be
loaded in a minimal configuration.
A python library and scripts to access this device can be found at [1].
[1]: https://github.com/linux-surface/surface-aggregator-module/tree/master/scripts/ssam
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Link: https://lore.kernel.org/r/20201221183959.1186143-9-luzmaximilian@gmail.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
The Surface Aggregator EC provides varying functionality, depending on
the Surface device. To manage this functionality, we use dedicated
client devices for each subsystem or virtual device of the EC. While
some of these clients are described as standard devices in ACPI and the
corresponding client drivers can be implemented as platform drivers in
the kernel (making use of the controller API already present), many
devices, especially on newer Surface models, cannot be found there.
To simplify management of these devices, we introduce a new bus and
client device type for the Surface Aggregator subsystem. The new device
type takes care of managing the controller reference, essentially
guaranteeing its validity for as long as the client device exists, thus
alleviating the need to manually establish device links for that purpose
in the client driver (as has to be done with the platform devices).
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20201221183959.1186143-7-luzmaximilian@gmail.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
This commit adds error injection hooks to the Surface Serial Hub
communication protocol implementation, to:
- simulate simple serial transmission errors,
- drop packets, requests, and responses, simulating communication
failures and potentially trigger retransmission timeouts, as well as
- inject invalid data into submitted and received packets.
Together with the trace points introduced in the previous commit, these
facilities are intended to aid in testing, validation, and debugging of
the Surface Aggregator communication layer.
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Link: https://lore.kernel.org/r/20201221183959.1186143-6-luzmaximilian@gmail.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Add trace points to the Surface Aggregator subsystem core. These trace
points can be used to track packets, requests, and allocations. They are
further intended for debugging and testing/validation, specifically in
combination with the error injection capabilities introduced in the
subsequent commit.
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Link: https://lore.kernel.org/r/20201221183959.1186143-5-luzmaximilian@gmail.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Event items are used for completing Surface Aggregator EC events, i.e.
placing event command data and payload on a workqueue for later
processing to avoid doing said processing directly on the receiver
thread. This means that event items are allocated for each incoming
event, regardless of that event being transmitted via sequenced or
unsequenced packets.
On the Surface Book 3 and Surface Laptop 3, touchpad HID input events
(unsequenced), can constitute a larger amount of traffic, and therefore
allocation of event items. This warrants caching event items to reduce
memory fragmentation. The size of the cached objects is specifically
tuned to accommodate keyboard and touchpad input events and their
payloads on those devices. As a result, this effectively also covers
most other event types. In case of a larger event payload, event item
allocation will fall back to kzalloc().
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20201221183959.1186143-4-luzmaximilian@gmail.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Surface Serial Hub communication is, in its core, packet based. Each
sequenced packet requires to be acknowledged, via an ACK-type control
packet. In case invalid data has been received by the driver, a NAK-type
(not-acknowledge/negative acknowledge) control packet is sent,
triggering retransmission.
Control packets are therefore a core communication primitive and used
frequently enough (with every sequenced packet transmission sent by the
embedded controller, including events and request responses) that it may
warrant caching their allocations to reduce possible memory
fragmentation.
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20201221183959.1186143-3-luzmaximilian@gmail.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Add Surface System Aggregator Module core and Surface Serial Hub driver,
required for the embedded controller found on Microsoft Surface devices.
The Surface System Aggregator Module (SSAM, SAM or Surface Aggregator)
is an embedded controller (EC) found on 4th and later generation
Microsoft Surface devices, with the exception of the Surface Go series.
This EC provides various functionality, depending on the device in
question. This can include battery status and thermal reporting (5th and
later generations), but also HID keyboard (6th+) and touchpad input
(7th+) on Surface Laptop and Surface Book 3 series devices.
This patch provides the basic necessities for communication with the SAM
EC on 5th and later generation devices. On these devices, the EC
provides an interface that acts as serial device, called the Surface
Serial Hub (SSH). 4th generation devices, on which the EC interface is
provided via an HID-over-I2C device, are not supported by this patch.
Specifically, this patch adds a driver for the SSH device (device HID
MSHW0084 in ACPI), as well as a controller structure and associated API.
This represents the functional core of the Surface Aggregator kernel
subsystem, introduced with this patch, and will be expanded upon in
subsequent commits.
The SSH driver acts as the main attachment point for this subsystem and
sets-up and manages the controller structure. The controller in turn
provides a basic communication interface, allowing to send requests from
host to EC and receiving the corresponding responses, as well as
managing and receiving events, sent from EC to host. It is structured
into multiple layers, with the top layer presenting the API used by
other kernel drivers and the lower layers modeled after the serial
protocol used for communication.
Said other drivers are then responsible for providing the (Surface model
specific) functionality accessible through the EC (e.g. battery status
reporting, thermal information, ...) via said controller structure and
API, and will be added in future commits.
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Link: https://lore.kernel.org/r/20201221183959.1186143-2-luzmaximilian@gmail.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
All Microsoft Surface platform-specific device drivers depend on ACPI,
but the gatekeeper symbol SURFACE_PLATFORMS does not. Hence when the
user is configuring a kernel without ACPI support, he is still asked
about Microsoft Surface drivers, even though this question is
irrelevant.
Fix this by moving the dependency on ACPI from the individual driver
symbols to SURFACE_PLATFORMS.
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Reviewed-by: Maximilian Luz <luzmaximilian@gmail.com>
Link: https://lore.kernel.org/r/20201216133752.1321978-1-geert@linux-m68k.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Fix build warnings when CONFIG_PM_SLEEP is not enabled and these
functions are not used:
../drivers/platform/surface/surface_gpe.c:189:12: warning: ‘surface_gpe_resume’ defined but not used [-Wunused-function]
static int surface_gpe_resume(struct device *dev)
^~~~~~~~~~~~~~~~~~
../drivers/platform/surface/surface_gpe.c:184:12: warning: ‘surface_gpe_suspend’ defined but not used [-Wunused-function]
static int surface_gpe_suspend(struct device *dev)
^~~~~~~~~~~~~~~~~~~
Fixes: 274335f1c5 ("platform/surface: Add Driver to set up lid GPEs on MS Surface device")
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Cc: Maximilian Luz <luzmaximilian@gmail.com>
Cc: Hans de Goede <hdegoede@redhat.com>
Cc: platform-driver-x86@vger.kernel.org
Reviewed-by: Maximilian Luz <luzmaximilian@gmail.com>
Link: https://lore.kernel.org/r/20201214233336.19782-1-rdunlap@infradead.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
In addition to a 13" version, there is also a 15" (business) version of
the Surface Laptop 3 based on Intel CPUs. This version also handles
wakeup by lid via (unmarked) GPEs, so add support for it as well.
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Link: https://lore.kernel.org/r/20201113223935.2073847-1-luzmaximilian@gmail.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Conventionally, wake-up events for a specific device, in our case the
lid device, are managed via the ACPI _PRW field. While this does not
seem strictly necessary based on ACPI spec, the kernel disables GPE
wakeups to avoid non-wakeup interrupts preventing suspend by default and
only enables GPEs associated via the _PRW field with a wake-up capable
device. This behavior has been introduced in commit f941d3e41d ("ACPI:
EC / PM: Disable non-wakeup GPEs for suspend-to-idle") and is described
in more detail in its commit message.
Unfortunately, on MS Surface devices, there is no _PRW field present on
the lid device, thus no GPE is associated with it, and therefore the GPE
responsible for sending the status-change notification to the lid gets
disabled during suspend, making it impossible to wake the device via the
lid.
This patch introduces a pseudo-device and respective driver which, based
on some DMI matching, marks the corresponding GPE of the lid device for
wake and enables it during suspend. The behavior of this driver models
the behavior of the ACPI/PM core for normal wakeup GPEs, properly
declared via the _PRW field.
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Link: https://lore.kernel.org/r/20201028105427.1593764-1-luzmaximilian@gmail.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Move the Surface Pro 3 Button driver from platform/x86 to the newly
created platform/surface directory.
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Acked-by: Chen Yu <yu.c.chen@intel.com>
Link: https://lore.kernel.org/r/20201009141128.683254-6-luzmaximilian@gmail.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Move the Surface 3 Power operation region driver from platform/x86 to
the newly created platform/surface directory.
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20201009141128.683254-5-luzmaximilian@gmail.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Move the Surface 3 Button driver from platform/x86 to the newly created
platform/surface directory.
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20201009141128.683254-4-luzmaximilian@gmail.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Move the Surface 3 WMI driver from platform/x86 to the newly created
platform/surface directory.
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20201009141128.683254-3-luzmaximilian@gmail.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
It may make sense to split the Microsoft Surface hardware platform
drivers out to a separate subdirectory, since some of it may be shared
between ARM and x86 in the future (regarding devices like the Surface
Pro X).
Further, newer Surface devices will require additional platform drivers
for fundamental support (mostly regarding their embedded controller),
which may also warrant this split from a size perspective.
This commit introduces a new platform/surface subdirectory for the
Surface device family, with subsequent commits moving existing Surface
drivers over from platform/x86.
A new MAINTAINERS entry is added for this directory. Patches to files in
this directory will be taken up by the platform-drivers-x86 team (i.e.
Hans de Goede and Mark Gross) after they have been reviewed by
Maximilian Luz.
Signed-off-by: Maximilian Luz <luzmaximilian@gmail.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20201009141128.683254-2-luzmaximilian@gmail.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Maximilian Luz
Hans de Goede
Mauro Carvalho Chehab
Colin Ian King
Geert Uytterhoeven
Randy Dunlap