Support for multiple instances of a class implementation,
and the ability to register an instance to a configuration
at runtime, requires a mechanism to add a string descriptor
and update its index based on the total number of descriptors.
We also need to handle some special string descriptors like
Product or Serial Number provided by the application.
Marked as such by using specific macros, these descriptors
can be sorted out by the stack and the device descriptor
indexes are updated automatically.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
Add and use specific macros for manufacturer, product, and
serial number string descriptors.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
While this does not provide much of an advantage yet,
it will allow us to add descriptors and assign an index
more easily in the next commit.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
When we enable USBD shell support with samples like cdc_acm or mass,
there would be another USB device context besides the one provided
by the shell. This patch introduces a new command to select an
alternate context to be used by USBD shell commands.
Also fix the type where "add" command should be "remove".
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
If we try to reuse a string descriptor node, the content will
be corrupted because the device stack assumes it is still ASCII7
encoded. Add a flag to indicate that a descriptor node contains
UTF16LE encoded content.
And while we are at it, add a flag to indicate that the SN string
should not be obtained from the hwinfo API.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
There may be more than one context using the same device, for example,
if the shell module is enabled. Check if another context that uses
the same device is initialized.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
At the shutdown of USB device stack we have to cleanup and
remove all registered class instances and string descriptors
from a configuration.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
Foobaz interface will be renamed and added as a standalone part
to serve as an API example in a subsequent commit.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
Until now iterable sections APIs have been part of the toolchain
(common) headers. They are not strictly related to a toolchain, they
just rely on linker providing support for sections. Most files relied on
indirect includes to access the API, now, it is included as needed.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Universal Serial Bus Mass Storage Specification For Bootability requires
that device shall accept Bulk-Only-Transport bCBWCBLength of 12 for all
commands documented in the specification. Allow padding for commands
documented in Bootability specification, but leave others intact.
This fixes Request Sense on Windows 11 which previously failed due to
bCBWCBLength being 12 instead of 6.
Signed-off-by: Tomasz Moń <tomasz.mon@nordicsemi.no>
A request to set alternate zero for a non-existent interface
would succeed because of the inappropriate order of checks.
Move the check if the desired alternate is the same as the
current one after the check if interface exists.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
Correct the return type of the helper function to get/set
current interface alternate value.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
Check that the length of the packet is at least the size of
the descriptor head. Otherwise there may be unpleasant side effects.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
Add CDC Ethernet Control Model class implementation for the
new experimental USB device support based on the existing
implementation subsys/usb/device/class/netusb/function_ecm.c.
The implementation forms virtual ethernet connection between
USB host and USB device and requires two corresponding MAC
addresses, one for the virtual interface on the device side,
and other for the host which is described by a string descriptor.
With upcoming changes it should also possible to use a real
ethernet controller as media access on the device side.
CDC ECM implementation supports multiple instances which are
specified using DT. The number of instances is limited by the
number of endpoints on the controller, two to three should usually
be possible.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
Transmit errors must be treated differently than Protocol Errors.
This change sets a flag that informs the stack of a Message
Transmit error.
Signed-off-by: Sam Hurst <sbh1187@gmail.com>
The possible control endpoint MPS for USB 2.0 FS devices is
8, 16, 32, or 64 bytes. Typically, USB2.0 compliant devices support MPS
up to 64 bytes, and we have not had the need to support other MPS.
This patch implements a mechanism to fall back to the minimum allowed
MPS when a controller is likely a USB 1.1 compliant device and does
not support control endpoint MPS of 64 bytes.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
This reverts commit cd7461495f,
introduced to fix possible deadlock as result of changes in
commit f206170c65
("usb: device: Do not call callback when transfer is cancelled")
which is now reverted.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
This reverts commit f206170c65
introduced as workaround for nRF USBD device controller in PR
https://github.com/zephyrproject-rtos/zephyr/pull/16193.
This commit may be reverted due to changes made in
commit e326c58399
("usb: device: Do not cancel transfers on suspend").
Signed-off-by: Nickolas Lapp <nickolaslapp@gmail.com>
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
The init infrastructure, found in `init.h`, is currently used by:
- `SYS_INIT`: to call functions before `main`
- `DEVICE_*`: to initialize devices
They are all sorted according to an initialization level + a priority.
`SYS_INIT` calls are really orthogonal to devices, however, the required
function signature requires a `const struct device *dev` as a first
argument. The only reason for that is because the same init machinery is
used by devices, so we have something like:
```c
struct init_entry {
int (*init)(const struct device *dev);
/* only set by DEVICE_*, otherwise NULL */
const struct device *dev;
}
```
As a result, we end up with such weird/ugly pattern:
```c
static int my_init(const struct device *dev)
{
/* always NULL! add ARG_UNUSED to avoid compiler warning */
ARG_UNUSED(dev);
...
}
```
This is really a result of poor internals isolation. This patch proposes
a to make init entries more flexible so that they can accept sytem
initialization calls like this:
```c
static int my_init(void)
{
...
}
```
This is achieved using a union:
```c
union init_function {
/* for SYS_INIT, used when init_entry.dev == NULL */
int (*sys)(void);
/* for DEVICE*, used when init_entry.dev != NULL */
int (*dev)(const struct device *dev);
};
struct init_entry {
/* stores init function (either for SYS_INIT or DEVICE*)
union init_function init_fn;
/* stores device pointer for DEVICE*, NULL for SYS_INIT. Allows
* to know which union entry to call.
*/
const struct device *dev;
}
```
This solution **does not increase ROM usage**, and allows to offer clean
public APIs for both SYS_INIT and DEVICE*. Note that however, init
machinery keeps a coupling with devices.
**NOTE**: This is a breaking change! All `SYS_INIT` functions will need
to be converted to the new signature. See the script offered in the
following commit.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
init: convert SYS_INIT functions to the new signature
Conversion scripted using scripts/utils/migrate_sys_init.py.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
manifest: update projects for SYS_INIT changes
Update modules with updated SYS_INIT calls:
- hal_ti
- lvgl
- sof
- TraceRecorderSource
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
tests: devicetree: devices: adjust test
Adjust test according to the recently introduced SYS_INIT
infrastructure.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
tests: kernel: threads: adjust SYS_INIT call
Adjust to the new signature: int (*init_fn)(void);
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
This allows us to get the result of synchronous transfer without
any hacks, just from the net_buf structure.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
USB Mass Storage Specification for Bootability requires MODE SENSE(10)
command. MODE SENSE(6) and MODE SENSE(10) generally access the same data
but differ in the maximum allocated length and LLBAA support. However
there is no point in extracting common handling because there no mode
pages are supported now.
Fail MODE SENSE requests if asking for anything other than supported
page codes (to which the essentially hardcoded response is valid).
Signed-off-by: Tomasz Moń <tomasz.mon@nordicsemi.no>
Mass Storage Class allows device to either silently discard OUT endpoint
data after receiving invalid CBW or to STALL the OUT endpoint (and keep
it STALLed regardless of host clearing halt) until Reset Recovery.
Switch from silently discarding to STALL in order to allow host realize
that device waits for Reset Recovery without transferring complete OUT
data.
Signed-off-by: Tomasz Moń <tomasz.mon@nordicsemi.no>
The asserts in the USB-C stack functions were causing compilation
errors when asserts were enabled in Kconfig.
This was caused because the asserts checked if the functions
parameters were within limits specified by the state machine arrays.
At the position of these asserts, these arrays were only declared
without specifying the size of them, what prohibited the asserts from
knowing the size of arrays and checking if specified value is within
limit of it.
Signed-off-by: Michał Barnaś <mb@semihalf.com>
Track LBA rather than byte offset in order to simplify the code and
support devices larger than 4GiB.
Signed-off-by: JP Sugarbroad <jpsugar@amazon.com>
set_device can be called in interrupt context at least with
stm32u5 soc. Calling k_usleep from there causes the isr
to run forever.
Signed-off-by: Miika Karanki <miika.karanki@vaisala.com>
Introduce new USB Mass Storage Bulk-Only Transport implementation
written from scratch. Main goal behind new implementation was to
separate USB and SCSI parts as clearly as possible.
Limited set of SCSI disk commands is implemented in separate source code
file that is internal to USB device stack. While it should be possible
to use the SCSI implementation by other components there is currently no
other user besides USB MSC and therefore SCSI header is kept private.
Signed-off-by: Tomasz Moń <tomasz.mon@nordicsemi.no>
Use a consistent nil_desc name for the descriptor that
terminates a class (function) descriptor.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
Changed incrementing `for` loop counters to `size_t` from `int` to
eliminate warning, "warning: comparison of integer expressions of
different signedness: 'uint32_t' {aka 'unsigned int'} and 'int'
[-Wsign-compare]"
Signed-off-by: Zachary J. Fields <zachary_fields@yahoo.com>
Add USB device class API to notify class instances that an endpoint
has been halted or cleared due to a feature request.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
Use proper format for size_t type eliminating warnings of type:
...
error: format '%u' expects argument of type 'unsigned int', but
argument 2 has type 'size_t' {aka 'long unsigned int'}
[-Werror=format=]
...
Signed-off-by: Andrei Emeltchenko <andrei.emeltchenko@intel.com>
Any Power Delivery message state that expects a reply must
start a sender response timer. This change addes a super
state that implements the Sender Response Timer
functionality, from which Power Deleiver messages states
can inherit from.
Signed-off-by: Sam Hurst <sbh1187@gmail.com>
Allows receive packets that are bigger than
CONFIG_RNDIS_BULK_EP_MPS.
Considering:
* MaxPacketsPerTransfer is 1
* MaxTransferSize is 1558
* PacketAlignmentFactor is 0 (20=1)
The rndis_bulk_out shall be able to receive transfer
up to 1558 bytes (inclusive).
Signed-off-by: Georgij Cernysiov <geo.cgv@gmail.com>
Add Bluetooth HCI USB transport layer implementation for the new
experimental USB device support based on the existing implementation
subsys/usb/device/class/bluetooth.c.
This implementation, unlike existing one, contains the interface
descriptors for isochronous endpoints. Since we do not support voice
channels, these are just there to avoid issues with Linux kernel btusb
driver when this implementation is part of a composite configuration.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
USB stack did leak memory on every SET ADDRESS request. UDC control
allocator could cope with up to 13 leaked allocations. Therefore,
issuing bus reset 13 times in a row (in addition to automatic reset
after connecting the USB cable) was enough to exhaust memory and
thus drive USB stack inoperable.
Signed-off-by: Tomasz Moń <tomasz.mon@nordicsemi.no>
Cancelling transfers on suspend contradicts Universal Serial Bus
Specification Revision 2.0, 9.1.1.6 Suspended:
* When suspended, the USB device maintains any internal status,
including its address and configuration.
The internal status definitely includes any pending USB transfers.
If there is a class that wants to cancel transfer on suspend, then the
cancel should be initiated by the class, not the device stack itself.
Update hal_nordic to a version that does not abort all endpoints at
suspend. It seems that aborting endpoints on suspend in nrfx driver was
the actual reason why transfers were canceled on suspend.
Remove transfer retriggering on resume from CDC ACM and Bluetooth class
implementations because transfers are no longer cancelled on suspend.
Other classes do not have any suspend related workarounds implemented.
Signed-off-by: Tomasz Moń <tomasz.mon@nordicsemi.no>
