A number of ESP-IDF header files that are needed to compile RIOT include the header file `driver/gpio.h` only because of the definition of the type `gpio_num_t`. However, this header file contains the entire GPIO API definition of the ESP-IDF, which conflicts with that of RIOT.
The solution was to use a wrapper library that does not need to include the `driver/gpio.h` file of the ESP-IDF during compilation of RIOT code.
This commit provides another approach which does not require such a wrapper library. It just provides its own `driver/gpio.h` in RIOT that is included by ESP-IDF header files instead of the original `driver/gpio.h` in ESP-IDF. It defines only the required `gpio_num_t` if RIOT code is compiled but includes the original `driver/gpio.h` of ESP-IDF if ESP-IDF code is compiled. This avoids to create a wrapper library for each module.
19420: cpu/esp32: use ets_printf instead of puts in startup r=maribu a=gschorcht
### Contribution description
This PR provides a workaround that fixes the problem that restarting an application automatically after flashing it in download mode via USB Serial/JTAG doesn't work and requires a hard reset by pressing the RESET button before it starts.
The reason that the application doesn't restart automatically after flashing it is that an exception occurs if `puts` or `printf` is called during startup before the first interrupt driven context switch in `thread_yield_higher`. The console seems to hange after bootloader:
```
EESP-ROM:esp32c3-api1-20210207
Build:Feb 7 2021
rst:0x15 (USB_UART_CHIP_RESET),boot:0xd (SPI_FAST_FLASH_BOOT)
Saved PC:0x40380786
SPIWP:0xee
mode:DIO, clock div:1
load:0x3fcd6100,len:0x6c
load:0x403ce000,len:0x7ec
load:0x403d0000,len:0x2170
entry 0x403ce000
Pro cpu up.
```
However, the system stucks in a exception/printf loop. ESP32-C3 and ESP32-S3 are affected.
### Testing procedure
Flash a ESP32-C3 or ESP32-S3 board that don't have a USB-to-UART chip with reset logic on board, for example
```
BOARD=hip-badge make -j8 -C tests/shell flash
```
or
```
BOARD=esp32s3-pros3 make -j8 -C tests/shell flash
```
Connect a terminal to the the board. Without the PR, the console doesn't seem to work and the RESET button has to pressed explicitly to get it working. With the PR, the console should work.
The problem can also be caused when using
```
dist/tools/esptools/espreset.py -p /dev/ttyACM0
```
while connected with a terminal to the board. Without the PR, the console output stops after
```
ESP-ROM:esp32c3-api1-20210207
Build:Feb 7 2021
rst:0x15 (USB_UART_CHIP_RESET),boot:0xd (SPI_FAST_FLASH_BOOT)
Saved PC:0x40380786
SPIWP:0xee
mode:DIO, clock div:1
load:0x3fcd6100,len:0x6c
load:0x403ce000,len:0x7ec
load:0x403d0000,len:0x2170
entry 0x403ce000
Pro cpu up.
```
while it continues with the PR as following:
```
main(): This is RIOT! (Version: 2023.04-devel-713-gcb721-boards/
test_shell.
>
```
### Issues/PRs references
Co-authored-by: Gunar Schorcht <gunar@schorcht.net>
19397: drivers/usbdev_synopsys_dwc2: fix and reenable DMA mode r=benpicco a=gschorcht
### Contribution description
This PR fixes the DMA mode for all STM32 USB OTG HS cores (including that for STM32F4xx CID 1.xxx) and reenables it. It fixes remaining problems in issue #19359.
This PR includes also includes some changes that are needed to use the DMA mode:
- EP number is used as defined in CMSIS (if defined) for STM32
- `periph_usbdev_hs` feature is added in Kconfig
- `periph_usbdev_hs` feature is added in board definition of `stm32f429i-disc1`
- largest number of available EPs is used for STM32 instead of the smallest number (to be able to use all EPs of HS peripheral)
- `stm32f429i-disco` is removed from blacklist in `tests/usbus_cdc_ecm` since it uses the HS peripheral
### Testing procedure
The following tests should work
```python
USEMODULE=stdio_cdc_acm BOARD=stm32f429i-disc1 make -j8 -C tests/usbus_cdc_ecm flash
```
<details>
<summary>Test results</summary>
```python
[526755.875691] usb 1-2.2: new full-speed USB device number 106 using xhci_hcd
[526755.977853] usb 1-2.2: config 1 interface 3 altsetting 1 endpoint 0x84 has invalid maxpacket 512, setting to 64
[526755.977856] usb 1-2.2: config 1 interface 3 altsetting 1 endpoint 0x2 has invalid maxpacket 512, setting to 64
[526755.978762] usb 1-2.2: New USB device found, idVendor=1209, idProduct=7d01, bcdDevice= 1.00
[526755.978764] usb 1-2.2: New USB device strings: Mfr=3, Product=2, SerialNumber=4
[526755.978766] usb 1-2.2: Product: stm32f429i-disc1
[526755.978768] usb 1-2.2: Manufacturer: RIOT-os.org
[526755.978769] usb 1-2.2: SerialNumber: 7C156425A950A8EB
[526755.991190] cdc_acm 1-2.2:1.0: ttyACM1: USB ACM device
[526755.998131] cdc_ether 1-2.2:1.2 usb0: register 'cdc_ether' at usb-0000:00:14.0-2.2, CDC Ethernet Device, a6:f6:4a:85:1d:c9
[526756.044150] cdc_ether 1-2.2:1.2 enp0s20f0u2u2i2: renamed from usb0
```
</details>
```python
USEMODULE='stdio_cdc_acm periph_usbdev_hs_utmi' BOARD=stm32f723e-disco make -j8 -C tests/usbus_cdc_ecm flash
```
<details>
<summary>Test results</summary>
```python
[528733.480207] usb 1-4.3.4: reset high-speed USB device number 32 using xhci_hcd
[528733.707800] usb 1-4.4: new high-speed USB device number 111 using xhci_hcd
[528733.808257] usb 1-4.4: config 1 interface 0 altsetting 0 endpoint 0x81 has an invalid bInterval 255, changing to 11
[528733.808260] usb 1-4.4: config 1 interface 1 altsetting 0 bulk endpoint 0x1 has invalid maxpacket 64
[528733.808263] usb 1-4.4: config 1 interface 1 altsetting 0 bulk endpoint 0x82 has invalid maxpacket 64
[528733.808642] usb 1-4.4: New USB device found, idVendor=1209, idProduct=7d01, bcdDevice= 1.00
[528733.808645] usb 1-4.4: New USB device strings: Mfr=3, Product=2, SerialNumber=4
[528733.808647] usb 1-4.4: Product: stm32f723e-disco
[528733.808649] usb 1-4.4: Manufacturer: RIOT-os.org
[528733.808651] usb 1-4.4: SerialNumber: A6BAC4E1B1E0806B
[528733.811988] cdc_acm 1-4.4:1.0: ttyACM1: USB ACM device
[528733.814456] cdc_ether 1-4.4:1.2 usb0: register 'cdc_ether' at usb-0000:00:14.0-4.4, CDC Ethernet Device, e6:75:97:3a:74:ba
[528733.854371] cdc_ether 1-4.4:1.2 enp0s20f0u4u4i2: renamed from usb0
```
</details>
```python
USEMODULE='stdio_cdc_acm periph_usbdev_hs_ulpi' BOARD=stm32f746g-disco make -j8 -C tests/usbus_cdc_ecm flash
```
<details>
<summary>Test results</summary>
```python
[529000.944482] usb 1-4.3.4: reset high-speed USB device number 32 using xhci_hcd
[529003.728260] usb 1-4.4: new high-speed USB device number 114 using xhci_hcd
[529003.833107] usb 1-4.4: config 1 interface 0 altsetting 0 endpoint 0x81 has an invalid bInterval 255, changing to 11
[529003.833111] usb 1-4.4: config 1 interface 1 altsetting 0 bulk endpoint 0x1 has invalid maxpacket 64
[529003.833113] usb 1-4.4: config 1 interface 1 altsetting 0 bulk endpoint 0x82 has invalid maxpacket 64
[529003.833743] usb 1-4.4: New USB device found, idVendor=1209, idProduct=7d00, bcdDevice= 1.00
[529003.833747] usb 1-4.4: New USB device strings: Mfr=3, Product=2, SerialNumber=4
[529003.833749] usb 1-4.4: Product: stm32f746g-disco
[529003.833751] usb 1-4.4: Manufacturer: RIOT-os.org
[529003.833753] usb 1-4.4: SerialNumber: 66FE8934D1A363E0
[529003.837143] cdc_acm 1-4.4:1.0: ttyACM1: USB ACM device
[529003.839755] cdc_ether 1-4.4:1.2 usb0: register 'cdc_ether' at usb-0000:00:14.0-4.4, CDC Ethernet Device, 6a:88:1f:1f:b1:f0
[529003.879025] cdc_ether 1-4.4:1.2 enp0s20f0u4u4i2: renamed from usb0```
```
</details>
### Issues/PRs references
Fixes#19359
19416: cpu/rpx0xx/cmsis: Update vendor header files r=benpicco a=maribu
### Contribution description
Generated new vendor header files from upstream SVD files using:
./SVDConv "$PICO_SDK_DIR"/src/rp2040/hardware_regs/rp2040.svd \
--generate=header --fields=macro --fields=enum
Note: The missing `--fields=struct` flag resulted in the header no longer containing bit-fields to represent different fields within registers. While this would generally ease writing code, the RP2040 has the unpleasant feature of corrupting the remaining bits of the register when a write access that is not word-sized occurs in the memory mapped I/O area. This could happen e.g. when a bit field is byte-sized and byte-aligned.
### Testing procedure
No binary changes (hopefully).
### Issues/PRs references
This adds a few additional vendor defines, notably for USB. If anyone were to implement USB, this would be a requirement.
19418: cpu/gd32v: fix gpio_read in periph_gpio r=benpicco a=gschorcht
### Contribution description
This PR fixes a bug in `gpio_read` which made `gpio_read` completely unusable!
A small bug with big consequences. In `gpio_read` the combined port | pin_num parameter `pin` was used instead of the pin number `pin_num` for the call of `_pin_is_input`. This caused the problem that for example instead of accessing GPIOA->CTL0 with address 0x40010800, address 0x60018c00 was accessed. As a result, a pin was randomly detected as input or output and thus a result was arbitrarily returned. Approx. 50% of all inputs always returned LOW.
I found this error by coincidence when I tried to find out why the BOOT0 button on a Sipeed Longan Nano is not usable as a button in RIOT.
### Testing procedure
Flash `tests/periph_gpio`
```
BOARD=sipeed-longan-nano make -j8 -C tests/periph_gpio flash
```
and use commands
```
init_in 0 8
read 0 8
```
Without this PR, the pin is always LOW. With the PR, the pin should be HIGH when the BOOT button is pressed.
### Issues/PRs references
19419: boards/sipeed-longan-nano: add BOOT as user button r=benpicco a=gschorcht
### Contribution description
This PR makes the BOOT button usable as a user button.
### Testing procedure
The test requires PR #19418 to work.
Flash and test:
```
BOARD=sipeed-longan-nano make -j8 -C tests/saul flash term
```
The output
```
Dev: BOOT Type: SENSE_BTN
Data: 0
```
should change to
```
Dev: BOOT Type: SENSE_BTN
Data: 1
```
when the BOOT button is pressed.
### Issues/PRs references
Depends on PR #19418
Co-authored-by: Gunar Schorcht <gunar@schorcht.net>
Co-authored-by: Marian Buschsieweke <marian.buschsieweke@ovgu.de>
Generated new vendor header files from upstream SVD files using:
./SVDConv "$PICO_SDK_DIR"/src/rp2040/hardware_regs/rp2040.svd \
--generate=header --fields=macro --fields=enum
Note: The missing `--fields=struct` flag resulted in the header no
longer containing bit-fields to represent different fields
within registers. While this would generally ease writing code,
the RP2040 has the unpleasant feature of corrupting the
remaining bits of the register when a write access that is not
word-sized occurs in the memory mapped I/O area. This could
happen e.g. when a bit field is byte-sized and byte-aligned.
19407: cpu/stm32/periph: Implement GPIO LL for STM32F1 without IRQ support (yet) r=gschorcht a=maribu
### Contribution description
This implements GPIO LL support for the STM32F1 in the first commit. IRQ support is added with https://github.com/RIOT-OS/RIOT/pull/19412.
This sneaks in a second commit replacing the `expect()` calls in `tests/periph_gpio_ll` with a trivial five-liner that doesn't `panic()`, so that stdio output will still be delivered on high level stdio implementations. The tests provides a lot of useful output to aid debugging, so its a great usability improvement if the test makes sure to actually deliver that output.
### Testing procedure
<details><summary><code>make -C tests/periph_gpio_ll BOARD=nucleo-f103rb flash term</code></summary>
```
2023-03-17 18:55:09,188 # Help: Press s to start test, r to print it is ready
s
2023-03-17 18:55:10,299 # START
2023-03-17 18:55:10,307 # main(): This is RIOT! (Version: 2023.04-devel-683-g9c3812-cpu/stm32/periph/gpio_ll)
2023-03-17 18:55:10,309 # Test / Hardware Details:
2023-03-17 18:55:10,310 # ========================
2023-03-17 18:55:10,311 # Cabling:
2023-03-17 18:55:10,313 # (INPUT -- OUTPUT)
2023-03-17 18:55:10,315 # P2.10 (PC10) -- P2.2 (PC2)
2023-03-17 18:55:10,318 # P2.12 (PC12) -- P2.3 (PC3)
2023-03-17 18:55:10,322 # Number of pull resistor values supported: 1
2023-03-17 18:55:10,325 # Number of drive strengths supported: 1
2023-03-17 18:55:10,328 # Number of slew rates supported: 3
2023-03-17 18:55:10,330 # Valid GPIO ports:
2023-03-17 18:55:10,332 # - PORT 0 (PORT A)
2023-03-17 18:55:10,333 # - PORT 1 (PORT B)
2023-03-17 18:55:10,335 # - PORT 2 (PORT C)
2023-03-17 18:55:10,336 # - PORT 3 (PORT D)
2023-03-17 18:55:10,338 # - PORT 4 (PORT E)
2023-03-17 18:55:10,338 #
2023-03-17 18:55:10,341 # Testing gpio_port_pack_addr()
2023-03-17 18:55:10,343 # =============================
2023-03-17 18:55:10,343 #
2023-03-17 18:55:10,344 # All OK
2023-03-17 18:55:10,344 #
2023-03-17 18:55:10,346 # Testing gpip_ng_init()
2023-03-17 18:55:10,348 # ======================
2023-03-17 18:55:10,348 #
2023-03-17 18:55:10,354 # Testing is_gpio_port_num_valid() is true for PORT_OUT and PORT_IN:
2023-03-17 18:55:10,354 #
2023-03-17 18:55:10,358 # Testing input configurations for PIN_IN_0:
2023-03-17 18:55:10,361 # Support for input with pull up: yes
2023-03-17 18:55:10,366 # state: in, pull: up, schmitt trigger: off, value: on
2023-03-17 18:55:10,369 # Support for input with pull down: yes
2023-03-17 18:55:10,374 # state: in, pull: down, schmitt trigger: off, value: off
2023-03-17 18:55:10,378 # Support for input with pull to bus level: no
2023-03-17 18:55:10,383 # Support for floating input (no pull resistors): yes
2023-03-17 18:55:10,388 # state: in, pull: none, schmitt trigger: off, value: off
2023-03-17 18:55:10,388 #
2023-03-17 18:55:10,392 # Testing output configurations for PIN_OUT_0:
2023-03-17 18:55:10,397 # Support for output (push-pull) with initial value of LOW: yes
2023-03-17 18:55:10,401 # state: out-pp, slew: slowest, value: off
2023-03-17 18:55:10,404 # Output is indeed LOW: yes
2023-03-17 18:55:10,408 # state: out-pp, slew: slowest, value: on
2023-03-17 18:55:10,411 # Output can be pushed HIGH: yes
2023-03-17 18:55:10,417 # Support for output (push-pull) with initial value of HIGH: yes
2023-03-17 18:55:10,420 # state: out-pp, slew: slowest, value: on
2023-03-17 18:55:10,424 # Output is indeed HIGH: yes
2023-03-17 18:55:10,430 # Support for output (open drain with pull up) with initial value of LOW: no
2023-03-17 18:55:10,437 # Support for output (open drain with pull up) with initial value of HIGH: no
2023-03-17 18:55:10,443 # Support for output (open drain) with initial value of LOW: yes
2023-03-17 18:55:10,449 # state: out-od, slew: slowest, pull: none, schmitt trigger: off, value: off
2023-03-17 18:55:10,452 # Output is indeed LOW: yes
2023-03-17 18:55:10,458 # Support for output (open drain) with initial value of HIGH: yes
2023-03-17 18:55:10,465 # state: out-od, slew: slowest, pull: none, schmitt trigger: off, value: on
2023-03-17 18:55:10,470 # state: in, pull: down, schmitt trigger: off, value: off
2023-03-17 18:55:10,474 # Output can indeed be pulled LOW: yes
2023-03-17 18:55:10,478 # state: in, pull: up, schmitt trigger: off, value: on
2023-03-17 18:55:10,483 # Output can indeed be pulled HIGH: yes
2023-03-17 18:55:10,488 # Support for output (open source) with initial value of LOW: no
2023-03-17 18:55:10,494 # Support for output (open source) with initial value of HIGH: no
2023-03-17 18:55:10,501 # Support for output (open source with pull up) with initial value of HIGH: no
2023-03-17 18:55:10,508 # Support for output (open source with pull up) with initial value of LOW: no
2023-03-17 18:55:10,511 # Support for disconnecting GPIO: yes
2023-03-17 18:55:10,515 # Output can indeed be pulled LOW: yes
2023-03-17 18:55:10,519 # Output can indeed be pulled HIGH: yes
2023-03-17 18:55:10,519 #
2023-03-17 18:55:10,523 # Testing Reading/Writing GPIO Ports
2023-03-17 18:55:10,526 # ==================================
2023-03-17 18:55:10,526 #
2023-03-17 18:55:10,529 # testing initial value of 0 after init
2023-03-17 18:55:10,531 # ...OK
2023-03-17 18:55:10,535 # testing setting both outputs_optional simultaneously
2023-03-17 18:55:10,537 # ...OK
2023-03-17 18:55:10,541 # testing clearing both outputs_optional simultaneously
2023-03-17 18:55:10,543 # ...OK
2023-03-17 18:55:10,547 # testing toggling first output (0 --> 1)
2023-03-17 18:55:10,548 # ...OK
2023-03-17 18:55:10,552 # testing toggling first output (1 --> 0)
2023-03-17 18:55:10,553 # ...OK
2023-03-17 18:55:10,557 # testing toggling second output (0 --> 1)
2023-03-17 18:55:10,558 # ...OK
2023-03-17 18:55:10,562 # testing toggling second output (1 --> 0)
2023-03-17 18:55:10,563 # ...OK
2023-03-17 18:55:10,569 # testing setting first output and clearing second with write
2023-03-17 18:55:10,570 # ...OK
2023-03-17 18:55:10,575 # testing setting second output and clearing first with write
2023-03-17 18:55:10,576 # ...OK
2023-03-17 18:55:10,580 # All input/output operations worked as expected
2023-03-17 18:55:10,580 #
2023-03-17 18:55:10,580 #
2023-03-17 18:55:10,582 # TEST SUCCEEDED
2023-03-17 18:55:10,588 # { "threads": [{ "name": "main", "stack_size": 1536, "stack_used": 456 }]}
```
</details>
<details><summary><code>make -C tests/bench_periph_gpio_ll BOARD=nucleo-f103rb flash term</code></summary>
```
2023-03-17 18:55:42,192 # Help: Press s to start test, r to print it is ready
s
2023-03-17 18:55:44,616 # START
2023-03-17 18:55:44,624 # main(): This is RIOT! (Version: 2023.04-devel-683-g9c3812-cpu/stm32/periph/gpio_ll)
2023-03-17 18:55:44,624 #
2023-03-17 18:55:44,626 # Benchmarking GPIO APIs
2023-03-17 18:55:44,628 # ======================
2023-03-17 18:55:44,628 #
2023-03-17 18:55:44,632 # estimating loop overhead for compensation
2023-03-17 18:55:44,635 # -----------------------------------------
2023-03-17 18:55:44,642 # 4168 us for 50000 iterations
2023-03-17 18:55:44,642 #
2023-03-17 18:55:44,647 # periph/gpio: Using 2x gpio_set() and 2x gpio_clear()
2023-03-17 18:55:44,651 # ---------------------------------------------------
2023-03-17 18:55:44,706 # 50000 iterations took 45840 us (50008 us uncompensated)
2023-03-17 18:55:44,713 # Two square waves pins at 1090750 Hz ( 999840 Hz uncompensated)
2023-03-17 18:55:44,719 # ~66 CPU cycles per square wave period (~72 cycles uncompensated)
2023-03-17 18:55:44,719 # :'-(
2023-03-17 18:55:44,719 #
2023-03-17 18:55:44,724 # periph/gpio_ll: Using gpio_ll_set() and gpio_ll_clear()
2023-03-17 18:55:44,729 # -------------------------------------------------------
2023-03-17 18:55:44,738 # 50000 iterations took 695 us (4863 us uncompensated)
2023-03-17 18:55:44,745 # Two square waves pins at 71942446 Hz ( 10281719 Hz uncompensated)
2023-03-17 18:55:44,750 # ~1 CPU cycles per square wave period (~7 cycles uncompensated)
2023-03-17 18:55:44,751 # :-D
2023-03-17 18:55:44,751 #
2023-03-17 18:55:44,755 # periph/gpio: Using 4x gpio_toggle()
2023-03-17 18:55:44,757 # -----------------------------------
2023-03-17 18:55:44,965 # 50000 iterations took 198646 us (202814 us uncompensated)
2023-03-17 18:55:44,972 # Two square waves pins at 251704 Hz ( 246531 Hz uncompensated)
2023-03-17 18:55:44,977 # ~286 CPU cycles per square wave period (~292 cycles uncompensated)
2023-03-17 18:55:44,978 # :'-(
2023-03-17 18:55:44,978 #
2023-03-17 18:55:44,982 # periph/gpio_ll: Using 2x gpio_ll_toggle()
2023-03-17 18:55:44,985 # -----------------------------------------
2023-03-17 18:55:45,010 # 50000 iterations took 15972 us (20140 us uncompensated)
2023-03-17 18:55:45,017 # Two square waves pins at 3130478 Hz ( 2482621 Hz uncompensated)
2023-03-17 18:55:45,023 # ~23 CPU cycles per square wave period (~29 cycles uncompensated)
2023-03-17 18:55:45,023 # :'-(
2023-03-17 18:55:45,023 #
2023-03-17 18:55:45,026 # periph/gpio: Using 4x gpio_write()
2023-03-17 18:55:45,029 # ----------------------------------
2023-03-17 18:55:45,097 # 50000 iterations took 58345 us (62513 us uncompensated)
2023-03-17 18:55:45,103 # Two square waves pins at 856971 Hz ( 799833 Hz uncompensated)
2023-03-17 18:55:45,109 # ~84 CPU cycles per square wave period (~90 cycles uncompensated)
2023-03-17 18:55:45,109 # :'-(
2023-03-17 18:55:45,110 #
2023-03-17 18:55:45,113 # periph/gpio_ll: Using 2x gpio_ll_write()
2023-03-17 18:55:45,117 # ----------------------------------------
2023-03-17 18:55:45,128 # 50000 iterations took 2777 us (6945 us uncompensated)
2023-03-17 18:55:45,135 # Two square waves pins at 18005041 Hz ( 7199424 Hz uncompensated)
2023-03-17 18:55:45,141 # ~4 CPU cycles per square wave period (~10 cycles uncompensated)
2023-03-17 18:55:45,141 # :-)
2023-03-17 18:55:45,141 #
2023-03-17 18:55:45,141 #
2023-03-17 18:55:45,142 # TEST SUCCEEDED
2023-03-17 18:55:45,149 # { "threads": [{ "name": "main", "stack_size": 1536, "stack_used": 448 }]}
```
</details>
### Issues/PRs references
None
Co-authored-by: Marian Buschsieweke <marian.buschsieweke@ovgu.de>
Use the largest instead of the smallest number of available EPs for this definition. This became necessary to be able to use all EPs of a USB OTG HS peripheral if enabled.
19371: sys/usbus: check for the number of required and provided EPs in static configurations r=dylad a=gschorcht
### Contribution description
This PR provides a static check at compile time whether the number of EPs required in a static configuration does not exceed the number of EPs provided by the USB device.
#### Background
In issue #19359 the problem was reported that `usbus_cdc_ecm` didn't work together with `stdio_cdc_acm` on some STM32 boards. The reason for some of the boards was simply that the application tried to allocate more EPs than available and simply ignored this and just didn't work.
#### Solution
Since `auto_init_usb` uses a static configuration with exactly one USBUS stack instance and one USB device, at least in case `auto_init` is used a static check can be carried out to make sure that the number of EPs required by the application doesn't exceed the number of EPs provided by the USB device. For this purpose, each `usbus_*` module defines the number of IN and OUT EPs required by that module. Each USB device driver defines the number of EPs provided by USB device if it differs from the default of 8 EPs. During the auto initialization the total number of required IN and OUT EPs is then compared with the number of EPs provided by the USB device using a static assert.
### Testing procedure
1. Green CI
2. Compilation of
```python
USEMODULE='stdio_cdc_acm' BOARD=nucleo-f439zi make -j8 -C tests/usbus_cdc_ecm
```
should lead to compilation error
```python
sys/auto_init/usb/auto_init_usb.c:81:1: error: static assertion failed: "Number of required IN endpoints exceeded"
_Static_assert(USBUS_EP_IN_REQUIRED_NUMOF <= USBDEV_NUM_ENDPOINTS,
^~~~~~~~~~~~~~
Makefile.base:146: recipe for target 'tests/usbus_cdc_ecm/bin/nucleo-f439zi/auto_init_usbus/auto_init_usb.o' failed
```
while compilation of
```
USEMODULE='stdio_cdc_acm' BOARD=nucleo-f767zi make -j8 -C tests/usbus_cdc_ecm
```
should work.
### Issues/PRs references
Fixes issue #19359 partially.
19382: tests/pkg_nanors: use static allocation r=benpicco a=benpicco
Co-authored-by: Gunar Schorcht <gunar@schorcht.net>
Co-authored-by: Benjamin Valentin <benpicco@beuth-hochschule.de>
The STM32 periph_timer driver reads the timer's status flags, then
clears them all. It is possible that a timer interrupt could occur
between reading the flag and clearing it. This would lead to a lost
interrupt.
The timer's status flags can be cleared by software, but can only be set
by the hardware. This patch takes advantage of this by only clearing the
flags it knows are set. The rest of the flags are set, which doesn't
actually change their state.
17086: usbdev: Add dedicated stall functions r=benpicco a=bergzand
### Contribution description
This PR adds dedicated stall functions for usbdev peripherals. Two
functions are added. The first function (usbdev_ep_stall) to enable and
disable the stall condition on generic endpoints. The second function is
a dedicated function to set the stall condition on endpoint zero in both
directions. This status can only be set and should automatically be
cleared by the usbdev implementation (or hardware) after a new setup
request is received from the host.
### Testing procedure
- examples/usbus_minimal should still enumerate correctly on the host side.
- #17085 can be used to demonstrate the ep0_stall function with the `tests/usbus_cdc_acm_stdio/` test
### Issues/PRs references
None
Co-authored-by: Koen Zandberg <koen@bergzand.net>
Co-authored-by: Gunar Schorcht <gunar@schorcht.net>
19343: ztimer: add ztimer_stopwatch convenience functions r=benpicco a=benpicco
19349: cpu/native: Switch to ztimer for gettimeofday r=benpicco a=MrKevinWeiss
### Contribution description
A xtimer is somewhat taken over by ztimer this explicitly uses ztimer instead of relying on the compatibility layer.
### Testing procedure
`make all test -C tests/cpp11_mutex/`
and green murdock I guess.
### Issues/PRs references
19353: doc: add quicklink to boards in navbar r=benpicco a=OlegHahm
### Contribution description
Finding a list of supported boards and how to use them is an essential information. Currently this list is somewhat hidden under "Modules" which is not very intuitive. Hence, I propose to (at least) put a link in the side menu to this overview page.
### Testing procedure
1. Call `make doc`
2. Check the sidebar `${RIOT_BASE}/doc/doxygen/html/index.html` for an entry "Supported Boards"
19361: nanocoap_sock: ensure response address is the same as request address r=benpicco a=benpicco
19363: Fix stm32 timer periodic r=benpicco a=Enoch247
### Contribution description
From the commit msg:
> cpu/stm32/periph/timer: remove unneeded header
>
> I see no reason this header should be included. It does not exist in
> RIOT's source tree. This patch removes the include.
and
> cpu/stm32/periph/timer: fix execution flow
>
> The implmentation of `timer_set_absolute()` has The following problems.
> First, it attempts to restore the auto reload register (ARR) to it's
> default if the ARR was previosly set by `timer_set_periodic()` by
> comparing it to the channel's capture compare (CC) register _after_ it
> has already set the CC register. Secondly, it clears spurious IRQs
> _after_ the CC register has been set. If the value being set is equal to
> the timer's current count (or the two become equal before the supurios
> IRQ clearing happens), this could cause a legitimate IRQ to be cleared.
>
> The implmentation of `timer_set()` has the same error in handling the
> ARR as described above.
>
> This patch reorders the operations of both functions to do:
>
> 1. handle ARR
> 2. clear spurious IRQs
> 3. set channel's CC
> 4. enable IRQ
>
> Additionally, the calulation of `value` in `timer_set()` is moved
> earlier in the function's exec path as a pedantic measure.
### Testing procedure
I tested by doing the following:
1. `make -C tests/periph_timer BOARD=nucleo-f767zi all flash term`
2. press s
3. press [ENTER]
4. observe test passes
5. `make -C tests/periph_timer_periodic BOARD=nucleo-f767zi all flash term`
6. press s
7. press [ENTER]
8. observe test passes
9. `make -C tests/periph_timer_short_relative_set BOARD=nucleo-f767zi all flash term`
10. press s
11. press [ENTER]
12. observe test passes
### Issues/PRs references
- none known
Co-authored-by: Benjamin Valentin <benpicco@beuth-hochschule.de>
Co-authored-by: MrKevinWeiss <weiss.kevin604@gmail.com>
Co-authored-by: Oleg Hahm <oleg@hobbykeller.org>
Co-authored-by: Benjamin Valentin <benjamin.valentin@bht-berlin.de>
Co-authored-by: Joshua DeWeese <jdeweese@primecontrols.com>
The implmentation of `timer_set_absolute()` has The following problems.
First, it attempts to restore the auto reload register (ARR) to it's
default if the ARR was previosly set by `timer_set_periodic()` by
comparing it to the channel's capture compare (CC) register _after_ it
has already set the CC register. Secondly, it clears spurious IRQs
_after_ the CC register has been set. If the value being set is equal to
the timer's current count (or the two become equal before the supurios
IRQ clearing happens), this could cause a legitimate IRQ to be cleared.
The implmentation of `timer_set()` has the same error in handling the
ARR as described above.
This patch reorders the operations of both functions to do:
1. handle ARR
2. clear spurious IRQs
3. set channel's CC
4. enable IRQ
Additionally, the calulation of `value` in `timer_set()` is moved
earlier in the function's exec path as a pedantic measure.
19341: Picolibc updates r=benpicco a=keith-packard
### Contribution description
Minor updates to picolibc support for newer versions of picolibc, including 1.8.
### Testing procedure
Here's the current build result for a board I happen to have on my bench right now:
```
$ make -C examples/blinky BOARD=nucleo-f103rb FEATURES_REQUIRED=picolibc
...
/home/keithp/src/RIOT/sys/picolibc_syscalls_default/syscalls.c:319:1: error: unknown type name '_READ_WRITE_RETURN_TYPE'
319 | _READ_WRITE_RETURN_TYPE read(int fd, void *dest, size_t count)
| ^~~~~~~~~~~~~~~~~~~~~~~
/home/keithp/src/RIOT/sys/picolibc_syscalls_default/syscalls.c:351:1: error: unknown type name '_READ_WRITE_RETURN_TYPE'
351 | _READ_WRITE_RETURN_TYPE write(int fd, const void *src, size_t count)
| ^~~~~~~~~~~~~~~~~~~~~~~
...
```
### Description of changes
1. Fix the read/write return values (Picolibc 1.8 uses POSIX standard types now)
2. Add new symbols to linker scripts (Picolibc needs help dealing with TLS alignment)
3. Align stack and thread local storage block during thread setup.
19345: cpu/riscv_common: convert to uword_t usage r=benpicco a=dylad
### Contribution description
This PR makes use of `uword_t` type in a few places in `cpu/riscv_common`.
This should not have any effect on produced binaries.
### Testing procedure
CI should be enough.
### Issues/PRs references
Quickly adapt from #16994. It will help for 64 bits support one day ;)
Co-authored-by: Keith Packard <keithp@keithp.com>
Co-authored-by: Dylan Laduranty <dylan.laduranty@mesotic.com>
Newer picolibc versions require some additional symbols defined in the
linker script to correctly manage alignment constraints on thread
local storage.
Signed-off-by: Keith Packard <keithp@keithp.com>
19329: MAINTAINING.md: add some notes on Bors r=benpicco a=benpicco
19330: native/stdio: Explicitly provide getchar r=chrysn a=chrysn
### Contribution description
This ensures that even when libc does not implement getchar through getc, any custom stdio is still in the loop when getchar is used.
Frankly, I don't know when this broke -- I'm pretty sure custom stdio worked just a few days ago -- but either way, without this patch RIOT on native currently bypasses a configured stdio for me.
### Testing procedure
* `make -C examples/saul all debug`
* `break stdio_read`
* `run`
Without this patch, observe how the shell runs w/o ever breaking. After, lots of breakpoint hits.
This is the way it behaves for me (Debian sid, libc6:i386 2.36-8). If it works for you before this patch, we might start bisecting the differences between the systems, but we may also accept that libcs may imlpement getchar in different ways, and not all of them pass by the getc which we're patching.
### Issues/PRs references
This is needed for testing #19289.
The implementation stems from the `fgetc(3)` man page, which states that "getchar() is equivalent to getc(stdin)".
Co-authored-by: Benjamin Valentin <benpicco@beuth-hochschule.de>
Co-authored-by: chrysn <chrysn@fsfe.org>
This allows automatically moving format strings to flash, provided that
code previously compiled fine with `-Wformat-nonliteral` (which in RIOT
is the case due to `-Wformat=2`).
This adds a layer of convenience abstraction over classical Harvard
architectures (like most AVRs) that do not map the flash memory into
the data address space and modern Harvard architectures or von-Neumann
architectures that do so. The motivation is to safe a lot of RAM for
AVR by storing constant strings into flash.
