Add new target: DISCO_STM32H573I (STM32H573I-DK)

[wdx04]

Summary of changes

This PR adds new target DISCO_STM32H573I (STM32H573I-DK) and the following features for the STM32H5 family:
QSPI with DMA (read speed increased from 4MB/s without DMA to 39MB/s with [email protected])
OSPI with DMA (read speed increased from 4MB/s without DMA to 60MB/s with [email protected])
USB

Impact of changes

Mbed CE now officially supports DISCO_STM32H573I (STM32H573I-DK).
QSPI, OSPI and USB features for the STM32H5 family are enabled.
OSPI DMA is also enabled and tested on B-U585I-IOT02A and STM32L4R9I-DISCO.

Migration actions required

Documentation

No documentation at the moment.

---

Pull request type

[] Patch update (Bug fix / Target update / Docs update / Test update / Refactor)
[X] Feature update (New feature / Functionality change / New API)
[] Major update (Breaking change E.g. Return code change / API behaviour change)

---

Test results

[] No Tests required for this change (E.g docs only update)
[] Covered by existing mbed-os tests (Greentea or Unittest)
[X] Tests / results supplied as part of this PR

greentea-log-h573.txt

The following tests FAILED:
21 - test-mbed-usb-device-basic (Failed)
22 - test-mbed-usb-device-hid (Failed)
101 - test-mbed-connectivity-netsocket-nanostack-dns (Failed)
104 - test-mbed-connectivity-netsocket-nanostack-tcp (Failed)
105 - test-mbed-connectivity-netsocket-lwipstack-tls (Failed)
106 - test-mbed-connectivity-netsocket-nanostack-tls (Failed)
108 - test-mbed-connectivity-netsocket-nanostack-udp (Failed)
109 - test-mbed-connectivity-network-emac (Failed)
111 - test-mbed-connectivity-netsocket-nanostack-network-interface (Failed)

Apart from the tests related to netsocket, I'm unable to pass 'usb-device-basic' and 'usb-device-hid' tests. On boards known to have good USB support, such as NUCLEO-F413ZH, these tests also failed. Maybe these tests require a specific version of libusb to be installed?

---

[JohnK1987]

Would be cool to hold the content of PR around its original idea. The name of PR is about add a STM32H5 dev board but inside is a lot of fixes for DMA and XSPI.

The 'usb-device-basic' is failing with every STM32 target and usually it was caused by a Endpoint Abort part, if I remember.
For usb-device-hid try to look for readme in USB driver folder and missing python modules.

[wdx04]

Would be cool to hold the content of PR around its original idea. The name of PR is about add a STM32H5 dev board but inside is a lot of fixes for DMA and XSPI.

The 'usb-device-basic' is failing with every STM32 target and usually it was caused by a Endpoint Abort part, if I remember. For usb-device-hid try to look for readme in USB driver folder and missing python modules.

No other STM32 board has actually used XSPI yet. I'm adding it here mainly to provide a complete featureset for this new target, while not breaking existing boards.

[JohnK1987]

Ah, ok, my bad.

I did the GT test of USBHID with Nucleo-H563ZI and it pass - #437 (comment)

[wdx04]

I did the GT test of USBHID with Nucleo-H563ZI and it pass - #437 (comment)

I have already installed hidapi 0.14.0.post2, so that is not the issue.

The error message of the USB HID test is:

22: [+969ms][CONN][RXD] >>> Running case #2: 'Configuration descriptor, keyboard'...
22: [+980ms][CONN][INF] found KV pair in stream: {{__testcase_start;Configuration descriptor, keyboard}}, queued...
22: [+991ms][CONN][INF] found KV pair in stream: {{test_get_desc_cfg;0}}, queued...
22: [+3119ms][HTST][INF] TEST ERROR: Failed with "The device has no langid (permission issue, no string descriptors supported or device error)". Tried 20 times.

For the USB basic test, the error message is:

21: [+536ms][CONN][RXD] >>> Running case #1: 'usb control basic test'...
21: [+536ms][CONN][INF] found KV pair in stream: {{__testcase_start;usb control basic test}}, queued...
21: [+827ms][CONN][INF] found KV pair in stream: {{control_basic_test;0123456789 3368 517}}, queued...
21: <<< get_set_configuration_test >>>
21: [+1569ms][HTST][INF] TEST FAILED: Operation not supported or unimplemented on this platform

Both of these errors are related to pyusb.

[JohnK1987]

My setup is

• pyusb-1.3.1
• hidapi-0.14.0.post2
• libusb0-1.4.0 (installed over Zidag with Mbed configs)

[multiplemonomials]

OK reviewed the first half of the change (up through USBPhy_STM32.cpp)!

[multiplemonomials]

Still reviewing, not done yet but more comments!

[multiplemonomials]

Nice work on this! Looks good, just a few last tiny things.