ChibiOS Free Embedded RTOS

Ah i see. It has been fixed in 18.2.2.

Thanks!

Hi Giovanni,

in STM32L4xx mcuconf.h there is STM32_LSCOSEL but in hal_lld.c it is not taken into account when setting up the clocks.

Best regards

Vinz

Hi Giovanni,

I have bought a nucleo board and tried to reproduce the problem there without result.
In the end it was an issue with my hardware.

Thanks for your time!

Vinz

Can I suggest you set up the scope to look at those signals in differential mode? That way you'll have a better idea of the actual logic levels seen by the CPU. It may be that the data is always one or zero; or maybe there are edges which cause the data to be misinterpreted.

Hi Steved,

thanks ...

Attached you will find a screenshot from sigroc logic analyzer as well as from my scope.
It looks like after the data0 of the setup there is garbage on the line.
I can't tell if this is caused by the peripheral not doing the right thing or a hardware problem.

Here its "SYNC ERR" to the right ...

Could this also be caused by faulty hardware e.g. bad usb signalling? Does it make sense to investigate in that direction e.g. check the signals and make a trace using a usb protocol analyzer?

DOEPINT_13 seems to be "nak":
https://www.intel.com/content/www/us/en ... 93052.html

Maybe this makes sense to you.

Trace of different error on different host port:


...

lld_serve_interrupt(sts = 0x00000008 /* GINTSTS_SOF */)

lld_serve_interrupt(sts = 0x00000008 /* GINTSTS_SOF */)

lld_serve_interrupt(sts = 0x00000008 /* GINTSTS_SOF */)

lld_serve_interrupt(sts = 0x00000010 /* GINTSTS_RXFLVL ...

Trace of failure:


...

lld_serve_interrupt(sts = 0x00000008 /* GINTSTS_SOF */)

lld_serve_interrupt(sts = 0x00000008 /* GINTSTS_SOF */)

lld_serve_interrupt(sts = 0x00000008 /* GINTSTS_SOF */)

lld_serve_interrupt(sts = 0x00080000 /* GINTSTS_OEPINT */)
OEPINT(src = 0x00010000)
otg ...

Giovanni wrote:Could you give a try to trunk code? there is a significant difference in the OTGv1 driver.

Giovanni

Did test against trunk. Same behavior.

Vinz