TF Riser Revision 0 Arrives

[arkadiusz.makarenko]

The Xilinx chips are hard to kill. Can you ID it?

Yes. I did read chain. Then I did wrire+verify, failed on verify. Did it again and all went throught.

I hope to test it today or tomorrow.

[arkadiusz.makarenko]

I am reading more on stm usb, a stm32 it seems that there are commercial 3rd party supporting hubs... https://community.st.com/s/question/0D5 ... st-library
So this suggest that it is library limitation, not hardware.

I bought this for tests. https://www.argos.co.uk/product/8239299

Only keyboard works, as mouse is on second configuration. Library doesn't support multiple configurations, but I was wondering if I could try and hard code interface and endpoints swap after each successful data read. Just to check if library can be "hacked" to read both.

[terriblefire]

I am reading more on stm usb, a stm32 it seems that there are commercial 3rd party supporting hubs... https://community.st.com/s/question/0D5 ... st-library
So this suggest that it is library limitation, not hardware.

Yeah i'd also read this and wondered if using the MiST code was maybe a better approach. Or abstracting it up into a library that could be used with any hardware interface.

I bought this for tests. https://www.argos.co.uk/product/8239299

Only keyboard works, as mouse is on second configuration. Library doesn't support multiple configurations, but I was wondering if I could try and hard code interface and endpoints swap after each successful data read. Just to check if library can be "hacked" to read both.

I have a similar keyboard and mouse and its the same deal. Only the keyboard works.

[arkadiusz.makarenko]

I managed to connect it to CD32, and with programmed CPLD mouse on external CD32 port doesn't work at all, but it does with blanked CPLD.
Keyboard seems not to be responding as well... I need to have a look if I have ARM->CPLD communication set up correctly.

So I think my CPLD does work at least partially.
I will try to check if you left on your github older cpld firmware to test bare minimum.

[terriblefire]

I managed to connect it to CD32, and with programmed CPLD mouse on external CD32 port doesn't work at all, but it does with blanked CPLD.
Keyboard seems not to be responding as well... I need to have a look if I have ARM->CPLD communication set up correctly.

So I think my CPLD does work at least partially.
I will try to check if you left on your github older cpld firmware to test bare minimum.

Let me see if i can connect back to the barn and check if the firmware is updated.

The basic premise is that

with NSS high the SPI port talks direct keyboard.

With NSS low it talks SPI to the "MiST" type interface.

The byte sequence to update the mouse is

0x04 JOY0DATH JOY0DATL BUTTONS

4 bytes in total

Simple as that really.

[arkadiusz.makarenko]

I managed to connect it to CD32, and with programmed CPLD mouse on external CD32 port doesn't work at all, but it does with blanked CPLD.
Keyboard seems not to be responding as well... I need to have a look if I have ARM->CPLD communication set up correctly.

So I think my CPLD does work at least partially.
I will try to check if you left on your github older cpld firmware to test bare minimum.

Let me see if i can connect back to the barn and check if the firmware is updated.

The basic premise is that

with NSS high the SPI port talks direct keyboard.

With NSS low it talks SPI to the "MiST" type interface.

The byte sequence to update the mouse is

0x04 JOY0DATH JOY0DATL BUTTONS

4 bytes in total

Simple as that really.

It is not cpld. I had one spare and changed it, just to be 100% sure. I can see exactly the same behaviour.
I did try older firmware from other commits, and I can see the same issue.

Anyway, I can progress other things. I will need to write hid report descriptor parser, and some kind of auto mapping routine. I am planning adapt code from mister.
And try to solve issue with mouse+keyboard composite device.

[terriblefire]

Its likely you have setup the SPI port for hardware not software. That will cause the issue you are seeing.

[arkadiusz.makarenko]

Its likely you have setup the SPI port for hardware not software. That will cause the issue you are seeing.

Should I see this issue even If I halt arm cpu by setting boot jumper?

[terriblefire]

Should I see this issue even If I halt arm cpu by setting boot jumper?

Its about whether the SPI port is set to software or hardware. If its set to hardware you might not be able to send to the mouse or keyboard.

[arkadiusz.makarenko]

Should I see this issue even If I halt arm cpu by setting boot jumper?

Its about whether the SPI port is set to software or hardware. If its set to hardware you might not be able to send to the mouse or keyboard.

Thank you. It was SPI port configuration!

Mouse still doesn't work, but it is irrelevant, as usb keyboard does!

So I have dev board!