zyp

  • Joined on Nov 14, 2009
Jul 11, 2016
zyp commented on pull request libopencm3/libopencm3#672
@zyp

When DOEPTSIZx.PKTCNT is always set to 1, every packet is always the last one, so that shouldn't really matter. I also believe that when it's confi…

Jul 11, 2016
zyp commented on pull request libopencm3/libopencm3#672
@zyp

I made some comments about @H2OBrain's comment in IRC earlier today. I think I might be onto something so I'm attaching it here too for reference. 1…

Jul 7, 2016
zyp commented on issue libopencm3/libopencm3#668
@zyp

No, it won't affect other endpoints. I considered your approach as well, but it'll be messier and it just masks the root cause instead of solving it.

Jul 7, 2016
zyp commented on issue libopencm3/libopencm3#668
@zyp

I agree with your analysis. Device is sitting with control_state.state = LAST_DATA_IN, TX completion should call _usbd_control_in() which changes s…

Jul 5, 2016
zyp commented on issue libopencm3/libopencm3#668
@zyp

After reading out a SETUP packet, you could check if there's more contents in the RXFIFO and if so read GRXSTSR to check what it is. If it's anothe…

Jul 5, 2016
zyp commented on issue libopencm3/libopencm3#668
@zyp

Ah, I think I see the problem. stm32fx07_ep_write_packet() looks at DIEPTSIZ.PKTCNT to check if the endpoint is already busy. I think a better chec…

Jul 5, 2016
zyp commented on issue libopencm3/libopencm3#668
@zyp

That sounds like a shot in the dark, yes. According to the RM you're not supposed to touch that register when EP0 IN is enabled, and when it's not …

Jul 4, 2016
zyp commented on issue libopencm3/libopencm3#668
@zyp

The libopencm3 USB stack itself shouldn't have any requirements on how often usb_poll() needs to be called, but it obviously can't do any work whil…

Jun 22, 2016
zyp commented on issue blacksphere/blackmagic#99
@zyp

That doesn't explain why my code were hardfaulting. When the bootloader is mapped in, it should run before anything gets to the entry point of my c…