Skip to content
This repository has been archived by the owner on Sep 28, 2019. It is now read-only.

"Remote failure reply: E0E" when OpenOCD is piped #18

Closed
ivankravets opened this issue Aug 10, 2017 · 8 comments
Closed

"Remote failure reply: E0E" when OpenOCD is piped #18

ivankravets opened this issue Aug 10, 2017 · 8 comments
Assignees
Labels

Comments

@ivankravets
Copy link

Hi @ilg-ul,

I don't know where is better to post this issue, would be thankful for a help. I can't reproduce this issue with 20170607-2132-dev but with v0.10.0-2-20170622-1535 I can.

Just pipe OpenOCD to GDB:

(gdb) target remote | openocd -c "gdb_port pipe" ...
...
undefined debug reason 7 - target needs reset
Remote failure reply: E0E
@ilg-ul
Copy link
Contributor

ilg-ul commented Aug 11, 2017

I'll investigate.

ilg-ul pushed a commit that referenced this issue Aug 26, 2017
riscv: Ensure that hart is halted before attempting to examine it.
@TommyMurphyTM1234
Copy link

TommyMurphyTM1234 commented Sep 2, 2017

Edit: sorry - ignore my post as I just saw the pull request dealing with the issue above!

_This may be due to the GDB/OpenOCD startup/init "race condition" whereby GDB gets started and tries to make a GDB Remote Serial Protocol connection before OpenOCD is quite ready to handle it? This can happen with the socket/port connection but maybe the same issue can arise with the pipe/stdio connection too? See here for more on this issue when using the socket/port connection:

eclipse-embed-cdt/eclipse-plugins#186

If you can launch OpenOCD standalone and then get GDB to connect to it OK then that may be evidence in support of this race condition being the problem.

Hope this helps._

@ilg-ul
Copy link
Contributor

ilg-ul commented Sep 2, 2017

I know there were some changes in OpenCOD that I did not study thoroughly, so it might be so.

as far as I know, there are no changes in GNU MCU Eclipse OpenOCD compared to master related to this issue, and if ther are, they probably come from merging the RISC-V code.

@JojoS62
Copy link

JojoS62 commented Sep 3, 2017

I get this error message also after exporting a project from mbed. Target is a nrf51822, debug probe a LPC-LINK2 with cmsis-dap firmware. Calling oocd 0.10.0-3-20170826-1813-dev manually for flashing works, but starting to debug gives the e0e error code. Also when I try to start gdb manually and to connect with 'target remote localhost:3333'.
When I use oocd 0.10.0 compiled from F. Chopin the debugger starts without errors.

@ilg-ul
Copy link
Contributor

ilg-ul commented Sep 25, 2017

looks the same as riscv-collab/riscv-openocd#115

@ilg-ul ilg-ul self-assigned this Sep 25, 2017
@ilg-ul ilg-ul added the bug label Sep 25, 2017
@ilg-ul
Copy link
Contributor

ilg-ul commented Jan 14, 2018

can you confirm if this was fixed in the latest binary release?

@ivankravets
Copy link
Author

Re-tested. It works with the latest GNU MCU Eclipse OpenOCD v0.10.0-6-20180112

Thanks and Happy New Year!

@ilg-ul
Copy link
Contributor

ilg-ul commented Jan 15, 2018

you're welcome, happy new year!

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
Projects
None yet
Development

No branches or pull requests

4 participants