Testers needed: IR Toy firmware v20 test 10

IR Toy firmware v20 is nearly ready for release. This version uses the new open source USB stack written by JTR and Honken.

The transmit, and receive functions have been completely reworked. The transmitter can handle 1000s of bytes of test signals without a glitch thanks to the new USB double-buffer.

There’s only one bug left before a release candidate is ready. Transmit data in multiples of 8 bytes crashes the IR Toy on our test setup. The first time it works, the second time the test hangs trying to open the serial port. Packet sizes not divisible by 8 work perfectly every time.

We could use your help testing the bug on different computers. There’s a package in the forum with everything you need to install the test firmware and verify the multiple of 8 bug. Please report your findings in the comments or the forum.

Primary tests:

  • Play-ok.bat – plays 5586bytes ok. Run it twice, it works fine both times.
  • play-fail.bat – plays 5588bytes ok. Run it twice, it works once, the second time the app fails. Replug the IR Toy to reset

Secondary tests:

  • play-ok-short.bat – plays 12 byte file ok, run it twice. there is an error, but IR Toy does not freeze
  • play-fail-short.bat – plays 8byte file. Run it twice, it works once, the second time the app fails. Replug the IR Toy to reset

Firmware v20 is otherwise working well, and should work (untested!) with the WinLIRC plugin, as long as transmit data is not on an 8byte boundary

Join the Conversation

5 Comments

  1. I updated the firmware and ran the tests as described above. Results are as described.
    Need more info or want me to run some additional tests? Just let me know.

    1. Thanks Menno,

      Any more general-use testing would be great. I’ve been using the latest versions in my HTPC and they have been working great. Any real-world use would be great.

      1. Well, my IR Toy hasn’t any real-world use yet. Other projects first.
        I’d be willing to run any (semi-)automated and pre-defined test though.

  2. I added details of my results to the forum post, summary is:

    Test results Summary – IRT v1.1, on Win XP
    Primary test 1 – Play-ok.bat: run 1 passed, run 2 had errors, but IRT did not hang.
    Primary test 2 – Play-fail.bat: run 1 & 2 errors, IRT hang on 2nd test (as expected?)
    Secondary test 1 – Play-ok-short.bat run 1 & 2 errors
    Secondary test 2 – Play-fail-short.bat: run 1
    WinLirc 1st send OK, 2nd – no LED, IRT seems hung.
    I have no idea if codes I sent contained the 8 byte boundary.

    Ubuntu 11
    GtkTerm – no response until sent 5 x0H
    R, V, T = OK
    E – no LED, no text sent from IRT, no response after sent 5 x0H
    un/replugged – noticed teh power on LED for ~1sec = good. recent test fw version bad behaviour.
    same as above, except did not have to send initial 5 x0H.
    In both cases exiting GtkTerm – it hung = past symptom of irt/coms issues
    un/replugged – r, t, v, s = OK.

    These results are repeatable on my config.

    I have not yet tried JTR_IRToyRecPlay_test9, sorry out of time today.

    QUERY: – should the IRT have been un/plugged after failure of test 1, or even before every test run?

    Also a big thanks for the test exe’s and batch files, makes testing a lot easier!

Leave a comment

Your email address will not be published. Required fields are marked *

Notify me of followup comments via e-mail. You can also subscribe without commenting.