Attention is currently required from: daniel, fixeria.

View Change

6 comments:

• File src/input/ipaccess.c:

  • Patch Set #1, Line 494: OSMO_ASSERT(cli);

    Why are you so assert()ive? ;) […]

    If you dive into the code you'd like to put even more asserts!
    After work on the BSC side I think I indeed can remove this one, will do.

  • Patch Set #1, Line 545: osmo_stream_cli_send

    Why not returning `rc` of this function?

    because osmo_stream_cli_send() returns void.
    YES, I KNOW, BIG FAIL!!!!! Remember this case next time you see a new API added returning void 😊

    I'm also thinking about creating a osmo_stream_cli_send2() which returns int, and deprectate the other one. That can be done later on though.

    For osmo_fd backend that was not much of a problem (because we were not checking for a max queue size at the time), but for osmo_io backend it can fail when reaching max length of its internal queue...

  • Patch Set #1, Line 928: *msgb_put(nmsg2, 1) = IPAC_MSGT_ID_ACK;

    `msgb_v_put(nmsg2, IPAC_MSGT_ID_ACK)`

    Done

  • Patch Set #1, Line 944: uint8_t *data = msgb_l2(msg);

    `const`?

    Done

  • Patch Set #1, Line 992:

    err:
    	return -1;
    

    `goto`s can be replaced with `return -1`?

    Yes, but I prefer having all error paths easily visible this way.

  • Patch Set #1, Line 1014: e1i_ts->line

    `s/e1i_ts->line/line/`

    Done

To view, visit change 38927. To unsubscribe, or for help writing mail filters, visit settings.