Attention is currently required from: dexter, pespin.

View Change

1 comment:

• File src/gprs_rlcmac_sched.cpp:

  • Patch Set #1, Line 486: * way we help BTS energy saving (on TRX!=C0) by sending nothing

    @fixeria bear in mind the "skip_idle && trx != 0" was only applied inside the #ifdef DIRECT_PHY [...]

    It's getting more and more complicated, but it seems I have a more complete picture in my mind. The `#ifdef DIRECT_PHY` block was removed in a recent patch (I7a08d8cc670fa14f7206ffffdbc22351f3668a17), and now this patch removes the `trx != 0` constraint.

    And that was done in order to make it send stuff on C0 because those osmo-bts lowerlayers are not filling the block otherwise.

    Ack. The `trx != 0` constraint was specifically needed for those DSP based models.

    I realize now though, that when proposing the cleanup in this patch, I assumed those were 1 TRX only, but lc15 has several of them right?

    I am not familiar with the lc15 PHY, but AFAIR there also was a sysmoBTS model with two TRX. Also, I don't know how those DSP based models behave when getting an empty DATA.req for `trx!=0`. Was this ever tested?

To view, visit change 35151. To unsubscribe, or for help writing mail filters, visit settings.