Attention is currently required from: falconia.
pespin has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-bts/+/32098 )
Change subject: common: implement rtp always-output mode ......................................................................
Patch Set 2:
(3 comments)
File src/common/l1sap.c:
https://gerrit.osmocom.org/c/osmo-bts/+/32098/comment/d308483b_46cafad6 PS2, Line 1250: /* A zero-length payload is never valid, it is merely a BFI marker Can you explain a bit more this change in logic here? It's not explained in the commit description so it's not clear why you added this.
https://gerrit.osmocom.org/c/osmo-bts/+/32098/comment/929c99d1_37b7f337 PS2, Line 1253: if (!resp_msg->len) == 0 would be clearer here, since len is an int and not a bool.
File src/common/vty.c:
https://gerrit.osmocom.org/c/osmo-bts/+/32098/comment/5ab42375_db8e1976 PS2, Line 425: vty_out(vty, " rtp always-output%s", VTY_NEWLINE); what about "rtp tx-lost" or "rtp tx-forged-lost" or similar? "always-output" is a bit difficult to understand its meaning.