11:17 <karlp> heh, I was tryign to figure out if oocd's rtt impl had a bug in the way it handled initial states of poitners,

11:17 <karlp> I always get some "old" data when I connect, then the "present" data turns up.

11:17 <karlp> but no, it turns out the RTT target code has no "just keep writing and wrapping" option

11:18 <karlp> so it literally writes old data, then ignores writes until the host reads out the buffer, then it allows new writes.

11:18 <karlp> so oocd is doing the best it can.

11:18 <karlp> and the three modes available are "don't write anything if will fill up" "write to the end, trimming, but still no wrap" and "block"

11:19 <karlp> there's no "wrap around, make it the host's problem for not reading enough, and we'll make sure that the most recent is in the buffer"

18:18 <zapb_> karlp, puh, no bug :)

18:18 <zapb_> karlp, there is a RTT rust implementation which is quite nice and may have different modes