From: Jonathan W. <jw...@ph...> - 2008-12-04 23:29:46
|
Hi Pieter Over on the ffado-users list (in the "MOTU Ultralite" thread) I'm trying to work out why a few users don't seem to be able to get reliable streaming. Since I know you're busy (and because this is more a developer issue than a user one) I've distilled the questions I have for you here in case it's easier for you to deal with right now. The system's firewire chipset is reported as this: 0a:09.0 FireWire (IEEE 1394): Ricoh Co Ltd R5C832 IEEE 1394 Controller (rev 05) Do you know off-hand whether this is one of the "bad" ones? [ As an aside, we probably need to collect all the "known bad" chipsets (and maybe the "known good" as well) and put them on the web. I might see if I can start something in the next few weeks if no one else does it first. This would help immensely in situations like this. ] On this particular system things seem to start OK (and audio is produced), but after a short period of time we see a series of messages like this: getPacket: reconstructed CTR counter discrepancy : getPacket: diffcy = -111 waitForPeriod: wait extended since period not ready... getPacket: (0x1915ff0) OHCI issue on cycle 7055 (dropped_cycles=0, last_cycle=7055, dropped=1, skipped: 0) getPacket: reconstructed CTR counter discrepancy : getPacket: diffcy = -110 : getPacket: diffcy = -109 jackd then closes down in response to an xrun. Is this the signature of an infamous "dropped packet" (suggested by the "OHCI issue" message) - and therefore a bad firewire chipset? Or is it something more beneign like USB shared interrupts, or not having a "low latency desktop" kernel? What exactly does the "reconstructed CTR counter discrepancy" message mean? >From another user (to whom I haven't replied yet) I'm seeing messages like Execute: (0x8071cf8) have to retry CTR read, diff unrealistic from CycleTimerHelper.cpp)[ 389]. Is this anything to worry about? This system uses an on-board SiS firewire chipset - do we have any experience with those? Regards jonathan |