Re: [Jack-Devel] Place for bug reporting
Please see below
On 03/06/2014 15:28, TimKa wrote:
> Dear Leonardo and Adrian, thanks for reply!
> I'm very sorry for long following text but I will try to repeat all my
> caveats during experience with jackd for my demo system.
> I tried to work with jackd1 and netone and I found next problem:
>
> --------------part of my mail from 21.05-----------
> /I have been cross-comiled jack1 0.124.1 from source and patch
> time.c, because i don't have bogomips in /proc/cpuinfo. I run jackd
> master and slave both on two AM335x-EVMSK boards. When I try to run
> jackd_netsource -H <master_IP> I get a lot of "nextrun amount 21ms"
> and less 106 or 61ms on master console. Unfortunately I can not find
> any information on Web by this issue so trying to find right way by
> mail list. //
> /
I can't find your exact EVM board, but I assume it can run up to 1GHz.
Did you run these tests with CPU scaling governor in performance mode? I
see you did in the last tests.
I haven't tried myself but I work on a similar processor and I know
jackd can consume a lot of CPU power if the period size is low. What is
the period size you set on the master machine? Anything from 256 up
should be fine. Remember the period size is imposed by the master.
I assume you run the board headless, just a console, no X, no desktop
manager, etc...
> ///
> //... //
> //(for jackd -d netone) //
> //JackEngine::XRun: client = alsa_in was not finished, state =
> Triggered //
> //JackEngine::XRun: client = alsa_out was not finished, state =
> Triggered //
> //JackAudioDriver::ProcessGraphAsyncMaster: Process error //
> //JackEngine::XRun: client = alsa_in was not finished, state =
> Triggered //
> //JackEngine::XRun: client = alsa_out was not finished, state =
> Triggered //
> //JackAudioDriver::ProcessGraphAsyncMaster: Process error //
> //... //
> //and CPU became highly loaded //
> /--------------end-----------
When XRUNs start to happen all together you get a domino effect and
nothing will work.
>
> So right now I do not have working solution based on jackd and
> thinking about writing my own alsaclient( with similar functionality.
>
Before moving to this option you may get a look at jacktrip. On my
system is quite stable, a bit CPU hungry but ok, and very easy to use.
The 1.0.5 ships with regular Linux distros, the 1.1.0 solves a few bugs.
In case you'd like to go on the ALSA way please remember that you will
not get synch between ALSA buffers and your application or other
additional application.
Regards
1401873363.12393_0.ltw:2,a <538EE3CD.4060101 at univpm dot it>