Subject: Re: [ecasound] JACK plugin added to ecasound
From: S. Massy (theanaloguekid_AT_tak.net.dhis.org)
Date: Sun Dec 16 2001 - 09:03:14 EET
On Sun, 16 Dec 2001, Kai Vehmanen <k_AT_eca.cx> wrote:
> Committed to CVS. Experimental but works.
Could you please give a brief overview of what JACK is for those of us
not following LAD? Even a google search didn't turn up much... :(
>
> ---
> Setup:
>
> To use it, update ecasound tree from CVS, and give --with-jack=/path/to
> option to configure. I've installed jack in ~/apps/jack so I use
> --with-jack=/home/user/apps/jack; with default jack install you'd use
> --with-jack=/usr/local.
>
> ---
> And actual use:
>
> ./jackd -p 128 -r 44100 &
> ecasound -b:128 -i foo.wav -o jack,ecaoutput1
> ecasound -b:128 -i jack,ecarec -o foo.wav
>
> ---
> Known limitations:
>
> - you must manually set buffersize to match jackd's buffersize;
> ie. 'jackd -p x ...' --> 'ecasound -b:x ...'
> - you must manually set sample rate
> ie. 'jackd -r 48000 ...' --> 'ecasound -sr:48000 ...'
> - don't use multiple JACK plugins in the same chainsetup (hangs
> in UNIX-socket poll() somewhere in JACK client code)
> - only mono output and input supported; depending on direction,
> ecasound makes a connection to either 'ALSA I/O:Input 1' or
> 'ALSA I/O:Output 1'
>
> ---
> Performance
>
> Playback and recording works without xruns even with '-p 64', but is very
> sensitive to outside events. I'm not running a ll-patched kernel so this
> is to be expected. Also, the ecasound JACK plugin has lots of room for
> optimization. As it is now, two pthread_cond+pthread_signal combinations
> are issued per period.
>
> There are lots of misc problems. When something goes wrong you usually
> have to "killall -9 jackd ; rm -f /tmp/jack* ; ./jackd".
>
> ---
> Patches&stuff
>
> I'm using ALSA0.9 ens1371 driver. With the current JACK CVS-tree I
> couldn't run jackd with '-r 44100'. ens1371 (unlike ens1370) has a
> continuous range of supported sample rates. This also means that if you
> request for 44100 you don't get that exact rate. Or at least this is my
> interpretation. In any case I get an error from
> 'snd_pcm_hw_param_set_rate()'. I got around this with the following
> change:
>
> --cut--
> diff -u -r1.9 alsa_driver.c
> --- alsa_driver.c 2001/12/08 18:11:38 1.9
> +++ alsa_driver.c 2001/12/16 01:18:09
> @@ -219,7 +219,7 @@
> }
> }
>
> - if ((err = snd_pcm_hw_params_set_rate (handle, hw_params, driver->frame_rate, 0)) < 0) {
> + if ((err = snd_pcm_hw_params_set_rate_near (handle, hw_params, driver->frame_rate, 0)) < 0) {
> --cut--
>
> And another very, very important patch :)...
>
> --cut--
> --- AUTHORS 2001/11/12 04:08:38 1.1.1.1
> +++ AUTHORS 2001/12/16 01:18:09
> @@ -5,7 +5,7 @@
> Paul Davis
> David Olofson
> Benno Sennoner
> - Kai Vehamen
> + Kai Vehmanen
> --cut--
>
> --
> http://www.eca.cx
> Audio software for Linux!
>
> --
> To unsubscribe send message 'unsubscribe' in the body of the
> message to <ecasound-list-request_AT_wakkanet.fi>.
>
>
-- To unsubscribe send message 'unsubscribe' in the body of the message to <ecasound-list-request_AT_wakkanet.fi>.
This archive was generated by hypermail 2b28 : Sun Dec 16 2001 - 08:55:28 EET