3 Jan
2011
3 Jan
'11
3:12 p.m.
On Sat, Jan 01, 2011 at 11:03:11PM +0100, Robin Gareus wrote:
On 12/31/2010 10:03 PM, Philipp Überbacher wrote:
I'd be interested in this as well.
`man 2 mlock` - preventing memory from being paged to the swap area.
that's needed to prevent stalling audio-applications because they're
swapped out.
What's the difference between jack1 and jack2 mlocking? Announcing this
as feature for tschack hints that jack1's implementation is "better" - why?
jack2 only mlocks the port memory.
but it never calls mlockall, which makes sure, that everything is locked
into memory.
Excerpts from torbenh's message of 2010-12-30
16:30:33 +0100:
[snip]
>>>
>>> get it here:
>>> http://hochstrom.endofinternet.org/files/tschack-0.120.1.tar.gz
>>> http://hochstrom.endofinternet.org/files/PyJackd-0.1.0.tar.gz
>>
>> Not sure what half of that means :)
>> One thing I think you forgot to mention but might be of interest is
>> jack-session support
>
> right. jack-session is supported. but thats what jack1 and jack2 also
> support nowadays.
There's support for it in both jack1 and jack2 SVN, but there's no
official release where it is enabled by default.
well....
which of the features is unclear ?
Oh, I thought there is no jack1 release with jack-session yet, dunno
about jack2.
>> features:
>> - jack1 mlocking
Means for a user?
TIA,
robin
>>>> - controlapi which works even when libjackserver.so is loaded
RTLD_LOCAL
>
> Means for a user?
>
>>>> - smp aware
>>>> - backendswitching
>
> Means for a user?
>
>>>> - strictly synchronous like jack1. (-> no latency penalty)
>
> Async (jack2?) has a latency penalty?
>
>>>> - clickless connections.
>>>> - shuts down audio processing when cpu is overloaded for too long.
>
> I guess most users don't know very much about the technical details of
> the jack implementations.
>
> _______________________________________________
> Linux-audio-user mailing list
> Linux-audio-user(a)lists.linuxaudio.org
> http://lists.linuxaudio.org/listinfo/linux-audio-user
--
torben Hohn