12 Jul
2004
12 Jul
'04
1:06 p.m.
Hi,
I have a program with some (p)threads, and a jack ringbuffer. One
thread is writing to the ringbuffer, and another is reading from it.
The question is: is it (thread-)safe to have a _third_ thread that
looks at the ringbuffer via jack_ringbuffer_read_space() at random
times to determine how much data is in the ringbuffer? This third
thread does not actually read nor write any data to/from the
ringbuffer, just wants to know the amount of data in it.
So far I didn't see any crashes/lockups, is it really safe to do this
or just my dirty luck? :)
Thanks,
Tom
12 Jul
12 Jul
4:21 p.m.
I have a program with some (p)threads, and a jack
ringbuffer. One
thread is writing to the ringbuffer, and another is reading from it.
The question is: is it (thread-)safe to have a _third_ thread that
looks at the ringbuffer via jack_ringbuffer_read_space() at random
times to determine how much data is in the ringbuffer? This third
thread does not actually read nor write any data to/from the
ringbuffer, just wants to know the amount of data in it.
it cannot accurately measure the available data or space. so it
depends what you plan to do with the data.
4:22 p.m.
On Mon, 12 Jul 2004 16:52:40 +0200
Tom Szilagyi <tszilagyi(a)users.sourceforge.net> wrote:
Hi,
I have a program with some (p)threads, and a jack ringbuffer. One
thread is writing to the ringbuffer, and another is reading from it.
The question is: is it (thread-)safe to have a _third_ thread that
looks at the ringbuffer via jack_ringbuffer_read_space() at random
times to determine how much data is in the ringbuffer? This third
thread does not actually read nor write any data to/from the
ringbuffer, just wants to know the amount of data in it.
So far I didn't see any crashes/lockups, is it really safe to do this
or just my dirty luck? :)
Thanks,
As long as the third thread really never reads/writes any data from the ringbuffer it
should be ok afaik.. But what is the purpose of that third thread? Do maybe the results of
this third thread have side effects on the other two threads? Depending on what you do
with the fill information of the ringbuffer you might get unexpected results..
Flo
--
Palimm Palimm!
http://affenbande.org/~tapas/wiki
6:12 p.m.
On Mon, Jul 12, 2004 at 08:20:08PM +0200, Florian Schmidt wrote:
On Mon, 12 Jul 2004 16:52:40 +0200
Tom Szilagyi <tszilagyi(a)users.sourceforge.net> wrote:
Not really important, it's not even audio-related. (Just some private
play-around to learn new things...) I was just using the jack
ringbuffer interface, as it is a really appealing way of doing data
transfer between threads.
As Paul pointed out, the measurement is not very accurate, but that's
not a problem. I just need to make sure it never crashes/locks up.
So thanks for the answers,
Tom
[...]
I have a program with some (p)threads, and a jack ringbuffer. One
thread is writing to the ringbuffer, and another is reading from it.
The question is: is it (thread-)safe to have a _third_ thread that
looks at the ringbuffer via jack_ringbuffer_read_space() at random
times to determine how much data is in the ringbuffer? This third
thread does not actually read nor write any data to/from the
ringbuffer, just wants to know the amount of data in it.
[...]
As long as the third thread really never reads/writes any data from
the ringbuffer it should be ok afaik.. But what is the purpose of
that third thread? Do maybe the results of this third thread have
side effects on the other two threads? Depending on what you do with
the fill information of the ringbuffer you might get unexpected
results..
7533
days inactive
7533
days old
linux-audio-dev@lists.linuxaudio.org
3 comments
3 participants
participants (3)
-
Florian Schmidt -
Paul Davis -
Tom Szilagyi