On 4/9/07, *Dragan Noveski* &lt;perodog(a)gmx.net <mailto:perodog@gmx.net>> wrote: hi list, i read some threads in the ardour forum about optimizing the jackd build, so i would like to find out, if i can build jackd here with more optimizing ...since i think that more optimizing is better than less optimizing?? the way i am building jack until now is simply: ./configure --with-default-tmpdir=/dev/shm/ add --enable-dynsimd to turn on SSE. Helps a LOT. when i run configure like this, i turns: jack-audio-connection-kit 0.103.0 : | Build with ALSA support............................... : true | Build with FreeBob support............................ : false | Build with OSS support................................ : true | Build with CoreAudio support.......................... : false | Build with PortAudio support.......................... : true | Compiler optimization flags........................... : -DREENTRANT -O3 -fomit-frame-pointer -ffast-math -funroll-loops -march=i686 -mmmx -msse -mfpmath=sse

| Compiler full flags................................... : -I$(top_srcdir)/config -I$(top_srcdir) -I$(top_srcdir) -D_REENTRANT -D_POSIX_PTHREAD_SEMANTICS -Wall -g -g -O2 -I/usr/include/nptl | Install dir for libjack + backends.................... : ${exec_prefix}/lib/jack | | Default driver backend................................ : "alsa" | Shared memory interface............................... : "System V" | Install prefix........................................ : /usr/local the processor type in the kernel config here is 'Pentium M' but i don't know if the '-march=i686' could be changed with something like 'pentium M'. since i am using a self made kernel, only a little help in this issue would help me a lot. cheers, doc _______________________________________________ Linux-audio-user mailing list Linux-audio-user(a)lists.linuxaudio.org <mailto:Linux-audio-user@lists.linuxaudio.org> http://lists.linuxaudio.org/mailman/listinfo.cgi/linux-audio-user -- Mike Taht PostCards From the Bleeding Edge http://the-edge.blogspot.com <http://the-edge.blogspot.com> ------------------------------------------------------------------------ _______________________________________________ Linux-audio-user mailing list Linux-audio-user(a)lists.linuxaudio.org http://lists.linuxaudio.org/mailman/listinfo.cgi/linux-audio-user

does jack say it's running SSE on startup?

On 4/9/07, *Dragan Noveski* &lt;perodog(a)gmx.net <mailto:perodog@gmx.net>> wrote: Mike Taht wrote: <mailto:perodog@gmx.net> <mailto:perodog@gmx.net <mailto:perodog@gmx.net>>> jackd here with than less false -march=i686 -mmmx but for me with this line, sse seems to be turned on already? i just added the option you mentioned, but the configure output is the same. cheers, doc -D_REENTRANT "alsa" "System V" /usr/local but i this issue <mailto:Linux-audio-user@lists.linuxaudio.org> <mailto:Linux-audio-user@lists.linuxaudio.org>> http://lists.linuxaudio.org/mailman/listinfo.cgi/linux-audio-user <http://the-edge.blogspot.com> ------------------------------------------------------------------------ > _______________________________________________ > Linux-audio-user mailing list > Linux-audio-user(a)lists.linuxaudio.org <mailto:Linux-audio-user@lists.linuxaudio.org> > http://lists.linuxaudio.org/mailman/listinfo.cgi/linux-audio-user _______________________________________________ Linux-audio-user mailing list Linux-audio-user(a)lists.linuxaudio.org <mailto:Linux-audio-user@lists.linuxaudio.org> http://lists.linuxaudio.org/mailman/listinfo.cgi/linux-audio-user -- Mike Taht PostCards From the Bleeding Edge http://the-edge.blogspot.com ------------------------------------------------------------------------ _______________________________________________ Linux-audio-user mailing list Linux-audio-user(a)lists.linuxaudio.org http://lists.linuxaudio.org/mailman/listinfo.cgi/linux-audio-user

Hand written assembler is still many orders faster than what gcc is capable of doing. In Ardour peak computation (for both metering and waveform displaying) is written in SSE (the first part in pure assembly, the second in a C-level abstraction which is almost 1:1 assembly). Both functions are more than 20x faster in raw performance than what gcc 4.1 can do.

btw, is there a reason, why ardour is using assembler code instead of compiler intrinsics?

the micro-benchmarks i did for pnpd/nova indicated an additional performance boost around 40%) ...

-- tim(a)klingt.org ICQ: 96771783 http://tim.klingt.org After one look at this planet any visitor from outer space would say "I want to see the manager." William S. Burroughs _______________________________________________ Linux-audio-user mailing list Linux-audio-user(a)lists.linuxaudio.org http://lists.linuxaudio.org/mailman/listinfo.cgi/linux-audio-user

Hi, I am having lots (thousands) of xruns in qjackctl and need some help to address this issue. Any tips on what should I enable/disable, etc. are appreciated. This happens under kernel 2.6.20 either generic or lowlatency. I believe realtime patches are already applied in the lowlatency version, isn't it? After googling a bit I found the appropriate way to have realtime is to add the lines, @audio - rtprio 99 @audio - nice -10 @audio - memlock 500000 to the /etc/security/limits.conf. But this hasn't help at all. I am running Ubuntu Feisty Fawn on a Macbook pro. More info follows below. Thanks in advance, eduard ******** .jackdrc ******** /usr/bin/jackd -R -dalsa -dhw:0 -r44100 -p1024 -n2 -i2 -o2 -zs *************************************************** dmesg | grep "Linux version" outputs the following: *************************************************** [ 0.000000] Linux version 2.6.20-14-lowlatency (root@rothera) (gcc version 4.1.2 (Ubuntu 4.1.2-0ubuntu4)) #2 SMP PREEMPT Mon Apr 2 20:41:03 UTC 2007 (Ubuntu Unofficial) ************* groups output ************* eaylon adm disk dialout cdrom floppy audio dip video plugdev netdev lpadmin powerdev scanner admin ***************************** cat /etc/security/limits.conf ***************************** # /etc/security/limits.conf # #Each line describes a limit for a user in the form: # #<domain> <type> <item> <value> # #Where: #<domain> can be: # - an user name # - a group name, with @group syntax # - the wildcard *, for default entry # - the wildcard %, can be also used with %group syntax, # for maxlogin limit # #<type> can have the two values: # - "soft" for enforcing the soft limits # - "hard" for enforcing hard limits # #<item> can be one of the following: # - core - limits the core file size (KB) # - data - max data size (KB) # - fsize - maximum filesize (KB) # - memlock - max locked-in-memory address space (KB) # - nofile - max number of open files # - rss - max resident set size (KB) # - stack - max stack size (KB) # - cpu - max CPU time (MIN) # - nproc - max number of processes # - as - address space limit # - maxlogins - max number of logins for this user # - maxsyslogins - max number of logins on the system # - priority - the priority to run user process with # - locks - max number of file locks the user can hold # - sigpending - max number of pending signals # - msgqueue - max memory used by POSIX message queues (bytes) # - nice - max nice priority allowed to raise to # - rtprio - max realtime priority # #<domain> <type> <item> <value> # #* soft core 0 #* hard rss 10000 #@student hard nproc 20 #@faculty soft nproc 20 #@faculty hard nproc 50 #ftp hard nproc 0 #@student - maxlogins 4 @audio - rtprio 99 @audio - nice -10 #@audio - memlock 4000000 @audio - memlock 500000 # End of file _______________________________________________ Linux-audio-user mailing list Linux-audio-user(a)lists.linuxaudio.org http://lists.linuxaudio.org/mailman/listinfo.cgi/linux-audio-user

Using 3 (and more) periods does improve it, but I still get a lot of xruns.

Using 3 (and more) periods does improve it, but I still get a lot of xruns. So. Pau and Dimitry, you think the problem comes from the internal soundcard? I do have an external one, but it's RME's Fireface400, which I believe it is not supported by freebob, isn't it? If anyone knows how can I use it under Linux that'll be just wonderful!!

At present the major bottlenecks left for ardour and jack are very much down in the noise floor. A typical user now spends more cpu time in plugins than in those two core programs. (thus I've been oprofiling a few plugins heavily, among other things, now have a SSE'd comb_run routine... and hopefully will announce some sped up plugins soon)