On Wed, 2004-12-29 at 16:21 -0800, Fernando Lopez-Lezcano wrote: This was on the CCRMA list? Got a URL? Lee

Hi guys, I'm off this week so I decided to test some mics (I'll post results later). But in preparation I had to build a shock-mount for my Rode NT-1, shure SM57 and Octava 012. I'm a DIY kinda guy but I never think to photograph the project as it develops... well almost never, this time I did. Enjoy (and please comment as you feel appropriate). http://ruinaudio.com/shockmount/shockmount.html -ry

Nice mod Ryan. I want one for the studio and one for the shower. I won't be happy 'til my shower head is a modified microphone. ron --- Ryan Gallagher <ruinaudio(a)comcast.net> wrote: __________________________________ Do you Yahoo!? Yahoo! Mail - Find what you need with new enhanced search. http://info.mail.yahoo.com/mail_250

On Wed, 29 Dec 2004 21:29:55 -0500, Lee Revell <rlrevell(a)joe-job.com> wrote: Are the LSM patches not available for 2.6.10? I have downloaded the kernel but probably wouldn't run it if I cannot run as a normal user. If they are available can you point me to the appropriate patch for 2.6.10? Thanks, Mark

On Wed, 2004-12-29 at 18:29, Lee Revell wrote: I guess he was referring to the kernel he built (without the realtime-lsm). I'll clarify this point in the Planet CCRMA list... I just booted into 2.6.10 + realtime-lsm, not bad at all in some very informal jack tests (run jack at 2x128 + ardour + hydrogen + freqtweak). I'll leave a full build chugging along all night long, should have fresh packages to test/release tomorrow :-) -- Fernando

Jack O'Quin wrote: Sorry for being stupid, but by build in do you mean build in like "always on" or "can be selected from make xconfig"? Because I don't see anything like that under "security options"... -- peace, love & harmony Atte http://www.atte.dk

Cournapeau David wrote: Ok. I downloaded the patch, copied it to /usr/src/, and ran [atte@aarhus src]$ patch -p0 < linux-2.6.0-rt2.patch patching file linux-2.6.10-rc2-mm3/Documentation/realtime-lsm.txt can't find file to patch at input line 47 Perhaps you used the wrong -p or --strip option? The text leading up to this was: -------------------------- |diff -ruN -X /home/joq/bin/kdiff.exclude linux-2.6.10-rc2-mm3/security/Kconfig linux-2.6.10-rc2-mm3-rt2/security/Kconfig |--- linux-2.6.10-rc2-mm3/security/Kconfig Wed Nov 24 09:35:44 2004 |+++ linux-2.6.10-rc2-mm3-rt2/security/Kconfig Wed Nov 24 09:58:29 2004 -------------------------- File to patch: How do I apply the patch? In fact is there a howto somewhere that explains which patches (and how) to apply to, say, 2.6.10 for realtime audio purposes? -- peace, love & harmony Atte http://www.atte.dk

I read that I need a »mm«-kernel. Is this a vanilla from kernel.org, or do I need special sources? I have already configured and built kernels from time to time, but as you can see I'm still a n00b concerning kernel knowledge ;-) . Thanks for any hint, ce

On Thu, 2004-12-30 at 13:38 +0100, Christoph Eckert wrote: This might not be the case anymore. The vanilla kernel is getting a lot better. As always, YMMV. You still need at least 2.6.10 + realtime LSM patch to get low latency. If the latency is not good enough then you might try -mm + Ingo patches. Lee

/etc/init.d has "local" service which looks for /etc/conf.d/local.[start/stop] (I use ~x86) Andrew ======= On Friday 31 December 2004 00:40, Christoph Eckert wrote: ======= so it gets automatically executed at boottime? In other Distros there's an /etc/init.d/boot.local, but not in Gentoo. Thanks & best regards ce

On Thu, Dec 30, 2004 at 10:40:37PM +0100, Christoph Eckert wrote: On gentoo just put in the file /etc/modules.autoload/kernel-2.6 realtime any=1 allcaps=1 and you will have the module loaded at boot time :) Regards Philippe

Funny, any longer I hadn't put allcaps=1 only alsaplayer whas abble to connect to jackd, when I put this parameters all clients where abble to connect... by this thread, I've been convinced that there is a lot of confusion about how to use the kernel 2.6 for low latency audio, sounds that a clear howto would be welcome for not kernel experts like me ! Regards Philippe

Am Freitag 31 Dezember 2004 00:20 schrieb Dubphil: changed; thanks a lot! Best regards ce

Atte André Jensen &lt;atte(a)ballbreaker.dk&gt; writes: By "built in" I mean configured as "Y" rather than the usual "M". I mentioned that because Nando was having some problems doing that on CCRMA with FC3 due to conflicts with the built in capability LSM. Currently, LSM stacking does not work smoothly. You have to patch first before the security options will appear. To apply the patch... $ cd /usr/src # (or wherever you like) $ tar xjf linux-2.6.10.tar.bz2 # (from kernel.org) $ cd linux-2.6.10 $ patch -p1 < /path/to/patch/file Configure and build the kernel... $ make xconfig $ make bzImage $ make modules Then (as root) install and update LILO or GRUB. The parameters are explained in `Documentation/realtime-lsm.txt'. With the LSM built in, you'll probably want to set parameters via sysfs (as root), using either... # echo 1 > /sys/module/realtime/parameters/any or... # echo 29 > /sys/module/realtime/parameters/gid Pick any group ID you like (29 is the Debian `audio' group). -- joq

"Rui Nuno Capela" &lt;rncbc(a)rncbc.org&gt; writes: Here are my results running vanilla 2.6.10. They support your conclusion, but also the idea that the vanilla kernel is really quite usable. Not sure what system statistics we should collect for this. My system is Debian woody with 2.6.10 and realtime-lsm on an Athlon 1800+ XP with 256MB main memory and M-Audio Delta-66 sound card. I imagine that long 10 msec xrun delay probably occurred during the graph sort after one of the clients disconnected. If so, that's more of a JACK implementation artifact than a kernel or system problem. ************* SUMMARY RESULT **************** Total seconds ran . . . . . . : 300 Number of clients . . . . . . : 20 Ports per client . . . . . . : 4 Frames per buffer . . . . . . : 64 ********************************************* Timeout Count . . . . . . . . :( 1) XRUN Count . . . . . . . . . : 2 Delay Count (>spare time) . . : 0 Delay Count (>1000 usecs) . . : 0 Delay Maximum . . . . . . . . : 10258 usecs Cycle Maximum . . . . . . . . : 825 usecs Average DSP Load. . . . . . . : 32.4 % Average CPU System Load . . . : 7.3 % Average CPU User Load . . . . : 24.1 % Average CPU Nice Load . . . . : 0.0 % Average CPU I/O Wait Load . . : 1.4 % Average CPU IRQ Load . . . . : 0.7 % Average CPU Soft-IRQ Load . . : 0.0 % Average Interrupt Rate . . . : 1689.4 /sec Average Context-Switch Rate . : 11771.0 /sec ********************************************* Very nice test, BTW. I had to hack it a bit to work on my system (mainly due to an old GCC 2.95.4 compiler). I would love to include some version of this as a `make test' option with the JACK distribution. -- joq

On Thu, 2004-12-30 at 16:16 +0000, Rui Nuno Capela wrote: I think this fix went upstream and is in 2.6.10. Also do your tests with Ingo's patch use PREEMPT_DESKTOP or PREEMPT_RT? Lee

Lee Revell writes: Yes, its been CONFIG_PREEMPT_RT all the time. Using CONFIG_PREEMPT_DESKTOP (aka Voluntary-Preempt, IIUC), which I've tested more than a couple of times before RT were booting reliably on both of my machines, the results were something in between vanilla and RT. I have to agree with Fernando when he says that when a RT configured kernel boots (and those IRQ threading gets properly tuned) its The linux-audio-dream come true. It just makes everything else look like a joke ;) Bye now. -- rncbc aka Rui Nuno Capela rncbc(a)rncbc.org

Russell Hanaghan writes: Guess what? I do run my "own" kernels since 2.6 appeared on the radar. Even since Mdk 9.1 IIRC. Never used the Mdk supplied ones, unless during installation/upgrade process, where you have no choice ;). Oh, and this "attitude" of mine also applies to jackd: I do my own jackd builds. Dunno why, but Thacs jackit(-realtime) rpms never ran quite to my satisfaction, whether the underlying kernel is wired for multimedia-realtime-low-latency-whatever or not. Poor me, I guess :) Its true that I had completely trashed my laptop several times in a row, and that surely will happen again, but blessing my DBA education, I always do and keep sane backups ;) Cheers. -- rncbc aka Rui Nuno Capela rncbc(a)rncbc.org

On Thu, 2004-12-30 at 03:19, Rui Nuno Capela wrote: Yes, when it works it is fantastic. Regretfully it does not always work. Several Planet CCRMA users have had problems with a very recent release featuring 0.7.33-04 (ie: machines not making it to the end of the boot process[*]). I've had lockups on my smp test machine, no clue left behind (and when compiled with full debugging it would not even boot). Anyway, it _is_ amazing work! Kudos to all the gurus involved! Now we have both alternatives, a "bleeding edge" kernel and a fairly decent one, almost stock (2.6.10). -- Fernando [*] "Bug: Scheduling while atomic: IRQ14/0x00000001/297" (IRQ14 is the hard disk :-), another one dying with: "gdb_hook_interrupt: Could not clear IER, not a UART!", this one solved through doing: "Hmm, I thought, interrupt handling? Low and behold my BIOS was set with PIC rather than APIC. It now boots fine." Of course these could be just kernel bugs unrelated to, or uncovered by, the realtime preempt patch, maybe fixed in 2.6.10... We'll see today how 2.6.10 fares when I release packages.

On Sat, 2005-01-01 at 18:23 -0500, Lee Revell wrote: Here are my results with "dbench 8" running in the background. ************* SUMMARY RESULT **************** Total seconds ran . . . . . . : 300 Number of clients . . . . . . : 16 Ports per client . . . . . . : 4 ********************************************* Timeout Count . . . . . . . . :( 0) XRUN Count . . . . . . . . . : 8 Delay Count (>spare time) . . : 0 Delay Count (>1000 usecs) . . : 0 Delay Maximum . . . . . . . . : 0 usecs Cycle Maximum . . . . . . . . : 311 usecs Average DSP Load. . . . . . . : 2.6 % Average CPU System Load . . . : 41.4 % Average CPU User Load . . . . : 30.7 % Average CPU Nice Load . . . . : 0.0 % Average CPU I/O Wait Load . . : 2.6 % Average CPU IRQ Load . . . . : 2.8 % Average CPU Soft-IRQ Load . . : 0.0 % Average Interrupt Rate . . . : 2354.8 /sec Average Context-Switch Rate . : 2183.4 /sec ********************************************* Lee

using 2.6.10-rc3-mm with Ingo's latest Real-Time Preempt on a pentium M 1.6ghz laptop with a gig of ram. The test was done as a normal user using the realtime-lsm patch and I have Rui's rtirq script up and running. One of the three tests was completed while compiling gnuplot in the background :). *********** CONSOLIDATED RESULTS ************ Total seconds ran . . . . . . : 1200 Number of clients . . . . . . : 20 Ports per client . . . . . . : 4 Frames per buffer . . . . . . : 64 ********************************************* Summary Result Count . . . . : 3 / 3 ********************************************* Timeout Rate . . . . . . . . :( 0.0)/hour XRUN Rate . . . . . . . . . . : 4.0 /hour Delay Rate (>spare time) . . : 85477.0 /hour Delay Rate (>1000 usecs) . . : 0.0 /hour Delay Maximum . . . . . . . . : 0 usecs Cycle Maximum . . . . . . . . : 0 usecs Average DSP Load. . . . . . . : 34.0 % Average CPU System Load . . . : 25.5 % Average CPU User Load . . . . : 27.7 % Average CPU Nice Load . . . . : 0.0 % Average CPU I/O Wait Load . . : 0.1 % Average CPU IRQ Load . . . . : 0.0 % Average CPU Soft-IRQ Load . . : 0.0 % Average Interrupt Rate . . . : 1881.0 /sec Average Context-Switch Rate . : 42065.8 /sec ********************************************* I ran it 3 times. twice at 300s and once for 600 seconds. However I dont think It actually ran for 5 minutes or 10 minutes. Every time I ran it I got a similar warning/error (maybe I am supposed to get this?) $ ./jack_test3_run.sh 600 Launching jack_test3_nmeter...done. Launching 20 ./jack_test3_client(s) instance(s)...done. Running jackd...1...2...3...4...5...6...7...8...9...10...11...12...13...14...15...16..../jack_test3_run.sh: line 48: 32325 Killed ${CMD} >>${LOG} 2>&1 done. ./jack_test3_run.sh: line 48: 1007 Broken pipe ${CMD} >>${LOG} 2>&1 ./jack_test3_run.sh: line 48: 1389 Broken pipe ${CMD} >>${LOG} 2>&1 ./jack_test3_run.sh: line 48: 1784 Broken pipe ${CMD} >>${LOG} 2>&1 ./jack_test3_run.sh: line 48: 2262 Broken pipe ${CMD} >>${LOG} 2>&1 ./jack_test3_run.sh: line 48: 2419 Broken pipe ${CMD} >>${LOG} 2>&1 ./jack_test3_run.sh: line 48: 2598 Broken pipe ${CMD} >>${LOG} 2>&1 ./jack_test3_run.sh: line 48: 2901 Broken pipe ${CMD} >>${LOG} 2>&1 ./jack_test3_run.sh: line 48: 3226 Broken pipe ${CMD} >>${LOG} 2>&1 ./jack_test3_run.sh: line 48: 3834 Broken pipe ${CMD} >>${LOG} 2>&1 ./jack_test3_run.sh: line 48: 4257 Broken pipe ${CMD} >>${LOG} 2>&1 ./jack_test3_run.sh: line 48: 4348 Broken pipe ${CMD} >>${LOG} 2>&1 ./jack_test3_run.sh: line 48: 4381 Broken pipe ${CMD} >>${LOG} 2>&1 ./jack_test3_run.sh: line 48: 4666 Broken pipe ${CMD} >>${LOG} 2>&1 ./jack_test3_run.sh: line 48: 5115 Broken pipe ${CMD} >>${LOG} 2>&1 Killing jack_test3_nmeter..../jack_test3_run.sh: line 144: 31774 Terminated ( ${NMETER} ${NMETER_ARGS} >>${LOG} 2>&1 ) OK. 17...************* SUMMARY RESULT **************** Total seconds ran . . . . . . : 600 Number of clients . . . . . . : 20 Ports per client . . . . . . : 4 Frames per buffer . . . . . . : 64 ********************************************* Timeout Count . . . . . . . . :( 0) XRUN Count . . . . . . . . . : 2 Delay Count (>spare time) . . : 29224 Delay Count (>1000 usecs) . . : 0 Delay Maximum . . . . . . . . : 0 usecs Cycle Maximum . . . . . . . . : 0 usecs Average DSP Load. . . . . . . : 36.7 % Average CPU System Load . . . : 38.8 % Average CPU User Load . . . . : 61.0 % Average CPU Nice Load . . . . : 0.0 % Average CPU I/O Wait Load . . : 0.2 % Average CPU IRQ Load . . . . : 0.0 % Average CPU Soft-IRQ Load . . : 0.0 % Average Interrupt Rate . . . : 1857.8 /sec Average Context-Switch Rate . : 43960.2 /sec ********************************************* Also my context-switch rate is WAY higher than other folks. Am I missing a tuning step? over all I have been super happy with the latest ingo patch. m. On Thu, 30 Dec 2004 11:19:14 -0000 (WET), Rui Nuno Capela &lt;rncbc(a)rncbc.org&gt; wrote: