#linuxcnc-devel | Logs for 2014-08-14

Back
[00:00:34] -!- Nick001-shop has quit [Quit: ChatZilla 0.9.90.1 [Firefox 30.0/20140605174243]]
[00:03:27] -!- anth0ny has quit [Quit: anth0ny]
[00:04:32] <CaptHindsight> all results are in
[00:04:57] <andypugh> The judges seem inconsistent :-)
[00:07:17] -!- andypugh has quit [Quit: andypugh]
[00:07:46] -!- sudobangbang has quit [Ping timeout: 250 seconds]
[00:16:46] -!- theorbtwo has quit [Read error: No route to host]
[00:27:50] -!- PetefromTn_ has quit [Quit: I'm Outta here!!]
[00:33:17] -!- The_Ball has quit [Remote host closed the connection]
[00:43:02] -!- knownasilya has quit [Quit: Connection closed for inactivity]
[00:44:50] -!- sylphiae has quit [Read error: Connection reset by peer]
[01:02:14] -!- sirdancealot has quit [Ping timeout: 272 seconds]
[01:05:37] -!- syyl_ has quit [Ping timeout: 245 seconds]
[01:06:48] -!- anth0ny has quit [Quit: anth0ny]
[01:35:15] -!- JT-Shop has quit [Remote host closed the connection]
[01:41:44] -!- SquirrelCZECH has quit [Quit: No Ping reply in 180 seconds.]
[01:41:56] -!- tjb1 has quit [Read error: Connection reset by peer]
[01:48:52] <jepler> seb_kuzminsky: thank you!
[01:50:06] <jepler> CaptHindsight: with all the numbers in, I'm back to thinking the best theory is: isolcpus numbers OR rtai cpuid numbers are off by one
[01:50:25] <jepler> CaptHindsight: I am confident that rtai kern/latency intends to run on CPU 1, and linuxcnc on CPU 3
[01:50:36] <jepler> CaptHindsight: but the best results for kern/latency come when CPU 0 is isolated
[01:50:46] <jepler> and the best results for linuxcnc latency-test come when CPU 2 is isolated
[01:55:04] <jepler> this is an interesting development: http://www.parallax.com/microcontrollers/propeller-1-open-source
[01:55:11] <jepler> > The Propeller 1 (P8X32A) is now a 100% open multicore microcontroller, including all of the hardware and tools: Verilog code, Spin interpreter, PropellerIDE and SimpleIDE programming tools, and compilers. The Propeller 1 may be the most open chip in its class.
[01:56:50] -!- tjb1 has quit [Quit: Leaving]
[01:58:23] -!- Tecan has quit [Changing host]
[02:00:18] <cradek> neato!
[02:01:42] -!- automata__ has quit [Ping timeout: 260 seconds]
[02:09:17] -!- SquirrelCZECH has quit [Quit: No Ping reply in 180 seconds.]
[02:19:29] -!- Benjamin23 has quit [Ping timeout: 255 seconds]
[02:22:55] -!- jepler has quit [Remote host closed the connection]
[02:36:02] <skunkworks> I had a mdadm array I made a few years ago with 2tb drives (6) before I knew / or fdisk knew about the 'advanced formated' drives. They where all miss-aligned. I have been removing each drive and rebuilding the array with correctly re-partitioned drives. Each time the rebuild speed increases. It started at about 20mb/sec and it is up to 42mb/sec and I still have 2 drives to go.
[02:36:16] -!- chopper791 has quit [Ping timeout: 255 seconds]
[02:36:47] <skunkworks> really wasn't a big deal as it was offsite backup. so it was still faster than the internet connection.
[02:42:10] -!- tmcw has quit [Client Quit]
[02:42:36] -!- terabyte- has quit [Quit: terabyte-]
[02:45:49] -!- jepler [jepler!~jepler@emc/developer/pdpc.professional.jepler] has joined #linuxcnc-devel
[02:46:02] <jepler> whoops, pressed C-A-D on the wrong computer
[02:50:52] <jepler> anyway, with uspace on a 2-core system, I didn't find it made much difference to specify isolcpus=. specifying idle=poll seemed to help latency a bit. but I did not run any extended tests / stressing while testing
[02:55:11] <skunkworks> I haven't used isolcpus in a while.. I should try it.. But I do find better latency with idle=poll.. like it can help from 30us latency down to 10us
[02:56:20] -!- zeeshan|2 has quit [Ping timeout: 250 seconds]
[02:58:16] -!- amiri_ has quit [Quit: Lost terminal]
[02:58:16] -!- amiri has quit [Quit: Lost terminal]
[03:05:34] -!- Jeebiss has quit [Ping timeout: 240 seconds]
[03:06:12] -!- Thetawaves has quit [Quit: This computer has gone to sleep]
[03:07:03] -!- heyman has quit [Ping timeout: 240 seconds]
[03:07:28] -!- sumpfralle has quit [Ping timeout: 244 seconds]
[03:11:37] -!- asdfasd has quit [Ping timeout: 272 seconds]
[03:18:47] -!- zzolo has quit [Quit: zzolo]
[03:18:53] -!- chally_ has quit [Read error: Connection reset by peer]
[03:19:01] -!- the_wench has quit [Ping timeout: 255 seconds]
[03:21:35] -!- archivist has quit [Ping timeout: 255 seconds]
[03:36:51] -!- archivist [archivist!~archivist@host81-149-189-98.in-addr.btopenworld.com] has joined #linuxcnc-devel
[03:43:39] -!- larryone has quit [Quit: This computer has gone to sleep]
[03:50:09] -!- jepler_ [jepler_!~jepler@emc/developer/pdpc.professional.jepler] has joined #linuxcnc-devel
[03:50:10] -!- Benjamin23 has quit [Remote host closed the connection]
[03:50:42] -!- FreezingAlt has quit [Ping timeout: 272 seconds]
[03:52:45] -!- bluemaex has quit [Ping timeout: 250 seconds]
[03:52:46] -!- SkramX___ has quit [Ping timeout: 250 seconds]
[03:52:49] -!- chopper791 has quit [Ping timeout: 250 seconds]
[03:52:52] -!- MarkusBec has quit [Ping timeout: 250 seconds]
[03:52:52] -!- WyrM has quit [Ping timeout: 250 seconds]
[03:53:04] -!- jerryitt has quit [Ping timeout: 250 seconds]
[03:53:04] -!- jepler has quit [Ping timeout: 250 seconds]
[03:53:04] -!- automata_ has quit [Ping timeout: 250 seconds]
[03:53:10] -!- MrSunshine has quit [Ping timeout: 250 seconds]
[03:53:15] -!- phragment has quit [Ping timeout: 250 seconds]
[03:53:17] -!- steves_logging has quit [Ping timeout: 250 seconds]
[03:53:19] -!- uwe_mobile__ has quit [Ping timeout: 250 seconds]
[03:53:27] _WyrM is now known as WyrM
[03:53:31] -!- MarkusBec_ [MarkusBec_!znc@80.69.39.131] has joined #linuxcnc-devel
[03:53:33] MarkusBec_ is now known as MarkusBec
[03:54:10] jerryitt_ is now known as jerryitt
[03:54:15] -!- the_wench [the_wench!~the_wench@host81-149-189-98.in-addr.btopenworld.com] has joined #linuxcnc-devel
[04:03:24] -!- terabyte- has quit [Quit: terabyte-]
[04:10:09] -!- larryone has quit [Quit: This computer has gone to sleep]
[04:15:30] -!- Thetawaves has quit [Quit: This computer has gone to sleep]
[04:20:19] -!- asah has quit [Quit: asah]
[04:21:06] <CaptHindsight> I can try a 2 core APU on the same board later today/tomorrow
[04:21:21] -!- SquirrelCZECH has quit [Quit: No Ping reply in 180 seconds.]
[04:26:08] -!- automata__ has quit [Ping timeout: 250 seconds]
[04:32:51] -!- anth0ny has quit [Quit: anth0ny]
[04:35:24] -!- toner has quit [Ping timeout: 255 seconds]
[04:37:25] -!- FinboySlick has quit [Quit: Leaving.]
[04:39:14] -!- terabyte- has quit [Quit: terabyte-]
[04:41:33] -!- kwallace2 [kwallace2!~kwallace@smb-10.sonnet.com] has parted #linuxcnc-devel
[04:55:33] -!- asah has quit [Ping timeout: 240 seconds]
[05:03:24] -!- Fox_Muldr has quit [Ping timeout: 250 seconds]
[05:04:24] -!- phantomD has quit [Read error: Connection reset by peer]
[05:10:32] SpeedEvil is now known as Guest25457
[05:10:57] -!- t12 has quit [Write error: Broken pipe]
[05:10:58] -!- Teme has quit [Write error: Broken pipe]
[05:10:58] -!- BlaDe^ has quit [Write error: Broken pipe]
[05:10:58] -!- radish has quit [Write error: Connection reset by peer]
[05:10:59] -!- Komzzpa has quit [Write error: Broken pipe]
[05:11:00] -!- SquirrelCZECH has quit [Write error: Broken pipe]
[05:11:08] -!- Guest25457 has quit [Remote host closed the connection]
[05:11:50] radish_ is now known as radish
[05:19:33] -!- Tecan has quit [Ping timeout: 240 seconds]
[05:24:24] -!- sudobangbang has quit [Quit: ChatZilla 0.9.90.1 [Firefox 31.0/20140716183446]]
[05:26:08] phantom is now known as phantoxeD
[05:33:15] -!- Cylly has quit [Ping timeout: 246 seconds]
[05:53:02] -!- knownasilya has quit [Quit: Connection closed for inactivity]
[05:58:10] -!- anth0ny has quit [Quit: anth0ny]
[06:07:24] -!- amiri has quit [Ping timeout: 260 seconds]
[06:17:20] -!- jthornton has quit [Read error: Connection reset by peer]
[06:18:24] -!- jthornton [jthornton!~john@172.243.165.194] has joined #linuxcnc-devel
[06:42:25] -!- rob_h [rob_h!~robh@2.124.121.72] has joined #linuxcnc-devel
[06:51:52] -!- SquirrelCZECH_ has quit [Quit: No Ping reply in 180 seconds.]
[06:59:52] -!- The_Ball has quit [Remote host closed the connection]
[07:02:46] -!- jerryitt has quit [Quit: Connection closed for inactivity]
[07:47:58] -!- The_Ball has quit [Ping timeout: 250 seconds]
[08:01:01] -!- b_b has quit [Changing host]
[08:14:43] -!- balestrino has quit [Remote host closed the connection]
[08:18:03] -!- PetefromTn_andro has quit [Ping timeout: 240 seconds]
[08:23:39] -!- balestrino has quit [Ping timeout: 244 seconds]
[08:35:33] -!- quiqua has quit [Ping timeout: 240 seconds]
[08:39:32] -!- quiqua_ has quit [Ping timeout: 260 seconds]
[08:55:29] BitEvil is now known as SpeedEvil
[09:01:33] -!- PetefromTn_andro has quit [Ping timeout: 240 seconds]
[09:37:28] -!- dan2k3k4 has quit [Ping timeout: 255 seconds]
[09:55:44] -!- tinkerer [tinkerer!~tinkerer@mail.play-pla.net] has joined #linuxcnc-devel
[10:06:14] -!- ashcan_ has quit [Client Quit]
[10:10:37] -!- sirdancealot has quit [Ping timeout: 245 seconds]
[10:32:00] -!- skunkworks has quit [Ping timeout: 260 seconds]
[10:33:57] -!- PetefromTn_andro has quit [Quit: Bye]
[10:43:31] -!- Thetawaves has quit [Quit: This computer has gone to sleep]
[10:56:47] Teme2 is now known as teme
[10:57:12] -!- sumpfralle has quit [Ping timeout: 260 seconds]
[11:00:36] -!- skunkworks [skunkworks!~skunkwork@68-115-41-210.static.eucl.wi.charter.com] has joined #linuxcnc-devel
[11:18:39] -!- sumpfralle has quit [Ping timeout: 272 seconds]
[11:22:10] -!- syyl_ has quit [Ping timeout: 260 seconds]
[11:25:42] -!- skunkworks_ [skunkworks_!~chatzilla@68-115-41-210.static.eucl.wi.charter.com] has joined #linuxcnc-devel
[11:26:25] -!- MC500 has quit [Ping timeout: 246 seconds]
[11:30:06] teme is now known as Teme
[11:34:00] -!- Valen has quit [Quit: Leaving.]
[11:59:33] <jepler_> two more tests occurred to me: (A) you can of course modify linuxcnc (or, with a bit more difficulty, rtai kernl/latency) to bind to a different CPUID
[12:00:33] <jepler_> (B) you can write a hal component that hogs about half the time (i.e., if period is 1ms then it busy-waits 500ms before returning), then (using taskset to run a compute-bound program on each core) you can see what linux cpu number an rtai cpu number corresponds to
[12:07:33] -!- skunkworks_ has quit [Ping timeout: 240 seconds]
[12:12:11] -!- Klaus-Krefeld has quit [Remote host closed the connection]
[12:14:07] -!- theorbtwo has quit [Remote host closed the connection]
[12:18:11] -!- terabyte- has quit [Quit: terabyte-]
[12:21:20] -!- larryone has quit [Ping timeout: 244 seconds]
[12:34:16] -!- jthornton has quit [Ping timeout: 260 seconds]
[12:35:56] -!- skunkworks_ [skunkworks_!~skunkwork@68-115-41-210.static.eucl.wi.charter.com] has joined #linuxcnc-devel
[12:37:08] -!- skunkworks__ [skunkworks__!~chatzilla@68-115-41-210.static.eucl.wi.charter.com] has joined #linuxcnc-devel
[12:38:50] -!- skunkworks has quit [Ping timeout: 250 seconds]
[12:39:02] skunkworks__ is now known as skunkworks
[12:42:35] -!- skunkworks__ [skunkworks__!~skunkwork@68-115-41-210.static.eucl.wi.charter.com] has joined #linuxcnc-devel
[12:45:04] <archivist> could our process measure its start time and then delay for diff to reduce peak variation
[12:45:14] -!- skunkworks_ has quit [Ping timeout: 260 seconds]
[12:51:08] -!- skunkworks__ has quit [Read error: Connection reset by peer]
[12:51:08] -!- skunkworks has quit [Read error: Connection reset by peer]
[12:52:07] -!- skunkworks [skunkworks!~skunkwork@68-115-41-210.static.eucl.wi.charter.com] has joined #linuxcnc-devel
[12:52:33] -!- syyl has quit [Ping timeout: 240 seconds]
[12:56:01] -!- likevinyl has quit [Quit: likevinyl]
[13:00:12] -!- skunkworks_ [skunkworks_!~skunkwork@68-115-41-210.static.eucl.wi.charter.com] has joined #linuxcnc-devel
[13:01:39] <jepler_> I performed test (B), again on my qemu -smp 4 machine
[13:01:49] <jepler_> component "busywait" busy-waits for a half a period
[13:02:15] <jepler_> then I ran a userspace program that takes about 2.5s normally on each cpu: for c in 0 1 2 3; do echo; echo $c; taskset -c $c time python shamble.py; done
[13:02:32] <jepler_> for c=3 the elapsed time was doubled
[13:03:00] -!- skunkworks has quit [Ping timeout: 255 seconds]
[13:03:18] <jepler_> so it feels like the rtai cpuid numbering and the userspace cpuid numbering match
[13:05:05] <jepler_> http://pastie.org/9473257 http://pastie.org/9473258
[13:05:22] -!- terabyte- has quit [Quit: terabyte-]
[13:09:55] -!- skunkworks [skunkworks!~chatzilla@68-115-41-210.static.eucl.wi.charter.com] has joined #linuxcnc-devel
[13:12:02] <jepler_> taskset shows that a fair number of processes are still allowed to schedule on core 3. For example, kpsmoused's affinity list is 0-3.
[13:18:16] -!- _balestrino has quit [Ping timeout: 250 seconds]
[13:25:02] <CaptHindsight> I'm considering setting up a hardware latency test rig to measure actual IO latency
[13:34:08] -!- kwallace [kwallace!~kwallace@smb-36.sonnet.com] has joined #linuxcnc-devel
[13:34:10] -!- Benjamin23 has quit [Remote host closed the connection]
[13:46:58] <CaptHindsight> I'll try comparing idle=poll as well later
[13:47:34] -!- theorbtwo has quit [Remote host closed the connection]
[13:52:13] <CaptHindsight> how much tuning is worth the effort to reduce latency if it doesn't impact actual IO?
[13:52:21] <cradek> none
[13:53:23] -!- jenia has quit [Read error: Connection reset by peer]
[13:54:27] <CaptHindsight> I wonder how close we are
[13:55:42] <CaptHindsight> on a 4 core cpu I don't see why RTAI should be sharing a core with any other processes
[14:04:11] -!- SquirrelCZECH has quit [Remote host closed the connection]
[14:07:34] -!- theorbtwo has quit [Remote host closed the connection]
[14:14:12] -!- lair82_ [lair82_!616b5c34@gateway/web/freenode/ip.97.107.92.52] has joined #linuxcnc-devel
[14:21:49] SkramX____ is now known as SkramX
[14:25:15] <jepler_> CaptHindsight: I think it would be interesting to modify linuxcnc so that it binds rtai cpuid 2 and test on your system with isolcpus=2 and isolcpus=3
[14:26:19] <jepler_> CaptHindsight: I can provide a patch to do that, or you can just find the call to rt_task_init_cpuid and change the arugment from rt_cpu to 2
[14:28:03] <jepler_> because the other theory that fits the observations nearly as well as "off by one" is that it's helpful to isolate the other core that shares the same module
[14:28:29] <jepler_> since those two cores share L1i cache and decode resources
[14:28:58] <jepler_> this theory would be confirmed if linuxcnc on cpuid 2 performs better with isolcpus=3 than with isolcpus=2
[14:29:37] <jepler_> and the off-by-one theory would be confirmed if it performs better with isolcpus=1 than with isolcpus=2
[14:31:46] <CaptHindsight> jepler_: I'll be happy to test any patch, if it's not lots of work for you.
[14:32:09] <jepler_> CaptHindsight: I was hoping my prose description was enough .. hold on
[14:33:04] <jepler_> http://pastie.org/9473416 -- obviously not for pushing
[14:33:37] -!- balestrino has quit [Ping timeout: 245 seconds]
[14:34:52] <CaptHindsight> the test system has a drive with 5-6 Linuxcnc installs on different distros
[14:35:53] <CaptHindsight> no tool chains or dev tools on them except for what the install might have added
[14:39:08] <KGB-linuxcnc> 03Jeff Epler 05jepler/hardcode-cpuid-for-ch ad8d425 06linuxcnc 10src/rtapi/rtai_rtapi.c Hard-code cpuid (CPU#) for rtai realtime * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=ad8d425
[14:39:26] <jepler_> CaptHindsight: eventually, buildbot will build this branch and provide debs you can install
[14:40:08] <jepler_> the deb'll go in here: http://buildbot.linuxcnc.org/dists/wheezy/scratch-rt/binary-i386/
[14:41:10] <jepler_> seb_kuzminsky: I just noticed that the buildbot front page doesn't list wheezy scratch-rt as an option
[14:42:03] <jepler_> seb_kuzminsky: also, it doesn't mention wheezy-rt for 2.6, which should probably be moved up above 2.5 and marked as "stable", while 2.5 should be downgraded to "old release"
[14:42:07] <jepler_> nit: picked
[14:50:41] <cradek> we sure need some web stuff updated
[14:51:15] <cradek> andy has some thoughts too (and of course he's right)
[14:52:15] -!- dr0w has quit [Quit: Leaving]
[14:56:48] -!- kwallace2 [kwallace2!~kwallace@smb-114.sonnet.com] has joined #linuxcnc-devel
[14:58:52] -!- kwallace has quit [Ping timeout: 272 seconds]
[15:01:02] -!- mle has quit [Ping timeout: 245 seconds]
[15:03:29] -!- syyl_ has quit [Ping timeout: 272 seconds]
[15:09:39] -!- txp has quit [Remote host closed the connection]
[15:10:32] <linuxcnc-build> build #122 of 1401.rip-wheezy-rtai-i386 is complete: Failure [4failed compile runtests] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1401.rip-wheezy-rtai-i386/builds/122 blamelist: Jeff Epler <jepler@unpythonic.net>
[15:12:36] -!- dan2k3k4 has quit [Ping timeout: 255 seconds]
[15:13:43] -!- Simooon has quit [Quit: Leaving]
[15:16:46] asheppard is now known as sheppard
[15:17:14] <linuxcnc-build> build #2315 of 0000.checkin is complete: Failure [4failed] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/2315 blamelist: Jeff Epler <jepler@unpythonic.net>
[15:17:42] <seb_kuzminsky> i've got a bad feeling about this
[15:18:30] <seb_kuzminsky> not sure why that bonked
[15:18:57] <seb_kuzminsky> i was fearing a kernel panic, but the machine is up and seems fine
[15:19:04] <cradek> if you bind to cpu 2 but you don't have a cpu 2, what happens?
[15:19:20] <seb_kuzminsky> fuck if i know
[15:19:40] <seb_kuzminsky> that may be the only rtai buildslave with 2 cpus, i bet all the others have 1
[15:20:02] <cradek> I think 2 would be the third cpu
[15:20:59] <seb_kuzminsky> yeah, my lucid & precise rtai buildslaves have 1 cpu, wheezy-rtai has 2
[15:21:39] <seb_kuzminsky> logger[psha]:
[15:21:48] <cradek> http://git.linuxcnc.org/gitweb?p=linuxcnc.git;a=commitdiff;h=ad8d4255d6bd543c9cff69d345d8865b8b72b911
[15:22:36] <cradek> so it might just be broken for <3 cpus
[15:22:40] <seb_kuzminsky> jepler_: i've made a note to update the buildbot links, may get to it tonight
[15:23:01] <seb_kuzminsky> cradek: but work on UP
[15:23:06] <seb_kuzminsky> yeah, could be
[15:23:08] <seb_kuzminsky> bbl
[15:23:47] amnesic_away is now known as amnesic
[15:25:51] -!- steves_logging [steves_logging!~Steve@wsip-70-168-134-18.dc.dc.cox.net] has joined #linuxcnc-devel
[15:26:58] -!- automata__ has quit [Ping timeout: 264 seconds]
[15:28:40] -!- andreas_ has quit [Remote host closed the connection]
[15:35:01] -!- tinkerer1 [tinkerer1!~tinkerer@mail.play-pla.net] has joined #linuxcnc-devel
[15:38:43] -!- tinkerer has quit [Ping timeout: 240 seconds]
[15:40:29] -!- liedman has quit [Ping timeout: 245 seconds]
[15:43:51] <seb_kuzminsky> the buildslave had a bunch of rtai kernel modules loaded after it terminated the test, i rebooted it
[15:44:08] <seb_kuzminsky> wheezy-rtai
[15:47:33] -!- Klaus-Krefeld has quit [Remote host closed the connection]
[15:54:50] -!- Deejay has quit [Quit: bye]
[16:00:18] -!- zzolo has quit [Quit: zzolo]
[16:01:04] -!- terabyte- has quit [Quit: terabyte-]
[16:08:13] <jepler_> argh, well, I guess buildbot's not going to give a package for that
[16:08:24] <jepler_> CaptHindsight: so you'll have to build linuxcnc yourself if you want to test
[16:08:58] <jepler_> seb_kuzminsky: sorry to cause the trouble
[16:09:23] <jepler_> or I could try writing rt_data->cpu > 2 ? 2 : rt_data->cpu
[16:10:36] -!- cncbasher [cncbasher!~kvirc@cpc8-hart9-2-0-cust254.11-3.cable.virginm.net] has joined #linuxcnc-devel
[16:13:43] <CaptHindsight> jepler_: ok, testing with idle=poll right now
[16:26:52] dr0w_ is now known as dr0w
[16:28:58] -!- lair82_ has quit [Quit: Page closed]
[16:29:00] -!- lair82 has quit [Quit: Page closed]
[16:36:40] -!- PetefromTn_ has quit [Quit: I'm Outta here!!]
[16:41:23] -!- md-2 has quit [Remote host closed the connection]
[16:43:33] -!- jduhls has quit [Ping timeout: 240 seconds]
[16:47:05] <CaptHindsight> idle=poll results http://wiki.linuxcnc.org/cgi-bin/wiki.pl?The_Isolcpus_Boot_Parameter_And_GRUB2#Latency_Testing
[16:52:30] <CaptHindsight> skunkworks: have you used any 4 or more cpus with Linuxcnc and tested isolcpus vs idle=poll?
[16:52:57] -!- Benjamin23 has quit [Ping timeout: 246 seconds]
[16:54:09] -!- cncbasher has quit [Quit: KVIrc 4.1.3 Equilibrium http://www.kvirc.net/]
[16:54:27] <KGB-linuxcnc> 03Jeff Epler 05jepler/hardcode-cpuid-for-ch e65eaf1 06linuxcnc 10src/rtapi/rtai_rtapi.c Hard-code cpuid (CPU#) for rtai realtime * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=e65eaf1
[16:55:03] <jepler_> I haven't ever had a machine beefy enough to have 4 cores running rtai
[16:55:10] -!- rob_h has quit [Ping timeout: 264 seconds]
[16:55:17] <jepler_> nor even 4 threads
[16:58:57] <CaptHindsight> heh, I meant 4 or more cpu cores, it's been type day for me :o
[17:00:02] -!- i_tarzan has quit [Ping timeout: 260 seconds]
[17:01:54] <linuxcnc-build> build #1511 of 1301.rip-precise-rtai-i386 is complete: Failure [4failed compile] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1301.rip-precise-rtai-i386/builds/1511 blamelist: dummy, Jeff Epler <jepler@unpythonic.net>
[17:02:11] <linuxcnc-build> build #2310 of 1201.rip-lucid-rtai-i386 is complete: Failure [4failed compile] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1201.rip-lucid-rtai-i386/builds/2310 blamelist: dummy, Jeff Epler <jepler@unpythonic.net>
[17:03:51] <linuxcnc-build> build #123 of 1401.rip-wheezy-rtai-i386 is complete: Failure [4failed compile] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1401.rip-wheezy-rtai-i386/builds/123 blamelist: dummy, Jeff Epler <jepler@unpythonic.net>
[17:04:38] <linuxcnc-build> build #2309 of 1900.clang-lucid-rtai-i386 is complete: Failure [4failed compile] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1900.clang-lucid-rtai-i386/builds/2309 blamelist: dummy, Jeff Epler <jepler@unpythonic.net>
[17:06:30] FreezingAlt is now known as FreezingCold
[17:08:42] -!- larryone1 has quit [Ping timeout: 272 seconds]
[17:26:07] <linuxcnc-build> build #2316 of 0000.checkin is complete: Failure [4failed] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/2316 blamelist: dummy, Jeff Epler <jepler@unpythonic.net>
[17:27:32] -!- automata_ has quit [Client Quit]
[17:55:46] <jepler_> I fail
[17:55:53] jepler_ is now known as jepler
[17:56:42] -!- gonzo__ has quit [Quit: Leaving]
[17:59:31] amnesic is now known as amnesic_away
[17:59:59] -!- prtksxna has quit [Excess Flood]
[18:02:08] <seb_kuzminsky> your branch is bad and you should feel bad
[18:06:03] -!- logger[psha] has quit [Ping timeout: 240 seconds]
[18:07:18] -!- logger[psha] [logger[psha]!~loggerpsh@195.135.238.205] has joined #linuxcnc-devel
[18:08:33] <seb_kuzminsky> jepler's branch is so broken it made logger[psha] quit
[18:10:54] -!- brianmorel99 [brianmorel99!~brianmore@71-13-81-38.static.bycy.mi.charter.com] has joined #linuxcnc-devel
[18:11:12] -!- `Nerobro has quit [Ping timeout: 250 seconds]
[18:11:22] -!- i_tarzan has quit [Ping timeout: 264 seconds]
[18:11:47] <cradek> jepler: I bet you can easily get it in 3 tries
[18:11:59] <cradek> that's still pretty good.
[18:12:57] -!- md-2 has quit [Remote host closed the connection]
[18:17:42] -!- md-2 has quit [Ping timeout: 250 seconds]
[18:34:54] -!- IchGuckLive has quit [Quit: ChatZilla 0.9.87 [Firefox 20.0/20130329043827]]
[18:38:56] -!- gambakufu has quit []
[18:46:19] -!- kfoltman has quit [Quit: Ex-Chat]
[18:47:09] -!- gambakufu has quit []
[18:47:53] <cradek> seb_kuzminsky: did you update linux-kbuild and linux-tools?
[18:53:18] -!- rob_h [rob_h!~robh@2.124.121.72] has joined #linuxcnc-devel
[18:57:37] -!- zeeshan has quit [Ping timeout: 244 seconds]
[19:09:33] <KGB-linuxcnc> 03Jeff Epler 05jepler/hardcode-cpuid-for-ch c4fcac6 06linuxcnc 10src/rtapi/rtai_rtapi.c Hard-code cpuid (CPU#) for rtai realtime * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=c4fcac6
[19:10:24] -!- zeitue has quit [Quit: Leaving]
[19:11:50] <seb_kuzminsky> cradek: no, i just touched them in a vain attempt to get apt-ftparchive to notice that they changed
[19:12:22] <seb_kuzminsky> what finally did the trick was adding APT::FTPArchive::AlwaysStat=True, and regenerating the Packages files
[19:13:04] -!- Flipp has quit [Remote host closed the connection]
[19:13:22] <seb_kuzminsky> i'm gonna push a new rtai-modules with the rtai_16550a driver included
[19:13:40] <seb_kuzminsky> it should be totally backwards compatible with the current rtai-modules, so no one should notice or care
[19:13:48] <seb_kuzminsky> except that one guy who wanted that driver
[19:15:34] <cradek> cool
[19:16:30] -!- syyl_ws has quit [Quit: Verlassend]
[19:17:27] <linuxcnc-build> build #1512 of 1301.rip-precise-rtai-i386 is complete: Failure [4failed compile] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1301.rip-precise-rtai-i386/builds/1512 blamelist: dummy, Jeff Epler <jepler@unpythonic.net>
[19:17:48] <linuxcnc-build> build #2311 of 1201.rip-lucid-rtai-i386 is complete: Failure [4failed compile] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1201.rip-lucid-rtai-i386/builds/2311 blamelist: dummy, Jeff Epler <jepler@unpythonic.net>
[19:18:07] <cradek> jepler: good grief, man
[19:18:24] <jepler> geez
[19:18:30] <cradek> jepler: do you want me to do it for you?
[19:18:38] -!- skorasaurus has quit [Ping timeout: 260 seconds]
[19:18:48] <linuxcnc-build> build #124 of 1401.rip-wheezy-rtai-i386 is complete: Failure [4failed compile] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1401.rip-wheezy-rtai-i386/builds/124 blamelist: dummy, Jeff Epler <jepler@unpythonic.net>
[19:19:56] <linuxcnc-build> build #2310 of 1900.clang-lucid-rtai-i386 is complete: Failure [4failed compile] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1900.clang-lucid-rtai-i386/builds/2310 blamelist: dummy, Jeff Epler <jepler@unpythonic.net>
[19:20:04] <jepler> cradek: at this point, yes
[19:24:25] <cradek> so you want it to use 0 if there is only cpu 0; 1 if there are 0,1; 2 if there are 0,1,2 or more
[19:24:38] <jepler> I want it to use cpu 2 on CaptHindsight's one machine
[19:24:47] <jepler> but I was punished for writing that
[19:24:54] <jepler> as were we all
[19:24:58] <jepler> especially seb
[19:25:14] <cradek> I think it was probably worst for you
[19:25:18] <jepler> but fundamentally this is CaptHindsight's fault for not being set up to apply the first simple one-liner change
[19:25:32] <jepler> nah, seb had to interrupt his actual life to reboot a machine I left in an unusable state after that first commit
[19:28:11] <KGB-linuxcnc> 03Jeff Epler 05jepler/hardcode-cpuid-for-ch 2edbb6b 06linuxcnc 10src/rtapi/rtai_rtapi.c Hard-code cpuid (CPU#) for rtai realtime * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=2edbb6b
[19:28:41] <cradek> (I didn't test building it either)
[19:29:04] <jepler> cradek: hah
[19:34:41] <seb_kuzminsky> i like these later commits that blow up in the compile phase instead of the runtests phase ;-)
[19:34:53] <seb_kuzminsky> way more convenient for me
[19:36:50] <cradek> much faster too
[19:38:59] <CaptHindsight> my humble apologies to everyone here for not having installed the tools on the test system
[19:39:20] <jepler> CaptHindsight: eh, just giving you a hard time
[19:40:09] <CaptHindsight> jepler: I forgot to set the sarcasm flag on my last comment ;p
[19:40:27] <jepler> oh my irc client filters out html anyway
[19:40:54] <linuxcnc-build> build #2317 of 0000.checkin is complete: Failure [4failed] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/2317 blamelist: dummy, Jeff Epler <jepler@unpythonic.net>
[19:47:12] -!- gambakufu has quit []
[19:49:32] -!- zeeshan|2 has quit [Ping timeout: 240 seconds]
[19:50:24] -!- _balestrino has quit [Ping timeout: 272 seconds]
[20:06:20] -!- skunkworks has quit [Ping timeout: 240 seconds]
[20:09:10] <CaptHindsight> we have 3,6 and 8 core cpus as well that are getting to the point of retirement and possibly finding their way into machine controllers
[20:11:03] <CaptHindsight> jepler: are first 3 cores the current limit of where RTAI will run?
[20:12:03] -!- i_tarzan has quit [Ping timeout: 240 seconds]
[20:12:46] -!- jerryitt has quit [Quit: Connection closed for inactivity]
[20:13:57] -!- shaun413 has quit [Ping timeout: 245 seconds]
[20:15:21] -!- johtso has quit [Quit: Connection closed for inactivity]
[20:19:28] -!- brianmorel99 has quit [Quit: Leaving]
[20:24:06] -!- PetefromTn_ has quit [Quit: I'm Outta here!!]
[20:34:19] -!- logger[psha] [logger[psha]!~loggerpsh@195.135.238.205] has joined #linuxcnc-devel
[20:35:00] -!- mhaberler has quit [Quit: mhaberler]
[20:35:23] <jepler> CaptHindsight: in unmodified linuxcnc, the intent is to run all realtime threads on the highest numbered CPU. It uses just one CPU in order to provide the "rate monotonic scheduling" guarantee.
[20:35:48] <jepler> so on your 4-core system, that means all linuxcnc code is expected to run on (zero-based) CPU #3
[20:36:00] <jepler> with the change that has taken all day to get right, the intent is to run on (zero-based) CPU #2
[20:36:32] -!- i_tarzan has quit [Ping timeout: 245 seconds]
[20:37:07] -!- theorbtwo has quit [Read error: Connection reset by peer]
[20:37:43] <jepler> which allows some tests which may discriminate between "on bulldozer+ it's important to isolcpus the other core in the same module" and "there's an off-by-one error in rtai cpuid numbering or kernel isolcpus="
[20:41:10] <CaptHindsight> I thought I had noticed something in the source yesterday limiting it to 4 cores, but I am mistaken
[20:41:55] <CaptHindsight> limiting the search for an isolated core to the first 4
[20:49:09] <jepler> I am not aware of any routine in rtai or rtapi which "search[es] for an isolated core".
[20:49:56] <jepler> when using the CPU mask APIs in rtai, it looks for the CPU in the mask with the fewest realtime tasks and chooses that one if possible. but rtapi/linuxcnc doesn't use this API.
[20:50:43] <jepler> kern/latency uses this API, passing in a mask of 0x3, but it only creates one realtime task so picking the CPU with the fewest realtime tasks is moot
[20:50:45] <seb_kuzminsky> rtai only handles up to 8 CPUs, any more than that it can't see
[20:51:05] <seb_kuzminsky> (i believe this from tracing rtai's configure's --enable-cpus=$NUMBER)
[20:51:23] <jepler> oh, yeah, I think I saw the same thing
[20:51:37] <jepler> /proc/rtai/scheduler says "Number of RT CPUs in system: 4 (sized for 8)"
[20:51:50] <seb_kuzminsky> meh
[20:53:00] <jepler> .. which means that in rtapi we should probably limit rt_cpu to be NR_RT_CPUS - 1 at most
[20:53:23] <seb_kuzminsky> it's spelled RTAI_NR_CPUS now
[20:53:36] <cradek> I recall an old smell of something breaking if you had too many CPUs
[20:53:46] <cradek> I wonder if it was that
[20:53:54] <jepler> could have been
[20:54:01] <jepler> why does rtai even configure itself for fewer CPUs than linux?
[20:54:03] <cradek> do you remember that? was it mozmck?
[20:55:15] <jepler> I don't remember specifically
[20:56:20] <jepler> somebody should make an implementation of rtapi that works directly on ipipe
[20:56:26] <jepler> though then you'd have to code up a math library, yuck
[20:56:34] <cradek> that would be really neat
[20:59:11] -!- skunkworks [skunkworks!~chatzilla@str-broadband-ccmts-ws-26.dsl.airstreamcomm.net] has joined #linuxcnc-devel
[21:01:50] <jepler> I think the list of math functions actually used in linuxcnc is: sin, cos, tan, sqrt, fabs, atan, atan2, asin, acos, pow, round, ceil, floor
[21:02:19] <jepler> (and of course sincos)
[21:02:29] -!- ashcan_ has quit [Remote host closed the connection]
[21:04:27] <jepler> cradek could copy those from freebsd libm in an afternoon
[21:04:58] <jepler> CaptHindsight: so I think http://buildbot.linuxcnc.org/dists/wheezy/scratch-rt/binary-i386/linuxcnc_2.6.2~jepler.hardcode.cpuid.for.ch~2edbb6b_i386.deb is finally ready
[21:05:00] <seb_kuzminsky> what does rtai "add" on top of ipipe? just a bunch of headache and heartbreak, seems to me
[21:05:21] <jepler> seb_kuzminsky: userspace / kernelspace shared memory; and math
[21:05:30] <jepler> and a 16550a driver, apparently
[21:05:36] <seb_kuzminsky> yeah woot
[21:06:32] <seb_kuzminsky> arent there things in vanilla linux that provide user/kernel shm?
[21:06:52] <seb_kuzminsky> debugfs does that
[21:08:00] <jepler> yeah
[21:08:30] <jepler> machinekit implemented something called "shmdrv"
[21:08:39] <CaptHindsight> jepler: ok, good...right before I reconfigure this hard drive
[21:14:24] -!- amiri_ has quit [Ping timeout: 250 seconds]
[21:16:30] <jepler> CaptHindsight: well if you still have time to gather some results .. I think isolcpus=1 =2 and =3 are the interesting cases
[21:16:43] <seb_kuzminsky> bummer, debugfs doesnt support mmap
[21:17:06] <jepler> and I'm betting =3 is best performance, meaning that on these bulldozer+ CPUs you want to isolate the other core in the module
[21:17:19] <CaptHindsight> jepler: will do, I'm going to drop Mint and Precise off the drive
[21:17:47] <CaptHindsight> looks like Wheezy and Gentoo from here on
[21:19:45] <jepler> I know I don't miss Ubuntu
[21:23:25] <CaptHindsight> isolcpus doesn't behave the same on Precise, not sure what is happening but the desktop bogs down and video playback gets really choppy with settings that were fine in Wheezy
[21:24:09] -!- afiber__ has quit [Quit: Konversation terminated!]
[21:24:10] -!- skorasaurus has quit [Read error: Connection reset by peer]
[21:26:45] -!- tjb1 has quit [Ping timeout: 240 seconds]
[21:29:28] tjb11 is now known as tjb1
[21:31:32] -!- rob_h has quit [Ping timeout: 240 seconds]
[21:45:34] <CaptHindsight> isolcpus=3, 1ms thread 12,800, 25 uS thread 7,800, rtai kern/latency 32,400
[21:48:56] <jepler> (rtai kern/latency result should be unchanged, since the only modified component of the system is linuxcnc)
[21:51:27] <CaptHindsight> isolcpus=2, 1ms 200,000, 25uS 43,600
[21:53:53] -!- micges [micges!~captain_p@aehy230.neoplus.adsl.tpnet.pl] has joined #linuxcnc-devel
[21:57:52] <CaptHindsight> jepler: any other tests you like to see?
[21:58:08] -!- spatialbrew has quit [Remote host closed the connection]
[21:59:01] <CaptHindsight> let me test isolcpus=1 brb
[21:59:40] BlaDe^_ is now known as BLaDe^
[22:04:03] <jepler> CaptHindsight: yes, isolcpus=1
[22:06:44] <CaptHindsight> isolcpus=1 1mS 146,000, 25uS 28,500
[22:07:25] <CaptHindsight> jepler: ^^
[22:07:28] <jepler> thank you
[22:07:39] <CaptHindsight> no problem
[22:08:23] <jepler> this seems to confirm that it's due to the bulldozer+ paired cores architecture
[22:09:38] <jepler> so the advice for anyone on bulldozer+ would be to isolate both CPUs in the pair
[22:10:17] <jepler> but there's also this general caveat that the rtai kern/latency test selects CPUs differently, so it shouldn't be used when testing isolcpus= performance
[22:10:33] <CaptHindsight> yeah, that was news to me
[22:11:07] <jepler> we could potentially fix it in linuxcnc.org's rtai package to use the same algorithm to select a cpuid to run on..
[22:11:18] <jepler> anyway, see you later. time to stir fry some tofu & veggies
[22:12:00] <CaptHindsight> after 5 here already
[22:12:27] <mozmck> cradek: I could never get my 6-core phenomII to work with the kernel I made for 10.04
[22:12:33] -!- FinboySlick has quit [Quit: Leaving.]
[22:12:43] <mozmck> if that's what you are thinking about.
[22:14:55] <CaptHindsight> mozmck: the kernel on the 10.04 livecd?
[22:16:32] <CaptHindsight> I'm pretty sure that works on a 6-core phenomII
[22:16:37] <mozmck> Yes
[22:16:52] <mozmck> I never could get the realtime to actually work here.
[22:17:14] <CaptHindsight> I'll try it here later. I'm typing on it right now
[22:17:27] <mozmck> It would boot, but if I tried to run the latency test or linuxcnc it would just sit there at 0
[22:17:41] <mozmck> I think linuxcnc would hang.
[22:18:03] <seb_kuzminsky> mozmck: is that 6 cores with 2 hyperthreads each? what does nproc say?
[22:18:11] <seb_kuzminsky> or lshw, or cat /proc/cpuinfo
[22:18:15] <mozmck> Seems like jepler had problems with it on his 4-core too, but I don't remember for sure. I know someone else had the same problem.
[22:18:41] <mozmck> seb_kuzminsky: no, it's 6 real cores. I don't have it in my computer now.
[22:18:51] <CaptHindsight> the 10.04 ISO works great on the 4 core APU's
[22:19:31] <mozmck> ok
[22:19:55] <CaptHindsight> ah we fixed that in BIOS settings
[22:20:15] <mozmck> I have an FX-8350 8-core now. I think the cores in this thing share more stuff than the 6-core did though.
[22:20:22] <CaptHindsight> we had to turn off all the core boosting, core unlocking fancy power stuff
[22:21:09] <CaptHindsight> we has the same problems when configuring phenomII's with 3-6 cores
[22:21:20] <CaptHindsight> but they all worked
[22:21:29] <mozmck> interesting. I don't remember what all I tried. The funny thing was that there were no error messages, the realtime just didn't run.
[22:21:46] <CaptHindsight> yeah, it would just sit at 0
[22:22:07] <CaptHindsight> memleak just said the same thing
[22:22:52] <mozmck> so do you have to do the same bios modifications to run the new rtai kernel?
[22:23:31] <CaptHindsight> well we turn off anything that has to do with speed changing or powering down
[22:24:02] <CaptHindsight> http://wiki.linuxcnc.org/cgi-bin/wiki.pl?TroubleShooting#AMD_APU_UEFI_settings
[22:24:21] <CaptHindsight> turn off any virtualization, C6 etc etc
[22:25:02] <CaptHindsight> all these tests we done with those settings http://wiki.linuxcnc.org/cgi-bin/wiki.pl?The_Isolcpus_Boot_Parameter_And_GRUB2#Latency_Testing
[22:25:32] -!- tinkerer1 has quit [Quit: Leaving.]
[22:25:55] <mozmck> Thanks. I haven't tried it on this machine yet, and it will not be running a machine except maybe for some testing.
[22:26:33] -!- ravenlock has quit [Quit: Leaving]
[22:26:40] <mozmck> But my 6-core board might could come back as a machine controller. Hate to use it for that when it is a pretty good desktop PC!
[22:27:19] <CaptHindsight> Ideally I'd like to run Linuxcnc with GPU hardware acceleration and have the other cores for image processing
[22:28:08] <CaptHindsight> rtai on one core, the GPU doing what it's good at and the other cores for machine vision
[22:28:41] <CaptHindsight> is that too much to ask? :)
[22:29:09] <CaptHindsight> the new GPU hardware drivers clobber latency
[22:30:08] <CaptHindsight> if kernel mode settings are on it won't even run on Wheezy, just a scrambled screen
[22:30:32] <CaptHindsight> I have to install from the wheezy hybrid ISO to confirm
[22:30:55] <CaptHindsight> I use wheezy LXDE and then installed the Linuxcnc debs
[22:32:14] <CaptHindsight> the integrated GPU on the early APU's are fine with the hardware driver and KMS on
[22:34:12] <mozmck> seems like CPUs/GPUs are not getting and more RT friendly
[22:34:22] -!- Klaus-Krefeld has quit [Quit: ChatZilla 0.9.90.1 [Firefox 20.0/20130329043827]]
[22:34:25] <mozmck> any more
[22:36:31] <CaptHindsight> and xenomai is not going to have the kernel API's anymore for v3
[22:36:50] <CaptHindsight> so it will slow down
[22:51:03] -!- kfoltman has quit [Quit: Ex-Chat]
[22:53:58] -!- sylphiae has quit [Ping timeout: 255 seconds]
[23:06:20] -!- micges has quit [Quit: Leaving]
[23:11:07] <CaptHindsight> mozmck: according to the preempt_rt tests at osadl the AMD A10's and the Intel i7's have the lowest latency, but we haven't been able to reproduce their results with Linuxcnc and preempt_rt
[23:11:21] <CaptHindsight> https://www.osadl.org/Latency-plot-of-system-in-rack-2-slot.qa-latencyplot-r2s7.0.html
[23:11:29] <CaptHindsight> https://www.osadl.org/Latency-plot-of-system-in-rack-0-slot.qa-latencyplot-r0s7.0.html
[23:12:29] <CaptHindsight> the Xeons do great as well
[23:26:08] -!- asdfasd has quit [Ping timeout: 250 seconds]
[23:27:15] -!- theorb has quit [Client Quit]
[23:39:56] <seb_kuzminsky> i added a link to the live+install email on the wlo front page
[23:43:02] -!- knownasilya has quit [Quit: Connection closed for inactivity]
[23:43:24] <seb_kuzminsky> and i tried to clean up the Downloading page here: http://linuxcnc.org/index.php/english/download
[23:44:12] <seb_kuzminsky> bbl
[23:45:48] -!- grummund has quit [Ping timeout: 272 seconds]