Back
[00:04:26] -!- mercuryrising has quit [Client Quit]
[00:13:44] -!- gmag has quit [Ping timeout: 256 seconds]
[00:15:12] -!- mhaberler has quit [Quit: mhaberler]
[00:22:39] -!- carper64 [carper64!carper64@027878c8.bb.sky.com] has joined #linuxcnc-devel
[00:23:45] -!- krusty_ar has quit [Ping timeout: 245 seconds]
[00:24:35] -!- micges [micges!~zenek@egh175.neoplus.adsl.tpnet.pl] has joined #linuxcnc-devel
[00:25:31] -!- carper64 has quit [Client Quit]
[00:25:39] -!- carper64 [carper64!carper64@027878c8.bb.sky.com] has joined #linuxcnc-devel
[00:26:40] -!- carper64 has quit [Client Quit]
[00:26:49] -!- carper64 [carper64!carper64@027878c8.bb.sky.com] has joined #linuxcnc-devel
[00:29:35] <jepler> if our EXPORT_SYMBOLs are wrong, why does kernelspace work?
[00:32:25] <jepler> (that branch has 81 commits since rtos-integration-preview3 and 488 since linuxcnc.org master .. I don't know what I am supposed to be looking at)
[00:40:05] -!- carper64 has quit []
[00:41:20] -!- carper64 [carper64!carper64@027878c8.bb.sky.com] has joined #linuxcnc-devel
[00:43:14] -!- carper64 [carper64!carper64@027878c8.bb.sky.com] has parted #linuxcnc-devel
[00:57:53] -!- sumpfralle has quit [Ping timeout: 240 seconds]
[01:29:06] -!- sumpfralle has quit [Read error: Operation timed out]
[01:38:10] -!- chopper79 has quit [Ping timeout: 256 seconds]
[02:22:01] -!- tmcw has quit [Remote host closed the connection]
[02:30:11] <jepler> http://emergent.unpythonic.net/files/sandbox/sim-symbol-fixes.patches.mbox
[02:30:35] <jepler> this makes two (newly-created) tests pass that didn't before: function symbols and global variable symbols don't leak from one component to a subsequently-loaded component.
[02:30:54] <jepler> mhaberler: ^^^ this may be relevant to you (I hope it is)
[02:38:30] -!- micges has quit [Quit: Leaving]
[02:45:46] -!- cbn_ has quit [Quit: Leaving]
[02:46:14] -!- linuxcnc-build has quit [Remote host closed the connection]
[02:47:45] -!- coors [coors!coors@c-71-193-105-131.hsd1.in.comcast.net] has joined #linuxcnc-devel
[02:48:33] <coors> hi, is this channel for hardware as well or just software?
[02:48:35] -!- hm2-buildmaster_ has quit [Remote host closed the connection]
[02:53:54] -!- Keknom has quit [Quit: Leaving.]
[02:54:08] <KimK> coors: I would say hardware and software if LinuxCNC-development-related. For hardware and software user-LinuxCNC-problems there's #linuxcnc. There's a larger audience there too. But nevermind, what's your trouble?
[02:56:26] <coors> looking for linear motion upgrades to buy for a homebuilt cnc
[02:56:31] <coors> so purely hardware
[02:57:16] <KimK> OK, like surplus rails and trucks?
[02:59:04] <coors> basically ballscrews and guides
[02:59:19] <coors> or surplus modules could work, too
[03:00:04] <coors> oh, that's this one in non-dev
[03:00:04] -!- RangerRick has quit [Remote host closed the connection]
[03:00:06] <coors> i see :)
[03:02:48] <KimK> Ha, I only knew VXB as a supplier of low-cost ball bearings, but it seems they have linear motion stuff too, maybe this will help:
http://www.vxb.com/page/bearings/CTGY/LM?gclid=CLOy6ab_qrYCFROnPAod4ioAog
[03:05:08] <coors> do you have experience with those?
[03:05:27] <coors> i am looking for people who use them and can tell me what kind of precision to expect from different series
[03:13:17] <KimK> The "ball-shells"(?) on round rails have "moderate" precision, and perhaps a little more play, but they're a lot cheaper. The linear rails and ball-trucks are much higher precision and near-zero play, but are a lot more money. Does that help?
[03:14:13] <coors> sort of
[03:14:23] <coors> that part i know and i checked out prizes a bit, too
[03:14:34] <coors> but i don't have enough experience to know exactly what i need
[03:14:47] <coors> so someone who builds machines all the time would be great to talk to
[03:16:12] <KimK> Well, first, what would you like to do, and how much do you expect to spend?
[03:16:29] -!- vladimirek [vladimirek!~vladimire@95.105.250.72] has joined #linuxcnc-devel
[03:16:53] -!- Mikegg has quit [Ping timeout: 240 seconds]
[03:17:44] <coors> money is the biggest issue
[03:17:58] <coors> i am ok doing machining and setting up things myself
[03:18:17] <coors> so i figured if i could get ballscrews and guides for "cheap", i can make it work
[03:18:37] <coors> question is just what kind of mechanism you can get for "cheap"
[03:18:46] <coors> i am thinking about $200 per axis
[03:19:02] <coors> just mechanism and aluminum, i got steppers and electronics already
[03:19:17] -!- zzolo has quit [Quit: zzolo]
[03:25:52] <KimK> What kind of machine do you want to end up with? Or perhaps what materials do you want to be able to machine?
[03:26:56] <coors> mostly PCBs, some hdpe and wood
[03:27:05] <coors> maybe aluminum, but i'm ok passing on that one
[03:27:09] <coors> if too expensive
[03:33:22] <KimK> OK. And this is to be a router or small mill type of machine? How big a workpiece?
[03:34:15] <coors> router or mill doesnt matter... atm it's a router with gantry though
[03:34:26] <coors> working area should be approx 8"x10" or so
[03:34:32] <coors> so the smallest you find
[03:35:00] <KimK> There are a lot of guys who have built MDF or plywood router frames. Or, you can start with a tabletop mill and a lot of your problems go away.
[03:36:01] <coors> yeah, do you think the plywood ones are good enough though?
[03:36:05] <KimK> There are many ways to skin the proverbial cat. I also should mention #diycnc if that helps.
[03:36:10] <coors> i would need <1 thou backlash/slop
[03:36:26] <coors> i joined that one, but there's no active people atm
[03:36:32] <coors> i'll keep my eyes open though
[03:37:02] <KimK> Ha, I should have looked first, sorry
[03:37:38] -!- zzolo has quit [Quit: zzolo]
[03:38:21] <coors> oh didnt notice you're active there, too
[03:38:32] <coors> i talked to theos very briefly in the morning
[03:38:41] <coors> i think i wanna talk to him again soon
[03:38:47] <coors> maybe he's got some experience
[03:39:07] -!- erictheise has quit [Quit: erictheise]
[03:39:20] <coors> ideally, i will find someone who built/upgraded recently and can tell what works/doesn't work
[03:39:26] <KimK> You can make a stiff enough plywood frame, but what kind of ways will the frame support, that's the question. Something like simple bearing rollers on metal flats is cheap to make, but hard to get all the play out (if you're serious about the .001")
[03:39:32] <coors> i mean i looked around and got some good ideas what i can get
[03:39:42] <coors> but they all claim to be backlash free :>
[03:40:11] <coors> yeah, i will have to be 0.001"
[03:40:33] <coors> even thats kinda borderline for some of the bigger surface mount packages
[03:40:38] <coors> but i can live with that
[03:40:51] <coors> i am trying to be as realistic as possible here
[03:41:35] <KimK> Yes, that's another way, if you have some kind of preloading spring on the rollers, but the problem is that wood tends to warp over time, that's how they make rocking chair feet, lol. So that trick works better against a cast iron frame.
[03:43:00] <pcw_home> Yeah for the smallish forces needed for PCB routing, a anti--backlash spring might be OK
[03:43:35] <coors> i do have one
[03:43:45] <coors> not neccessarily a good one i think though
[03:43:47] <KimK> Yes, if you don't mind a little fiddling from time to time (Hi Peter!)
[03:43:57] <coors> hi peter, thanks for chipping in
[03:44:19] <coors> i have a basically a hardware store type threaded rod with springloaded nuts
[03:44:28] <coors> and it's close, but not quite good enough
[03:44:42] <coors> so i am looking to either get acme screws or low-end ballscrews
[03:45:45] <pcw_home> so theres still backlash with the spring loaded nuts?
[03:46:51] <coors> i am not entirely sure if it's backlash or just mechanical slop from the bad supports
[03:46:59] <KimK> pcw_home: I saw your 8-thermocouple board, recommended it to Stuart, he said he ordered some, even though they're not yet available. I have to run to the shop, back in a bit.
[03:47:07] <coors> but with a dial gauge, i see 3-5 mil slop
[03:47:12] <pcw_home> It all adds up unfortunately
[03:49:20] <coors> yeah, that's pure z motion though
[03:49:29] <coors> not moving anything else but one axis
[03:49:36] <coors> so i'm sure i wanna get something better
[03:49:50] <coors> but i am trying to see who can recommend something reasonably cheap
[04:05:13] -!- AR_ has quit [Ping timeout: 256 seconds]
[04:12:11] -!- Tecan has quit [Quit: Live Long And Phosphor!]
[04:28:11] -!- coors has quit []
[04:29:17] -!- FinboySlick has quit [Quit: Leaving.]
[04:38:21] tandoori_Chicken is now known as tandoori
[04:38:34] -!- tandoori has quit [Changing host]
[04:48:47] -!- jfire has quit [Quit: Leaving.]
[04:50:27] -!- skorasaurus has quit [Remote host closed the connection]
[04:58:33] -!- Valen has quit [Read error: No route to host]
[05:00:57] -!- mhaberler [mhaberler!~mhaberler@extern-186.stiwoll.mah.priv.at] has joined #linuxcnc-devel
[05:02:27] -!- Fox_Muldr has quit [Ping timeout: 240 seconds]
[05:08:45] <mhaberler> Jeff - thanks, the RTLD_NOW on module load makes sense, had'nt thought of that
[05:09:30] <mhaberler> re branch: what I suggest to look at is src/rtapi/Submakefile: how rtapi_app is linked now
[05:11:09] <mhaberler> also, to verify there's no symbol leakage from rtapi_app into comp space I've added these two pointers as NULL in rtapi:
http://git.mah.priv.at/gitweb?p=emc2-dev.git;a=blob;f=src/rtapi/sim_rtapi_app.cc;h=d95d8b52e3c7e3f9cd6c2cdfb194f3cd8c998f71;hb=refs/heads/dynload-rtapi#l102
[05:11:53] <mhaberler> in case they leak into comp space they'd immediately cause a segfault (which it did)
[05:17:40] <mhaberler> I still need to fully understand the linker semantics for ../rtlib/%.so implies ad if that works for me
[05:17:57] frewsxcv94709 is now known as frewsxcv
[05:18:23] <mhaberler> one tricky part was an interaction between libpthread and .so's - namely a thread being started from within a loaded .so
[05:21:05] <mhaberler> this resulted in segfaults unless the comp was compiled (and I think linked) with -pthread
[05:26:32] <mhaberler> trying to parse the Makefile fragment ;) I assume:
[05:27:20] <mhaberler> $(Q)ld --no-export-dynamic -d -r -o objects/$*.tmp $^ - translates into 'minimally needed export of symbols'?
[05:28:35] <mhaberler> this: $(Q)objcopy -j .rtapi_export -O binary objects/$*.tmp /dev/stdout \
[05:28:35] <mhaberler> + | tr '\0' '\n' | grep . | sort -u > objects/$*.exp.tmp
[05:29:10] <mhaberler> I assume you're collecting all syms tagged with EXPORT_SYMBOL(sym);
[05:31:30] <mhaberler> then you localize all not explicitly exported syms here (?): $(Q)objcopy --keep-global-symbols=objects/$*.exp.tmp --localize-hidden objects/$*.tmp objects/$*.tmp2
[05:33:49] <mhaberler> re the question: why kernel space works - that I have no good answer to; I'll try to reproduce how that happened but I am fairly sure it was a missing EXPORT_SYMBOL(sym)
[05:34:48] -!- jfire has quit [Quit: Leaving.]
[05:36:00] <mhaberler> I'll use this make fragment on the dynload-rtapi branch and report the results; I routinely test that branch on ARM, and with xenomai-kernel and rt-preempt threads too - manually, since the buildbot aint there yet
[05:36:42] -!- tjtr33 has quit [Quit: Leaving]
[05:37:18] <mhaberler> thanks for giving thought to it, it was a major time consumer and I think it's contained now
[05:39:07] -!- kwallace [kwallace!~kwallace@smb-38.sonnet.com] has parted #linuxcnc-devel
[06:05:53] <mhaberler> re screenshots:
http://imgur.com/rDhhgEz - some comments are timeless; I didnt know the term "percussive maintenance"
[06:22:44] -!- JT-Shop has quit [Ping timeout: 252 seconds]
[06:22:53] -!- jthornton has quit [Ping timeout: 256 seconds]
[06:26:29] -!- psha[work] [psha[work]!~psha@195.135.238.205] has joined #linuxcnc-devel
[06:28:14] -!- mhaberler has quit [Quit: mhaberler]
[06:32:10] -!- mhaberler [mhaberler!~mhaberler@macbook.stiwoll.mah.priv.at] has joined #linuxcnc-devel
[06:49:34] -!- tomate__ [tomate__!~quassel@2a01:e34:ee61:a5f0:e0d2:5115:8e1a:be1c] has joined #linuxcnc-devel
[06:57:32] <mhaberler> jepler: tried your linker makefile fragment; it seems we need to keep __i686.get_pc_thunk.bx syms around: `__i686.get_pc_thunk.bx' referenced in section `.text' of objects/watchdog.tmp2: defined in discarded section `.text.__i686.get_pc_thunk.bx[__i686.get_pc_thunk.bx]' of objects/watchdog.tmp2
[07:00:28] <mhaberler> not sure how to massage those in
[07:05:54] -!- JT-Shop [JT-Shop!~John@162.72.171.228] has joined #linuxcnc-devel
[07:06:03] -!- jthornton [jthornton!~john@162.72.171.228] has joined #linuxcnc-devel
[07:06:36] -!- emel has quit [Excess Flood]
[07:11:51] -!- linuxcnc-build [linuxcnc-build!~linuxcnc-@174-29-90-87.hlrn.qwest.net] has joined #linuxcnc-devel
[07:19:01] -!- pingufan has quit [Quit: Konversation terminated!]
[07:19:03] -!- toastyde1th has quit [Read error: Connection reset by peer]
[07:19:09] -!- JT-Shop has quit [Read error: Connection reset by peer]
[07:19:09] -!- jthornton has quit [Read error: Connection reset by peer]
[07:26:53] -!- cevad has quit [Ping timeout: 240 seconds]
[07:27:18] -!- carper64 [carper64!carper64@027878c8.bb.sky.com] has joined #linuxcnc-devel
[07:29:47] -!- carper64 [carper64!carper64@027878c8.bb.sky.com] has parted #linuxcnc-devel
[07:33:54] -!- maximilian_h has quit [Quit: Leaving.]
[07:36:45] -!- hjhart has quit [Ping timeout: 276 seconds]
[07:37:25] -!- carper64 [carper64!carper64@027878c8.bb.sky.com] has joined #linuxcnc-devel
[07:37:53] -!- carper64 has quit [Client Quit]
[07:43:37] -!- JT-Shop [JT-Shop!~John@162.72.171.228] has joined #linuxcnc-devel
[07:43:38] -!- jthornton [jthornton!~john@162.72.171.228] has joined #linuxcnc-devel
[07:57:27] <KGB-linuxcnc> 03chrisinnanaimo 05master 8e56dc2 06linuxcnc 10configs/sim/gscreen_custom/industrial.glade * gscreen config -fix extra axes showing when not use XY or Z
[07:57:27] <KGB-linuxcnc> 03chrisinnanaimo 05master bc4a73d 06linuxcnc 10configs/sim/gscreen_custom/gscreen.glade 10configs/sim/gscreen_custom/industrial.glade 10src/emc/usr_intf/gscreen/gscreen.glade 10src/emc/usr_intf/gscreen/gscreen.py * gscreen - add button to load Classicladder if available
[08:02:35] -!- emel has quit [Excess Flood]
[08:08:54] -!- Chicharo has quit [Remote host closed the connection]
[08:09:28] -!- jthornton has quit [Read error: Connection reset by peer]
[08:09:29] -!- JT-Shop has quit [Read error: Connection reset by peer]
[08:17:06] -!- zultron has quit [Read error: Operation timed out]
[08:18:48] -!- vladimirek has quit [Ping timeout: 256 seconds]
[08:21:04] Elwell_ is now known as Elwell
[08:21:12] -!- vladimirek [vladimirek!~vladimire@95.105.250.72] has joined #linuxcnc-devel
[08:30:01] -!- jthornton [jthornton!~john@162.72.171.228] has joined #linuxcnc-devel
[08:30:06] -!- JT-Shop [JT-Shop!~John@162.72.171.228] has joined #linuxcnc-devel
[08:39:20] -!- JT-Shop has quit [Read error: Connection reset by peer]
[08:39:20] -!- jthornton has quit [Read error: Connection reset by peer]
[08:40:49] -!- JT-Shop [JT-Shop!~John@162.72.171.228] has joined #linuxcnc-devel
[08:40:50] -!- jthornton [jthornton!~john@162.72.171.228] has joined #linuxcnc-devel
[08:45:21] -!- rob_h [rob_h!~rob_h@027c16a0.bb.sky.com] has joined #linuxcnc-devel
[08:57:13] -!- carper64 [carper64!carper64@027878c8.bb.sky.com] has joined #linuxcnc-devel
[08:59:04] -!- carper64 has quit [Client Quit]
[09:05:33] -!- jpk has quit [Ping timeout: 240 seconds]
[09:05:37] -!- carper64 [carper64!~carper64@027878c8.bb.sky.com] has joined #linuxcnc-devel
[09:05:48] -!- carper64 has quit [Client Quit]
[09:14:14] -!- rob_h has quit [Quit: Leaving]
[09:19:00] -!- JT-Shop has quit [Read error: Connection reset by peer]
[09:19:00] -!- jthornton has quit [Read error: Connection reset by peer]
[09:19:36] -!- JT-Shop [JT-Shop!~John@162.72.171.228] has joined #linuxcnc-devel
[09:19:37] -!- jthornton [jthornton!~john@162.72.171.228] has joined #linuxcnc-devel
[09:41:11] -!- vladimirek has quit [Remote host closed the connection]
[09:53:40] -!- V0idExp has quit [Read error: No route to host]
[10:27:38] -!- mhaberler has quit [Ping timeout: 256 seconds]
[10:34:42] -!- carper64 [carper64!carper64@027878c8.bb.sky.com] has joined #linuxcnc-devel
[10:48:39] -!- skunkworks has quit [Remote host closed the connection]
[11:17:14] -!- b_b has quit [Changing host]
[11:19:03] -!- zultron [zultron!~zultron@99-190-134-148.lightspeed.austtx.sbcglobal.net] has joined #linuxcnc-devel
[11:19:34] -!- sumpfralle has quit [Ping timeout: 246 seconds]
[11:21:33] -!- Mikegg has quit [Ping timeout: 256 seconds]
[11:29:30] -!- skunkworks [skunkworks!~skunkwork@68-115-41-210.static.eucl.wi.charter.com] has joined #linuxcnc-devel
[11:46:15] <jepler> mhaberler: hm, I didn't have any problems that seemed related to __i686.get_pc_thunk.bx but (A) of course hm2 is not building and (B) I'm on amd64
[11:56:33] <jepler> and whatever is going on I can't build for 32-bits on this machine
[11:56:39] <jepler> $ CC="gcc -m32" CXX="g++ -m32" ./configure
[11:56:39] <jepler> checking libgl1-mesa-dri workaround... test for libgl1-mesa-dri workaround failed, please file a bug
[11:58:53] <jepler> the first failure is because that script respect $(CXX) in one instance but not $(CC) or $(CXX) in the other instances
[12:00:01] <jepler> but the second failure is because I don't have enough i386 devel stuff installed, and unfortunately something in the dependencies of libglu1-mesa-dev isn't multiarch enough to install at the same time for 32-bit and 64-bit systems
[12:09:36] -!- coors [coors!coors@c-71-193-105-131.hsd1.in.comcast.net] has joined #linuxcnc-devel
[12:27:46] -!- ravenlock has quit [Remote host closed the connection]
[12:53:06] <jepler> aha, I threw out the baby with the bathwater on that last set of commits
[12:53:32] <jepler> I started from building with comp/makefile.modinc but of course that was totally broken (it didn't have the old objcopy magic)
[12:58:12] -!- Valen has quit [Quit: Leaving.]
[13:00:06] -!- asdfasd has quit [Ping timeout: 256 seconds]
[13:00:46] <jepler> mhaberler: for your consideration:
http://emergent.unpythonic.net/files/sandbox/sim-symbol-fixes-v2.patches.mbox
[13:26:43] -!- psha[work] has quit [Quit: Lost terminal]
[13:38:54] -!- stsydow has quit [Quit: Leaving]
[14:05:23] <jepler> yay: underhanded C contest is back. boo: server killed by excess traffic
[14:23:19] -!- mhaberler [mhaberler!~mhaberler@089144206116.atnat0015.highway.a1.net] has joined #linuxcnc-devel
[14:37:05] -!- erictheise has quit [Quit: erictheise]
[14:49:03] -!- Tecan has quit [Changing host]
[14:54:37] -!- mhaberler has quit [Quit: mhaberler]
[15:11:21] -!- AphelionZ has quit [Quit: Page closed]
[15:23:14] -!- zzolo has quit [Quit: zzolo]
[15:25:42] -!- mhaberler [mhaberler!~mhaberler@178-191-249-145.adsl.highway.telekom.at] has joined #linuxcnc-devel
[15:26:32] <mhaberler> Jeff - thanks, just saw round#2 of patches, will give them a spin lateron
[15:27:50] <mhaberler> assuming this works out, I would think the patch should go into v2.5_branch because that's quite fundamental, and v2.5_branch is merged upward into rtos-integration-preview3 - what do you think?
[15:33:42] -!- AR_ has quit [Ping timeout: 264 seconds]
[15:43:43] -!- mhaberler has quit [Quit: mhaberler]
[15:50:05] -!- hewball has quit [Ping timeout: 256 seconds]
[15:51:18] -!- erictheise has quit [Quit: erictheise]
[15:52:12] -!- asdfasd has quit [Read error: Connection reset by peer]
[15:53:49] <seb_kuzminsky> mhaberler: why do you think this should go in 2.5?
[15:54:30] -!- ve7it [ve7it!~LawrenceG@S01060014d19d0b68.pk.shawcable.net] has joined #linuxcnc-devel
[16:04:23] -!- carper64 has quit []
[16:06:41] -!- kwallace [kwallace!~kwallace@smb-92.sonnet.com] has joined #linuxcnc-devel
[16:12:34] -!- cnc_yqp has quit [Quit: ChatZilla 0.9.90 [Firefox 19.0.2/20130307023931]]
[16:16:45] -!- mephux has quit [Excess Flood]
[16:35:26] -!- mackerski has quit [Quit: mackerski]
[16:37:53] -!- L84Supper has quit [Ping timeout: 240 seconds]
[16:39:03] -!- L84Supper [L84Supper!~Larch@unaffiliated/l84supper] has joined #linuxcnc-devel
[16:45:43] -!- tomate__ has quit [Remote host closed the connection]
[16:49:20] <jepler> mhaberler: with cradek's approval and some positive testing .. I think these changes may be fixing real problems.
[16:55:12] -!- dway has quit [Quit: NOOOOOOooooooooo……]
[17:11:21] -!- mephux has quit [Excess Flood]
[17:12:53] -!- andypugh [andypugh!~andy2@cpc16-basl9-2-0-cust685.20-1.cable.virginmedia.com] has joined #linuxcnc-devel
[17:16:27] -!- V0idExp has quit [Quit: Leaving.]
[17:16:28] -!- jfire has quit [Quit: Leaving.]
[17:19:23] -!- erictheise has quit [Quit: erictheise]
[17:39:57] -!- skorasaurus has quit [Ping timeout: 276 seconds]
[17:56:37] -!- rob_h [rob_h!~rob_h@027c16a0.bb.sky.com] has joined #linuxcnc-devel
[18:03:52] -!- RangerRick has quit [Remote host closed the connection]
[18:13:05] -!- IchGuckLive has quit [Quit: ChatZilla 0.9.87 [Firefox 18.0/20130108033621]]
[18:14:44] <jepler> seb_kuzminsky: the real problem case is tests/symbols.1, where two components (possibly relevant that they're compiled by comp, not by the regular 'make' process) which happen to have global variables of the same name without associated EXPORT_SYMBOLs
[18:14:56] <jepler> seb_kuzminsky: changes to the global in the one component are seen in the other component(!)
[18:15:03] -!- skorasaurus has quit [Ping timeout: 276 seconds]
[18:15:05] <jepler> also it may be relevant that the globals are not initialized (they are in bss)
[18:16:04] <jepler> though I have no evidence any developer has hit this problem
[18:19:03] -!- uncrtnmind has quit [Remote host closed the connection]
[18:20:00] -!- erictheise has quit [Quit: erictheise]
[18:20:59] -!- vladimirek [vladimirek!~vladimire@95.105.250.72] has joined #linuxcnc-devel
[18:23:07] <cradek> jepler: how is ppc different? I'm pizzled by the 8af8 change
[18:23:11] <cradek> haha pizzled
[18:23:25] <jepler> cradek: I have NFC about the powerpc parts of the change. they are copy&paste from the main Makefile.
[18:23:54] <jepler> back in the day we had someone developing on powerpc for whom all that black magic didn't work; his proposed fix was simply to skip it on powerpc
[18:24:36] -!- AR_ has quit [Ping timeout: 245 seconds]
[18:24:44] <cradek> oh ok, so it is "tested"
[18:25:45] <cradek> I have no idea how to evaluate that part of the change except to be happy that it fixes your runtest
[18:26:16] <cradek> even man ld about -d doesn't illuminate me
[18:29:04] <jepler> in C, a declaration like this: int i;
[18:29:20] <jepler> becomes an assembler directive which reserves enough bytes for 'i' in the 'common symbol area': .commi,4,4
[18:29:33] <jepler> err there should have been a space before 'i'
[18:29:54] <jepler> the linker merges all the common symbol declarations with the same name
[18:30:01] <jepler> (and so they'd better also share the size and alignment)
[18:30:52] <jepler> anyway, 'ld -d -r' stops those common definitions in the input files from being merged with common definitions in a subsequent link, such as by the dynamic linker.
[18:31:13] <jepler> specifically fixing the symbols.1 test
[18:31:52] <cradek> aha, thanks
[18:39:45] -!- hashfail has quit [Ping timeout: 276 seconds]
[18:40:24] -!- stsydow has quit [Ping timeout: 264 seconds]
[18:53:19] -!- ravenlock has quit [Ping timeout: 252 seconds]
[18:57:03] -!- tmcw has quit [Read error: Connection reset by peer]
[19:03:26] -!- sumpfralle has quit [Ping timeout: 240 seconds]
[19:04:30] -!- syyl_ws has quit [Quit: Verlassend]
[19:29:52] -!- tmcw_ has quit [Remote host closed the connection]
[19:31:56] -!- hjhart has quit [Ping timeout: 260 seconds]
[19:35:30] -!- skorasaurus has quit [Quit: left the building.]
[19:36:57] -!- AR_ has quit [Ping timeout: 252 seconds]
[19:37:39] -!- ler_hydra has quit [Remote host closed the connection]
[19:38:51] -!- stsy has quit [Client Quit]
[19:40:05] -!- chopper79 has quit [Quit: Leaving.]
[19:57:38] -!- mephux has quit [Excess Flood]
[20:05:26] -!- elchicharo has quit [Ping timeout: 256 seconds]
[20:07:12] -!- gimpswork has quit []
[20:13:48] -!- bedah has quit [Remote host closed the connection]
[20:14:13] -!- skunkworks has quit [Read error: Connection reset by peer]
[20:16:53] <jepler> so to test this on buildbot I push it with a branch name of my choice, then once the tests are done I delete it? or is there some other magic?
[20:16:56] <jepler> I don't do that often enough
[20:18:12] -!- ravenlock_ has quit [Ping timeout: 264 seconds]
[20:18:31] -!- stsydow has quit [Quit: Verlassend]
[20:20:20] -!- vladimirek has quit [Remote host closed the connection]
[20:21:31] -!- b_b has quit [Changing host]
[20:22:30] <seb_kuzminsky> jepler: that's it
[20:26:46] <KGB-linuxcnc> 03jepler 05jepler-modlinking d2c3e87 06linuxcnc 10src/hal/utils/comp.g * comp: fix communication of make exit code
[20:26:47] <KGB-linuxcnc> 03jepler 05jepler-modlinking ef306d1 06linuxcnc 10src/hal/utils/comp.g * comp: it's an error if cd fails
[20:26:50] <KGB-linuxcnc> 03jepler 05jepler-modlinking 113fb02 06linuxcnc 10tests/ 10(12 files in 2 dirs) * New tests (currently failing in sim)
[20:26:57] <KGB-linuxcnc> 03jepler 05jepler-modlinking 60c8c4c 06linuxcnc 10src/Makefile.modinc.in * build: use $(Q) to quiet external module building
[20:27:03] <KGB-linuxcnc> 03jepler 05jepler-modlinking ef119f5 06linuxcnc 10src/Makefile 10src/Makefile.modinc.in 04tests/symbols.1/xfail * sim: don't mix common symbols from two .so modules
[20:27:11] <KGB-linuxcnc> 03jepler 05jepler-modlinking 5da4256 06linuxcnc 10src/rtapi/sim_rtapi_app.cc 04tests/symbols.0/xfail * sim: Bind all symbols at load time
[20:27:18] <KGB-linuxcnc> 03jepler 05jepler-modlinking 89271c1 06linuxcnc 10src/rtapi/sim_rtapi_app.cc * sim: don't export symbols from sim_rtapi_app to components
[20:36:44] <seb_kuzminsky> linuxcnc-build: watch checkin
[20:36:45] <linuxcnc-build> watching build checkin #877 until it finishes..
[20:36:59] <jepler> seb_kuzminsky: it looks like you intervened manually for me there..
[20:37:04] <jepler> thanks
[20:37:13] -!- ekolojik has quit [Quit: Konversation terminated!]
[20:38:02] -!- syyl has quit [Quit: Leaving]
[20:39:16] zq is now known as bzzzz
[20:41:56] -!- _1SheYode has quit [Ping timeout: 260 seconds]
[20:47:25] <seb_kuzminsky> yeah there was a hung build from last night, not sure what that was about
[20:47:41] <seb_kuzminsky> i was backing up the buildbot at about that time, maybe that's related... i'll keep an eye on it
[20:55:25] <jepler> seb_kuzminsky: when buildbot does the package builds it will build the pdf documentation. what do you think of removing the pdf documentation building from the non-package builds (--enable-build-documentation=html)? sorry if this a retread of a past discussion...
[20:58:41] <jepler> I am glad that the buildbots are building the html even if (I think) we're still not packaging it
[20:59:04] -!- chopper79 has quit [Quit: Leaving.]
[21:01:37] -!- DJ9DJ has quit [Quit: bye]
[21:05:44] -!- sumpfralle has quit [Ping timeout: 260 seconds]
[21:11:36] -!- FinboySlick has quit [Quit: Leaving.]
[21:12:10] <seb_kuzminsky> there's basically 4 kinds of builds on the buildbot: rip & test (no docs), clang (no docs), docs (
http://buildbot.linuxcnc.org/doc/), and full packaging
[21:12:57] <seb_kuzminsky> oh wait, i'm wrong
[21:13:02] <seb_kuzminsky> the rip build also builds docs
[21:13:03] <seb_kuzminsky> oops
[21:13:15] -!- joe9 has quit [Remote host closed the connection]
[21:13:15] <seb_kuzminsky> i wonder if i meant to do that
[21:13:52] <seb_kuzminsky> i think maybe i did? so if the docs fail to build, the build fails & stops?
[21:14:45] <seb_kuzminsky> are you suggesting to turn off pdf (but not html) on bbot rip builds for speed reasons?
[21:14:55] -!- chillly has quit [Quit: Leaving]
[21:17:07] <seb_kuzminsky> oops, i forgot to say jepler to make his terminal beep
[21:18:51] -!- mhaberler [mhaberler!~mhaberler@178-191-243-18.adsl.highway.telekom.at] has joined #linuxcnc-devel
[21:19:09] <linuxcnc-build> build #80 of precise-i386-realtime-rip is complete: Failure [4failed runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/precise-i386-realtime-rip/builds/80 blamelist: Jeff Epler <jepler@unpythonic.net>
[21:20:36] <linuxcnc-build> build #879 of hardy-amd64-realtime-rip is complete: Failure [4failed runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/hardy-amd64-realtime-rip/builds/879 blamelist: Jeff Epler <jepler@unpythonic.net>
[21:21:22] -!- skunkworks [skunkworks!~chatzilla@str-broadband-ccmts-ws-26.dsl.airstreamcomm.net] has joined #linuxcnc-devel
[21:22:10] <seb_kuzminsky> jepler: test_use: Unknown symbol testuse (err 0)
[21:23:53] <linuxcnc-build> build #879 of hardy-i386-realtime-rip is complete: Failure [4failed runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/hardy-i386-realtime-rip/builds/879 blamelist: Jeff Epler <jepler@unpythonic.net>
[21:25:11] <linuxcnc-build> build #879 of lucid-i386-realtime-rip is complete: Failure [4failed runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/lucid-i386-realtime-rip/builds/879 blamelist: Jeff Epler <jepler@unpythonic.net>
[21:28:30] <mhaberler> this test should fail and it does since the symbols have not been EXPORT_SYMBOL()'d
[21:30:25] <seb_kuzminsky> linuxcnc-build: notify list
[21:30:26] <linuxcnc-build> The following events are being notified: ['failure']
[21:30:54] <seb_kuzminsky> mhaberler: ah, ok, maybe there's a missing "expect failure" file?
[21:31:05] <mhaberler> probably
[21:39:24] -!- Wildhoney has quit [Ping timeout: 252 seconds]
[21:41:50] -!- chopper79 has quit [Quit: Leaving.]
[21:43:45] -!- jpk has quit [Quit: Leaving.]
[21:50:05] -!- mackerski has quit [Quit: mackerski]
[21:56:55] <linuxcnc-build> build #877 of checkin is complete: Failure [4failed] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/checkin/builds/877 blamelist: Jeff Epler <jepler@unpythonic.net>
[21:58:34] -!- zzolo has quit [Quit: zzolo]
[22:08:48] <jepler> hmmm the test is not written properly for rt
[22:08:57] <jepler> it is expected that the symbol will not be available for linking
[22:09:30] <jepler> but the test is of halrun's stderr (?) and a very stupid grep
[22:09:41] <jepler> I'll have to revisit the test
[22:10:21] -!- skorasaurus has quit [Ping timeout: 276 seconds]
[22:10:25] <jepler> insmod: error inserting '/home/buildslave/emc2-buildbot/precise-rtai-x86/precise-i386-realtime-rip/build/rtlib/test_use.ko': -1 Unknown symbol in module
[22:11:14] -!- Loetmichel has quit [Ping timeout: 256 seconds]
[22:11:21] <jepler> seb_kuzminsky: red is stderr, black is stdout?
[22:20:58] -!- kjn has quit [Quit: Quit]
[22:21:56] -!- dway has quit [Quit: dway]
[22:26:17] -!- ravenlock has quit [Remote host closed the connection]
[22:28:21] -!- hjhart has quit [Ping timeout: 245 seconds]
[22:31:46] -!- mhaberler has quit [Quit: mhaberler]
[22:34:06] -!- jfire has quit [Read error: Connection reset by peer]
[22:36:34] -!- chopper79 has quit [Quit: Leaving.]
[22:38:54] -!- tmcw has quit [Remote host closed the connection]
[22:54:59] -!- motioncontrol has quit [Quit: Sto andando via]
[22:56:00] -!- coors has quit [Read error: Connection reset by peer]
[22:56:10] -!- coors [coors!coors@c-71-193-105-131.hsd1.in.comcast.net] has joined #linuxcnc-devel
[22:56:22] <seb_kuzminsky> jepler: yes
[22:56:34] -!- PCW_ has quit [Remote host closed the connection]
[23:05:03] zenek is now known as micges
[23:06:42] -!- sumpfralle has quit [Ping timeout: 264 seconds]
[23:17:56] -!- phillipadsmith_ has quit [Ping timeout: 240 seconds]
[23:47:12] -!- rob_h has quit [Ping timeout: 276 seconds]
[23:59:47] -!- hjhart has quit [Ping timeout: 258 seconds]
[23:59:52] -!- Nick001-Shop has quit [Quit: ChatZilla 0.9.90 [Firefox 19.0.2/20130307023931]]