#emc | Logs for 2011-10-08

[00:00:15] -!- Eartaker [Eartaker!~Eartaker@c-67-185-76-100.hsd1.wa.comcast.net] has joined #emc
[00:05:33] -!- ries [ries!~ries@] has joined #emc
[00:13:24] -!- factor [factor!~factor@] has joined #emc
[00:14:06] -!- Loetmichel has quit [Ping timeout: 258 seconds]
[00:16:13] -!- Guest12014 [Guest12014!~tom_itx@c-98-244-134-232.hsd1.ga.comcast.net] has joined #emc
[00:17:10] -!- robh__ has quit [Ping timeout: 260 seconds]
[00:19:08] -!- Loetmichel [Loetmichel!Cylly@] has joined #emc
[00:40:04] -!- Eartaker has quit [Read error: Connection reset by peer]
[00:40:28] -!- Eartaker [Eartaker!~Eartaker@c-67-185-76-100.hsd1.wa.comcast.net] has joined #emc
[00:42:13] -!- nullie has quit [Remote host closed the connection]
[00:46:14] -!- PCW has quit [Quit: ChatZilla 0.9.87 [Firefox 3.6.13/20101203075014]]
[00:47:06] -!- factor has quit [Read error: Connection reset by peer]
[00:52:22] -!- Guest12014 has quit []
[00:56:52] <elmo40> http://tromsite.com/
[00:57:59] -!- elmo40 has quit [Read error: Connection reset by peer]
[01:01:07] -!- cevad has quit [Quit: Leaving]
[01:02:01] -!- factor [factor!~factor@] has joined #emc
[01:05:13] <Jymmm> http://abcnews.go.com/Business/bank-america-moves-foreclose-florida-couple-paying-mortgage/story?id=14355558
[01:08:26] <andypugh> 70/78 retired couple, one terminally ill, owing $177,000 on a mortgage? How the heck is that suposed to work?
[01:09:29] <andypugh> They are paying the mortgage off at $1000 per month, so only 15 years to go, assuming zero intrest payments.
[01:16:38] <FinboySlick> andypugh: When interests are low enough and banks have run out of sane loans to give, they dump debt onto everything and sucker investment funds into buying it.
[01:17:54] <FinboySlick> In the US, when that comes appart, they also sucker the tax payer into buying it.
[01:19:56] <andypugh> Same in the UK. (In fact, the same dodgy US mortgages)
[01:20:30] <FinboySlick> It's the 'unforeseen' side effect of keeping interest rates low to stimulate the economy.
[01:20:38] <FinboySlick> Mr Greenspan, take note.
[01:27:14] -!- seb_kuzminsky [seb_kuzminsky!~seb@174-16-124-56.hlrn.qwest.net] has joined #emc
[01:28:14] <andypugh> is int **b; identical to in *b[] ?
[01:28:57] -!- ries has quit [Read error: Connection reset by peer]
[01:29:44] -!- ries [ries!~ries@] has joined #emc
[01:33:15] <FinboySlick> I'd assume it's equivalent.
[01:34:04] <FinboySlick> It'd be identical to *b[0] I think.
[01:34:42] <FinboySlick> Huh, sorry, you're declaring there, right?
[01:44:45] -!- cevad [cevad!~davec@host-174-45-229-40.bln-mt.client.bresnan.net] has joined #emc
[02:04:31] <andypugh> Header file declaration. Trying to figure out which part of what I changed made the whole machine bomb out.
[02:04:56] <cradek> pasetbin a diff?
[02:04:57] <cradek> e
[02:05:27] <andypugh> Can I diff between saves?
[02:06:08] -!- kljsdfhklj has quit [Ping timeout: 248 seconds]
[02:07:09] <cradek> umm
[02:07:25] <cradek> sometimes I'll save then undo all the way to the previous save, and write that to a new file
[02:07:42] <andypugh> Oddly, I was just looking at one of the very first programs I wrote. Back in 1983 I wrote a game that you can still play online in a simulator. Though I am not sure you would want to. I am not sure I have leaned much in the meantime. http://www.worldofspectrum.org/infoseek.cgi?regexp=%5EPrivateer$&pub=%5EMC+Lothlorien+Ltd$&loadpics=1
[02:07:58] <cradek> (whether you can do that depends on whether your editor is at least as capable as emacs...)
[02:08:15] <cradek> DO NOT STOP THE TAPE YET
[02:08:18] <cradek> whee :-)
[02:08:27] <andypugh> Is anything as capable as emacs?
[02:08:41] <cradek> in some ways, I'm sure
[02:08:57] <andypugh> (the Java cup icons take you to an online emulator)
[02:09:30] <cradek> Please upgrade to Netscape v4.0+ or Internet Explorer v4.0+
[02:09:44] <cradek> I bet your game worked better in '83...
[02:10:32] <andypugh> Arguably.
[02:10:33] -!- kljsdfhklj [kljsdfhklj!~ln@] has joined #emc
[02:12:03] <andypugh> I actually looked it up as part of an argument elsewhere about Steve Jobs being "a marketer who never invented anything". I suspect a lack of perspective on their part on just how rubbish computers used to be before Apple decided they should be easy to use.
[02:13:23] <cradek> http://stallman.org/archives/2011-jul-oct.html#06_October_2011_(Steve_Jobs)
[02:14:13] <andypugh> I still remember the magazine reviews of the Lisa. The journailsts seeme baffled as to why we would want a GUI, but I thought it was cool.
[02:15:36] <cradek> I think it's a bit revisionist to think apple was the "easy to use" force in the market. I remember the apple II and the C 64, and they were roughly equivalently difficult for an untrained person to use.
[02:16:42] <cradek> (of course it's also revisionist to think apple invented GUI/WIMP, word processing, spreadsheets, etc)
[02:17:55] <cradek> apple were/are the king of polish, and in more recent days, the king of using jails for profit
[02:17:56] <andypugh> Definitely. That is sort-of my point, it was Jobs who really pushed for the Lisa and Mac to have GUIs. And he got the idea from Xerox-PARC,but they lacked his megalomaniacal zeal so never tried to sell anything with one.
[02:18:20] <cradek> dammit, I wasn't going to argue with anyone about SJ. I could see it coming :-)
[02:18:50] <andypugh> I don't think we are arguing.
[02:18:53] <cradek> ... and it sounds like we agree anyway - terrible argument.
[02:19:00] <cradek> right
[02:19:32] <FinboySlick> I briefly thought I'd mac-fanboy my way into it to add some sport... But I'd fail miserably.
[02:19:50] <cradek> heh
[02:20:15] <andypugh> There' s a lot of Apple code in Ubuntu. They are a bit confused about Free.
[02:20:29] <FinboySlick> I have a lot of fun taunting my coworkers about their "little fetish objects that they need to stroke" or "testosterone sucking devices".
[02:20:41] <FinboySlick> Yes, that includes our CEO.
[02:21:05] <andypugh> I freely admit to being a Fanboi. I _will_ be having an iPhone 4S.
[02:21:34] <cradek> I have a G3 clamshell... my only apple product I think
[02:21:41] <FinboySlick> I'm a few generations behind. I can't even get myself to carry a mobile phone, nevermind it being smart.
[02:21:46] <cradek> it still works great, including the battery
[02:21:48] <andypugh> My Ubuntu machines were the first non-mac PCs I have bought since the Spectrum.
[02:23:01] <FinboySlick> The current terms of mobile phone ownership for me are 1- Work pays for it. 2- it stays on my office desk.
[02:23:01] -!- syyl_ has quit [Quit: Leaving]
[02:23:19] <cradek> I can't believe I'm still coherent, being up since 4am
[02:23:30] <andypugh> it's a bit wierd how the same folk who say "it's just a computer/phone/MP3 player, but prettier and twice the price" will also buy a BMW when a Kia would do the job.
[02:23:52] <andypugh> cradek: What time is it now?
[02:23:57] <cradek> I'm just as likely to buy a new car as a new computer - that is, not at all
[02:24:02] <cradek> 2124
[02:24:45] <andypugh> Well, was squashing bugs at 0230, up at 0700, and it's now 0324. Sleep is for the weak.
[02:25:10] <andypugh> (The last 3 hours have been spent inventing new bugs)
[02:25:34] <FinboySlick> andypugh: Working on EMC itself or some other project?
[02:25:54] <andypugh> I don't think I should be allowed to use languages without automatic memory mnagement.
[02:26:02] <andypugh> EMC2
[02:26:48] * FinboySlick bows his respects and offers the Joe Dirt encouragement: "Keep on keeping on."
[02:27:27] <andypugh> I am not writing many program lines, but I am playing with rather a complicated structure of dynamically allocated memory to support cards with unknown pin structures, and I am at the limits of my competence. It's fun though.
[02:28:32] <cradek> I hear you can teach a kid to swim by tossing him in the water and yelling "now, SWIM!"
[02:29:05] <andypugh> For a large enough sample of kids it will work for every survivor.
[02:29:25] <cradek> I'm impressed that you've tackled learning hal, emc, and C at the same time
[02:29:41] <FinboySlick> andypugh: I'm very inexperienced when it comes to coding any real projects but I also love pushing myself that way too. Nothing like OCD and way more time than I should spend on something.
[02:30:03] <cradek> it's unfortunate you're necessarily in kernel land while doing it; that just makes it harder.
[02:31:26] <andypugh> I am making it unnecessarily harder stlll by doing the coding on my Mac, and compiling through ssh to the Ubuntu machine. It does make for a lot of trips upstairs or to the workshop to reset the machine.
[02:32:04] <cradek> ouch
[02:32:06] <FinboySlick> andypugh: Doesn't the mac have some sort of sshfs? That could save you a bit of trouble.
[02:32:25] <FinboySlick> andypugh: Well, not with that...
[02:33:20] <andypugh> Have you managed to crash any kernel code recently? the dmesg output seems to have changed and I can't get code lines from the eip any more. I put the -g flag back in the makefile.
[02:34:01] <cradek> many moons ago when I was working on X, I had a serial terminal on the machine, so I could stop X in the debugger.
[02:34:29] <cradek> I believe you can attach a remote gdb (over serial maybe?) to a running kernel on linux, but I've never actually done it.
[02:35:06] <andypugh> If I were to study hard enough I think that I should be able to do master-slave (kgdb) debugging between the Mac and the Ubuntu machine.
[02:35:07] <cradek> andypugh: no, I haven't done it forever.
[02:35:42] <andypugh> We need to stop typing the same thing at the same time
[02:36:10] <cradek> if a shot of this really good bourbon would help, I'll share it with you if you come by.
[02:36:37] <andypugh> I have quite enough South African Red to hand :-)
[02:37:09] <cradek> sadly for my checkbook, I discovered a really great local record store
[02:37:28] -!- i_tarzan_ [i_tarzan_!~i_tarzan@] has joined #emc
[02:37:36] -!- i_tarzan has quit [Ping timeout: 255 seconds]
[02:37:59] <andypugh> records? Is that like iTunes?
[02:38:02] <andypugh> :-)
[02:38:24] <cradek> yeah except cheaper, and they don't all disappear if your computer crashes
[02:39:11] -!- Eartaker has quit [Read error: Connection reset by peer]
[02:39:28] -!- Eartaker [Eartaker!~Eartaker@c-67-185-76-100.hsd1.wa.comcast.net] has joined #emc
[02:39:34] <cradek> what's your crashy dmesg output look like now?
[02:39:41] <andypugh> I am not a great fan of iTunes the software, but I think it is possibly unfair to blame Apple for the licensing stuff, I think that was negotiated by the record companies, who really still don't get it.
[02:39:55] <cradek> haha "record companies"
[02:40:23] -!- FinboySlick has quit [Quit: Leaving.]
[02:40:52] <andypugh> "IP rights aggregators" I meant, of course.
[02:42:03] -!- pcw_home [pcw_home!~chatzilla@ip-66-80-167-54.sjc.megapath.net] has joined #emc
[02:42:22] <andypugh> http://pastebin.com/zHADhBjz
[02:42:42] <andypugh> (Let me restart, then try the addr2line thing.)
[02:42:48] <cradek> EIP is at hm2_sserial_auto_make_pins+0xc0/0x212
[02:43:02] <cradek> it's already deciphered for you...?
[02:43:30] <andypugh> Yes, I was expecting
[02:43:31] <andypugh> RTAPI: Task 1[c2800000]: Fault with vec=14, signo=11 ip=c93dc01a.
[02:44:10] <andypugh> Does that mean I can use addr2line on 0xC0?
[02:44:20] <cradek> I don't know what addr2line is
[02:44:35] <andypugh> Neither do I, really,
[02:44:41] <cradek> you can disassemble that function and see what instruction is c0 bytes in
[02:44:59] <cradek> ... somehow
[02:44:59] <andypugh> src/docs/rtfaults.txt
[02:45:31] <andypugh> Sorry, src/docs/src/rtfaults.doc
[02:45:42] <andypugh> No, not that either.
[02:45:44] <cradek> found it
[02:45:53] <cradek> docs/rtfaults.doc
[02:45:59] <cradek> docs/rtfaults.txt
[02:46:00] <cradek> jeez
[02:46:54] <andypugh> Alcohol, a marvelous invention if you aspire to a single-digit IQ :-)
[02:47:07] <andypugh> (Gets another glass)
[02:47:58] <cradek> your EIP is f9ddb29d, it shows it undecoded at the bottom
[02:53:30] <andypugh> Running again, I get an offset into hostmot.ko of 0x929d
[02:54:07] <andypugh> andypugh@mill:~/emc2-dev/src$ addr2line -e hostmot2.ko 0x929d
[02:54:07] <andypugh> ??:0
[02:54:50] <andypugh> (which is the code for Eh?)
[02:55:13] <cradek> what is /sys/module/hostmot2/sections/.text?
[02:55:49] <andypugh> andypugh@mill:~/emc2-dev/src$ cat /sys/module/hostmot2/sections/.text
[02:55:49] <andypugh> 0xf9c2d000
[02:56:06] <cradek> and what's the faulting EIP?
[02:56:41] <andypugh> [ 371.941348] EIP: [<f9c3629d>] hm2_sserial_auto_make_pins+0xc0/0x212 [hostmot2] SS:ESP 0068:eacddb74
[02:57:58] -!- Eartaker has quit [Quit: OOOOOO Whats that....]
[02:58:34] <andypugh> There is 112k of Hostmot2, so it is in that file
[02:58:48] <cradek> hm ok, I also get 0x929d
[02:59:29] <andypugh> Maybe my -g flag is in the wrong place?
[03:00:10] <cradek> do you have to be in the right directory for addr2line to find the hostmot2.ko?
[03:00:26] <andypugh> Maybe, but i was
[03:00:39] <cradek> oh right, they're dumped in the toplevel, sorry
[03:01:10] <andypugh> (which seems odd, by the way)
[03:01:18] -!- ries has quit [Quit: ries]
[03:01:42] <cradek> unforunately it's a "feature" of kbuild, which we must use
[03:01:48] <cradek> it's irritating
[03:02:50] <andypugh> Hmm, can you look at Makefile line 833?
[03:02:53] -!- Eartaker [Eartaker!~kvirc@c-67-185-76-100.hsd1.wa.comcast.net] has joined #emc
[03:02:56] <cradek> nm --debug-syms hostmot2.ko
[03:03:12] <cradek> which branch are you on?
[03:03:17] <andypugh> 2.5
[03:03:56] <andypugh> It looks like you only get the EXTRA CFLAGS if BUILD_SYS is sim.
[03:04:20] <cradek> oh did you add it there?
[03:05:19] <andypugh> That's where rtfaults said to, but that makes no sense now I actually look
[03:05:36] <cradek> see 618
[03:06:41] -!- Eartaker has quit [Client Quit]
[03:06:44] <cradek> just stick it under that big one, or add your -g on to the end
[03:06:47] <andypugh> Wierd. Something very odd with my search function, it was only finding one instance. Now it finds 14. I am confused by that
[03:08:14] -!- Eartaker [Eartaker!~kvirc@c-67-185-76-100.hsd1.wa.comcast.net] has joined #emc
[03:08:45] <andypugh> I thought it was a long way in.
[03:09:52] -!- ve7it has quit [Remote host closed the connection]
[03:09:58] <andypugh> It's nice when somebody points out the far-too-obvious.
[03:10:18] <cradek> does that fix it?
[03:10:35] <andypugh> Need to rebuild, but I expect so.
[03:10:58] <andypugh> Any reason not to have -g in the normal build?
[03:11:07] <cradek> it makes everything much bigger
[03:11:28] <andypugh> Ah, that's a good reason.
[03:11:33] <cradek> we don't strip anything, so there are symbols, which is often good enough
[03:11:45] <cradek> like how you got that nice backtrace
[03:11:48] -!- i_tarzan_ has quit [Ping timeout: 255 seconds]
[03:12:28] <andypugh> Pretty: http://www.ebay.com/itm/NICE-WOHLHAUPTER-1-2-BORING-HEAD-4MT-SHANK-UPA2-/360399346876
[03:13:13] <andypugh> There is a UPA1 too. I could have the whole family. (I have a 3 and 4)
[03:16:44] -!- i_tarzan [i_tarzan!~i_tarzan@] has joined #emc
[03:18:29] <Jymmm> andypugh and his new instant family!!!
[03:18:46] <andypugh> My boring family
[03:18:51] <Jymmm> HAHAHA
[03:18:56] <Jymmm> andypugh++
[03:19:57] <andypugh> I put in the -g flag in the right place, I changed hal_bit_t **bit_pins; tp hal_bit_t *bit_pins[]; and it doesn't crash any more.
[03:20:18] -!- FinboySlick [FinboySlick!~FinboySli@] has joined #emc
[03:20:22] <andypugh> So I need to change the code back to re-create the crash :-)
[03:25:07] <andypugh> The internet says that type_t **foo and type_t *foo[] are the same. But one crashes and one doesn't. Discuss.
[03:25:26] <cradek> are you sure the pattern you think you see is a pattern?
[03:26:00] <andypugh> 4 compiles, 100% correlation. 66% confidence?
[03:26:06] <cradek> hmm
[03:27:27] <andypugh> Neither work. But one gives: [ 919.813354] BUG: unable to handle kernel NULL pointer dereference at (null)
[03:27:50] <andypugh> The other gives: [ 431.054283] HAL: ERROR: pin_new(hm2_5i25.0.7i76.0.0.Input-00) called with already-initialized memory
[03:31:15] -!- Eartaker has quit [Quit: KVIrc 4.0.4 Insomnia http://www.kvirc.net/]
[03:31:33] -!- Eartaker [Eartaker!~Eartaker@c-67-185-76-100.hsd1.wa.comcast.net] has joined #emc
[03:33:28] <andypugh> The compiler flag worked.
[03:33:31] <andypugh> :-)
[03:33:36] <cradek> yay!
[03:33:42] <andypugh> andypugh@mill:~/emc2-dev/src$ addr2line -e hostmot2.ko 0x929D
[03:33:55] <andypugh> home/andypugh/emc2-dev/src/hal/drivers/mesa-hostmot2/mesa_autoconf.c:200
[03:34:26] <andypugh> But why **foo =/= *foo[] is interesting.
[03:45:34] -!- mhaberler [mhaberler!~mhaberler@] has joined #emc
[03:46:51] <andypugh> Night all
[03:46:59] <cradek> g'night andypugh
[03:47:01] <mhaberler> cu
[03:47:09] -!- andypugh [andypugh!~andy2@cpc2-basl1-0-0-cust1037.basl.cable.virginmedia.com] has parted #emc
[04:08:56] -!- kljsdfhklj has quit [Ping timeout: 260 seconds]
[04:15:30] -!- kljsdfhklj [kljsdfhklj!~ln@] has joined #emc
[04:22:59] -!- toastydeath [toastydeath!~toast@c-69-140-223-139.hsd1.md.comcast.net] has joined #emc
[04:25:46] -!- toastyde1th has quit [Ping timeout: 258 seconds]
[04:49:59] -!- elmo40 [elmo40!~Elmo40@CPE0017ee09ce6d-CM001bd7a89c28.cpe.net.cable.rogers.com] has joined #emc
[04:54:23] -!- Danimal_garage has quit [Read error: Connection reset by peer]
[04:54:47] -!- Danimal_garage [Danimal_garage!~kvirc@ip70-179-52-97.sd.sd.cox.net] has joined #emc
[04:56:33] -!- Eartaker has quit [Quit: OOOOOO Whats that....]
[04:57:20] -!- Eartaker [Eartaker!~kvirc@c-67-185-76-100.hsd1.wa.comcast.net] has joined #emc
[05:26:13] -!- theos has quit [Ping timeout: 245 seconds]
[05:27:39] -!- FinboySlick has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
[07:25:11] -!- Eartaker has quit [Quit: KVIrc 4.0.4 Insomnia http://www.kvirc.net/]
[07:58:06] -!- Calyp has quit [Ping timeout: 260 seconds]
[08:14:22] -!- WalterN has quit [Quit: Leaving]
[08:18:15] -!- adb [adb!~Moldovean@] has joined #emc
[08:29:03] -!- bootnecklad_ [bootnecklad_!~bootneckl@host-78-149-230-117.as13285.net] has joined #emc
[09:19:20] <alex_joni> yay ssh and irssi work on my new phone
[09:22:06] <awallin_> you didn't wait for a nokia n9? :)
[09:22:13] <Jymmm> Yeah, but can you make a phone call?
[09:24:32] <alex_joni> oddly i can
[09:24:55] <Jymmm> heh
[09:37:50] -!- robh__ [robh__!~robert@5ace7030.bb.sky.com] has joined #emc
[09:40:37] -!- theos [theos!~theos@unaffiliated/theos] has joined #emc
[09:41:04] -!- pjm has quit [Read error: Connection reset by peer]
[09:41:12] -!- pjm [pjm!~pjm@uhfsatcom.plus.com] has joined #emc
[09:59:42] -!- Valen [Valen!~Valen@2001:44b8:3199:9100:21e:8cff:fe63:55df] has joined #emc
[10:26:11] -!- robh__ has quit [Quit: Leaving]
[11:11:17] -!- jthornton has quit [Quit: ChatZilla 0.9.87 [Firefox 3.6.23/20110921065534]]
[11:11:38] -!- jthornton [jthornton!~chatzilla@216-41-156-49.semo.net] has joined #emc
[11:18:42] -!- kljsdfhklj has quit [Quit: kljsdfhklj]
[11:19:17] -!- kljsdfhklj [kljsdfhklj!~ln@] has joined #emc
[11:59:51] -!- ries [ries!~ries@] has joined #emc
[12:04:29] -!- robh__ [robh__!~robert@5ace7030.bb.sky.com] has joined #emc
[12:20:30] -!- Mjolinor [Mjolinor!~Mjolinor@cpc1-burn3-0-0-cust572.10-1.cable.virginmedia.com] has joined #emc
[12:21:33] <Mjolinor> help, emc is crashing on start. Was fine for the last few days since I installed it but today it jsut will not run
[12:21:58] <Mjolinor> reinstalled adn it is the same, nothign meaningful in emc_debug.txt
[12:23:19] <Mjolinor> I have " /usr/bin/emc line 654 4719 segmentation fault
[12:24:32] <Valen> are there any lines above that in the log?
[12:24:39] <Valen> /var/log/syslog perhaps?
[12:26:54] <Mjolinor> error 6 in r128_dri.so
[12:27:44] <Mjolinor> I iwll install xchat on it then I can paste thigns, this is a different computer
[12:29:58] -!- Mjolinor has quit [Quit: Leaving]
[12:30:50] -!- Mjolinor [Mjolinor!~jd@cpc1-burn3-0-0-cust572.10-1.cable.virginmedia.com] has joined #emc
[12:31:21] <Mjolinor> Oct 8 13:26:14 jd-HP kernel: [13348.784020] axis[5137]: segfault at 78 ip 00fde718 sp bfcf9488 error 6 in r128_dri.so[fbc000+244000]
[12:31:34] <Mjolinor> that is my error from syslog
[12:31:59] <Valen> sounds like a problem with your video card/drivers
[12:32:09] <Valen> http://www.google.com.au/search?gcx=w&sourceid=chrome&ie=UTF-8&q=error+6+in+r128_dri.so
[12:32:13] <Mjolinor> i was sort fo wondering that but couldnt see how
[12:32:56] <Mjolinor> cheers, I will gop and read and no doubt be back when I fail :)
[12:33:18] <Mjolinor> it is only emc that is not working which I wouldnt really expect if it were video drivers but I iwll give it a go
[12:33:28] <Valen> try glxgears
[12:33:47] <Mjolinor> I jsut did
[12:33:50] <Mjolinor> seg fault :)
[12:33:59] <Mjolinor> looks like video to me
[12:34:05] <Mjolinor> bloody thing
[12:34:26] <Valen> good luck
[12:34:34] <Valen> i have nfi about ATI drivers ;->
[12:35:14] <Mjolinor> I hate ati drovers, personal vendetta against me they have
[12:35:41] <Valen> you don't have on board video in any form?
[12:35:50] <Mjolinor> it is onboard
[12:35:55] <Valen> oh...
[12:36:00] <Valen> thats.... unfortunate ;->
[12:36:06] <Mjolinor> otherewise I wouldnt ahve an ati :)
[12:36:08] <Valen> you can try the vesa driver
[12:40:54] -!- Mjolinor has quit [Quit: Leaving]
[13:43:22] -!- syyl [syyl!~syyl@p4FD12DB5.dip.t-dialin.net] has joined #emc
[13:53:58] -!- robh__ has quit [Quit: Leaving]
[14:01:57] -!- Valen has quit [Quit: Leaving.]
[14:14:38] -!- Mjolinor [Mjolinor!~Mjolinor@cpc1-burn3-0-0-cust572.10-1.cable.virginmedia.com] has joined #emc
[14:34:19] -!- Mjolinor has quit [Quit: Leaving]
[15:10:18] -!- bootnecklad_ has quit [Read error: Connection reset by peer]
[15:10:42] -!- bootnecklad [bootnecklad!~bootneckl@host-78-149-230-117.as13285.net] has joined #emc
[15:20:23] -!- skunkworks [skunkworks!~chatzilla@184-158-81-121.dyn.centurytel.net] has joined #emc
[15:27:41] -!- nullie [nullie!~nullie@nullie.telenet.ru] has joined #emc
[15:31:15] -!- syyl_ [syyl_!~syyl@] has joined #emc
[15:33:49] -!- syyl has quit [Ping timeout: 240 seconds]
[15:55:25] -!- mhaberler has quit [Quit: mhaberler]
[16:01:18] -!- vladimirek [vladimirek!~vladimire@bband-dyn243.178-40-47.t-com.sk] has joined #emc
[16:08:56] -!- tom3p [tom3p!~tomp@74-93-88-241-Illinois.hfc.comcastbusiness.net] has joined #emc
[16:16:49] -!- isssy [isssy!EMCIRCUser@landmark.tea.bg] has joined #emc
[16:48:06] <skunkworks> chirp chirp
[16:48:27] skunkworks is now known as skunkKandT
[16:55:18] -!- nullie has quit [Quit: Ex-Chat]
[16:56:35] -!- ve7it [ve7it!~LawrenceG@S0106001c10b7770f.pk.shawcable.net] has joined #emc
[17:04:56] <syyl_> *tumbleweed rolls trough the channel*
[17:07:13] -!- isssy has quit [Ping timeout: 252 seconds]
[17:09:07] -!- pjm__ [pjm__!~pjm@uhfsatcom.plus.com] has joined #emc
[17:09:09] -!- pjm__ has quit [Client Quit]
[17:09:56] -!- IchGuckLive [IchGuckLive!~chatzilla@95-89-108-110-dynip.superkabel.de] has joined #emc
[17:10:01] <IchGuckLive> Hi all
[17:10:25] -!- syyl_ has quit [Ping timeout: 260 seconds]
[17:10:30] -!- syyl [syyl!~syyl@p4FD15E41.dip.t-dialin.net] has joined #emc
[17:12:13] -!- pjm has quit [Ping timeout: 240 seconds]
[17:18:08] -!- isssy [isssy!EMCIRCUser@landmark.tea.bg] has joined #emc
[17:28:31] -!- pjm [pjm!~pjm@uhfsatcom.plus.com] has joined #emc
[17:31:39] -!- nullie [nullie!~nullie@nullie.telenet.ru] has joined #emc
[17:36:29] -!- jv4779 [jv4779!62c90637@gateway/web/freenode/ip.] has joined #emc
[17:38:53] <jv4779> I have been using Axis "run from here" to skip around multi-part gcode files. I can see that emc2 tracks the G61/G64 mode and if I run from a part that is inside a G61 it will swithc to G61. I stared trying to use G64 P0.001 and G64 P0.1 in different parts and "run from here" doesn't seem to get the P value correct.
[17:42:20] <IchGuckLive> G64 is not modal
[17:42:45] <IchGuckLive> G61 is in the system fixet
[17:46:25] <jv4779> I had some gcode taht was set to G61, I tried to stop it, use MDI to set G64, then do a run from here. that reverted to G61. this was caused by run from here using some system default ?
[17:46:59] <IchGuckLive> yes G61 is default
[17:47:43] <jv4779> so run from here is just incompatable with setting G64 ? it will always run with G61 until it would hit the G63 again in the gcode ?
[17:48:09] <jv4779> sorry, G63=G64
[17:48:13] <IchGuckLive> yes
[17:49:04] <IchGuckLive> you can stop the programm and give it a G64 at the stop value and then run from here remind ther are 3 lines reading back
[17:49:05] <jv4779> do things like G20/21 stick ?
[17:49:56] <jv4779> 3 lines reading back ? not sure what that means
[17:51:24] <IchGuckLive> if you start at N10 axis will look at N7 and maybe run to this point first
[17:52:19] <IchGuckLive> the default is actualy G64 P0
[17:53:24] <jv4779> I wasn't clear on if P0 mean't no extra tolerance like G61 or default G64 tolerance
[17:53:57] <IchGuckLive> jv4779: http://www.linuxcnc.org/docview/html/common_User_Concepts.html#r1_2_1
[17:54:44] <IchGuckLive> default G64 without P is the G2 error limit
[17:55:35] <jv4779> This is all coming up because last night I was cutting a dxf with tons of short segment polylines and doing G61 was taking forver. I was playing with my CAM to do a roughing pass with G64 P0.1 and giving it 0.1 roughing clearance, then a finish pass.
[17:56:18] <jv4779> doing it this way was then making "run form here" behave weird
[17:56:57] <IchGuckLive> agree
[17:57:36] <IchGuckLive> but as i said stop the programm ged it edit with the G64 P0.1 at the new start and restart from there
[17:58:04] <jv4779> as long as I start my run from here at a G64 line it will work better
[17:58:36] <IchGuckLive> yes you may need also a z safe and a z depth word
[17:58:47] <IchGuckLive> at the new restart
[17:59:22] <jv4779> I would raise the spindle, then start at the G0 leading to the new pocket, that saved running the cutter across the work
[17:59:49] <IchGuckLive> a good post can also give you a restart point every 50 lines 100 liines or whatever
[18:01:11] <IchGuckLive> as you raise the spindel give the G64 to the line bevore G0
[18:02:20] <jv4779> what would a good restart point look like in the gcode ?
[18:02:24] -!- tom3p [tom3p!~tomp@74-93-88-241-Illinois.hfc.comcastbusiness.net] has parted #emc
[18:02:36] <IchGuckLive> jv4779: there is no devel here thay woudt no where to change this
[18:03:33] <jv4779> I am not really trying to get anything changed, just understand how best to use "run from here"
[18:04:24] <IchGuckLive> agree
[18:05:05] <IchGuckLive> i did also charshed some parts with older versions as the G64 wars set to higher values
[18:05:34] <IchGuckLive> heekscad is abel to get a restart point on every sketch
[18:11:35] -!- pjm has quit [Quit: TTFO]
[18:21:44] <jv4779> I did a little testing and having G61 in the start of the gcode, will put it in G61 when a run from here is done no matter if G64 is active in the MDI or not. The opposite also happens. I am guessing that it is only the P values of G64 that are not tracked and restored on a run form here
[18:22:02] -!- isssy has quit [Ping timeout: 252 seconds]
[18:22:45] <jv4779> is there anyway to query what the current P value of an active G64 is ? MDI just shows G64
[18:27:07] -!- pjm [pjm!~pjm@uhfsatcom.plus.com] has joined #emc
[18:29:40] <IchGuckLive> 0.0002 inch or 0.002 millimeter
[18:30:03] -!- IchGuckLive has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.22/20110905191234]]
[18:34:44] <jv4779> Thank you very much for the help.
[18:34:54] -!- maximilian_h [maximilian_h!~bonsai@f051149030.adsl.alicedsl.de] has joined #emc
[18:41:44] -!- isssy [isssy!EMCIRCUser@landmark.tea.bg] has joined #emc
[18:53:22] -!- maximilian_h [maximilian_h!~bonsai@f051149030.adsl.alicedsl.de] has parted #emc
[18:58:42] -!- vladimirek has quit [Remote host closed the connection]
[18:59:38] -!- andypugh [andypugh!~andy2@cpc2-basl1-0-0-cust1037.basl.cable.virginmedia.com] has joined #emc
[19:34:10] -!- PasstScho [PasstScho!~PasstScho@dslb-188-104-124-196.pools.arcor-ip.net] has joined #emc
[19:34:19] <PasstScho> hi
[19:34:48] <PasstScho> i was playing around with sub-procs today and maybe i didn't get the syntax right
[19:35:05] <PasstScho> how should a do-while loop with break look like?
[19:35:09] <PasstScho> o100 do
[19:35:13] <PasstScho> o100 break
[19:35:21] <PasstScho> o100 while [1]
[19:36:16] <PasstScho> would this be a correct loop (useless, of course...)? or is break intended to be used another way? i don't really understand why these o100 (or whatever names) are needed
[19:40:43] <nullie> http://linuxcnc.org/docs/html/gcode_main.html#r3_2 - have you read this?
[19:41:23] <nullie> labels are needed to mark a loop, because there can be nested loops
[19:41:44] <PasstScho> well i've read parts of it ;)
[19:42:53] <nullie> and you loop looks ok to me
[19:42:54] <PasstScho> ok, so all do/while/if/whatever "commands" should always have the same name/number if they belong together
[19:43:10] <PasstScho> so the 3-liner a few lines above would be correct...
[19:44:03] <PasstScho> i don't like the word bug... lets say i have some strange behaviour ;)
[19:44:04] <PasstScho> http://nopaste.info/d44eb82412.html
[19:45:03] <PasstScho> break in the while loop works as i am used to from C and every other language... but break in the do-while does strange stuff... it is more like continue
[19:45:08] <nullie> labels don't match on second loop
[19:45:20] <PasstScho> hm well i tried around a bit
[19:45:27] <PasstScho> it doesn't work when they match ;)
[19:45:30] <nullie> you have o102 in the beginning and o103 elsewhere
[19:45:41] <PasstScho> just cant start emc atm cuz of problems with vmware :/
[19:45:58] <PasstScho> just imagine those labels would match ;)
[19:46:10] <nullie> then it should print first statement of second loop
[19:46:23] <PasstScho> and then it should print endsub
[19:46:27] <PasstScho> but try it ;)
[19:46:38] <PasstScho> endless "start of do-while"....
[19:46:56] -!- slowfuse has quit [Ping timeout: 248 seconds]
[19:46:57] <PasstScho> wait a second... ill correct it again
[19:47:06] <nullie> it should be because you have wrong label on break
[19:47:15] <nullie> hm
[19:47:37] <PasstScho> this should be how i tried it http://nopaste.info/512cb05657.html
[19:47:41] <PasstScho> forget the first link
[19:48:24] -!- slowfuse [slowfuse!~slowfuse@slowfuse.com] has joined #emc
[19:50:12] <nullie> someone should write online g code interpreter
[19:50:15] -!- PasstScho2 [PasstScho2!~checker@dslb-188-104-124-196.pools.arcor-ip.net] has joined #emc
[19:50:32] -!- PasstScho has quit [Remote host closed the connection]
[19:50:53] <PasstScho2> doing a reboot of the vm... i hope emc will at least start again :/
[19:51:35] <nullie> you ran emc2 in a vm?
[19:52:32] <PasstScho2> yep... but just for programming
[19:52:42] <PasstScho2> the interpreter still should work right ;)
[19:58:00] <PasstScho2> so, any ideas?
[19:58:19] <nullie> nope
[19:58:32] <PasstScho2> hm
[19:58:43] <PasstScho2> do i get a cookie if it is a bug? ;)
[19:59:46] <nullie> dunno, emc2 development community looks alienate to me
[19:59:47] <Jymmm> s/cookie/kick in the ass/
[19:59:53] <nullie> yeah
[20:00:22] <PasstScho2> hehe
[20:04:04] PasstScho2 is now known as PasstScho
[20:17:59] <andypugh> I didn't even know there was a "break"
[20:19:11] <PasstScho> as long as you are not the developer of it... ;)
[20:20:37] -!- bootnecklad` [bootnecklad`!~bootneckl@host-78-151-83-252.as13285.net] has joined #emc
[20:23:01] -!- bootnecklad has quit [Ping timeout: 240 seconds]
[20:24:04] <andypugh> No, but I do occasonally look at the source code.
[20:24:52] <skunkKandT> developers usually don't remember what they did...
[20:24:54] <skunkKandT> ;)
[20:25:12] <PasstScho> i know that problem too well ;)
[20:25:51] -!- kljsdfhklj has quit [Ping timeout: 248 seconds]
[20:30:07] <Jymmm> skunkKandT: YOU CAN PANDUIT!
[20:30:32] <andypugh> PasstScho: I think you found a bug.
[20:31:02] <skunkKandT> Jymmm: still running great!
[20:31:23] -!- kljsdfhklj [kljsdfhklj!~ln@] has joined #emc
[20:31:25] <Jymmm> skunkKandT: LIES! All lies!
[20:31:37] <andypugh> PasstScho: Do you read C++?
[20:31:53] <PasstScho> ?
[20:32:07] <PasstScho> well i understand c & c++
[20:32:11] <PasstScho> (usually ;))
[20:32:15] <andypugh> That was what I meant.
[20:33:20] <andypugh> It all happens from line 646 onwards at http://git.linuxcnc.org/gitweb?p=emc2.git;a=blob;f=src/emc/rs274ngc/interp_o_word.cc;h=866516b8b531a9e3dbfe4737f99d534a5f4d8c34;hb=a4f10e63e9d4f4c0155dc400d24237af3acac199
[20:34:37] <andypugh> "break" sets the mode to "skipping" , but "while" always resets the mode. Basically it looks like it can't tell the difference between a do-while and a while-endwhile when it gets to the while
[20:37:03] -!- bootnecklad` has quit [Read error: Connection reset by peer]
[20:37:06] <PasstScho> ah ok :)
[20:38:14] <andypugh> You probably ought o report it to the bug tracker: http://sourceforge.net/tracker/?func=add&group_id=6744&atid=106744
[20:39:02] <PasstScho> can someone pls give me my link again? i am kinda fighting with my vm atm... ;)
[20:39:45] -!- kljsdfhklj has quit [Ping timeout: 260 seconds]
[20:39:53] <skunkKandT> logger[ps
[20:39:56] <skunkKandT> logger[psha]:
[20:40:03] <PasstScho> ty :)
[20:40:18] <skunkKandT> give a person a fish...
[20:40:45] <PasstScho> i just need cookies :P
[20:45:24] -!- kljsdfhklj [kljsdfhklj!~ln@] has joined #emc
[20:46:08] <PasstScho> bug reported
[20:46:16] <PasstScho> thanks andypugh
[20:55:05] <PasstScho> time for a reboot :/
[20:55:12] <PasstScho> brb
[20:55:15] -!- PasstScho has quit []
[21:07:06] -!- PasstScho [PasstScho!~checker@dslb-188-104-124-196.pools.arcor-ip.net] has joined #emc
[21:08:00] -!- ries has quit [Ping timeout: 248 seconds]
[21:16:40] <skunkKandT> damn.. Cnc is nice. everything lines up.. go figure
[21:17:21] -!- ries [ries!~ries@] has joined #emc
[21:25:24] -!- elmo40 has quit [Quit: Leaving.]
[21:39:39] -!- bootnecklad [bootnecklad!~bootneckl@host-78-151-83-252.as13285.net] has joined #emc
[21:46:36] -!- nullie has quit [Quit: Ex-Chat]
[21:57:29] -!- Eartaker [Eartaker!~Eartaker@c-67-185-76-100.hsd1.wa.comcast.net] has joined #emc
[22:10:54] -!- seb_kuzminsky has quit [Remote host closed the connection]
[22:11:34] -!- seb_kuzminsky [seb_kuzminsky!~seb@174-16-124-56.hlrn.qwest.net] has joined #emc
[22:30:25] -!- robh__ [robh__!~robert@5ace7030.bb.sky.com] has joined #emc
[22:41:16] -!- isssy has quit [Quit: Visitor from www.linuxcnc.org]
[22:44:12] -!- isssy [isssy!EMCIRCUser@landmark.tea.bg] has joined #emc
[22:44:13] -!- isssy has quit [Client Quit]
[22:58:31] -!- SWPadnos_ [SWPadnos_!~Me@74-92-8-208-NewEngland.hfc.comcastbusiness.net] has joined #emc
[23:02:08] -!- SWPadnos has quit [Ping timeout: 248 seconds]
[23:02:09] SWPadnos_ is now known as SWPadnos
[23:05:13] -!- bootnecklad has quit [Ping timeout: 255 seconds]
[23:06:12] * Jymmm lol @ skunkKandT
[23:08:26] -!- chester88 has quit [Quit: Leaving.]
[23:14:47] <cradek> skunkKandT: I wish cnc was all it took
[23:17:01] <skunkKandT> cradek: well - a well built cnc and a lucky programmer.. ;)
[23:21:19] <andypugh> I just hooked up 5 sslbp boards to my 5i23 and 7i44 . That made a _long_ list of pins.
[23:23:05] <skunkKandT> heh
[23:24:50] <andypugh> (calculates) 136 lines of IO.
[23:25:21] <alex4nder> hey
[23:25:37] <cradek> skunkKandT: one time I carefully programmed everything, and got a perfect part that was the mirror image of what I needed.
[23:26:59] -!- JT-Shop has quit [Remote host closed the connection]
[23:27:35] -!- JT-Shop [JT-Shop!~chatzilla@216-41-156-49.semo.net] has joined #emc
[23:27:48] <PasstScho> btw, short question about modbus... is the modbus communication started in a rt-thread? or in other words... would it already be possible to use modbus for time-critical communication?
[23:28:08] <cradek> no and no
[23:28:14] <PasstScho> damn ;)
[23:28:29] <cradek> modbus is very slow - I don't think it's good for anything time critical
[23:28:44] <skunkKandT> cradek: heh - oops
[23:28:45] <cradek> its strength is that you can control lots of hardware with a simple interface on the pc
[23:28:49] <PasstScho> well modbus isn't limited to any baudrate
[23:28:58] <andypugh> alex4nder just reminded me of last weekend, when we were touring round the Essex countryside with my parents. They live a few hundred miles away, where the history is a bit different (Norse place names, not Saxon). Local to them there are two houses called "Mean Hey" and "Upper Hey". My mum saw a sign by a farm gate saying "Bunny Hay" and asked if that was really the place name. :-)
[23:29:07] <skunkKandT> the question is - how do I get this bearing out I just tapped in to the hole I milled..
[23:29:08] <cradek> even at the very fastest serial baud rates, you can't get much done in a millisecond
[23:29:45] <PasstScho> cheapest rs232 cards already do 1 mbit
[23:29:55] <JT-Shop> Sam heat?
[23:30:10] <andypugh> skunkKandT: Bearing into a blind hole?
[23:30:16] <PasstScho> it would at least be enought for 6 axis position data
[23:30:24] <andypugh> Sealed bearing or open?
[23:30:24] <skunkKandT> well - in a vise... sorta blind
[23:30:49] <PasstScho> and a 4-channel rs232 card is 19 ... so 4 mbit ;)
[23:30:59] <cradek> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Emc2HardwareDesign#RS232_Serial
[23:31:05] <skunkKandT> actually the bearing is a used one - so no great shakes
[23:31:26] <syyl> do you still need the bearing?
[23:31:29] <andypugh> One good tip for a sealed bearing is to fill the hole with oil, insert a drift of the right size to be a nice fit in the bore, then hot the drift with a hammer. The hydraulic pressure pushes the bearing out (Also woks for bushes)
[23:31:49] <cradek> I don't even know what the modbus overhead is, so I can't do the math
[23:31:50] <andypugh> (hit the drift with a hammer)
[23:31:52] <JT-Shop> kinda hard on seals but might work
[23:31:52] <syyl> destroy the cage, pull out the balls, cage and inner ring
[23:32:01] <syyl> then putt a few welding beads in the outer ring
[23:32:02] <cradek> is it academic or are you trying to do a specific thing?
[23:32:13] <syyl> the ring shrinks while the welding beads cool down
[23:32:25] <syyl> ring fall out
[23:32:51] <PasstScho> cradek: 11 bytes overhead + useful data
[23:32:51] <skunkKandT> got it - dad has more patience
[23:32:55] <cpresser> neat trick, syyl. ill remember it :)
[23:32:58] <JT-Shop> I've done similar to that syyl
[23:33:06] <JT-Shop> and it works :)
[23:33:23] <andypugh> If you want to connect IO and drives with Ethernet cable I am rather liking the Mesa format at the moment. (that's 2.5Mbit, but the comms is done by the FPGA)
[23:33:28] <PasstScho> would be 35 bytes for 6 axes with 32bit each
[23:33:38] <syyl> used it once for a total stuck sitting morsetaper sleve..
[23:33:42] <syyl> even there it worked
[23:34:28] <andypugh> bead of weld is a common solution to getting the bottom head race out of a motorbike
[23:36:32] <PasstScho> the idea of using cheapo hardware for servo stuff is just too cool to drop it ;)
[23:37:16] <andypugh> How cheap?
[23:37:26] -!- ve7it has quit [Remote host closed the connection]
[23:37:42] <PasstScho> 19 for 4 x 1mbit
[23:38:13] <PasstScho> http://www.ebay.de/itm/150360546140 <== like that one
[23:38:22] <andypugh> That's cheap in UKP, so must be even cheaper in other currencies.
[23:38:33] <PasstScho> yep
[23:38:59] <PasstScho> i already have some closed-loop-stepper-dreams with those cards ;)
[23:39:02] <andypugh> But how much for the stuff that they drive?
[23:39:33] <PasstScho> i'd like to build a modbus -> stepdir device
[23:39:42] <skunkKandT> yeck
[23:39:48] <PasstScho> BUT
[23:39:56] <skunkKandT> sorry - did i say that out loud?
[23:40:06] <PasstScho> too early :P
[23:40:08] <andypugh> They wouldn't use the existing Modbus driver in EMC2, I think that uses the serial port.
[23:40:22] <PasstScho> using it just as a "frequency generator" output
[23:40:46] <PasstScho> yep... it won't work with the current modbus module :-(
[23:40:49] <andypugh> You would need a realtime PCI HAL driver. Which actually makes things rather better.
[23:41:35] <PasstScho> building the modbus device with glass-scale interface is not really a problem - but i have no clue where to start making such an emc module ;)
[23:42:15] <andypugh> Writing drivers for EMC2 is pretty easy, if you have the register layout. I have knocked one up in an afternoon for hardware I never even saw. (that was ISA, but the principle is the same)
[23:42:35] <PasstScho> hmm :)
[23:42:49] <PasstScho> but i guess you have basic linux knowledge? ;)
[23:43:10] <PasstScho> i only use linux for my cnc...
[23:45:36] <andypugh> When I started my CNC machine I had never used Linux, or C or HAL. It is possible to learn.
[23:45:57] <PasstScho> well i know it is possible to learn :)
[23:46:10] <PasstScho> i just don't have too much time for everything :-/
[23:46:16] <andypugh> This thead (which rambles a bit, so is rather long) is the process that led to me writing that driver I mentioned. http://www.linuxcnc.org/component/option,com_kunena/Itemid,20/func,view/id,10910/catid,38/limit,6/limitstart,0/lang,english/
[23:46:41] <PasstScho> ah thanks
[23:46:59] <PasstScho> thats gonna take some days for me to read ;)
[23:47:12] <PasstScho> but hm maybe its worth :>
[23:49:11] -!- micges has quit [Quit: Ex-Chat]
[23:53:05] -!- SWPadnos has quit [Changing host]
[23:53:05] -!- SWPadnos [SWPadnos!~Me@emc/developer/SWPadnos] has joined #emc
[23:59:54] <andypugh> Night all