#emc-devel | Logs for 2011-03-22

Back
[00:04:14] -!- theorbtwo has quit [Ping timeout: 246 seconds]
[00:04:17] theorb is now known as theorbtwo
[00:14:46] -!- robh__ has quit [Ping timeout: 240 seconds]
[00:20:56] -!- OoBIGeye has quit [Ping timeout: 276 seconds]
[00:23:28] -!- mhaberler has quit [Quit: mhaberler]
[00:30:06] -!- OoBIGeye has quit [Ping timeout: 260 seconds]
[00:34:54] -!- findux has quit [Client Quit]
[00:36:36] -!- mozmck1 has quit [Ping timeout: 250 seconds]
[00:51:29] -!- mozmck [mozmck!~moses@client-173.225.233.219.dfwtx.partnershipbroadband.com] has joined #emc-devel
[00:55:19] -!- Tom_shop has quit []
[00:58:38] -!- OoBIGeye has quit [Ping timeout: 276 seconds]
[01:05:44] -!- crazy_imp has quit [Read error: Operation timed out]
[01:05:52] -!- tom3p [[email protected]] has joined #emc-devel
[01:24:21] -!- mikeggg has quit [Ping timeout: 250 seconds]
[01:39:41] -!- skunkworks [skunkworks!~chatzilla@str-bb-cable-south2-static-6-412.dsl.airstreamcomm.net] has joined #emc-devel
[01:50:17] <andypugh> Night all.
[01:50:31] -!- andypugh has quit [Quit: andypugh]
[01:53:29] -!- tlab has quit [Quit: Leaving]
[01:54:32] -!- OoBIGeye has quit [Ping timeout: 276 seconds]
[01:56:46] -!- Dallur has quit [Read error: Connection reset by peer]
[01:59:56] -!- OoBIGeye has quit [Ping timeout: 260 seconds]
[02:29:38] -!- OoBIGeye has quit [Ping timeout: 276 seconds]
[02:31:11] <jepler> night all
[02:31:28] <cradek> g'night
[02:42:06] -!- tom3p [[email protected]] has parted #emc-devel
[02:42:31] -!- SteveStallings [[email protected]] has joined #emc-devel
[02:43:39] -!- mikeggg has quit [Ping timeout: 250 seconds]
[02:45:44] SteveStallings is now known as steves_logging
[02:49:13] -!- rooks has quit [Quit: So long, and thanks for all the fish.]
[02:51:44] -!- OoBIGeye has quit [Ping timeout: 276 seconds]
[03:39:57] -!- Valen has quit [Quit: Leaving.]
[03:45:09] -!- Tom_shop has quit []
[03:50:53] -!- OoBIGeye has quit [Ping timeout: 276 seconds]
[04:05:09] X704 is now known as WalterN
[04:05:33] -!- jv4779 [jv4779!62c90637@gateway/web/freenode/ip.98.201.6.55] has parted #emc-devel
[04:08:16] -!- OoBIGeye has quit [Ping timeout: 260 seconds]
[05:08:19] -!- kb8wmc has quit [Ping timeout: 240 seconds]
[05:17:00] -!- Valen has quit [Quit: Leaving.]
[05:23:18] -!- mhaberler [[email protected]] has joined #emc-devel
[05:39:44] -!- swamidog_ has quit [Quit: Leaving]
[06:30:04] -!- ve7it has quit [Remote host closed the connection]
[07:47:34] -!- robh__ [[email protected]] has joined #emc-devel
[07:59:03] -!- The_Ball has quit [Remote host closed the connection]
[08:27:16] -!- cradek has quit [Ping timeout: 260 seconds]
[08:28:21] -!- cradek [[email protected]] has joined #emc-devel
[08:31:06] -!- Birdman3131 has quit [Ping timeout: 250 seconds]
[08:38:01] -!- cradek has quit [Ping timeout: 246 seconds]
[08:39:40] -!- cradek [[email protected]] has joined #emc-devel
[09:10:22] -!- psha[work] [psha[work][email protected]] has joined #emc-devel
[09:41:14] -!- Valen has quit [Remote host closed the connection]
[09:42:48] -!- alex_chally has quit [Ping timeout: 240 seconds]
[09:51:33] <mhaberler> psha: hi!
[09:54:03] -!- danimal_garage has quit [Ping timeout: 248 seconds]
[09:57:50] <psha[work]> hi
[10:19:17] -!- awallin_ [awallin_!~quassel@2001:708:110:1020:224:7eff:feda:7c7d] has joined #emc-devel
[10:29:27] -!- Techrat has quit [Quit: Leaving]
[11:16:46] -!- mikeggg has quit [Ping timeout: 240 seconds]
[11:51:23] -!- skunkworks has quit [Ping timeout: 248 seconds]
[12:07:23] -!- mhaberler has quit [Quit: mhaberler]
[12:11:40] -!- mhaberler [[email protected]] has joined #emc-devel
[12:20:20] -!- Poincare has quit [Read error: Operation timed out]
[12:23:09] -!- mhaberler has quit [Quit: mhaberler]
[12:43:47] -!- JT-Shop has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.15/20110303024726]]
[12:44:26] -!- Valen has quit [Quit: Leaving.]
[12:49:15] -!- mhaberler [[email protected]] has joined #emc-devel
[12:49:58] -!- skunkworks [skunkworks!447329d2@gateway/web/freenode/ip.68.115.41.210] has joined #emc-devel
[13:17:24] -!- Poincare has quit [Quit: changing servers]
[13:28:29] Tech_Talk is now known as Birdman3131
[13:37:16] -!- cjdavis has quit [Read error: Connection reset by peer]
[13:43:11] -!- alex_joni has quit [Ping timeout: 250 seconds]
[13:43:29] -!- alex_joni [alex_joni!~alex_joni@emc/board-of-directors/alexjoni] has joined #emc-devel
[13:55:38] -!- Poincare has quit [Quit: changing servers]
[14:41:21] -!- tris has quit [Excess Flood]
[14:47:47] Dreiselfift is now known as pcw_home
[14:55:59] -!- KimK has quit [Ping timeout: 250 seconds]
[14:57:23] -!- toastydeath has quit [Ping timeout: 264 seconds]
[15:03:38] -!- psha[work] has quit [Quit: Lost terminal]
[15:09:12] -!- KimK [[email protected]] has joined #emc-devel
[15:30:41] -!- kb8wmc [[email protected]] has joined #emc-devel
[15:31:40] -!- psha has quit [Client Quit]
[15:33:50] -!- psha [[email protected]] has joined #emc-devel
[15:46:19] <mhaberler> JT-work: here's a preview of the revised gcode/overview section: http://emc.mah.priv.at/docs/html/gcode/overview.html - all parameter related info is there, and I think the conversion issues are all fixed
[15:48:22] <skunkworks> I am in awe of people that can write manuals in a non native lanquage.
[15:48:50] <mhaberler> this is NOT ready for commit because Pavel or myselves need to get in the source-highlight changes which are needed for things like this (note GCode highlighting): http://emc.mah.priv.at/docs/html/gcode/overview.html#EXISTS-Function
[15:49:00] -!- ve7it [[email protected]] has joined #emc-devel
[15:49:28] <mhaberler> thanks ;-) my father started feeding me with Texsas Instruments manuals before high school ;-)
[15:50:40] <skunkworks> heh - great dad.
[15:50:48] <archivist> I have noticed how well german speakers type english, better than natives
[15:51:00] <mhaberler> this source-highlight stuff will enable proper highlighting of halcmd, ini files and ngc files - unfortunately for html only at this point
[15:51:36] <mhaberler> note that we are separated from Germany by the same language ;-)
[15:51:56] <archivist> as I am from the americans :)
[15:52:21] <mhaberler> yeah, we share that issue with Candaians and Kiwis
[15:53:17] <mhaberler> oversized, overly funny neighbours
[15:53:47] -!- nullie has quit [Quit: Ex-Chat]
[15:55:14] <archivist> we should re assert control of our empire :)
[15:55:43] <mhaberler> forget it, them colonies are loose
[15:55:57] <archivist> loose canons
[15:58:36] -!- awallin_ has quit [Remote host closed the connection]
[16:04:59] -!- pcw_home [[email protected]] has joined #emc-devel
[16:09:48] -!- psha has quit [Quit: -a-]
[16:10:05] -!- psha [[email protected]] has joined #emc-devel
[16:12:44] <mhaberler> I forgot the Irish
[16:23:14] -!- psha has quit [Quit: -a-]
[16:54:16] -!- ries has quit [Read error: Connection reset by peer]
[16:54:22] -!- ries [[email protected]] has joined #emc-devel
[16:55:28] <cradek> mhaberler: the parameter file format stuff is wrong - there are no headers or blank lines. I don't know whether comments are accepted. If so, I doubt they are preserved.
[16:56:45] <mhaberler> I can imagine; this is old stuff copied over from 'general machining center info' - I'll see what the code actually does and use that. Thanks!
[16:58:55] <mhaberler> cradek: could you do me a favor and check in the source-highlight changes so th revised files will build?
[16:59:19] <cradek> what branch do I fetch for that?
[16:59:39] <cradek> I haven't kept up on any of the highlighting stuff
[17:00:01] <mhaberler> ok, the branch is http://git.mah.priv.at/gitweb/emc2-dev.git/shortlog/refs/heads/asciidoc-highlighting
[17:00:13] <mhaberler> please dont merge right away... the issue is:
[17:00:30] <cradek> bbl, lunch, I will read back
[17:00:43] <mhaberler> bon apettite!
[17:02:22] <mhaberler> cradek: pls read the first 13 lines of http://git.mah.priv.at/gitweb/emc2-dev.git/blob/7b2bf699daa7a497c0ff48cbef136d2baeb88aee:/share/source-highlight/README - this describes what needs to be done: copy files, plus edit the existing lang.map file
[17:03:06] <mhaberler> I was unsure how to massage this into the build/install process
[17:03:42] <mhaberler> btw this needs to be done even if run-in-place because source-highlight is inflexible wrt pathnames
[17:39:39] <mhaberler> parameter file format updated in http://emc.mah.priv.at/docs/html/gcode/overview.html#cap:Parameter-File-Format
[17:52:16] -!- kljsdfhklj has quit [Ping timeout: 250 seconds]
[17:57:11] -!- OoBIGeye has quit [Ping timeout: 260 seconds]
[18:00:27] -!- kb8wmc has quit [Quit: ChatZilla 0.9.86.1 [Firefox 3.6.15/20110303170030]]
[18:01:55] <cradek> mhaberler: nice update of appendix A - very precise
[18:02:24] <mhaberler> ah, so you wrote save_parameters() then ;-)
[18:03:01] <cradek> about source-highlight - can you say exactly what you want me to do please?
[18:03:13] <mhaberler> pls read the first 13 lines of http://git.mah.priv.at/gitweb/emc2-dev.git/blob/7b2bf699daa7a497c0ff48cbef136d2baeb88aee:/share/source-highlight/README - this describes what needs to be done: copy files, plus edit the existing lang.map file
[18:03:34] <cradek> ok I did read it
[18:04:12] <cradek> I still don't understand what you want me to do with this information
[18:04:18] <mhaberler> I am unure where that should go, share/source-highlight aint optimal, maybe docs/src/source-highlight
[18:05:08] <mhaberler> then the 'make install' process needs to be modified to copy docs/src/source-highlight/*.lang to /usr/share/source-highight (step1)
[18:05:13] <cradek> I don't even have these files on my system
[18:05:23] <cradek> is there a new dependency?
[18:05:41] <mhaberler> no, source-highlight is in deps already
[18:06:05] <cradek> I do not have it installed yet I have been building docs
[18:06:34] <mhaberler> you havent seen my source files yet... these refer to the new language defs, eg ngc
[18:06:44] <mhaberler> .txt i mean
[18:07:13] <mhaberler> hold on, I'll give an ida
[18:07:15] <mhaberler> idea
[18:08:08] <mhaberler> see http://emc.mah.priv.at/docs/src/gcode/overview.txt, search for source,{ngc}
[18:08:55] <cradek> ok, you are trying to make quoted gcode colorful in the docs?
[18:09:01] <mhaberler> yes
[18:09:32] <mhaberler> in html for now. I am not yet heroic enough to write a latex style file which would be requred
[18:09:43] <mhaberler> I rather read core dumps for breakfeast
[18:10:09] <mhaberler> the same goes for halcmd files, and highlighting .ini files
[18:11:43] <cradek> can you run source-highlight with --lang-map=filename --lang-def=filename etc? it is quite terrible that we would have to edit the system lang.map
[18:12:03] <mhaberler> missing step 2: edit /usr/share/source-highight/lang.map to contain the lines hal = hal.lang and ngc = ngc.lang
[18:12:07] <cradek> not being in /etc, it is probably not even marked by dpkg as a config file, so it will get porked at next update
[18:12:18] <mhaberler> no, unfortunately
[18:12:58] <mhaberler> oh, maybe yes. Stop those presses.
[18:13:05] <mhaberler> Blush.
[18:13:33] <mhaberler> Ok, you're off the hook for now, sorry..
[18:13:42] <cradek> oh good
[18:14:00] <cradek> it wasn't looking good for a minute there.
[18:14:42] <cradek> source-highlight should use a .d directory so a package can add support for a language by simply adding files
[18:16:03] <mhaberler> well, tough part - source-highlight is called from layer #129 or so beneaths asciidoc, so I ned to figure how to massage that
[18:16:22] <cradek> yuck.
[18:16:44] <mhaberler> anyway. My problem for now.
[18:16:57] <mhaberler> bbl, dinner time - thanks!
[18:17:06] <cradek> :-)
[18:17:56] <mhaberler> aha, only now I understand what this .d directory convention is for.. thx
[18:19:12] <cradek> I verified that lang.map is NOT marked as a conffile, so we really want to avoid editing it
[18:24:37] <mhaberler> ok, I'll out myself as n00b: how did you verify this? is it tagged in debia archives somehow?
[18:25:24] <cradek> I had to madly read docs until I found out about dpkg --status source-highlight
[18:25:46] -!- OoBIGeye has quit [Ping timeout: 260 seconds]
[18:27:16] <mhaberler> so whatever is in the Conffiles: section is fair game to fiddle with when installing, I guess
[18:27:26] <mhaberler> thanks - that helps me for the future
[18:27:51] <cradek> well it won't be overwritten by updates. it bugs the user and asks what he wants to do about it - keep old, install new, try to merge, edit manually, etc
[18:28:04] <mhaberler> ah. that is the magic.
[18:28:07] <cradek> so it's _still_ best to not mess with them if possible
[18:28:44] <mhaberler> super!
[18:28:48] <cradek> like badly fitting pants, just causes irritation :-)
[18:29:22] -!- jv4779 [jv4779!62c90637@gateway/web/freenode/ip.98.201.6.55] has joined #emc-devel
[18:29:53] -!- mhaberler has quit [Quit: mhaberler]
[18:34:31] -!- OoBIGeye has quit [Ping timeout: 260 seconds]
[18:50:26] -!- MattyMatt has quit [Ping timeout: 255 seconds]
[18:53:28] <jepler> it's terrible if a configuration file detail would affect how emc2 builds
[18:53:56] <jepler> the build should run as non-root, so putting a setting in /etc just can't be done, even if the concern about being wiped away by updates is ignored
[18:54:26] <cradek> jepler: my understanding is he's working on a better answer
[18:54:47] <jepler> Yes, and I'm glad of that
[18:54:48] <jepler> !
[18:55:02] <cradek> yes.
[18:58:38] -!- EDocTooR has quit [Read error: Connection reset by peer]
[18:59:35] -!- OoBIGeye has quit [Ping timeout: 276 seconds]
[19:02:18] -!- EDocTooR [[email protected]] has joined #emc-devel
[19:11:14] -!- IchGuckLive has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100423140709]]
[19:12:15] -!- mhaberler [[email protected]] has joined #emc-devel
[19:13:46] -!- psha [[email protected]] has joined #emc-devel
[19:16:09] -!- kb8wmc [[email protected]] has joined #emc-devel
[19:18:23] -!- andypugh [andypugh!~andy2@cpc2-basl1-0-0-cust1037.basl.cable.virginmedia.com] has joined #emc-devel
[19:19:29] <andypugh> What happens if I kmalloc the same thing twice?
[19:21:21] <mhaberler> duck.. and cover!
[19:21:25] <andypugh> Experiments indicate that it keeps the same address. I wonder if that is always the case? Would it be safer to kfree it first?
[19:21:47] <cradek> what do you mean kmalloc the same thing twice?
[19:22:02] <archivist> you would allocate another copy I would have thought, a memory leak
[19:22:18] <jepler> each kmalloc has to have a matching kfree, or you leak memory.
[19:22:27] <cradek> type *a; a = kmalloc(1); a = kmalloc(1);
[19:22:35] <andypugh> OK, no redefining to a different size then?
[19:22:46] <cradek> yes that's a leak
[19:23:02] <cradek> is there a krealloc? if not, you have to malloc a new one, copy, then free the original
[19:23:22] <andypugh> I think it is safe to lose the contents
[19:23:32] <cradek> oh, then just free it
[19:24:55] -!- jv4779 has quit [Ping timeout: 251 seconds]
[19:24:55] -!- skunkworks has quit [Ping timeout: 251 seconds]
[19:25:06] <andypugh> Hostmot2.c calls hm2_register_tram_regions before it finishes inserting the realtime module. I have another module that wants to add a bunch of other items to the lists, then register the new list.
[19:26:03] <andypugh> There is no way for hostmot2 to know that I later intend to run mesa_7i65, so it can't not bother with the tram register.
[19:27:31] -!- skunkworks_ [skunkworks_!447329d2@gateway/web/freenode/ip.68.115.41.210] has joined #emc-devel
[19:27:43] <andypugh> Is it enough to have if (pointer != null) kfree(pointer) ?
[19:28:22] <cradek> that's sometimes used, but there is no guarantee that a freed (or so-far unalloced) pointer is NULL unless you make it so
[19:28:40] <cradek> people who want to depend on that need to write free(pointer); pointer = NULL;
[19:29:55] <cradek> beware that while double malloc leaks memory, double free is a crashy operation - it's important to match them 1:1
[19:30:22] <andypugh> It's the inverse problem, if it isn't NULL, is it OK to kfree it? (the problem I am trying to avoid is kfree-ing a null pointer, because that is bad)
[19:30:25] <Jymmmm> leaks memory? c++ ???
[19:30:54] <cradek> kfreeing anything that doesn't come from kmalloc is the bad idea - NULL is only one case of this
[19:32:02] <andypugh> So, perhaps I need a static flag in the function, saying "we have already been here once"
[19:32:41] <andypugh> Jymmmm: Nothing so high-falutin' as C++.
[19:32:44] <cradek> I'm not sure of the details, but yes if you need to keep track separately, you have to just do that
[19:33:19] <cradek> or, you can use the =NULL paradigm, just be sure to set it to NULL at the right times (initially before malloc, and after free)
[19:33:30] <andypugh> So, what you are saying is that it is possible for a pointer not to be null and for it still to be unsafe to kfree it?
[19:33:38] <Jymmmm> I just didn't realize that c types had inhertited memory leaks
[19:33:50] <cradek> oh yeah that's not just possible, it's the general case
[19:34:09] <jepler> char c; char *d = &c; free(d);
[19:34:11] <cradek> type *a; if(a) free(a); <- very likely crash
[19:34:34] <Jymmmm> why is that?
[19:34:48] <cradek> nothing set a to NULL
[19:34:56] <andypugh> Isn't if (a == NULL) a bit more specific than if(a)?
[19:34:58] <jepler> my example is freeing a pointer that is not returned from an earlier malloc
[19:35:04] <cradek> andypugh: no, they are equivalent
[19:35:25] <Jymmmm> so unsetting an undef causes havoc (basically)?
[19:35:26] <cradek> er, opposite
[19:35:28] <jepler> cradek's example is freeing a pointer with an undefined value
[19:35:29] <cradek> you meant !=
[19:36:05] <cradek> freeing is not "unsetting"
[19:36:13] <Jymmmm> oh
[19:36:29] <Jymmmm> then I don't know what "freeing" is
[19:36:45] <andypugh> Sounds like the 80 byte memory leak hit (sometimes) might actually be the safest solution?
[19:36:49] <jepler> going commando is pretty freeing
[19:36:57] <Jymmmm> unallocating an undef?
[19:37:06] <jepler> do you know what "going commando" is?
[19:37:07] <cradek> andypugh: no, the safest solution is to keep track and make sure your malloc and free calls are 1:1
[19:37:32] <Jymmmm> jepler: When you've turned your undies insde out for the third time?
[19:37:33] <cradek> andypugh: like I said, you can do that with =NULL for unallocated, just be sure you code it instead of assuming it
[19:37:48] <jepler> unlike regular userspace programs, where the memory they allocate is freed when the program exits, leaked kernel memory isn't freed until you reboot
[19:38:06] <Jymmmm> ew
[19:38:24] <cradek> well, the kernel doesn't exit until you reboot
[19:38:25] <Jymmmm> that's like zombies sorta kinda
[19:38:34] <cradek> it's the same
[19:38:39] <jepler> if everybody writing kernel code says "a 40 byte leak doesn't matter", you'd actually run out of your 8GB memory pretty quickly
[19:38:52] <Jymmmm> Need a super kill == shotgun
[19:40:38] <andypugh> jepler: I agree in principle, in practice the code runs once or fewer times per EMC2 startup.
[19:41:03] <cradek> andypugh: but -- when in doubt, do it right
[19:41:18] <andypugh> With _my_ reputation? Are you mad?
[19:41:31] <andypugh> Yeah, I am asking what "right" is.
[19:42:16] <cradek> I think I've answered it a few times - you must have one free per malloc. you have to keep track somehow. using =NULL is a perfectly fine way.
[19:43:05] <andypugh> Yes, and thanks. I am now pondering how to do that.
[19:43:10] <cradek> ok :-)
[19:43:17] <cradek> maybe I don't understand the remaining questions, if there are some
[19:43:43] <andypugh> Well, how to ensure that the pointer is initialised to NULL.
[19:43:55] <cradek> you just initialize it
[19:43:58] <andypugh> I thought they always were, but now I am not sure)
[19:44:03] <cradek> type *p = NULL;
[19:44:38] <andypugh> And again after the kfree? (Google has indicated that is wise)
[19:45:01] <cradek> yes if you intend to guard against a later free by it being NULL, you must set it to NULL
[19:45:56] -!- pingufan has quit [Remote host closed the connection]
[19:48:00] <jepler> speaking specifically of the code in hm2_allocate_tram_regions, the memory where the pointer is stored is within a hostmot2_t, which is allocated with kmalloc and then set to all zeros with memset
[19:48:15] <jepler> so from there until the first assignment to any particular field, that field's value is 0.
[19:51:16] <jepler> two notes: there is a krealloc, and what it does if the old pointer is 0 is defined (and useful)
[19:51:34] <jepler> http://tomoyo.sourceforge.jp/cgi-bin/lxr/source/mm/util.c#L137
[19:52:12] <jepler> similarly, what kfree does with NULL is documented and useful.
[19:52:13] <jepler> http://tomoyo.sourceforge.jp/cgi-bin/lxr/source/mm/slab.c#L3809
[19:52:38] <jepler> so you could write ptr = krealloc(ptr, newsize)
[19:53:04] <andypugh> OK, let me look at those.
[19:53:07] <jepler> as long as ptr is either NULL or a previously returned value from krealloc, until it is finally kfreed somewhere (e.g., module removal time)
[19:53:21] <andypugh> Another interesting puzzle: http://pastebin.com/66SNraE5
[19:59:08] <jepler> is there a use-case for finding stepgen instances by name, or are you just curious?
[19:59:40] <jepler> I don't think there's a beautiful way to do it..
[20:00:22] <jepler> as to that function, I'm worried about it, because it only returns the index of the bspi instance within a particular card
[20:00:34] <jepler> how do you tell which card the instance is on?
[20:00:49] <jepler> oh, wait, maybe I see it now
[20:00:52] <jepler> *hm2 = list_entry...
[20:04:57] <andypugh> Getting the right card is the main point of the function.
[20:06:10] <andypugh> I don't want to search for a specific stepgen, but I can see that in the future I might want to search for a Simple SPI, Decoded BSPI, and maybe even a Smart Serial if this way of working proves to be useful (the jury is still out)
[20:10:59] -!- Athlocatle has quit [Ping timeout: 255 seconds]
[20:11:12] <andypugh> And to flit to another subject, when someone describes a commit as "610a4726b3", how does one find that on gitweb?
[20:11:43] <cradek> first, close the web browser, then at the command line, gitk 610a4726b3 or git log 610a4726b3
[20:12:01] <cradek> that important first step is the most useful to learn for using gitweb
[20:12:17] <andypugh> I was rather hoping to send someone a link?
[20:12:38] <cradek> if you pick some commit, then edit the URL, maybe it will work
[20:12:47] <cradek> (gitweb really sucks)
[20:13:13] <andypugh> Yes, but it is a very easy way to send folk links to the sourcecode.
[20:14:37] <jepler> to the best of my knowledge there's no place in the gitweb user interface to help you find a specific revision given its hash
[20:14:58] <jepler> you can construct them manually. http://git.linuxcnc.org/gitweb?p=emc2.git;a=commit;h=610a472
[20:16:06] <andypugh> I git it (that was a typo, but seemed apposite)
[20:16:13] <cradek> ha
[20:16:46] <andypugh> Not sure what radek is going on about here: http://www.linuxcnc.org/component/option,com_kunena/Itemid,20/func,view/catid,10/id,7986/limit,6/limitstart,6/lang,english/#8000
[20:16:58] <andypugh> (which was the context of the query)
[20:17:30] -!- theorbtwo has quit [Read error: Operation timed out]
[20:18:20] <andypugh> (it must be intensely frustrating trying to describe bugs in a language you are not very good at. It is certainly very frustrating trying to understand the descriptions)
[20:18:39] <cradek> grr, zip of source files is not very useful
[20:21:33] <jepler> http://pastebin.com/3c3VgkZ3
[20:21:33] <cradek> ohh
[20:21:33] <cradek> I truly can't imagine what that has to do with radek's problem
[20:29:22] -!- logger[psha] [logger[psha][email protected]] has joined #emc-devel
[20:36:43] -!- e-jones [[email protected]] has joined #emc-devel
[20:43:32] <mhaberler> cradek: an asciidoc command line option (extra filter config) will fix it
[20:54:45] -!- psha has quit [Quit: leaving]
[20:55:37] -!- awallin has quit [Read error: Operation timed out]
[20:56:07] <andypugh> I wonder why insmod is failing with my comp module.
[20:56:25] <andypugh> No errors, nothing in dmesg, all the pins are there...
[20:56:26] -!- awallin [[email protected]] has joined #emc-devel
[20:57:15] <andypugh> It's been like this from the beginning but now it is bugging me.
[21:24:16] -!- Tom_shop has quit []
[21:33:58] -!- ries has quit [Read error: Connection reset by peer]
[21:37:04] -!- motioncontrol has quit [Quit: Sto andando via]
[21:37:44] -!- ries [[email protected]] has joined #emc-devel
[21:38:45] -!- skunkworks_ has quit [Ping timeout: 252 seconds]
[21:40:29] -!- e-jones has quit [Quit: Ex-Chat]
[21:53:23] -!- tlab has quit [Quit: Leaving]
[21:59:17] -!- G38point4 has quit [Ping timeout: 252 seconds]
[22:07:16] <andypugh> Is there a way to define a path to a #included file which is independent of where the source file is?
[22:08:04] <andypugh> (an absolute link to the src/ directory, for example)
[22:18:17] -!- atom1 has quit []
[22:30:33] -!- acemi has quit [Quit: WeeChat 0.3.2]
[22:42:01] -!- DaViruz has quit [Read error: Operation timed out]
[22:54:21] -!- Fox_Muldr has quit [Ping timeout: 276 seconds]
[23:01:22] -!- tom3p [[email protected]] has joined #emc-devel
[23:02:59] -!- EDocTooR has quit [Read error: Connection reset by peer]
[23:04:56] -!- JT-Shop [[email protected]] has joined #emc-devel
[23:06:23] -!- EDocTooR [[email protected]] has joined #emc-devel
[23:23:05] -!- DaViruz has quit [Client Quit]
[23:23:13] <mhaberler> cradek: done
[23:23:35] <mhaberler> I'll ask Pavel to review it before I go ahead
[23:31:56] -!- kljsdfhklj has quit [Ping timeout: 276 seconds]
[23:32:52] -!- JT-Shop has quit [Read error: Connection reset by peer]
[23:32:58] Crow is now known as Guest75615
[23:33:24] -!- JT-Shop [[email protected]] has joined #emc-devel
[23:38:12] Guest75615 is now known as bluntz
[23:41:22] -!- mhaberler has quit [Quit: mhaberler]
[23:41:39] -!- skunkworks [skunkworks!~chatzilla@str-bb-cable-south2-static-6-412.dsl.airstreamcomm.net] has joined #emc-devel
[23:49:02] -!- brendanjerwin has quit [Remote host closed the connection]
[23:56:50] -!- bluntz has quit [Remote host closed the connection]