#emc-devel | Logs for 2011-08-14

Back
[00:00:13] <jepler> OK.
[00:00:36] <andypugh> Ah, yes, that looks more sensible.
[00:00:55] <jepler> now, what branch is marked with '*' in the output of 'git branch'?
[00:01:26] <andypugh> 2.5_branch
[00:01:34] <andypugh> Ahead of you there :-)
[00:01:54] <andypugh> Is there a way to git am blob-by-blob?
[00:02:12] <andypugh> Or is it simpler to manually edit the patch file?
[00:02:53] -!- mikegg has quit [Quit: Leaving]
[00:03:53] -!- theorbtwo has quit [Ping timeout: 250 seconds]
[00:03:59] theorb is now known as theorbtwo
[00:04:20] <andypugh> Oh, and another thing: Linux line ends are a bare "LF" I think? I have a problem where all my patches give a trailing whitespace error for every added line, and I can't figure out why as my editor is set of "LF" and the patches never leave the Ubuntu box.
[00:05:46] <jepler> hm, it doesn't sound like you should have any problems, then
[00:07:00] <jepler> can you easily get me a copy of this patch(set) you're trying to apply?
[00:07:04] <andypugh> git am --ignore-whitespace stops the error messages, but does that mean the trailing whitespace is still there?
[00:07:14] <jepler> I think that you want --whitespace=fix
[00:07:39] <andypugh> Yes, they all were posted to the dev list last week (with git send-email)
[00:08:46] <jepler> what's the subject line?
[00:08:56] <jepler> I'll try to "git am" it here and see what result I get
[00:09:37] <andypugh> They (for some reason) all start [PATCH 1/3]
[00:10:36] <andypugh> http://news.gmane.org/find-root.php?message_id=%3cCAN1%2bYZWFpfhx9go46FUvKaYDEUDKeePM74r9U2gADat6z0Wo1Q%40mail.gmail.com%3e
[00:10:54] <andypugh> I selected the text!
[00:11:15] <andypugh> http://news.gmane.org/find-root.php?message_id=%3cCAN1%2bYZWFpfhx9go46FUvKaYDEUDKeePM74r9U2gADat6z0Wo1Q%40mail.gmail.com%3e
[00:11:44] <jepler> it looks like there are 3 messages in there, all stuck inside one single message. Part way down there's this inside:
[00:11:47] <jepler> >From 18553272786ca369bc397d1651a15f9236f12a94 Mon Sep 17 00:00:00 2001
[00:11:50] <jepler> From: Andy Pugh <andy@...>
[00:11:52] <jepler> Date: Thu, 28 Jul 2011 02:58:29 +0100
[00:11:55] <jepler> Subject: [PATCH 2/3] Switch to config strings.
[00:12:11] <andypugh> Wierd. It appears in the text-entry box as the email subject, and gets converted to a URL when I press enter.
[00:12:49] <andypugh> Yes, those are git format-patch patches, so are (I assume) trying to conserve the history?
[00:13:18] <andypugh> I can dcc you the squashed version?
[00:13:29] <jepler> I have no idea whether dcc works, but it's worth a try
[00:13:37] <jepler> I'm having trouble unscrambling the one from the mailing list, unfortunately ;(
[00:14:14] <jepler> (could be my fault or the list's fault, but it ended up converted to MIME with lots of stray =20s)
[00:14:35] <andypugh> I was only using git send-email because you said my previous approach was too hard..
[00:15:01] -!- jbunch has quit [Read error: Connection reset by peer]
[00:16:38] <jepler> OK, I got these messages from git am, and it dumped me back at the shell: http://pastebin.com/vgAW7C5D
[00:16:56] <jepler> is that about where you are now?
[00:17:50] <jepler> (huh? pastebin says that paste looks like spam)
[00:19:10] <andypugh> Yes, it fails because it is trying to patch a file that is only on master (which will be removed anyway, as it has been overtaken by events)
[00:19:13] <jepler> The "whitespace" errors are not because of a CRLF line ending, but because of a plain old space (chr(32)) at the end of some of the added/changed lines of documentation.
[00:19:41] <andypugh> Ah, OK.
[00:20:05] <andypugh> That doesn't seem like an error, if the lines might be re-wrapped
[00:20:31] <jepler> you're not the only one to feel that way, but for better or for worse git warns about it by default
[00:20:39] <jepler> (I think Linus is intensely anal about that kind of thing...)
[00:21:38] <andypugh> What's the "EMC Policy" on spaces in wrapped, triple-quoted lines?
[00:22:24] <jepler> it's not a huge sin
[00:22:32] <jepler> but if you catch them before you push them, it's nice to fix them
[00:24:16] <jepler> so of the changes in the first patch of 3, are there any that should be applied to v2.5_branch?
[00:24:24] <andypugh> OK, if it counts as "fix" then I will do it. I actually put them there deliberately
[00:27:09] <jepler> my editor doesn't require space at the end of the line in order to re-flow a paragraph, so it wouldn't have occurred to me to do that.
[00:27:50] <jepler> and neither should the .9 or .comp documentation formats, when formatting docs for display
[00:28:41] <andypugh> I just don't necessarily see a line-break as a word break.
[00:29:24] <andypugh> It's a pain if you re-wrap to a different width and words merge.
[00:30:07] <andypugh> I just thought it better to make the word-breaks explicit. I can easily take them out.
[00:33:51] <jepler> I'd rather you left them out, but it's not a huge deal
[00:34:37] <andypugh> Consider them gone.
[00:34:53] Tech_Talk is now known as Birdman3131
[00:34:55] <jepler> ok.
[00:35:27] <jepler> so now we can get back to the question of how best to get the relevant parts of the stuff in sserial_updates.patch into v2.5_branch
[00:36:09] <jepler> have you used "git mergetool" yet?
[00:36:39] <andypugh> No
[00:36:56] <andypugh> Let me study it
[00:38:41] <jepler> when you have a merge conflict, it is a program that goes file-by-file and asks what you want to do. Usually (when the conflict is in the text of the file), that action is just to launch a tool that helps you resolve the merge. some people like to use a graphical merge tool like meld, and others (like me) like to use a special mode of their editor like gvimdiff.
[00:38:59] <jepler> (if you're not already a vi user, I can't recommend gvimdiff once; on the other hand, I don't really know which of the graphical tools are "good")
[00:39:42] <andypugh> I have never used vi
[00:41:33] <jepler> then install the package 'meld' and configure git to use it: git config --global merge.tool meld
[00:42:33] <andypugh> Are you suggesting that hand-editing the path files until they "take" isn't the best way?
[00:42:57] <jepler> it's a matter of taste .. editing patch files by hand can be .. tricky
[00:42:57] <andypugh> :-)
[00:43:14] <andypugh> Yeah, I noticed. But you do get what youwant
[00:43:31] <jepler> well, if you want to do it that way, you sure can
[00:43:45] -!- jbunch_ has quit [Read error: Connection reset by peer]
[00:43:54] <andypugh> No, I just didn't know there were options.
[00:43:57] <jepler> certainly if all you want to do is delete hunks from the patches, that's fine
[00:44:22] <jepler> In that case, I'll continue walking you through using mergetool and then you can decide what to do in the future
[00:44:36] <jepler> if you were already in the middle of a failing 'git am', abort it: $ git am --abort
[00:44:53] <jepler> now, run it in the way that makes it create conflicts instead of just giving up: $ git am -3 sserial_updates.patch
[00:44:55] <andypugh> Yes, I found that one
[00:45:23] <jepler> now run mergetool: $ git mergetool
[00:45:37] <andypugh> (abort that is, not the -3 )
[00:45:40] <jepler> ok
[00:46:02] <jepler> you should get this:
[00:46:02] <jepler> Normal merge conflict for 'src/hal/drivers/mesa-hostmot2/hostmot2.h':
[00:46:02] <jepler> {local}: modified
[00:46:03] <jepler> {remote}: modified
[00:46:03] <jepler> Hit return to start merge resolution tool (meld):
[00:46:08] <jepler> hit enter and the merge gui will pop up
[00:48:04] <jepler> are you with me?
[00:48:35] <andypugh> That's lucky, it works. Graphical stuff is a bit hit-and-miss. I make life more interesting by sitting at a Mac when working on EMC2.
[00:49:10] <andypugh> Yes, I see three columns, with pink highlights
[00:49:20] <jepler> yes. it's the pink you need to look for
[00:49:41] <jepler> that means "there are differences in this area, and the automatic method for resolving the differences didn't work"
[00:51:20] <andypugh> I am not that clear what each column represents
[00:52:26] <jepler> the middle is the file you're working on
[00:52:49] <jepler> the left is the file in the v2.5_branch before trying to apply the patch
[00:53:00] <jepler> the right is the file from master from before this change
[00:53:17] <andypugh> Ah, OK. Why mention master?
[00:54:15] <jepler> because hidden in the patch file it says that the patch was originally made on master (it says it by ref, not by name; I'm just assuming the name of the branch it was originally made on was master, based on the discussion we've been having)
[00:54:17] <andypugh> Maybe it makes sense.
[00:54:43] <andypugh> Ah, yes. That would be why.
[00:55:04] <jepler> so in another window, I'm also looking at the patch file.
[00:55:18] <jepler> from that, I can see that this change is adding a field to hal_other_t, but that struct doesn't exist in 2.5.
[00:55:24] <andypugh> I basically want to ignore that hink
[00:56:26] <jepler> in that case, you want to highlight all those red/pink lines in the middle column, then press backspace -- that'll delete them.
[00:57:13] <andypugh> Yes, the whole other_sserial thing (raw register access to unknown hardware) is only in master. Every module needs the extra register adding, and so the patch included other_sserial. (I actually thought that other_sserial was in 2.5, to be honest)
[00:57:27] <jepler> the other green sections in the middle will show you that you did keep some changes to that file -- adding reg_2_read and so on
[00:58:52] <andypugh> Yes, that has made it all green.
[00:59:08] <andypugh> I guess I hit "save" then
[00:59:22] <jepler> yes, click save and then close the window
[00:59:34] <jepler> now back in your terminal you should get this:
[00:59:34] <jepler> Deleted merge conflict for 'src/hal/drivers/mesa-hostmot2/mesa_other_sserial.c':
[00:59:37] <jepler> {local}: deleted
[00:59:40] <jepler> {remote}: modified
[00:59:42] <jepler> Use (m)odified or (d)eleted file, or (a)bort?
[00:59:47] <jepler> I think you want to use the (d)eleted file, so type 'd' and hit enter
[01:00:02] <jepler> "deleted" really just means "doesn't exist" in this case
[01:01:02] <andypugh> "modified" seems like a much more obvious choice to me.
[01:01:30] <jepler> maybe I made the wrong choice
[01:01:38] <andypugh> I just modified a file, I would like to use it... I seem to not have a linux mindset
[01:01:54] <jepler> did you look what file it's talking about?
[01:01:59] <jepler> it's not talking about hostmot2.h anymore
[01:02:31] <andypugh> You see, Linux actually expects you to pay attention and read stuff..
[01:02:48] <jepler> that commit changed mesa_other_sserial, which exists in master but not v2.5_branch
[01:03:17] <jepler> if you said (m)odified, then it would put the whole file from master into v2.5_branch. If that's what you want, that's fine, but my first guess was that you didn't..
[01:04:20] <andypugh> No, I want it to stay absent. Maybe "use deleted file" could be a little clearer though.
[01:04:55] <jepler> yes, it's yet another of those things that I've learned to just accept as it is, but the first time around it's confusing nomenclature
[01:04:58] <andypugh> The overall effect looks correct
[01:05:01] <jepler> OK
[01:05:11] <jepler> then you: $ git am -3 --resolved
[01:05:20] -!- WalterN has quit [Ping timeout: 240 seconds]
[01:05:29] <jepler> that'll go ahead and create the commit, and go on and do the other commits from the same file
[01:05:37] <jepler> .. both of which went off without a hitch here
[01:05:46] <andypugh> And here.
[01:05:58] <andypugh> Why -3 ?
[01:06:17] <jepler> to be honest, I'm not sure if it's needed when you say --resolved
[01:06:31] <jepler> but that's what git am told me to type earlier and long-since scrolled away
[01:06:57] <andypugh> Or, to phrase the question better, why is the "create conflict data" option represented by -3 ?
[01:07:03] <jepler> in the first place, 'git am -3 sserial_updates.patch' tells git that you want conflicts (which was what enabled us to use 'git mergetool')
[01:08:25] <andypugh> man explains, it means "use 3-way merge"
[01:08:42] <jepler> the program that creates a file with the conflict markers (the <<< === >>> that you may have noticed when deleting the pink lines) needs 3 files: your original file, the other original file, and the modified file
[01:09:22] <jepler> once again I'll happily agree that it is at least obtuse, if not downright opaque
[01:09:49] <jepler> but anyway, at this point we are in the happy position of having 3 commits which may even be the commits we want
[01:11:10] <jepler> at the very least, you should now test that the software still builds. and if you have a procedure to test whether it's right with respect to these changes, you should do that too
[01:11:38] <jepler> (it does build for me)
[01:12:37] <andypugh> Aye, I need to compile, start, and unplug hardware to force comms errors to be sure.
[01:12:38] -!- jbunch has quit [Read error: Connection reset by peer]
[01:13:04] <jepler> but at this point you can also try a 'git push --dry-run' and see what git wants to push
[01:14:18] <jepler> when you 'push' without any additional arguments, what git does depends on how you have configured it. The default is "matching", which means that for each branch that also exists at origin with the same name, it'll push that branch.
[01:14:31] <jepler> personally, I find that undesirable, so I've done this: $ git config --global push.default current
[01:14:39] <jepler> that pushes only the currently-checked-out branch by default
[01:15:32] <andypugh> I guess mergetool also lets me fix whitespace? There are two lines where my editor has auto-indented nothing..
[01:15:34] <jepler> but no matter what, always push with --dry-run to make sure it's what you want. A push should be considered "not undoable"
[01:16:08] <andypugh> Yes, I got that impression.
[01:17:06] -!- jbunch_ has quit [Ping timeout: 258 seconds]
[01:17:06] <andypugh> It seems to want to push to 2.5, and that works, and also to master, which fails.
[01:17:23] <jepler> (accidentally pushing an untended merge is the worst of all, because it can't be git-revert'ed (I've done it :() and bad enough that we'll go behind git's back to undo it)
[01:17:24] <andypugh> So I think I need to copy that option
[01:18:23] <andypugh> There are reasons that I didn't ask for push access, I knew I was incompetent. At least now I know where I am incompetent.
[01:19:22] <jepler> (speaking from experience) it's when you get cocky that the problems actually start
[01:19:35] <andypugh> If I miss-type --dry-run what happens? Say, git push dry-run?
[01:20:18] <andypugh> Do I get an error, or do I create a remote branch called dry-run, and get laughed-at?
[01:21:00] <jepler> 'git push dry-run' fails because there's no remote named dry-run
[01:21:10] <jepler> 'git push --dyr-run' fails because there's no option 'dyr-run'
[01:21:12] <cradek> andypugh: you could try a git push --dry-run dry-run and see
[01:21:37] <jepler> 'git push --dry-ru' does a dry-run push because it's OK to abbreviate (some) options
[01:24:08] <andypugh> pushing dry rum seems more apt right now
[01:24:13] <jepler> JT-Shop: I just scrolled back to see your question -- I was on master branch
[01:24:35] <jepler> I'm about to try again on v2.5_branch
[01:25:09] -!- jbunch has quit [Ping timeout: 258 seconds]
[01:25:32] <jepler> .. where it fails
[01:25:39] <jepler> lucid, rtai, in a virtual machine
[01:28:30] -!- robh__ has quit [Ping timeout: 240 seconds]
[01:52:06] <jepler> hm, I'm not seeing the failure on a non-virtual machine configured for sim
[02:10:06] -!- nicko has quit [Ping timeout: 255 seconds]
[02:23:47] -!- Loetmichel has quit [Ping timeout: 252 seconds]
[02:24:32] <jepler> .. and I'm repeatedly failing at using 'git bisect' to find where the problem was introduced on the system where I am seeing it
[02:28:01] <andypugh> As in it always fails, or always passes?
[02:28:38] <jepler> as in .. it says it finds the first bad commit, but when I test the prior commit it's bad too
[02:38:10] -!- uwe_ has quit [Ping timeout: 240 seconds]
[02:40:06] <andypugh> Good job you like a puzzle, then?
[02:43:47] <jepler> 4b37e1ea145821fb4fdaf95d886ac493e43a6020 is the first bad commit
[02:43:52] <jepler> Author: Jeff Epler <[email protected]>
[02:43:52] <jepler> Date: Wed Oct 13 10:55:31 2010 -0500
[02:43:53] <jepler> halmodule: add const and casts to silence warnings
[02:43:58] <jepler> OK, it's said this twice in a row now
[02:44:47] <jepler> OK, that's because I specified a "good" starting point that was actually bad
[02:45:39] <jepler> andypugh: something like that
[02:57:46] -!- gene76 has quit [Read error: Operation timed out]
[03:05:28] <andypugh> andypugh@mill:~/emc2-dev/src$ git push --dry-run
[03:05:28] <andypugh> To ssh://[email protected]/git/emc2.git
[03:05:29] <andypugh> 30a6a0d..7787039 HEAD -> v2.5_branch
[03:05:39] <andypugh> does that look right?
[03:05:52] <jepler> one last double check: git log --oneline 30a6a0d..7787039
[03:05:56] <jepler> it should be exactly the commits you want
[03:06:18] <andypugh> Yes, that looks good
[03:06:41] <andypugh> Should I rebase -i to squash them into one commit?
[03:07:08] <jepler> at each step, does the software work at least as well as at the stop before?
[03:07:29] <jepler> (not that you necessarily tested each one the same amount)
[03:08:31] <andypugh> if that is the criterion they ought to be squashed
[03:08:51] <jepler> if the software works worse during an intermediate step, then it should probably be squahsed
[03:10:03] <andypugh> I think they all work, but two are really very trivial.
[03:11:01] <jepler> trivial is just fine
[03:11:02] <jepler> broken is not
[03:11:44] <andypugh> No, they do all work. (and were all tested)
[03:12:11] <jepler> then push it as 3 commits
[03:12:25] <andypugh> OK
[03:12:54] <andypugh> From the gui or the command line?
[03:13:04] <andypugh> Or is that a matter of taste?
[03:13:36] <jepler> I would do it from the commandline, but that doesn't mean it's not a matter of taste
[03:15:11] <andypugh> is it just git push? That looked OK in dry-run
[03:15:23] <jepler> yes. if you remove --dry-run it will actually do what it said it would do
[03:15:45] <andypugh> <crosses fingers>
[03:15:57] <CIA-11> EMC: 03andypugh 07v2.5_branch * r91a64a3c03e7 10/ (3 files in 2 dirs): Add support for second user register in Smart Serial, and improve the handling of errors.
[03:15:58] <CIA-11> EMC: 03andypugh 07v2.5_branch * r7d345542fe95 10/src/hal/drivers/mesa-hostmot2/ (mesa_8i20.c sserial.c): Switch to config strings.
[03:15:59] <CIA-11> EMC: 03andypugh 07v2.5_branch * r77870396fada 10/src/hal/drivers/mesa-hostmot2/ (hostmot2.h sserial.c): Tidy up 8i20 fault handling and support second user register
[03:16:06] <jepler> whee
[03:16:43] <andypugh> And now to sleep.
[03:16:59] <cradek> yay!
[03:17:56] <andypugh> Anyone care to pull and make, just to put my mind at rest?
[03:18:47] <andypugh> (only about three people can actually test at the moment, it needs hardware to work at all)
[03:19:11] <jepler> I'll at least do a build
[03:20:09] <jepler> all looks OK on build
[03:20:16] <andypugh> I think that Pete is hoping that the other 8 cards can get into 2.5. Does that seem likely?
[03:20:37] <andypugh> It's all fairly isolated code.
[03:21:20] <jepler> cradek's the one to ask about that
[03:22:08] <andypugh> logger[psha]:
[03:23:27] <cradek> dir
[03:23:29] <cradek> dangit
[03:24:07] <cradek> andypugh: are you pretty sure it doesn't affect other existing cards?
[03:24:47] <andypugh> I will be, as I have both existing cards to test with
[03:25:22] <cradek> new drivers that don't affect old drivers are usually a safe bet for the stable branch. we've done that a lot.
[03:25:23] <andypugh> And it will definitely only affect smart-serial
[03:26:23] <andypugh> One thing I am planning to push is that ISA card driver (PCL720). That's just a single comp, so ought to be 100% safe?
[03:27:01] <andypugh> it would be nice if the one user didn't lose it if he upgrades :-)
[03:27:10] <cradek> yep, adding a comp is no problem. the only possible glitch is packaging, so ideally you should test building and installing a package
[03:29:01] <andypugh> Looks like I missed the deadline for 2.5 :-)
[03:29:04] <andypugh> [20:57:58] <andypugh> cradek: Give me a deadline for 8i20/sserial bugfixes. I can't work witout deadlines
[03:29:04] <andypugh> [20:58:32] <cradek> andypugh: July 22 23:59UTC
[03:30:54] <andypugh> Right, sleep time. Thanks jepler, thanks cradek
[03:30:58] <cradek> g'night
[03:31:02] -!- andypugh has quit [Quit: andypugh]
[03:35:15] <jepler> man, I am just too stupid to 'git bisect' tonight
[03:35:26] <cradek> jepler: it's too late for that kind of thing
[03:36:19] -!- ve7it has quit [Remote host closed the connection]
[03:39:17] <jepler> it shouldn't have been when I started, 3 hours ago..
[03:39:58] <cradek> well I'm backing up files via sneakernet...
[03:44:06] <cradek> cp: reading `MATRIX.ASM': Input/output error
[03:46:06] <jepler> floppies are as dependable as ever, I see.
[03:50:07] <cradek> I need another floppy
[03:50:41] <cradek> or, it's looking like it might be worth setting up networking.
[03:50:50] <cradek> because really, how hard can it be?
[04:07:20] <jepler> well, I've successfully chased my bug from this universe into an unreachable parallel universe where it belongs..
[04:07:23] <jepler> 'night
[04:19:41] -!- pjm__ has quit [Ping timeout: 260 seconds]
[04:26:16] -!- ve7it [[email protected]] has joined #emc-devel
[05:04:33] -!- Technicus has quit [Quit: Leaving.]
[05:16:22] -!- factor has quit [Ping timeout: 264 seconds]
[05:34:45] -!- Eartaker has quit [Quit: Not again...]
[05:42:20] -!- factor has quit [Quit: Leaving]
[06:04:31] -!- ve7it has quit [Remote host closed the connection]
[06:14:57] -!- krushia has quit [Ping timeout: 258 seconds]
[06:48:14] -!- theorbtwo has quit [Read error: Operation timed out]
[06:54:07] -!- pjm has quit [Read error: Connection reset by peer]
[07:16:26] -!- FinboySlick has quit [Quit: Leaving.]
[07:42:01] -!- capricorn_one has quit [Remote host closed the connection]
[08:30:07] -!- Eartaker has quit [Quit: Not again...]
[08:54:21] -!- robh__ [[email protected]] has joined #emc-devel
[10:55:14] -!- nicko has quit [Quit: Visitor from www.linuxcnc.org]
[11:18:32] -!- theos has quit [Disconnected by services]
[11:24:43] -!- maximilian_h [[email protected]] has joined #emc-devel
[11:25:22] -!- geo01005 has quit [Ping timeout: 264 seconds]
[11:26:16] -!- jbunch has quit [Read error: Connection reset by peer]
[11:27:16] -!- maximilian_h has quit [Read error: Connection reset by peer]
[11:27:34] -!- maximilian_h [[email protected]] has joined #emc-devel
[11:31:10] -!- jbunch has quit [Read error: Connection reset by peer]
[11:35:50] -!- jbunch_ has quit [Read error: Connection reset by peer]
[11:41:35] -!- jbunch has quit [Read error: Connection reset by peer]
[11:43:11] -!- fragalot has quit [Ping timeout: 240 seconds]
[11:43:55] fragalot is now known as Guest83000
[12:22:12] -!- jbunch_ has quit [Ping timeout: 276 seconds]
[12:33:50] -!- jbunch has quit [Read error: Connection reset by peer]
[12:36:41] -!- jbunch has quit [Read error: Connection reset by peer]
[12:39:31] -!- jbunch_ has quit [Read error: Connection reset by peer]
[12:47:13] -!- jbunch has quit [Read error: Connection reset by peer]
[12:48:11] -!- gene76 has quit [Ping timeout: 240 seconds]
[12:49:53] -!- jbunch_ has quit [Read error: Connection reset by peer]
[12:50:28] <jthornton> I get this error when building 2.5 with docs "Required image file ../docs/src/code/emc2-motion-homing-diag.ps does not exist. Stop."
[13:03:17] -!- jbunch has quit [Read error: Connection reset by peer]
[13:11:36] <jepler> try a make clean. It looks like that file was moved in recent memory (june 7 or so)
[13:12:04] <jthornton> thanks, I forgot to do a make clean
[13:13:08] <jthornton> and I should have known that as many times as I moved stuff around...
[13:15:00] <jthornton> speaking of beer I was pointed to this place for a neat spreadsheet for beer water calculations http://sites.google.com/site/brunwater/
[13:50:21] -!- sumpfralle has quit [Ping timeout: 258 seconds]
[14:01:08] <jthornton> I get 23 failed tests on master with a pull and make clean and make today http://pastebin.com/QTHWJYgn
[14:01:18] <jthornton> but I don't see the exact error you did
[14:03:36] -!- sumpfralle1 has quit [Ping timeout: 246 seconds]
[14:10:00] <jepler> looks like all the tests that require realtime failed
[14:10:06] <jepler> does emc run?
[14:10:23] <JT-Shop> I'll run down and see
[14:13:43] <jthornton> no, this is the output http://pastebin.com/dJiSjRS0
[14:13:59] <jepler> module_helper is not setuid root
[14:14:07] <jepler> you forgot to 'sudo make setuid'
[14:14:14] <jthornton> ok
[14:14:58] <jepler> so far I've just brewed with tap water .. this looks crazy complicated.
[14:15:05] <jepler> something else I have to worry about in my beer? on no.
[14:15:13] <jthornton> it does at first look complicated
[14:16:39] <jthornton> Runtest: 47 tests run, 47 successful, 0 failed + 0 expected
[14:16:57] <jthornton> it skipped two tests motion/g0 and save.1
[14:20:06] <jthornton> I just use RO water and add a couple of things to get the profile I want
[14:25:39] <jepler> what unit of measurement is "meg/L"?
[14:27:02] <jepler> aha, https://secure.wikimedia.org/wikipedia/en/wiki/Equivalent_%28chemistry%29
[14:36:17] -!- Valen has quit [Quit: Leaving.]
[14:41:18] -!- betta_y_omega has quit [Ping timeout: 276 seconds]
[14:43:18] -!- betta_y_omega [[email protected]] has joined #emc-devel
[14:44:32] <CIA-11> EMC: 03jepler 07v2.5_branch * rb78674f925ad 10/scripts/decode_dmesg: add copyright notice
[14:48:31] <CIA-11> EMC: 03jepler 07master * r0a152eadf3ce 10/src/hal/drivers/mesa-hostmot2/ (12 files): Remove rtapi_app where it is unneeded
[14:49:38] <CIA-11> EMC: 03jepler 07master * r91a64a3c03e7 10/ (3 files in 2 dirs): Add support for second user register in Smart Serial, and improve the handling of errors.
[14:49:39] <CIA-11> EMC: 03jepler 07master * r7d345542fe95 10/src/hal/drivers/mesa-hostmot2/ (mesa_8i20.c sserial.c): Switch to config strings.
[14:49:40] <CIA-11> EMC: 03jepler 07master * r77870396fada 10/src/hal/drivers/mesa-hostmot2/ (hostmot2.h sserial.c): Tidy up 8i20 fault handling and support second user register
[14:49:41] <CIA-11> EMC: 03jepler 07master * rb78674f925ad 10/scripts/decode_dmesg: add copyright notice
[14:49:42] <CIA-11> EMC: 03jepler 07master * rf76c05baf342 10/ (5 files in 3 dirs): Merge remote branch 'origin/v2.5_branch'
[14:59:40] -!- A2Sheds has quit [Quit: puff of smoke]
[15:08:34] -!- Birdman3131 has quit [Read error: Connection reset by peer]
[15:13:38] <cradek> ooh, look at those nice easy merges
[15:18:30] -!- Nick001 has quit [Ping timeout: 240 seconds]
[15:18:50] <CIA-11> EMC: 03seb 07v2.5_branch * r6f704093b79e 10/debian/update-dch-from-git: change debian version number format, again
[15:29:02] -!- mikegg has quit [Read error: Connection reset by peer]
[15:48:02] <jepler> hm, turns out it is documented that rtapi_sprintf doesn't handle long longs
[15:48:08] <jepler> (in rtapi.h)
[16:00:39] <jepler> .. and with pretty good reason: after doing the obvious thing to add support for long longs, formatting a 3-digit number takes about 70ns longer
[16:01:01] Guest83000 is now known as fragalot
[16:01:09] -!- fragalot has quit [Changing host]
[16:32:41] -!- maximilian_h [[email protected]] has parted #emc-devel
[16:33:44] -!- syyl has quit [Quit: Leaving]
[16:57:26] -!- mhaberler [[email protected]] has joined #emc-devel
[17:02:48] -!- factor has quit [Read error: Connection reset by peer]
[17:06:23] -!- IchGuckLive has quit [Remote host closed the connection]
[17:12:44] -!- syyl_ has quit [Quit: Verlassend]
[17:22:36] -!- isssy has quit [Client Quit]
[17:26:35] -!- jbunch_ has quit [Remote host closed the connection]
[17:39:26] -!- maximilian_h [[email protected]] has joined #emc-devel
[17:49:13] -!- ries has quit [Quit: ries]
[17:49:49] -!- ries has quit [Client Quit]
[18:01:00] -!- isssy has quit [Quit: Visitor from www.linuxcnc.org]
[18:06:58] -!- ve7it [[email protected]] has joined #emc-devel
[18:12:46] -!- Birdman3131 has quit [Ping timeout: 240 seconds]
[18:24:25] -!- psha [[email protected]] has joined #emc-devel
[18:44:34] -!- factor has quit [Ping timeout: 264 seconds]
[18:46:46] <CIA-11> EMC: 03jepler 07master * r4ab5a6b3767b 10/ (3 files in 3 dirs): rtapi_byteorder: what's architecture's byte order?
[18:46:47] <CIA-11> EMC: 03jepler 07master * r55f370ffcc6f 10/src/rtapi/rtapi_math.h: rtapi_math: provide unions that access parts of ieee doubles
[18:46:49] <CIA-11> EMC: 03jepler 07master * rb28f06e4a021 10/src/rtapi/vsnprintf.h: rtapi_vsnprintf: fix up a constness gripe
[18:46:50] <CIA-11> EMC: 03jepler 07master * r1f897e8cf8be 10/src/rtapi/vsnprintf.h: rtapi_vsnprintf: if buf+size-1 wraps around, fail immediately
[18:46:51] <CIA-11> EMC: 03jepler 07master * rc9fc8ff995f9 10/src/rtapi/vsnprintf.h: rtapi_vsnprintf: make the compiler like it more
[18:46:51] <CIA-11> EMC: 03jepler 07master * rdfa3b7211e69 10/src/rtapi/vsnprintf.h: rtapi_vsnprintf: Match rtapi.h declaration
[18:46:52] <CIA-11> EMC: 03jepler 07master * r18af9d76611b 10/ (5 files in 3 dirs): rtapi_vsnprintf: provide a test program
[18:46:52] <CIA-11> EMC: 03jepler 07master * r25bab2df54db 10/src/rtapi/vsnprintf.h: rtapi_vsnprintf: integer version of %A for ieee
[18:46:53] <CIA-11> EMC: 03jepler 07master * ra4971ede0503 10/ (4 files in 3 dirs): rtapi_math64.h: division and remainder of 64-bit quantities
[18:46:54] <CIA-11> EMC: 03jepler 07master * r1446aa6f2859 10/docs/man/man3/rtapi_snprintf.3rtapi: rtapi_vsnprintf: document the limitations
[18:50:11] -!- Roguish [[email protected]] has joined #emc-devel
[18:55:43] -!- Roguish has quit [Quit: ChatZilla 0.9.86.1 [Firefox 3.6.17/20110422072923]]
[18:56:25] -!- Roguish [[email protected]] has joined #emc-devel
[18:58:07] <Roguish> afternoon all. any hostmod knowledge on today?
[18:59:09] <Roguish> I have a 5i20 to a 7i37 (opto-isolator) but cannot seem to get the outputs to work.
[19:00:14] <Roguish> inputs on the 7i37 are good to go, but the outputs on the 7i37 do not change state. the hal pins and signals all seem fine.
[19:00:48] <JT-Shop> have you configured them for outputs?
[19:01:45] <Roguish> yes, i believe so. using setp....is_output
[19:02:22] <JT-Shop> I used that same setup on my Hardinge http://gnipsel.com/shop/hardinge/hardinge.xhtml
[19:02:38] <JT-Shop> wiring diagrams etc on that link
[19:02:56] <JT-Shop> well crap the link is hosed
[19:03:19] <JT-Shop> give me a minute to fix it
[19:04:35] <Roguish> looking at your .hal file. why do you invert the pins?
[19:05:51] <jthornton> cause they are all on if you don't
[19:06:52] <Roguish> ok. i actually have that also.
[19:07:15] <Roguish> and i have tried ....is_opendrain
[19:08:13] <jthornton> ok, refresh that page and the wiring diagrams will work
[19:10:45] <Roguish> did, thanks. i am trying to fire a simple relay to turn on/off the spindle.
[19:13:33] <JT-Shop> np hope it helps you
[19:13:54] <JT-Shop> I guess I should label the rails :/
[19:14:03] <Roguish> i see you're using solid state relays. i have a simple 24v dc coil type.
[19:14:54] <JT-Shop> just make sure the coil doesn't draw too much current
[19:15:03] <JT-Shop> they are a bit power hungry
[19:15:07] <Roguish> help a LOT.
[19:15:35] <Roguish> it's a small relay, going to a large 220 contactor.
[19:16:19] <Roguish> besides, i have 30A 24v ps
[19:16:40] <JT-Shop> whats the limit for the 7i37?
[19:17:14] <Roguish> the outputs ar limited to 48v and 1
[19:17:16] -!- sumpfralle has quit [Ping timeout: 258 seconds]
[19:17:16] <Roguish> aa
[19:17:19] <Roguish> amp
[19:17:30] <Roguish> pretty small
[19:17:37] <Roguish> little fets
[19:18:11] <JT-Shop> should pull in a relay
[19:18:17] <JT-Shop> the relay works?
[19:19:11] <JT-Shop> time to get back to the kitchen floor :/
[19:26:05] -!- Tom_itx has quit [Ping timeout: 258 seconds]
[19:26:25] <Roguish> yes the relay works fine. THANKS FOR THE HELP.
[19:29:06] -!- e-ndy [[email protected]] has joined #emc-devel
[19:32:55] <JT-Shop> did you get it to work?
[19:45:44] -!- El_Matarife has quit [Quit: Nettalk6 - www.ntalk.de]
[19:53:05] -!- nullie has quit [Quit: Ex-Chat]
[19:53:22] -!- pixpop [[email protected]] has joined #emc-devel
[20:13:04] <Roguish> JT-Shop: well, not really, but yes. i triple checked my setup and wiring and everything looks ok. still did not work. so, i unwrapped a new 7i37 and put it in and it works. got no idea what i did to the first board.
[20:13:40] <JT-Shop> at least you have it working...
[20:13:41] -!- pixpop_ [[email protected]] has joined #emc-devel
[20:14:46] <Roguish> yes. better now. guess i'll be speaking with peter w next week. thanks again for the help. been driving me nuts for a too many hours.
[20:15:31] <JT-Shop> np
[20:16:16] -!- pixpop_ has quit [Client Quit]
[20:16:46] -!- pixpop has quit [Ping timeout: 250 seconds]
[20:17:08] -!- Roguish has quit [Remote host closed the connection]
[20:18:31] -!- psha has quit [Quit: Lost terminal]
[20:19:41] -!- Tom_garage has quit [Ping timeout: 240 seconds]
[20:21:45] -!- Calyp has quit [Quit: Leaving]
[20:34:52] -!- andypugh [andypugh!~andy2@cpc2-basl1-0-0-cust1037.basl.cable.virginmedia.com] has joined #emc-devel
[21:09:27] -!- betta_y_omega has quit [Ping timeout: 252 seconds]
[21:11:25] -!- betta_y_omega [[email protected]] has joined #emc-devel
[21:40:40] -!- maximilian_h has quit [Quit: Leaving.]
[21:43:57] -!- maximilian_h [[email protected]] has joined #emc-devel
[21:47:30] -!- Techrat has quit [Ping timeout: 276 seconds]
[21:53:33] -!- Fox_Muldr has quit [Ping timeout: 276 seconds]
[22:12:03] -!- Tech_Talk has quit [Ping timeout: 250 seconds]
[22:20:29] -!- e-ndy has quit [Quit: Ex-Chat]
[22:52:30] -!- anonimasu has quit [Ping timeout: 276 seconds]
[22:53:39] -!- mikegg has quit [Ping timeout: 250 seconds]
[23:17:35] -!- uwe_ has quit [Ping timeout: 260 seconds]
[23:28:20] <CIA-11> EMC: 03andypugh 07v2.5_branch * rd3d577ba1de5 10/src/hal/drivers/pcl720.comp: Add driver for PCL720 64ch IO card http://www.advantech.com
[23:30:24] -!- cwheeler has quit [Remote host closed the connection]
[23:40:22] -!- mendelbuild has quit [Ping timeout: 264 seconds]
[23:43:51] -!- elmo40 has quit [Ping timeout: 276 seconds]