Re: [Machinekit] hb04 Pendant integration

[schoone...@btinternet.com]

On 20/05/17 03:10, Mike Cinquino wrote: Installed libgcc1_4.9.2-10_armhf.deb I had to installed gdebi to get it done as the command line methods I could not get to work. It said it installed and I had no errors.  Confirmed libgcc_s.so.1 is in both 4.9 and 4.9.2 directory. Those are just symlinks Confirmed it is also in /lib/arm-linux-gnueabihf That is the real file, installed at last! $ sudo ldconfig -p | grep libgcc_s.so.1 returned : libgcc_s.so.1 (libc6,hard-float) => /lib/arm-linux-grueahihf/libgcc_s.so.1 sudo ldconfig -p saw the loads of stuff as you said. Attempted to launch with same error. Attempted sudo ldconfig multiple times. Same error. If you search this error it seems to occur in lots of different applications, for no good reason, when the library is installed. There has been a problem with python calls closing a file descriptor used by dlopen and leaving the library unreachable and loads of other similar but unresolved stuff. Now that you actually have it installed, I come to the end of what I can test. I don't use a BBB and just downloaded the image and burnt it to a SD card so I could see what was in it. The pendant and library loading works just fine on a x86_64 platform The possible fixes that suggest themselves would be tortuously difficult to try to explain given your level of linux-fu and I cannot test them first.   My best suggestion is get hold of Simon Mali, who has a working XHC-HB04 config on a BBB as per this thread https://groups.google.com/forum/#!topicsearchin/machinekit/xhc-hb04;context-place=forum/machinekit/machinekit/8mMHqZ_Zb80 and ask him to clone his SD card image and post it somewhere so you can use it. On Fri, May 19, 2017 at 9:36 AM, schoone...@btinternet.com <schoone...@btinternet.com> wrote: On 19/05/17 13:56, Mike Cinquino wrote: I agree. And I truly appreciate your patience with me.  This is what I have done to this point. Red is what I have pulled from the above string that you have asked me to do. 1- Confirmed- You need to physically look in /usr/lib/gcc/arm-linux-gnueabihf/4.9  for libgcc_s.so.1 (4.9 has the libgcc_s.so.1 file in it) Does it, or is it a symlink which does not lead anywhere? The file in /usr/lib/gcc/arm-linux-gnueabihf/4.9 is just a symlink The actual file is a /lib/arm-linux-gnueabihf If you are using some dumb file manager which can't tell the difference, that is part of your problem The symlink is installed as part of another gcc package, but its presence does not mean that what it points to is present. 2- Not sure here - Now that you have installed libgcc1 package and can find the file libgcc_s.so.1, does this file appear in ldconfig cache? $ sudo ldconfig -p | grep libgcc_s.so.1 Is the above command supposed to tell me if the file is in ldconfig cache? Because when I run it I get nothing. Just returns to command prompt no errors. I test Machinekit and fail with error as described. If it returns nothing, that is because there is no entry in its cache relating to libgcc_s.so.1 Run sudo ldconfig -p and it will return loads of stuff, piping to grep is just a way to sort out what you are looking for So you haven't installed it. You must install libgcc1 and you will have the file https://packages.debian.org/jessie/armhf/libgcc1/filelist It is in libgcc1_4.9.2-10_armhf.deb Until you install the package it will not work. 3- Not sure here either - This command should put the file in the ldconfig cache - If not run $sudo ldconfig and try again    (I run this command and get nothing in return, no error, just new command prompt?)  I test Machinekit and fail with error as described. So after running sudo ldconfig I go back to item 2 and run $ sudo ldconfig -p | grep libgcc_s.so.1 again with same result. I test Machinekit and fail with error as described. ******************************************************************* Now in the directory /usr/lib/gcc/arm-linux-gnueabihf there appear to be multiple versions. 4.9.2, 4.8.4, 4.8 and 4.9. These are all folders in folders are in /usr/lib/gcc/arm-linux-gnueabihf directory. 4.9 and 4.9.2 have libgcc_s.so.1 in them with multiple other files. The other 2 are empty. How do I know what directory (or version) is being looked at or used? You identify 4.9 as being the one to look in. But how do we know it's not one of the others? On Fri, May 19, 2017 at 3:00 AM, schoone...@btinternet.com <schoone...@btinternet.com> wrote: You have exactly the same error as before libgcc_s.so.1 must be installed for pthread_cancel to work /usr/bin/linuxcnc: line 753:  5138 Aborted                 haltcl -i "$INIFILE" $CFGFILE 5069 Until you get to grips with actually installing the lib and testing that it is present, you will just go around in circles. I have detailed what to do, but even when you say you have done it , you don't elaborate or give any output to show what resulted, so it is hard to help you. On 19/05/17 00:59, Mike Cinquino wrote: Here is the error list after the last changes. On Thu, May 18, 2017 at 4:35 PM, Mike Cinquino <mcinqu...@gmail.com> wrote: Is there an older image that is known to work with BBB and the hb04? I might be better off going that way if there is? I have not been in front of the machine much the last couple days. Now that I know the difference between L and i ... I got through those commands.  Still get an error. I believe it is slightly different. It does not say the file is missing it says it needs to be installed. I will send the error log once I get back to the machine. On Wed, May 17, 2017 at 9:23 AM, Mike Cinquino <mcinqu...@gmail.com> wrote: Oops ...I think I am using a "i" rather than an "L" Capital I looks like I .... lower case L looks like l...same same.  On Wed, May 17, 2017 at 9:17 AM, Mike Cinquino <mcinqu...@gmail.com> wrote: I think I was looking at before and after I ran the commands. Looks the same now. It tells me Idconfig: command not found On Wed, May 17, 2017 at 9:09 AM, schoone...@btinternet.com <schoone...@btinternet.com> wrote: On 17/05/17 13:19, Mike Cinquino wrote: I ran the commands above. Then...... I found something that I had not noticed.  I was browsing to the file location usr/lib/gcc... But I was doing it under rootfs. I was not using the filesystem folder. Now when I find it under the filesystem folder I have 4.9.2, 4.8.4, 4.8 and 4.9. 4.9.2 and 4.9 both have the files in them. The 4.8.4 and 4.8 are empty. Could machinekit also be looking at the wrong location? Afraid I don't understand what you are saying, rootfs is root filesystem Now that you have installed libgcc1 package and can find the file libgcc_s.so.1, does this file appear in ldconfig cache? $ sudo ldconfig -p | grep libgcc_s.so.1 If not run $sudo ldconfig and try again If found run the sim config, does it work? On Tue, May 16, 2017 at 2:15 AM, schoone...@btinternet.com <schoone...@btinternet.com> wrote: On 16/05/2017 02:31, Mike Cinquino wrote: So I don't have a 4.9 directory. I have 4.7, 4.6, 4.63 and 4.72. You said "The image I burned in relation to last message was: bone-debian-8.7-machinekit-armhf-2017-02-12-4gb.img" that should be Jessie and have gcc-4.9.2 No matter, just run sudo apt-get update sudo apt-get install libgcc1 That should install libgcc1_4.7.2 if that is your gcc version 4.6 and 4.62 have the file. The other 2 directories are empty.  I am using synaptic to install the .deb but any internet activity on the BBB is painfully slow. Took over an hour to search for the file. Looks like it found it but it is still thinking about it.  Hopefully it will allow me to install it and then I can try.  On Mon, May 15, 2017 at 9:03 AM, schoone...@btinternet.com <schoone...@btinternet.com> wrote: On 15/05/17 13:36, Mike Cinquino wrote: Do I have to be in any particular directory to run the commands? I ran: sudo ldconfig -p | grep libgcc_s.so.1 got no file found. then ran: sudo ldconfig  got no file found. No, if the library is not in the ldconfig cache, running sudo ldconfig should put it there if it exists. You need to physically look in /usr/lib/gcc/arm-linux-gnueabihf/4.9  for libgcc_s.so.1 The stock image does not appear to contain it, just a dead symlink to it Then you need to install libgcc1_4.9.2-10_armhf.deb which contains it Then run sudo ldconfig -p | grep libgcc_s.so.1 and it should report back showing it and the symlink to it Now you will be able to run the sim hopefully On Mon, May 15, 2017 at 3:56 AM, schoone...@btinternet.com <schoone...@btinternet.com> wrote: On 14/05/17 21:04, Mike Cinquino wrote: Can I just add that file? If so is it specific to my setup or is it a generic file? You need to do what I outlined below. Look to see if it is physically in your system. If it is, look at the ldconfig cache to see if it is listed, if not add it by running ldconfig again and recheck. On May 14, 2017 12:05 PM, "schoone...@btinternet.com" <schoone...@btinternet.com> wrote: On 14/05/17 16:37, Mike Cinquino wrote: Is it an issue that I do not have a machinekit folder in /usr/src/ ? No, this is a development machine, not a BBB running an image. The issue appears to be that you do not have libgcc_s.so.1 or it is not being found, as per below. On Sun, May 14, 2017 at 6:01 AM, schoone...@btinternet.com <schoone...@btinternet.com> wrote: Unfortunately I am not able to replicate your error. I have taken the zip config folder you attached, installed it into my configs and run that and it still runs OK. The xhc-hb04 binary is linked against libgcc_s.so.1 The binary is loaded from line 343 of xhc-hb04.tcl, which I assume is why the error occurs during the tcl file loads On my system the lib is in /usr/lib/gcc/x86_64-linux-gnu/4.9/  but yours will be slightly different but similar, you need to ensure it is physically present. Then run the command as below to see if it is in the ldconfig cache and you should get something similar user@INTEL-i7:/usr/src/machinekit/configs/sim/axis/xhc-hb04$ sudo ldconfig -p | grep libgcc_s.so.1     libgcc_s.so.1 (libc6,x32) => /usr/libx32/libgcc_s.so.1     libgcc_s.so.1 (libc6,x86-64) => /lib/x86_64-linux-gnu/libgcc_s.so.1     libgcc_s.so.1 (libc6) => /lib/i386-linux-gnu/libgcc_s.so.1     libgcc_s.so.1 (libc6) => /usr/lib32/libgcc_s.so.1 If you don't see it, run sudo ldconfig and then run the above again If you search on this error you will see it seems to occur quite often, when it appears the correct package is installed. It also turns up here https://groups.google.com/forum/#!topicsearchin/machinekit/libgcc_s.so.1;context-place=forum/machinekit/machinekit/fWz0MPrQeGw in a completely unrelated post, but again on a BBB, which makes me wonder. On 13/05/17 14:47, Mike Cinquino wrote: I ran sudo apt-get install libgcc1 and it said it was installed and nothing to add. It said to "autoremove" liburiparser1 so I did? I learned some usermod commands and checked and added machinekit user to "users" I created the .rules file as you described. Because I was getting the same error I experimented with the readme file version of the .rules file. I noticed it had ATTR vs SYSFS. The readme had the name as 90-xhc.rules. I tried both combinations of file names with ATTR and SYSFS. Rebooted a few times in the process and still the same error. libgcc_s.so.1 must be installed for pthread_cancel to work On Sat, May 13, 2017 at 7:32 AM, schoone...@btinternet.com <schoone...@btinternet.com> wrote: On 12/05/17 18:15, Mike Cinquino wrote: Is the attached what you are looking for? The pendant is being detected OK and the calls to 'halcmd show pexists <pin-name> are working The error is here libgcc_s.so.1 must be installed for pthread_cancel to work /usr/bin/linuxcnc: line 753:  9932 Aborted                 haltcl -i "$INIFILE" $CFGFILE That line of linuxcnc script deals with the reading in of tcl config files If they are the same as the ones in your zip earlier, they appear OK, so I would install libgcc1 sudo apt-get install libgcc1 check that your /etc/udev/rules.d/ contains 90-xhc-hb04.rules as per SYSFS{idProduct}=="eb70", SYSFS{idVendor}=="10ce", MODE="666", OWNER="root", GROUP="users" (unless your pendant comes back as different product:vendor codes) and that your user is a member of group 'users' Then try again. Works fine here -- website: http://www.machinekit.io blog: http://blog.machinekit.io github: https://github.com/machinekit --- You received this message because you are subscribed to a topic in the Google Groups "Machinekit" group. To unsubscribe from this topic, visit https://groups.google.com/d/topic/machinekit/OvxRsCF3auI/unsubscribe. To unsubscribe from this group and all its topics, send an email to machinekit+unsubscribe@googlegroups.com. Visit this group at https://groups.google.com/group/machinekit. For more options, visit https://groups.google.com/d/optout. -- website: http://www.machinekit.io blog: http://blog.machinekit.io github: https://github.com/machinekit --- You received this message because you are subscribed to the Google Groups "Machinekit" group. To unsubscribe from this group and stop receiving emails from it, send an email to machinekit+unsubscribe@googlegroups.com. Visit this group at https://groups.google.com/group/machinekit. For more options, visit https://groups.google.com/d/optout. -- website: http://www.machinekit.io blog: http://blog.machinekit.io github: https://github.com/machinekit --- You received this message because you are subscribed to a topic in the Google Groups "Machinekit" group. To unsubscribe from this topic, visit https://groups.google.com/d/topic/machinekit/OvxRsCF3auI/unsubscribe. To unsubscribe from this group and all its topics, send an email to machinekit+unsubscribe@googlegroups.com. Visit this group at https://groups.google.com/group/machinekit. For more options, visit https://groups.google.com/d/optout. -- website: http://www.machinekit.io blog: http://blog.machinekit.io github: https://github.com/machinekit --- You received this message because you are subscribed to a topic in the Google Groups "Machinekit" group. To unsubscribe from this topic, visit https://groups.google.com/d/topic/machinekit/OvxRsCF3auI/unsubscribe. To unsubscribe from this group and all its topics, send an email to machinekit+unsubscribe@googlegroups.com. Visit this group at https://groups.google.com/group/machinekit. For more options, visit https://groups.google.com/d/optout. -- website: http://www.machinekit.io blog: http://blog.machinekit.io github: https://github.com/machinekit --- You received this message because you are subscribed to the Google Groups "Machinekit" group. To unsubscribe from this group and stop receiving emails from it, send an email to machinekit+unsubscribe@googlegroups.com. Visit this group at https://groups.google.com/group/machinekit. For more options, visit https://groups.google.com/d/optout. -- website: http://www.machinekit.io blog: http://blog.machinekit.io github: https://github.com/machinekit --- You received this message because you are subscribed to a topic in the Google Groups "Machinekit" group. To unsubscribe from this topic, visit https://groups.google.com/d/topic/machinekit/OvxRsCF3auI/unsubscribe. To unsubscribe from this group and all its topics, send an email to machinekit+unsubscribe@googlegroups.com. Visit this group at https://groups.google.com/group/machinekit. For more options, visit https://groups.google.com/d/optout. -- website: http://www.machinekit.io blog: http://blog.machinekit.io github: https://github.com/machinekit --- You received this message because you are subscribed to the Google Groups "Machinekit" group. To unsubscribe from this group and stop receiving emails from it, send an email to machinekit+unsubscribe@googlegroups.com. Visit this group at https://groups.google.com/group/machinekit. For more options, visit https://groups.google.com/d/optout. -- website: http://www.machinekit.io blog: http://blog.machinekit.io github: https://github.com/machinekit --- You received this message because you are subscribed to a topic in the Google Groups "Machinekit" group. To unsubscribe from this topic, visit https://groups.google.com/d/topic/machinekit/OvxRsCF3auI/unsubscribe. To unsubscribe from this group and all its topics, send an email to machinekit+unsubscribe@googlegroups.com. Visit this group at https://groups.google.com/group/machinekit. For more options, visit https://groups.google.com/d/optout. -- website: http://www.machinekit.io blog: http://blog.machinekit.io github: https://github.com/machinekit --- You received this message because you are subscribed to the Google Groups "Machinekit" group. To unsubscribe from this group and stop receiving emails from it, send an email to machinekit+unsubscribe@googlegroups.com. Visit this group at https://groups.google.com/group/machinekit. For more options, visit https://groups.google.com/d/optout. -- website: http://www.machinekit.io blog: http://blog.machinekit.io github: https://github.com/machinekit --- You received this message because you are subscribed to a topic in the Google Groups "Machinekit" group. To unsubscribe from this topic, visit https://groups.google.com/d/topic/machinekit/OvxRsCF3auI/unsubscribe. To unsubscribe from this group and all its topics, send an email to machinekit+unsubscribe@googlegroups.com. Visit this group at https://groups.google.com/group/machinekit. For more options, visit https://groups.google.com/d/optout. -- website: http://www.machinekit.io blog: http://blog.machinekit.io github: https://github.com/machinekit --- You received this message because you are subscribed to the Google Groups "Machinekit" group. To unsubscribe from this group and stop receiving emails from it, send an email to machinekit+unsubscr...@googlegroups.com. Visit this group at https://groups.google.com/group/machinekit. For more options, visit https://groups.google.com/d/optout.