Re: [Emc-users] Emc-users Digest, Vol 105, Issue 33
I didn't say that latency would be any good, JT was asking about building the nvidia module regards On 19/01/2015 22:04, emc-users-requ...@lists.sourceforge.net wrote: Schooner: Please check your latency-test results with the NVidia drivers loaded and working. I found it very hard to backtrack to the nouveau or vesa drivers when my 8 or 9 microsecond latency-test results were replaced by several hundred milliseconds of lag. I was limited to about 2 ipm while doing sw stepping on an older athlon single core board. The motors sounded very ragged, and were simply were visibly stopping while glxgears was running. -- New Year. New Location. New Benefits. New Data Center in Ashburn, VA. GigeNET is offering a free month of service with a new server in Ashburn. Choose from 2 high performing configs, both with 100TB of bandwidth. Higher redundancy.Lower latency.Increased capacity.Completely compliant. http://p.sf.net/sfu/gigenet ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] Wheezy and Nvidia
Hi John No, you are right there is something screwy about the Wheezy distro and nvidia I ran the build manually on the 3.4.9-rtai-pae install, after the package install using DKMS failed (log attached) The crucial bit is at the end 'FATAL: modpost: GPL-incompatible module nvidia.ko uses GPL-only symbol 'ipipe_percpu'' If you google it, it can be caused by quite a few things from licence collisions to extra debug symbols interfering but basically it is a headers / kernel issue, they are not compatible with the nvidia module code This happens when there is at least one symbol in the kernel that is only usable for modules with the license GPL while the nvidia kernel module reports its license as nvidia This is not the case for the normal kernel but happens when additional debugging is enabled and seemingly also for the rt patch. You can build the nvidia module on the stock Debian rt-preempt kernel quite happily, it is not a rt 'per se' thing, depends entirely on the kernel config and symbols regards make -C /lib/modules/3.4-9-rtai-686-pae/source O=/lib/modules/3.4-9-rtai-686-pae/build M=`/bin/pwd` modules make[1]: Entering directory `/usr/src/linux-headers-3.4-9-common-rtai' CC [M] /usr/src/nvidia-304.125/nv.o In file included from /usr/src/linux-headers-3.4-9-common-rtai/include/linux/kernel.h:19:0, from /usr/src/linux-headers-3.4-9-common-rtai/include/linux/sched.h:55, from /usr/src/linux-headers-3.4-9-common-rtai/include/linux/utsname.h:35, from /usr/src/nvidia-304.125/nv-linux.h:42, from /usr/src/nvidia-304.125/nv.c:13: /usr/src/linux-headers-3.4-9-common-rtai/include/linux/bitops.h: In function ‘hweight_long’: /usr/src/linux-headers-3.4-9-common-rtai/include/linux/bitops.h:66:26: warning: signed and unsigned type in conditional expression [-Wsign-compare] In file included from /usr/src/linux-headers-3.4-9-common-rtai/arch/x86/include/asm/uaccess.h:575:0, from /usr/src/linux-headers-3.4-9-common-rtai/arch/x86/include/asm/sections.h:5, from /usr/src/linux-headers-3.4-9-common-rtai/arch/x86/include/asm/hw_irq.h:26, from /usr/src/linux-headers-3.4-9-common-rtai/include/linux/irq.h:374, from /usr/src/linux-headers-3.4-9-common-rtai/include/linux/ipipe.h:28, from /usr/src/linux-headers-3.4-9-common-rtai/include/linux/sched.h:65, from /usr/src/linux-headers-3.4-9-common-rtai/include/linux/utsname.h:35, from /usr/src/nvidia-304.125/nv-linux.h:42, from /usr/src/nvidia-304.125/nv.c:13: /usr/src/linux-headers-3.4-9-common-rtai/arch/x86/include/asm/uaccess_32.h: In function ‘copy_from_user’: /usr/src/linux-headers-3.4-9-common-rtai/arch/x86/include/asm/uaccess_32.h:208:6: warning: comparison between signed and unsigned integer expressions [-Wsign-compare] In file included from /usr/src/linux-headers-3.4-9-common-rtai/arch/x86/include/asm/elf.h:89:0, from /usr/src/linux-headers-3.4-9-common-rtai/include/linux/elf.h:7, from /usr/src/linux-headers-3.4-9-common-rtai/include/linux/module.h:14, from /usr/src/nvidia-304.125/nv-linux.h:81, from /usr/src/nvidia-304.125/nv.c:13: /usr/src/linux-headers-3.4-9-common-rtai/arch/x86/include/asm/desc.h: In function ‘alloc_system_vector’: /usr/src/linux-headers-3.4-9-common-rtai/arch/x86/include/asm/desc.h:358:3: warning: comparison between signed and unsigned integer expressions [-Wsign-compare] CC [M] /usr/src/nvidia-304.125/nv-acpi.o In file included from /usr/src/linux-headers-3.4-9-common-rtai/include/linux/kernel.h:19:0, from /usr/src/linux-headers-3.4-9-common-rtai/include/linux/sched.h:55, from /usr/src/linux-headers-3.4-9-common-rtai/include/linux/utsname.h:35, from /usr/src/nvidia-304.125/nv-linux.h:42, from /usr/src/nvidia-304.125/nv-acpi.c:15: /usr/src/linux-headers-3.4-9-common-rtai/include/linux/bitops.h: In function ‘hweight_long’: /usr/src/linux-headers-3.4-9-common-rtai/include/linux/bitops.h:66:26: warning: signed and unsigned type in conditional expression [-Wsign-compare] In file included from /usr/src/linux-headers-3.4-9-common-rtai/arch/x86/include/asm/uaccess.h:575:0, from /usr/src/linux-headers-3.4-9-common-rtai/arch/x86/include/asm/sections.h:5, from /usr/src/linux-headers-3.4-9-common-rtai/arch/x86/include/asm/hw_irq.h:26, from /usr/src/linux-headers-3.4-9-common-rtai/include/linux/irq.h:374, from /usr/src/linux-headers-3.4-9-common-rtai/include/linux/ipipe.h:28, from /usr/src/linux-headers-3.4-9-common-rtai/include/linux/sched.h:65, from /usr/src/linux-headers-3.4-9-common-rtai/include/linux/utsname.h:35, from
Re: [Emc-users] Wheezy and Nvidia
Hi John
I have a Nvidia GT630 GPU card working fine on Wheezy 32 64 bit
Works with nouveau and if you download the nvidia-glx package, DKMS
builds the driver fine (you need headers and a build environment)
What you then need to do however is add
/etc/X11/xorg.conf.d/20-nvidia.conf as attached
Then check that the driver has been installed to
/lib/modules/{kernel-version}/extra/nvidia.ko
and /etc/modprobe.d/nvidia-kernel-common.conf is present blacklisting
nouveau
The only problem I have had is when doing a load of stuff at the same
time after building the driver and re-making initrd-img, which somehow
screwed things up, but
redoing everything sorted it
regards
Mick
Section Device
Identifier My GPU
Driver nvidia
EndSection
alias char-major-195* nvidia
options nvidia NVreg_DeviceFileUID=0 NVreg_DeviceFileGID=44
NVreg_DeviceFileMode=0660
# To enable FastWrites and Sidebus addressing, uncomment these lines
# options nvidia NVreg_EnableAGPSBA=1
# options nvidia NVreg_EnableAGPFW=1
# see #580894
blacklist nouveau
--
New Year. New Location. New Benefits. New Data Center in Ashburn, VA.
GigeNET is offering a free month of service with a new server in Ashburn.
Choose from 2 high performing configs, both with 100TB of bandwidth.
Higher redundancy.Lower latency.Increased capacity.Completely compliant.
http://p.sf.net/sfu/gigenet___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] Regression in pyvcp's radiobutton from 2.5 to 2.6.5, nothing selected in the hal layer on startup
Hi I have tested your code on 2.6.3, 2.7 and 2.8 All of them produce a pyVCP panel with button 3 selected But no, the respective pin is not selected, I think I may know why. When I made a lot of alterations a while back to pyVCP, I changed the update code, so that it did not keep cycling through updates every call unless there was a change. This will have suppressed the first update. I expect no-one has noticed until now, because the initval will generally reflect the default setting rather than actually set it. Thereafter all clicks upon radio buttons will work as anticipated I have written a fix, which seems to work OK. Will do a bit more testing and submit a patch to correct this regards -- New Year. New Location. New Benefits. New Data Center in Ashburn, VA. GigeNET is offering a free month of service with a new server in Ashburn. Choose from 2 high performing configs, both with 100TB of bandwidth. Higher redundancy.Lower latency.Increased capacity.Completely compliant. http://p.sf.net/sfu/gigenet ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] Regression in pyvcp's radiobutton from 2.5 to 2.6.5, nothing selected in the hal layer on startup
Further to below If you want an immediate fix /usr/share/pyshared/pyvcp_widgets.py line 617 below self.selected = initval insert pycomp[self.halpins[initval]]=1 at same indentation That will force an update of the initval selected pin or defaults setting pin 0 if initval is not set, which is the default radio button setting at initialisation regards Hi I have tested your code on 2.6.3, 2.7 and 2.8 All of them produce a pyVCP panel with button 3 selected But no, the respective pin is not selected, I think I may know why. When I made a lot of alterations a while back to pyVCP, I changed the update code, so that it did not keep cycling through updates every call unless there was a change. This will have suppressed the first update. I expect no-one has noticed until now, because the initval will generally reflect the default setting rather than actually set it. Thereafter all clicks upon radio buttons will work as anticipated I have written a fix, which seems to work OK. Will do a bit more testing and submit a patch to correct this regards -- New Year. New Location. New Benefits. New Data Center in Ashburn, VA. GigeNET is offering a free month of service with a new server in Ashburn. Choose from 2 high performing configs, both with 100TB of bandwidth. Higher redundancy.Lower latency.Increased capacity.Completely compliant. http://p.sf.net/sfu/gigenet ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] My experiment with the wheezy based download may be over
Suggestions? Just run the 64 bit stock rt-preempt debian kernel on Wheezy and either build a RIP of the uspace build of LCNC or build MachineKit I have this on all my non-controller machines and it works fine. No need for a simulator build, just do a full build and run sim configs regards -- New Year. New Location. New Benefits. New Data Center in Ashburn, VA. GigeNET is offering a free month of service with a new server in Ashburn. Choose from 2 high performing configs, both with 100TB of bandwidth. Higher redundancy.Lower latency.Increased capacity.Completely compliant. http://p.sf.net/sfu/gigenet ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] Persistently storing HAL pin value
Hi Alex Seb wrote: If you're making a userspace component (as opposed to a realtime component), you can just use fopen/fwrite/fclose to persist the values when they change, and fopen/fread/fclose at startup to restore them. My component was called paramsaver, that is what it did, except that it created pins to match the data types, which could be linked to pyVCP widgets or whatever and it both saved and initialised values through the linked pins You can find it at mgwareDOTcoDOTuk HAL Components Paramsaver regards -- Dive into the World of Parallel Programming! The Go Parallel Website, sponsored by Intel and developed in partnership with Slashdot Media, is your hub for all things parallel software development, from weekly thought leadership blogs to news, videos, case studies, tutorials and more. Take a look and join the conversation now. http://goparallel.sourceforge.net ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] linuxcnc simulator on mint 17
After compiling, I execute the command ./scripts/linuxcnc inside the directory ~/linuxcnc-dev You should not have to use the full path Did you run make setuid after the build and then cd to ~/linuxcnc-dev and run . ./scripts/rip-environment ? Sounds like a problem with environment not being set. regards -- Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server from Actuate! Instantly Supercharge Your Business Reports and Dashboards with Interactivity, Sharing, Native Excel Exports, App Integration more Get technology previously reserved for billion-dollar corporations, FREE http://pubads.g.doubleclick.net/gampad/clk?id=164703151iu=/4140/ostg.clktrk ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] Not able to use comp (command not found)
Name changed to halcompile to prevent clashes with other packages and better describe what it does regards -- Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server from Actuate! Instantly Supercharge Your Business Reports and Dashboards with Interactivity, Sharing, Native Excel Exports, App Integration more Get technology previously reserved for billion-dollar corporations, FREE http://pubads.g.doubleclick.net/gampad/clk?id=164703151iu=/4140/ostg.clktrk ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
Re: [Emc-users] building a linuxcnc computer
1 of: ADATA Premier Series DDR3 1333Mhz 8 GB Laptop Memory Kit 2 x 4 GB CL9 Dual Channel SO-DIMM AD3S1333C4G9-2 Bearing in mind however that if you are running an x86 distro of linuxcnc, 4.75MB ish of that memory will not be able to be seen or used -- ___ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
