You are not logged in.
- Topics: Active | Unanswered
#326 Re: Hardware & System Configuration » [SOLVED] Configuring CUDA support » 2018-09-07 16:26:25
I've found out a bit more. Adding --verbose to nvcc's parms makes it print the commands it runs to invoke lower level programs that do the work. Comparing what it does when called with -ccbin clang and with -ccbin clang-3.8 the only significant difference is the last line of output:
#$ clang++ -fPIC -fvisibility=hidden -O3 -m64 -o "sort_engine.so" -Wl,--start-group "/tmp/tmpxft_000002fc_00000000-18_sort_engine_dlink.o" "/tmp/tmpxft_000002fc_00000000-16_sort_engine.o" -shared -L/usr/lib/x86_64-linux-gnu/stubs -lcudadevrt -lcudart_static -lrt -lpthread -ldl -Wl,--end-group #$ clang-3.8 -fPIC -fvisibility=hidden -O3 -m64 -o "sort_engine.so" -Wl,--start-group "/tmp/tmpxft_0000029c_00000000-18_sort_engine_dlink.o" "/tmp/tmpxft_0000029c_00000000-16_sort_engine.o" -shared -L/usr/lib/x86_64-linux-gnu/stubs -lcudadevrt -lcudart_static -lrt -lpthread -ldl -Wl,--end-group It seems that when it calls clang++ the code works and when it calls clang-3.8 it fails. Which is reasonable if it's c++ code.
So I think it's nvcc that's going wrong. Not part of Devuan.
Chris
#327 Re: Devuan » We're migrating our distributed network to Devuan » 2018-09-05 18:11:22
* Umounting encrypted drives suddenly became intermittent. Something not reported in lsof sometimes has a hold on the mount point. Worked fine in jessie and now my automatic provisioners trip intermittently.
That could be something accessing the mount point for short intervals. So it's gone when you look for it with lsof. But finding out what it is could be difficult.
You might be able to get round that with a script that retries the unmount a few times if it fails. But that should not be necessary.
Chris
#328 Re: Desktop and Multimedia » In search of a privacy oriented browser » 2018-09-02 19:30:22
I've tried the version of chromium in Synaptic, it looks to be googled in that it keeps suggesting I sign on to google services.
But I'm not intending to use it. It's obviously designed to be mouse driven and lacks keyboard shortcuts. Which makes it very annoying when used with a full size keyboard in front of me and the mouse off to one side. At least in Konqueror I can press alt-B for bookmarks etc without needing to take my hands off the keyboard. All it needs is a cache to make it do all I want from a browser.
Chris
PS. iceweasel was a fork of firefox, it's been folded back now, the version in Synaptic would be firefox in all but name.
#329 Re: Installation » Services start issue » 2018-09-02 19:16:19
By default sudo will ask for a password the first time you use it, then allow you to use it without asking for a password for 15 minutes. The sudo service clears that so you always get asked for a password the first time you use sudo after a reboot.
You can change how long it remembers by setting timestamp_timeout in /etc/sudoers. Setting it to 0 should make sudo always ask for a password when you use it.
It's a good idea to have another way to get to root (eg su) when experimenting with sudo. So a typo in /etc/sudoers won't lock you out. And keep one terminal session running as root and use another to test sudo as your id.
I'm not sure why you had to add a % to the line in /etc/sudoers.d/[file_name] (mentioned in the other thread). In the line you showed changing groucho to %groucho should make it match members of the group called groucho instead of user groucho. But it's been several years since I've had to use sudo to do more than changing to root.
Chris
#330 Re: Installation » [Solved] Permisison issue with winecfg » 2018-08-31 06:50:50
Read the man page for sudoers. That will tell you everything you can do with it. Useful options include timestamp_timeout which sets how often it asks for your password. And it includes several examples.
For editing /etc/sudoers try EDITOR=jed visudo which should let you edit /etc/sudoers with jed and have it checked.
Chris
#331 Re: Installation » [Solved] Permisison issue with winecfg » 2018-08-30 18:27:56
Look in /etc/sudoers, there should be something like this:
# Allow members of group sudo to execute any command
%sudo ALL=(ALL:ALL) ALLIf not run visudo as root to add it (visudo edits a copy of /etc/sudoers with vi, then syntax checks your changes and copies it back if they are OK). If you don't know how to use vi then copying the above lines into /etc/sudoers with your favourite editor should work (I'd only recommend this if you can get to root without using sudo, else you could lock yourself out).
Then try sudo -l to list what sudo will let you do. And sudo -i to get to a root shell.
If it still doesn't work post output from sudo -l (lower case L), groups and id.
Chris
#332 Re: Desktop and Multimedia » [SOLVED] Nvidia error at boot » 2018-08-28 17:05:06
All I can suggest is raising a bug against udevd because the error message should say what the error was. Not just:
[ 0.861399] udevd[97]: Error running install command for nvidiaIn my previous career as a MVS systems programmer I went to some effort to ensure messages put out by programs I wrote had enough information to give a reasonable chance of working out what was wrong. It saved me a lot of work in the long run.
Chris
#333 Re: Hardware & System Configuration » [SOLVED] Configuring CUDA support » 2018-08-27 08:09:49
Hello,
I've found that the CUDA programs I'm using only work if complied with clang instead of clang-3.8. Which is surprising because they are the same program:
chris@rigel:~$ which clang
/usr/bin/clang
chris@rigel:~$ which clang-3.8
/usr/bin/clang-3.8
chris@rigel:~$ ls -l /usr/bin/clang /usr/bin/clang-3.8
lrwxrwxrwx 1 root root 25 Apr 18 2017 /usr/bin/clang -> ../lib/llvm-3.8/bin/clang
lrwxrwxrwx 1 root root 25 Jun 2 2017 /usr/bin/clang-3.8 -> ../lib/llvm-3.8/bin/clangComparing two test cases (test4 fails, test3 works):
chris@rigel:~/msieve-svn1022.test4/trunk/cub$ diff Makefile ~/msieve-svn1022.test3/trunk/cub/Makefile
20c20
< -ccbin clang-3.8 -Xcompiler -fPIC -Xcompiler -fvisibility=hidden
---
> -ccbin clang -Xcompiler -fPIC -Xcompiler -fvisibility=hiddenchris@rigel:~/msieve-svn1022.test4/trunk/cub$ ldd sort_engine.so
linux-vdso.so.1 (0x00007ffc90ffe000)
librt.so.1 => /lib/x86_64-linux-gnu/librt.so.1 (0x00007fa9cf8d2000)
libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 (0x00007fa9cf6b5000)
libdl.so.2 => /lib/x86_64-linux-gnu/libdl.so.2 (0x00007fa9cf4b1000)
libgcc_s.so.1 => /lib/x86_64-linux-gnu/libgcc_s.so.1 (0x00007fa9cf29a000)
libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007fa9ceefb000)
/lib64/ld-linux-x86-64.so.2 (0x00007fa9cfe30000)chris@rigel:~/msieve-svn1022.test3/trunk/cub$ ldd sort_engine.so
linux-vdso.so.1 (0x00007ffd177ef000)
librt.so.1 => /lib/x86_64-linux-gnu/librt.so.1 (0x00007f47b6618000)
libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 (0x00007f47b63fb000)
libdl.so.2 => /lib/x86_64-linux-gnu/libdl.so.2 (0x00007f47b61f7000)
libstdc++.so.6 => /usr/lib/x86_64-linux-gnu/libstdc++.so.6 (0x00007f47b5e75000)
libm.so.6 => /lib/x86_64-linux-gnu/libm.so.6 (0x00007f47b5b71000)
libgcc_s.so.1 => /lib/x86_64-linux-gnu/libgcc_s.so.1 (0x00007f47b595a000)
libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007f47b55bb000)
/lib64/ld-linux-x86-64.so.2 (0x00007f47b6b76000)The error message is:
cannot load library '/home/chris/msieve-svn1022.cuda/trunk/cub/sort_engine.so': /home/chris/msieve-svn1022.cuda/trunk/cub/sort_engine.so: undefined symbol: _ZNSt8ios_base4InitD1EvSo why does compiling with clang-3.8 produce different output from compiling with clang?
Chris
#334 Re: Desktop and Multimedia » In search of a privacy oriented browser » 2018-08-24 16:04:37
I've tested NetSurf, it has the same flaw as Konqueror in that it doesn't cache web pages you visit. So if you go back a page it re-fetches it, which is slow and loses your place on the page. Especially if you search for something in your favourite search engine, click one hit to read it, then go back the search is re-run.
In Konqueror I've taken to right-clicking and selecting open in new window so I just need to close the window to go back quickly. But that's a clumsy way to do it.
Firefox at least has a cache. But needs a lot of work and add-ons to make it respect privacy and security (don't trust javascript from arbitrary sites, don't leak what I'm doing to Mozilla, etc).
Chris
#335 Re: Hardware & System Configuration » Microcode to fight Spectre and Meltdown cpu flaws » 2018-08-24 15:52:39
The only really effective counter to Spectre is not to allow any untrustworthy code to run on your system. Or assume that any code running on it can read (but not update) everything in memory on it. There is no CPU on the market now where you can guarantee there is no exploitable side channel that would leak memory contents.
Chris
#336 Re: Desktop and Multimedia » In search of a privacy oriented browser » 2018-08-21 12:52:49
I've searched Synaptic package manager for web browsers and found:
chromium - web browser
conkeror - keyboard focused web browser with Emacs look and feel
dillo - Small and fast web browser
firefox-esr - Mozilla Firefox web browser - Extended Support Release (ESR) <--- installed on my system
hv3 - Lightweight web browser
iceweasel - Web browser based on Firefox - Transitional package
konqueror - advanced file manager, web browser and document viewer <--- I'm using this to make this post
links2 - Web browser running in both graphics and text mode
luakit - fast and small web browser extensible by Lua
midori - fast, lightweight graphical web browser
netsurf - small web browser with CSS support - transitional package
qupzilla - lightweight web browser based on libqtwebkit
surf2 - Simple web browser by suckless community. Version 2.
xombrero - minimalist web browser
And that's ignoring text based browsers.
So I'm a bit spoilt for choice looking for a privacy oriented browser. Which would be worth considering? Or should I ask which are not worth considering for privacy?
Chris
#337 Re: Hardware & System Configuration » [SOLVED] Configuring CUDA support » 2018-08-17 16:54:18
I've managed to get gmp-ecm GPU support working by recompiling it all with clang. See http://mersenneforum.org/showthread.php?t=23561 post 10.
So this issue is solved. How do I set the topic to say that?
Chris
#338 Re: Hardware & System Configuration » [SOLVED] Configuring CUDA support » 2018-08-15 17:06:44
Upgrading the OS from 4.9.0-6-amd64 to 4.9.0-7-amd64 fixed the driver problems. Now nvidia-smi works OK.
Testing nvcc I found it won't work with gcc 6.3.0. But it will work with clang 3.8 (and probably 3.9 or 4.0).
Now all I need to do is to get the apps I want to run under CUDA to work.
Chris
#339 Re: Hardware & System Configuration » Nvidia warning at boot » 2018-08-15 07:45:00
Thank you. The clue was you are running 4.9.0-7-amd64 while I was running 4.9.0-6-amd64. After apt-get update and apt-get dist-upgrade I get
$ nvidia-smi
Wed Aug 15 08:34:48 2018
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 384.130 Driver Version: 384.130 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 GeForce GTX 460 Off | 00000000:01:00.0 N/A | N/A |
| 40% 39C P3 N/A / N/A | 89MiB / 963MiB | N/A Default |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| 0 Not Supported |
+-----------------------------------------------------------------------------+Which is what I should get.
Chris
#340 Re: Hardware & System Configuration » [SOLVED] Configuring CUDA support » 2018-08-14 15:53:34
Hello,
Has anyone managed to get CUDA working on Devuan? If I can't get it working I'll reluctantly have to switch to another Linux distribution.
Chris
#341 Re: Hardware & System Configuration » Nvidia warning at boot » 2018-08-12 15:59:34
Could anyone running the nvidia driver without problems (either on ascii or an earlier release) please run the following commands (they don't need to be run as root) and post the output:
dpkg-query --search nvidia-current-drm
dpkg-query --search nvidia-current-uvm
dpkg-query --search nvidia-current-modeset
That should tell us what package provides them.
Chris
#342 Hardware & System Configuration » [SOLVED] Configuring CUDA support » 2018-08-11 14:13:11
- chris2be8
- Replies: 6
Hello,
I'm trying to get CUDA working on a system running Devuan. The GPU should support it:
$ nvidia-detect
Detected NVIDIA GPUs:
01:00.0 VGA compatible controller [0300]: NVIDIA Corporation GF104 [GeForce GTX 460] [10de:0e22] (rev a1)
Checking card: NVIDIA Corporation GF104 [GeForce GTX 460] (rev a1)
Your card is supported by all driver versions.
It is recommended to install the
nvidia-driver
package.nvidia-driver version 384.130-1 is installed.
But something seems to be missing:
# nvidia-smi
NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running.When rebooting the system I got messages about nvidia-current-modeset and nvidia-current-drm not being in /lib/modules/4.9.0-6-amd64 (I had to pause the boot with control-S, write them down, then control-Q to continue the boot). Similar messages come from:
# modprobe nvidia
modprobe: FATAL: Module nvidia-current not found in directory /lib/modules/4.9.0-6-amd64
modprobe: ERROR: ../libkmod/libkmod-module.c:977 command_do() Error running install command for nvidia
modprobe: ERROR: could not insert 'nvidia': Operation not permittedWhat packages provide nvidia-current-modeset and nvidia-current-drm?
Chris
Board footer
