I have access to a server machine, with 3 CUDA enabled GPUs in it, and I would like to use NVidia Parallel Nsight, to remotly debug on the machine.
This works just find.
Now, is it possibble, to start another debug session (possibbly by another developer), on the same machine, but on another GPGPU?
Is it possibble, to do this, if I use gdb on linux?
Thanks,
krisy
Krisy, yes this is possible.
However this case/scenario that you mentioned has not been actively tested internally by the Nsight team yet. I tried this our real quick on a system with a similar setup as the one you mentioned and I was able to debug 2 different instances of CUDA app simulataneously (provided each app runs on a different unique device that is not connected to any output display).
The stability of this is not guaranteed. From what I've tried so far, this worked for me and it should work in theory as well but there were instances where I experienced sluggish behavior on my system.
For other developers who are interested to know more about this, please take a look at: http://forums.nvidia.com/index.php?showtopic=201211
Related
I am beginner. I want install freebsd on VM and test open source world! I want to write a small function and to put it into kernel of OS, and then I want to use it in another program as a system function.
I just installed freebsd11 on VM. I know a command line environment! I should use a GUI? Where is kernel?
FreeBSD is a wonderful beast once you know all its capabilities, if you want to play with jails, ZFS and build your own kernels, probably this already build image can be a good starting point:
https://fabrik.red/post/test/ more info can be found here including scripts about how to create your own images/jails (custom kernel): https://github.com/fabrik-red/images
There is no GUI on the images, and maybe you will never need one, unless you want to use FreeBSD as a desktop, but FreeBSD shines more on headless systems (no GUI).
Update:
For GUI probably you can give a try to trueos.org or you can install your own Xorg, desk environments
Sounds like you should learn a bit more about what the operating system does before trying to modify the kernel. There are lots of resources about the unix kernels. See the developers handbook, https://www.freebsd.org/doc/en/books/developers-handbook/ or just google and you will find many resources.
I have recently updated to Fedora 21 Beta, and updated all packages. As the final release is scheduled early next month, I don't expect anything big to change. So I think the problem I am facing may well persist into the production release.
The problem is that when using some terminal based applications, the terminal responses very slowly. For example, say I edit a file with Vim in terminal, after a few minutes, it becomes increasingly difficult to use. Every time I hit a keystroke, the cursor will wait like a second to respond. Edit the same file (which is of only a dozen of lines) in GVim, everything works as smoothly as expected. Other terminal based applications shows the same slow response. However, using the terminal itself as an interactive shell has no problem at all.
I understand it is very hard to come by an answer to why it is experiencing this kind of slow response based only on my vague description. However, if anyone can point to where I should look for diagnostics of the problem, some log files etc., it will highly appreciated.
Below are some more details of the system.
The computer is a Dell T3500 workstation with Xeon W3550 CPU, NVIDIA Quadro 2000 GPU. I am using the latest NVIDIA binary driver (both the long live version 340.58 as well as the beta series 346.16 are tried). The open-source driver was uninstalled and its kernel module were blacklisted (lsmod | grep nouveau shows nothing, as expected). The desktop environment is GNOME Classic.
I believe the issue is with the Nvidia driver configuration. I know this question is old, but I was looking for a solution to the same problem in Fedora 22. I experienced this issue when I was using Fedora 21, but I mostly put up with it.
The solution for me was to uncheck Sync to VBlank in the Nvidia X Server settings.
Are you running in a VM whose storage grows as you need it? Perhaps the problem is that the VM is taking time to allocate space as your storage needs grow. If that's the case, after a while you won't have the problem anymore, as your virtual disk will have grown as big as it needs to be.
As for diagnostics, try running "top", look for paging activity and resident set size.
Maybe something has a memory leak. It might not be in the terminal, but in one of its dependencies.
I'm writing a program that manages data entered by users. I plan to open a test version to the public and have no idea how many users there may be.
I want my program to test when memory is getting low so that I know when to buy more server space and so that I can automatically restrict data entry when necessary. What's a good way to detect memory shortage? Allocate garbage space temporarily to get the exception? Is there a better way?
This may be best accomplished outside of your application using a performance monitoring tool. Windows Server can be configured to do this for you; see this question. There are other tools out there that help you monitor your servers, and I advise you to use an existing system unless you absolutely have to do this with Python.
If you must absolutely do this using Python, then have a look at the psutil library:
psutil (python system and process utilities) is a cross-platform
library for retrieving information on running processes and system
utilization (CPU, memory, disks, network) in Python. It is useful
mainly for system monitoring, profiling and limiting process resources
and management of running processes. It implements many
functionalities offered by command line tools such as: ps, top, lsof,
netstat, ifconfig, who, df, kill, free, nice, ionice, iostat, iotop,
uptime, pidof, tty, taskset, pmap. It currently supports Linux,
Windows, OSX, FreeBSD and Sun Solaris, both 32-bit and 64-bit
architectures, with Python versions from 2.4 to 3.4. Pypi is also
known to work.
You may combine this with the email package to send the alerts.
I'm building a node-webkit app that needs to run on all 3 main desktop environments (windows, mac and linux) I need my app to connect to a plugged in USB device and I'm having a bit of trouble working out exactly how to go about this.
Is there an npm that would work across all OS's? Could I get one built in C++ that would work? Is there anything built in to node-webkit for interacting with devices (Devices API?)
Thanks in advance.
You're almost certain to need a C/C++ module to make this happen, which means a build process is required. The good news is that you can just do the build on your machine and distribute the resulting binaries (x86 and x64) for Windows and OS X. Linux might be a little trickier, but on the other hand it's more likely to have a working build environment.
There are a few approaches, in order of ease:
If your GPS device can present itself as a simple serial device (a COM port on Windows, or /dev/tty-usbserial on *nix), then you can just use serialport to connect to the device and receive raw data. There's a nmea module that can handle parsing the data. serialport works out of the box on all 3 platforms.
If the drivers don't have an official option to present the device as a serial device, you may be able to find an unsupported way to make it happen with some searching.
If the GPS device's drivers don't expose the device as a serial device, you'll need to write a C++ module that interfaces with the driver. You'll need to write code for all three platforms; the device manufacturer should provide documentation/an SDK. (See here for advice about getting started with native modules on Windows.)
Last resort: you may be able to use the usb module to communicate directly with the device. This will probably involve reverse engineering the protocol that the device uses to communicate with the computer.
I recommend you to execute a command line script for that. Caution, it is platform-specific.
Actually you don't need any module you can use chrome usb api its already available for node webkit and electron applications https://developer.chrome.com/apps/usb
Nvidia has released extended eclipse for CUDA 5. They have Nsight plugin for VS2010 also. In VS2010 we can stop program execution at breakpoint in kernel but how to achieve this functionality in eclipse on Linux? I don't see any nsight specific keys to stop execution. I tried changing perspective but it debugs as a normal C/C++ application. I'm using Tesla C2070, Intel Xeon 8 core machine with Linux.
I'm from Nsight Eclipse Edition team.
Our goal is specifically for the application to be debugged as a normal C/C++ application. This means that you can set breakpoints, use "run to line", etc. regardless of whether you debug host or device code.
Basically, the process is quite standard for Eclipse:
Create a project (you can also import existing executable)
Click debug button
Debugger will run and by default will break in the main function. Note that no device code posted on the device so you will only see the host thread.
Set a breakpoint in the device code and hit resume (note that Breakpoints view toolbar also allows you setting breakpoint on any CUDA kernel launch)
Debugger will break when device code reaches the breakpoint. You can inspect your application state using visual debugger UI.
Couple things, and not sure which solved the issue. Drivers updated to latest ones with RC5.0, but I chose to run VNC server instead of native X server. Then the CUDA card(s) are dedicated to my apps and debugging, and it works like a charm, and now accessible from everywhere.
Eugene,
I just installed Cuda 5, and I wasn't able to break in any kernel code. It was a clean install of centos 5.5, with a fresh download of cuda-5, and i am running on a asus g71x laptop which has a gtx260m installed.
I thought maybe you cant run display and dedbug on one device still, so i switched to non-nv x display, but still had same issue, cant stop in the kernel code.
Have you tried CUDA 5.0 RC1? It is available now. You can download and try it. And I have tried the Nsight in it, it works well for debugging.
Best regards!
The 304.43 NVIDIA Driver does not let users other than root debug their CUDA application.
That problem is not present in any past or future public releases. The CUDA documentation recommends using only drivers listed in the CUDA DevZone. The 304.43 driver is not one of them.
That may or may not be the issue you are hitting. But I thought it was worth mentioning.