Untrusted GPGPU code (OpenCL etc) - is it safe? What risks?

Posted by Grzegorz Wierzowiecki on Stack Overflow See other posts from Stack Overflow or by Grzegorz Wierzowiecki
Published on 2011-01-08T19:40:35Z Indexed on 2011/01/08 19:54 UTC
Read the original article Hit count: 321

Filed under:

security

|

cuda

|

sandbox

|

opencl

|

GPGPU

There are many approaches when it goes about running untrusted code on typical CPU : sandboxes, fake-roots, virtualization...

What about untrusted code for GPGPU (OpenCL,cuda or already compiled one) ?

Assuming that memory on graphics card is cleared before running such third-party untrusted code,

are there any security risks?
What kind of risks?
Any way to prevent them ? (Possible sandboxing on gpgpu or other technique?)

P.S. I am more interested in gpu binary code level security rather than hight-level gpgpu programming language security (But those solutions are welcome as well). What I mean is that references to gpu opcodes (a.k.a machine code) are welcome.

© Stack Overflow or respective owner

Related posts about security

sudo apt-get update errors

as seen on Ask Ubuntu - Search for 'Ask Ubuntu'
Here is what I get on my terminal when running sudo apt-get update errors. I dont know if the issue is from my sources.list or my proxy setup(have not made any changes to proxies). Thank you for any help in advanced. Ign http://security.ubuntu.com oneiric-security Release.gpg Ign http://security… >>> More
The Security-Developer Gap: Why IT Security Can't Fix Your Security Problems Alone

as seen on Devx - Search for 'Devx'
Two well-known white hats explain how hackers take advantage of security holes left by enterprise application developers. >>> More
The Security-Developer Gap: Why IT Security Can't Fix Your Security Problems Alone

as seen on Internet.com - Search for 'Internet.com'
Two well-known white hats explain how hackers take advantage of security holes left by enterprise application developers. >>> More
StackOverFlowError while creating Mac object on AS400/Java

as seen on Stack Overflow - Search for 'Stack Overflow'
Hello all, I am a newbie to AS400-Java programming. I am trying to create my first program to test the implementation of Message Authentication Code (MAC). I am trying to use the HMACSHA1 hash function. My (Java 1.4) program runs fine on a dev box (V5R4).But fails terribly on the QA box (V5R3). My… >>> More
Google App Engine - Spring Security Issue (java.security.AccessControlException)

as seen on Stack Overflow - Search for 'Stack Overflow'
I'm currently getting the AccessControlException below when I deploy to app engine (I don't see it when I run in my local environment). I'm using GAE 1.3.1, Spring 3.0.1, and Spring Security 3.0.2. Any ideas how to get around this issue? It appears to be an issue with Spring Security trying to get… >>> More

Related posts about cuda

CUDA Driver API vs. CUDA runtime

as seen on Stack Overflow - Search for 'Stack Overflow'
When writing CUDA applications, you can either work at the driver level or at the runtime level as illustrated on this image (The libraries are CUFFT and CUBLAS for advanced math): I assume the tradeoff between the two are increased performance for the low-evel API but at the cost of increased… >>> More
Updating a Cuda 4.0 project to Cuda 4.2

as seen on Stack Overflow - Search for 'Stack Overflow'
I have a VS2010 project that was tested with CUDA 4.0, today I installed CUDA 4.2 and I want to update this project, the problem is that when I try to run the project it asks me for cudart32_40_17.dll, but since this is CUDA 4.2 I only have on my folders (C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v4… >>> More
How to solve CUDA crash when run CUDA example fluidsGL?

as seen on Ask Ubuntu - Search for 'Ask Ubuntu'
I use ubuntu 12.04 64 bits with GTX560Ti. I install CUDA by following instruction: wget http: //developer.download.nvidia.com/compute/cuda/4_2/rel/toolkit/cudatoolkit_4.2.9_lin ux_64_ubuntu11.04.run wget http: //developer.download.nvidia.com/compute/cuda/4_2/rel/drivers/devdriver_4… >>> More
Context migration in CUDA.NET

as seen on Stack Overflow - Search for 'Stack Overflow'
I'm currently using CUDA.NET library by GASS. I need to initialize cuda arrays (actually cublas vectors, but it doesn't matters) in one CPU thread and use them in other CPU thread. But CUDA context which holding all initialized arrays and loaded functions, can be attached to only one CPU thread. There… >>> More
CUDA on GeForce 8600GT

as seen on Super User - Search for 'Super User'
I have got the cuda driver, toolkit and sdk installed in Ubuntu 10.04. I'm using nVidia Geforce 8600 GT card. Official website says my card is CUDA supported. But on running the deviceQuery that comes with the cuda sdk, I'm getting the following output. ./deviceQuery Starting... CUDA Device Query… >>> More