98 questions
1
vote
1
answer
69
views
Local atomics causes GPU to crash
I am writing a OpenCL kernel that uses atomics. As I only need to synchronize groups of 192 threads, I figured using local atomics would be ideal. However, the change from global to local atomics ...
1
vote
0
answers
41
views
Unresolved extern function '__write_pipe_2' when building an OpenCL program
I'm using the OpenCL clBuildProgram() API function on a program created from a source string. The source is:
kernel void foo(int val, write_only pipe int outPipe)
{
write_pipe(outPipe, &val);
}...
- 137k
0
votes
1
answer
86
views
OpenCL createProgramWithSource doesn't work with a c-string declared in either global or function scope
I'm trying to run a basic kernel in OpenCL. See the snipped attached
const char kernel_source[] = "__kernel void matmul(__global float* A, __global float* B, __global float* C) { int row = ...
- 61
0
votes
1
answer
49
views
OpenCL 2.0 full profile, without atomic_store & atomic_load? Is this possible?
I use the OpenCL.NET C# wrapper for OpenCL.
My GPU from GPU-Z is AMD Radeon Barcelo, and specific for OpenCL:
Platform Version: OpenCL 2.1 AMD-APP (3570.0)
Device Name: gfx90c
Device Profile: ...
- 2,229
2
votes
1
answer
104
views
Why is fp32 performance only two percent of theoretical maximum of my RTX 2070 in my OpenCL program?
I've written a few OpenCL programs which perform integer arithmetic, and I usually manage to get somewhere near, sometimes more, than the theoretical maximum 32 bit 'flops' of the GPU for 32 bit ...
- 1,997
0
votes
1
answer
80
views
Can't seem to achieve anywhere near my GPU global memory bandwidth in OpenCL
Using opencl on my AMD GPU, I've only been able to achieve 4% (15 GB/sec) of the GPU global-memory bandwidth reported by clpeak (375 GB/sec).
Before resigning myself to this, I want to make sure I'm ...
- 115
2
votes
1
answer
85
views
OpenCL - How to suppress build errors from going to Standard Error?
In my application, I have a single OpenCL C program that gets built with several swappable modules that change parts of the code—notably, it changes some macros so that different arithmetic types are ...
- 19.9k
0
votes
1
answer
82
views
Safe GPU Programming
I recently learned how to program my AMD GPU using OpenCL in C. However, if I give the GPU a task that is too demanding, my entire system stops working properly and I have to reboot. I am using linux (...
- 3
1
vote
1
answer
185
views
OpenCL hangs forever when trying to read kernel output
This is a follow up to OpenCL Host ran out of Memory in trivial Kernel results that after applying the corrections and taking another errors I got that my program hangs at the EnqueueReadBuffer for an ...
- 353
2
votes
1
answer
67
views
OpenCL Host ran out of Memory in trivial Kernel
I am trying to do the kernel from in a sample learning program with 1024 entries of input buffer and output buffer both arrays of floating point entries of 32bits.
https://anteru.net/blog/2012/getting-...
- 353
1
vote
1
answer
109
views
Why does vectorialization of this simple openCl kernel make it slower?
I am currently implementing a naive parallelized index sum on gpu using OpenCl. This is part of a project that needs continuous stream compaction on large arrays, so I thought that it would be a good ...
- 23
1
vote
0
answers
48
views
Multi GPU clCreateBuffer failure on single context
I have two Intel ARC A770 GPUs and I am seeing this behavior (test code attached).
I am trying to allocate memory using clCreateBuffer (1 MB each).
I have two GPU devices.
The test code creates a ...
- 31
0
votes
1
answer
46
views
Summing a stream of records using OpenCL
We acquire data from an external device in the form of a continuous stream of "records" of 16-bit values. The record length is configured before the streaming starts and can vary between 20,...
- 10.3k
1
vote
0
answers
49
views
Opencl Program block occurred after the program running for a long time
We met a weird problem：The program worked as expected at start then got block after running for a long time.(10000 times or more).
Command queue：
clEnqueueWriteBuffer(CL_TRUE) * 3 times;
...
- 11
0
votes
1
answer
43
views
Problems additionVector with OpenCL
I want to learn OpenCL so i read a tutorial with a simple vector addition https://www.eriksmistad.no/getting-started-with-opencl-and-gpu-computing/
Im working with ubuntu
Distributor ID: Ubuntu
...
