Skip to main content
23 questions
Newest Active Bountied Unanswered
Filter by
Sorted by
Tagged with
1 vote
0 answers
87 views

Suppose I want to copy some memory between different CUDA contexts (possibly on different devices). The CUDA Driver API offers me: cuMemcpyPeer - for plain old device global memory cuMemcpy3DPeer - ...
einpoklum's user avatar
  • 137k
0 votes
2 answers
335 views

I'm trying to use the CUDA Driver API to copy data into a 2D array, in the program listed below, but am getting an "invalid value" error when I pass my copy parameters. What value in them is ...
einpoklum's user avatar
  • 137k
0 votes
1 answer
934 views

I am playing around with NVDEC H.264 decoder from NVIDIA CUDA samples, one thing I've found out is once frame is decoded, it's converted from NV12 to BGRA buffer which is allocated on CUDA's side, ...
Alex's user avatar
  • 582
0 votes
1 answer
1k views

I can find many examples online that use CUDA texture references, but not so many that rely on texture objects. I am trying to understand why my code below always fetches 0 rather than my input ...
Touloudou's user avatar
  • 2,293
0 votes
1 answer
600 views

Say you have a cuArray for binding a surface object. Something of the form: // These are inputs to a function really. cudaArray* d_cuArrSurf cudaSurfaceObject_t * surfImage; const cudaExtent extent ...
Ander Biguri's user avatar
  • 35.7k
2 votes
1 answer
618 views

I'm fairly new to julia and I'm currently trying out some deep convolution networks with recurrent structures. I'm training the networks on a GPU using CuArrays(CUDA Version 9.0). Having two ...
SeVe's user avatar
  • 432
16 votes
1 answer
4k views

What is the correct way of using a Vulkan VkImage as a CUDA cuArray? I've been trying to follow some examples, however I get a CUDA_ERROR_INVALID_VALUE on a call to ...
jsantander's user avatar
  • 5,142
1 vote
1 answer
529 views

I'm trying to process a video, frame by frame. To do this I want to create a texture containing the current frame and pass it to the kernel. The frames are 1440*1080 pixel with each pixel being ...
octavio's user avatar
  • 486
0 votes
1 answer
78 views

I have a JCuda project that's encountering an access violation whenever it tries to create a texture object using the driver API. Java HotSpot claims that the error is coming from nvcuda.dll. The ...
cfuqua's user avatar
  • 3
-1 votes
1 answer
245 views

I just want to put the test.jpg into the texture memory, and then through the text2D () to show the picture, but the results are very strange. The "result" should be the same as "gray".Click there to ...
biaodiluer's user avatar
  • 121
0 votes
1 answer
876 views

I have setup the cudaArray, and have bound it to a texture: texture<float, 2, cudaReadModeElementType> tex; cudaChannelFormatDesc channelDesc = cudaCreateChannelDesc(32, 0, 0, 0,...
Nick X Tsui's user avatar
  • 2,922
0 votes
1 answer
2k views

I wrote following code to see how to use texture memory for 1D array.but tex1D function is not fetching the value from array for corresponding thread id.Please correct this code and tell me how to use ...
Pramod Yelmewad's user avatar
  • 5
2 votes
2 answers
1k views

I am new to CUDA. I have figured out how to do 1D and 2D textures in CUDA. However, I am struggling with how to use a 1D layered texture. The output of my kernel which uses the texture is all zeros, ...
Rehman Ali's user avatar
  • 51
15 votes
2 answers
3k views

How to deal with OpenGL cube map textures in CUDA? When one want to use OpenGL textures in CUDA kernel one of the things to do is to retrieve a CUDA array from registered image and mapped resource, ...
Dorota Kadłubowska's user avatar
  • 675
1 vote
2 answers
1k views

Attempting to define a class with a per-instance texture. Yes, the number of instances of that class will be small. To work around the restriction that CUDA texture must be a global variable, I tried ...
user3367017's user avatar
  • 21
0 votes
1 answer
1k views

Non-normalized linear interpolation from a CUDA texture object bound to a CUDA array appears to be returning incorrect results. It appears that the interpolated values are a factor of 0.5 smaller than ...
nwhsvc's user avatar
  • 652
6 votes
1 answer
6k views

I'm learning cuda texture memory. Now, I got a opencv Iplimage, and I get its imagedata. Then I bind a texture to this uchar array, like below: Iplimage *image = cvCreateImage(cvSize(width, height), ...
hakunami's user avatar
  • 2,451
2 votes
2 answers
2k views

Im developing a small cuda lib, stuck by this annoying tex ref issue. This is the sample code from Cuda C Programming Guide, Page43~44: texture<float, cudaTextureType2D,cudaReadModeElementType>...
Defd's user avatar
  • 480
1 vote
2 answers
6k views

Is there a way to read the values in a cudaArray from the device without wrapping it in a texture reference/object? All of the examples I've looked at use cudaArrays exclusively for creating textures. ...
agrippa's user avatar
  • 379
0 votes
1 answer
2k views

I seem to have an issue with the function cudaMemcpyToArray. I have the following commands: float *h_data = new float[bmp.width * bmp.height]; ... cudaChannelFormatDesc channelDesc = ...
wolfPack88's user avatar
  • 4,213
2 votes
1 answer
1k views

I have a CUarray that I got from my OpenGL-context via cuGraphicsSubResourceGetMappedArray(). Is there a possiblity to use it with cuMemset*()?
morph's user avatar
  • 325
1 vote
4 answers
7k views

I have already read the following thread , but I couldn't get my code to work. I am trying to allocate a 2D array on GPU, fill it with values, and copy it back to the CPU. My code is as follows: ...
lina's user avatar
  • 1,697
7 votes
2 answers
9k views

Ok, so far, I can create an array on the host computer (of type float), and copy it to the gpu, then bring it back to the host as another array (to test if the copy was successful by comparing to the ...
hwrd's user avatar
  • 2,180