Shared GPU memory is computed by CPU memory ÷ 2 And I want it to be equal to CPU memory. I need it for my pinned memory with cudaMallocHost.

I have a vague recollection that this 50% limit is baked into Windows, at least with the WDDM driver model where the operating system is in charge of GPU memory allocation. Note that when we speak about “shared memory” in the context of CUDA programming, we usually mean something entirely different…

How do I increase the shared GPU memory allocation multiplicator?

Accelerated Computing CUDA CUDA Programming and Performance

Topic		Replies	Views
How do I increase the VRAM capacity programmatically? CUDA Programming and Performance	4	2113	October 12, 2021
size of shrare memory between CPU and GPU Jetson TX2	7	1319	October 18, 2021
cudaMallocManaged do not allocate on shared VRAM, but on dedicated VRAM CUDA Programming and Performance	2	502	October 12, 2021
Ram become shared VEAM in ubuntu CUDA Setup and Installation cuda	0	289	October 22, 2024
Problem with cudaMallocHost CUDA Programming and Performance	3	7916	April 23, 2009
Dose cudaMalloc() increase host memory? CUDA Programming and Performance	0	694	February 28, 2014
pinned memory cannot be freed on one of multi-GPUs CUDA Programming and Performance	3	826	September 13, 2019
cudaMallocManaged allocating more memory than requested CUDA Programming and Performance	7	3219	July 13, 2018
sth wierd about managed memory and free GPU memeory size CUDA Programming and Performance	2	650	November 25, 2019
Multi GPU, Windows 10 pagefile and global memory issues CUDA Programming and Performance	3	2255	July 31, 2018

How do I increase the shared GPU memory allocation multiplicator?

Related topics