Solve on the GPU #3
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Compute the solution on the GPU, one thread/kernel execution for every cell. The main solution code is in the `gpu` directory, this is the part compiled for the GPU. The `cpu` directory contains the host code that reads the input file, copies it to GPU memory and launches the GPU kernel with enough threads. Compiling --- The `ROCM_PATH` environment variable must be set to an installation of ROCm. When compiling the GPU crate, the hardware architecture must be specified through an environment variable, e.g. `CARGO_BUILD_RUSTFLAGS='-Ctarget-cpu=gfx900'`. Replace `gfx900` with your GPU as printed by e.g. `rocminfo` or the `cpu` executable. E.g. ```bash $ cd gpu $ CARGO_BUILD_RUSTFLAGS='-Ctarget-cpu=gfx900' cargo build --release $ cd ../cpu $ cargo run ../gpu/target/amdgcn-amd-amdhsa/release/gpu.elf ```
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Compute the solution on the GPU, one thread/kernel execution for every cell.
The main solution code is in the
gpudirectory, this is the part compiled for the GPU.The
cpudirectory contains the host code that reads the input file, copies it to GPU memory and launches the GPU kernel with enough threads.Compiling
The
ROCM_PATHenvironment variable must be set to an installation of ROCm.When compiling the GPU crate, the hardware architecture must be specified through an environment variable, e.g.
CARGO_BUILD_RUSTFLAGS='-Ctarget-cpu=gfx900'. Replacegfx900with your GPU as printed by e.g.rocminfoor thecpuexecutable.See also here for instructions: https://github.com/Flakebi/amdgpu-rs
E.g.