codeflash-admin/codeflash
1
0
Fork 0
mirror of https://github.com/codeflash-ai/codeflash.git synced 2026-05-04 18:25:17 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

ready to review

Browse source
This commit is contained in:
aseembits93 2026-01-26 16:54:04 -08:00
parent 754eb6cc5e
commit ec3eed6b8a
1 changed files with 4 additions and 4 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

8
docs/codeflash-concepts/benchmarking-gpu-code.mdx
View file

@ -110,8 +110,8 @@ With synchronize: 152.277 ms
# How Codeflash measures execution time involving GPUs
Codeflash automatically inserts synchronization barriers before measuring performance. It currently supports GPU code written in `Pytorch`, `Tensorflow` and `JAX` for NVIDIA GPUs (CUDA) and MacOS Metal Performance Shaders (MPS).
Codeflash automatically inserts synchronization barriers before measuring performance. It currently supports GPU code written in `Pytorch`, `Tensorflow` and `JAX` for NVIDIA GPUs (`CUDA`) and MacOS Metal Performance Shaders (`MPS`).
- **PyTorch**: Uses `torch.cuda.synchronize()` (CUDA) or `torch.mps.synchronize()` (MPS) depending on the device.
- **JAX**: Uses `jax.block_until_ready()` to wait for computation to complete. It works for both CUDA and MPS devices.
- **TensorFlow**: Uses `tf.test.experimental.sync_devices()` for device synchronization. It works for both CUDA and MPS devices.
- **PyTorch**: Uses `torch.cuda.synchronize()` (`CUDA`) or `torch.mps.synchronize()` (`MPS`) depending on the device.
- **JAX**: Uses `jax.block_until_ready()` to wait for computation to complete. It works for both `CUDA` and `MPS` devices.
- **TensorFlow**: Uses `tf.test.experimental.sync_devices()` for device synchronization. It works for both `CUDA` and `MPS` devices.