xla

xla copied to clipboard

PR #32738: [XLA:GPU] Allow cuDNN scaled dot fusions in the gemm autotuner Imported from GitHub PR https://github.com/openxla/xla/pull/32738 📝 Summary of Changes Allow selecting cuDNN gemm configs when autotuning scaled dot...

copybara-service[bot]

📝 Summary of Changes This PR adds a functional test checking for a particular failed gemm config on rocm platform. But this might also be useful on cuda. To detect...

pemeliya

[ROCM] Splitting gridDim.x for very large reductions / loop fusion kernels

5

We had several failures on ROCM side due to very large gridDim.x values for fusion kernels. Here is the proposal to split gridDim.x into gridDim.x/y in these cases. 1. added...

pemeliya