Troy Alderson
Troy Alderson
As previously discussed, I have been experimenting with optimizing BDSQR by using multiple kernel launches, with device synchronizations to determine the iterative loop's stopping condition. Broadly speaking, I have made...
This PR removes the loop unrolls from the small-size getri and trtri kernels, reducing the library size and allowing them to be built even if the -n flag is passed...
This is a prototype of my proposed changes to workspace management, which has been applied to getf2, getrf, getrs, gesv, geblttrf, geblttrs, and getri_outofplace. I recommend looking at getrf and...