Benjamin Maxwell
Benjamin Maxwell
Using the path rasterizer here is much slower than simply drawing four lines. This also more accurately shows the (real) clip as the bounding box is truncated to an int...
See related PR: https://github.com/llvm/llvm-project/pull/101645
This reduces the default AArch64 matmul tile sizes from (8, 16, 1) to (6, 16, 1). Originally, (8, 16, 1) was chosen to attempt to use all available vector registers...
**Demo Video:** This demo shows my intended workflow (which previously required a fair bit of copy/pasting): https://github.com/user-attachments/assets/0a01846f-15bf-48ba-81a8-b156e4be4b09 This adds (hacky) support for chaining the output from one compiler to the...
This enables the LLVM IR filters for the output of `mlir-translate` (with `-mlir-to-llvmir`). The filters are still disabled for `mlir-opt` (which only outputs MLIR) as there's currently no support for...
In streaming[-compatible] functions, use SVE for scalar FP conversions to/from integer types. This can help avoid moves between FPRs and GRPs, which could be costly. This patch also updates definitions...
See commits for details: Progression `rotated.pdf`: **Before**: **After**:
The previous SelectionDAG lowering is still available via `-aarch64-new-sme-abi=false` (this will stay around until at least LLVM 23). In tests that contained `CHECK-NEWLOWERING` the checks have been updated so: *...