mirror of
https://github.com/intel/intel-graphics-compiler.git
synced 2025-11-04 08:21:06 +08:00
6b8e1e040c
Move the PromoteToPredicatedMemoryAccess pass to the optimization stage of the compiler. This allows to keep standard LLVM passes to optimize the IR before the predication pass is applied. Change the pass to support scalarized loads and stores. Add a new pass to hoist conversion operations to the common dominator to unblock the predication pass. Fix generation of predicated stores, in case address is uniform and stored value is not.