- Migrate BoundCheckers and Simplify - Migrate RewriteUnsafeSelect and RemoveNoOp - Migrate UnrollLoop and StorageRewrite - Migrate InjectDoubleBuffer and InjectVirtualThread - Migrate LoopPartition and Vectorize - Migrate CoProcSync, LiftAttrScope, InjectCopyIntrin We still keep ir_pass registerations for now. Need a separate PR to refactor the parts before the StorageFlatten.
Name |
Last commit
|
Last update |
---|---|---|
.. | ||
arg_binder.cc | Loading commit data... | |
arg_binder.h | Loading commit data... | |
detect_device.cc | Loading commit data... | |
ffi_api.cc | Loading commit data... | |
hoist_if_then_else.cc | Loading commit data... | |
inject_prefetch.cc | Loading commit data... | |
inline.cc | Loading commit data... | |
ir_util.cc | Loading commit data... | |
ir_util.h | Loading commit data... | |
simple_passes.cc | Loading commit data... | |
ssa.cc | Loading commit data... | |
storage_access.cc | Loading commit data... | |
storage_access.h | Loading commit data... | |
storage_flatten.cc | Loading commit data... | |
tensor_core.cc | Loading commit data... | |
verify_compact_buffer.cc | Loading commit data... | |
verify_gpu_code.cc | Loading commit data... |