# AscVF passes
* [`-ascvf-eliminate-common-mask`](#ascvf-eliminate-common-mask)
* [`-ascvf-eliminate-data-transfer`](#ascvf-eliminate-data-transfer)
* [`-ascvf-find-vf-group`](#ascvf-find-vf-group)
* [`-ascvf-fuse-vf-for`](#ascvf-fuse-vf-for)
* [`-ascvf-insert-local-mem-bar`](#ascvf-insert-local-mem-bar)
* [`-ascvf-lower-to-micro`](#ascvf-lower-to-micro)
* [`-ascvf-materialize-load-store`](#ascvf-materialize-load-store)
* [`-ascvf-reorder-ops-in-vec-scope`](#ascvf-reorder-ops-in-vec-scope)
<!-- Autogenerated by mlir-tblgen; don't manually edit -->
## `-ascvf-eliminate-common-mask`

_Eliminates duplicate mask operations within VFGroupOp_

Removes redundant UpdateMaskOp and CreateMaskOp operations inside VFGroupOp
regions by replacing uses of duplicate masks with the first occurrence and
deleting the duplicates.
## `-ascvf-eliminate-data-transfer`

_Eliminates redundant load/store operations in VFGroupOp_

Performs three optimizations: eliminates loads after stores by replacing
the loaded value with the stored value, removes stores that are immediately
overwritten, and merges repeated loads from the same memory location.
## `-ascvf-find-vf-group`

_Finds and groups L2 operations into VFGroupOp for vector fusion_

Identifies groups of fusible L2 operations (binary, unary, reduce, duplicate)
that share the same calCount and element type, and wraps them into VFGroupOp
operations for subsequent vector fusion optimization.
## `-ascvf-fuse-vf-for`

_Fuses consecutive VFForOp loops within VecScopeOp_

Merges adjacent VFForOp loops into a single loop by copying operations from
subsequent loops into the first loop, reducing loop overhead.
## `-ascvf-insert-local-mem-bar`

_Inserts local memory barriers between consecutive VFForOp loops_

Inserts LocalMemBarOp operations between consecutive VFForOp loops within
VFGroupOp to synchronize vector store and vector load operations, ensuring
proper memory ordering for vector fusion execution.
## `-ascvf-lower-to-micro`

_Lowers L2 operations to micro-level operations with loops_

Converts L2 operations (binary, unary, reduce, duplicate) into micro-level
operations by wrapping them in VecScopeOp, creating VFForOp loops, allocating
register tensors, and inserting load/store micro operations.
## `-ascvf-materialize-load-store`

_Materializes abstract load/store micro ops into data copy operations_

Converts LoadMicroOp and StoreMicroOp operations into concrete data copy
operations with physical addresses, creating LocalTensorGetPhyAddrV2Op and
DataCopyLoadOp/DataCopyStoreOp operations.
## `-ascvf-reorder-ops-in-vec-scope`

_Hoists and reorders operations within VecScopeOp blocks_

Reorders operations inside VecScopeOp blocks by moving constants, register
tensors, variables, mask operations, and duplicate operations to the
beginning in a specific order for better code generation.