Aart Bik 86e9bc1a34 [mlir][sparse] add option for 32-bit indices in scatter/gather ...

Controlled by a compiler option, if 32-bit indices can be handled
with zero/sign-extention alike (viz. no worries on non-negative
indices), scatter/gather operations can use the more efficient
32-bit SIMD version.

Reviewed By: bixia

Differential Revision: https://reviews.llvm.org/D103632

2021-06-04 16:57:12 -07:00

..

conversion.mlir

[mlir][sparse] add full dimension ordering support

2021-05-21 12:35:13 -07:00

invalid_encoding.mlir

[mlir][sparse] complete migration to sparse tensor type

2021-05-10 12:55:22 -07:00

invalid.mlir

[mlir][sparse] complete migration to sparse tensor type

2021-05-10 12:55:22 -07:00

roundtrip_encoding.mlir

[mlir][sparse] complete migration to sparse tensor type

2021-05-10 12:55:22 -07:00

roundtrip.mlir

[mlir][sparse] complete migration to sparse tensor type

2021-05-10 12:55:22 -07:00

sparse_1d.mlir

[mlir][sparse] complete migration to sparse tensor type

2021-05-10 12:55:22 -07:00

sparse_2d.mlir

[mlir][sparse] complete migration to sparse tensor type

2021-05-10 12:55:22 -07:00

sparse_3d.mlir

[mlir][sparse] skip sparsification for unannotated (or unhandled) cases

2021-05-19 13:49:28 -07:00

sparse_lower_col.mlir

[mlir][sparse] add full dimension ordering support

2021-05-21 12:35:13 -07:00

sparse_lower_inplace.mlir

[mlir][sparse] add full dimension ordering support

2021-05-21 12:35:13 -07:00

sparse_lower.mlir

[mlir][sparse] add full dimension ordering support

2021-05-21 12:35:13 -07:00

sparse_nd.mlir

[mlir][sparse] complete migration to sparse tensor type

2021-05-10 12:55:22 -07:00

sparse_parallel.mlir

[mlir][sparse] skip sparsification for unannotated (or unhandled) cases

2021-05-19 13:49:28 -07:00

sparse_storage.mlir

[mlir][sparse] complete migration to sparse tensor type

2021-05-10 12:55:22 -07:00

sparse_vector.mlir

[mlir][sparse] add option for 32-bit indices in scatter/gather

2021-06-04 16:57:12 -07:00