tree 949d787c5ae14b21fbd15bfddf7e7baf500114b8
parent 7fa8c205a0e8952a76268bc66975d69a813ee910
author Scott Todd <scotttodd@google.com> 1629915358 -0700
committer GitHub <noreply@github.com> 1629915358 -0700
gpgsig -----BEGIN PGP SIGNATURE-----
 
 wsBcBAABCAAQBQJhJojeCRBK7hj4Ov3rIwAAuHUIAKrq4InT6LiZE9pDsZOar1PT
 uy5gBs0IVdRdiX00C5EoeiZJ/ZyztUbkvFei6BiUsRg7LkppwkRAhthj0xKL8ODa
 ce2DCvtnDveEqNsPYxplPW3f8IWL01Nq0Erj1k3oPGfNSrKPi2280ux65df/ixp5
 qQvwCnn/aARPLLMV2JUtZ2Xt1c16krtWP59Na1Sp/lnBI31IlUhs9ESQaxnqQZp/
 L/5EUnano4rB0P+UP+oVDR7Cr9meSx011fFDEcdV39/wpyUr0rJ4E4mCdNfA2IFB
 Syi0reSub1uzGh70b7z8X2VjWxWtd1I7Z8GuoTQxLi8KwGU2WVFKIjp3zFqcqVA=
 =ROlL
 -----END PGP SIGNATURE-----
 

Remove PromoteTensorLoads pass, convert ExtractOp in TensorToFlow. (#6852)

Fixes https://github.com/google/iree/issues/6756 (the [`tosa if.mlir`](https://github.com/google/iree/blob/main/iree/test/e2e/tosa_ops/if.mlir) file compiles successfully using `-iree-flow-enable-linalg-detensorize` with this change)

The `PromoteTensorLoads` pass was converting `i1` loads to `i8` loads using `ZeroExtendIOp` and `TruncateIOp`. That was producing weird cycles during compilation when detensoring was applied, and `flow` ops should be fine with i1 types. We still need to handle `i1` types when going to the HAL (since storage is incompatible) on the outside (external interface) and inside (codegen).