Lines Matching refs:bufferize

48     analyze the entire IR and make bufferization decisions. Then, bufferize
70 then bufferize the remaining IR.
77 copied. The [bufferize] method of the op interface inspects analysis results and
123 Tensor ops that are not in destination-passing style always bufferize to a
193 [`-one-shot-bufferize`](https://mlir.llvm.org/docs/Passes/#-one-shot-bufferize-one-shot-bufferize)
212 One-Shot Bufferize can be configured to bufferize only ops from a set of
275 the `scf.if` example above would bufferize to IR similar to the following:
376 *   `bufferize`: Rewrite the op with the given rewriter. Ops should be replaced
394 1.  Due to a RaW conflict, it is not safe to bufferize in-place. I.e., the
409 can be used to bufferize a program in a single pass, as long as each op
424 run the existing `--func-bufferize` bufferization pass after One-Shot Bufferize.
486 be a pass `X-bufferize` that knows how to bufferize all the ops in that dialect.
487 By running pass `X-bufferize` for each dialect `X` in the program, all the ops
498 materializations. The pass `finalizing-bufferize` provides a minimal pass that
505 bufferize ops and eliminate materializations. This has a number of disadvantages
530 passes are bracketed by `arith-bufferize` and `func-bufferize`, which are module
535 passes are provided as part of the upstream MLIR distribution and bufferize
536 their respective dialects (e.g. `scf-bufferize` bufferizes the `scf` dialect).
537 The `tcp-bufferize` pass is an exception -- it is a partial bufferization pass
538 used to bufferize the downstream `tcp` dialect, and fits in perfectly with all
543 that the upstream `finalizing-bufferize` pass can be used as the finalizing
558 `tensor-bufferize` pass
560 …-project/blob/bc8acf2ce8ad6e8c9b1d97b2e02d3f4ad26e1d9d/mlir/test/Dialect/Tensor/bufferize.mlir#L1))
634 -   `scf-bufferize`
636 …lvm-project/blob/bc8acf2ce8ad6e8c9b1d97b2e02d3f4ad26e1d9d/mlir/test/Dialect/SCF/bufferize.mlir#L1))
639     -   This is an example of how to bufferize ops that implement
648 -   `func-bufferize`
650 …oject/blob/2f5715dc78328215d51d5664c72c632a6dac1046/mlir/test/Dialect/Func/func-bufferize.mlir#L1))
653     -   This is an example of how to bufferize ops that have multi-block
663 The easiest way to write a finalizing bufferize pass is to not write one at all!
664 MLIR provides a pass `finalizing-bufferize` which eliminates the
673 pattern, bug in the code, etc.), `finalizing-bufferize` will emit a nice clean
674 error, and the IR seen by `finalizing-bufferize` will only contain only one
680 simultaneously bufferize everything at once. Thus, one might see code in
689 -   `func-bufferize` was changed to be a partial conversion pass, and there is a
690     new `finalizing-bufferize` which serves as a general finalizing