5 Commits

Author SHA1 Message Date
Rob Suderman
cbd475040f [mlir][mlprogram] Add mlprogram-pipeline-globals optimization pass
Added pass optimizes MLProgram global operations by reducing to only
the minimal load/store operations for global tensors. This avoids
unnecessary global operations throughout a program and potentially
improves operation gusion.

Reviewed By: jpienaar

Differential Revision: https://reviews.llvm.org/D159228
2023-09-18 17:11:29 -07:00
Jacques Pienaar
d30c0221cf [mlir] Split MLProgram global load and store to Graph variants
* Split ops into X_graph variants as discussed;
* Remove tokens from non-Graph region variants and rely on side-effect
  modelling there while removing side-effect modelling from Graph
  variants and relying on explicit ordering there;
* Make tokens required to be produced by Graph variants - but kept
  explicit token type specification given previous discussion on this
  potentially being configurable in future;

This results in duplicating some code. I considered adding helper
functions but decided against adding an abstraction there early given
size of duplication and creating accidental coupling.

Differential Revision: https://reviews.llvm.org/D127813
2022-06-16 20:01:54 -07:00
Stella Laurenzo
3bb7999339 [mlir] Add global_load and global_store ops to ml_program.
* Adds simple, non-atomic, non-volatile, non-synchronized direct load/store ops.

Differential Revision: https://reviews.llvm.org/D126230
2022-06-01 11:32:15 -07:00
Stella Laurenzo
2bb252852c [mlir] Add GlobalOp, GlobalLoadConstOp to ml_program.
The approach I took was to define a dialect 'extern' attribute that a GlobalOp can take as a value to signify external linkage. I think this approach should compose well and should also work with wherever the OpaqueElements work goes in the future (since that is just another kind of attribute). I special cased the GlobalOp parser/printer for this case because it is significantly easier on the eyes.

In the discussion, Jeff Niu had proposed an alternative syntax for GlobalOp that I ended up not taking. I did try to implement it but a) I don't think it made anything easier to read in the common case, and b) it made the parsing/printing logic a lot more complicated (I think I would need a completely custom parser/printer to do it well). Please have a look at the common cases where the global type and initial value type match: I don't think how I have it is too bad. The less common cases seem ok to me.

I chose to only implement the direct, constant load op since that is non side effecting and there was still discussion pending on that.

Differential Revision: https://reviews.llvm.org/D124318
2022-05-18 23:08:28 -07:00
Stella Laurenzo
61352a580a [mlir] Introduce ml_program dialect.
Differential Revision: https://reviews.llvm.org/D120203
2022-04-13 21:38:14 -07:00