Update dma-generate pass to (1) work on blocks of instructions (instead of just

loops), (2) take into account fast memory space capacity and lower 'dmaDepth'
to fit, (3) add location information for debug info / errors

- change dma-generate pass to work on blocks of instructions (start/end
  iterators) instead of 'for' loops; complete TODOs - allows DMA generation for
  straightline blocks of operation instructions interspersed b/w loops
- take into account fast memory capacity: check whether memory footprint fits
  in fastMemoryCapacity parameter, and recurse/lower the depth at which DMA
  generation is performed until it does fit in the provided memory
- add location information to MemRefRegion; any insufficient fast memory
  capacity errors or debug info w.r.t dma generation shows location information
- allow DMA generation pass to be instantiated with a fast memory capacity
  option (besides command line flag)

- change getMemRefRegion to return unique_ptr's
- change getMemRefFootprintBytes to work on a 'Block' instead of 'ForInst'
- other helper methods; add postDomInstFilter option for
  replaceAllMemRefUsesWith; drop forInst->walkOps, add Block::walkOps methods

Eg. output

$ mlir-opt  -dma-generate -dma-fast-mem-capacity=1 /tmp/single.mlir
/tmp/single.mlir:9:13: error: Total size of all DMA buffers' for this block exceeds fast memory capacity

        for %i3 = (d0) -> (d0)(%i1) to (d0) -> (d0 + 32)(%i1) {
            ^

$ mlir-opt -debug-only=dma-generate  -dma-generate -dma-fast-mem-capacity=400 /tmp/single.mlir
/tmp/single.mlir:9:13: note: 8 KiB of DMA buffers in fast memory space for this block

        for %i3 = (d0) -> (d0)(%i1) to (d0) -> (d0 + 32)(%i1) {

PiperOrigin-RevId: 232297044

1 files changed, 2 insertions, 3 deletions

diff --git a/mlir/lib/Transforms/MemRefDataFlowOpt.cpp b/mlir/lib/Transforms/MemRefDataFlowOpt.cpp
index 4191a9cc279..e6ce273b532 100644
--- a/mlir/lib/Transforms/MemRefDataFlowOpt.cpp
+++ b/mlir/lib/Transforms/MemRefDataFlowOpt.cpp
@@ -178,9 +178,8 @@ void MemRefDataFlowOpt::visitOperationInst(OperationInst *opInst) {
       // is trivially loading from a single location at that depth; so there
       // isn't a need to call isRangeOneToOne.
       if (getNestingDepth(*storeOpInst) < loadOpDepth) {
-        MemRefRegion region;
-        getMemRefRegion(loadOpInst, nsLoops, &region);
-        if (!region.getConstraints()->isRangeOneToOne(
+        auto region = getMemRefRegion(loadOpInst, nsLoops);
+        if (!region->getConstraints()->isRangeOneToOne(
                 /*start=*/0, /*limit=*/loadOp->getMemRefType().getRank()))
           break;
       }