[NFC] Add comment to specify status of FusedKernelMatcher pass.

We're porting over this pass based on usage/need, so I'm updating the comments to make this clear.

PiperOrigin-RevId: 316553704
Change-Id: I5db9a4637edd3bcb27b4db6259b7113ae0f8dd0c
This commit is contained in:
Lucy Fox 2020-06-15 15:08:04 -07:00 committed by TensorFlower Gardener
parent 176ab11d0a
commit b59b2a10b1

View File

@ -32,11 +32,15 @@ namespace TF {
namespace {
// Note: This implements fusions performed in the old Remapper Grappler pass.
// That pass has specific cases for GPU and based on different target
// configurations on both CPU and GPU (Intel MKL, ROCm, etc.). This MLIR pass
// covers the general CPU case and at the moment does not account for any
// target-specific configurations.
// Note: This implements the fusions performed in the old Remapper Grappler
// pass. That pass has specific cases for GPU and based on different
// target configurations on both CPU and GPU (Intel MKL, ROCm, etc.). This MLIR
// pass covers (some of) the general CPU case and at the moment does not account
// for any target-specific configurations.
// This pass is being ported over from the Grappler Remapper pass based on
// need/usage. File a bug to request porting over additional fusions.
// TODO(b/158265178): Support GPU-specific fusions.
// TODO(b/158266710): Support CPU MKL configurations.