Minor API fixes for TTNN encoding ettribute #1390
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
mtopalovicTT
requested review from
sdjordjevicTT,
svuckovicTT,
rpavlovicTT,
nobradovictt,
jserbedzijaTT and
jnie-TT
as code owners
svuckovicTT
approved these changes
Nov 26, 2024
Comment on lines
112
to
114
| Here is how tensor will look like after layout tensor<32x32x64xf32, #ttnn.ttnn_layout<linear, grid, memref, mem_layout>> | ||
| Lets break down what each parameter means: | ||
| - affine_map: An affine map that defines how the logical tensor dimensions map to physical space. |
lib/Dialect/TTNN/IR/TTNNOps.cpp
Outdated
| @@ -190,25 +190,13 @@ ::mlir::LogicalResult mlir::tt::ttnn::EmptyOp::verify() { | |||
|
|
|||
| // DataType and Layout | |||
| // | |||
| mlir::MemRefType memref = layoutAttr.getMemref(); | |||
| Type elementType = memref.getElementType(); | |||
| // Type elementType = layoutAttr.getElementType(); | |||
There was a problem hiding this comment.
Yep.. Tnx for noticing
lib/Dialect/TTNN/IR/TTNNOpsAttrs.cpp
Outdated
Comment on lines
39
to
43
| if (isTiled()) { | ||
| return Layout::Tile; | ||
| } | ||
|
|
||
| return Layout::RowMajor; |
There was a problem hiding this comment.
Consider packing into a one-liner.
lib/Dialect/TTNN/IR/TTNNOpsAttrs.cpp
Outdated
| // Example: memref<128x128xf32> -> { 128, 128 } | ||
| // Example: memref<2x3!tt.tile<32x32xf32>> -> { 64, 96 } | ||
| // | ||
| // /return The scalar shape of the shard. |
There was a problem hiding this comment.
I'm not familiar with this /return, is it a docs thing?
There was a problem hiding this comment.
But I now see that I ill formed comments. Doxygen requires /// for comments in order to generate documentation.
Also return param should be prefixed with \ not /.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR adds couple convenience methods to TTNN tensor encoding attribute and also removes redundant utils functions.
Renaming/Adding some new functions...
getDataTypeto get scalar data type:memref<2x2x!tt.tile<32x32xf32>>returns float data typememref<128x128xi32>returns int data typegetElementTypeto get type from memref:memref<2x2x!tt.tile<32x32xf32>>returns TileTypememref<128x128xi32>returns IntegerTypegetLayout- gets layout of encoding i.e Tile/RowMajorgetShardShape:memref<2x2x!tt.tile<32x32xf32>>returns(2, 2)memref<128x128xi32>returns(128, 128)getScalarShardShape:memref<2x2x!tt.tile<32x32xf32>>returns(64, 64)memref<128x128xi32>returns(128, 128)