[onert] Introduce capabilities to find operands which can share memory #14228
Conversation
This commit adds capabilities to find operands linked which Tensors which can share memory buffers. It's applicable for ops like Reshape, Squeeze and ExpandsDims (where only shape is changed and data is not modified). ONE-DCO-1.0-Signed-off-by: Mateusz Bencer m.bencer@partner.samsung.com
| { | ||
| shared_memory_operand_map[op.getOutputs().at(0)] = op.getInputs().at(0); | ||
| } |
There was a problem hiding this comment.
(optional) How about adding assertions here to check is_memory_sharing_allowed only captures single input - single output operation?
Before reading line 53 (ops_with_possible_memory_sharing), I didn't notice that it targets single input, single output operation.
| { | |
| shared_memory_operand_map[op.getOutputs().at(0)] = op.getInputs().at(0); | |
| } | |
| { | |
| assert(op.getInputs().size() == 1); | |
| assert(op.getOutputs().size() == 1); | |
| shared_memory_operand_map[op.getOutputs().at(0)] = op.getInputs().at(0); | |
| } |
There was a problem hiding this comment.
added (note that for Reshape/ExpandDims number of inputs is 2)
There was a problem hiding this comment.
added (note that for Reshape/ExpandDims number of inputs is 2)
Oh, I see. I missed that. 👍
| for (auto [shared_ind, source_ind] : shared_memory_operand_map) | ||
| { | ||
| bool other_source_found = false; | ||
| auto it = std::end(shared_memory_operand_map); | ||
| while ((it = shared_memory_operand_map.find(source_ind)) != std::end(shared_memory_operand_map)) | ||
| { | ||
| source_ind = shared_memory_operand_map[source_ind]; | ||
| other_source_found = true; | ||
| } | ||
| if (other_source_found) | ||
| { | ||
| shared_memory_operand_map[shared_ind] = source_ind; | ||
| } | ||
| } |
There was a problem hiding this comment.
If shared_memory_operand_map = {{1, 2}, {2, 3}, {2, 4}} is given, is this works well..?
There was a problem hiding this comment.
As I understand 2 cannot indicate other operand two times. (I've made some experiments about this implementation here - https://godbolt.org/z/11KKr9xqY)
There was a problem hiding this comment.
Oh I see. I misunderstood the details.
Thank you for your implementations :)
My original question was that I it is safe with below graphs :
-> reshape -> output operand
/
operand --
\
-> other operation -> output operand
There was a problem hiding this comment.
ohh, I see, definitely it makes sense to write such unit test, thank you!
There was a problem hiding this comment.
I added such a test:
and it seems to work correctly (the memory is sharing)
| } | ||
| } // namespace | ||
|
|
||
| TEST(SharedMemoryOperands, no_shared_memory_graph) |
There was a problem hiding this comment.
| TEST(SharedMemoryOperands, no_shared_memory_graph) | |
| TEST(SharedMemoryOperands, neg_shared_memory_graph) |
@mbencer Please follow onert's unittest naming policy: negative test should have naming prefix neg_.
And negative test count should same or more than positive test.
There was a problem hiding this comment.
done, used naming prefix neg_ + added more tests. Now there are 4 positive and 4 negative tests
This commit adds capabilities to find operands linked which Tensors which can share memory buffers. It's applicable for ops like Reshape, Squeeze and ExpandsDims (where only shape is changed and data is not modified).
ONE-DCO-1.0-Signed-off-by: Mateusz Bencer m.bencer@partner.samsung.com
Draft: #14057