MLBEDSW-2388: Bug fix cascaded pooling
Kernel height was not correctly calculated for pooling
operations in rolling_buffer_dims_from_passes.
Change-Id: I48763b4b3276538c111e6699f66636327e569705
Signed-off-by: Louis Verhaard <louis.verhaard@arm.com>
diff --git a/ethosu/vela/npu_performance.py b/ethosu/vela/npu_performance.py
index eda8e42..57a72a6 100644
--- a/ethosu/vela/npu_performance.py
+++ b/ethosu/vela/npu_performance.py
@@ -29,6 +29,7 @@
from .nn_graph import PassPlacement
from .nn_graph import SchedulerRewrite
from .operation import NpuBlockType
+from .register_command_stream_generator import get_op_kernel
from .tensor import MemArea
from .tensor import shape_num_elements
from .tensor import TensorBlockTraversal
@@ -36,43 +37,24 @@
def rolling_buffer_dims_from_passes(arch, ps1, block_config_ps1, ps2, block_config_ps2):
- ps2_strides = (1, 1, 1, 1)
- ps2_dilation = (1, 1, 1, 1)
- for op in ps2.ops:
- if "strides" in op.attrs:
- ps2_strides = op.attrs["strides"]
- if "dilation" in op.attrs:
- ps2_dilation = op.attrs["dilation"]
-
- ifm_idx, _, weight_idx, _, _ = op.get_ifm_ifm2_weight_bias_ofm_indices()
-
- rolling_buffer_sizes = []
-
- weight_tensor = op.inputs[weight_idx]
-
ofm_block = Block(block_config_ps2[-3], block_config_ps2[-4], block_config_ps2[-1])
- kernel = Kernel(
- weight_tensor.shape[1], weight_tensor.shape[0], ps2_strides[2], ps2_strides[1], ps2_dilation[2], ps2_dilation[1]
- )
- kernel_block = Block(weight_tensor.shape[1], weight_tensor.shape[0], 65536)
+ kernel = get_op_kernel(ps2)
if ps2.npu_block_type in set((NpuBlockType.ConvolutionMxN, NpuBlockType.VectorProduct)):
+ op = ps2.primary_op
+ ifm_idx, _, _, _, _ = op.get_ifm_ifm2_weight_bias_ofm_indices()
ifm_block_depth = arch.calc_ifm_block_depth(
op.inputs[ifm_idx].shape[-1], op.inputs[ifm_idx].dtype.size_in_bits()
)
else:
ifm_block_depth = block_config_ps2[-1]
- ifm_block = arch.get_ifm_block_size(ifm_block_depth, ofm_block, kernel, kernel_block)
+ ifm_block = arch.get_ifm_block_size(ifm_block_depth, ofm_block, kernel, arch.ofm_block_max)
# The performed height calculation is for worst case
height = numeric_util.round_up(ifm_block.height + block_config_ps1[0], block_config_ps1[0])
width = ifm_block.width
-
- rolling_buffer_sizes.append(height)
- rolling_buffer_sizes.append(width)
-
- return rolling_buffer_sizes
+ return [height, width]
class PassCycles(enum.IntEnum):