QNN Partitioner - error with MatMul op

[metinsuloglu]

🐛 Describe the bug

When calling executorch.exir.backend.backend_api.to_backend with torch.matmul and the QnnPartitioner, it raises a KeyError.

import torch
from executorch.backends.qualcomm.utils.utils import (
    capture_program,
    generate_htp_compiler_spec,
    generate_qnn_executorch_compiler_spec,
)
from executorch.backends.qualcomm.partition.qnn_partitioner import (
    QnnPartitioner
)
from executorch.backends.qualcomm.serialization.qc_schema import QcomChipset
from executorch.exir import to_edge
from executorch.exir.backend.backend_api import to_backend


class MatMulModel(torch.nn.Module):
    def forward(self, inputs):
        return torch.matmul(*inputs)


def export(model, example_inputs, fname):
    model = torch.export.export(model, example_inputs, strict=True).module()

    edge_program = capture_program(model, example_inputs)

    backend_options = generate_htp_compiler_spec(use_fp16=True)
    qnn_partitioner = QnnPartitioner(
        generate_qnn_executorch_compiler_spec(
            soc_model=QcomChipset.SM8550,
            backend_options=backend_options,
        )
    )

    exported_program = to_backend(
        edge_program.exported_program,
        qnn_partitioner
    )
    exec_prog = to_edge(exported_program).to_executorch()

    with open(fname, "wb") as file:
        file.write(exec_prog.buffer)


model = MatMulModel().eval()
sample_inputs = ((torch.randn((1, 4, 16, 16)), torch.randn((16, 16))),)
export(model, sample_inputs, "qnn_matmul.pte")

Output:

[INFO] [Qnn ExecuTorch]: create QNN Logger with log_level 2
[WARNING] [Qnn ExecuTorch]:  <W> Initializing HtpProvider

[INFO] [Qnn ExecuTorch]: Initialize Qnn backend parameters for Qnn executorch backend type 2
[INFO] [Qnn ExecuTorch]: Caching: Caching is in SAVE MODE.
[WARNING] [Qnn ExecuTorch]:  <W> Performance Estimates unsupported

[WARNING] [Qnn ExecuTorch]:  <W> Arch 68 set by custom config is different from arch associated with SoC 43, will overwrite it to 73

[INFO] [Qnn ExecuTorch]: Running level=3 optimization.
[QNN Partitioner Op Support]: aten.view_copy.default | True
Traceback (most recent call last):
  File "/home/metsul01/code/executorch_bug/./py_scripts/matmul_ticket_2.py", line 45, in <module>
    export(model, sample_inputs, "qnn_matmul.pte")
  File "/home/metsul01/code/executorch_bug/./py_scripts/matmul_ticket_2.py", line 33, in export
    exported_program = to_backend(
  File "/usr/lib/python3.10/functools.py", line 889, in wrapper
    return dispatch(args[0].__class__)(*args, **kw)
  File "/home/metsul01/code/executorch_bug/executorch/exir/backend/backend_api.py", line 375, in _
    partitioner_result = partitioner_instance(fake_edge_program)
  File "/home/metsul01/code/executorch_bug/executorch/exir/backend/partitioner.py", line 66, in __call__
    return self.partition(exported_program)
  File "/home/metsul01/code/executorch_bug/executorch/backends/qualcomm/partition/qnn_partitioner.py", line 164, in partition
    partitions = self.generate_partitions(edge_program)
  File "/home/metsul01/code/executorch_bug/executorch/backends/qualcomm/partition/qnn_partitioner.py", line 133, in generate_partitions
    return generate_partitions_from_list_of_nodes(
  File "/home/metsul01/code/executorch_bug/executorch/exir/backend/canonical_partitioners/pattern_op_partitioner.py", line 50, in generate_partitions_from_list_of_nodes
    partition_list = capability_partitioner.propose_partitions()
  File "/home/metsul01/code/executorch_bug/venv/lib/python3.10/site-packages/torch/fx/passes/infra/partitioner.py", line 220, in propose_partitions
    if self.__is_node_supported(node) and node not in assignment:
  File "/home/metsul01/code/executorch_bug/venv/lib/python3.10/site-packages/torch/fx/passes/infra/partitioner.py", line 83, in __is_node_supported
    return self.operator_support.is_node_supported(
  File "/home/metsul01/code/executorch_bug/executorch/backends/qualcomm/partition/qnn_partitioner.py", line 83, in is_node_supported
    op_wrapper = self.node_visitors[node.target.__name__].define_node(
KeyError: 'aten.mm.default'
[INFO] [Qnn ExecuTorch]: Destroy Qnn backend parameters
[INFO] [Qnn ExecuTorch]: Destroy Qnn context
[INFO] [Qnn ExecuTorch]: Destroy Qnn device
[INFO] [Qnn ExecuTorch]: Destroy Qnn backend

Versions

Collecting environment information...
PyTorch version: 2.6.0.dev20250104+cpu
Is debug build: False
CUDA used to build PyTorch: Could not collect
ROCM used to build PyTorch: N/A

OS: Ubuntu 24.04.1 LTS (x86_64)
GCC version: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0
Clang version: 18.1.3 (1ubuntu1)
CMake version: version 3.31.4
Libc version: glibc-2.39

Python version: 3.10.16 (main, Dec 4 2024, 08:53:38) [GCC 13.2.0] (64-bit runtime)
Python platform: Linux-6.8.0-51-generic-x86_64-with-glibc2.39
Is CUDA available: False
CUDA runtime version: 12.0.140
CUDA_MODULE_LOADING set to: N/A
GPU models and configuration:
GPU 0: NVIDIA GeForce RTX 2080 Ti
GPU 1: NVIDIA GeForce RTX 2080 Ti

Nvidia driver version: 550.120
cuDNN version: Probably one of the following:
/usr/lib/x86_64-linux-gnu/libcudnn.so.8.9.7
/usr/lib/x86_64-linux-gnu/libcudnn.so.9.4.0
/usr/lib/x86_64-linux-gnu/libcudnn_adv.so.9.4.0
/usr/lib/x86_64-linux-gnu/libcudnn_adv_infer.so.8.9.7
/usr/lib/x86_64-linux-gnu/libcudnn_adv_train.so.8.9.7
/usr/lib/x86_64-linux-gnu/libcudnn_cnn.so.9.4.0
/usr/lib/x86_64-linux-gnu/libcudnn_cnn_infer.so.8.9.7
/usr/lib/x86_64-linux-gnu/libcudnn_cnn_train.so.8.9.7
/usr/lib/x86_64-linux-gnu/libcudnn_engines_precompiled.so.9.4.0
/usr/lib/x86_64-linux-gnu/libcudnn_engines_runtime_compiled.so.9.4.0
/usr/lib/x86_64-linux-gnu/libcudnn_graph.so.9.4.0
/usr/lib/x86_64-linux-gnu/libcudnn_heuristic.so.9.4.0
/usr/lib/x86_64-linux-gnu/libcudnn_ops.so.9.4.0
/usr/lib/x86_64-linux-gnu/libcudnn_ops_infer.so.8.9.7
/usr/lib/x86_64-linux-gnu/libcudnn_ops_train.so.8.9.7
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

CPU:
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Address sizes: 46 bits physical, 48 bits virtual
Byte Order: Little Endian
CPU(s): 20
On-line CPU(s) list: 0-19
Vendor ID: GenuineIntel
Model name: Intel(R) Core(TM) i9-10900X CPU @ 3.70GHz
CPU family: 6
Model: 85
Thread(s) per core: 2
Core(s) per socket: 10
Socket(s): 1
Stepping: 7
CPU(s) scaling MHz: 37%
CPU max MHz: 4700.0000
CPU min MHz: 1200.0000
BogoMIPS: 7399.70
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 cdp_l3 ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm mpx rdt_a avx512f avx512dq rdseed adx smap clflushopt clwb intel_pt avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local dtherm ida arat pln pts hwp hwp_act_window hwp_epp hwp_pkg_req vnmi avx512_vnni md_clear flush_l1d arch_capabilities
Virtualization: VT-x
L1d cache: 320 KiB (10 instances)
L1i cache: 320 KiB (10 instances)
L2 cache: 10 MiB (10 instances)
L3 cache: 19.3 MiB (1 instance)
NUMA node(s): 1
NUMA node0 CPU(s): 0-19
Vulnerability Gather data sampling: Mitigation; Microcode
Vulnerability Itlb multihit: KVM: Mitigation: VMX disabled
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Mmio stale data: Mitigation; Clear CPU buffers; SMT vulnerable
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed: Mitigation; Enhanced IBRS
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2: Mitigation; Enhanced / Automatic IBRS; IBPB conditional; RSB filling; PBRSB-eIBRS SW sequence; BHI SW loop, KVM SW loop
Vulnerability Srbds: Not affected
Vulnerability Tsx async abort: Mitigation; TSX disabled

Versions of relevant libraries:
[pip3] executorch==0.6.0a0+a5c7609
[pip3] numpy==2.2.2
[pip3] torch==2.6.0.dev20250104+cpu
[pip3] torchao==0.8.0+git11333ba2
[pip3] torchaudio==2.6.0.dev20250104+cpu
[pip3] torchsr==1.0.4
[pip3] torchvision==0.22.0.dev20250104+cpu
[pip3] triton==3.2.0
[conda] Could not collect

cc @cccclai @winskuo-quic @shewu-quic @digantdesai @freddan80 @per @zingo @oscarandersson8218

[digantdesai]

@ashutosh-arm - can you try to assign this to yourself?

[ashutosh-arm]

Thanks @digantdesai - it worked this time around 👍

[robell]

I've added the partner: qualcomm tag as this looks to be on the qnn backend.

[haowhsu-quic]

Hi @metinsuloglu, thank you for reporting this issue. Could you apply following patch and try again?

index c9215d11b..577bcb12a 100644
--- a/backends/qualcomm/builders/op_matmul.py
+++ b/backends/qualcomm/builders/op_matmul.py
@@ -15,7 +15,7 @@ from .qnn_constants import OpMatMul, QNN_OP_PACKAGE_NAME_QTI_AISW

 @register_node_visitor
 class Matmul(NodeVisitor):
-    target = ["aten.matmul.default"]
+    target = ["aten.matmul.default", "aten.mm.default"]

     def __init__(self, *args) -> None:
         super().__init__(*args)

[metinsuloglu]

Thanks @haowhsu-quic, it works with that patch.