Failed to load delegate from libedgetpu.so.1 M.2 Accelerator with Dual Edge TPU #611

tehniemer · 2022-06-14T20:08:29Z

Description

The image classifier test from https://www.coral.ai/docs/m2/get-started/ fails. I had this coral device installed and working in Frigate for months before it quit.

Hi @techniemer! Could you follow these things and let us know.

Ensure the Coral device is plugged in while inference is happening.

Check the permissions on your /dev/apex_0 device.

Add the user to plugdev group.

Attached issues 1 and 2 with similar error stack trace for reference.

Please post on Coral Edge Tpu support for further assistance.
Thank you!

Originally posted by @mohantym in tensorflow/tensorflow#56451 (comment)

This is an M.2 TPU and it is connected.
Permissions are 660, output below.
Added the user to that group with the same end result.

Click to expand!

Issue Type

Support

Operating System

Linux

Coral Device

M.2 Accelerator with dual Edge TPU

Other Devices

No response

Programming Language

Python 3.9

Relevant Log Output

user@nvr:~/coral/pycoral$ python3 examples/classify_image.py --model test_data/mobilenet_v2_1.0_224_inat_bird_quant_edgetpu.tflite --labels test_data/inat_bird_labels.txt --input test_data/parrot.jpg
Traceback (most recent call last):
  File "/usr/lib/python3/dist-packages/tflite_runtime/interpreter.py", line 160, in load_delegate
    delegate = Delegate(library, options)
  File "/usr/lib/python3/dist-packages/tflite_runtime/interpreter.py", line 119, in __init__
    raise ValueError(capture.message)
ValueError

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/user/coral/pycoral/examples/classify_image.py", line 121, in <module>
    main()
  File "/home/user/coral/pycoral/examples/classify_image.py", line 71, in main
    interpreter = make_interpreter(*args.model.split('@'))
  File "/usr/lib/python3/dist-packages/pycoral/utils/edgetpu.py", line 87, in make_interpreter
    delegates = [load_edgetpu_delegate({'device': device} if device else {})]
  File "/usr/lib/python3/dist-packages/pycoral/utils/edgetpu.py", line 52, in load_edgetpu_delegate
    return tflite.load_delegate(_EDGETPU_SHARED_LIB, options or {})
  File "/usr/lib/python3/dist-packages/tflite_runtime/interpreter.py", line 162, in load_delegate
    raise ValueError('Failed to load delegate from {}\n{}'.format(
ValueError: Failed to load delegate from libedgetpu.so.1

user@nvr:~/coral/pycoral$ ls /dev/apex_0
/dev/apex_0

user@nvr:/root$ lspci -nn | grep 089a
01:00.0 System peripheral [0880]: Global Unichip Corp. Coral Edge TPU [1ac1:089a]

root@nvr:~# dmesg | grep apex
[    3.313709] apex 0000:01:00.0: enabling device (0100 -> 0102)
[    3.314038] apex 0000:01:00.0: Couldn't initialize interrupts: -22
[    8.471106] apex 0000:01:00.0: Apex performance not throttled due to temperature

root@nvr:~# dmesg | grep 089a
[    0.315384] pci 0000:01:00.0: [1ac1:089a] type 00 class 0x0000ff

root@nvr:~# id user
uid=1000(user) gid=1000(user) groups=1000(user),24(cdrom),25(floppy),27(sudo),29(audio),30(dip),44(video),46(plugdev),107(render),109(netdev),1001(apex)

root@nvr:~# ls -l /dev/apex_0 
crw-rw---- 1 root apex 120, 0 Jun 13 19:28 /dev/apex_0

hjonnala · 2022-06-14T20:42:06Z

Hello @tehniemer can you please add the below lines here and share the logs.

from pycoral.pybind._pywrap_coral import SetVerbosity as set_verbosity
set_verbosity(10)

tehniemer · 2022-06-14T21:19:57Z

root@nvr:~/coral/pycoral# python3 examples/classify_image.py --model test_data/mobilenet_v2_1.0_224_inat_bird_quant_edgetpu.tf
lite --labels test_data/inat_bird_labels.txt --input test_data/parrot.jpg
I tflite/edgetpu_manager_direct.cc:453] No matching device is already opened for shared ownership.
I driver/usb/local_usb_device.cc:944] EnumerateDevices: vendor:0x1a6e, product:0x89a
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[2] port[0]
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[1] port[1]
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[1] port[0]
I driver/usb/local_usb_device.cc:944] EnumerateDevices: vendor:0x18d1, product:0x9302
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[2] port[0]
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[1] port[1]
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[1] port[0]
I tflite/edgetpu_context_direct.cc:106] USB always DFU: False (default)
I tflite/edgetpu_context_direct.cc:128] USB bulk-in queue capacity: default
I tflite/edgetpu_context_direct.cc:67] Performance expectation: Max (default)
I ./driver/mmio/host_queue.h:266] Starting in normal mode
I driver/kernel/kernel_registers.cc:83] Opening /dev/apex_0. read_only=0
I driver/kernel/kernel_registers.cc:97] mmap_offset=0x0000000000040000, mmap_size=4096
I driver/kernel/kernel_registers.cc:108] Got map addr at 0x0x7fb7ea1f0000
I driver/kernel/kernel_registers.cc:97] mmap_offset=0x0000000000044000, mmap_size=4096
I driver/kernel/kernel_registers.cc:108] Got map addr at 0x0x7fb7ea1ef000
I driver/kernel/kernel_registers.cc:97] mmap_offset=0x0000000000048000, mmap_size=4096
I driver/kernel/kernel_registers.cc:108] Got map addr at 0x0x7fb7e4dad000
I driver/kernel/kernel_registers.cc:211] Read: offset = 0x00000000000486f0, value: = 0x0000000000000000
I driver/kernel/kernel_registers.cc:190] Write: offset = 0x00000000000487a8, value = 0x0000000000000000
I driver/kernel/kernel_registers.cc:122] Closing /dev/apex_0. mmap_offset=0x0000000000040000, mmap_size=4096, read_only=0
I driver/kernel/kernel_registers.cc:122] Closing /dev/apex_0. mmap_offset=0x0000000000044000, mmap_size=4096, read_only=0
I driver/kernel/kernel_registers.cc:122] Closing /dev/apex_0. mmap_offset=0x0000000000048000, mmap_size=4096, read_only=0
I tflite/edgetpu_context_direct.cc:401] Failed to open device [Apex (PCIe)] at [/dev/apex_0]: Failed precondition: Could not p
artition page table. : 5 (Operation not permitted)
Traceback (most recent call last):
  File "/usr/lib/python3/dist-packages/tflite_runtime/interpreter.py", line 160, in load_delegate
    delegate = Delegate(library, options)
  File "/usr/lib/python3/dist-packages/tflite_runtime/interpreter.py", line 119, in __init__
    raise ValueError(capture.message)
ValueError

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/root/coral/pycoral/examples/classify_image.py", line 122, in <module>
    main()
  File "/root/coral/pycoral/examples/classify_image.py", line 72, in main
    interpreter = make_interpreter(*args.model.split('@'))
  File "/usr/lib/python3/dist-packages/pycoral/utils/edgetpu.py", line 87, in make_interpreter
    delegates = [load_edgetpu_delegate({'device': device} if device else {})]
  File "/usr/lib/python3/dist-packages/pycoral/utils/edgetpu.py", line 52, in load_edgetpu_delegate
    return tflite.load_delegate(_EDGETPU_SHARED_LIB, options or {})
  File "/usr/lib/python3/dist-packages/tflite_runtime/interpreter.py", line 162, in load_delegate
    raise ValueError('Failed to load delegate from {}\n{}'.format(
ValueError: Failed to load delegate from libedgetpu.so.1

hjonnala · 2022-06-14T21:33:51Z

can you please try the demo after performing step 3 and 4 at: https://coral.ai/docs/m2/get-started/#2a-on-linux

tehniemer · 2022-06-14T21:36:23Z

I followed that guide from start to finish, including the troubleshooting steps.

tehniemer · 2022-06-14T23:14:17Z

Here's the results from running under the user user

user@nvr:~/coral/pycoral$ python3 examples/classify_image.py --model test_data/mobilenet_v2_1.0_224_inat_bird_quant_edgetpu.tflite --labels test_data/inat_bird_labels.txt --input test_data/parrot.jpg
I tflite/edgetpu_manager_direct.cc:453] No matching device is already opened for shared ownership.
I driver/usb/local_usb_device.cc:944] EnumerateDevices: vendor:0x1a6e, product:0x89a
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[2] port[0]
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[1] port[1]
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[1] port[0]
I driver/usb/local_usb_device.cc:944] EnumerateDevices: vendor:0x18d1, product:0x9302
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[2] port[0]
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[1] port[1]
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[1] port[0]
I tflite/edgetpu_context_direct.cc:106] USB always DFU: False (default)
I tflite/edgetpu_context_direct.cc:128] USB bulk-in queue capacity: default
I tflite/edgetpu_context_direct.cc:67] Performance expectation: Max (default)
I ./driver/mmio/host_queue.h:266] Starting in normal mode
I driver/kernel/kernel_registers.cc:83] Opening /dev/apex_0. read_only=0
I driver/kernel/kernel_registers.cc:97] mmap_offset=0x0000000000040000, mmap_size=4096
I tflite/edgetpu_context_direct.cc:401] Failed to open device [Apex (PCIe)] at [/dev/apex_0]: Internal: Could not mmap: Operation not permitted
Traceback (most recent call last):
  File "/usr/lib/python3/dist-packages/tflite_runtime/interpreter.py", line 160, in load_delegate
    delegate = Delegate(library, options)
  File "/usr/lib/python3/dist-packages/tflite_runtime/interpreter.py", line 119, in __init__
    raise ValueError(capture.message)
ValueError

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/user/coral/pycoral/examples/classify_image.py", line 122, in <module>
    main()
  File "/home/user/coral/pycoral/examples/classify_image.py", line 72, in main
    interpreter = make_interpreter(*args.model.split('@'))
  File "/usr/lib/python3/dist-packages/pycoral/utils/edgetpu.py", line 87, in make_interpreter
    delegates = [load_edgetpu_delegate({'device': device} if device else {})]
  File "/usr/lib/python3/dist-packages/pycoral/utils/edgetpu.py", line 52, in load_edgetpu_delegate
    return tflite.load_delegate(_EDGETPU_SHARED_LIB, options or {})
  File "/usr/lib/python3/dist-packages/tflite_runtime/interpreter.py", line 162, in load_delegate
    raise ValueError('Failed to load delegate from {}\n{}'.format(
ValueError: Failed to load delegate from libedgetpu.so.1

hjonnala · 2022-06-15T17:08:21Z

can you please try the below options:

sudo python3 examples/classify_image.py
adding root to apex and plugdev groups.
sudo adduser root apex
sudo usermod -aG plugdev root

tehniemer · 2022-06-15T17:57:38Z

Same errors

user@nvr:~/coral/pycoral$ sudo python3 examples/classify_image.py 
usage: classify_image.py [-h] -m MODEL -i INPUT [-l LABELS] [-k TOP_K] [-t THRESHOLD] [-c COUNT] [-a INPUT_MEAN]
                         [-s INPUT_STD]
classify_image.py: error: the following arguments are required: -m/--model, -i/--input

user@nvr:~/coral/pycoral$ sudo python3 examples/classify_image.py --model test_data/mobilenet_v2_1.0_224_inat_bird_quant
_edgetpu.tflite --labels test_data/inat_bird_labels.txt --input test_data/parrot.jpg
[sudo] password for user: 
I tflite/edgetpu_manager_direct.cc:453] No matching device is already opened for shared ownership.
I driver/usb/local_usb_device.cc:944] EnumerateDevices: vendor:0x1a6e, product:0x89a
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[2] port[0]
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[1] port[1]
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[1] port[0]
I driver/usb/local_usb_device.cc:944] EnumerateDevices: vendor:0x18d1, product:0x9302
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[2] port[0]
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[1] port[1]
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[1] port[0]
I tflite/edgetpu_context_direct.cc:106] USB always DFU: False (default)
I tflite/edgetpu_context_direct.cc:128] USB bulk-in queue capacity: default
I tflite/edgetpu_context_direct.cc:67] Performance expectation: Max (default)
I ./driver/mmio/host_queue.h:266] Starting in normal mode
I driver/kernel/kernel_registers.cc:83] Opening /dev/apex_0. read_only=0
I driver/kernel/kernel_registers.cc:97] mmap_offset=0x0000000000040000, mmap_size=4096
I driver/kernel/kernel_registers.cc:108] Got map addr at 0x0x7f49d0417000
I driver/kernel/kernel_registers.cc:97] mmap_offset=0x0000000000044000, mmap_size=4096
I driver/kernel/kernel_registers.cc:108] Got map addr at 0x0x7f49d0416000
I driver/kernel/kernel_registers.cc:97] mmap_offset=0x0000000000048000, mmap_size=4096
I driver/kernel/kernel_registers.cc:108] Got map addr at 0x0x7f49ccfd4000
I driver/kernel/kernel_registers.cc:211] Read: offset = 0x00000000000486f0, value: = 0x0000000000000000
I driver/kernel/kernel_registers.cc:190] Write: offset = 0x00000000000487a8, value = 0x0000000000000000
I driver/kernel/kernel_registers.cc:122] Closing /dev/apex_0. mmap_offset=0x0000000000040000, mmap_size=4096, read_only=
0
I driver/kernel/kernel_registers.cc:122] Closing /dev/apex_0. mmap_offset=0x0000000000044000, mmap_size=4096, read_only=
0
I driver/kernel/kernel_registers.cc:122] Closing /dev/apex_0. mmap_offset=0x0000000000048000, mmap_size=4096, read_only=
0
I tflite/edgetpu_context_direct.cc:401] Failed to open device [Apex (PCIe)] at [/dev/apex_0]: Failed precondition: Could
 not partition page table. : 5 (Operation not permitted)
Traceback (most recent call last):
  File "/usr/lib/python3/dist-packages/tflite_runtime/interpreter.py", line 160, in load_delegate
    delegate = Delegate(library, options)
  File "/usr/lib/python3/dist-packages/tflite_runtime/interpreter.py", line 119, in __init__
    raise ValueError(capture.message)
ValueError

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/user/coral/pycoral/examples/classify_image.py", line 122, in <module>
    main()
  File "/home/user/coral/pycoral/examples/classify_image.py", line 72, in main
    interpreter = make_interpreter(*args.model.split('@'))
  File "/usr/lib/python3/dist-packages/pycoral/utils/edgetpu.py", line 87, in make_interpreter
    delegates = [load_edgetpu_delegate({'device': device} if device else {})]
  File "/usr/lib/python3/dist-packages/pycoral/utils/edgetpu.py", line 52, in load_edgetpu_delegate
    return tflite.load_delegate(_EDGETPU_SHARED_LIB, options or {})
  File "/usr/lib/python3/dist-packages/tflite_runtime/interpreter.py", line 162, in load_delegate
    raise ValueError('Failed to load delegate from {}\n{}'.format(
ValueError: Failed to load delegate from libedgetpu.so.1

root@nvr:~/coral/pycoral# id root
uid=0(root) gid=0(root) groups=0(root),46(plugdev),1001(apex)

root@nvr:~/coral/pycoral# python3 examples/classify_image.py --model test_data/mobilenet_v2_1.0_224_inat_bird_quant_edge
tpu.tflite --labels test_data/inat_bird_labels.txt --input test_data/parrot.jpg
I tflite/edgetpu_manager_direct.cc:453] No matching device is already opened for shared ownership.
I driver/usb/local_usb_device.cc:944] EnumerateDevices: vendor:0x1a6e, product:0x89a
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[2] port[0]
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[1] port[1]
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[1] port[0]
I driver/usb/local_usb_device.cc:944] EnumerateDevices: vendor:0x18d1, product:0x9302
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[2] port[0]
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[1] port[1]
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[1] port[0]
I tflite/edgetpu_context_direct.cc:106] USB always DFU: False (default)
I tflite/edgetpu_context_direct.cc:128] USB bulk-in queue capacity: default
I tflite/edgetpu_context_direct.cc:67] Performance expectation: Max (default)
I ./driver/mmio/host_queue.h:266] Starting in normal mode
I driver/kernel/kernel_registers.cc:83] Opening /dev/apex_0. read_only=0
I driver/kernel/kernel_registers.cc:97] mmap_offset=0x0000000000040000, mmap_size=4096
I driver/kernel/kernel_registers.cc:108] Got map addr at 0x0x7fdc5ae78000
I driver/kernel/kernel_registers.cc:97] mmap_offset=0x0000000000044000, mmap_size=4096
I driver/kernel/kernel_registers.cc:108] Got map addr at 0x0x7fdc5ae77000
I driver/kernel/kernel_registers.cc:97] mmap_offset=0x0000000000048000, mmap_size=4096
I driver/kernel/kernel_registers.cc:108] Got map addr at 0x0x7fdc57a35000
I driver/kernel/kernel_registers.cc:211] Read: offset = 0x00000000000486f0, value: = 0x0000000000000000
I driver/kernel/kernel_registers.cc:190] Write: offset = 0x00000000000487a8, value = 0x0000000000000000
I driver/kernel/kernel_registers.cc:122] Closing /dev/apex_0. mmap_offset=0x0000000000040000, mmap_size=4096, read_only=
0
I driver/kernel/kernel_registers.cc:122] Closing /dev/apex_0. mmap_offset=0x0000000000044000, mmap_size=4096, read_only=
0
I driver/kernel/kernel_registers.cc:122] Closing /dev/apex_0. mmap_offset=0x0000000000048000, mmap_size=4096, read_only=
0
I tflite/edgetpu_context_direct.cc:401] Failed to open device [Apex (PCIe)] at [/dev/apex_0]: Failed precondition: Could
 not partition page table. : 5 (Operation not permitted)
Traceback (most recent call last):
  File "/usr/lib/python3/dist-packages/tflite_runtime/interpreter.py", line 160, in load_delegate
    delegate = Delegate(library, options)
  File "/usr/lib/python3/dist-packages/tflite_runtime/interpreter.py", line 119, in __init__
    raise ValueError(capture.message)
ValueError

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/root/coral/pycoral/examples/classify_image.py", line 122, in <module>
    main()
  File "/root/coral/pycoral/examples/classify_image.py", line 72, in main
    interpreter = make_interpreter(*args.model.split('@'))
  File "/usr/lib/python3/dist-packages/pycoral/utils/edgetpu.py", line 87, in make_interpreter
    delegates = [load_edgetpu_delegate({'device': device} if device else {})]
  File "/usr/lib/python3/dist-packages/pycoral/utils/edgetpu.py", line 52, in load_edgetpu_delegate
    return tflite.load_delegate(_EDGETPU_SHARED_LIB, options or {})
  File "/usr/lib/python3/dist-packages/tflite_runtime/interpreter.py", line 162, in load_delegate
    raise ValueError('Failed to load delegate from {}\n{}'.format(
ValueError: Failed to load delegate from libedgetpu.so.1

hjonnala · 2022-06-16T21:01:36Z

[    8.471106] apex 0000:01:00.0: Apex performance not throttled due to temperature

can you please share the EdgeTPU junction temperature here using the Command below on LInux.

cat /sys/class/apex/apex_0/temp

tehniemer · 2022-06-16T21:09:06Z

root@nvr:~# cat /sys/class/apex/apex_0/temp
45050

hjonnala · 2022-06-16T21:20:01Z

Thanks! Also please share the output of below commands..

cat /sys/class/apex/apex_0/hw_temp_warn2
cat /sys/class/apex/apex_0/hw_temp_warn2_en

tehniemer · 2022-06-16T21:23:36Z

root@nvr:~# cat /sys/class/apex/apex_0/hw_temp_warn2
99800
root@nvr:~# cat /sys/class/apex/apex_0/hw_temp_warn2_en
1

hjonnala · 2022-06-16T21:30:58Z

root@nvr:~# cat /sys/class/apex/apex_0/hw_temp_warn2_en
1

can you try the demo changing this value to 0.

tehniemer · 2022-06-16T22:37:18Z

Sorry, how do I change that value?

hjonnala · 2022-06-16T23:01:45Z

Please try the command below as super user:
echo 0 > /sys/class/apex/apex_0/hw_temp_warn2_en

tehniemer · 2022-06-17T01:41:49Z

Looks like the same results.

root@nvr:~# cat /sys/class/apex/apex_0/hw_temp_warn2_en
0
root@nvr:~# cd coral/pycoral/
root@nvr:~/coral/pycoral# python3 examples/classify_image.py --model test_data/m                                                                           obilenet_v2_1.0_224_inat_bird_quant_edgetpu.tflite --labels test_data/inat_bird_                                                                           labels.txt --input test_data/parrot.jpg
I tflite/edgetpu_manager_direct.cc:453] No matching device is already opened for                                                                            shared ownership.
I driver/usb/local_usb_device.cc:944] EnumerateDevices: vendor:0x1a6e, product:0                                                                           x89a
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[2] port[0]
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[1] port[1]
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[1] port[0]
I driver/usb/local_usb_device.cc:944] EnumerateDevices: vendor:0x18d1, product:0                                                                           x9302
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[2] port[0]
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[1] port[1]
I driver/usb/local_usb_device.cc:979] EnumerateDevices: checking bus[1] port[0]
I tflite/edgetpu_context_direct.cc:106] USB always DFU: False (default)
I tflite/edgetpu_context_direct.cc:128] USB bulk-in queue capacity: default
I tflite/edgetpu_context_direct.cc:67] Performance expectation: Max (default)
I ./driver/mmio/host_queue.h:266] Starting in normal mode
I driver/kernel/kernel_registers.cc:83] Opening /dev/apex_0. read_only=0
I driver/kernel/kernel_registers.cc:97] mmap_offset=0x0000000000040000, mmap_siz                                                                           e=4096
I driver/kernel/kernel_registers.cc:108] Got map addr at 0x0x7f823456c000
I driver/kernel/kernel_registers.cc:97] mmap_offset=0x0000000000044000, mmap_siz                                                                           e=4096
I driver/kernel/kernel_registers.cc:108] Got map addr at 0x0x7f823456b000
I driver/kernel/kernel_registers.cc:97] mmap_offset=0x0000000000048000, mmap_siz                                                                           e=4096
I driver/kernel/kernel_registers.cc:108] Got map addr at 0x0x7f8231129000
I driver/kernel/kernel_registers.cc:211] Read: offset = 0x00000000000486f0, valu                                                                           e: = 0x0000000000000000
I driver/kernel/kernel_registers.cc:190] Write: offset = 0x00000000000487a8, val                                                                           ue = 0x0000000000000000
I driver/kernel/kernel_registers.cc:122] Closing /dev/apex_0. mmap_offset=0x0000                                                                           000000040000, mmap_size=4096, read_only=0
I driver/kernel/kernel_registers.cc:122] Closing /dev/apex_0. mmap_offset=0x0000                                                                           000000044000, mmap_size=4096, read_only=0
I driver/kernel/kernel_registers.cc:122] Closing /dev/apex_0. mmap_offset=0x0000                                                                           000000048000, mmap_size=4096, read_only=0
I tflite/edgetpu_context_direct.cc:401] Failed to open device [Apex (PCIe)] at [                                                                           /dev/apex_0]: Failed precondition: Could not partition page table. : 5 (Operatio                                                                           n not permitted)
Traceback (most recent call last):
  File "/usr/lib/python3/dist-packages/tflite_runtime/interpreter.py", line 160,                                                                            in load_delegate
    delegate = Delegate(library, options)
  File "/usr/lib/python3/dist-packages/tflite_runtime/interpreter.py", line 119,                                                                            in __init__
    raise ValueError(capture.message)
ValueError

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/root/coral/pycoral/examples/classify_image.py", line 122, in <module>
    main()
  File "/root/coral/pycoral/examples/classify_image.py", line 72, in main
    interpreter = make_interpreter(*args.model.split('@'))
  File "/usr/lib/python3/dist-packages/pycoral/utils/edgetpu.py", line 87, in ma                                                                           ke_interpreter
    delegates = [load_edgetpu_delegate({'device': device} if device else {})]
  File "/usr/lib/python3/dist-packages/pycoral/utils/edgetpu.py", line 52, in lo                                                                           ad_edgetpu_delegate
    return tflite.load_delegate(_EDGETPU_SHARED_LIB, options or {})
  File "/usr/lib/python3/dist-packages/tflite_runtime/interpreter.py", line 162,                                                                            in load_delegate
    raise ValueError('Failed to load delegate from {}\n{}'.format(
ValueError: Failed to load delegate from libedgetpu.so.1

hjonnala · 2022-06-17T18:59:34Z

Okay, please revert revert the hw_temp_warn2_en value to 1.

Can you please try the demo after unplug and plugin the device back to host machine, and restarting the machine.

Also please share the output of below command:

lspci -vvv|grep -i MSI-X

tehniemer · 2022-06-20T22:10:38Z

Same results after re-seating the device.

That command outputs nothing

user@nvr:~/coral/pycoral$ lspci -vvv|grep -i MSI-X
user@nvr:~/coral/pycoral$

tehniemer · 2022-06-20T22:11:20Z

nevermind, forgot to sudo

user@nvr:~/coral/pycoral$ sudo lspci -vvv|grep -i MSI-X
        Capabilities: [d0] MSI-X: Enable- Count=128 Masked-
        Capabilities: [b0] MSI-X: Enable- Count=4 Masked-
pcilib: sysfs_read_vpd: read failed: Input/output error
        Capabilities: [b0] MSI-X: Enable- Count=9 Masked-

hjonnala · 2022-06-20T22:28:27Z

I am not sure how to enable MSI-X support for your host system. It seems MSI-X is disabled on your host system, which is required for M.2/PCIE-e products:
https://docs.kernel.org/PCI/msi-howto.html#how-to-tell-whether-msi-msi-x-is-enabled-on-a-device

tehniemer · 2022-06-20T23:06:33Z

That was it. I had pci=nomsi in my grub.cfg file, removed that and all is well.

google-coral-bot · 2022-06-20T23:06:36Z

Are you satisfied with the resolution of your issue?
Yes
No

google-coral-bot bot assigned hjonnala Jun 14, 2022

google-coral-bot bot added comp:model Model related isssues Hardware:M.2 Accelerator with dual Edge TPU Coral M.2 Accelerator with Dual Edge TPU issues subtype:ubuntu/linux Ubuntu/Linux Build/installation issues type:support Support question or issue labels Jun 14, 2022

tehniemer mentioned this issue Jun 14, 2022

Failed to load delegate from libedgetpu.so.1 M.2 Accelerator with Dual Edge TPU tensorflow/tensorflow#56451

Closed

tehniemer closed this as completed Jun 20, 2022

tehniemer mentioned this issue Jul 6, 2022

[Support]: No EdgeTPU was detected, ValueError: Failed to load delegate from libedgetpu.so.1.0. Was working previously. blakeblackshear/frigate#3116

Closed

tehniemer mentioned this issue Aug 14, 2022

Second TPU on dual M.2 Accelerator not working, Failed to load delegate from libedgetpu.so.1 #641

Closed

hjonnala mentioned this issue Apr 26, 2023

ValueError: Failed to load delegate from libedgetpu.so.1 google-coral/tflite#63

Closed

adamkewley mentioned this issue May 11, 2023

S1: Fix libedgetpu not found PortableBalanceLab/PBL#66

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Failed to load delegate from libedgetpu.so.1 M.2 Accelerator with Dual Edge TPU #611

Failed to load delegate from libedgetpu.so.1 M.2 Accelerator with Dual Edge TPU #611

tehniemer commented Jun 14, 2022 •

edited

Loading

Issue Type

Operating System

Coral Device

Other Devices

Programming Language

Relevant Log Output

hjonnala commented Jun 14, 2022

tehniemer commented Jun 14, 2022

hjonnala commented Jun 14, 2022

tehniemer commented Jun 14, 2022

tehniemer commented Jun 14, 2022

hjonnala commented Jun 15, 2022

tehniemer commented Jun 15, 2022

hjonnala commented Jun 16, 2022

tehniemer commented Jun 16, 2022

hjonnala commented Jun 16, 2022

tehniemer commented Jun 16, 2022

hjonnala commented Jun 16, 2022

tehniemer commented Jun 16, 2022

hjonnala commented Jun 16, 2022

tehniemer commented Jun 17, 2022

hjonnala commented Jun 17, 2022

tehniemer commented Jun 20, 2022

tehniemer commented Jun 20, 2022

hjonnala commented Jun 20, 2022

tehniemer commented Jun 20, 2022

google-coral-bot bot commented Jun 20, 2022

Failed to load delegate from libedgetpu.so.1 M.2 Accelerator with Dual Edge TPU #611

Failed to load delegate from libedgetpu.so.1 M.2 Accelerator with Dual Edge TPU #611

Comments

tehniemer commented Jun 14, 2022 • edited Loading

Description

Issue Type

Operating System

Coral Device

Other Devices

Programming Language

Relevant Log Output

hjonnala commented Jun 14, 2022

tehniemer commented Jun 14, 2022

hjonnala commented Jun 14, 2022

tehniemer commented Jun 14, 2022

tehniemer commented Jun 14, 2022

hjonnala commented Jun 15, 2022

tehniemer commented Jun 15, 2022

hjonnala commented Jun 16, 2022

tehniemer commented Jun 16, 2022

hjonnala commented Jun 16, 2022

tehniemer commented Jun 16, 2022

hjonnala commented Jun 16, 2022

tehniemer commented Jun 16, 2022

hjonnala commented Jun 16, 2022

tehniemer commented Jun 17, 2022

hjonnala commented Jun 17, 2022

tehniemer commented Jun 20, 2022

tehniemer commented Jun 20, 2022

hjonnala commented Jun 20, 2022

tehniemer commented Jun 20, 2022

google-coral-bot bot commented Jun 20, 2022

tehniemer commented Jun 14, 2022 •

edited

Loading