ncnn install on RPi4 Bullseye 32bit

[EvanBrownVTM]

Hi Qengineering,
I need your help understanding this issue when running make -j4 under ncnn/build:

/usr/bin/ld: ../src/libncnn.a(cpu.cpp.o): undefined reference to symbol 'dlopen@@GLIBC_2.4'

Thank you!!

[Qengineering]

Add -ldl to your link

[EvanBrownVTM]

Hi, thank you for your reply! I got the error running with the -ldl flag as in: cmake -D PI3=ON \ -D NCNN_DISABLE_RTTI=OFF \ -D NCNN_BUILD_TOOLS=ON \ -D CMAKE_EXE_LINKER_FLAGS=-ldl \ -D CMAKE_TOOLCHAIN_FILE=../toolchains/pi3.toolchain.cmake ..

…

On Tue, May 17, 2022 at 12:05 PM Q-engineering ***@***.***> wrote: Add -ldl to your link — Reply to this email directly, view it on GitHub <#5 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AXONPOOEVCGE56UGIP5K2TDVKPUXDANCNFSM5WFZRL4A> . You are receiving this because you authored the thread.Message ID: <Qengineering/Face-detection-Raspberry-Pi-32-64-bits/issues/5/1129218326@ github.com>

[Qengineering]

#3807
Just use

cmake -D PI3=ON \
-D NCNN_DISABLE_RTTI=OFF \
-D NCNN_BUILD_TOOLS=ON \
-D CMAKE_EXE_LINKER_FLAGS=-ldl  ..

[EvanBrownVTM]

So I tried running cmake with this command, but I still get the same output: /usr/bin/ld: ../src/libncnn.a(cpu.cpp.o): undefined reference to symbol 'dlopen@@GLIBC_2.4'

…

On Wed, May 18, 2022 at 4:24 AM Q-engineering ***@***.***> wrote: #3807 <Tencent/ncnn#3807> Just use cmake -D PI3=ON \ -D NCNN_DISABLE_RTTI=OFF \ -D NCNN_BUILD_TOOLS=ON \ -D CMAKE_EXE_LINKER_FLAGS=-ldl .. — Reply to this email directly, view it on GitHub <#5 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AXONPONEDIKZTL7ORKUG7C3VKTHO7ANCNFSM5WFZRL4A> . You are receiving this because you authored the thread.Message ID: <Qengineering/Face-detection-Raspberry-Pi-32-64-bits/issues/5/1129888258@ github.com>

[Qengineering]

Strange.
This morning I flashed a fresh Bullseye 32-bit OS onto an SD. Install it on a Raspberry Pi 4 with 2 GB of RAM.
Then followed the tutorial and got your error message.
Contacted Tencent/ncnn and followed nihui's instructions. It works well for me.
Tip: remove the whole ncnn folder and start all over again. $ sudo rm -rf ncnn

[EvanBrownVTM]

Thank you so much for your help. Everything is working now, ncnn and the face-landmark detector!!

…

On Wed, May 18, 2022 at 12:18 PM Q-engineering ***@***.***> wrote: Strange. This morning I flashed a fresh Bullseye 32-bit OS onto an SD. Install it on a Raspberry Pi 4 with 2 GB of RAM. Then followed the tutorial <https://qengineering.eu/install-ncnn-on-raspberry-pi-4.html> and got your error message. Contacted Tencent/ncnn and followed nihui's instructions. It works well for me. Tip: remove the whole ncnn folder and start all over again. $ sudo rm -rf ncnn — Reply to this email directly, view it on GitHub <#5 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AXONPOMH2T77VVK7XKN6UQ3VKU667ANCNFSM5WFZRL4A> . You are receiving this because you authored the thread.Message ID: <Qengineering/Face-detection-Raspberry-Pi-32-64-bits/issues/5/1130411866@ github.com>