[Bug] Problem generating Docker image for VitisAI

[fPecc]

I noticed that there is a problem with the docker image for Vitis AI, which I also noticed was detected by @areusch in #13435, but I saw there is no specific issue opened for this, so I wanted to open one with the problem I am experiencing:

Following the official guide from the documentation for the Vitis AI Integration (https://tvm.apache.org/docs/how_to/deploy/vitis_ai.html), the building of the docker container results in several errors. Steps completed so far:

git clone --recursive https://github.com/apache/tvm.git
cd tvm
./docker/build.sh demo_vitis_ai bash

First error is in line "RUN apt-get update --fix-missing", which requires the public key for kitware to be added to the apt key manager. This can be done by adding the command RUN apt-key adv --keyserver keyserver.ubuntu.com --recv-keys 6AF7F09730B3F0A4 before this line. The errors mentioned below were produced after adding this command:

Line 36: "RUN bash /install/ubuntu_install_python_package.sh" results in a dependency conflict:

ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
poetry 1.2.0b1 requires packaging<21.0,>=20.4, but you have packaging 23.0 which is incompatible.

This error doesn't interrupt the script.

Line 54: "RUN bash /install/ubuntu_install_python_package.sh" is run again, this time inside a virtual environment, which is activated in the line before with "RUN conda activate vitis-ai-tensorflow". Because this time it results in different dependency conflicts compared to the error mentioned above, we inspected the python version. Outside of the virtual environment python 3.7.5 is used, inside python 3.6.13 (Checked with "RUN python3 --version" right before and right after the "RUN conda activate vitis-ai-tensorflow" command in line 53). The python version 3.6 in the virtual environment results in following dependency conflicts:

• numpy :

  ERROR: Could not find a version that satisfies the requirement numpy==1.21.* (from versions: 1.3.0, 1.4.1, 1.5.0, 1.5.1, 1.6.0, 1.6.1, 1.6.2, 1.7.0, 1.7.1, 1.7.2, 1.8.0, 1.8.1, 1.8.2, 1.9.0, 1.9.1, 1.9.2, 1.9.3, 1.10.0.post2, 1.10.1, 1.10.2, 1.10.4, 1.11.0, 1.11.1, 1.11.2, 1.11.3, 1.12.0, 1.12.1, 1.13.0, 1.13.1, 1.13.3, 1.14.0, 1.14.1, 1.14.2, 1.14.3, 1.14.4, 1.14.5, 1.14.6, 1.15.0, 1.15.1, 1.15.2, 1.15.3, 1.15.4, 1.16.0, 1.16.1, 1.16.2, 1.16.3, 1.16.4, 1.16.5, 1.16.6, 1.17.0, 1.17.1, 1.17.2, 1.17.3, 1.17.4, 1.17.5, 1.18.0, 1.18.1, 1.18.2, 1.18.3, 1.18.4, 1.18.5, 1.19.0, 1.19.1, 1.19.2, 1.19.3, 1.19.4, 1.19.5)
  ERROR: No matching distribution found for numpy==1.21.*

• Pillow :

  ERROR: Could not find a version that satisfies the requirement Pillow==9.1.0 (from versions: 1.0, 1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7.0, 1.7.1, 1.7.2, 1.7.3, 1.7.4, 1.7.5, 1.7.6, 1.7.7, 1.7.8, 2.0.0, 2.1.0, 2.2.0, 2.2.1, 2.2.2, 2.3.0, 2.3.1, 2.3.2, 2.4.0, 2.5.0, 2.5.1, 2.5.2, 2.5.3, 2.6.0, 2.6.1, 2.6.2, 2.7.0, 2.8.0, 2.8.1, 2.8.2, 2.9.0, 3.0.0, 3.1.0rc1, 3.1.0, 3.1.1, 3.1.2, 3.2.0, 3.3.0, 3.3.1, 3.3.2, 3.3.3, 3.4.0, 3.4.1, 3.4.2, 4.0.0, 4.1.0, 4.1.1, 4.2.0, 4.2.1, 4.3.0, 5.0.0, 5.1.0, 5.2.0, 5.3.0, 5.4.0, 5.4.1, 6.0.0, 6.1.0, 6.2.0, 6.2.1, 6.2.2, 7.0.0, 7.1.0, 7.1.1, 7.1.2, 7.2.0, 8.0.0, 8.0.1, 8.1.0, 8.1.1, 8.1.2, 8.2.0, 8.3.0, 8.3.1, 8.3.2, 8.4.0)
  ERROR: No matching distribution found for Pillow==9.1.0

• pydot:

  ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
  pyxir 0.3.1 requires pydot==1.4.1, but you have pydot 1.4.2 which is incompatible.

• byoc:vitis_ai

[mshr-h]

Hi, I've met the same problem. I guess we have to ask Xilinx dev team to help to solve the problem because the below dependency hell.

• TVM supports Python 3.7 or above.
• TVM's Vitis-AI backend depends on pyxir which is maintained by Xilinx.
• pyxir depends on the decent_q module only in Xilinx's custom TensorFlow 1.15.
• Xilinx's custom TensorFlow 1.15 only works in Python 3.6.

There's a related issue in the pyxir repo. https://github.com/Xilinx/pyxir/issues/94