alter-xp opened a new pull request, #12614: URL: https://github.com/apache/tvm/pull/12614
Integrate CSI-NN2 and add unit test for Conv2D [RFC] https://discuss.tvm.apache.org/t/coordination-of-risc-v-integration-in-tvm/13133 # 1. cmake Whether to build with CSI-NN2 codegen support Possible values: - OFF: disable csinn runtime support - ON: add support for compiling CSI-NN2 codegen. - path/to/CSI-NN2: use a specific version of the CSI-NN2 compute library set(USE_CSINN OFF) Whether to build with csinn runtime for device Possible values: - OFF: disable csinn runtime support - X86: compiling CSI-NN2 runtime for x86 device. - C906: cross-compiling CSI-NN2 runtime for C906 device. set(USE_CSINN_DEVICE_RUNTIME OFF) # 2. document - docs/how_to/deploy/csinn_compute_lib.rst - gallery/how_to/deploy_models/deploy_model_on_d1.py # 3. code - [relay op] python/tvm/relay/op/contrib/csinn.py - [backend] src/relay/backend/contrib/csinn/codegen.cc - [runtime] src/runtime/contrib/csinn/csinn_json_runtime.cc # 4. unit test tests/python/contrib/test_csinn/test_conv2d.py # 5. ci - [build csinn for x86] tests/scripts/task_config_build_csinn2.sh - [build csinn for c906] tests/scripts/task_config_build_c906.sh -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
