Hello,
I am running the deploy_model_on_android.py example on android, with vulkan enabled. Despite some error messages (that should be ignored), I get an output, but it is not the one expected:
[15:58:15] /home/SERILOCAL/n.perto/Documents/tvm/src/runtime/vulkan/vulkan_device_api.cc:693: Cannot initialize vulkan: [15:58:15] /home/SERILOCAL/n.perto/Documents/tvm/src/runtime/vulkan/vulkan_device_api.cc:614: Check failed: __e == VK_SUCCESS: Vulan Error, code=-3: VK_ERROR_INITIALIZATION_FAILED
Stack trace:
[bt] (0) /home/SERILOCAL/n.perto/Documents/tvm/build/libtvm.so(+0xa0b0ab) [0x7fe9810ee0ab]
[bt] (1) /home/SERILOCAL/n.perto/Documents/tvm/build/libtvm.so(+0xa0bf9b) [0x7fe9810eef9b]
[bt] (2) /home/SERILOCAL/n.perto/Documents/tvm/build/libtvm.so(+0xa14712) [0x7fe9810f7712]
[bt] (3) /home/SERILOCAL/n.perto/Documents/tvm/build/libtvm.so(+0x93497b) [0x7fe98101797b]
[bt] (4) /home/SERILOCAL/n.perto/Documents/tvm/build/libtvm.so(+0x2811ff) [0x7fe9809641ff]
[bt] (5) /home/SERILOCAL/n.perto/Documents/tvm/build/libtvm.so(+0x29d88e) [0x7fe98098088e]
[bt] (6) /home/SERILOCAL/n.perto/Documents/tvm/build/libtvm.so(+0x2a1471) [0x7fe980984471]
[bt] (7) /home/SERILOCAL/n.perto/Documents/tvm/build/libtvm.so(tvm::build(tvm::Map<tvm::Target, tvm::Array<tvm::LoweredFunc, void>, void, void> const&, tvm::Target const&, tvm::BuildConfig const&)+0x295) [0x7fe98098c255]
[bt] (8) /home/SERILOCAL/n.perto/Documents/tvm/build/libtvm.so(tvm::build(tvm::Map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, tvm::Array<tvm::LoweredFunc, void>, void, void> const&, tvm::Target const&, tvm::BuildConfig const&)+0x349) [0x7fe98098e899]
You can still compile vulkan module but cannot run locally
TVM prediction top-1: tench, Tinca tinca
Evaluate inference time cost...
Mean inference time (std dev): 693.35 ms (45.77 ms)
Most of the time I get tench but I sometimes have gotten something else. The expected result is cat.