masahi commented on issue #8294:
URL: https://github.com/apache/tvm/issues/8294#issuecomment-865432698
Got this error when running `test_convert_single_conv()` with cuda or opencl:
``` File
"/home/masa/projects/dev/tvm/python/tvm/topi/cuda/conv2d_winograd.py", line
100, in <lambda>
kernel[co][ci][r_kh][r_kw] * G[eps][r_kh] * G[nu][r_kw], axis=[r_kh,
r_kw]
File "/home/masa/projects/dev/tvm/python/tvm/tir/expr.py", line 77, in
__mul__
return _generic.multiply(self, other)
File "/home/masa/projects/dev/tvm/python/tvm/topi/generic_op_impl.py",
line 83, in _tensor_bop_impl
return orig_bop(lhs, rhs)
File "/home/masa/projects/dev/tvm/python/tvm/tir/generic.py", line 86, in
multiply
return _ffi_api._OpMul(lhs, rhs, span)
File "/home/masa/projects/dev/tvm/python/tvm/_ffi/_ctypes/packed_func.py",
line 237, in __call__
raise get_last_ffi_error()
3: TVMFuncCall
2: std::_Function_handler<void (tvm::runtime::TVMArgs,
tvm::runtime::TVMRetValue*), tvm::runtime::TypedPackedFunc<tvm::PrimExpr
(tvm::PrimExpr, tvm::PrimExpr,
tvm::Span)>::AssignTypedLambda<tvm::{lambda(tvm::PrimExpr, tvm::PrimExpr,
tvm::Span)#5}>(tvm::{lambda(tvm::PrimExpr, tvm::PrimExpr, tvm::Span)#5},
std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char>
>)::{lambda(tvm::runtime::TVMArgs const&,
tvm::runtime::TVMRetValue*)#1}>::_M_invoke(std::_Any_data const&,
tvm::runtime::TVMArgs&&, tvm::runtime::TVMRetValue*&&)
1: tvm::mul(tvm::PrimExpr, tvm::PrimExpr, tvm::Span)
0: tvm::BinaryOpMatchTypes(tvm::PrimExpr&, tvm::PrimExpr&, tvm::Span)
File "../src/tir/op/op.cc", line 144
TVMError: Cannot match type float16 vs float32
```
--
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.
For queries about this service, please contact Infrastructure at:
[email protected]
