Problem with ADReal and std::pow

[joe61vette]

Check these boxes if you have followed the posting rules.

• I have consulted the Posting Guidelines.
• I have searched the Discussions Forum and MOOSE Framework Troubleshooting and have not found what I was looking for
• Q&A Getting Started is the most appropriate category for my question (trouble installing, beginner user, ...)

Issue or question about MOOSE

Just started working with MOOSE again after a 3 year hiatus. Using an M4 MacBook Air with OS 26.3. Just updated MOOSE & Conda, compiled and successfully ran all test problems. However, in my app, I tried to use some coding that used to work and I get a compile error. The coding is:

ADReal AirFluidProperties::k_from_p_T(const ADReal & /* p */, const ADReal & T) const { return _k_0 * std::pow(T, _k_exp); }

Note that _k_exp is Real. Also, AirFluidProperties inherits from SinglePhaseFluidProperties and the module was included in my makefile.

The error messages are:

In file included from /Users/joe/projects/sam_md/build/unity_src/fluidprops_Unity.C:2:
/Users/joe/projects/sam_md/src/fluidprops/AirFluidProperties.C:61:17: error: no matching function for call to 'pow'
61 | return _k_0 * std::pow(T, _k_exp);
| ^~~~~~~~
/Users/joe/miniforge/envs/moose/bin/../include/c++/v1/__math/exponential_functions.h:152:30: note: candidate function not viable: no known conversion from 'const ADReal' (aka 'const DualNumber<double, SemiDynamicSparseNumberArrayGeneric<double, unsigned long long, MetaPhysicL::NWrapper<64>, DynamicArrayWrapper<array, double, 64UL>>, true>') to 'double' for 1st argument
152 | _LIBCPP_HIDE_FROM_ABI double pow(double __x, double __y) _NOEXCEPT {
| ^ ~~~~~~~~~~
/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include/math.h:453:15: note: candidate function not viable: no known conversion from 'const ADReal' (aka 'const DualNumber<double, SemiDynamicSparseNumberArrayGeneric<double, unsigned long long, MetaPhysicL::NWrapper<64>, DynamicArrayWrapper<array, double, 64UL>>, true>') to 'double' for 1st argument
453 | extern double pow(double, double);
| ^ ~~~~~~
/Users/joe/miniforge/envs/moose/bin/../include/c++/v1/__math/exponential_functions.h:149:36: note: candidate function not viable: no known conversion from 'const ADReal' (aka 'const DualNumber<double, SemiDynamicSparseNumberArrayGeneric<double, unsigned long long, MetaPhysicL::NWrapper<64>, DynamicArrayWrapper<array, double, 64UL>>, true>') to 'float' for 1st argument
149 | inline _LIBCPP_HIDE_FROM_ABI float pow(float __x, float __y) _NOEXCEPT { return __builtin_powf(__x, __y); }
| ^ ~~~~~~~~~
/Users/joe/miniforge/envs/moose/bin/../include/c++/v1/__math/exponential_functions.h:156:42: note: candidate function not viable: no known conversion from 'const ADReal' (aka 'const DualNumber<double, SemiDynamicSparseNumberArrayGeneric<double, unsigned long long, MetaPhysicL::NWrapper<64>, DynamicArrayWrapper<array, double, 64UL>>, true>') to 'long double' for 1st argument
156 | inline _LIBCPP_HIDE_FROM_ABI long double pow(long double __x, long double __y) _NOEXCEPT {
| ^ ~~~~~~~~~~~~~~~
/Users/joe/projects/moose/framework/build/header_symlinks/Units.h:22:12: note: candidate function not viable: no known conversion from 'const ADReal' (aka 'const DualNumber<double, SemiDynamicSparseNumberArrayGeneric<double, unsigned long long, MetaPhysicL::NWrapper<64>, DynamicArrayWrapper<array, double, 64UL>>, true>') to 'const MooseUnits' for 1st argument
22 | MooseUnits pow(const MooseUnits &, int);
| ^ ~~~~~~~~~~~~~~~~~~
/Users/joe/miniforge/envs/moose/bin/../include/c++/v1/__math/exponential_functions.h:161:65: note: candidate template ignored: requirement 'is_arithmetic<MetaPhysicL::DualNumber<double, MetaPhysicL::SemiDynamicSparseNumberArrayGeneric<double, unsigned long long, MetaPhysicL::NWrapper<64>, MetaPhysicL::DynamicArrayWrapper<std::array, double, 64>>, true>>::value' was not satisfied [with _A1 = ADReal, _A2 = Real]
161 | inline _LIBCPP_HIDE_FROM_ABI typename __promote<_A1, _A2>::type pow(_A1 __x, _A2 __y) _NOEXCEPT {
| ^
/Users/joe/miniforge/envs/moose/bin/../include/c++/v1/complex:1098:43: note: candidate template ignored: could not match 'complex' against 'DualNumber'
1098 | inline _LIBCPP_HIDE_FROM_ABI complex<_Tp> pow(const complex<_Tp>& __x, const complex<_Tp>& __y) {
| ^
/Users/joe/miniforge/envs/moose/bin/../include/c++/v1/complex:1104:1: note: candidate template ignored: could not match 'complex' against 'DualNumber'
1104 | pow(const complex<_Tp>& __x, const complex<_Up>& __y) {
| ^
/Users/joe/miniforge/envs/moose/bin/../include/c++/v1/complex:1110:74: note: candidate template ignored: could not match 'complex' against 'DualNumber'
1110 | inline _LIBCPP_HIDE_FROM_ABI complex<typename __promote<_Tp, _Up>::type> pow(const complex<_Tp>& __x, const _Up& __y) {
| ^
/Users/joe/miniforge/envs/moose/bin/../include/c++/v1/complex:1116:74: note: candidate template ignored: could not match 'const complex<_Up>' against 'const Real' (aka 'const double')
1116 | inline _LIBCPP_HIDE_FROM_ABI complex<typename __promote<_Tp, _Up>::type> pow(const _Tp& __x, const complex<_Up>& __y) {
| ^
/Users/joe/miniforge/envs/moose/bin/../include/c++/v1/valarray:3266:1: note: candidate template ignored: requirement '__is_val_expr<MetaPhysicL::DualNumber<double, MetaPhysicL::SemiDynamicSparseNumberArrayGeneric<double, unsigned long long, MetaPhysicL::NWrapper<64>, MetaPhysicL::DynamicArrayWrapper<std::array, double, 64>>, true>>::value' was not satisfied [with _Expr1 = ADReal, _Expr2 = Real]
3266 | pow(const _Expr1& __x, const _Expr2& __y) {
| ^
/Users/joe/miniforge/envs/moose/bin/../include/c++/v1/valarray:3275:1: note: candidate template ignored: requirement '__is_val_expr<MetaPhysicL::DualNumber<double, MetaPhysicL::SemiDynamicSparseNumberArrayGeneric<double, unsigned long long, MetaPhysicL::NWrapper<64>, MetaPhysicL::DynamicArrayWrapper<std::array, double, 64>>, true>>::value' was not satisfied [with _Expr = ADReal]
3275 | pow(const _Expr& __x, const typename _Expr::value_type& __y) {
| ^
/Users/joe/miniforge/envs/moose/bin/../include/c++/v1/valarray:3284:1: note: candidate template ignored: requirement '__is_val_expr::value' was not satisfied [with _Expr = Real]
3284 | pow(const typename _Expr::value_type& __x, const _Expr& __y) {

I found similar coding in HeliumFluidProperties (part of fluid properties module):

ADReal HeliumFluidProperties::p_from_v_e(const ADReal & v, const ADReal & e) const { using std::pow; const ADReal T = T_from_v_e(v, e); return T / (48.14 * v - 0.4446 / pow(T, 0.2)) * 1.0e5; }

Compiled and tested fluidproperties. No compile error. So, I am stumped. Thanks for the help,
Joe Kelly

(Optional) code in question / simulation log / errors

No response

Encountering Errors? Please include diagnostic output

No response

[lindsayad]

MetaPhysicL no longer puts math function overloads in the std namespace. Instead they are in the MetaPhysicL namespace so that argument dependent lookup works. For the code you pasted here, you can just remove the std:: qualification and it will just work

[joe61vette]

Thanks Alex. Never would have thought of that after training myself to always use std.

[lindsayad]

No problem! What we were doing previously is actually "illegal" according to the C++ standard. You are not supposed to add to namespace std