| Summary: | Template partial ordering considered non-ambiguous with deduced and non-deduced parameter packs | ||
|---|---|---|---|
| Product: | gcc | Reporter: | Barry Revzin <barry.revzin> |
| Component: | c++ | Assignee: | Not yet assigned to anyone <unassigned> |
| Status: | RESOLVED FIXED | ||
| Severity: | normal | CC: | daniel.kruegler, ed, marxin, webrown.cpp |
| Priority: | P3 | Keywords: | accepts-invalid |
| Version: | 7.1.0 | ||
| Target Milestone: | 10.2 | ||
| See Also: |
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67928 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85612 |
||
| Host: | Target: | ||
| Build: | Known to work: | 10.2.0, 11.1.0 | |
| Known to fail: | 10.1.0 | Last reconfirmed: | 2022-01-25 00:00:00 |
Got hit again today, while trying to write an `overload(...)` class:
template <typename... Ts>
struct overload : Ts...
{
template <typename... TFwds>
constexpr overload(TFwds&&... xs)
: Ts{FWD(xs)}...
{
}
using Ts::operator()...;
};
template <typename... TFwds>
overload(TFwds&&...) -> overload<std::decay_t<TFwds>...>;
With the code above
auto o = overload{[]{}};
fails to compile. See: https://godbolt.org/g/Rb3ZKt
See also #80438 Sorry. See also https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80438 Confirmed, though seems fixed for GCC 10.2.0 and 11.1.0; I suspect r11-1571 fixed it. Fixed with r11-1571-g57b4daf8dc4ed7b6. Fixed in GCC 10.2/11. |
Consider this example: #ifdef VARIADIC template <class... Ts, class... Us> constexpr int foo(Us... ) { return 1; } template <class... Us> constexpr int foo(Us... ) { return 2; } #else template <class... Ts, class U> constexpr int foo(U ) { return 1; } template <class U> constexpr int foo(U ) { return 2; } #endif int main() { static_assert(foo(1) == 1, "!"); } Without -DVARIADIC, the call to foo(1) is considered ambiguous and the program fails to compile. But with -DVARIADIC, the program compiles. The call should be ambiguous, since Ts... does not participate in partial ordering.