Added some math funcs implementation

[kuzminrobin]

The first step for #501
Resolves #508

[cgranade]

This looks reasonable to me, but I'm not an LLVM expert per se so I'll leave it to others to approve. In any case, it looks like the tests are all at the C level, but it would be good to have integration tests similar to

qsharp-runtime/src/QirRuntime/test/QIR-dynamic/qir-test-random-lnx.ll

Line 90 in 0cf4a5e

define i64 @Microsoft__Quantum__Testing__QIR__QuantumRandomNumberGenerator__body() #0 {

that ensure that the new intrinsic works correctly when called from QIR files generated by the Q# compiler.

[kuzminrobin]

This looks reasonable to me, but I'm not an LLVM expert per se so I'll leave it to others to approve. In any case, it looks like the tests are all at the C level, but it would be good to have integration tests similar to

qsharp-runtime/src/QirRuntime/test/QIR-dynamic/qir-test-random-lnx.ll

Line 90 in 0cf4a5e

define i64 @Microsoft__Quantum__Testing__QIR__QuantumRandomNumberGenerator__body() #0 {

that ensure that the new intrinsic works correctly when called from QIR files generated by the Q# compiler.

Moved the tests to .qs file.

[guenp]

Just some comments; thanks for implementing this!

[IrinaYatsenko]

Looks good to me on the QirRuntime's side of things. Signing off modulo Chris' comments on the semantics of arctan2.

[cgranade]

Approving to unblock @guenp, but we should as quickly as possible:

• Ensure exclusive language is not used in codebase
• Extend QSharpFoundation unit tests to cover what you noticed about ArcTan2, so that we can prevent regressions that would cause QIR and C# runtimes to disagree

[kuzminrobin]

Approving to unblock @guenp, but we should as quickly as possible:

• Ensure exclusive language is not used in codebase
• Extend QSharpFoundation unit tests to cover what you noticed about ArcTan2, so that we can prevent regressions that would cause QIR and C# runtimes to disagree

Filed #508.
Added to #506:

Extend QSharpFoundation unit tests to cover what has been noticed about ArcTan2, so that we can prevent regressions that would cause QIR and C# runtimes to disagree