[WIP] [runtime] adding mx and my support for qarrays

[cabreraam]

Description

This is a work in progress, but I wanted to go ahead and try to address issue #2219.

Upon testing this, I found that the issue arises here:

/usr/local/cudaq/include/cudaq/qis/qubit_qis.h:760:23: note: candidate function not viable: no known conversion from 'cudaq::qarray<10UL>' to 'qubit &' (aka 'qudit<2> &') for 1st argument
inline measure_result mx(qubit &q) {
                      ^

So, I implemented support for mx and my. I'm posting this before adding any test cases just to see if I'm on the right track, here.

Feedback appreciated!

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

[github-actions]

CLA Assistant Lite bot All Contributors have signed the CLA.

[cabreraam]

I have read the Contributor License Agreement and I hereby accept the Terms.

[amccaskey]

Here I think you don't want the basis change operations, since you are calling the single qubit my which will apply the basis change already.

[cabreraam]

Will do; thanks for the tip!

[amccaskey]

Thanks for looking into this issue @cabreraam . Would you mind also adding a test to unittests/ and a FileCheck test to the test/AST-Quake folder? You can just use the failing snippet from Issue tracker.

[amccaskey]

Also while you're here, there are other mz overloads that we'll likely want to implement for mx, my. Lines 777-827

[cabreraam]

Thanks for looking into this issue @cabreraam . Would you mind also adding a test to unittests/ and a FileCheck test to the test/AST-Quake folder? You can just use the failing snippet from Issue tracker.

I will do this!

Also while you're here, there are other mz overloads that we'll likely want to implement for mx, my. Lines 777-827

Sounds good; I'll look into it.

[schweitzpgi]

/ok to test

Command Bot: Processing...

[cabreraam]

@amccaskey @schweitzpgi

Added a couple of lit tests. I tried to remove all of the superfluous stuff and just check for the most important bit of cudaq-quake's result. Let me know if you want to include all of the other setup MLIR stuff as well.

Additionally, #2219 wants to lower to OpenQASM 2.0. However, I'm not sure of a corresponding OQ2 instruction that measures with respect to y and z. (I'm using the OpenQASM 2.0 spec and the relevant cuda-quantum source code to try to deduce if I can lower this.)

Also while you're here, there are other mz overloads that we'll likely want to implement for mx, my. Lines 777-827

I will address the other mx and my cases here in a different PR for modularity.

With all of that being said, what more needs to be done to land this PR?

[khalatepradnya]

/ok to test

Command Bot: Processing...