RMKirkpatrick

I was looking at [Qemu's documentation concerning ARM emulation](https://www.qemu.org/docs/master/system/target-arm.html#choosing-a-board-model) over the weekend. Would testing on a virtual armhf system merely entail configuring Qemu to emulate some arbitrary 32-bit ARM board...

So, what needs to be done to support checkpointing with `mxTryHard*()`? I guess it needs to be able to accept a `checkpoint` argument, and pass its value to its calls...

> Do you have the link handy? It's in the OP of this issue.

I'll volunteer to tackle this issue, unless you want to, @jpritikin .

> >Yeah, I recommend always using strict=TRUE with mxRestore > This suggests to me that it should be the default. I'm inclined to agree.

SRMR as used in MPlus: http://www.statmodel.com/download/SRMR2.pdf

FWIW, I'd like to point out a script I wrote, [/models/passing/GeneralizedEstimatingEquations.R](https://github.com/OpenMx/OpenMx/blob/master/inst/models/passing/GeneralizedEstimatingEquations.R), as an example of a test script that is also a readable example script.

@tbates I don't know about "moving" them, if that means taking them out of the test suite. Symlinking them, maybe?

@tbates There's quite a bit of information in the [Supplementary Appendices](https://static-content.springer.com/esm/art%3A10.1007%2Fs10519-015-9757-z/MediaObjects/10519_2015_9757_MOESM1_ESM.pdf) as well.

I just committed dab591c. It won't fix your issue with the getOpenMx.R script, but you should now be able to build OpenMx from a clone of the OpenMx git repository....