8088 8D instruction (LEA) test has LEA r, r forms.

[GonBDLH]

According to the readme, the form of the LEA instruction "LEA r, r" should not appear in the tests because it is not defined. However, there are tests of this form, such as "LEA dx, dx".

e.g.:

{
    "name": "lea dx, dx",
    "bytes": [38, 141, 210],
    "initial": {
        "regs": {
            "ax": 53822,
            "bx": 42257,
            "cx": 16234,
            "dx": 58498,
            "cs": 18747,
            "ss": 29741,
            "ds": 58434,
            "es": 25500,
            "sp": 46446,
            "bp": 58072,
            "si": 22081,
            "di": 10332,
            "ip": 40845,
            "flags": 64642
        },
        "ram": [
            [340797, 38],
            [340798, 141],
            [340799, 210],
            [340800, 144]
        ],
        "queue": []
    },
    "final": {
        "regs": {
            "ax": 53822,
            "bx": 42257,
            "cx": 16234,
            "dx": 19841,
            "cs": 18747,
            "ss": 29741,
            "ds": 58434,
            "es": 25500,
            "sp": 46446,
            "bp": 58072,
            "si": 22081,
            "di": 10332,
            "ip": 40848,
            "flags": 64642
        },
        "ram": [
            [340797, 38],
            [340798, 141],
            [340799, 210],
            [340800, 144]
        ],
        "queue": []
    },
    "cycles": [
        ["-", 340798, "CS", "R--", "---", 0, "CODE", "T2", "F", 38],
        ["-", 340798, "CS", "R--", "---", 141, "PASV", "T3", "-", 0],
        ["-", 340798, "CS", "---", "---", 0, "PASV", "T4", "-", 0],
        ["A", 340799, "--", "---", "---", 0, "CODE", "T1", "-", 0],
        ["-", 340799, "CS", "R--", "---", 0, "CODE", "T2", "F", 141],
        ["-", 340799, "CS", "R--", "---", 210, "PASV", "T3", "-", 0],
        ["-", 340799, "CS", "---", "---", 0, "PASV", "T4", "-", 0],
        ["A", 340800, "--", "---", "---", 0, "CODE", "T1", "-", 0],
        ["-", 340800, "CS", "R--", "---", 0, "CODE", "T2", "S", 210],
        ["-", 340800, "CS", "R--", "---", 144, "PASV", "T3", "-", 0],
        ["-", 340800, "CS", "---", "---", 0, "PASV", "T4", "-", 0],
        ["A", 340801, "--", "---", "---", 0, "CODE", "T1", "-", 0]
    ]
},

[dbalsom]

Apologies, my mistake. I did have a correct 8D.json but I committed the wrong one. I will submit a PR to fix.

I've had some suggestions to leave in the reg,reg forms, but the way they work is they use the value of the last calculated effective address. The register set up program on the Arduino changes this in a way that your emulator can't really replicate; but it might be possible to set it up specifically, and then include the value of EA in the test initial state. Thoughts?

[GonBDLH]

No worries. And btw, thank you very much for this tests, i had some bugs in my emulator that i wasn't able to fix before finding them.

I think including the value of the ea in the initial state is a good solution.

[dbalsom]

https://github.com/TomHarte/ProcessorTests/pull/57

[dbalsom]

this issue can be closed as the pull request has been merged