<table border="1" cellspacing="0" cellpadding="8">
    <tr>
        <th>Issue</th>
        <td>
            <a href=https://github.com/llvm/llvm-project/issues/59569>59569</a>
        </td>
    </tr>

    <tr>
        <th>Summary</th>
        <td>
            [x86] Intel scheduler models' instruction MicroOps counts need to consistently use the fused domain
        </td>
    </tr>

    <tr>
      <th>Labels</th>
      <td>
            backend:X86,
            tools:llvm-exegesis,
            tools:llvm-mca
      </td>
    </tr>

    <tr>
      <th>Assignees</th>
      <td>
      </td>
    </tr>

    <tr>
      <th>Reporter</th>
      <td>
          RKSimon
      </td>
    </tr>
</table>

<pre>
    The microops counts need to be compatible with ROB / MicroOpBufferSize - while the pipe resource counts need to match unfused uop numbers to correctly model pipe occupancy.

The PFM uop counters might need checking to ensure they match the retire_slots counts (the "any"/"all" counts are the unfused domain counts).
</pre>
<img width="1px" height="1px" alt="" src="http://email.email.llvm.org/o/eJxsUsFunDAQ_RpzsbKCYWGXA4ekK6SqilIlPfRWmWEW3Bgbeewm26-vDJu0ansBy8_z5s2bp5j1aIlaUd2J6pSpGCbn28dPT3p2NuvdcGm_TCRnjd65hSW6aANLSzTI4GRPEt28qKB7Q_JFh0k-PtxJAZ28TyUPy108n8k_6Z8kb-TLpA3JMJFc9ELSE7vokf5mnVXASUZ7jkyDjG6RNs49eU4gOu8Jg7nI2Q1kNiaHGBdl8bIT-Unkt9s3Kf_c3a8Ma4tEMetxClsrnAiftR0TLVmOftV2ufZPMj0F7ekbGxfeZxdwTJAAUPYiAAR06WyMAHh7ozaq9xkGNyttr6iAZpcNbTk0ZaMyaov6UJT5sS4gm1qkPapS9XikI5VNg8W-2de4P1dNUQx1nekWcoACigPkUJawU9XhQLkq6-E8lIe-F_ucZqXNzpgf8875MdPMkdqqqeomM6onw-u-AXqFz2QHUd5-PdbrKB8EQHDOsChvU_0NvdJIrPn_6IwqAdUp8-160ceRxT43mgP_FhB0MGvGXo-1qE7yow1kJONEQzTkt02ygIPUloOPGLSzbwn6J3ToLGsOZFMIIm9O_-lzFr1ppxCWpHNdTzfqMMV-h24W0CVd19_N4t13wiCgW11iAd1q1K8AAAD__016Cz0">