<table border="1" cellspacing="0" cellpadding="8">
    <tr>
        <th>Issue</th>
        <td>
            <a href=https://github.com/llvm/llvm-project/issues/62952>62952</a>
        </td>
    </tr>

    <tr>
        <th>Summary</th>
        <td>
            [X86] Remove X86Subtarget::isAtom and replace with tuning flags
        </td>
    </tr>

    <tr>
      <th>Labels</th>
      <td>
            backend:X86
      </td>
    </tr>

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

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

<pre>
    We still have special case handling for Atom (Bonnell) CPUs, behind a ST->isAtom() check - some of these might be entirely redundant (disabling vector interleaving might be better now with decent cost tables?), while others should be replaced with tuning flags:

Disable promotion of i16 comparisons
Use ILP scheduling
Optimal AntiDepBreakMode (#62875)

This would rid us of the last remaining CPU specific feature flag (ProcIntelAtom)
</pre>
<img width="1px" height="1px" alt="" src="http://email.email.llvm.org/o/eJxkUstu6zYQ_RpqM7AhU9FroYUT18BFW9S4uUGzpcixyIYiBc7IRv6-oJw-gLuRAGLmPOYcReSmgDiI-lnUp0KtbGMavv_66uYYijGaz-FPBGLnPVh1Q6AFtVMetCIEq4LxLkxwjQmOHGcQsnuOIaD3QvbwcnkjIV9gROuCAQWvP3ai-sVRnhWyyzPaov6AHVCcEeIV2CIhzG6yDCMCBnYJ_SckNGswKnDmMI7UuDHfUHNM4AJj8qhu-e3f5RGZMUGId7g7tmBQY2DQkRhYjR5JVGch-6zxbp1HiGwxEZCNqzcZIuHilUbzAOA1bHa9mkhUR1GeRPn1PW2SEJYU58guhmzGHRrQcV5UchQDPSbfCOHbbxcgbdGs2cXj_Y-F3aw8HAO7Ey7PCdXH79EgbJeqGtm1dRb7P9If1hHcN63JGVjp64LgFTEknJXbBL9c3h7JXZ2GKypeE24uMvYlRf0tMPpHKn1hhsr0Va8KHA5NV5d9e-iawg5t0yM2ZYctmrIbu66VvcH2qa2N7vu-LdwgS1mVtWwOZVWVzd4cqnJUpm10r6-qOoinMmvye-9v8z6mqXBEKw6N7GtZeDWip62LUo5Kf2Awojq-d42QMtczDXlvN64TiafSO2L6D4kd-63Ieb4-wXec4w3hvWte15FVmpBzZNXxUT9QwfyT7s_hFmvyg2VetpjlWcjz5Niu417nG50z69dvt6T4F2oW8ryZISHPm5-_AwAA___pzxrT">