<table border="1" cellspacing="0" cellpadding="8">
    <tr>
        <th>Issue</th>
        <td>
            <a href=https://github.com/llvm/llvm-project/issues/62552>62552</a>
        </td>
    </tr>

    <tr>
        <th>Summary</th>
        <td>
            [X86] Allow shuffle combining to use the hasNoDomainDelayShuffle/hasNoDomainDelayBlend controls
        </td>
    </tr>

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

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

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

<pre>
    As many recent targets (or old ones like atom....) don't exhibit any stall when using cross-domain shuffles we should allow shuffle combining to combine to any vector domain when the subtarget supports it.

Much of this work will be adding suitable test coverage without excessive bloat like for variable per/cross-lane shuffles.
</pre>
<img width="1px" height="1px" alt="" src="http://email.email.llvm.org/o/eJx0Ur2O2zwQfBqqWZwgU6bPLlT4YKj58KXINWkpciUxR3EN7tLOvX0g2UmAAGkkcn9mdzhjmcOUEDtl3pTWX_97DwslpbUyl8oWmSl3z2A1kP_szgyLTZ-Q0WESEJsnFAalj5SBogdKyBDDB4IVWuq6rpU-gV9BXwXwxxyGILBCsNgY4T5jgsIhTeAyMb94WmxIwHMZx4gMdwSeqUQPNka6_0qAo2UIae0Tel5wPa7QN3RCGZ5Q2wiZEbgMj4WBy_VKWRiC1Kq5qOb8-P5f3Aw0gsyB4U75A-4hRhgQrPfrLC5B7BARBFnA0Q2znRDuQWYqKz-HzOGGMESy8niIkTLcbA5b3xWz0v2DarQJfxOtK9-1_tSebIXd7nBs9e6oX001dziOrh29Px52BrV13px8i0avwcYeXBU63ei2Mc1-d2hfjantOBwMmtEaZ4ed3qt9g4sNsY7xttSUpyowF-wO2hhdRTtg5KcFBus-MHnVnr8dD08j5G7texnKxGrfxMDCf5AkSNz8s9abC5z_rVJh3HSYLX-hyybOBaP9fH8UK93_nXmLmDw4SpIpclVy7GaRK6v2rHSvdD8FmctQO1qU7tednr-Xa6bv6ETpfqPKSvcb258BAAD__5npAfk">