<table border="1" cellspacing="0" cellpadding="8">
    <tr>
        <th>Issue</th>
        <td>
            <a href=https://github.com/llvm/llvm-project/issues/120878>120878</a>
        </td>
    </tr>

    <tr>
        <th>Summary</th>
        <td>
            [AArch64][GISel] Optimize bitreverse intrinsic for vector types
        </td>
    </tr>

    <tr>
      <th>Labels</th>
      <td>
            new issue
      </td>
    </tr>

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

    <tr>
      <th>Reporter</th>
      <td>
          madhur13490
      </td>
    </tr>
</table>

<pre>
    2 x s32, 2 x i64, 4 x s16, 4 x s32, 8 x s16 types are going through suboptimal lowering. Looking at SDAG, `revXX` followed by `rbit` is more optimal.
godbolt: https://godbolt.org/z/nP8jssKh8

</pre>
<img width="1" height="1" alt="" src="http://email.email.llvm.org/o/eJxcUcFu3CAU_Bp8eYqFn8FrHzg4XTmqWqmVcskV7GebFJsV4E2Tr6_YrFqpJ0bzhnkwo2O0y06kmHxk8lzoI60-qE1P6xGqWnS8MH56Vwi_IdbI8AtkaBuRochs1fyFn4L2k4X0fqEIOhAs3u4LpDX4Y1khHsZfkt20A-ffKNh9KeG797-ySCd4PvdP2Yc1PND15YU1HGbvsnYC837jjU2ZthE2HwjufiXj_eIn411idQ9rSpfI6p7hwHC4D0ofFobDB8Nh_9m-xvhtbRnvi0nVU1d3uiBVnWrRSsmFLFY165MkwlaLhtddh2LqppNp-ChEM1dIhVXIUVSIyNuKC1kaPZtaUm3QzFI0mglOm7audO665fWFjfEgVSFvT23htCEXbwUg7vQGtylDzH0ElS89mGOJTHBnY4r_bJJN7tZc34dxbQSTZyYfn74-k2PyDD9yKPaDwNgU6EohEtg9BbtHO8LsA1xpTD58FlUcwan_ErNpPUw5-o3hkLfej4dL8K80JobD7a2R4XD_zFXhnwAAAP__bnG7Ew">