<table border="1" cellspacing="0" cellpadding="8">
    <tr>
        <th>Issue</th>
        <td>
            <a href=https://github.com/llvm/llvm-project/issues/133431>133431</a>
        </td>
    </tr>

    <tr>
        <th>Summary</th>
        <td>
            [AMDGPU] SIPeepholeSDWA: Add AMDGPU::V_CNDMASK_B32 support
        </td>
    </tr>

    <tr>
      <th>Labels</th>
      <td>
            backend:AMDGPU,
            missed-optimization
      </td>
    </tr>

    <tr>
      <th>Assignees</th>
      <td>
            frederik-h
      </td>
    </tr>

    <tr>
      <th>Reporter</th>
      <td>
          frederik-h
      </td>
    </tr>
</table>

<pre>
    The `isConvertibleToSDWA` function form `llvm/lib/Target/AMDGPU/SIPeepholeSDWA.cpp` contains this FIXME:
```c++
// FIXME: has SDWA but require handling of implicit VCC use
if (Opc == AMDGPU::V_CNDMASK_B32_e32)
    return false;
```
</pre>
<img width="1" height="1" alt="" src="http://email.email.llvm.org/o/eJx0kkFv1D4QxT_N5GJ15Yyd7e4hh-zmn78qVKjUUrhVjjNJTJ042M5K8OmRl0DhgOSTNXrz5v2eCsEMM1EJxQkQe08defN6MwIiFHWm1jg6X779Z63rvpVPIzHYcxPObr6Qj6a19OQe608V7Dnr11lH42bWOz-lOWsvE2BjTQvYPCk_UARsqvv6_4ePgM3j3QPRMjpLSWGnlyWpaDdHZebA4mgCa-4-3_8HogKeVvx8GvCUHq8AG8Dm9xAbVWBJi7VrZJ6-rsYTG9XcWTMPzPXMTIs12kT2fD6zNRDwyvQM8PBh0QxEDaJmmz9RgaieX87v6_vq8d3LSeALCQQ8Aq8YY8xTXP3MemUDgTj9aTDrStEdxVFlVOa3UuR73MtjNpaFlrrXpAqRIy8E59R28oCy16qn_HafmRI5FlzgIc8LLo-7w20h9hKPPXLMpZQgOU3K2F3Kduf8kJkQVipzIaTIM6tasmGj2ir9SnMHovoVOQKeAXEyIVB345ZoJvNdJWYbd18m3Zt2HQJIbk2I4W1TNNFeG7PJFTX7G2FCUHXdvxNkYV0W52O2eluOMS4hDV0hDiaOa7vT7tqYrTjJy-LdF9KpONdLA2CzHXsp8UcAAAD__45v1s4">