<table border="1" cellspacing="0" cellpadding="8">
    <tr>
        <th>Issue</th>
        <td>
            <a href=https://github.com/llvm/llvm-project/issues/87393>87393</a>
        </td>
    </tr>

    <tr>
        <th>Summary</th>
        <td>
            [HLSL] Diagnostics for semantic annotations
        </td>
    </tr>

    <tr>
      <th>Labels</th>
      <td>
            HLSL
      </td>
    </tr>

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

    <tr>
      <th>Reporter</th>
      <td>
          llvm-beanz
      </td>
    </tr>
</table>

<pre>
    ### Description

In HLSL there are certain combinations of semantics that are valid for inputs and outputs of shaders as defined by the shader stage. @bob80905 began writing a proposal for how we can perform this analysis and verification during semantic analysis: https://github.com/microsoft/hlsl-specs/pull/156

### Acceptance Criteria

Implement robust diagnostic for combinations of semantics on inputs and outputs as defined in the specification above. 
</pre>
<img width="1px" height="1px" alt="" src="http://email.email.llvm.org/o/eJx8ks2OnDoQhZ_GbErTMgYaWLDoe1utRJpdnqBsCqjI2Mg2PZo8fWR6_hZJJCQsuVx16jsHY-TZEQ2i-U801wL3tPgwWHtfnzSh-1VoP74OQlWPD64UTeAtsXdCXoS8fHfw7fnHM6SFAgEGAkMhITswftXsMJdG8BNEWtElNhHSgukovaPlESYfgN22pwjoRvB7Os75yYIjhQgYYaSJHY2gX_OotxuICWc6gail9rqTvWxA04wOXgIndjMgbMFvPqI9xiz-BV4IDDrYKEw-rJAWznPRvkZ-CLhT4InNoRzGPeQ-7-I_KkV1gSWlLR-Eugl1mzktuz4Zvwp1W9kEH_2UhLotNtqnuJGJQt223VqhbmVzfvD7RHsxhraEzhD8HzhRYHxDvG6WVnIJgtd7TDAyzs7HLCcv9XfQ3v0J7Bea7B40NzKfK6P2dzpBMQ7V2Fc9FjSUbalkV9V1WyxDfVYtnivZlf25nTSNJXW66Rs1qVKVUhY8KKlqWUtVdnUt21N_1tQ07Vjqsq1UN4pa0opsTzloJx_mgmPcaejaqq8Ki5psPCKpVA6XUCqH8z2X-xxFLS3HFD8bJE72iPHxoLnC9QNSPCh9cdD59KBV7MEO_3Axd3_7PW3B_yST_Ty0Zi8Pub8DAAD__5UAGQ0">