<table border="1" cellspacing="0" cellpadding="8">
    <tr>
        <th>Issue</th>
        <td>
            <a href=https://github.com/llvm/llvm-project/issues/55932>55932</a>
        </td>
    </tr>

    <tr>
        <th>Summary</th>
        <td>
            [WebAssembly][SIMD] Optimize codegen for relaxed SIMD instructions
        </td>
    </tr>

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

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

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

<pre>
    Currently the relaxed SIMD instructions are generated only from llvm intrinsics. They should additionally be selected based on normal code patterns so that e.g. autovectorization optimizations can take advantage of them.
</pre>
<img width="1px" height="1px" alt="" src="http://email.email.llvm.org/o/eJx1Uctu3DAM_Br7QsTwyqt1fPAhzSJADkUPLdAzLdG2Gj0Wkryo-_Wl3KDNpYAAkho-Zsgp6H183mIkn-0OeSWIZPEnafj6-vkKxqccN5VN8AkwEizkKWJmPHgumGNwYO3dcWaOnG1UauDbSjukNWxWA2ptSjlaTp8IEllSpX7CdHQBH6JDCypoghvmTJFnpcBkMAM1SwO45XDnqhDNLyzNINyyce9BAoUeMr4RD7ujz7gQhLmIcU2tx04P3YB1NtnSWMlP32l6SoncZPdKXvmjKGUPvvxpSgcVFgpziP9fR71FO64531LVPVXihd9i8rpNjQqOg7KVd_Nwi-EH8-fQpLRRYkfKoRP1Ok6dRDw_yk7SoAZSRAOeqJVDr7TqLlhbnMimQrwSgvyKXpHjc3FUiWdgM6F6I6-Zxkdpf2HHM0k_fFxZAeW1NqNohWgvbS-E7E99g_2lfWxPfTe353m4zNW5JYfGNkVFE-JSx_EQNG1LYtCalNM_EFMyi6djy6U_320NcczW3Mnu9aF9PIT_Bikb2Bo">