<table border="1" cellspacing="0" cellpadding="8">
    <tr>
        <th>Issue</th>
        <td>
            <a href=https://github.com/llvm/llvm-project/issues/140135>140135</a>
        </td>
    </tr>

    <tr>
        <th>Summary</th>
        <td>
            svptest_any(svptrue_b64(), x)
        </td>
    </tr>

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

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

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

<pre>
    svptest_any(svptrue_b64(), x) is equivalent to svptest_any(x, x) but the codegen is different:
https://godbolt.org/z/1EExvceje


The stack overflow discussion here:
 https://stackoverflow.com/questions/79623202/arm-sve-passing-same-mask-twice-svptest-anym-m-or-svptest-anysvptrue-m?noredirect=1#comment140429481_79623202

P.S. Peter Cordes mentioned similar issues for ptest on x86 - though I didn't look into that.
</pre>
<img width="1" height="1" alt="" src="http://email.email.llvm.org/o/eJyMUsFupDoQ_BpzaTEybRjgwCEvCdK7RXrvsqfI4B5wBuyJ25CZ_foVM7PZzZ5WQsK2uqq7q0oz28ERNaL4RxRPiV7i6EPzRM7yNx08j7Ty0V6SzptLw-spEsdX7S4Cq-0WFnrt9rnASmAt8BHOAmuwDPS-2FVP5CJED1-B58_CbokQR4LeGxrIbUBjDwcK5KJQD0I-jDGeeDtiK7AdvOn8FHc-DALb7wLb7Pn5vPb0RkI-3L7_RwKOuj-CXykcJv8BxnK_MFvvYKRAN2b4Sn2F_ETsej8LbN8X4mi9Y4FtWe9RoUSBrQ5zyiulp009N6SsZ0pnzcc0ftie0vu2qXaXOZ1TH35_uauWzkK1zgcyNlAfhXrKBKrezzO5mOUyxzqvstfPrtfVXnb_7eCFIgV49MEQw1ZtvSMDbGc76QCWeSGGgw9w7QnewbnaQwpx9Mswwr9grHECywiT90ewLnqIo467xDTK1KrWCTVZmZcqzyqFydjoLOvqikpz2Kt9LfdY6E5VJJUp8VAUJrENSixkkRUoc1Xgrje1VFR2melKI6kUuaRZ22k3Teu82Zdc52yyXGaqSCbd0cTXECI6-rhtIRC3TIZmA6XdMrDI5WQ58i-aaONEfxvMZAlT80eibByX7u73xnr_pafg3zZjsL0pumXtNuza4I8AAAD___KHDPA">