<html>
    <head>
      <base href="https://llvm.org/bugs/" />
    </head>
    <body><table border="1" cellspacing="0" cellpadding="8">
        <tr>
          <th>Bug ID</th>
          <td><a class="bz_bug_link 
          bz_status_NEW "
   title="NEW --- - [ppc] bad scheduling for special register access instructions"
   href="https://llvm.org/bugs/show_bug.cgi?id=25685">25685</a>
          </td>
        </tr>

        <tr>
          <th>Summary</th>
          <td>[ppc] bad scheduling for special register access instructions
          </td>
        </tr>

        <tr>
          <th>Product</th>
          <td>libraries
          </td>
        </tr>

        <tr>
          <th>Version</th>
          <td>trunk
          </td>
        </tr>

        <tr>
          <th>Hardware</th>
          <td>PC
          </td>
        </tr>

        <tr>
          <th>OS</th>
          <td>Linux
          </td>
        </tr>

        <tr>
          <th>Status</th>
          <td>NEW
          </td>
        </tr>

        <tr>
          <th>Severity</th>
          <td>normal
          </td>
        </tr>

        <tr>
          <th>Priority</th>
          <td>P
          </td>
        </tr>

        <tr>
          <th>Component</th>
          <td>Backend: PowerPC
          </td>
        </tr>

        <tr>
          <th>Assignee</th>
          <td>unassignedbugs@nondot.org
          </td>
        </tr>

        <tr>
          <th>Reporter</th>
          <td>carrot@google.com
          </td>
        </tr>

        <tr>
          <th>CC</th>
          <td>llvm-bugs@lists.llvm.org
          </td>
        </tr>

        <tr>
          <th>Classification</th>
          <td>Unclassified
          </td>
        </tr></table>
      <p>
        <div>
        <pre>This is a very common instruction sequence. 

One example is Perl_sv_setsv_flags from perlbench. On power8, llvm generated
Perl_sv_setsv_flags consumes 4.26% of run time, gcc generated function consumes
3.5% of run time.

perf annotate shows following instructions and cycles of llvm generated code:

      │      00000000100c67d0 <Perl_sv_setsv_flags>:                            
       │        addis  r2,r12,26                                                
       │        addi   r2,r2,31576                                              
  4.95 │        mfcr   r12                                                      
  4.18 │        mflr   r0                                                       
  0.07 │        std    r31,-8(r1)                                               
       │        std    r0,16(r1)                                                
  0.02 │        stw    r12,8(r1) 
                ...                         

The two instructions mfcr/mflr consume many time, these are slow instructions,
but the results are used immediately, causes stalling.

For comparison following is gcc generated code:

       │      00000000100d02f0 <Perl_sv_setsv_flags>:                           
       │        addis  r2,r12,24                                                
       │        addi   r2,r2,21112                                              
  1.63 │        mflr   r0                                                       
       │        cmpld  cr7,r4,r3                                                
  1.57 │        std    r30,-16(r1)                                              
  0.01 │        std    r31,-8(r1)                                               
  1.66 │        mfocrf r12,8                                                    
       │        std    r26,-48(r1)                                              
       │        std    r27,-40(r1)                                              
  0.02 │        mr     r31,r3                                                   
       │        mr     r30,r4                                                   
       │        std    r28,-32(r1)                                              
  2.64 │        std    r29,-24(r1)                                              
       │        stw    r12,8(r1)                                                
       │        std    r0,16(r1)   
                ...                        

It has much better scheduling of mflr/mfocrf and corresponding usage
instructions.</pre>
        </div>
      </p>
      <hr>
      <span>You are receiving this mail because:</span>
      
      <ul>
          <li>You are on the CC list for the bug.</li>
      </ul>
    </body>
</html>