LI77629: INEFFICIENT CODE FOR 64-BIT VEC_CMPGT

XL C/C++ for Linux Fix Pack 8 (October 2015 Update) for 12.1
XL C/C++ for Linux Fix Pack 4 (October 2013 Update) for 12.1
XL C/C++ for Linux Fix Pack 5 (December 2013 Update) for 12.1
XL C/C++ for Linux Fix Pack 6 (February 2014 Update) for 12.1
XL C/C++ for Linux Fix Pack 7 (May 2014 Update) for 12.1

APAR status

• Closed as program error.

Error description

• Sub-optimal code is generated for vec_cmpgt built-in for long
  long input.
  

Local fix

• N/A
  

Problem summary

• PROBLEM DESCRIPTION:
  For the vec_cmpgt vector built-in function, when used with
  signed long long values, if the first operand is zero, it is
  not necessary to check both first and second words of each
  element.
  
  USERS AFFECTED:
  Users using the vector built-in functions vec_cmpgt/vec_cmplt
  when compared with a zero vector.
  

Problem conclusion

• When the first operand of vcmpgtsd is vector zero, the compiler
  will only compare the first word of each element.
  

Temporary fix

Comments

APAR Information

• APAR is sysrouted FROM one or more of the following:

  IV40035

• APAR is sysrouted TO one or more of the following:

Fix information

• Fixed component name

  XL C/C++ FOR LI

• Fixed component ID

  5725C7300

Applicable component levels

• RC10 PSN IV40035

     UP06/09/13