[Bug other/16798] New: PowerPC - Opportunity to use recording form instruction.

[Bug other/16798] New: PowerPC - Opportunity to use recording form instruction.

[gcc-bugzilla at gcc dot gnu dot org]

Description:
A non-optimal code sequence is illustraded.  The recording form of an instruction could be used allowing for the elimination of a compare.  Duplicate using gcc 3.5 and command line:

gcc -O3 -m64 -c test.c


Testcase:
long int i, j;
int foo ()
{

  j = (i>>52);

  if (j< 0) {
    i = 10;
  } else {
    i = 20;
  };

  return i;

}

Assembly:
.foo:
	ld 11,.LC0@toc(2)
	ld 9,.LC1@toc(2)
	ld 0,0(11)
	sradi 0,0,52 <-- Use the recording form of sradi
	cmpdi 7,0,0  <-- and eliminate this compare.
	std 0,0(9)
	blt- 7,.L7
	li 0,20
	std 0,0(11)
	lwa 3,4(11)
	blr
.L7:
	li 0,10
	std 0,0(11)
	lwa 3,4(11)
	blr



-- 
           Summary: PowerPC - Opportunity to use recording form instruction.
           Product: gcc
           Version: 3.5.0
            Status: UNCONFIRMED
          Severity: enhancement
          Priority: P1
         Component: other
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: steinmtz at us dot ibm dot com
                CC: gcc-bugs at gcc dot gnu dot org,steinmtz at us dot ibm
                    dot com
 GCC build triplet: powerpc64-linux
  GCC host triplet: powerpc64-linux
GCC target triplet: powerpc64-linux


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16798

[Bug target/16798] PowerPC - Opportunity to use recording form instruction.

[bangerth at dealii dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|other                       |target


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16798

[Bug target/16798] PowerPC - Opportunity to use recording form instruction.

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-08-01 00:54 -------
Note IIRC this is really only a size win rather than a performance win but I could be wrong.
Confirmed but I cannot understand why not:
(define_insn "*ashrdi3_internal1" 
  [(set (match_operand:DI 0 "gpc_reg_operand" "=r")
        (ashiftrt:DI (match_operand:DI 1 "gpc_reg_operand" "r")
                     (match_operand:SI 2 "reg_or_cint_operand" "ri")))]
  "TARGET_POWERPC64"
  "srad%I2 %0,%1,%H2") 
....
(define_insn "*ashrdi3_internal3"
  [(set (match_operand:CC 3 "cc_reg_operand" "=x,?y")
        (compare:CC (ashiftrt:DI (match_operand:DI 1 "gpc_reg_operand" "r,r")
                                 (match_operand:SI 2 "reg_or_cint_operand" "ri,ri"))
                    (const_int 0)))
   (set (match_operand:DI 0 "gpc_reg_operand" "=r,r")
        (ashiftrt:DI (match_dup 1) (match_dup 2)))]
  "TARGET_64BIT"
  "@
   srad%I2. %0,%1,%H2
   #"
  [(set_attr "type" "delayed_compare")
   (set_attr "length" "4,8")])

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|                            |1
   Last reconfirmed|0000-00-00 00:00:00         |2004-08-01 00:54:58
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16798

[Bug target/16798] PowerPC - Opportunity to use recording form instruction.

[pinskia at gcc dot gnu dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |missed-optimization
   Last reconfirmed|2004-08-01 00:54:58         |2004-10-31 04:58:03
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16798

[Bug target/16798] PowerPC - Opportunity to use recording form instruction.

[nathan at gcc dot gnu dot org]

------- Additional Comments From nathan at gcc dot gnu dot org  2004-11-09 12:25 -------
The problem here is that combine only combines chains of instructions that have
no other uses of the intermediate results.  In this case, the intermediate
result is stored into a global variable.

One solution would be to add a peephole, but that would only trigger when (a)
the shift and compare were adjacent, and (b) the compare used CR0.  You'd also
need peepholers for all the instructions that can implicitly set CR0.

Really combine, which is a generic peepholer, needs some kind of templatizing on
the target machine.  It currently functions by generating promising patterns and
 then seeing if they exist.  This is wasteful, in that it generates many
patterns  that don't exist on the particular target, and it is blind in that it
doesn't spot any special cases the target might have (like this case).  What's
needed is something like pre-reg-alloc peepholes.  That of course would be a
large amount of work.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16798