llvm-svn: 28130

generated:

        movl 8(%esp), %eax
        movl %eax, %edx
        addl $4316, %edx
        cmpb $1, %cl
        ja LBB1_2       #cond_false
LBB1_1: #cond_true
        movl L_QuantizationTables720$non_lazy_ptr, %ecx
        movl %ecx, (%edx)
        movl L_QNOtoQuantTableShift720$non_lazy_ptr, %edx
        movl %edx, 4460(%eax)
        ret
...

Now we generate:

        movl 8(%esp), %eax
        cmpb $1, %cl
        ja LBB1_2       #cond_false
LBB1_1: #cond_true
        movl L_QuantizationTables720$non_lazy_ptr, %ecx
        movl %ecx, 4316(%eax)
        movl L_QNOtoQuantTableShift720$non_lazy_ptr, %ecx
        movl %ecx, 4460(%eax)
        ret

... which uses one fewer register.

llvm-svn: 28129

llvm-svn: 28124

  // fold (and (sext x), (sext y)) -> (sext (and x, y))
  // fold (or  (sext x), (sext y)) -> (sext (or  x, y))
  // fold (xor (sext x), (sext y)) -> (sext (xor x, y))
  // fold (and (aext x), (aext y)) -> (aext (and x, y))
  // fold (or  (aext x), (aext y)) -> (aext (or  x, y))
  // fold (xor (aext x), (aext y)) -> (aext (xor x, y))

llvm-svn: 28123

        mov EAX, DWORD PTR [ESP + 4]
        mov ECX, DWORD PTR [EAX]
        mov EDX, ECX
        add EDX, EDX
        or EDX, ECX
        and EDX, -2147483648
        and ECX, 2147483647
        or EDX, ECX
        mov DWORD PTR [EAX], EDX
        ret

instead of:

        sub ESP, 4
        mov DWORD PTR [ESP], ESI
        mov EAX, DWORD PTR [ESP + 8]
        mov ECX, DWORD PTR [EAX]
        mov EDX, ECX
        add EDX, EDX
        mov ESI, ECX
        and ESI, -2147483648
        and EDX, -2147483648
        or EDX, ESI
        and ECX, 2147483647
        or EDX, ECX
        mov DWORD PTR [EAX], EDX
        mov ESI, DWORD PTR [ESP]
        add ESP, 4
        ret

llvm-svn: 28122

llvm-svn: 28121

  // fold (and (trunc x), (trunc y)) -> (trunc (and x, y))
  // fold (or  (trunc x), (trunc y)) -> (trunc (or  x, y))
  // fold (xor (trunc x), (trunc y)) -> (trunc (xor x, y))

llvm-svn: 28120

llvm-svn: 28117

of cross-block live ranges, and allows the bb-at-a-time selector to always
coallesce these away, at isel time.

This reduces the load on the coallescer and register allocator.  For example
on a codec on X86, we went from:

   1643 asm-printer           - Number of machine instrs printed
    419 liveintervals         - Number of loads/stores folded into instructions
   1144 liveintervals         - Number of identity moves eliminated after coalescing
   1022 liveintervals         - Number of interval joins performed
    282 liveintervals         - Number of intervals after coalescing
   1304 liveintervals         - Number of original intervals
     86 regalloc              - Number of times we had to backtrack
1.90232 regalloc              - Ratio of intervals processed over total intervals
     40 spiller               - Number of values reused
    182 spiller               - Number of loads added
    121 spiller               - Number of stores added
    132 spiller               - Number of register spills
      6 twoaddressinstruction - Number of instructions commuted to coalesce
    360 twoaddressinstruction - Number of two-address instructions

to:

   1636 asm-printer           - Number of machine instrs printed
    403 liveintervals         - Number of loads/stores folded into instructions
   1155 liveintervals         - Number of identity moves eliminated after coalescing
   1033 liveintervals         - Number of interval joins performed
    279 liveintervals         - Number of intervals after coalescing
   1312 liveintervals         - Number of original intervals
     76 regalloc              - Number of times we had to backtrack
1.88998 regalloc              - Ratio of intervals processed over total intervals
      1 spiller               - Number of copies elided
     41 spiller               - Number of values reused
    191 spiller               - Number of loads added
    114 spiller               - Number of stores added
    128 spiller               - Number of register spills
      4 twoaddressinstruction - Number of instructions commuted to coalesce
    356 twoaddressinstruction - Number of two-address instructions

On this testcase, this change provides a modest reduction in spill code,
regalloc iterations, and total instructions emitted.  It increases the number
of register coallesces.

llvm-svn: 28115

llvm-svn: 28110

scheduler can go into a "vertical mode" (i.e. traversing up the two-address
chain, etc.) when the register pressure is low.
This does seem to reduce the number of spills in the cases I've looked at. But
with x86, it's no guarantee the performance of the code improves.
It can be turned on with -sched-vertically option.

llvm-svn: 28108

llvm-svn: 28107

llvm-svn: 28105

llvm-svn: 28104

llvm-svn: 28102

llvm-svn: 28099

thing that can be in it.  Remove a dead method.

llvm-svn: 28098

llvm-svn: 28094

llvm-svn: 28093

llvm-svn: 28092

llvm-svn: 28091

simplifies the MachineCodeEmitter interface just a little bit and makes
BasicBlocks work like constant pools and jump tables.

llvm-svn: 28082

not be 100% dense.  Increase the minimum threshold for the number of cases
in a switch statement from 4 to 6 in order to create a jump table.

llvm-svn: 28079

the heuristic to further reduce spills for several test cases. (Note, it may
not necessarily translate to runtime win!)

llvm-svn: 28076

Refactor TargetMachine, pushing handling of TargetData into the target-specific subclasses.  This has one caller-visible change: getTargetData() now returns a pointer instead of a reference.

This fixes PR 759.

llvm-svn: 28074

llvm-svn: 28069

1. Change several methods in the MachineCodeEmitter class to be pure virtual.
2. Suck emitConstantPool/initJumpTableInfo into startFunction, removing them
   from the MachineCodeEmitter interface, and reducing the amount of target-
   specific code.
3. Change the JITEmitter so that it allocates constantpools and jump tables
   *right* next to the functions that they belong to, instead of in a separate
   pool of memory.  This makes all memory for a function be contiguous, and
   means the JITEmitter only tracks one block of memory now.

llvm-svn: 28065

just have the JIT malloc them.

llvm-svn: 28062

code emission location into the base class, instead of being in the derived classes.

This change means that low-level methods like emitByte/emitWord now are no longer
virtual (yaay for speed), and we now have a framework to support growable code
segments.  This implements feature request #1 of PR469.

llvm-svn: 28059

llvm-svn: 28057

llvm-svn: 28055

useful for debugging.

llvm-svn: 28051

basic block labels, consolidating the code to do so in one place for each
target.

llvm-svn: 28050

llvm-svn: 28047

llvm-svn: 28046

llvm-svn: 28044

instructions in the virtregfolded map that were deleted.  Because they
were deleted, newly allocated instructions could end up at the same address,
magically finding themselves in the map.  The solution is to remove entries
from the map when we delete the instructions.

llvm-svn: 28041

instruction folded with spill code, make sure the remove the load from
the virt reg folded map.

llvm-svn: 28040

llvm-svn: 28039

llvm-svn: 28035