Turns them into calls to memset / memcpy if 1) buffer(s) are not DWORD aligned,
2) size is not known to be greater or equal to some minimum value (currently 128).

llvm-svn: 26224

unswitch this loop on 2 before sweating to unswitch on 1/3.

void test4(int N, int i, int C, int*P, int*Q) {
  int j;
  for (j = 0; j < N; ++j) {
    switch (C) {                // general unswitching.
    default: P[i+j] = 0; break;
    case 1: Q[i+j] = 0; break;
    case 3: P[i+j] = Q[i+j]; break;
    case 2: break;              //  TRIVIAL UNSWITCH on C==2
    }
  }
}

llvm-svn: 26223

llvm-svn: 26222

llvm-svn: 26221

this for example:

  for (j = 0; j < N; ++j) {     // trivial unswitch
    if (C)
      P[i+j] = 0;
  }

turning it into the obvious code without bothering to duplicate an empty loop.

llvm-svn: 26220

llvm-svn: 26219

llvm-svn: 26218

llvm-svn: 26217

MachineDebugInfo tables.

llvm-svn: 26216

Y = seteq bool X, true
instead of just using X :)

llvm-svn: 26215

llvm-svn: 26213

llvm-svn: 26211

llvm-svn: 26209

llvm-svn: 26208

llvm-svn: 26207

llvm-svn: 26206

like the flex stuff, which actually works when people do cvs updates and
get conflicts in the updated checked in file.

llvm-svn: 26205

it, because this:

.bss
X:
.byte 0

results in the assembler warning: "initialization in bss segment".  Annoying.

llvm-svn: 26204

llvm-svn: 26203

The ABI specifies that there is a register save area at the bottom of the
stack, which means the actual used pointer needs to be an offset from
the subtracted value.

llvm-svn: 26202

GCC's syntax for auto-dependency generation stuff. This should
be changed to be disabling dependency stuff unless GCC/ICC is
found.

llvm-svn: 26201

llvm-svn: 26200

llvm-svn: 26199

go through, but we do want to know if we're using GCC/ICC since they
share certain funky command line options (for dependency generation
stuff)

llvm-svn: 26198

llvm-svn: 26197

llvm-svn: 26196

ops.

llvm-svn: 26195

llvm-svn: 26194

llvm-svn: 26193

llvm-svn: 26192

llvm-svn: 26191

llvm-svn: 26190

can provide work for the outer loop to canonicalize.

This fixes a case that breaks unswitching.

llvm-svn: 26189

llvm-svn: 26188

block in the appropriate loop nest.

Third time is the charm, right?

llvm-svn: 26187

llvm-svn: 26186

emit it (instead of .space), saving a bit of space in the .o file.

For example:
int foo[100];
int bar[100] = {};

when compiled with C++ or -fno-common results in shrinkage from 1160 to 360
bytes of space.  The X86 backend can also do this on darwin.

llvm-svn: 26185

llvm-svn: 26184

llvm-svn: 26183

XMM registers were not being handled.

llvm-svn: 26182