Reverting the quote patch.  For some reason, this breaks the building of
llvm/runtime (the shell doesn't like it for some reason).  I might play
with it to see if I can get the quotes done in such a way that the shell
like it, but no promises.

llvm-svn: 19275

llvm-svn: 19274

llvm-svn: 19273

Make it HTML 4.01 Strict compliant
Provide better explanations of the dependency information.

llvm-svn: 19272

llvm-svn: 19271

llvm-svn: 19270

llvm-svn: 19269

llvm-svn: 19268

llvm-svn: 19267

and creates issues on other platforms like Cygwin and MingW.

llvm-svn: 19266

it breaks "gmake check".

llvm-svn: 19265

llvm-svn: 19264

-  unsigned TrueValue = getReg(TrueVal, BB, BB->begin());
+  unsigned TrueValue = getReg(TrueVal);

Fixes the PPC regressions from last night.

The other hunk is just a clarity improvement.

llvm-svn: 19263

llvm-svn: 19262

llvm-svn: 19261

llvm-svn: 19260

llvm-svn: 19259

llvm-svn: 19258

correctly.

llvm-svn: 19257

exceptions to abort() in cases where it should not.

Many thanks to Duraid Madina for doing the heavy lifting on the analysis
of this problem.

llvm-svn: 19256

Cygwin).

llvm-svn: 19255

llvm-svn: 19254

llvm-svn: 19253

to get Visual Studio to link in X86.lib to the executables that need it.  There
is another way of doing it.

llvm-svn: 19252

addition to being three address, LEA's don't set the flags.

This fixes 186.crafty.

llvm-svn: 19251

llvm-svn: 19250

llvm-svn: 19249

llvm-svn: 19248

branches.

llvm-svn: 19247

1. Add new instructions for checking parity flags: JP, JNP, SETP, SETNP.
2. Set the isCommutable and isPromotableTo3Address bits on several
   instructions.

llvm-svn: 19246

1. If we are two-addressing a commutable instruction and the LHS is not the
   last use of the variable, see if the instruction is the last use of the
   RHS.  If so, commute the instruction, allowing us to avoid a
   register-register copy in many cases for common instructions like ADD, OR,
   AND, etc on X86.
2. If #1 doesn't hold, and if this is an instruction that also existing in
   3-address form, promote the instruction to a 3-address instruction to
   avoid the register-register copy.  We can do this for several common
   instructions in X86, including ADDrr, INC, DEC, etc.

This patch implements test/Regression/CodeGen/X86/commute-two-addr.ll,
overlap-add.ll, and overlap-shift.ll when I check in the X86 support for it.

llvm-svn: 19245

llvm-svn: 19244

llvm-svn: 19243

they are dense.  Add a virtual method that targets can choose to implement.

llvm-svn: 19242

llvm-svn: 19241

llvm-svn: 19240

This prevents bytecode splats with usage like: cat file.ll | llvm-as -

llvm-svn: 19239

into lib/Support so it can be used with other tools.

llvm-svn: 19238

a terminal/console.

llvm-svn: 19237

console or not.

llvm-svn: 19236