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: 19241

llvm-svn: 19240

a terminal/console.

llvm-svn: 19237

console or not.

llvm-svn: 19236

console or not.

llvm-svn: 19233

  * We can now fold cast instructions into select instructions that
    have at least one constant operand.
  * We now optimize expressions more aggressively based on bits that are
    known to be zero.  These optimizations occur a lot in code that uses
    bitfields even in simple ways.
  * We now turn more cast-cast sequences into AND instructions.  Before we
    would only do this if it if all types were unsigned.  Now only the
    middle type needs to be unsigned (guaranteeing a zero extend).
  * We transform sign extensions into zero extensions in several cases.

This corresponds to these test/Regression/Transforms/InstCombine testcases:
  2004-11-22-Missed-and-fold.ll
  and.ll: test28-29
  cast.ll: test21-24
  and-or-and.ll
  cast-cast-to-and.ll
  zeroext-and-reduce.ll

llvm-svn: 19220

While we're at it, improve codegen of select instructions.  For this
testcase:

int %test(bool %C, int %A, int %B) {
  %D = select bool %C, int %A, int %B
  ret int %D
}

We used to generate this code:

_test:
        cmpwi cr0, r3, 0
        bne .LBB_test_2 ;
.LBB_test_1:    ;
        b .LBB_test_3   ;
.LBB_test_2:    ;
        or r5, r4, r4
.LBB_test_3:    ;
        or r3, r5, r5
        blr

Now we emit:

_test:
        cmpwi cr0, r3, 0
        bne .LBB_test_2 ;
.LBB_test_1:    ;
        or r4, r5, r5
.LBB_test_2:    ;
        or r3, r4, r4
        blr

-Chris

llvm-svn: 19214

In particular, instead of compiling this:

bool %test(int %A, int %B) {
  %C = setlt int %A, %B
  ret bool %C
}

to this:

test:
        save %sp, -96, %sp
        subcc %i0, %i1, %g0
        bl .LBBtest_1   !
        nop
        ba .LBBtest_2   !
        nop
.LBBtest_1:     !
        or %g0, 1, %i0
        ba .LBBtest_3   !
        nop
.LBBtest_2:     !
        or %g0, 0, %i0
        ba .LBBtest_3   !
        nop
.LBBtest_3:     !
        restore %g0, %g0, %g0
        retl
        nop

We now compile it to this:

test:
        save %sp, -96, %sp
        subcc %i0, %i1, %g0
        or %g0, 1, %i0
        bl .LBBtest_2   !
        nop
.LBBtest_1:     !
        or %g0, %g0, %i0
.LBBtest_2:     !
        restore %g0, %g0, %g0
        retl
        nop

llvm-svn: 19213

SimplifyCFG is one of those passes that we use for final cleanup: it should
not rely on other passes to clean up its garbage.  This fixes the "why are
trivially dead setcc's in the output of gccas" problem.

llvm-svn: 19212

llvm-svn: 19210

llvm-svn: 19203

llvm-svn: 19202

llvm-svn: 19201

llvm-svn: 19199

llvm-svn: 19198

llvm-svn: 19190

llvm-svn: 19180

Fix testcase CodeGen/CBackend/2004-12-28-LogicalConstantExprs.ll

llvm-svn: 19176

llvm-svn: 19172

portability, the --track-space option was inadvertently ignored. This patch
fixes that so that sys::Process::GetMallocUsage() is only invoked if the
--track-spaces option is given. Apparently the mallinfo() call that
GetMallocUsage() uses is *very* slow, especially when processing very large
modules like projects/llvm-test/MultiSource/Applications/kimwitu++.

llvm-svn: 19163

* Ensure #includes are wrapped with appropriate HAVE_ guards
* Account for variations in time.h and sys/time.h inclusion.
* Add macros for getting wait(2) exit status correctly (per autoconf sugg.)
* Implement ThrowErrno in terms of strerror_r, strerror or none based on
  what is available on the platform.

llvm-svn: 19161

* Remove redundant #includes that are now in Unix.h

llvm-svn: 19160

* Move implementation of sys::PreventCoreFiles function to this file from
  the now defunct SysConfig abstraction.

llvm-svn: 19159

* Ensure #includes are wrapped with appropriate HAVE_ guards
* Ensure variations in "dirent" structure are accounted for.

llvm-svn: 19158

* Ensure #includes are wrapped with appropriate HAVE_ guards
* Consolidate implementation from operating system specific directory.

llvm-svn: 19157

* Ensure #includes are wrapped with appropriate HAVE_ guards

llvm-svn: 19156

* Consolidate implementation for Unix systems into Unix/TimeValue.cpp
* Avoid use of symbolic link to #include platform-specific implementation.

llvm-svn: 19155

* Consolidate implementation for Unix systems into Unix/Program.cpp
* Avoid use of symbolic link to #include platform-specific implementation.

llvm-svn: 19154

* Consolidate implementation for Unix systems into Unix/Signals.cpp
* Avoid use of symbolic link to #include platform-specific implementation.

llvm-svn: 19153

* Consolidate implementation for Unix systems into Unix/Memory.cpp
* Avoid use of symbolic link to #include platform-specific implementation.

llvm-svn: 19152

* Consolidate implementation for Unix systems into Unix/MappedFile.cpp
* Avoid use of symbolic link to #include platform-specific implementation.

llvm-svn: 19151

* Consolidate implementation for Unix systems into Unix/Process.cpp
* Avoid use of symbolic link to #include platform-specific implementation.

llvm-svn: 19150

SysConf abstraction was pointless because it had a single function in it that
pertained only to the current process. So merge it into the Process abstraction
and remove the files completely.

llvm-svn: 19149

Old versions of the C runtime somehow get loaded into the process.  Make
sure they aren't searched for symbols.

llvm-svn: 19141

llvm-svn: 19140

ltdl's LGPL license would infect all of LLVM.

llvm-svn: 19137

Merge implementations of isValid and GetTemporaryDirectory into this file. There
is not any point having the operating system specific files for such little
variation between the Unix family of systems.

llvm-svn: 19134

This implementation is no longer needed, its been merged to Unix/Path.cpp since
there is such little variation between the platforms.

llvm-svn: 19133

Use the LLVM_ON_UNIX and LLVM_ON_WIN32 #defines to include the implementation
instead of relying upon the "platform" link

llvm-svn: 19132