hold your nose!)

llvm-svn: 17869

already been emitted, we don't have to remember it and deal with it later,
just emit it directly.

llvm-svn: 17868

* Get rid of "emitMaybePCRelativeValue", either we want to emit a PC relative
  value or not: drop the maybe BS.  As it turns out, the only places where
  the bool was a variable coming in, the bool was a dynamic constant.

llvm-svn: 17867

set up.

llvm-svn: 17862

llvm-svn: 17861

llvm-svn: 17750

llvm-svn: 17714

shld is a very high latency operation. Instead of emitting it for shifts of
two or three, open code the equivalent operation which is faster on athlon
and P4 (by a substantial margin).

For example, instead of compiling this:

long long X2(long long Y) { return Y << 2; }

to:

X3_2:
        movl 4(%esp), %eax
        movl 8(%esp), %edx
        shldl $2, %eax, %edx
        shll $2, %eax
        ret

Compile it to:

X2:
        movl 4(%esp), %eax
        movl 8(%esp), %ecx
        movl %eax, %edx
        shrl $30, %edx
        leal (%edx,%ecx,4), %edx
        shll $2, %eax
        ret

Likewise, for << 3, compile to:

X3:
        movl 4(%esp), %eax
        movl 8(%esp), %ecx
        movl %eax, %edx
        shrl $29, %edx
        leal (%edx,%ecx,8), %edx
        shll $3, %eax
        ret

This matches icc, except that icc open codes the shifts as adds on the P4.

llvm-svn: 17707

llvm-svn: 17706

long long X3_2(long long Y) { return Y+Y; }
int X(int Y) { return Y+Y; }

into:

X3_2:
        movl 4(%esp), %eax
        movl 8(%esp), %edx
        addl %eax, %eax
        adcl %edx, %edx
        ret
X:
        movl 4(%esp), %eax
        addl %eax, %eax
        ret

instead of:

X3_2:
        movl 4(%esp), %eax
        movl 8(%esp), %edx
        shldl $1, %eax, %edx
        shll $1, %eax
        ret

X:
        movl 4(%esp), %eax
        shll $1, %eax
        ret

llvm-svn: 17705

It's stosl (l for long == 32 bit).

llvm-svn: 17658

llvm-svn: 17488

llvm-svn: 17484

llvm-svn: 17431

llvm-svn: 17406

llvm-svn: 17286

llvm-svn: 17167

llvm-svn: 17155

llvm-svn: 17136

llvm-svn: 17126

last night. :)  bork!

llvm-svn: 17093

double %test(uint %X) {
        %tmp.1 = cast uint %X to double         ; <double> [#uses=1]
        ret double %tmp.1
}

into:

test:
        sub %ESP, 8
        mov %EAX, DWORD PTR [%ESP + 12]
        mov %ECX, 0
        mov DWORD PTR [%ESP], %EAX
        mov DWORD PTR [%ESP + 4], %ECX
        fild QWORD PTR [%ESP]
        add %ESP, 8
        ret

... which basically zero extends to 8 bytes, then does an fild for an
8-byte signed int.

Now we generate this:


test:
        sub %ESP, 4
        mov %EAX, DWORD PTR [%ESP + 8]
        mov DWORD PTR [%ESP], %EAX
        fild DWORD PTR [%ESP]
        shr %EAX, 31
        fadd DWORD PTR [.CPItest_0 + 4*%EAX]
        add %ESP, 4
        ret

        .section .rodata
        .align  4
.CPItest_0:
        .quad   5728578726015270912

This does a 32-bit signed integer load, then adds in an offset if the sign
bit of the integer was set.

It turns out that this is substantially faster than the preceeding sequence.
Consider this testcase:

unsigned a[2]={1,2};
volatile double G;

void main() {
    int i;
    for (i=0; i<100000000; ++i )
        G += a[i&1];
}

On zion (a P4 Xeon, 3Ghz), this patch speeds up the testcase from 2.140s
to 0.94s.

On apoc, an athlon MP 2100+, this patch speeds up the testcase from 1.72s
to 1.34s.

Note that the program takes 2.5s/1.97s on zion/apoc with GCC 3.3 -O3
-fomit-frame-pointer.

llvm-svn: 17083

us to use index registers for CPI's

llvm-svn: 17082

index reg and scale

llvm-svn: 17081

  %X = and Y, constantint
  %Z = setcc %X, 0

instead of emitting:

        and %EAX, 3
        test %EAX, %EAX
        je .LBBfoo2_2   # UnifiedReturnBlock

We now emit:

        test %EAX, 3
        je .LBBfoo2_2   # UnifiedReturnBlock

This triggers 581 times on 176.gcc for example.

llvm-svn: 17080

now compile:

'foo() {}' into "ret" instead of "mov EAX, 0; ret"

llvm-svn: 17049

case:

int C[100];
int foo() {
  return C[4];
}

We now codegen:

foo:
        mov %EAX, DWORD PTR [C + 16]
        ret

instead of:

foo:
        mov %EAX, OFFSET C
        mov %EAX, DWORD PTR [%EAX + 16]
        ret

Other impressive features may be coming later.

This patch is contributed by Jeff Cohen!

llvm-svn: 17011

contributed by Jeff Cohen!

llvm-svn: 17010

constant displacements from global variables.  Patch by Jeff Cohen!

llvm-svn: 17009

by Jeff Cohen!

llvm-svn: 17008

llvm-svn: 16950

llvm-svn: 16893

the -sse* options (to avoid misleading people).

Also, the stack alignment of the target doesn't depend on whether SSE is
eventually implemented, so remove a comment.

llvm-svn: 16860

which prevented setcc's from being folded into branches.  It appears that
conditional branchinst's CC operand is actually operand(2), not operand(0)
as we might expect. :(

llvm-svn: 16859

instcombine xform, which is why we didn't notice it before.

llvm-svn: 16840

the JIT had last night.

llvm-svn: 16766

t:
        mov %EDX, DWORD PTR [%ESP + 4]
        mov %ECX, 2
        mov %EAX, %EDX
        sar %EDX, 31
        idiv %ECX
        mov %EAX, %EDX
        ret

Generate:
t:
        mov %ECX, DWORD PTR [%ESP + 4]
***     mov %EAX, %ECX
        cdq
        and %ECX, 1
        xor %ECX, %EDX
        sub %ECX, %EDX
***     mov %EAX, %ECX
        ret

Note that the two marked moves are redundant, and should be eliminated by the
register allocator, but aren't.

Compare this to GCC, which generates:

t:
        mov     %eax, DWORD PTR [%esp+4]
        mov     %edx, %eax
        shr     %edx, 31
        lea     %ecx, [%edx+%eax]
        and     %ecx, -2
        sub     %eax, %ecx
        ret

or ICC 8.0, which generates:

t:
        movl      4(%esp), %ecx                                 #3.5
        movl      $-2147483647, %eax                            #3.25
        imull     %ecx                                          #3.25
        movl      %ecx, %eax                                    #3.25
        sarl      $31, %eax                                     #3.25
        addl      %ecx, %edx                                    #3.25
        subl      %edx, %eax                                    #3.25
        addl      %eax, %eax                                    #3.25
        negl      %eax                                          #3.25
        subl      %eax, %ecx                                    #3.25
        movl      %ecx, %eax                                    #3.25
        ret                                                     #3.25

We would be in great shape if not for the moves.

llvm-svn: 16763

s:   ;; X / 4
        mov %EAX, DWORD PTR [%ESP + 4]
        mov %ECX, %EAX
        sar %ECX, 1
        shr %ECX, 30
        mov %EDX, %EAX
        add %EDX, %ECX
        sar %EAX, 2
        ret

When we really meant:

s:
        mov %EAX, DWORD PTR [%ESP + 4]
        mov %ECX, %EAX
        sar %ECX, 1
        shr %ECX, 30
        add %EAX, %ECX
        sar %EAX, 2
        ret

Hey, this also reduces register pressure too :)

llvm-svn: 16761

instead of:

s:   ;; X / 2
        movl 4(%esp), %eax
        movl %eax, %ecx
        shrl $31, %ecx
        movl %eax, %edx
        addl %ecx, %edx
        sarl $1, %eax
        ret

t:   ;; X / -2
        movl 4(%esp), %eax
        movl %eax, %ecx
        shrl $31, %ecx
        movl %eax, %edx
        addl %ecx, %edx
        sarl $1, %eax
        negl %eax
        ret

Emit:

s:
        movl 4(%esp), %eax
        cmpl $-2147483648, %eax
        sbbl $-1, %eax
        sarl $1, %eax
        ret

t:
        movl 4(%esp), %eax
        cmpl $-2147483648, %eax
        sbbl $-1, %eax
        sarl $1, %eax
        negl %eax
        ret

llvm-svn: 16760

llvm-svn: 16759