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

* Update comments
* Rearrange code a bit
* Finally ELIMINATE the GAS workaround emitter for Intel mode.  woot!

llvm-svn: 16647

may now choose their output format with the -x86-asm-syntax={intel|att} flag.

llvm-svn: 16646

llvm-svn: 16645

old and broken AT&T syntax assemblers.  The problem with this hack is that
*SOME* forms of the fdiv and fsub instructions have the 'r' bit inverted.
This was a real pain to figure out, but is trivially easy to support: thus
we are now bug compatible with gas and gcc.

llvm-svn: 16644

llvm-svn: 16642

llvm-svn: 16641

llvm-svn: 16640

Intel and AT&T style assembly language.  The ultimate goal of this is to
eliminate the GasBugWorkaroundEmitter class, but for now AT&T style emission
is not fully operational.

llvm-svn: 16639

hopefully lead to the death of the 'GasBugWorkaroundEmitter'.  This also
includes changes to wrap the whole file to 80 columns! Woot! :)

Note that the AT&T style output has not been tested at all.

llvm-svn: 16638

are only used by the stackifier when transforming FPn register
allocations to the real stack file x87 registers.

llvm-svn: 16472

front-end in gcc does not mangle classes in anonymous namespaces correctly.

llvm-svn: 16469