Programming language benchmark

[bearophile]

Don:

Sorry for my slow answer, I was quite busy for days.


> I've never heard that claim before. Do you have evidence for that?

I compare/convert code to D every day, so I am aware that D code compiled with DMD is often slower than C/C++ code compiled with GCC. Since some years I even keep a collection of snippets of slow code.

But I am also aware that the low performance has many different causes, like some missing inlining, missing loop unrolling, etc, so spotting a clear and small case of integer arithmetic code that causes a slow down, to give you evidence, is not easy. So I am sorry for my overly broad claim. 


>If it is true, there's a strong possibility that it's a small, fixable issue (for example, DMD used to have terrible performance for ulong multiplication).<

You are right, the case I'm going to show is a precise problem that's fixable. 

-----------------------

// C code
#include "limits.h"
#include "stdio.h"

int divideBySeven(int x) {
    return x / 7;
}

int main() {
    int i = INT_MAX;
    int r;
    while (i--)
        r = divideBySeven(i);

    printf("%d\n", r);
    return 0;
}

-----------------------

// D code
int divideBySeven(int x) {
    return x / 7;
}

void main() {
    int i = int.max;
    int r;
    while (i--)
        r = divideBySeven(i);
    printf("%d\n", r);
}

-----------------------

Asm from the C version:

_divideBySeven:
	pushl	%ebx
	movl	$-1840700269, %ebx
	movl	8(%esp), %ecx
	movl	%ebx, %eax
	popl	%ebx
	imull	%ecx
	leal	(%edx,%ecx), %eax
	sarl	$31, %ecx
	sarl	$2, %eax
	subl	%ecx, %eax
	ret

_main:
	leal	4(%esp), %ecx
	andl	$-16, %esp
	pushl	-4(%ecx)
	pushl	%ebx
	movl	$-1840700269, %ebx
	pushl	%ecx
	subl	$20, %esp
	call	___main
	movl	$2147483646, %ecx
	.p2align 4,,10
L4:
	movl	%ecx, %eax
	imull	%ebx
	movl	%ecx, %eax
	addl	%ecx, %edx
	sarl	$31, %eax
	sarl	$2, %edx
	decl	%ecx
	subl	%eax, %edx
	cmpl	$-1, %ecx
	jne	L4
	movl	%edx, 4(%esp)
	movl	$LC0, (%esp)
	call	_printf
	addl	$20, %esp
	xorl	%eax, %eax
	popl	%ecx
	popl	%ebx
	leal	-4(%ecx), %esp
	ret
	.def	_printf;	.scl	2;	.type	32;	.endef

-----------------------

Asm from the D version:

_D9int_div_d13divideBySevenFiZi	comdat
		mov	ECX,7
		cdq
		idiv	ECX
		ret

__Dmain	comdat
L0:		push	EAX
		push	EBX
		mov	EBX,07FFFFFFFh
		push	ESI
		xor	ESI,ESI
		test	EBX,EBX
		lea	EBX,-1[EBX]
		je	L24
L11:		mov	EAX,EBX
		mov	ECX,7
		cdq
		idiv	ECX
		test	EBX,EBX
		mov	ESI,EAX
		lea	EBX,-1[EBX]
		jne	L11
L24:		push	ESI
		mov	EDX,offset FLAT:_DATA
		push	EDX
		call	near ptr _printf
		add	ESP,8
		xor	EAX,EAX
		pop	ESI
		pop	EBX
		pop	ECX
		ret

-----------------------

For a more real case see:
http://d.puremagic.com/issues/show_bug.cgi?id=5607

Bye,
bearophile