Compiler output gcc 4.8.1 vs clang +llvm 3.3 in armv7-a with -O.

sum.c

int sum(x)
{
	int sum = 0;
	int i;
	for (i = 1; i <= x; i++) {
		sum += i;
	}
	return sum;
}

#include <stdio.h>
int main()
{
	int n;
	n = 100;
	printf("sum(%d) = %d\n", n, sum(n));
}

sum.s.clang

	.syntax unified
	.cpu cortex-a8
	.eabi_attribute 6, 10
	.eabi_attribute 7, 65
	.eabi_attribute 8, 1
	.eabi_attribute 9, 2
	.eabi_attribute 10, 3
	.fpu vfpv3
	.eabi_attribute 20, 1
	.eabi_attribute 21, 1
	.eabi_attribute 23, 3
	.eabi_attribute 24, 1
	.eabi_attribute 25, 1
	.file	"sum.c"
	.text
	.globl	sum
	.align	2
	.type	sum,%function
sum:
	mov	r1, #0
	cmp	r0, #1
	blt	.LBB0_2
	sub	r1, r0, #2
	sub	r2, r0, #1
	umull	r1, r2, r2, r1
	and	r2, r2, #1
	lsrs	r2, r2, #1
	rrx	r1, r1
	add	r0, r1, r0, lsl #1
	sub	r1, r0, #1
.LBB0_2:
	mov	r0, r1
	bx	lr
.Ltmp0:
	.size	sum, .Ltmp0-sum

	.globl	main
	.align	2
	.type	main,%function
main:
	push	{r11, lr}
	movw	r0, :lower16:.L.str
	mov	r1, #100
	movt	r0, :upper16:.L.str
	movw	r2, #5050
	mov	r11, sp
	bl	printf
	mov	r0, #0
	pop	{r11, pc}
.Ltmp1:
	.size	main, .Ltmp1-main

	.type	.L.str,%object
	.section	.rodata.str1.1,"aMS",%progbits,1
.L.str:
	.asciz	 "sum(%d) = %d\n"
	.size	.L.str, 14

sum.s.gcc

	.syntax unified
	.arch armv7-a
	.eabi_attribute 27, 3
	.eabi_attribute 28, 1
	.fpu vfpv3-d16
	.eabi_attribute 20, 1
	.eabi_attribute 21, 1
	.eabi_attribute 23, 3
	.eabi_attribute 24, 1
	.eabi_attribute 25, 1
	.eabi_attribute 26, 1
	.eabi_attribute 30, 2
	.eabi_attribute 34, 1
	.eabi_attribute 18, 4
	.thumb
	.file	"sum.c"
	.text
	.align	2
	.global	sum
	.thumb
	.thumb_func
	.type	sum, %function
sum:
	@ args = 0, pretend = 0, frame = 0
	@ frame_needed = 0, uses_anonymous_args = 0
	@ link register save eliminated.
	cmp	r0, #0
	ble	.L4
	adds	r2, r0, #1
	movs	r3, #1
	movs	r0, #0
.L3:
	add	r0, r0, r3
	adds	r3, r3, #1
	cmp	r3, r2
	bne	.L3
	bx	lr
.L4:
	movs	r0, #0
	bx	lr
	.size	sum, .-sum
	.section	.text.startup,"ax",%progbits
	.align	2
	.global	main
	.thumb
	.thumb_func
	.type	main, %function
main:
	@ args = 0, pretend = 0, frame = 0
	@ frame_needed = 0, uses_anonymous_args = 0
	@ link register save eliminated.
	movs	r3, #1
	movs	r2, #0
.L9:
	add	r2, r2, r3
	adds	r3, r3, #1
	cmp	r3, #101
	bne	.L9
	movw	r0, #:lower16:.LC0
	movs	r1, #100
	movt	r0, #:upper16:.LC0
	b	printf
	.size	main, .-main
	.section	.rodata.str1.4,"aMS",%progbits,1
	.align	2
.LC0:
	.ascii	"sum(%d) = %d\012\000"
	.ident	"GCC: (GNU) 4.8.1"