Skip to content

Instantly share code, notes, and snippets.

@ranma42

ranma42/clang -std=c99 -c -m32 -O3 test.c -o test.o && objdump -S test.o

Created January 23, 2016 01:20
Show Gist options
  • Save ranma42/2a82c5d8f1f774bd003f to your computer and use it in GitHub Desktop.
Save ranma42/2a82c5d8f1f774bd003f to your computer and use it in GitHub Desktop.
Download ZIP
Raw
clang -std=c99 -c -m32 -march=i686 -O3 test.c -o test.o && objdump -S test.o
test.o: file format elf32-i386
Disassembly of section .text:
00000000 <add>:
0: dd 44 24 04 fldl 0x4(%esp)
4: dc 44 24 0c faddl 0xc(%esp)
8: c3 ret
9: 90 nop
a: 90 nop
b: 90 nop
c: 90 nop
d: 90 nop
e: 90 nop
f: 90 nop
00000010 <addv>:
10: 56 push %esi
11: 8b 44 24 10 mov 0x10(%esp),%eax
15: 85 c0 test %eax,%eax
17: 7e 24 jle 3d <addv+0x2d>
19: 8b 4c 24 0c mov 0xc(%esp),%ecx
1d: 8b 54 24 08 mov 0x8(%esp),%edx
21: 90 nop
22: 90 nop
23: 90 nop
24: 90 nop
25: 90 nop
26: 90 nop
27: 90 nop
28: 90 nop
29: 90 nop
2a: 90 nop
2b: 90 nop
2c: 90 nop
2d: 90 nop
2e: 90 nop
2f: 90 nop
30: 8b 31 mov (%ecx),%esi
32: 01 32 add %esi,(%edx)
34: 83 c1 04 add $0x4,%ecx
37: 83 c2 04 add $0x4,%edx
3a: 48 dec %eax
3b: 75 f3 jne 30 <addv+0x20>
3d: 5e pop %esi
3e: c3 ret
Raw
clang -std=c99 -c -m32 -O3 test.c -o test.o && objdump -S test.o
test.o: file format elf32-i386
Disassembly of section .text:
00000000 <add>:
0: 83 ec 0c sub $0xc,%esp
3: f2 0f 10 44 24 10 movsd 0x10(%esp),%xmm0
9: f2 0f 58 44 24 18 addsd 0x18(%esp),%xmm0
f: f2 0f 11 04 24 movsd %xmm0,(%esp)
14: dd 04 24 fldl (%esp)
17: 83 c4 0c add $0xc,%esp
1a: c3 ret
1b: 0f 1f 44 00 00 nopl 0x0(%eax,%eax,1)
00000020 <addv>:
20: 57 push %edi
21: 56 push %esi
22: 8b 44 24 14 mov 0x14(%esp),%eax
26: 85 c0 test %eax,%eax
28: 7e 63 jle 8d <addv+0x6d>
2a: 8b 54 24 10 mov 0x10(%esp),%edx
2e: 8b 4c 24 0c mov 0xc(%esp),%ecx
32: 31 ff xor %edi,%edi
34: 89 c6 mov %eax,%esi
36: 83 e6 f8 and $0xfffffff8,%esi
39: 74 37 je 72 <addv+0x52>
3b: 31 ff xor %edi,%edi
3d: 0f 1f 00 nopl (%eax)
40: f3 0f 6f 04 ba movdqu (%edx,%edi,4),%xmm0
45: f3 0f 6f 4c ba 10 movdqu 0x10(%edx,%edi,4),%xmm1
4b: f3 0f 6f 14 b9 movdqu (%ecx,%edi,4),%xmm2
50: f3 0f 6f 5c b9 10 movdqu 0x10(%ecx,%edi,4),%xmm3
56: 66 0f fe d0 paddd %xmm0,%xmm2
5a: 66 0f fe d9 paddd %xmm1,%xmm3
5e: f3 0f 7f 14 b9 movdqu %xmm2,(%ecx,%edi,4)
63: f3 0f 7f 5c b9 10 movdqu %xmm3,0x10(%ecx,%edi,4)
69: 83 c7 08 add $0x8,%edi
6c: 39 fe cmp %edi,%esi
6e: 75 d0 jne 40 <addv+0x20>
70: 89 f7 mov %esi,%edi
72: 39 c7 cmp %eax,%edi
74: 74 17 je 8d <addv+0x6d>
76: 8d 14 ba lea (%edx,%edi,4),%edx
79: 8d 0c b9 lea (%ecx,%edi,4),%ecx
7c: 29 f8 sub %edi,%eax
7e: 66 90 xchg %ax,%ax
80: 8b 32 mov (%edx),%esi
82: 01 31 add %esi,(%ecx)
84: 83 c2 04 add $0x4,%edx
87: 83 c1 04 add $0x4,%ecx
8a: 48 dec %eax
8b: 75 f3 jne 80 <addv+0x60>
8d: 5e pop %esi
8e: 5f pop %edi
8f: c3 ret
Raw
gcc -std=c99 -c -m32 -march=pentium4 -mfpmath=sse -O3 test.c -o test.o && objdump -S test.o
test.o: file format elf32-i386
Disassembly of section .text:
00000000 <add>:
0: 83 ec 0c sub $0xc,%esp
3: f2 0f 10 44 24 18 movsd 0x18(%esp),%xmm0
9: f2 0f 58 44 24 10 addsd 0x10(%esp),%xmm0
f: f2 0f 11 04 24 movsd %xmm0,(%esp)
14: dd 04 24 fldl (%esp)
17: 83 c4 0c add $0xc,%esp
1a: c3 ret
0000001b <addv>:
1b: 55 push %ebp
1c: 57 push %edi
1d: 56 push %esi
1e: 53 push %ebx
1f: 83 ec 1c sub $0x1c,%esp
22: 8b 44 24 30 mov 0x30(%esp),%eax
26: 8b 4c 24 34 mov 0x34(%esp),%ecx
2a: 8b 54 24 38 mov 0x38(%esp),%edx
2e: 85 d2 test %edx,%edx
30: 0f 8e eb 00 00 00 jle 121 <addv+0x106>
36: 89 c3 mov %eax,%ebx
38: 83 e3 0f and $0xf,%ebx
3b: c1 eb 02 shr $0x2,%ebx
3e: f7 db neg %ebx
40: 83 e3 03 and $0x3,%ebx
43: 39 d3 cmp %edx,%ebx
45: 89 de mov %ebx,%esi
47: 0f 47 f2 cmova %edx,%esi
4a: 89 d3 mov %edx,%ebx
4c: 83 fa 04 cmp $0x4,%edx
4f: 0f 87 d4 00 00 00 ja 129 <addv+0x10e>
55: 8b 31 mov (%ecx),%esi
57: 01 30 add %esi,(%eax)
59: 83 fb 01 cmp $0x1,%ebx
5c: 0f 86 de 00 00 00 jbe 140 <addv+0x125>
62: 8b 71 04 mov 0x4(%ecx),%esi
65: 01 70 04 add %esi,0x4(%eax)
68: 83 fb 02 cmp $0x2,%ebx
6b: 0f 86 d9 00 00 00 jbe 14a <addv+0x12f>
71: 8b 71 08 mov 0x8(%ecx),%esi
74: 01 70 08 add %esi,0x8(%eax)
77: 83 fb 03 cmp $0x3,%ebx
7a: 0f 86 b6 00 00 00 jbe 136 <addv+0x11b>
80: 8b 71 0c mov 0xc(%ecx),%esi
83: 01 70 0c add %esi,0xc(%eax)
86: be 04 00 00 00 mov $0x4,%esi
8b: 39 da cmp %ebx,%edx
8d: 0f 84 8e 00 00 00 je 121 <addv+0x106>
93: 89 d7 mov %edx,%edi
95: 29 df sub %ebx,%edi
97: 89 7c 24 08 mov %edi,0x8(%esp)
9b: 89 fd mov %edi,%ebp
9d: c1 ed 02 shr $0x2,%ebp
a0: 89 2c 24 mov %ebp,(%esp)
a3: 89 fd mov %edi,%ebp
a5: 83 e5 fc and $0xfffffffc,%ebp
a8: 89 6c 24 0c mov %ebp,0xc(%esp)
ac: 74 46 je f4 <addv+0xd9>
ae: 01 db add %ebx,%ebx
b0: 01 db add %ebx,%ebx
b2: 8d 2c 18 lea (%eax,%ebx,1),%ebp
b5: 8d 3c 19 lea (%ecx,%ebx,1),%edi
b8: 89 7c 24 04 mov %edi,0x4(%esp)
bc: 31 db xor %ebx,%ebx
be: 31 ff xor %edi,%edi
c0: 89 44 24 30 mov %eax,0x30(%esp)
c4: 8b 44 24 04 mov 0x4(%esp),%eax
c8: f3 0f 6f 04 18 movdqu (%eax,%ebx,1),%xmm0
cd: 66 0f fe 44 1d 00 paddd 0x0(%ebp,%ebx,1),%xmm0
d3: 66 0f 7f 44 1d 00 movdqa %xmm0,0x0(%ebp,%ebx,1)
d9: 83 c7 01 add $0x1,%edi
dc: 83 c3 10 add $0x10,%ebx
df: 39 3c 24 cmp %edi,(%esp)
e2: 77 e4 ja c8 <addv+0xad>
e4: 8b 44 24 30 mov 0x30(%esp),%eax
e8: 8b 7c 24 0c mov 0xc(%esp),%edi
ec: 01 fe add %edi,%esi
ee: 39 7c 24 08 cmp %edi,0x8(%esp)
f2: 74 2d je 121 <addv+0x106>
f4: 8d 1c 36 lea (%esi,%esi,1),%ebx
f7: 01 db add %ebx,%ebx
f9: 8b 3c 19 mov (%ecx,%ebx,1),%edi
fc: 01 3c 18 add %edi,(%eax,%ebx,1)
ff: 8d 5e 01 lea 0x1(%esi),%ebx
102: 39 da cmp %ebx,%edx
104: 7e 1b jle 121 <addv+0x106>
106: 01 db add %ebx,%ebx
108: 01 db add %ebx,%ebx
10a: 8b 3c 19 mov (%ecx,%ebx,1),%edi
10d: 01 3c 18 add %edi,(%eax,%ebx,1)
110: 83 c6 02 add $0x2,%esi
113: 39 f2 cmp %esi,%edx
115: 7e 0a jle 121 <addv+0x106>
117: 01 f6 add %esi,%esi
119: 01 f6 add %esi,%esi
11b: 8b 14 31 mov (%ecx,%esi,1),%edx
11e: 01 14 30 add %edx,(%eax,%esi,1)
121: 83 c4 1c add $0x1c,%esp
124: 5b pop %ebx
125: 5e pop %esi
126: 5f pop %edi
127: 5d pop %ebp
128: c3 ret
129: 85 f6 test %esi,%esi
12b: 75 27 jne 154 <addv+0x139>
12d: 31 db xor %ebx,%ebx
12f: 31 f6 xor %esi,%esi
131: e9 5d ff ff ff jmp 93 <addv+0x78>
136: be 03 00 00 00 mov $0x3,%esi
13b: e9 4b ff ff ff jmp 8b <addv+0x70>
140: be 01 00 00 00 mov $0x1,%esi
145: e9 41 ff ff ff jmp 8b <addv+0x70>
14a: be 02 00 00 00 mov $0x2,%esi
14f: e9 37 ff ff ff jmp 8b <addv+0x70>
154: 89 f3 mov %esi,%ebx
156: e9 fa fe ff ff jmp 55 <addv+0x3a>
Raw
gcc -std=c99 -c -m32 -march=pentium4 -O3 test.c -o test.o && objdump -S test.o
test.o: file format elf32-i386
Disassembly of section .text:
00000000 <add>:
0: 83 ec 0c sub $0xc,%esp
3: dd 44 24 18 fldl 0x18(%esp)
7: dc 44 24 10 faddl 0x10(%esp)
b: dd 1c 24 fstpl (%esp)
e: dd 04 24 fldl (%esp)
11: 83 c4 0c add $0xc,%esp
14: c3 ret
00000015 <addv>:
15: 55 push %ebp
16: 57 push %edi
17: 56 push %esi
18: 53 push %ebx
19: 83 ec 1c sub $0x1c,%esp
1c: 8b 44 24 30 mov 0x30(%esp),%eax
20: 8b 4c 24 34 mov 0x34(%esp),%ecx
24: 8b 54 24 38 mov 0x38(%esp),%edx
28: 85 d2 test %edx,%edx
2a: 0f 8e eb 00 00 00 jle 11b <addv+0x106>
30: 89 c3 mov %eax,%ebx
32: 83 e3 0f and $0xf,%ebx
35: c1 eb 02 shr $0x2,%ebx
38: f7 db neg %ebx
3a: 83 e3 03 and $0x3,%ebx
3d: 39 d3 cmp %edx,%ebx
3f: 89 de mov %ebx,%esi
41: 0f 47 f2 cmova %edx,%esi
44: 89 d3 mov %edx,%ebx
46: 83 fa 04 cmp $0x4,%edx
49: 0f 87 d4 00 00 00 ja 123 <addv+0x10e>
4f: 8b 31 mov (%ecx),%esi
51: 01 30 add %esi,(%eax)
53: 83 fb 01 cmp $0x1,%ebx
56: 0f 86 de 00 00 00 jbe 13a <addv+0x125>
5c: 8b 71 04 mov 0x4(%ecx),%esi
5f: 01 70 04 add %esi,0x4(%eax)
62: 83 fb 02 cmp $0x2,%ebx
65: 0f 86 d9 00 00 00 jbe 144 <addv+0x12f>
6b: 8b 71 08 mov 0x8(%ecx),%esi
6e: 01 70 08 add %esi,0x8(%eax)
71: 83 fb 03 cmp $0x3,%ebx
74: 0f 86 b6 00 00 00 jbe 130 <addv+0x11b>
7a: 8b 71 0c mov 0xc(%ecx),%esi
7d: 01 70 0c add %esi,0xc(%eax)
80: be 04 00 00 00 mov $0x4,%esi
85: 39 da cmp %ebx,%edx
87: 0f 84 8e 00 00 00 je 11b <addv+0x106>
8d: 89 d7 mov %edx,%edi
8f: 29 df sub %ebx,%edi
91: 89 7c 24 08 mov %edi,0x8(%esp)
95: 89 fd mov %edi,%ebp
97: c1 ed 02 shr $0x2,%ebp
9a: 89 2c 24 mov %ebp,(%esp)
9d: 89 fd mov %edi,%ebp
9f: 83 e5 fc and $0xfffffffc,%ebp
a2: 89 6c 24 0c mov %ebp,0xc(%esp)
a6: 74 46 je ee <addv+0xd9>
a8: 01 db add %ebx,%ebx
aa: 01 db add %ebx,%ebx
ac: 8d 2c 18 lea (%eax,%ebx,1),%ebp
af: 8d 3c 19 lea (%ecx,%ebx,1),%edi
b2: 89 7c 24 04 mov %edi,0x4(%esp)
b6: 31 db xor %ebx,%ebx
b8: 31 ff xor %edi,%edi
ba: 89 44 24 30 mov %eax,0x30(%esp)
be: 8b 44 24 04 mov 0x4(%esp),%eax
c2: f3 0f 6f 04 18 movdqu (%eax,%ebx,1),%xmm0
c7: 66 0f fe 44 1d 00 paddd 0x0(%ebp,%ebx,1),%xmm0
cd: 66 0f 7f 44 1d 00 movdqa %xmm0,0x0(%ebp,%ebx,1)
d3: 83 c7 01 add $0x1,%edi
d6: 83 c3 10 add $0x10,%ebx
d9: 39 3c 24 cmp %edi,(%esp)
dc: 77 e4 ja c2 <addv+0xad>
de: 8b 44 24 30 mov 0x30(%esp),%eax
e2: 8b 7c 24 0c mov 0xc(%esp),%edi
e6: 01 fe add %edi,%esi
e8: 39 7c 24 08 cmp %edi,0x8(%esp)
ec: 74 2d je 11b <addv+0x106>
ee: 8d 1c 36 lea (%esi,%esi,1),%ebx
f1: 01 db add %ebx,%ebx
f3: 8b 3c 19 mov (%ecx,%ebx,1),%edi
f6: 01 3c 18 add %edi,(%eax,%ebx,1)
f9: 8d 5e 01 lea 0x1(%esi),%ebx
fc: 39 da cmp %ebx,%edx
fe: 7e 1b jle 11b <addv+0x106>
100: 01 db add %ebx,%ebx
102: 01 db add %ebx,%ebx
104: 8b 3c 19 mov (%ecx,%ebx,1),%edi
107: 01 3c 18 add %edi,(%eax,%ebx,1)
10a: 83 c6 02 add $0x2,%esi
10d: 39 f2 cmp %esi,%edx
10f: 7e 0a jle 11b <addv+0x106>
111: 01 f6 add %esi,%esi
113: 01 f6 add %esi,%esi
115: 8b 14 31 mov (%ecx,%esi,1),%edx
118: 01 14 30 add %edx,(%eax,%esi,1)
11b: 83 c4 1c add $0x1c,%esp
11e: 5b pop %ebx
11f: 5e pop %esi
120: 5f pop %edi
121: 5d pop %ebp
122: c3 ret
123: 85 f6 test %esi,%esi
125: 75 27 jne 14e <addv+0x139>
127: 31 db xor %ebx,%ebx
129: 31 f6 xor %esi,%esi
12b: e9 5d ff ff ff jmp 8d <addv+0x78>
130: be 03 00 00 00 mov $0x3,%esi
135: e9 4b ff ff ff jmp 85 <addv+0x70>
13a: be 01 00 00 00 mov $0x1,%esi
13f: e9 41 ff ff ff jmp 85 <addv+0x70>
144: be 02 00 00 00 mov $0x2,%esi
149: e9 37 ff ff ff jmp 85 <addv+0x70>
14e: 89 f3 mov %esi,%ebx
150: e9 fa fe ff ff jmp 4f <addv+0x3a>
Raw
gcc -std=c99 -c -m32 -O3 test.c -o test.o && objdump -S test.o
test.o: file format elf32-i386
Disassembly of section .text:
00000000 <add>:
0: 83 ec 0c sub $0xc,%esp
3: dd 44 24 18 fldl 0x18(%esp)
7: dc 44 24 10 faddl 0x10(%esp)
b: dd 1c 24 fstpl (%esp)
e: dd 04 24 fldl (%esp)
11: 83 c4 0c add $0xc,%esp
14: c3 ret
15: 8d 74 26 00 lea 0x0(%esi,%eiz,1),%esi
19: 8d bc 27 00 00 00 00 lea 0x0(%edi,%eiz,1),%edi
00000020 <addv>:
20: 56 push %esi
21: 31 c0 xor %eax,%eax
23: 53 push %ebx
24: 8b 5c 24 14 mov 0x14(%esp),%ebx
28: 8b 54 24 0c mov 0xc(%esp),%edx
2c: 8b 74 24 10 mov 0x10(%esp),%esi
30: 85 db test %ebx,%ebx
32: 7e 11 jle 45 <addv+0x25>
34: 8d 74 26 00 lea 0x0(%esi,%eiz,1),%esi
38: 8b 0c 86 mov (%esi,%eax,4),%ecx
3b: 01 0c 82 add %ecx,(%edx,%eax,4)
3e: 83 c0 01 add $0x1,%eax
41: 39 d8 cmp %ebx,%eax
43: 75 f3 jne 38 <addv+0x18>
45: 5b pop %ebx
46: 5e pop %esi
47: c3 ret
Raw
test.c
double add(double x, double y)
{
return x+y;
}
void addv(int *restrict x, int *restrict y, int n)
{
for (int i = 0; i < n; i++)
x[i] += y[i];
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment