Re: Is je more effective than jg
- From: rwt <ratwithouttail@xxxxxxxxx>
- Date: Mon, 27 Nov 2006 19:39:54 +0100
I wonder is jne more effective than jbe, if so why?
The statement:
for(i=0;i<=10;++i)
{
}
seems slower than:
for(i=0;i!=10;++i)
{
}
?
int
main()
{
const int count = 10;
int i;
for ( i = 0; i <= count - 1; i++ ) {}
for ( i = 0; i != count; i++ ) {}
return 0;
}
$ cc -S -Os -momit-leaf-frame-pointer test.c
$ cc --version
cc (GCC) 3.3.5 (propolice)
main:
movl $9, %eax
..L6:
decl %eax
jns .L6 # <-- 1
xorl %eax, %eax
..L11:
incl %eax
cmpl $10, %eax # <-- 2
jne .L11 # <-- 2
xorl %eax, %eax
ret
--
/ qo |) :@=N%_g=v=a=g_eD_e=c()=d=8! =%!Gn@8're. w8in/ad
\ _x/ , ;h-%-a'hA'H4,X0'Xo~xo~xO,R`-%EXp01ITed: *-7/+eh
/ | ng `-%__%--'__%--'__%--~__%--^%B`/$qV3r[o; &GooMee
L o_O http://tech.groups.yahoo.com/group/opRWTng O_o L"EnOF"
.
- References:
- Is je more effective than jg
- From: leon800219
- Is je more effective than jg
- Prev by Date: Re: GeForce / ATI?
- Next by Date: Re: GeForce / ATI?
- Previous by thread: Re: Is je more effective than jg
- Next by thread: GeForce / ATI?
- Index(es):
Relevant Pages
|