Mercurial > projects > ldc
view tests/mini/asm6.d @ 978:6a32d2e18175
Fix a latent bug in the asm code.
I think that technically, using "*m0" instead of "*0" allows LLVM to pick
between using the same memory as output 0 and using a new memory location.
(So far I haven't been able to construct a testcase that actually breaks
because of this, though)
author | Frits van Bommel <fvbommel wxs.nl> |
---|---|
date | Wed, 18 Feb 2009 03:38:12 +0100 |
parents | 4ac97ec7c18e |
children | 4c524d80e6e1 |
line wrap: on
line source
extern(C) int printf(char*, ...); void main() { int a,b,c; a = int.max-1; b = 5; version (LLVM_InlineAsm_X86) { asm { mov EAX, a; mov ECX, b; add EAX, ECX; jo Loverflow; mov c, EAX; } } else version (LLVM_InlineAsm_X86_64) { asm { movq RDX, a; movq RAX, b; add RDX, RAX; jo Loverflow; movq c, RDX; } } printf("a == %d\n", a); printf("b == %d\n", b); printf("c == %d\n", c); assert(c == c); return; Loverflow: int y=0; //assert(0, "overflow"); }