On x86_64, gcc 4.5.0, this code generates bad assembly:
--- C code ---
typedef unsigned long size_t;
void *memcpy(void *dest, const void *src, size_t n);
void
buggy_init(void *ptr, size_t size)
{
const char *str = "Hello world!";
memcpy(ptr, &str, size);
}
--------------
Compiled with gcc -O -foptimize-sibling-calls, the generated assembly looks
like this:
--------------
buggy_init:
movq %rsi, %rdx
movq $.LC0, -16(%rsp)
leaq -16(%rsp), %rsi
jmp memcpy
--------------
which passes rsp-16 as memcpy's second argument. memcpy overwrites this part
of the stack, and copies the wrong value, which causes a crash later.
--
Summary: Wrong code with -foptimize-sibling-calls and memcpy on
x86_64
Product: gcc
Version: 4.5.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: c
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: tavianator at gmail dot com
GCC build triplet: x86_64-unknown-linux-gnu
GCC host triplet: x86_64-unknown-linux-gnu
GCC target triplet: x86_64-unknown-linux-gnu
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43904
