Question:
There is a cross compiler for the Sparc architecture. But it does not support Built-in functions for atomic memory access. I'm interested in the function __sync_val_compare_and_swap (long *ptr, long oldval long newval)
it performs an atomic compare and exchange operation: That is, if the current value of *ptr is oldval, then write newval into *ptr.
Here is the implementation for i386: Code:
struct __xchg_dummy { unsigned long a[100]; };
#define __xg(x) ((struct __xchg_dummy *)(x))
static inline unsigned long __cmpxchg(volatile void *ptr, unsigned long old,
unsigned long new, int size)
{
unsigned long prev;
switch (size) {
case 1:
__asm__ __volatile__(LOCK_PREFIX "cmpxchgb %b1,%2"
: "=a"(prev)
: "q"(new), "m"(*__xg(ptr)), "0"(old)
: "memory");
return prev;
case 2:
__asm__ __volatile__(LOCK_PREFIX "cmpxchgw %w1,%2"
: "=a"(prev)
: "q"(new), "m"(*__xg(ptr)), "0"(old)
: "memory");
return prev;
case 4:
__asm__ __volatile__(LOCK_PREFIX "cmpxchgl %1,%2"
: "=a"(prev)
: "q"(new), "m"(*__xg(ptr)), "0"(old)
: "memory");
return prev;
}
return old;
}
#define cmpxchg(ptr,o,n)\
((__typeof__(*(ptr)))__cmpxchg((ptr),(unsigned long)(o),\
(unsigned long)(n),sizeof(*(ptr))))
Is anyone familiar with writing assembler inserts (for C code) for the Sparc V8 architecture? I will be very grateful for help
Answer:
For Sparc, the same is done with instructions called CASA and CASXA. You can look for example in FreeBSD: functions : see casa
, casxa
, and their usage atomic_cas_32
, atomic_cas_64
.