make all objects used with atomic operations volatile

the memory model we use internally for atomics permits plain loads of
values which may be subject to concurrent modification without
requiring that a special load function be used. since a compiler is
free to make transformations that alter the number of loads or the way
in which loads are performed, the compiler is theoretically free to
break this usage. the most obvious concern is with atomic cas
constructs: something of the form tmp=*p;a_cas(p,tmp,f(tmp)); could be
transformed to a_cas(p,*p,f(*p)); where the latter is intended to show
multiple loads of *p whose resulting values might fail to be equal;
this would break the atomicity of the whole operation. but even more
fundamental breakage is possible.

with the changes being made now, objects that may be modified by
atomics are modeled as volatile, and the atomic operations performed
on them by other threads are modeled as asynchronous stores by
hardware which happens to be acting on the request of another thread.
such modeling of course does not itself address memory synchronization
between cores/cpus, but that aspect was already handled. this all
seems less than ideal, but it's the best we can do without mandating a
C11 compiler and using the C11 model for atomics.

in the case of pthread_once_t, the ABI type of the underlying object
is not volatile-qualified. so we are assuming that accessing the
object through a volatile-qualified lvalue via casts yields volatile
access semantics. the language of the C standard is somewhat unclear
on this matter, but this is an assumption the linux kernel also makes,
and seems to be the correct interpretation of the standard.

9 files changed, 18 insertions, 16 deletions

diff --git a/src/thread/pthread_atfork.c b/src/thread/pthread_atfork.c
index 95fce207..a40d7f63 100644
--- a/src/thread/pthread_atfork.c
+++ b/src/thread/pthread_atfork.c
@@ -8,7 +8,7 @@ static struct atfork_funcs {
 	struct atfork_funcs *prev, *next;
 } *funcs;
 
-static int lock[2];
+static volatile int lock[2];
 
 void __fork_handler(int who)
 {
diff --git a/src/thread/pthread_barrier_wait.c b/src/thread/pthread_barrier_wait.c
index e15abb84..bfeb3464 100644
--- a/src/thread/pthread_barrier_wait.c
+++ b/src/thread/pthread_barrier_wait.c
@@ -54,10 +54,10 @@ static int pshared_barrier_wait(pthread_barrier_t *b)
 
 struct instance
 {
-	int count;
-	int last;
-	int waiters;
-	int finished;
+	volatile int count;
+	volatile int last;
+	volatile int waiters;
+	volatile int finished;
 };
 
 int pthread_barrier_wait(pthread_barrier_t *b)
diff --git a/src/thread/pthread_cond_timedwait.c b/src/thread/pthread_cond_timedwait.c
index f5fd08c0..27b1a99a 100644
--- a/src/thread/pthread_cond_timedwait.c
+++ b/src/thread/pthread_cond_timedwait.c
@@ -29,8 +29,8 @@ int __pthread_setcancelstate(int, int *);
 
 struct waiter {
 	struct waiter *prev, *next;
-	int state, barrier;
-	int *notify;
+	volatile int state, barrier;
+	volatile int *notify;
 };
 
 /* Self-synchronized-destruction-safe lock functions */
@@ -67,7 +67,8 @@ enum {
 int __pthread_cond_timedwait(pthread_cond_t *restrict c, pthread_mutex_t *restrict m, const struct timespec *restrict ts)
 {
 	struct waiter node = { 0 };
-	int e, seq, *fut, clock = c->_c_clock, cs, shared=0, oldstate, tmp;
+	int e, seq, clock = c->_c_clock, cs, shared=0, oldstate, tmp;
+	volatile int *fut;
 
 	if ((m->_m_type&15) && (m->_m_lock&INT_MAX) != __pthread_self()->tid)
 		return EPERM;
@@ -175,7 +176,8 @@ done:
 int __private_cond_signal(pthread_cond_t *c, int n)
 {
 	struct waiter *p, *first=0;
-	int ref = 0, cur;
+	volatile int ref = 0;
+	int cur;
 
 	lock(&c->_c_lock);
 	for (p=c->_c_tail; n && p; p=p->prev) {
diff --git a/src/thread/pthread_key_create.c b/src/thread/pthread_key_create.c
index bfcd5976..198ae56e 100644
--- a/src/thread/pthread_key_create.c
+++ b/src/thread/pthread_key_create.c
@@ -3,7 +3,7 @@
 volatile size_t __pthread_tsd_size = sizeof(void *) * PTHREAD_KEYS_MAX;
 void *__pthread_tsd_main[PTHREAD_KEYS_MAX] = { 0 };
 
-static void (*keys[PTHREAD_KEYS_MAX])(void *);
+static void (*volatile keys[PTHREAD_KEYS_MAX])(void *);
 
 static void nodtor(void *dummy)
 {
diff --git a/src/thread/pthread_once.c b/src/thread/pthread_once.c
index df655ef9..a8f8aeb1 100644
--- a/src/thread/pthread_once.c
+++ b/src/thread/pthread_once.c
@@ -40,7 +40,7 @@ int __pthread_once(pthread_once_t *control, void (*init)(void))
 {
 	/* Return immediately if init finished before, but ensure that
 	 * effects of the init routine are visible to the caller. */
-	if (*control == 2) {
+	if (*(volatile int *)control == 2) {
 		a_barrier();
 		return 0;
 	}
diff --git a/src/thread/sem_open.c b/src/thread/sem_open.c
index 9a95d257..ab884a42 100644
--- a/src/thread/sem_open.c
+++ b/src/thread/sem_open.c
@@ -20,7 +20,7 @@ static struct {
 	sem_t *sem;
 	int refcnt;
 } *semtab;
-static int lock[2];
+static volatile int lock[2];
 
 #define FLAGS (O_RDWR|O_NOFOLLOW|O_CLOEXEC|O_NONBLOCK)
 
diff --git a/src/thread/sem_timedwait.c b/src/thread/sem_timedwait.c
index a7488df7..8132eb1b 100644
--- a/src/thread/sem_timedwait.c
+++ b/src/thread/sem_timedwait.c
@@ -19,7 +19,7 @@ int sem_timedwait(sem_t *restrict sem, const struct timespec *restrict at)
 		int r;
 		a_inc(sem->__val+1);
 		a_cas(sem->__val, 0, -1);
-		pthread_cleanup_push(cleanup, sem->__val+1);
+		pthread_cleanup_push(cleanup, (void *)(sem->__val+1));
 		r = __timedwait_cp(sem->__val, -1, CLOCK_REALTIME, at, sem->__val[2]);
 		pthread_cleanup_pop(1);
 		if (r && r != EINTR) {
diff --git a/src/thread/synccall.c b/src/thread/synccall.c
index 47d070b4..000ec4e3 100644
--- a/src/thread/synccall.c
+++ b/src/thread/synccall.c
@@ -14,8 +14,8 @@ static struct chain {
 	sem_t target_sem, caller_sem;
 } *volatile head;
 
-static int synccall_lock[2];
-static int target_tid;
+static volatile int synccall_lock[2];
+static volatile int target_tid;
 static void (*callback)(void *), *context;
 static volatile int dummy = 0;
 weak_alias(dummy, __block_new_threads);
diff --git a/src/thread/vmlock.c b/src/thread/vmlock.c
index aba9e311..125c6dc9 100644
--- a/src/thread/vmlock.c
+++ b/src/thread/vmlock.c
@@ -1,6 +1,6 @@
 #include "pthread_impl.h"
 
-static int vmlock[2];
+static volatile int vmlock[2];
 
 void __vm_lock(int inc)
 {