/* Copyright 1995-1998,2000-2014,2018-2019
Free Software Foundation, Inc.
This file is part of Guile.
Guile is free software: you can redistribute it and/or modify it
under the terms of the GNU Lesser General Public License as published
by the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
Guile is distributed in the hope that it will be useful, but WITHOUT
ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public
License for more details.
You should have received a copy of the GNU Lesser General Public
License along with Guile. If not, see
. */
#ifdef HAVE_CONFIG_H
# include
#endif
#include
#include
#include
#include
#include
#include
#include
#include /* for memset used by FD_ZERO on Solaris 10 */
#include
#include
#if HAVE_PTHREAD_NP_H
# include
#endif
#include "async.h"
#include "bdw-gc.h"
#include "boolean.h"
#include "continuations.h"
#include "deprecation.h"
#include "dynwind.h"
#include "eval.h"
#include "extensions.h"
#include "fluids.h"
#include "gc-inline.h"
#include "gc.h"
#include "gsubr.h"
#include "hashtab.h"
#include "init.h"
#include "iselect.h"
#include "jit.h"
#include "list.h"
#include "modules.h"
#include "numbers.h"
#include "pairs.h"
#include "ports.h"
#include "scmsigs.h"
#include "strings.h"
#include "symbols.h"
#include "variable.h"
#include "version.h"
#include "vm.h"
#include "threads.h"
#include
/* The GC "kind" for threads that allow them to mark their VM
stacks. */
static int thread_gc_kind;
static struct GC_ms_entry *
thread_mark (GC_word *addr, struct GC_ms_entry *mark_stack_ptr,
struct GC_ms_entry *mark_stack_limit, GC_word env)
{
int word;
struct scm_thread *t = (struct scm_thread *) addr;
if (SCM_UNPACK (t->handle) == 0)
/* T must be on the free-list; ignore. (See warning in
gc_mark.h.) */
return mark_stack_ptr;
/* Mark T. We could be more precise, but it doesn't matter. */
for (word = 0; word * sizeof (*addr) < sizeof (*t); word++)
mark_stack_ptr = GC_MARK_AND_PUSH ((void *) addr[word],
mark_stack_ptr, mark_stack_limit,
NULL);
/* The pointerless freelists are threaded through their first word,
but GC doesn't know to trace them (as they are pointerless), so we
need to do that here. See the comments at the top of libgc's
gc_inline.h. */
for (size_t n = 0; n < SCM_INLINE_GC_FREELIST_COUNT; n++)
{
void *chain = t->pointerless_freelists[n];
if (chain)
{
/* The first link is already marked by the thread itsel, so we
just have to mark the tail. */
while ((chain = *(void **)chain))
mark_stack_ptr = GC_mark_and_push (chain, mark_stack_ptr,
mark_stack_limit, NULL);
}
}
mark_stack_ptr = scm_i_vm_mark_stack (&t->vm, mark_stack_ptr,
mark_stack_limit);
return mark_stack_ptr;
}
static void
to_timespec (SCM t, scm_t_timespec *waittime)
{
if (scm_is_pair (t))
{
waittime->tv_sec = scm_to_ulong (SCM_CAR (t));
waittime->tv_nsec = scm_to_ulong (SCM_CDR (t)) * 1000;
}
else
{
double time = scm_to_double (t);
double sec = scm_c_truncate (time);
waittime->tv_sec = (long) sec;
waittime->tv_nsec = (long) ((time - sec) * 1000000000);
}
}
/*** Queues */
/* Note: We annotate with "GC-robust" assignments whose purpose is to avoid
the risk of false references leading to unbounded retained space as
described in "Bounding Space Usage of Conservative Garbage Collectors",
H.J. Boehm, 2001. */
/* Make an empty queue data structure.
*/
static SCM
make_queue ()
{
return scm_cons (SCM_EOL, SCM_EOL);
}
static scm_i_pthread_mutex_t queue_lock = SCM_I_PTHREAD_MUTEX_INITIALIZER;
/* Put T at the back of Q and return a handle that can be used with
remqueue to remove T from Q again.
*/
static SCM
enqueue (SCM q, SCM t)
{
SCM c = scm_cons (t, SCM_EOL);
scm_i_pthread_mutex_lock (&queue_lock);
if (scm_is_null (SCM_CDR (q)))
SCM_SETCDR (q, c);
else
SCM_SETCDR (SCM_CAR (q), c);
SCM_SETCAR (q, c);
scm_i_pthread_mutex_unlock (&queue_lock);
return c;
}
/* Remove the element that the handle C refers to from the queue Q. C
must have been returned from a call to enqueue. The return value
is zero when the element referred to by C has already been removed.
Otherwise, 1 is returned.
*/
static int
remqueue (SCM q, SCM c)
{
SCM p, prev = q;
scm_i_pthread_mutex_lock (&queue_lock);
for (p = SCM_CDR (q); !scm_is_null (p); p = SCM_CDR (p))
{
if (scm_is_eq (p, c))
{
if (scm_is_eq (c, SCM_CAR (q)))
SCM_SETCAR (q, scm_is_eq (prev, q) ? SCM_EOL : prev);
SCM_SETCDR (prev, SCM_CDR (c));
/* GC-robust */
SCM_SETCDR (c, SCM_EOL);
scm_i_pthread_mutex_unlock (&queue_lock);
return 1;
}
prev = p;
}
scm_i_pthread_mutex_unlock (&queue_lock);
return 0;
}
/* Remove the front-most element from the queue Q and return it.
Return SCM_BOOL_F when Q is empty.
*/
static SCM
dequeue (SCM q)
{
SCM c;
scm_i_pthread_mutex_lock (&queue_lock);
c = SCM_CDR (q);
if (scm_is_null (c))
{
scm_i_pthread_mutex_unlock (&queue_lock);
return SCM_BOOL_F;
}
else
{
SCM_SETCDR (q, SCM_CDR (c));
if (scm_is_null (SCM_CDR (q)))
SCM_SETCAR (q, SCM_EOL);
scm_i_pthread_mutex_unlock (&queue_lock);
/* GC-robust */
SCM_SETCDR (c, SCM_EOL);
return SCM_CAR (c);
}
}
/*** Thread smob routines */
static int
thread_print (SCM exp, SCM port, scm_print_state *pstate SCM_UNUSED)
{
/* On a Gnu system pthread_t is an unsigned long, but on mingw it's a
struct. A cast like "(unsigned long) t->pthread" is a syntax error in
the struct case, hence we go via a union, and extract according to the
size of pthread_t. */
union {
scm_i_pthread_t p;
unsigned short us;
unsigned int ui;
unsigned long ul;
uintmax_t um;
} u;
scm_thread *t = SCM_I_THREAD_DATA (exp);
scm_i_pthread_t p = t->pthread;
uintmax_t id;
u.p = p;
if (sizeof (p) == sizeof (unsigned short))
id = u.us;
else if (sizeof (p) == sizeof (unsigned int))
id = u.ui;
else if (sizeof (p) == sizeof (unsigned long))
id = u.ul;
else
id = u.um;
scm_puts ("#", port);
return 1;
}
/*** Blocking on queues. */
/* See also scm_system_async_mark_for_thread for how such a block is
interrputed.
*/
/* Put the current thread on QUEUE and go to sleep, waiting for it to
be woken up by a call to 'unblock_from_queue', or to be
interrupted. Upon return of this function, the current thread is
no longer on QUEUE, even when the sleep has been interrupted.
The caller of block_self must hold MUTEX. It will be atomically
unlocked while sleeping, just as with scm_i_pthread_cond_wait.
When WAITTIME is not NULL, the sleep will be aborted at that time.
The return value of block_self is an errno value. It will be zero
when the sleep has been successfully completed by a call to
unblock_from_queue, EINTR when it has been interrupted by the
delivery of a system async, and ETIMEDOUT when the timeout has
expired.
The system asyncs themselves are not executed by block_self.
*/
static int
block_self (SCM queue, scm_i_pthread_mutex_t *mutex,
const scm_t_timespec *waittime)
{
scm_thread *t = SCM_I_CURRENT_THREAD;
SCM q_handle;
int err;
if (scm_i_prepare_to_wait_on_cond (t, mutex, &t->sleep_cond))
return EINTR;
t->block_asyncs++;
q_handle = enqueue (queue, t->handle);
if (waittime == NULL)
err = scm_i_scm_pthread_cond_wait (&t->sleep_cond, mutex);
else
err = scm_i_scm_pthread_cond_timedwait (&t->sleep_cond, mutex, waittime);
/* When we are still on QUEUE, we have been interrupted. We
report this only when no other error (such as a timeout) has
happened above.
*/
if (remqueue (queue, q_handle) && err == 0)
err = EINTR;
t->block_asyncs--;
scm_i_wait_finished (t);
return err;
}
/* Wake up the first thread on QUEUE, if any. The awoken thread is
returned, or #f if the queue was empty.
*/
static SCM
unblock_from_queue (SCM queue)
{
SCM thread = dequeue (queue);
if (scm_is_true (thread))
scm_i_pthread_cond_signal (&SCM_I_THREAD_DATA(thread)->sleep_cond);
return thread;
}
/* Getting into and out of guile mode.
*/
/* Key used to attach a cleanup handler to a given thread. Also, if
thread-local storage is unavailable, this key is used to retrieve the
current thread with `pthread_getspecific ()'. */
scm_i_pthread_key_t scm_i_thread_key;
#ifdef SCM_HAVE_THREAD_STORAGE_CLASS
/* When thread-local storage (TLS) is available, a pointer to the
current-thread object is kept in TLS. Note that storing the thread-object
itself in TLS (rather than a pointer to some malloc'd memory) is not
possible since thread objects may live longer than the actual thread they
represent. */
/* SCM_INTERNAL fixes https://debbugs.gnu.org/60234 */
SCM_INTERNAL SCM_THREAD_LOCAL scm_thread *scm_i_current_thread = NULL;
#endif /* SCM_HAVE_THREAD_STORAGE_CLASS */
static scm_i_pthread_mutex_t thread_admin_mutex = SCM_I_PTHREAD_MUTEX_INITIALIZER;
static scm_thread *all_threads = NULL;
static int thread_count;
static SCM default_dynamic_state;
/* Perform first stage of thread initialisation, in non-guile mode.
*/
static void
guilify_self_1 (struct GC_stack_base *base, int needs_unregister)
{
scm_thread t;
/* We must arrange for SCM_I_CURRENT_THREAD to point to a valid value
before allocating anything in this thread, because allocation could
cause GC to run, and GC could cause finalizers, which could invoke
Scheme functions, which need the current thread to be set. */
memset (&t, 0, sizeof (t));
t.pthread = scm_i_pthread_self ();
t.handle = SCM_BOOL_F;
t.result = SCM_BOOL_F;
t.pending_asyncs = SCM_EOL;
t.block_asyncs = 1;
t.base = base->mem_base;
t.continuation_root = SCM_EOL;
t.continuation_base = t.base;
scm_i_pthread_cond_init (&t.sleep_cond, NULL);
scm_i_vm_prepare_stack (&t.vm);
if (pipe2 (t.sleep_pipe, O_CLOEXEC) != 0)
/* FIXME: Error conditions during the initialization phase are handled
gracelessly since public functions such as `scm_init_guile ()'
currently have type `void'. */
abort ();
t.exited = 0;
t.guile_mode = 0;
t.needs_unregister = needs_unregister;
/* The switcheroo. */
{
scm_thread *t_ptr = &t;
GC_disable ();
t_ptr = GC_generic_malloc (sizeof (*t_ptr), thread_gc_kind);
memcpy (t_ptr, &t, sizeof t);
scm_i_pthread_setspecific (scm_i_thread_key, t_ptr);
#ifdef SCM_HAVE_THREAD_STORAGE_CLASS
/* Cache the current thread in TLS for faster lookup. */
scm_i_current_thread = t_ptr;
#endif
scm_i_pthread_mutex_lock (&thread_admin_mutex);
t_ptr->next_thread = all_threads;
all_threads = t_ptr;
thread_count++;
scm_i_pthread_mutex_unlock (&thread_admin_mutex);
GC_enable ();
}
}
/* Perform second stage of thread initialisation, in guile mode.
*/
static void
guilify_self_2 (SCM dynamic_state)
{
scm_thread *t = SCM_I_CURRENT_THREAD;
t->guile_mode = 1;
SCM_NEWSMOB (t->handle, scm_tc16_thread, t);
t->continuation_root = scm_cons (t->handle, SCM_EOL);
t->continuation_base = t->base;
t->dynamic_state = scm_gc_typed_calloc (scm_t_dynamic_state);
t->dynamic_state->thread_local_values = scm_c_make_hash_table (0);
scm_set_current_dynamic_state (dynamic_state);
t->dynstack.base = scm_gc_malloc (16 * sizeof (scm_t_bits), "dynstack");
t->dynstack.limit = t->dynstack.base + 16;
t->dynstack.top = t->dynstack.base + SCM_DYNSTACK_HEADER_LEN;
t->block_asyncs = 0;
/* See note in finalizers.c:queue_finalizer_async(). */
GC_invoke_finalizers ();
}
static void
on_thread_exit (void *v)
{
/* This handler is executed in non-guile mode. Note that although
libgc isn't guaranteed to see thread-locals, for this thread-local
that isn't an issue as we have the all_threads list. */
scm_thread *t = (scm_thread *) v, **tp;
t->exited = 1;
close (t->sleep_pipe[0]);
close (t->sleep_pipe[1]);
t->sleep_pipe[0] = t->sleep_pipe[1] = -1;
scm_i_pthread_mutex_lock (&thread_admin_mutex);
for (tp = &all_threads; *tp; tp = &(*tp)->next_thread)
if (*tp == t)
{
*tp = t->next_thread;
/* GC-robust */
t->next_thread = NULL;
break;
}
thread_count--;
/* Prevent any concurrent or future marker from visiting this
thread. */
t->handle = SCM_PACK (0);
/* If there's only one other thread, it could be the signal delivery
thread, so we need to notify it to shut down by closing its read pipe.
If it's not the signal delivery thread, then closing the read pipe isn't
going to hurt. */
if (thread_count <= 1)
scm_i_close_signal_pipe ();
scm_i_pthread_mutex_unlock (&thread_admin_mutex);
/* Although this thread has exited, the thread object might still be
alive. Release unused memory. */
for (size_t n = 0; n < SCM_INLINE_GC_FREELIST_COUNT; n++)
t->freelists[n] = t->pointerless_freelists[n] = NULL;
t->dynamic_state = NULL;
t->dynstack.base = NULL;
t->dynstack.top = NULL;
t->dynstack.limit = NULL;
scm_i_vm_free_stack (&t->vm);
#if ENABLE_JIT
scm_jit_state_free (t->jit_state);
#endif
t->jit_state = NULL;
#ifdef SCM_HAVE_THREAD_STORAGE_CLASS
scm_i_current_thread = NULL;
#endif
#if SCM_USE_PTHREAD_THREADS
if (t->needs_unregister)
GC_unregister_my_thread ();
#endif
}
static scm_i_pthread_once_t init_thread_key_once = SCM_I_PTHREAD_ONCE_INIT;
static void
init_thread_key (void)
{
scm_i_pthread_key_create (&scm_i_thread_key, on_thread_exit);
}
/* Perform any initializations necessary to make the current thread
known to Guile (via SCM_I_CURRENT_THREAD), initializing Guile itself,
if necessary.
BASE is the stack base to use with GC.
DYNAMIC_STATE is the set of fluid values to start with.
Returns zero when the thread was known to guile already; otherwise
return 1.
Note that it could be the case that the thread was known
to Guile, but not in guile mode (because we are within a
scm_without_guile call). Check SCM_I_CURRENT_THREAD->guile_mode to
be sure. New threads are put into guile mode implicitly. */
static int
scm_i_init_thread_for_guile (struct GC_stack_base *base,
SCM dynamic_state)
{
scm_i_pthread_once (&init_thread_key_once, init_thread_key);
if (SCM_I_CURRENT_THREAD)
{
/* Thread is already known to Guile.
*/
return 0;
}
else
{
/* This thread has not been guilified yet.
*/
scm_i_pthread_mutex_lock (&scm_i_init_mutex);
if (scm_initialized_p == 0)
{
/* First thread ever to enter Guile. Run the full
initialization.
*/
scm_i_init_guile (base);
#if SCM_USE_PTHREAD_THREADS
/* Allow other threads to come in later. */
GC_allow_register_threads ();
#endif
scm_i_pthread_mutex_unlock (&scm_i_init_mutex);
}
else
{
int needs_unregister = 0;
/* Guile is already initialized, but this thread enters it for
the first time. Only initialize this thread.
*/
scm_i_pthread_mutex_unlock (&scm_i_init_mutex);
/* Register this thread with libgc. */
#if SCM_USE_PTHREAD_THREADS
if (GC_register_my_thread (base) == GC_SUCCESS)
needs_unregister = 1;
#endif
guilify_self_1 (base, needs_unregister);
guilify_self_2 (dynamic_state);
}
return 1;
}
}
void
scm_init_guile ()
{
struct GC_stack_base stack_base;
if (GC_get_stack_base (&stack_base) == GC_SUCCESS)
scm_i_init_thread_for_guile (&stack_base, default_dynamic_state);
else
{
fprintf (stderr, "Failed to get stack base for current thread.\n");
exit (EXIT_FAILURE);
}
}
struct with_guile_args
{
GC_fn_type func;
void *data;
SCM dynamic_state;
};
static void *
with_guile_trampoline (void *data)
{
struct with_guile_args *args = data;
return scm_c_with_continuation_barrier (args->func, args->data);
}
static void *
with_guile (struct GC_stack_base *base, void *data)
{
void *res;
int new_thread;
scm_thread *t;
struct with_guile_args *args = data;
new_thread = scm_i_init_thread_for_guile (base, args->dynamic_state);
t = SCM_I_CURRENT_THREAD;
if (new_thread)
{
/* We are in Guile mode. */
assert (t->guile_mode);
res = scm_c_with_continuation_barrier (args->func, args->data);
/* Leave Guile mode. */
t->guile_mode = 0;
}
else if (t->guile_mode)
{
/* Already in Guile mode. */
res = scm_c_with_continuation_barrier (args->func, args->data);
}
else
{
/* We are not in Guile mode, either because we are not within a
scm_with_guile, or because we are within a scm_without_guile.
This call to scm_with_guile() could happen from anywhere on the
stack, and in particular lower on the stack than when it was
when this thread was first guilified. Thus, `base' must be
updated. */
#if SCM_STACK_GROWS_UP
if (SCM_STACK_PTR (base->mem_base) < t->base)
t->base = SCM_STACK_PTR (base->mem_base);
#else
if (SCM_STACK_PTR (base->mem_base) > t->base)
t->base = SCM_STACK_PTR (base->mem_base);
#endif
t->guile_mode = 1;
res = GC_call_with_gc_active (with_guile_trampoline, args);
t->guile_mode = 0;
}
return res;
}
static void *
scm_i_with_guile (void *(*func)(void *), void *data, SCM dynamic_state)
{
struct with_guile_args args;
args.func = func;
args.data = data;
args.dynamic_state = dynamic_state;
return GC_call_with_stack_base (with_guile, &args);
}
void *
scm_with_guile (void *(*func)(void *), void *data)
{
return scm_i_with_guile (func, data, default_dynamic_state);
}
void *
scm_without_guile (void *(*func)(void *), void *data)
{
void *result;
scm_thread *t = SCM_I_CURRENT_THREAD;
if (t->guile_mode)
{
SCM_I_CURRENT_THREAD->guile_mode = 0;
result = GC_do_blocking (func, data);
SCM_I_CURRENT_THREAD->guile_mode = 1;
}
else
/* Otherwise we're not in guile mode, so nothing to do. */
result = func (data);
return result;
}
/*** Thread creation */
/* Because (ice-9 boot-9) loads up (ice-9 threads), we know that this
variable will get loaded before a call to scm_call_with_new_thread
and therefore no lock or pthread_once_t is needed. */
static SCM call_with_new_thread_var;
SCM
scm_call_with_new_thread (SCM thunk, SCM handler)
{
SCM call_with_new_thread = scm_variable_ref (call_with_new_thread_var);
if (SCM_UNBNDP (handler))
return scm_call_1 (call_with_new_thread, thunk);
return scm_call_2 (call_with_new_thread, thunk, handler);
}
typedef struct launch_data launch_data;
struct launch_data {
launch_data *prev;
launch_data *next;
SCM dynamic_state;
SCM thunk;
};
/* GC-protect the launch data for new threads. */
static launch_data *protected_launch_data;
static scm_i_pthread_mutex_t protected_launch_data_lock =
SCM_I_PTHREAD_MUTEX_INITIALIZER;
static void
protect_launch_data (launch_data *data)
{
scm_i_pthread_mutex_lock (&protected_launch_data_lock);
data->next = protected_launch_data;
if (protected_launch_data)
protected_launch_data->prev = data;
protected_launch_data = data;
scm_i_pthread_mutex_unlock (&protected_launch_data_lock);
}
static void
unprotect_launch_data (launch_data *data)
{
scm_i_pthread_mutex_lock (&protected_launch_data_lock);
if (data->next)
data->next->prev = data->prev;
if (data->prev)
data->prev->next = data->next;
else
protected_launch_data = data->next;
scm_i_pthread_mutex_unlock (&protected_launch_data_lock);
}
static void *
really_launch (void *d)
{
scm_thread *t = SCM_I_CURRENT_THREAD;
unprotect_launch_data (d);
/* The thread starts with asyncs blocked. */
t->block_asyncs++;
SCM_I_CURRENT_THREAD->result = scm_call_0 (((launch_data *)d)->thunk);
return 0;
}
static void *
launch_thread (void *d)
{
launch_data *data = (launch_data *)d;
scm_i_pthread_detach (scm_i_pthread_self ());
scm_i_with_guile (really_launch, d, data->dynamic_state);
return NULL;
}
SCM_INTERNAL SCM scm_sys_call_with_new_thread (SCM);
SCM_DEFINE (scm_sys_call_with_new_thread, "%call-with-new-thread", 1, 0, 0,
(SCM thunk), "")
#define FUNC_NAME s_scm_sys_call_with_new_thread
{
launch_data *data;
scm_i_pthread_t id;
int err;
SCM_ASSERT (scm_is_true (scm_thunk_p (thunk)), thunk, SCM_ARG1, FUNC_NAME);
GC_collect_a_little ();
data = scm_gc_typed_calloc (launch_data);
data->dynamic_state = scm_current_dynamic_state ();
data->thunk = thunk;
protect_launch_data (data);
err = scm_i_pthread_create (&id, NULL, launch_thread, data);
if (err)
{
errno = err;
scm_syserror (NULL);
}
return SCM_UNSPECIFIED;
}
#undef FUNC_NAME
SCM
scm_spawn_thread (scm_t_catch_body body, void *body_data,
scm_t_catch_handler handler, void *handler_data)
{
SCM body_closure, handler_closure;
body_closure = scm_c_make_thunk (body, body_data);
handler_closure = handler == NULL ? SCM_UNDEFINED :
scm_i_make_catch_handler (handler, handler_data);
return scm_call_with_new_thread (body_closure, handler_closure);
}
SCM_DEFINE (scm_yield, "yield", 0, 0, 0,
(),
"Move the calling thread to the end of the scheduling queue.")
#define FUNC_NAME s_scm_yield
{
return scm_from_bool (scm_i_sched_yield ());
}
#undef FUNC_NAME
static SCM cancel_thread_var;
SCM
scm_cancel_thread (SCM thread)
{
scm_call_1 (scm_variable_ref (cancel_thread_var), thread);
return SCM_UNSPECIFIED;
}
static SCM join_thread_var;
SCM
scm_join_thread (SCM thread)
{
return scm_call_1 (scm_variable_ref (join_thread_var), thread);
}
SCM
scm_join_thread_timed (SCM thread, SCM timeout, SCM timeoutval)
{
SCM join_thread = scm_variable_ref (join_thread_var);
if (SCM_UNBNDP (timeout))
return scm_call_1 (join_thread, thread);
else if (SCM_UNBNDP (timeoutval))
return scm_call_2 (join_thread, thread, timeout);
else
return scm_call_3 (join_thread, thread, timeout, timeoutval);
}
SCM_DEFINE (scm_thread_p, "thread?", 1, 0, 0,
(SCM obj),
"Return @code{#t} if @var{obj} is a thread.")
#define FUNC_NAME s_scm_thread_p
{
return SCM_I_IS_THREAD(obj) ? SCM_BOOL_T : SCM_BOOL_F;
}
#undef FUNC_NAME
/* We implement our own mutex type since we want them to be 'fair', we
want to do fancy things while waiting for them (like running
asyncs) and we might want to add things that are nice for
debugging.
*/
enum scm_mutex_kind {
/* A standard mutex can only be locked once. If you try to lock it
again from the thread that locked it to begin with (the "owner"
thread), it throws an error. It can only be unlocked from the
thread that locked it in the first place. */
SCM_MUTEX_STANDARD,
/* A recursive mutex can be locked multiple times by its owner. It
then has to be unlocked the corresponding number of times, and like
standard mutexes can only be unlocked by the owner thread. */
SCM_MUTEX_RECURSIVE,
/* An unowned mutex is like a standard mutex, except that it can be
unlocked by any thread. A corrolary of this behavior is that a
thread's attempt to lock a mutex that it already owns will block
instead of signalling an error, as it could be that some other
thread unlocks the mutex, allowing the owner thread to proceed.
This kind of mutex is a bit strange and is here for use by
SRFI-18. */
SCM_MUTEX_UNOWNED
};
struct scm_mutex {
scm_i_pthread_mutex_t lock;
/* The thread that owns this mutex, or #f if the mutex is unlocked. */
SCM owner;
/* Queue of threads waiting for this mutex. */
SCM waiting;
/* For SCM_MUTEX_RECURSIVE (and only SCM_MUTEX_RECURSIVE), the
recursive lock count. The first lock does not count. */
int level;
};
#define SCM_MUTEXP(x) SCM_SMOB_PREDICATE (scm_tc16_mutex, x)
#define SCM_MUTEX_DATA(x) ((struct scm_mutex *) SCM_SMOB_DATA (x))
#define SCM_MUTEX_KIND(x) ((enum scm_mutex_kind) (SCM_SMOB_FLAGS (x) & 0x3))
static int
scm_mutex_print (SCM mx, SCM port, scm_print_state *pstate SCM_UNUSED)
{
struct scm_mutex *m = SCM_MUTEX_DATA (mx);
scm_puts ("#", port);
return 1;
}
SCM_SYMBOL (allow_external_unlock_sym, "allow-external-unlock");
SCM_SYMBOL (recursive_sym, "recursive");
SCM_DEFINE (scm_make_mutex_with_kind, "make-mutex", 0, 1, 0,
(SCM kind),
"Create a new mutex. If @var{kind} is not given, the mutex\n"
"will be a standard non-recursive mutex. Otherwise pass\n"
"@code{recursive} to make a recursive mutex, or\n"
"@code{allow-external-unlock} to make a non-recursive mutex\n"
"that can be unlocked from any thread.")
#define FUNC_NAME s_scm_make_mutex_with_kind
{
enum scm_mutex_kind mkind = SCM_MUTEX_STANDARD;
struct scm_mutex *m;
scm_i_pthread_mutex_t lock = SCM_I_PTHREAD_MUTEX_INITIALIZER;
if (!SCM_UNBNDP (kind))
{
if (scm_is_eq (kind, allow_external_unlock_sym))
mkind = SCM_MUTEX_UNOWNED;
else if (scm_is_eq (kind, recursive_sym))
mkind = SCM_MUTEX_RECURSIVE;
else
SCM_MISC_ERROR ("unsupported mutex kind: ~a", scm_list_1 (kind));
}
m = scm_gc_malloc (sizeof (struct scm_mutex), "mutex");
/* Because PTHREAD_MUTEX_INITIALIZER is static, it's plain old data,
and so we can just copy it. */
memcpy (&m->lock, &lock, sizeof (m->lock));
m->owner = SCM_BOOL_F;
m->level = 0;
m->waiting = make_queue ();
return scm_new_smob (scm_tc16_mutex | (mkind << 16), (scm_t_bits) m);
}
#undef FUNC_NAME
SCM
scm_make_mutex (void)
{
return scm_make_mutex_with_kind (SCM_UNDEFINED);
}
SCM_DEFINE (scm_make_recursive_mutex, "make-recursive-mutex", 0, 0, 0,
(void),
"Create a new recursive mutex. ")
#define FUNC_NAME s_scm_make_recursive_mutex
{
return scm_make_mutex_with_kind (recursive_sym);
}
#undef FUNC_NAME
SCM
scm_lock_mutex (SCM mx)
{
return scm_timed_lock_mutex (mx, SCM_UNDEFINED);
}
static inline SCM
lock_mutex (enum scm_mutex_kind kind, struct scm_mutex *m,
scm_thread *current_thread, scm_t_timespec *waittime)
#define FUNC_NAME "lock-mutex"
{
scm_i_scm_pthread_mutex_lock (&m->lock);
if (scm_is_eq (m->owner, SCM_BOOL_F))
{
m->owner = current_thread->handle;
scm_i_pthread_mutex_unlock (&m->lock);
return SCM_BOOL_T;
}
else if (kind == SCM_MUTEX_RECURSIVE &&
scm_is_eq (m->owner, current_thread->handle))
{
m->level++;
scm_i_pthread_mutex_unlock (&m->lock);
return SCM_BOOL_T;
}
else if (kind == SCM_MUTEX_STANDARD &&
scm_is_eq (m->owner, current_thread->handle))
{
scm_i_pthread_mutex_unlock (&m->lock);
SCM_MISC_ERROR ("mutex already locked by thread", SCM_EOL);
}
else
while (1)
{
int err = block_self (m->waiting, &m->lock, waittime);
if (err == 0)
{
goto maybe_acquire;
}
else if (err == ETIMEDOUT)
{
scm_i_pthread_mutex_unlock (&m->lock);
return SCM_BOOL_F;
}
else if (err == EINTR)
{
scm_i_pthread_mutex_unlock (&m->lock);
scm_async_tick ();
scm_i_scm_pthread_mutex_lock (&m->lock);
goto maybe_acquire;
}
else
{
/* Shouldn't happen. */
scm_i_pthread_mutex_unlock (&m->lock);
errno = err;
SCM_SYSERROR;
}
maybe_acquire:
if (scm_is_eq (m->owner, SCM_BOOL_F))
{
m->owner = current_thread->handle;
scm_i_pthread_mutex_unlock (&m->lock);
return SCM_BOOL_T;
}
}
}
#undef FUNC_NAME
SCM_DEFINE (scm_timed_lock_mutex, "lock-mutex", 1, 1, 0,
(SCM mutex, SCM timeout),
"Lock mutex @var{mutex}. If the mutex is already locked, "
"the calling thread blocks until the mutex becomes available.")
#define FUNC_NAME s_scm_timed_lock_mutex
{
scm_t_timespec cwaittime, *waittime = NULL;
struct scm_mutex *m;
scm_thread *t = SCM_I_CURRENT_THREAD;
SCM ret;
SCM_VALIDATE_MUTEX (1, mutex);
m = SCM_MUTEX_DATA (mutex);
if (! SCM_UNBNDP (timeout) && ! scm_is_false (timeout))
{
to_timespec (timeout, &cwaittime);
waittime = &cwaittime;
}
/* Specialized lock_mutex implementations according to the mutex
kind. */
switch (SCM_MUTEX_KIND (mutex))
{
case SCM_MUTEX_STANDARD:
ret = lock_mutex (SCM_MUTEX_STANDARD, m, t, waittime);
break;
case SCM_MUTEX_RECURSIVE:
ret = lock_mutex (SCM_MUTEX_RECURSIVE, m, t, waittime);
break;
case SCM_MUTEX_UNOWNED:
ret = lock_mutex (SCM_MUTEX_UNOWNED, m, t, waittime);
break;
default:
abort ();
}
scm_remember_upto_here_1 (mutex);
return ret;
}
#undef FUNC_NAME
static void
lock_mutex_return_void (SCM mx)
{
(void) scm_lock_mutex (mx);
}
static void
unlock_mutex_return_void (SCM mx)
{
(void) scm_unlock_mutex (mx);
}
void
scm_dynwind_lock_mutex (SCM mutex)
{
scm_dynwind_unwind_handler_with_scm (unlock_mutex_return_void, mutex,
SCM_F_WIND_EXPLICITLY);
scm_dynwind_rewind_handler_with_scm (lock_mutex_return_void, mutex,
SCM_F_WIND_EXPLICITLY);
}
SCM
scm_try_mutex (SCM mutex)
{
return scm_timed_lock_mutex (mutex, SCM_INUM0);
}
/* This function is static inline so that the compiler can specialize it
against the mutex kind. */
static inline void
unlock_mutex (enum scm_mutex_kind kind, struct scm_mutex *m,
scm_thread *current_thread)
#define FUNC_NAME "unlock-mutex"
{
scm_i_scm_pthread_mutex_lock (&m->lock);
if (!scm_is_eq (m->owner, current_thread->handle))
{
if (scm_is_eq (m->owner, SCM_BOOL_F))
{
scm_i_pthread_mutex_unlock (&m->lock);
SCM_MISC_ERROR ("mutex not locked", SCM_EOL);
}
if (kind != SCM_MUTEX_UNOWNED)
{
scm_i_pthread_mutex_unlock (&m->lock);
SCM_MISC_ERROR ("mutex not locked by current thread", SCM_EOL);
}
}
if (kind == SCM_MUTEX_RECURSIVE && m->level > 0)
m->level--;
else
{
m->owner = SCM_BOOL_F;
/* Wake up one waiter. */
unblock_from_queue (m->waiting);
}
scm_i_pthread_mutex_unlock (&m->lock);
}
#undef FUNC_NAME
SCM_DEFINE (scm_unlock_mutex, "unlock-mutex", 1, 0, 0, (SCM mutex),
"Unlocks @var{mutex}. The calling thread must already hold\n"
"the lock on @var{mutex}, unless the mutex was created with\n"
"the @code{allow-external-unlock} option; otherwise an error\n"
"will be signalled.")
#define FUNC_NAME s_scm_unlock_mutex
{
struct scm_mutex *m;
scm_thread *t = SCM_I_CURRENT_THREAD;
SCM_VALIDATE_MUTEX (1, mutex);
m = SCM_MUTEX_DATA (mutex);
/* Specialized unlock_mutex implementations according to the mutex
kind. */
switch (SCM_MUTEX_KIND (mutex))
{
case SCM_MUTEX_STANDARD:
unlock_mutex (SCM_MUTEX_STANDARD, m, t);
break;
case SCM_MUTEX_RECURSIVE:
unlock_mutex (SCM_MUTEX_RECURSIVE, m, t);
break;
case SCM_MUTEX_UNOWNED:
unlock_mutex (SCM_MUTEX_UNOWNED, m, t);
break;
default:
abort ();
}
scm_remember_upto_here_1 (mutex);
return SCM_BOOL_T;
}
#undef FUNC_NAME
SCM_DEFINE (scm_mutex_p, "mutex?", 1, 0, 0,
(SCM obj),
"Return @code{#t} if @var{obj} is a mutex.")
#define FUNC_NAME s_scm_mutex_p
{
return SCM_MUTEXP (obj) ? SCM_BOOL_T : SCM_BOOL_F;
}
#undef FUNC_NAME
SCM_DEFINE (scm_mutex_owner, "mutex-owner", 1, 0, 0,
(SCM mx),
"Return the thread owning @var{mx}, or @code{#f}.")
#define FUNC_NAME s_scm_mutex_owner
{
SCM owner;
struct scm_mutex *m = NULL;
SCM_VALIDATE_MUTEX (1, mx);
m = SCM_MUTEX_DATA (mx);
scm_i_pthread_mutex_lock (&m->lock);
owner = m->owner;
scm_i_pthread_mutex_unlock (&m->lock);
return owner;
}
#undef FUNC_NAME
SCM_DEFINE (scm_mutex_level, "mutex-level", 1, 0, 0,
(SCM mx),
"Return the lock level of mutex @var{mx}.")
#define FUNC_NAME s_scm_mutex_level
{
SCM_VALIDATE_MUTEX (1, mx);
if (SCM_MUTEX_KIND (mx) == SCM_MUTEX_RECURSIVE)
return scm_from_int (SCM_MUTEX_DATA (mx)->level + 1);
else if (scm_is_eq (SCM_MUTEX_DATA (mx)->owner, SCM_BOOL_F))
return SCM_INUM0;
else
return SCM_INUM1;
}
#undef FUNC_NAME
SCM_DEFINE (scm_mutex_locked_p, "mutex-locked?", 1, 0, 0,
(SCM mx),
"Returns @code{#t} if the mutex @var{mx} is locked.")
#define FUNC_NAME s_scm_mutex_locked_p
{
SCM_VALIDATE_MUTEX (1, mx);
if (scm_is_eq (SCM_MUTEX_DATA (mx)->owner, SCM_BOOL_F))
return SCM_BOOL_F;
else
return SCM_BOOL_T;
}
#undef FUNC_NAME
struct scm_cond {
scm_i_pthread_mutex_t lock;
SCM waiting; /* the threads waiting for this condition. */
};
#define SCM_CONDVARP(x) SCM_SMOB_PREDICATE (scm_tc16_condvar, x)
#define SCM_CONDVAR_DATA(x) ((struct scm_cond *) SCM_SMOB_DATA (x))
static int
scm_cond_print (SCM cv, SCM port, scm_print_state *pstate SCM_UNUSED)
{
struct scm_cond *c = SCM_CONDVAR_DATA (cv);
scm_puts ("#", port);
return 1;
}
SCM_DEFINE (scm_make_condition_variable, "make-condition-variable", 0, 0, 0,
(void),
"Make a new condition variable.")
#define FUNC_NAME s_scm_make_condition_variable
{
struct scm_cond *c;
SCM cv;
c = scm_gc_malloc (sizeof (struct scm_cond), "condition variable");
c->waiting = SCM_EOL;
SCM_NEWSMOB (cv, scm_tc16_condvar, (scm_t_bits) c);
c->waiting = make_queue ();
return cv;
}
#undef FUNC_NAME
static inline SCM
timed_wait (enum scm_mutex_kind kind, struct scm_mutex *m, struct scm_cond *c,
scm_thread *current_thread, scm_t_timespec *waittime)
#define FUNC_NAME "wait-condition-variable"
{
scm_i_scm_pthread_mutex_lock (&m->lock);
if (!scm_is_eq (m->owner, current_thread->handle))
{
if (scm_is_eq (m->owner, SCM_BOOL_F))
{
scm_i_pthread_mutex_unlock (&m->lock);
SCM_MISC_ERROR ("mutex not locked", SCM_EOL);
}
if (kind != SCM_MUTEX_UNOWNED)
{
scm_i_pthread_mutex_unlock (&m->lock);
SCM_MISC_ERROR ("mutex not locked by current thread", SCM_EOL);
}
}
while (1)
{
int err = 0;
/* Unlock the mutex. */
if (kind == SCM_MUTEX_RECURSIVE && m->level > 0)
m->level--;
else
{
m->owner = SCM_BOOL_F;
/* Wake up one waiter. */
unblock_from_queue (m->waiting);
}
/* Wait for someone to signal the cond, a timeout, or an
interrupt. */
err = block_self (c->waiting, &m->lock, waittime);
/* We woke up for some reason. Reacquire the mutex before doing
anything else.
FIXME: We disable interrupts while reacquiring the mutex. If
we allow interrupts here, there's the risk of a nonlocal exit
before we reaquire the mutex, which would be visible to user
code.
For example the unwind handler in
(with-mutex m (wait-condition-variable c m))
that tries to unlock M could see M in an already-unlocked
state, if an interrupt while waiting on C caused the wait to
abort and the woke thread lost the race to reacquire M. That's
not great. Maybe it's necessary but for now we just disable
interrupts while reaquiring a mutex after a wait. */
current_thread->block_asyncs++;
if (kind == SCM_MUTEX_RECURSIVE &&
scm_is_eq (m->owner, current_thread->handle))
{
m->level++;
scm_i_pthread_mutex_unlock (&m->lock);
}
else
while (1)
{
if (scm_is_eq (m->owner, SCM_BOOL_F))
{
m->owner = current_thread->handle;
scm_i_pthread_mutex_unlock (&m->lock);
break;
}
block_self (m->waiting, &m->lock, waittime);
}
current_thread->block_asyncs--;
/* Now that we have the mutex again, handle the return value. */
if (err == 0)
return SCM_BOOL_T;
else if (err == ETIMEDOUT)
return SCM_BOOL_F;
else if (err == EINTR)
/* Let caller run scm_async_tick() and loop. */
return SCM_BOOL_T;
else
{
/* Shouldn't happen. */
errno = err;
SCM_SYSERROR;
}
}
}
#undef FUNC_NAME
SCM_DEFINE (scm_timed_wait_condition_variable, "wait-condition-variable", 2, 1, 0,
(SCM cond, SCM mutex, SCM timeout),
"Wait until condition variable @var{cv} has been signalled. While waiting, "
"mutex @var{mx} is atomically unlocked (as with @code{unlock-mutex}) and "
"is locked again when this function returns. When @var{t} is given, "
"it specifies a point in time where the waiting should be aborted. It "
"can be either a integer as returned by @code{current-time} or a pair "
"as returned by @code{gettimeofday}. When the waiting is aborted the "
"mutex is locked and @code{#f} is returned. When the condition "
"variable is in fact signalled, the mutex is also locked and @code{#t} "
"is returned. ")
#define FUNC_NAME s_scm_timed_wait_condition_variable
{
scm_t_timespec waittime_val, *waittime = NULL;
struct scm_cond *c;
struct scm_mutex *m;
scm_thread *t = SCM_I_CURRENT_THREAD;
SCM ret;
SCM_VALIDATE_CONDVAR (1, cond);
SCM_VALIDATE_MUTEX (2, mutex);
c = SCM_CONDVAR_DATA (cond);
m = SCM_MUTEX_DATA (mutex);
if (!SCM_UNBNDP (timeout))
{
to_timespec (timeout, &waittime_val);
waittime = &waittime_val;
}
/* Specialized timed_wait implementations according to the mutex
kind. */
switch (SCM_MUTEX_KIND (mutex))
{
case SCM_MUTEX_STANDARD:
ret = timed_wait (SCM_MUTEX_STANDARD, m, c, t, waittime);
break;
case SCM_MUTEX_RECURSIVE:
ret = timed_wait (SCM_MUTEX_RECURSIVE, m, c, t, waittime);
break;
case SCM_MUTEX_UNOWNED:
ret = timed_wait (SCM_MUTEX_UNOWNED, m, c, t, waittime);
break;
default:
abort ();
}
scm_remember_upto_here_2 (mutex, cond);
return ret;
}
#undef FUNC_NAME
SCM_DEFINE (scm_signal_condition_variable, "signal-condition-variable", 1, 0, 0,
(SCM cv),
"Wake up one thread that is waiting for @var{cv}")
#define FUNC_NAME s_scm_signal_condition_variable
{
struct scm_cond *c;
SCM_VALIDATE_CONDVAR (1, cv);
c = SCM_CONDVAR_DATA (cv);
unblock_from_queue (c->waiting);
return SCM_BOOL_T;
}
#undef FUNC_NAME
SCM_DEFINE (scm_broadcast_condition_variable, "broadcast-condition-variable", 1, 0, 0,
(SCM cv),
"Wake up all threads that are waiting for @var{cv}. ")
#define FUNC_NAME s_scm_broadcast_condition_variable
{
struct scm_cond *c;
SCM_VALIDATE_CONDVAR (1, cv);
c = SCM_CONDVAR_DATA (cv);
while (scm_is_true (unblock_from_queue (c->waiting)))
;
return SCM_BOOL_T;
}
#undef FUNC_NAME
SCM_DEFINE (scm_condition_variable_p, "condition-variable?", 1, 0, 0,
(SCM obj),
"Return @code{#t} if @var{obj} is a condition variable.")
#define FUNC_NAME s_scm_condition_variable_p
{
return SCM_CONDVARP(obj) ? SCM_BOOL_T : SCM_BOOL_F;
}
#undef FUNC_NAME
/*** Select */
struct select_args
{
int nfds;
fd_set *read_fds;
fd_set *write_fds;
fd_set *except_fds;
struct timeval *timeout;
int result;
int errno_value;
};
static void *
do_std_select (void *args)
{
struct select_args *select_args;
select_args = (struct select_args *) args;
select_args->result =
select (select_args->nfds,
select_args->read_fds, select_args->write_fds,
select_args->except_fds, select_args->timeout);
select_args->errno_value = errno;
return NULL;
}
int
scm_std_select (int nfds,
fd_set *readfds,
fd_set *writefds,
fd_set *exceptfds,
struct timeval *timeout)
{
fd_set my_readfds;
int res, eno, wakeup_fd;
scm_thread *t = SCM_I_CURRENT_THREAD;
struct select_args args;
if (readfds == NULL)
{
FD_ZERO (&my_readfds);
readfds = &my_readfds;
}
if (scm_i_prepare_to_wait_on_fd (t, t->sleep_pipe[1]))
{
eno = EINTR;
res = -1;
}
else
{
wakeup_fd = t->sleep_pipe[0];
FD_SET (wakeup_fd, readfds);
if (wakeup_fd >= nfds)
nfds = wakeup_fd+1;
args.nfds = nfds;
args.read_fds = readfds;
args.write_fds = writefds;
args.except_fds = exceptfds;
args.timeout = timeout;
/* Explicitly cooperate with the GC. */
scm_without_guile (do_std_select, &args);
res = args.result;
eno = args.errno_value;
scm_i_wait_finished (t);
if (res > 0 && FD_ISSET (wakeup_fd, readfds))
{
char dummy;
full_read (wakeup_fd, &dummy, 1);
FD_CLR (wakeup_fd, readfds);
res -= 1;
if (res == 0)
{
eno = EINTR;
res = -1;
}
}
}
errno = eno;
return res;
}
/* Convenience API for blocking while in guile mode. */
#if SCM_USE_PTHREAD_THREADS
/* It seems reasonable to not run procedures related to mutex and condition
variables within `GC_do_blocking ()' since, (i) the GC can operate even
without it, and (ii) the only potential gain would be GC latency. See
http://thread.gmane.org/gmane.comp.programming.garbage-collection.boehmgc/2245/focus=2251
for a discussion of the pros and cons. */
int
scm_pthread_mutex_lock (scm_i_pthread_mutex_t *mutex)
{
int res = scm_i_pthread_mutex_lock (mutex);
return res;
}
static void
do_unlock (void *data)
{
scm_i_pthread_mutex_unlock ((scm_i_pthread_mutex_t *)data);
}
void
scm_dynwind_pthread_mutex_lock (scm_i_pthread_mutex_t *mutex)
{
scm_i_scm_pthread_mutex_lock (mutex);
scm_dynwind_unwind_handler (do_unlock, mutex, SCM_F_WIND_EXPLICITLY);
}
int
scm_pthread_cond_wait (scm_i_pthread_cond_t *cond, scm_i_pthread_mutex_t *mutex)
{
return scm_i_pthread_cond_wait (cond, mutex);
}
int
scm_pthread_cond_timedwait (scm_i_pthread_cond_t *cond,
scm_i_pthread_mutex_t *mutex,
const scm_t_timespec *wt)
{
return scm_i_pthread_cond_timedwait (cond, mutex, wt);
}
#endif
static void
do_unlock_with_asyncs (void *data)
{
scm_i_pthread_mutex_unlock ((scm_i_pthread_mutex_t *)data);
SCM_I_CURRENT_THREAD->block_asyncs--;
}
void
scm_i_dynwind_pthread_mutex_lock_block_asyncs (scm_i_pthread_mutex_t *mutex)
{
SCM_I_CURRENT_THREAD->block_asyncs++;
scm_i_scm_pthread_mutex_lock (mutex);
scm_dynwind_unwind_handler (do_unlock_with_asyncs, mutex,
SCM_F_WIND_EXPLICITLY);
}
unsigned long
scm_std_usleep (unsigned long usecs)
{
struct timeval tv;
tv.tv_usec = usecs % 1000000;
tv.tv_sec = usecs / 1000000;
scm_std_select (0, NULL, NULL, NULL, &tv);
return tv.tv_sec * 1000000 + tv.tv_usec;
}
unsigned int
scm_std_sleep (unsigned int secs)
{
struct timeval tv;
tv.tv_usec = 0;
tv.tv_sec = secs;
scm_std_select (0, NULL, NULL, NULL, &tv);
return tv.tv_sec;
}
/*** Misc */
SCM_DEFINE (scm_current_thread, "current-thread", 0, 0, 0,
(void),
"Return the thread that called this function.")
#define FUNC_NAME s_scm_current_thread
{
return SCM_I_CURRENT_THREAD->handle;
}
#undef FUNC_NAME
static SCM
scm_c_make_list (size_t n, SCM fill)
{
SCM res = SCM_EOL;
while (n-- > 0)
res = scm_cons (fill, res);
return res;
}
SCM_DEFINE (scm_all_threads, "all-threads", 0, 0, 0,
(void),
"Return a list of all threads.")
#define FUNC_NAME s_scm_all_threads
{
/* We can not allocate while holding the thread_admin_mutex because
of the way GC is done.
*/
int n = thread_count;
scm_thread *t;
SCM list = scm_c_make_list (n, SCM_UNSPECIFIED), *l;
scm_i_pthread_mutex_lock (&thread_admin_mutex);
l = &list;
for (t = all_threads; t && n > 0; t = t->next_thread)
{
if (t != scm_i_signal_delivery_thread)
{
SCM_SETCAR (*l, t->handle);
l = SCM_CDRLOC (*l);
}
n--;
}
*l = SCM_EOL;
scm_i_pthread_mutex_unlock (&thread_admin_mutex);
return list;
}
#undef FUNC_NAME
SCM_DEFINE (scm_thread_exited_p, "thread-exited?", 1, 0, 0,
(SCM thread),
"Return @code{#t} iff @var{thread} has exited.\n")
#define FUNC_NAME s_scm_thread_exited_p
{
return scm_from_bool (scm_c_thread_exited_p (thread));
}
#undef FUNC_NAME
int
scm_c_thread_exited_p (SCM thread)
#define FUNC_NAME s_scm_thread_exited_p
{
scm_thread *t;
SCM_VALIDATE_THREAD (1, thread);
t = SCM_I_THREAD_DATA (thread);
return t->exited;
}
#undef FUNC_NAME
SCM_DEFINE (scm_total_processor_count, "total-processor-count", 0, 0, 0,
(void),
"Return the total number of processors of the machine, which\n"
"is guaranteed to be at least 1. A ``processor'' here is a\n"
"thread execution unit, which can be either:\n\n"
"@itemize\n"
"@item an execution core in a (possibly multi-core) chip, in a\n"
" (possibly multi- chip) module, in a single computer, or\n"
"@item a thread execution unit inside a core in the case of\n"
" @dfn{hyper-threaded} CPUs.\n"
"@end itemize\n\n"
"Which of the two definitions is used, is unspecified.\n")
#define FUNC_NAME s_scm_total_processor_count
{
return scm_from_ulong (num_processors (NPROC_ALL));
}
#undef FUNC_NAME
SCM_DEFINE (scm_current_processor_count, "current-processor-count", 0, 0, 0,
(void),
"Like @code{total-processor-count}, but return the number of\n"
"processors available to the current process. See\n"
"@code{setaffinity} and @code{getaffinity} for more\n"
"information.\n")
#define FUNC_NAME s_scm_current_processor_count
{
return scm_from_ulong (num_processors (NPROC_CURRENT));
}
#undef FUNC_NAME
static scm_i_pthread_cond_t wake_up_cond;
static int threads_initialized_p = 0;
/*** Initialization */
scm_i_pthread_mutex_t scm_i_misc_mutex;
#if SCM_USE_PTHREAD_THREADS
pthread_mutexattr_t scm_i_pthread_mutexattr_recursive[1];
#endif
void
scm_threads_prehistory (void *base)
{
#if SCM_USE_PTHREAD_THREADS
pthread_mutexattr_init (scm_i_pthread_mutexattr_recursive);
pthread_mutexattr_settype (scm_i_pthread_mutexattr_recursive,
PTHREAD_MUTEX_RECURSIVE);
#endif
scm_i_pthread_mutex_init (&scm_i_misc_mutex, NULL);
scm_i_pthread_cond_init (&wake_up_cond, NULL);
thread_gc_kind =
GC_new_kind (GC_new_free_list (),
GC_MAKE_PROC (GC_new_proc (thread_mark), 0),
0, 1);
guilify_self_1 ((struct GC_stack_base *) base, 0);
}
scm_t_bits scm_tc16_thread;
scm_t_bits scm_tc16_mutex;
scm_t_bits scm_tc16_condvar;
static void
scm_init_ice_9_threads (void *unused)
{
#include "threads.x"
cancel_thread_var =
scm_module_variable (scm_current_module (),
scm_from_latin1_symbol ("cancel-thread"));
join_thread_var =
scm_module_variable (scm_current_module (),
scm_from_latin1_symbol ("join-thread"));
call_with_new_thread_var =
scm_module_variable (scm_current_module (),
scm_from_latin1_symbol ("call-with-new-thread"));
}
void
scm_init_threads ()
{
scm_tc16_thread = scm_make_smob_type ("thread", sizeof (scm_thread));
scm_set_smob_print (scm_tc16_thread, thread_print);
scm_tc16_mutex = scm_make_smob_type ("mutex", sizeof (struct scm_mutex));
scm_set_smob_print (scm_tc16_mutex, scm_mutex_print);
scm_tc16_condvar = scm_make_smob_type ("condition-variable",
sizeof (struct scm_cond));
scm_set_smob_print (scm_tc16_condvar, scm_cond_print);
default_dynamic_state = SCM_BOOL_F;
guilify_self_2 (scm_i_make_initial_dynamic_state ());
threads_initialized_p = 1;
scm_c_register_extension ("libguile-" SCM_EFFECTIVE_VERSION,
"scm_init_ice_9_threads",
scm_init_ice_9_threads, NULL);
}
void
scm_init_threads_default_dynamic_state ()
{
default_dynamic_state = scm_current_dynamic_state ();
}