#include "qemu-common.h"
#include "qemu/timer.h"
-#include "slirp/slirp.h"
+#include "qemu/sockets.h" // struct in_addr needed for libslirp.h
+#include "sysemu/qtest.h"
+#include "slirp/libslirp.h"
#include "qemu/main-loop.h"
#include "block/aio.h"
sigaddset(&set, SIGIO);
sigaddset(&set, SIGALRM);
sigaddset(&set, SIGBUS);
+ /* SIGINT cannot be handled via signalfd, so that ^C can be used
+ * to interrupt QEMU when it is being run under gdb. SIGHUP and
+ * SIGTERM are also handled asynchronously, even though it is not
+ * strictly necessary, because they use the same handler as SIGINT.
+ */
pthread_sigmask(SIG_BLOCK, &set, NULL);
sigdelset(&set, SIG_IPI);
fcntl_setfl(sigfd, O_NONBLOCK);
- qemu_set_fd_handler2(sigfd, NULL, sigfd_handler, NULL,
- (void *)(intptr_t)sigfd);
+ qemu_set_fd_handler(sigfd, sigfd_handler, NULL, (void *)(intptr_t)sigfd);
return 0;
}
#endif
static AioContext *qemu_aio_context;
+static QEMUBH *qemu_notify_bh;
+
+static void notify_event_cb(void *opaque)
+{
+ /* No need to do anything; this bottom half is only used to
+ * kick the kernel out of ppoll/poll/WaitForMultipleObjects.
+ */
+}
+
+AioContext *qemu_get_aio_context(void)
+{
+ return qemu_aio_context;
+}
void qemu_notify_event(void)
{
if (!qemu_aio_context) {
return;
}
- aio_notify(qemu_aio_context);
+ qemu_bh_schedule(qemu_notify_bh);
}
-int qemu_init_main_loop(void)
+static GArray *gpollfds;
+
+int qemu_init_main_loop(Error **errp)
{
int ret;
GSource *src;
+ Error *local_error = NULL;
init_clocks();
- if (init_timer_alarm() < 0) {
- fprintf(stderr, "could not initialize alarm timer\n");
- exit(1);
- }
ret = qemu_signal_init();
if (ret) {
return ret;
}
- qemu_aio_context = aio_context_new();
+ qemu_aio_context = aio_context_new(&local_error);
+ qemu_notify_bh = qemu_bh_new(notify_event_cb, NULL);
+ if (!qemu_aio_context) {
+ error_propagate(errp, local_error);
+ return -EMFILE;
+ }
+ gpollfds = g_array_new(FALSE, FALSE, sizeof(GPollFD));
src = aio_get_g_source(qemu_aio_context);
g_source_attach(src, NULL);
g_source_unref(src);
+ src = iohandler_get_g_source();
+ g_source_attach(src, NULL);
+ g_source_unref(src);
return 0;
}
-static fd_set rfds, wfds, xfds;
-static int nfds;
-static GPollFD poll_fds[1024 * 2]; /* this is probably overkill */
-static int n_poll_fds;
static int max_priority;
#ifndef _WIN32
-static void glib_select_fill(int *max_fd, fd_set *rfds, fd_set *wfds,
- fd_set *xfds, uint32_t *cur_timeout)
+static int glib_pollfds_idx;
+static int glib_n_poll_fds;
+
+static void glib_pollfds_fill(int64_t *cur_timeout)
{
GMainContext *context = g_main_context_default();
- int i;
int timeout = 0;
+ int64_t timeout_ns;
+ int n;
g_main_context_prepare(context, &max_priority);
- n_poll_fds = g_main_context_query(context, max_priority, &timeout,
- poll_fds, ARRAY_SIZE(poll_fds));
- g_assert(n_poll_fds <= ARRAY_SIZE(poll_fds));
-
- for (i = 0; i < n_poll_fds; i++) {
- GPollFD *p = &poll_fds[i];
-
- if ((p->events & G_IO_IN)) {
- FD_SET(p->fd, rfds);
- *max_fd = MAX(*max_fd, p->fd);
- }
- if ((p->events & G_IO_OUT)) {
- FD_SET(p->fd, wfds);
- *max_fd = MAX(*max_fd, p->fd);
- }
- if ((p->events & G_IO_ERR)) {
- FD_SET(p->fd, xfds);
- *max_fd = MAX(*max_fd, p->fd);
- }
+ glib_pollfds_idx = gpollfds->len;
+ n = glib_n_poll_fds;
+ do {
+ GPollFD *pfds;
+ glib_n_poll_fds = n;
+ g_array_set_size(gpollfds, glib_pollfds_idx + glib_n_poll_fds);
+ pfds = &g_array_index(gpollfds, GPollFD, glib_pollfds_idx);
+ n = g_main_context_query(context, max_priority, &timeout, pfds,
+ glib_n_poll_fds);
+ } while (n != glib_n_poll_fds);
+
+ if (timeout < 0) {
+ timeout_ns = -1;
+ } else {
+ timeout_ns = (int64_t)timeout * (int64_t)SCALE_MS;
}
- if (timeout >= 0 && timeout < *cur_timeout) {
- *cur_timeout = timeout;
- }
+ *cur_timeout = qemu_soonest_timeout(timeout_ns, *cur_timeout);
}
-static void glib_select_poll(fd_set *rfds, fd_set *wfds, fd_set *xfds,
- bool err)
+static void glib_pollfds_poll(void)
{
GMainContext *context = g_main_context_default();
+ GPollFD *pfds = &g_array_index(gpollfds, GPollFD, glib_pollfds_idx);
- if (!err) {
- int i;
-
- for (i = 0; i < n_poll_fds; i++) {
- GPollFD *p = &poll_fds[i];
-
- if ((p->events & G_IO_IN) && FD_ISSET(p->fd, rfds)) {
- p->revents |= G_IO_IN;
- }
- if ((p->events & G_IO_OUT) && FD_ISSET(p->fd, wfds)) {
- p->revents |= G_IO_OUT;
- }
- if ((p->events & G_IO_ERR) && FD_ISSET(p->fd, xfds)) {
- p->revents |= G_IO_ERR;
- }
- }
- }
-
- if (g_main_context_check(context, max_priority, poll_fds, n_poll_fds)) {
+ if (g_main_context_check(context, max_priority, pfds, glib_n_poll_fds)) {
g_main_context_dispatch(context);
}
}
-static int os_host_main_loop_wait(uint32_t timeout)
+#define MAX_MAIN_LOOP_SPIN (1000)
+
+static int os_host_main_loop_wait(int64_t timeout)
{
- struct timeval tv, *tvarg = NULL;
int ret;
+ static int spin_counter;
+
+ glib_pollfds_fill(&timeout);
- glib_select_fill(&nfds, &rfds, &wfds, &xfds, &timeout);
+ /* If the I/O thread is very busy or we are incorrectly busy waiting in
+ * the I/O thread, this can lead to starvation of the BQL such that the
+ * VCPU threads never run. To make sure we can detect the later case,
+ * print a message to the screen. If we run into this condition, create
+ * a fake timeout in order to give the VCPU threads a chance to run.
+ */
+ if (!timeout && (spin_counter > MAX_MAIN_LOOP_SPIN)) {
+ static bool notified;
+
+ if (!notified && !qtest_driver()) {
+ fprintf(stderr,
+ "main-loop: WARNING: I/O thread spun for %d iterations\n",
+ MAX_MAIN_LOOP_SPIN);
+ notified = true;
+ }
- if (timeout < UINT32_MAX) {
- tvarg = &tv;
- tv.tv_sec = timeout / 1000;
- tv.tv_usec = (timeout % 1000) * 1000;
+ timeout = SCALE_MS;
}
- if (timeout > 0) {
+ if (timeout) {
+ spin_counter = 0;
qemu_mutex_unlock_iothread();
+ } else {
+ spin_counter++;
}
- ret = select(nfds + 1, &rfds, &wfds, &xfds, tvarg);
+ ret = qemu_poll_ns((GPollFD *)gpollfds->data, gpollfds->len, timeout);
- if (timeout > 0) {
+ if (timeout) {
qemu_mutex_lock_iothread();
}
- glib_select_poll(&rfds, &wfds, &xfds, (ret < 0));
+ glib_pollfds_poll();
return ret;
}
#else
FD_CONNECT | FD_WRITE | FD_OOB);
}
-static int os_host_main_loop_wait(uint32_t timeout)
+static int pollfds_fill(GArray *pollfds, fd_set *rfds, fd_set *wfds,
+ fd_set *xfds)
+{
+ int nfds = -1;
+ int i;
+
+ for (i = 0; i < pollfds->len; i++) {
+ GPollFD *pfd = &g_array_index(pollfds, GPollFD, i);
+ int fd = pfd->fd;
+ int events = pfd->events;
+ if (events & G_IO_IN) {
+ FD_SET(fd, rfds);
+ nfds = MAX(nfds, fd);
+ }
+ if (events & G_IO_OUT) {
+ FD_SET(fd, wfds);
+ nfds = MAX(nfds, fd);
+ }
+ if (events & G_IO_PRI) {
+ FD_SET(fd, xfds);
+ nfds = MAX(nfds, fd);
+ }
+ }
+ return nfds;
+}
+
+static void pollfds_poll(GArray *pollfds, int nfds, fd_set *rfds,
+ fd_set *wfds, fd_set *xfds)
+{
+ int i;
+
+ for (i = 0; i < pollfds->len; i++) {
+ GPollFD *pfd = &g_array_index(pollfds, GPollFD, i);
+ int fd = pfd->fd;
+ int revents = 0;
+
+ if (FD_ISSET(fd, rfds)) {
+ revents |= G_IO_IN;
+ }
+ if (FD_ISSET(fd, wfds)) {
+ revents |= G_IO_OUT;
+ }
+ if (FD_ISSET(fd, xfds)) {
+ revents |= G_IO_PRI;
+ }
+ pfd->revents = revents & pfd->events;
+ }
+}
+
+static int os_host_main_loop_wait(int64_t timeout)
{
GMainContext *context = g_main_context_default();
- int select_ret, g_poll_ret, ret, i;
+ GPollFD poll_fds[1024 * 2]; /* this is probably overkill */
+ int select_ret = 0;
+ int g_poll_ret, ret, i, n_poll_fds;
PollingEntry *pe;
WaitObjects *w = &wait_objects;
gint poll_timeout;
+ int64_t poll_timeout_ns;
static struct timeval tv0;
+ fd_set rfds, wfds, xfds;
+ int nfds;
/* XXX: need to suppress polling by better using win32 events */
ret = 0;
return ret;
}
+ FD_ZERO(&rfds);
+ FD_ZERO(&wfds);
+ FD_ZERO(&xfds);
+ nfds = pollfds_fill(gpollfds, &rfds, &wfds, &xfds);
+ if (nfds >= 0) {
+ select_ret = select(nfds + 1, &rfds, &wfds, &xfds, &tv0);
+ if (select_ret != 0) {
+ timeout = 0;
+ }
+ if (select_ret > 0) {
+ pollfds_poll(gpollfds, nfds, &rfds, &wfds, &xfds);
+ }
+ }
+
g_main_context_prepare(context, &max_priority);
n_poll_fds = g_main_context_query(context, max_priority, &poll_timeout,
poll_fds, ARRAY_SIZE(poll_fds));
poll_fds[n_poll_fds + i].events = G_IO_IN;
}
- if (poll_timeout < 0 || timeout < poll_timeout) {
- poll_timeout = timeout;
+ if (poll_timeout < 0) {
+ poll_timeout_ns = -1;
+ } else {
+ poll_timeout_ns = (int64_t)poll_timeout * (int64_t)SCALE_MS;
}
+ poll_timeout_ns = qemu_soonest_timeout(poll_timeout_ns, timeout);
+
qemu_mutex_unlock_iothread();
- g_poll_ret = g_poll(poll_fds, n_poll_fds + w->num, poll_timeout);
+ g_poll_ret = qemu_poll_ns(poll_fds, n_poll_fds + w->num, poll_timeout_ns);
+
qemu_mutex_lock_iothread();
if (g_poll_ret > 0) {
for (i = 0; i < w->num; i++) {
g_main_context_dispatch(context);
}
- /* Call select after g_poll to avoid a useless iteration and therefore
- * improve socket latency.
- */
-
- if (nfds >= 0) {
- select_ret = select(nfds + 1, &rfds, &wfds, &xfds, &tv0);
- if (select_ret != 0) {
- timeout = 0;
- }
- }
-
return select_ret || g_poll_ret;
}
#endif
{
int ret;
uint32_t timeout = UINT32_MAX;
+ int64_t timeout_ns;
if (nonblocking) {
timeout = 0;
}
/* poll any events */
+ g_array_set_size(gpollfds, 0); /* reset for new iteration */
/* XXX: separate device handlers from system ones */
- nfds = -1;
- FD_ZERO(&rfds);
- FD_ZERO(&wfds);
- FD_ZERO(&xfds);
-
#ifdef CONFIG_SLIRP
- slirp_update_timeout(&timeout);
- slirp_select_fill(&nfds, &rfds, &wfds, &xfds);
+ slirp_pollfds_fill(gpollfds, &timeout);
#endif
- qemu_iohandler_fill(&nfds, &rfds, &wfds, &xfds);
- ret = os_host_main_loop_wait(timeout);
- qemu_iohandler_poll(&rfds, &wfds, &xfds, ret);
+
+ if (timeout == UINT32_MAX) {
+ timeout_ns = -1;
+ } else {
+ timeout_ns = (uint64_t)timeout * (int64_t)(SCALE_MS);
+ }
+
+ timeout_ns = qemu_soonest_timeout(timeout_ns,
+ timerlistgroup_deadline_ns(
+ &main_loop_tlg));
+
+ ret = os_host_main_loop_wait(timeout_ns);
#ifdef CONFIG_SLIRP
- slirp_select_poll(&rfds, &wfds, &xfds, (ret < 0));
+ slirp_pollfds_poll(gpollfds, (ret < 0));
#endif
- qemu_run_all_timers();
+ /* CPU thread can infinitely wait for event after
+ missing the warp */
+ qemu_clock_warp(QEMU_CLOCK_VIRTUAL);
+ qemu_clock_run_all_timers();
return ret;
}
{
return aio_bh_new(qemu_aio_context, cb, opaque);
}
-
-bool qemu_aio_wait(void)
-{
- return aio_poll(qemu_aio_context, true);
-}
-
-#ifdef CONFIG_POSIX
-void qemu_aio_set_fd_handler(int fd,
- IOHandler *io_read,
- IOHandler *io_write,
- AioFlushHandler *io_flush,
- void *opaque)
-{
- aio_set_fd_handler(qemu_aio_context, fd, io_read, io_write, io_flush,
- opaque);
-}
-#endif
-
-void qemu_aio_set_event_notifier(EventNotifier *notifier,
- EventNotifierHandler *io_read,
- AioFlushEventNotifierHandler *io_flush)
-{
- aio_set_event_notifier(qemu_aio_context, notifier, io_read, io_flush);
-}