Merge branch 'jh/simple-ipc'

A simple IPC interface gets introduced to build services like
fsmonitor on top.

* jh/simple-ipc:
  t0052: add simple-ipc tests and t/helper/test-simple-ipc tool
  simple-ipc: add Unix domain socket implementation
  unix-stream-server: create unix domain socket under lock
  unix-socket: disallow chdir() when creating unix domain sockets
  unix-socket: add backlog size option to unix_stream_listen()
  unix-socket: eliminate static unix_stream_socket() helper function
  simple-ipc: add win32 implementation
  simple-ipc: design documentation for new IPC mechanism
  pkt-line: add options argument to read_packetized_to_strbuf()
  pkt-line: add PACKET_READ_GENTLE_ON_READ_ERROR option
  pkt-line: do not issue flush packets in write_packetized_*()
  pkt-line: eliminate the need for static buffer in packet_write_gently()

3 files changed, 1778 insertions, 0 deletions

diff --git a/compat/simple-ipc/ipc-shared.c b/compat/simple-ipc/ipc-shared.c
new file mode 100644
index 0000000000..1edec81595
--- /dev/null
+++ b/compat/simple-ipc/ipc-shared.c
@@ -0,0 +1,28 @@
+#include "cache.h"
+#include "simple-ipc.h"
+#include "strbuf.h"
+#include "pkt-line.h"
+#include "thread-utils.h"
+
+#ifdef SUPPORTS_SIMPLE_IPC
+
+int ipc_server_run(const char *path, const struct ipc_server_opts *opts,
+		   ipc_server_application_cb *application_cb,
+		   void *application_data)
+{
+	struct ipc_server_data *server_data = NULL;
+	int ret;
+
+	ret = ipc_server_run_async(&server_data, path, opts,
+				   application_cb, application_data);
+	if (ret)
+		return ret;
+
+	ret = ipc_server_await(server_data);
+
+	ipc_server_free(server_data);
+
+	return ret;
+}
+
+#endif /* SUPPORTS_SIMPLE_IPC */
diff --git a/compat/simple-ipc/ipc-unix-socket.c b/compat/simple-ipc/ipc-unix-socket.c
new file mode 100644
index 0000000000..38689b278d
--- /dev/null
+++ b/compat/simple-ipc/ipc-unix-socket.c
@@ -0,0 +1,999 @@
+#include "cache.h"
+#include "simple-ipc.h"
+#include "strbuf.h"
+#include "pkt-line.h"
+#include "thread-utils.h"
+#include "unix-socket.h"
+#include "unix-stream-server.h"
+
+#ifdef NO_UNIX_SOCKETS
+#error compat/simple-ipc/ipc-unix-socket.c requires Unix sockets
+#endif
+
+enum ipc_active_state ipc_get_active_state(const char *path)
+{
+	enum ipc_active_state state = IPC_STATE__OTHER_ERROR;
+	struct ipc_client_connect_options options
+		= IPC_CLIENT_CONNECT_OPTIONS_INIT;
+	struct stat st;
+	struct ipc_client_connection *connection_test = NULL;
+
+	options.wait_if_busy = 0;
+	options.wait_if_not_found = 0;
+
+	if (lstat(path, &st) == -1) {
+		switch (errno) {
+		case ENOENT:
+		case ENOTDIR:
+			return IPC_STATE__NOT_LISTENING;
+		default:
+			return IPC_STATE__INVALID_PATH;
+		}
+	}
+
+	/* also complain if a plain file is in the way */
+	if ((st.st_mode & S_IFMT) != S_IFSOCK)
+		return IPC_STATE__INVALID_PATH;
+
+	/*
+	 * Just because the filesystem has a S_IFSOCK type inode
+	 * at `path`, doesn't mean it that there is a server listening.
+	 * Ping it to be sure.
+	 */
+	state = ipc_client_try_connect(path, &options, &connection_test);
+	ipc_client_close_connection(connection_test);
+
+	return state;
+}
+
+/*
+ * Retry frequency when trying to connect to a server.
+ *
+ * This value should be short enough that we don't seriously delay our
+ * caller, but not fast enough that our spinning puts pressure on the
+ * system.
+ */
+#define WAIT_STEP_MS (50)
+
+/*
+ * Try to connect to the server.  If the server is just starting up or
+ * is very busy, we may not get a connection the first time.
+ */
+static enum ipc_active_state connect_to_server(
+	const char *path,
+	int timeout_ms,
+	const struct ipc_client_connect_options *options,
+	int *pfd)
+{
+	int k;
+
+	*pfd = -1;
+
+	for (k = 0; k < timeout_ms; k += WAIT_STEP_MS) {
+		int fd = unix_stream_connect(path, options->uds_disallow_chdir);
+
+		if (fd != -1) {
+			*pfd = fd;
+			return IPC_STATE__LISTENING;
+		}
+
+		if (errno == ENOENT) {
+			if (!options->wait_if_not_found)
+				return IPC_STATE__PATH_NOT_FOUND;
+
+			goto sleep_and_try_again;
+		}
+
+		if (errno == ETIMEDOUT) {
+			if (!options->wait_if_busy)
+				return IPC_STATE__NOT_LISTENING;
+
+			goto sleep_and_try_again;
+		}
+
+		if (errno == ECONNREFUSED) {
+			if (!options->wait_if_busy)
+				return IPC_STATE__NOT_LISTENING;
+
+			goto sleep_and_try_again;
+		}
+
+		return IPC_STATE__OTHER_ERROR;
+
+	sleep_and_try_again:
+		sleep_millisec(WAIT_STEP_MS);
+	}
+
+	return IPC_STATE__NOT_LISTENING;
+}
+
+/*
+ * The total amount of time that we are willing to wait when trying to
+ * connect to a server.
+ *
+ * When the server is first started, it might take a little while for
+ * it to become ready to service requests.  Likewise, the server may
+ * be very (temporarily) busy and not respond to our connections.
+ *
+ * We should gracefully and silently handle those conditions and try
+ * again for a reasonable time period.
+ *
+ * The value chosen here should be long enough for the server
+ * to reliably heal from the above conditions.
+ */
+#define MY_CONNECTION_TIMEOUT_MS (1000)
+
+enum ipc_active_state ipc_client_try_connect(
+	const char *path,
+	const struct ipc_client_connect_options *options,
+	struct ipc_client_connection **p_connection)
+{
+	enum ipc_active_state state = IPC_STATE__OTHER_ERROR;
+	int fd = -1;
+
+	*p_connection = NULL;
+
+	trace2_region_enter("ipc-client", "try-connect", NULL);
+	trace2_data_string("ipc-client", NULL, "try-connect/path", path);
+
+	state = connect_to_server(path, MY_CONNECTION_TIMEOUT_MS,
+				  options, &fd);
+
+	trace2_data_intmax("ipc-client", NULL, "try-connect/state",
+			   (intmax_t)state);
+	trace2_region_leave("ipc-client", "try-connect", NULL);
+
+	if (state == IPC_STATE__LISTENING) {
+		(*p_connection) = xcalloc(1, sizeof(struct ipc_client_connection));
+		(*p_connection)->fd = fd;
+	}
+
+	return state;
+}
+
+void ipc_client_close_connection(struct ipc_client_connection *connection)
+{
+	if (!connection)
+		return;
+
+	if (connection->fd != -1)
+		close(connection->fd);
+
+	free(connection);
+}
+
+int ipc_client_send_command_to_connection(
+	struct ipc_client_connection *connection,
+	const char *message, struct strbuf *answer)
+{
+	int ret = 0;
+
+	strbuf_setlen(answer, 0);
+
+	trace2_region_enter("ipc-client", "send-command", NULL);
+
+	if (write_packetized_from_buf_no_flush(message, strlen(message),
+					       connection->fd) < 0 ||
+	    packet_flush_gently(connection->fd) < 0) {
+		ret = error(_("could not send IPC command"));
+		goto done;
+	}
+
+	if (read_packetized_to_strbuf(
+		    connection->fd, answer,
+		    PACKET_READ_GENTLE_ON_EOF | PACKET_READ_GENTLE_ON_READ_ERROR) < 0) {
+		ret = error(_("could not read IPC response"));
+		goto done;
+	}
+
+done:
+	trace2_region_leave("ipc-client", "send-command", NULL);
+	return ret;
+}
+
+int ipc_client_send_command(const char *path,
+			    const struct ipc_client_connect_options *options,
+			    const char *message, struct strbuf *answer)
+{
+	int ret = -1;
+	enum ipc_active_state state;
+	struct ipc_client_connection *connection = NULL;
+
+	state = ipc_client_try_connect(path, options, &connection);
+
+	if (state != IPC_STATE__LISTENING)
+		return ret;
+
+	ret = ipc_client_send_command_to_connection(connection, message, answer);
+
+	ipc_client_close_connection(connection);
+
+	return ret;
+}
+
+static int set_socket_blocking_flag(int fd, int make_nonblocking)
+{
+	int flags;
+
+	flags = fcntl(fd, F_GETFL, NULL);
+
+	if (flags < 0)
+		return -1;
+
+	if (make_nonblocking)
+		flags |= O_NONBLOCK;
+	else
+		flags &= ~O_NONBLOCK;
+
+	return fcntl(fd, F_SETFL, flags);
+}
+
+/*
+ * Magic numbers used to annotate callback instance data.
+ * These are used to help guard against accidentally passing the
+ * wrong instance data across multiple levels of callbacks (which
+ * is easy to do if there are `void*` arguments).
+ */
+enum magic {
+	MAGIC_SERVER_REPLY_DATA,
+	MAGIC_WORKER_THREAD_DATA,
+	MAGIC_ACCEPT_THREAD_DATA,
+	MAGIC_SERVER_DATA,
+};
+
+struct ipc_server_reply_data {
+	enum magic magic;
+	int fd;
+	struct ipc_worker_thread_data *worker_thread_data;
+};
+
+struct ipc_worker_thread_data {
+	enum magic magic;
+	struct ipc_worker_thread_data *next_thread;
+	struct ipc_server_data *server_data;
+	pthread_t pthread_id;
+};
+
+struct ipc_accept_thread_data {
+	enum magic magic;
+	struct ipc_server_data *server_data;
+
+	struct unix_ss_socket *server_socket;
+
+	int fd_send_shutdown;
+	int fd_wait_shutdown;
+	pthread_t pthread_id;
+};
+
+/*
+ * With unix-sockets, the conceptual "ipc-server" is implemented as a single
+ * controller "accept-thread" thread and a pool of "worker-thread" threads.
+ * The former does the usual `accept()` loop and dispatches connections
+ * to an idle worker thread.  The worker threads wait in an idle loop for
+ * a new connection, communicate with the client and relay data to/from
+ * the `application_cb` and then wait for another connection from the
+ * server thread.  This avoids the overhead of constantly creating and
+ * destroying threads.
+ */
+struct ipc_server_data {
+	enum magic magic;
+	ipc_server_application_cb *application_cb;
+	void *application_data;
+	struct strbuf buf_path;
+
+	struct ipc_accept_thread_data *accept_thread;
+	struct ipc_worker_thread_data *worker_thread_list;
+
+	pthread_mutex_t work_available_mutex;
+	pthread_cond_t work_available_cond;
+
+	/*
+	 * Accepted but not yet processed client connections are kept
+	 * in a circular buffer FIFO.  The queue is empty when the
+	 * positions are equal.
+	 */
+	int *fifo_fds;
+	int queue_size;
+	int back_pos;
+	int front_pos;
+
+	int shutdown_requested;
+	int is_stopped;
+};
+
+/*
+ * Remove and return the oldest queued connection.
+ *
+ * Returns -1 if empty.
+ */
+static int fifo_dequeue(struct ipc_server_data *server_data)
+{
+	/* ASSERT holding mutex */
+
+	int fd;
+
+	if (server_data->back_pos == server_data->front_pos)
+		return -1;
+
+	fd = server_data->fifo_fds[server_data->front_pos];
+	server_data->fifo_fds[server_data->front_pos] = -1;
+
+	server_data->front_pos++;
+	if (server_data->front_pos == server_data->queue_size)
+		server_data->front_pos = 0;
+
+	return fd;
+}
+
+/*
+ * Push a new fd onto the back of the queue.
+ *
+ * Drop it and return -1 if queue is already full.
+ */
+static int fifo_enqueue(struct ipc_server_data *server_data, int fd)
+{
+	/* ASSERT holding mutex */
+
+	int next_back_pos;
+
+	next_back_pos = server_data->back_pos + 1;
+	if (next_back_pos == server_data->queue_size)
+		next_back_pos = 0;
+
+	if (next_back_pos == server_data->front_pos) {
+		/* Queue is full. Just drop it. */
+		close(fd);
+		return -1;
+	}
+
+	server_data->fifo_fds[server_data->back_pos] = fd;
+	server_data->back_pos = next_back_pos;
+
+	return fd;
+}
+
+/*
+ * Wait for a connection to be queued to the FIFO and return it.
+ *
+ * Returns -1 if someone has already requested a shutdown.
+ */
+static int worker_thread__wait_for_connection(
+	struct ipc_worker_thread_data *worker_thread_data)
+{
+	/* ASSERT NOT holding mutex */
+
+	struct ipc_server_data *server_data = worker_thread_data->server_data;
+	int fd = -1;
+
+	pthread_mutex_lock(&server_data->work_available_mutex);
+	for (;;) {
+		if (server_data->shutdown_requested)
+			break;
+
+		fd = fifo_dequeue(server_data);
+		if (fd >= 0)
+			break;
+
+		pthread_cond_wait(&server_data->work_available_cond,
+				  &server_data->work_available_mutex);
+	}
+	pthread_mutex_unlock(&server_data->work_available_mutex);
+
+	return fd;
+}
+
+/*
+ * Forward declare our reply callback function so that any compiler
+ * errors are reported when we actually define the function (in addition
+ * to any errors reported when we try to pass this callback function as
+ * a parameter in a function call).  The former are easier to understand.
+ */
+static ipc_server_reply_cb do_io_reply_callback;
+
+/*
+ * Relay application's response message to the client process.
+ * (We do not flush at this point because we allow the caller
+ * to chunk data to the client thru us.)
+ */
+static int do_io_reply_callback(struct ipc_server_reply_data *reply_data,
+		       const char *response, size_t response_len)
+{
+	if (reply_data->magic != MAGIC_SERVER_REPLY_DATA)
+		BUG("reply_cb called with wrong instance data");
+
+	return write_packetized_from_buf_no_flush(response, response_len,
+						  reply_data->fd);
+}
+
+/* A randomly chosen value. */
+#define MY_WAIT_POLL_TIMEOUT_MS (10)
+
+/*
+ * If the client hangs up without sending any data on the wire, just
+ * quietly close the socket and ignore this client.
+ *
+ * This worker thread is committed to reading the IPC request data
+ * from the client at the other end of this fd.  Wait here for the
+ * client to actually put something on the wire -- because if the
+ * client just does a ping (connect and hangup without sending any
+ * data), our use of the pkt-line read routines will spew an error
+ * message.
+ *
+ * Return -1 if the client hung up.
+ * Return 0 if data (possibly incomplete) is ready.
+ */
+static int worker_thread__wait_for_io_start(
+	struct ipc_worker_thread_data *worker_thread_data,
+	int fd)
+{
+	struct ipc_server_data *server_data = worker_thread_data->server_data;
+	struct pollfd pollfd[1];
+	int result;
+
+	for (;;) {
+		pollfd[0].fd = fd;
+		pollfd[0].events = POLLIN;
+
+		result = poll(pollfd, 1, MY_WAIT_POLL_TIMEOUT_MS);
+		if (result < 0) {
+			if (errno == EINTR)
+				continue;
+			goto cleanup;
+		}
+
+		if (result == 0) {
+			/* a timeout */
+
+			int in_shutdown;
+
+			pthread_mutex_lock(&server_data->work_available_mutex);
+			in_shutdown = server_data->shutdown_requested;
+			pthread_mutex_unlock(&server_data->work_available_mutex);
+
+			/*
+			 * If a shutdown is already in progress and this
+			 * client has not started talking yet, just drop it.
+			 */
+			if (in_shutdown)
+				goto cleanup;
+			continue;
+		}
+
+		if (pollfd[0].revents & POLLHUP)
+			goto cleanup;
+
+		if (pollfd[0].revents & POLLIN)
+			return 0;
+
+		goto cleanup;
+	}
+
+cleanup:
+	close(fd);
+	return -1;
+}
+
+/*
+ * Receive the request/command from the client and pass it to the
+ * registered request-callback.  The request-callback will compose
+ * a response and call our reply-callback to send it to the client.
+ */
+static int worker_thread__do_io(
+	struct ipc_worker_thread_data *worker_thread_data,
+	int fd)
+{
+	/* ASSERT NOT holding lock */
+
+	struct strbuf buf = STRBUF_INIT;
+	struct ipc_server_reply_data reply_data;
+	int ret = 0;
+
+	reply_data.magic = MAGIC_SERVER_REPLY_DATA;
+	reply_data.worker_thread_data = worker_thread_data;
+
+	reply_data.fd = fd;
+
+	ret = read_packetized_to_strbuf(
+		reply_data.fd, &buf,
+		PACKET_READ_GENTLE_ON_EOF | PACKET_READ_GENTLE_ON_READ_ERROR);
+	if (ret >= 0) {
+		ret = worker_thread_data->server_data->application_cb(
+			worker_thread_data->server_data->application_data,
+			buf.buf, do_io_reply_callback, &reply_data);
+
+		packet_flush_gently(reply_data.fd);
+	}
+	else {
+		/*
+		 * The client probably disconnected/shutdown before it
+		 * could send a well-formed message.  Ignore it.
+		 */
+	}
+
+	strbuf_release(&buf);
+	close(reply_data.fd);
+
+	return ret;
+}
+
+/*
+ * Block SIGPIPE on the current thread (so that we get EPIPE from
+ * write() rather than an actual signal).
+ *
+ * Note that using sigchain_push() and _pop() to control SIGPIPE
+ * around our IO calls is not thread safe:
+ * [] It uses a global stack of handler frames.
+ * [] It uses ALLOC_GROW() to resize it.
+ * [] Finally, according to the `signal(2)` man-page:
+ *    "The effects of `signal()` in a multithreaded process are unspecified."
+ */
+static void thread_block_sigpipe(sigset_t *old_set)
+{
+	sigset_t new_set;
+
+	sigemptyset(&new_set);
+	sigaddset(&new_set, SIGPIPE);
+
+	sigemptyset(old_set);
+	pthread_sigmask(SIG_BLOCK, &new_set, old_set);
+}
+
+/*
+ * Thread proc for an IPC worker thread.  It handles a series of
+ * connections from clients.  It pulls the next fd from the queue
+ * processes it, and then waits for the next client.
+ *
+ * Block SIGPIPE in this worker thread for the life of the thread.
+ * This avoids stray (and sometimes delayed) SIGPIPE signals caused
+ * by client errors and/or when we are under extremely heavy IO load.
+ *
+ * This means that the application callback will have SIGPIPE blocked.
+ * The callback should not change it.
+ */
+static void *worker_thread_proc(void *_worker_thread_data)
+{
+	struct ipc_worker_thread_data *worker_thread_data = _worker_thread_data;
+	struct ipc_server_data *server_data = worker_thread_data->server_data;
+	sigset_t old_set;
+	int fd, io;
+	int ret;
+
+	trace2_thread_start("ipc-worker");
+
+	thread_block_sigpipe(&old_set);
+
+	for (;;) {
+		fd = worker_thread__wait_for_connection(worker_thread_data);
+		if (fd == -1)
+			break; /* in shutdown */
+
+		io = worker_thread__wait_for_io_start(worker_thread_data, fd);
+		if (io == -1)
+			continue; /* client hung up without sending anything */
+
+		ret = worker_thread__do_io(worker_thread_data, fd);
+
+		if (ret == SIMPLE_IPC_QUIT) {
+			trace2_data_string("ipc-worker", NULL, "queue_stop_async",
+					   "application_quit");
+			/*
+			 * The application layer is telling the ipc-server
+			 * layer to shutdown.
+			 *
+			 * We DO NOT have a response to send to the client.
+			 *
+			 * Queue an async stop (to stop the other threads) and
+			 * allow this worker thread to exit now (no sense waiting
+			 * for the thread-pool shutdown signal).
+			 *
+			 * Other non-idle worker threads are allowed to finish
+			 * responding to their current clients.
+			 */
+			ipc_server_stop_async(server_data);
+			break;
+		}
+	}
+
+	trace2_thread_exit();
+	return NULL;
+}
+
+/* A randomly chosen value. */
+#define MY_ACCEPT_POLL_TIMEOUT_MS (60 * 1000)
+
+/*
+ * Accept a new client connection on our socket.  This uses non-blocking
+ * IO so that we can also wait for shutdown requests on our socket-pair
+ * without actually spinning on a fast timeout.
+ */
+static int accept_thread__wait_for_connection(
+	struct ipc_accept_thread_data *accept_thread_data)
+{
+	struct pollfd pollfd[2];
+	int result;
+
+	for (;;) {
+		pollfd[0].fd = accept_thread_data->fd_wait_shutdown;
+		pollfd[0].events = POLLIN;
+
+		pollfd[1].fd = accept_thread_data->server_socket->fd_socket;
+		pollfd[1].events = POLLIN;
+
+		result = poll(pollfd, 2, MY_ACCEPT_POLL_TIMEOUT_MS);
+		if (result < 0) {
+			if (errno == EINTR)
+				continue;
+			return result;
+		}
+
+		if (result == 0) {
+			/* a timeout */
+
+			/*
+			 * If someone deletes or force-creates a new unix
+			 * domain socket at our path, all future clients
+			 * will be routed elsewhere and we silently starve.
+			 * If that happens, just queue a shutdown.
+			 */
+			if (unix_ss_was_stolen(
+				    accept_thread_data->server_socket)) {
+				trace2_data_string("ipc-accept", NULL,
+						   "queue_stop_async",
+						   "socket_stolen");
+				ipc_server_stop_async(
+					accept_thread_data->server_data);
+			}
+			continue;
+		}
+
+		if (pollfd[0].revents & POLLIN) {
+			/* shutdown message queued to socketpair */
+			return -1;
+		}
+
+		if (pollfd[1].revents & POLLIN) {
+			/* a connection is available on server_socket */
+
+			int client_fd =
+				accept(accept_thread_data->server_socket->fd_socket,
+				       NULL, NULL);
+			if (client_fd >= 0)
+				return client_fd;
+
+			/*
+			 * An error here is unlikely -- it probably
+			 * indicates that the connecting process has
+			 * already dropped the connection.
+			 */
+			continue;
+		}
+
+		BUG("unandled poll result errno=%d r[0]=%d r[1]=%d",
+		    errno, pollfd[0].revents, pollfd[1].revents);
+	}
+}
+
+/*
+ * Thread proc for the IPC server "accept thread".  This waits for
+ * an incoming socket connection, appends it to the queue of available
+ * connections, and notifies a worker thread to process it.
+ *
+ * Block SIGPIPE in this thread for the life of the thread.  This
+ * avoids any stray SIGPIPE signals when closing pipe fds under
+ * extremely heavy loads (such as when the fifo queue is full and we
+ * drop incomming connections).
+ */
+static void *accept_thread_proc(void *_accept_thread_data)
+{
+	struct ipc_accept_thread_data *accept_thread_data = _accept_thread_data;
+	struct ipc_server_data *server_data = accept_thread_data->server_data;
+	sigset_t old_set;
+
+	trace2_thread_start("ipc-accept");
+
+	thread_block_sigpipe(&old_set);
+
+	for (;;) {
+		int client_fd = accept_thread__wait_for_connection(
+			accept_thread_data);
+
+		pthread_mutex_lock(&server_data->work_available_mutex);
+		if (server_data->shutdown_requested) {
+			pthread_mutex_unlock(&server_data->work_available_mutex);
+			if (client_fd >= 0)
+				close(client_fd);
+			break;
+		}
+
+		if (client_fd < 0) {
+			/* ignore transient accept() errors */
+		}
+		else {
+			fifo_enqueue(server_data, client_fd);
+			pthread_cond_broadcast(&server_data->work_available_cond);
+		}
+		pthread_mutex_unlock(&server_data->work_available_mutex);
+	}
+
+	trace2_thread_exit();
+	return NULL;
+}
+
+/*
+ * We can't predict the connection arrival rate relative to the worker
+ * processing rate, therefore we allow the "accept-thread" to queue up
+ * a generous number of connections, since we'd rather have the client
+ * not unnecessarily timeout if we can avoid it.  (The assumption is
+ * that this will be used for FSMonitor and a few second wait on a
+ * connection is better than having the client timeout and do the full
+ * computation itself.)
+ *
+ * The FIFO queue size is set to a multiple of the worker pool size.
+ * This value chosen at random.
+ */
+#define FIFO_SCALE (100)
+
+/*
+ * The backlog value for `listen(2)`.  This doesn't need to huge,
+ * rather just large enough for our "accept-thread" to wake up and
+ * queue incoming connections onto the FIFO without the kernel
+ * dropping any.
+ *
+ * This value chosen at random.
+ */
+#define LISTEN_BACKLOG (50)
+
+static int create_listener_socket(
+	const char *path,
+	const struct ipc_server_opts *ipc_opts,
+	struct unix_ss_socket **new_server_socket)
+{
+	struct unix_ss_socket *server_socket = NULL;
+	struct unix_stream_listen_opts uslg_opts = UNIX_STREAM_LISTEN_OPTS_INIT;
+	int ret;
+
+	uslg_opts.listen_backlog_size = LISTEN_BACKLOG;
+	uslg_opts.disallow_chdir = ipc_opts->uds_disallow_chdir;
+
+	ret = unix_ss_create(path, &uslg_opts, -1, &server_socket);
+	if (ret)
+		return ret;
+
+	if (set_socket_blocking_flag(server_socket->fd_socket, 1)) {
+		int saved_errno = errno;
+		unix_ss_free(server_socket);
+		errno = saved_errno;
+		return -1;
+	}
+
+	*new_server_socket = server_socket;
+
+	trace2_data_string("ipc-server", NULL, "listen-with-lock", path);
+	return 0;
+}
+
+static int setup_listener_socket(
+	const char *path,
+	const struct ipc_server_opts *ipc_opts,
+	struct unix_ss_socket **new_server_socket)
+{
+	int ret, saved_errno;
+
+	trace2_region_enter("ipc-server", "create-listener_socket", NULL);
+
+	ret = create_listener_socket(path, ipc_opts, new_server_socket);
+
+	saved_errno = errno;
+	trace2_region_leave("ipc-server", "create-listener_socket", NULL);
+	errno = saved_errno;
+
+	return ret;
+}
+
+/*
+ * Start IPC server in a pool of background threads.
+ */
+int ipc_server_run_async(struct ipc_server_data **returned_server_data,
+			 const char *path, const struct ipc_server_opts *opts,
+			 ipc_server_application_cb *application_cb,
+			 void *application_data)
+{
+	struct unix_ss_socket *server_socket = NULL;
+	struct ipc_server_data *server_data;
+	int sv[2];
+	int k;
+	int ret;
+	int nr_threads = opts->nr_threads;
+
+	*returned_server_data = NULL;
+
+	/*
+	 * Create a socketpair and set sv[1] to non-blocking.  This
+	 * will used to send a shutdown message to the accept-thread
+	 * and allows the accept-thread to wait on EITHER a client
+	 * connection or a shutdown request without spinning.
+	 */
+	if (socketpair(AF_UNIX, SOCK_STREAM, 0, sv) < 0)
+		return -1;
+
+	if (set_socket_blocking_flag(sv[1], 1)) {
+		int saved_errno = errno;
+		close(sv[0]);
+		close(sv[1]);
+		errno = saved_errno;
+		return -1;
+	}
+
+	ret = setup_listener_socket(path, opts, &server_socket);
+	if (ret) {
+		int saved_errno = errno;
+		close(sv[0]);
+		close(sv[1]);
+		errno = saved_errno;
+		return ret;
+	}
+
+	server_data = xcalloc(1, sizeof(*server_data));
+	server_data->magic = MAGIC_SERVER_DATA;
+	server_data->application_cb = application_cb;
+	server_data->application_data = application_data;
+	strbuf_init(&server_data->buf_path, 0);
+	strbuf_addstr(&server_data->buf_path, path);
+
+	if (nr_threads < 1)
+		nr_threads = 1;
+
+	pthread_mutex_init(&server_data->work_available_mutex, NULL);
+	pthread_cond_init(&server_data->work_available_cond, NULL);
+
+	server_data->queue_size = nr_threads * FIFO_SCALE;
+	CALLOC_ARRAY(server_data->fifo_fds, server_data->queue_size);
+
+	server_data->accept_thread =
+		xcalloc(1, sizeof(*server_data->accept_thread));
+	server_data->accept_thread->magic = MAGIC_ACCEPT_THREAD_DATA;
+	server_data->accept_thread->server_data = server_data;
+	server_data->accept_thread->server_socket = server_socket;
+	server_data->accept_thread->fd_send_shutdown = sv[0];
+	server_data->accept_thread->fd_wait_shutdown = sv[1];
+
+	if (pthread_create(&server_data->accept_thread->pthread_id, NULL,
+			   accept_thread_proc, server_data->accept_thread))
+		die_errno(_("could not start accept_thread '%s'"), path);
+
+	for (k = 0; k < nr_threads; k++) {
+		struct ipc_worker_thread_data *wtd;
+
+		wtd = xcalloc(1, sizeof(*wtd));
+		wtd->magic = MAGIC_WORKER_THREAD_DATA;
+		wtd->server_data = server_data;
+
+		if (pthread_create(&wtd->pthread_id, NULL, worker_thread_proc,
+				   wtd)) {
+			if (k == 0)
+				die(_("could not start worker[0] for '%s'"),
+				    path);
+			/*
+			 * Limp along with the thread pool that we have.
+			 */
+			break;
+		}
+
+		wtd->next_thread = server_data->worker_thread_list;
+		server_data->worker_thread_list = wtd;
+	}
+
+	*returned_server_data = server_data;
+	return 0;
+}
+
+/*
+ * Gently tell the IPC server treads to shutdown.
+ * Can be run on any thread.
+ */
+int ipc_server_stop_async(struct ipc_server_data *server_data)
+{
+	/* ASSERT NOT holding mutex */
+
+	int fd;
+
+	if (!server_data)
+		return 0;
+
+	trace2_region_enter("ipc-server", "server-stop-async", NULL);
+
+	pthread_mutex_lock(&server_data->work_available_mutex);
+
+	server_data->shutdown_requested = 1;
+
+	/*
+	 * Write a byte to the shutdown socket pair to wake up the
+	 * accept-thread.
+	 */
+	if (write(server_data->accept_thread->fd_send_shutdown, "Q", 1) < 0)
+		error_errno("could not write to fd_send_shutdown");
+
+	/*
+	 * Drain the queue of existing connections.
+	 */
+	while ((fd = fifo_dequeue(server_data)) != -1)
+		close(fd);
+
+	/*
+	 * Gently tell worker threads to stop processing new connections
+	 * and exit.  (This does not abort in-process conversations.)
+	 */
+	pthread_cond_broadcast(&server_data->work_available_cond);
+
+	pthread_mutex_unlock(&server_data->work_available_mutex);
+
+	trace2_region_leave("ipc-server", "server-stop-async", NULL);
+
+	return 0;
+}
+
+/*
+ * Wait for all IPC server threads to stop.
+ */
+int ipc_server_await(struct ipc_server_data *server_data)
+{
+	pthread_join(server_data->accept_thread->pthread_id, NULL);
+
+	if (!server_data->shutdown_requested)
+		BUG("ipc-server: accept-thread stopped for '%s'",
+		    server_data->buf_path.buf);
+
+	while (server_data->worker_thread_list) {
+		struct ipc_worker_thread_data *wtd =
+			server_data->worker_thread_list;
+
+		pthread_join(wtd->pthread_id, NULL);
+
+		server_data->worker_thread_list = wtd->next_thread;
+		free(wtd);
+	}
+
+	server_data->is_stopped = 1;
+
+	return 0;
+}
+
+void ipc_server_free(struct ipc_server_data *server_data)
+{
+	struct ipc_accept_thread_data * accept_thread_data;
+
+	if (!server_data)
+		return;
+
+	if (!server_data->is_stopped)
+		BUG("cannot free ipc-server while running for '%s'",
+		    server_data->buf_path.buf);
+
+	accept_thread_data = server_data->accept_thread;
+	if (accept_thread_data) {
+		unix_ss_free(accept_thread_data->server_socket);
+
+		if (accept_thread_data->fd_send_shutdown != -1)
+			close(accept_thread_data->fd_send_shutdown);
+		if (accept_thread_data->fd_wait_shutdown != -1)
+			close(accept_thread_data->fd_wait_shutdown);
+
+		free(server_data->accept_thread);
+	}
+
+	while (server_data->worker_thread_list) {
+		struct ipc_worker_thread_data *wtd =
+			server_data->worker_thread_list;
+
+		server_data->worker_thread_list = wtd->next_thread;
+		free(wtd);
+	}
+
+	pthread_cond_destroy(&server_data->work_available_cond);
+	pthread_mutex_destroy(&server_data->work_available_mutex);
+
+	strbuf_release(&server_data->buf_path);
+
+	free(server_data->fifo_fds);
+	free(server_data);
+}
diff --git a/compat/simple-ipc/ipc-win32.c b/compat/simple-ipc/ipc-win32.c
new file mode 100644
index 0000000000..8f89c02037
--- /dev/null
+++ b/compat/simple-ipc/ipc-win32.c
@@ -0,0 +1,751 @@
+#include "cache.h"
+#include "simple-ipc.h"
+#include "strbuf.h"
+#include "pkt-line.h"
+#include "thread-utils.h"
+
+#ifndef GIT_WINDOWS_NATIVE
+#error This file can only be compiled on Windows
+#endif
+
+static int initialize_pipe_name(const char *path, wchar_t *wpath, size_t alloc)
+{
+	int off = 0;
+	struct strbuf realpath = STRBUF_INIT;
+
+	if (!strbuf_realpath(&realpath, path, 0))
+		return -1;
+
+	off = swprintf(wpath, alloc, L"\\\\.\\pipe\\");
+	if (xutftowcs(wpath + off, realpath.buf, alloc - off) < 0)
+		return -1;
+
+	/* Handle drive prefix */
+	if (wpath[off] && wpath[off + 1] == L':') {
+		wpath[off + 1] = L'_';
+		off += 2;
+	}
+
+	for (; wpath[off]; off++)
+		if (wpath[off] == L'/')
+			wpath[off] = L'\\';
+
+	strbuf_release(&realpath);
+	return 0;
+}
+
+static enum ipc_active_state get_active_state(wchar_t *pipe_path)
+{
+	if (WaitNamedPipeW(pipe_path, NMPWAIT_USE_DEFAULT_WAIT))
+		return IPC_STATE__LISTENING;
+
+	if (GetLastError() == ERROR_SEM_TIMEOUT)
+		return IPC_STATE__NOT_LISTENING;
+
+	if (GetLastError() == ERROR_FILE_NOT_FOUND)
+		return IPC_STATE__PATH_NOT_FOUND;
+
+	return IPC_STATE__OTHER_ERROR;
+}
+
+enum ipc_active_state ipc_get_active_state(const char *path)
+{
+	wchar_t pipe_path[MAX_PATH];
+
+	if (initialize_pipe_name(path, pipe_path, ARRAY_SIZE(pipe_path)) < 0)
+		return IPC_STATE__INVALID_PATH;
+
+	return get_active_state(pipe_path);
+}
+
+#define WAIT_STEP_MS (50)
+
+static enum ipc_active_state connect_to_server(
+	const wchar_t *wpath,
+	DWORD timeout_ms,
+	const struct ipc_client_connect_options *options,
+	int *pfd)
+{
+	DWORD t_start_ms, t_waited_ms;
+	DWORD step_ms;
+	HANDLE hPipe = INVALID_HANDLE_VALUE;
+	DWORD mode = PIPE_READMODE_BYTE;
+	DWORD gle;
+
+	*pfd = -1;
+
+	for (;;) {
+		hPipe = CreateFileW(wpath, GENERIC_READ | GENERIC_WRITE,
+				    0, NULL, OPEN_EXISTING, 0, NULL);
+		if (hPipe != INVALID_HANDLE_VALUE)
+			break;
+
+		gle = GetLastError();
+
+		switch (gle) {
+		case ERROR_FILE_NOT_FOUND:
+			if (!options->wait_if_not_found)
+				return IPC_STATE__PATH_NOT_FOUND;
+			if (!timeout_ms)
+				return IPC_STATE__PATH_NOT_FOUND;
+
+			step_ms = (timeout_ms < WAIT_STEP_MS) ?
+				timeout_ms : WAIT_STEP_MS;
+			sleep_millisec(step_ms);
+
+			timeout_ms -= step_ms;
+			break; /* try again */
+
+		case ERROR_PIPE_BUSY:
+			if (!options->wait_if_busy)
+				return IPC_STATE__NOT_LISTENING;
+			if (!timeout_ms)
+				return IPC_STATE__NOT_LISTENING;
+
+			t_start_ms = (DWORD)(getnanotime() / 1000000);
+
+			if (!WaitNamedPipeW(wpath, timeout_ms)) {
+				if (GetLastError() == ERROR_SEM_TIMEOUT)
+					return IPC_STATE__NOT_LISTENING;
+
+				return IPC_STATE__OTHER_ERROR;
+			}
+
+			/*
+			 * A pipe server instance became available.
+			 * Race other client processes to connect to
+			 * it.
+			 *
+			 * But first decrement our overall timeout so
+			 * that we don't starve if we keep losing the
+			 * race.  But also guard against special
+			 * NPMWAIT_ values (0 and -1).
+			 */
+			t_waited_ms = (DWORD)(getnanotime() / 1000000) - t_start_ms;
+			if (t_waited_ms < timeout_ms)
+				timeout_ms -= t_waited_ms;
+			else
+				timeout_ms = 1;
+			break; /* try again */
+
+		default:
+			return IPC_STATE__OTHER_ERROR;
+		}
+	}
+
+	if (!SetNamedPipeHandleState(hPipe, &mode, NULL, NULL)) {
+		CloseHandle(hPipe);
+		return IPC_STATE__OTHER_ERROR;
+	}
+
+	*pfd = _open_osfhandle((intptr_t)hPipe, O_RDWR|O_BINARY);
+	if (*pfd < 0) {
+		CloseHandle(hPipe);
+		return IPC_STATE__OTHER_ERROR;
+	}
+
+	/* fd now owns hPipe */
+
+	return IPC_STATE__LISTENING;
+}
+
+/*
+ * The default connection timeout for Windows clients.
+ *
+ * This is not currently part of the ipc_ API (nor the config settings)
+ * because of differences between Windows and other platforms.
+ *
+ * This value was chosen at random.
+ */
+#define WINDOWS_CONNECTION_TIMEOUT_MS (30000)
+
+enum ipc_active_state ipc_client_try_connect(
+	const char *path,
+	const struct ipc_client_connect_options *options,
+	struct ipc_client_connection **p_connection)
+{
+	wchar_t wpath[MAX_PATH];
+	enum ipc_active_state state = IPC_STATE__OTHER_ERROR;
+	int fd = -1;
+
+	*p_connection = NULL;
+
+	trace2_region_enter("ipc-client", "try-connect", NULL);
+	trace2_data_string("ipc-client", NULL, "try-connect/path", path);
+
+	if (initialize_pipe_name(path, wpath, ARRAY_SIZE(wpath)) < 0)
+		state = IPC_STATE__INVALID_PATH;
+	else
+		state = connect_to_server(wpath, WINDOWS_CONNECTION_TIMEOUT_MS,
+					  options, &fd);
+
+	trace2_data_intmax("ipc-client", NULL, "try-connect/state",
+			   (intmax_t)state);
+	trace2_region_leave("ipc-client", "try-connect", NULL);
+
+	if (state == IPC_STATE__LISTENING) {
+		(*p_connection) = xcalloc(1, sizeof(struct ipc_client_connection));
+		(*p_connection)->fd = fd;
+	}
+
+	return state;
+}
+
+void ipc_client_close_connection(struct ipc_client_connection *connection)
+{
+	if (!connection)
+		return;
+
+	if (connection->fd != -1)
+		close(connection->fd);
+
+	free(connection);
+}
+
+int ipc_client_send_command_to_connection(
+	struct ipc_client_connection *connection,
+	const char *message, struct strbuf *answer)
+{
+	int ret = 0;
+
+	strbuf_setlen(answer, 0);
+
+	trace2_region_enter("ipc-client", "send-command", NULL);
+
+	if (write_packetized_from_buf_no_flush(message, strlen(message),
+					       connection->fd) < 0 ||
+	    packet_flush_gently(connection->fd) < 0) {
+		ret = error(_("could not send IPC command"));
+		goto done;
+	}
+
+	FlushFileBuffers((HANDLE)_get_osfhandle(connection->fd));
+
+	if (read_packetized_to_strbuf(
+		    connection->fd, answer,
+		    PACKET_READ_GENTLE_ON_EOF | PACKET_READ_GENTLE_ON_READ_ERROR) < 0) {
+		ret = error(_("could not read IPC response"));
+		goto done;
+	}
+
+done:
+	trace2_region_leave("ipc-client", "send-command", NULL);
+	return ret;
+}
+
+int ipc_client_send_command(const char *path,
+			    const struct ipc_client_connect_options *options,
+			    const char *message, struct strbuf *response)
+{
+	int ret = -1;
+	enum ipc_active_state state;
+	struct ipc_client_connection *connection = NULL;
+
+	state = ipc_client_try_connect(path, options, &connection);
+
+	if (state != IPC_STATE__LISTENING)
+		return ret;
+
+	ret = ipc_client_send_command_to_connection(connection, message, response);
+
+	ipc_client_close_connection(connection);
+
+	return ret;
+}
+
+/*
+ * Duplicate the given pipe handle and wrap it in a file descriptor so
+ * that we can use pkt-line on it.
+ */
+static int dup_fd_from_pipe(const HANDLE pipe)
+{
+	HANDLE process = GetCurrentProcess();
+	HANDLE handle;
+	int fd;
+
+	if (!DuplicateHandle(process, pipe, process, &handle, 0, FALSE,
+			     DUPLICATE_SAME_ACCESS)) {
+		errno = err_win_to_posix(GetLastError());
+		return -1;
+	}
+
+	fd = _open_osfhandle((intptr_t)handle, O_RDWR|O_BINARY);
+	if (fd < 0) {
+		errno = err_win_to_posix(GetLastError());
+		CloseHandle(handle);
+		return -1;
+	}
+
+	/*
+	 * `handle` is now owned by `fd` and will be automatically closed
+	 * when the descriptor is closed.
+	 */
+
+	return fd;
+}
+
+/*
+ * Magic numbers used to annotate callback instance data.
+ * These are used to help guard against accidentally passing the
+ * wrong instance data across multiple levels of callbacks (which
+ * is easy to do if there are `void*` arguments).
+ */
+enum magic {
+	MAGIC_SERVER_REPLY_DATA,
+	MAGIC_SERVER_THREAD_DATA,
+	MAGIC_SERVER_DATA,
+};
+
+struct ipc_server_reply_data {
+	enum magic magic;
+	int fd;
+	struct ipc_server_thread_data *server_thread_data;
+};
+
+struct ipc_server_thread_data {
+	enum magic magic;
+	struct ipc_server_thread_data *next_thread;
+	struct ipc_server_data *server_data;
+	pthread_t pthread_id;
+	HANDLE hPipe;
+};
+
+/*
+ * On Windows, the conceptual "ipc-server" is implemented as a pool of
+ * n idential/peer "server-thread" threads.  That is, there is no
+ * hierarchy of threads; and therefore no controller thread managing
+ * the pool.  Each thread has an independent handle to the named pipe,
+ * receives incoming connections, processes the client, and re-uses
+ * the pipe for the next client connection.
+ *
+ * Therefore, the "ipc-server" only needs to maintain a list of the
+ * spawned threads for eventual "join" purposes.
+ *
+ * A single "stop-event" is visible to all of the server threads to
+ * tell them to shutdown (when idle).
+ */
+struct ipc_server_data {
+	enum magic magic;
+	ipc_server_application_cb *application_cb;
+	void *application_data;
+	struct strbuf buf_path;
+	wchar_t wpath[MAX_PATH];
+
+	HANDLE hEventStopRequested;
+	struct ipc_server_thread_data *thread_list;
+	int is_stopped;
+};
+
+enum connect_result {
+	CR_CONNECTED = 0,
+	CR_CONNECT_PENDING,
+	CR_CONNECT_ERROR,
+	CR_WAIT_ERROR,
+	CR_SHUTDOWN,
+};
+
+static enum connect_result queue_overlapped_connect(
+	struct ipc_server_thread_data *server_thread_data,
+	OVERLAPPED *lpo)
+{
+	if (ConnectNamedPipe(server_thread_data->hPipe, lpo))
+		goto failed;
+
+	switch (GetLastError()) {
+	case ERROR_IO_PENDING:
+		return CR_CONNECT_PENDING;
+
+	case ERROR_PIPE_CONNECTED:
+		SetEvent(lpo->hEvent);
+		return CR_CONNECTED;
+
+	default:
+		break;
+	}
+
+failed:
+	error(_("ConnectNamedPipe failed for '%s' (%lu)"),
+	      server_thread_data->server_data->buf_path.buf,
+	      GetLastError());
+	return CR_CONNECT_ERROR;
+}
+
+/*
+ * Use Windows Overlapped IO to wait for a connection or for our event
+ * to be signalled.
+ */
+static enum connect_result wait_for_connection(
+	struct ipc_server_thread_data *server_thread_data,
+	OVERLAPPED *lpo)
+{
+	enum connect_result r;
+	HANDLE waitHandles[2];
+	DWORD dwWaitResult;
+
+	r = queue_overlapped_connect(server_thread_data, lpo);
+	if (r != CR_CONNECT_PENDING)
+		return r;
+
+	waitHandles[0] = server_thread_data->server_data->hEventStopRequested;
+	waitHandles[1] = lpo->hEvent;
+
+	dwWaitResult = WaitForMultipleObjects(2, waitHandles, FALSE, INFINITE);
+	switch (dwWaitResult) {
+	case WAIT_OBJECT_0 + 0:
+		return CR_SHUTDOWN;
+
+	case WAIT_OBJECT_0 + 1:
+		ResetEvent(lpo->hEvent);
+		return CR_CONNECTED;
+
+	default:
+		return CR_WAIT_ERROR;
+	}
+}
+
+/*
+ * Forward declare our reply callback function so that any compiler
+ * errors are reported when we actually define the function (in addition
+ * to any errors reported when we try to pass this callback function as
+ * a parameter in a function call).  The former are easier to understand.
+ */
+static ipc_server_reply_cb do_io_reply_callback;
+
+/*
+ * Relay application's response message to the client process.
+ * (We do not flush at this point because we allow the caller
+ * to chunk data to the client thru us.)
+ */
+static int do_io_reply_callback(struct ipc_server_reply_data *reply_data,
+		       const char *response, size_t response_len)
+{
+	if (reply_data->magic != MAGIC_SERVER_REPLY_DATA)
+		BUG("reply_cb called with wrong instance data");
+
+	return write_packetized_from_buf_no_flush(response, response_len,
+						  reply_data->fd);
+}
+
+/*
+ * Receive the request/command from the client and pass it to the
+ * registered request-callback.  The request-callback will compose
+ * a response and call our reply-callback to send it to the client.
+ *
+ * Simple-IPC only contains one round trip, so we flush and close
+ * here after the response.
+ */
+static int do_io(struct ipc_server_thread_data *server_thread_data)
+{
+	struct strbuf buf = STRBUF_INIT;
+	struct ipc_server_reply_data reply_data;
+	int ret = 0;
+
+	reply_data.magic = MAGIC_SERVER_REPLY_DATA;
+	reply_data.server_thread_data = server_thread_data;
+
+	reply_data.fd = dup_fd_from_pipe(server_thread_data->hPipe);
+	if (reply_data.fd < 0)
+		return error(_("could not create fd from pipe for '%s'"),
+			     server_thread_data->server_data->buf_path.buf);
+
+	ret = read_packetized_to_strbuf(
+		reply_data.fd, &buf,
+		PACKET_READ_GENTLE_ON_EOF | PACKET_READ_GENTLE_ON_READ_ERROR);
+	if (ret >= 0) {
+		ret = server_thread_data->server_data->application_cb(
+			server_thread_data->server_data->application_data,
+			buf.buf, do_io_reply_callback, &reply_data);
+
+		packet_flush_gently(reply_data.fd);
+
+		FlushFileBuffers((HANDLE)_get_osfhandle((reply_data.fd)));
+	}
+	else {
+		/*
+		 * The client probably disconnected/shutdown before it
+		 * could send a well-formed message.  Ignore it.
+		 */
+	}
+
+	strbuf_release(&buf);
+	close(reply_data.fd);
+
+	return ret;
+}
+
+/*
+ * Handle IPC request and response with this connected client.  And reset
+ * the pipe to prepare for the next client.
+ */
+static int use_connection(struct ipc_server_thread_data *server_thread_data)
+{
+	int ret;
+
+	ret = do_io(server_thread_data);
+
+	FlushFileBuffers(server_thread_data->hPipe);
+	DisconnectNamedPipe(server_thread_data->hPipe);
+
+	return ret;
+}
+
+/*
+ * Thread proc for an IPC server worker thread.  It handles a series of
+ * connections from clients.  It cleans and reuses the hPipe between each
+ * client.
+ */
+static void *server_thread_proc(void *_server_thread_data)
+{
+	struct ipc_server_thread_data *server_thread_data = _server_thread_data;
+	HANDLE hEventConnected = INVALID_HANDLE_VALUE;
+	OVERLAPPED oConnect;
+	enum connect_result cr;
+	int ret;
+
+	assert(server_thread_data->hPipe != INVALID_HANDLE_VALUE);
+
+	trace2_thread_start("ipc-server");
+	trace2_data_string("ipc-server", NULL, "pipe",
+			   server_thread_data->server_data->buf_path.buf);
+
+	hEventConnected = CreateEventW(NULL, TRUE, FALSE, NULL);
+
+	memset(&oConnect, 0, sizeof(oConnect));
+	oConnect.hEvent = hEventConnected;
+
+	for (;;) {
+		cr = wait_for_connection(server_thread_data, &oConnect);
+
+		switch (cr) {
+		case CR_SHUTDOWN:
+			goto finished;
+
+		case CR_CONNECTED:
+			ret = use_connection(server_thread_data);
+			if (ret == SIMPLE_IPC_QUIT) {
+				ipc_server_stop_async(
+					server_thread_data->server_data);
+				goto finished;
+			}
+			if (ret > 0) {
+				/*
+				 * Ignore (transient) IO errors with this
+				 * client and reset for the next client.
+				 */
+			}
+			break;
+
+		case CR_CONNECT_PENDING:
+			/* By construction, this should not happen. */
+			BUG("ipc-server[%s]: unexpeced CR_CONNECT_PENDING",
+			    server_thread_data->server_data->buf_path.buf);
+
+		case CR_CONNECT_ERROR:
+		case CR_WAIT_ERROR:
+			/*
+			 * Ignore these theoretical errors.
+			 */
+			DisconnectNamedPipe(server_thread_data->hPipe);
+			break;
+
+		default:
+			BUG("unandled case after wait_for_connection");
+		}
+	}
+
+finished:
+	CloseHandle(server_thread_data->hPipe);
+	CloseHandle(hEventConnected);
+
+	trace2_thread_exit();
+	return NULL;
+}
+
+static HANDLE create_new_pipe(wchar_t *wpath, int is_first)
+{
+	HANDLE hPipe;
+	DWORD dwOpenMode, dwPipeMode;
+	LPSECURITY_ATTRIBUTES lpsa = NULL;
+
+	dwOpenMode = PIPE_ACCESS_INBOUND | PIPE_ACCESS_OUTBOUND |
+		FILE_FLAG_OVERLAPPED;
+
+	dwPipeMode = PIPE_TYPE_MESSAGE | PIPE_READMODE_BYTE | PIPE_WAIT |
+		PIPE_REJECT_REMOTE_CLIENTS;
+
+	if (is_first) {
+		dwOpenMode |= FILE_FLAG_FIRST_PIPE_INSTANCE;
+
+		/*
+		 * On Windows, the first server pipe instance gets to
+		 * set the ACL / Security Attributes on the named
+		 * pipe; subsequent instances inherit and cannot
+		 * change them.
+		 *
+		 * TODO Should we allow the application layer to
+		 * specify security attributes, such as `LocalService`
+		 * or `LocalSystem`, when we create the named pipe?
+		 * This question is probably not important when the
+		 * daemon is started by a foreground user process and
+		 * only needs to talk to the current user, but may be
+		 * if the daemon is run via the Control Panel as a
+		 * System Service.
+		 */
+	}
+
+	hPipe = CreateNamedPipeW(wpath, dwOpenMode, dwPipeMode,
+				 PIPE_UNLIMITED_INSTANCES, 1024, 1024, 0, lpsa);
+
+	return hPipe;
+}
+
+int ipc_server_run_async(struct ipc_server_data **returned_server_data,
+			 const char *path, const struct ipc_server_opts *opts,
+			 ipc_server_application_cb *application_cb,
+			 void *application_data)
+{
+	struct ipc_server_data *server_data;
+	wchar_t wpath[MAX_PATH];
+	HANDLE hPipeFirst = INVALID_HANDLE_VALUE;
+	int k;
+	int ret = 0;
+	int nr_threads = opts->nr_threads;
+
+	*returned_server_data = NULL;
+
+	ret = initialize_pipe_name(path, wpath, ARRAY_SIZE(wpath));
+	if (ret < 0) {
+		errno = EINVAL;
+		return -1;
+	}
+
+	hPipeFirst = create_new_pipe(wpath, 1);
+	if (hPipeFirst == INVALID_HANDLE_VALUE) {
+		errno = EADDRINUSE;
+		return -2;
+	}
+
+	server_data = xcalloc(1, sizeof(*server_data));
+	server_data->magic = MAGIC_SERVER_DATA;
+	server_data->application_cb = application_cb;
+	server_data->application_data = application_data;
+	server_data->hEventStopRequested = CreateEvent(NULL, TRUE, FALSE, NULL);
+	strbuf_init(&server_data->buf_path, 0);
+	strbuf_addstr(&server_data->buf_path, path);
+	wcscpy(server_data->wpath, wpath);
+
+	if (nr_threads < 1)
+		nr_threads = 1;
+
+	for (k = 0; k < nr_threads; k++) {
+		struct ipc_server_thread_data *std;
+
+		std = xcalloc(1, sizeof(*std));
+		std->magic = MAGIC_SERVER_THREAD_DATA;
+		std->server_data = server_data;
+		std->hPipe = INVALID_HANDLE_VALUE;
+
+		std->hPipe = (k == 0)
+			? hPipeFirst
+			: create_new_pipe(server_data->wpath, 0);
+
+		if (std->hPipe == INVALID_HANDLE_VALUE) {
+			/*
+			 * If we've reached a pipe instance limit for
+			 * this path, just use fewer threads.
+			 */
+			free(std);
+			break;
+		}
+
+		if (pthread_create(&std->pthread_id, NULL,
+				   server_thread_proc, std)) {
+			/*
+			 * Likewise, if we're out of threads, just use
+			 * fewer threads than requested.
+			 *
+			 * However, we just give up if we can't even get
+			 * one thread.  This should not happen.
+			 */
+			if (k == 0)
+				die(_("could not start thread[0] for '%s'"),
+				    path);
+
+			CloseHandle(std->hPipe);
+			free(std);
+			break;
+		}
+
+		std->next_thread = server_data->thread_list;
+		server_data->thread_list = std;
+	}
+
+	*returned_server_data = server_data;
+	return 0;
+}
+
+int ipc_server_stop_async(struct ipc_server_data *server_data)
+{
+	if (!server_data)
+		return 0;
+
+	/*
+	 * Gently tell all of the ipc_server threads to shutdown.
+	 * This will be seen the next time they are idle (and waiting
+	 * for a connection).
+	 *
+	 * We DO NOT attempt to force them to drop an active connection.
+	 */
+	SetEvent(server_data->hEventStopRequested);
+	return 0;
+}
+
+int ipc_server_await(struct ipc_server_data *server_data)
+{
+	DWORD dwWaitResult;
+
+	if (!server_data)
+		return 0;
+
+	dwWaitResult = WaitForSingleObject(server_data->hEventStopRequested, INFINITE);
+	if (dwWaitResult != WAIT_OBJECT_0)
+		return error(_("wait for hEvent failed for '%s'"),
+			     server_data->buf_path.buf);
+
+	while (server_data->thread_list) {
+		struct ipc_server_thread_data *std = server_data->thread_list;
+
+		pthread_join(std->pthread_id, NULL);
+
+		server_data->thread_list = std->next_thread;
+		free(std);
+	}
+
+	server_data->is_stopped = 1;
+
+	return 0;
+}
+
+void ipc_server_free(struct ipc_server_data *server_data)
+{
+	if (!server_data)
+		return;
+
+	if (!server_data->is_stopped)
+		BUG("cannot free ipc-server while running for '%s'",
+		    server_data->buf_path.buf);
+
+	strbuf_release(&server_data->buf_path);
+
+	if (server_data->hEventStopRequested != INVALID_HANDLE_VALUE)
+		CloseHandle(server_data->hEventStopRequested);
+
+	while (server_data->thread_list) {
+		struct ipc_server_thread_data *std = server_data->thread_list;
+
+		server_data->thread_list = std->next_thread;
+		free(std);
+	}
+
+	free(server_data);
+}