16 files changed, 1566 insertions, 65 deletions

diff --git a/compat/.gitattributes b/compat/.gitattributes
new file mode 100644
index 0000000..40dbfb1
--- /dev/null
+++ b/compat/.gitattributes
@@ -0,0 +1 @@
+/zlib-uncompress2.c	whitespace=-indent-with-non-tab,-trailing-space
diff --git a/compat/fsmonitor/fsm-darwin-gcc.h b/compat/fsmonitor/fsm-darwin-gcc.h
new file mode 100644
index 0000000..1c75c3d
--- /dev/null
+++ b/compat/fsmonitor/fsm-darwin-gcc.h
@@ -0,0 +1,92 @@
+#ifndef FSM_DARWIN_GCC_H
+#define FSM_DARWIN_GCC_H
+
+#ifndef __clang__
+/*
+ * It is possible to #include CoreFoundation/CoreFoundation.h when compiling
+ * with clang, but not with GCC as of time of writing.
+ *
+ * See https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93082 for details.
+ */
+typedef unsigned int FSEventStreamCreateFlags;
+#define kFSEventStreamEventFlagNone               0x00000000
+#define kFSEventStreamEventFlagMustScanSubDirs    0x00000001
+#define kFSEventStreamEventFlagUserDropped        0x00000002
+#define kFSEventStreamEventFlagKernelDropped      0x00000004
+#define kFSEventStreamEventFlagEventIdsWrapped    0x00000008
+#define kFSEventStreamEventFlagHistoryDone        0x00000010
+#define kFSEventStreamEventFlagRootChanged        0x00000020
+#define kFSEventStreamEventFlagMount              0x00000040
+#define kFSEventStreamEventFlagUnmount            0x00000080
+#define kFSEventStreamEventFlagItemCreated        0x00000100
+#define kFSEventStreamEventFlagItemRemoved        0x00000200
+#define kFSEventStreamEventFlagItemInodeMetaMod   0x00000400
+#define kFSEventStreamEventFlagItemRenamed        0x00000800
+#define kFSEventStreamEventFlagItemModified       0x00001000
+#define kFSEventStreamEventFlagItemFinderInfoMod  0x00002000
+#define kFSEventStreamEventFlagItemChangeOwner    0x00004000
+#define kFSEventStreamEventFlagItemXattrMod       0x00008000
+#define kFSEventStreamEventFlagItemIsFile         0x00010000
+#define kFSEventStreamEventFlagItemIsDir          0x00020000
+#define kFSEventStreamEventFlagItemIsSymlink      0x00040000
+#define kFSEventStreamEventFlagOwnEvent           0x00080000
+#define kFSEventStreamEventFlagItemIsHardlink     0x00100000
+#define kFSEventStreamEventFlagItemIsLastHardlink 0x00200000
+#define kFSEventStreamEventFlagItemCloned         0x00400000
+
+typedef struct __FSEventStream *FSEventStreamRef;
+typedef const FSEventStreamRef ConstFSEventStreamRef;
+
+typedef unsigned int CFStringEncoding;
+#define kCFStringEncodingUTF8 0x08000100
+
+typedef const struct __CFString *CFStringRef;
+typedef const struct __CFArray *CFArrayRef;
+typedef const struct __CFRunLoop *CFRunLoopRef;
+
+struct FSEventStreamContext {
+    long long version;
+    void *cb_data, *retain, *release, *copy_description;
+};
+
+typedef struct FSEventStreamContext FSEventStreamContext;
+typedef unsigned int FSEventStreamEventFlags;
+#define kFSEventStreamCreateFlagNoDefer 0x02
+#define kFSEventStreamCreateFlagWatchRoot 0x04
+#define kFSEventStreamCreateFlagFileEvents 0x10
+
+typedef unsigned long long FSEventStreamEventId;
+#define kFSEventStreamEventIdSinceNow 0xFFFFFFFFFFFFFFFFULL
+
+typedef void (*FSEventStreamCallback)(ConstFSEventStreamRef streamRef,
+				      void *context,
+				      __SIZE_TYPE__ num_of_events,
+				      void *event_paths,
+				      const FSEventStreamEventFlags event_flags[],
+				      const FSEventStreamEventId event_ids[]);
+typedef double CFTimeInterval;
+FSEventStreamRef FSEventStreamCreate(void *allocator,
+				     FSEventStreamCallback callback,
+				     FSEventStreamContext *context,
+				     CFArrayRef paths_to_watch,
+				     FSEventStreamEventId since_when,
+				     CFTimeInterval latency,
+				     FSEventStreamCreateFlags flags);
+CFStringRef CFStringCreateWithCString(void *allocator, const char *string,
+				      CFStringEncoding encoding);
+CFArrayRef CFArrayCreate(void *allocator, const void **items, long long count,
+			 void *callbacks);
+void CFRunLoopRun(void);
+void CFRunLoopStop(CFRunLoopRef run_loop);
+CFRunLoopRef CFRunLoopGetCurrent(void);
+extern CFStringRef kCFRunLoopDefaultMode;
+void FSEventStreamScheduleWithRunLoop(FSEventStreamRef stream,
+				      CFRunLoopRef run_loop,
+				      CFStringRef run_loop_mode);
+unsigned char FSEventStreamStart(FSEventStreamRef stream);
+void FSEventStreamStop(FSEventStreamRef stream);
+void FSEventStreamInvalidate(FSEventStreamRef stream);
+void FSEventStreamRelease(FSEventStreamRef stream);
+
+#endif /* !clang */
+#endif /* FSM_DARWIN_GCC_H */
diff --git a/compat/fsmonitor/fsm-listen-darwin.c b/compat/fsmonitor/fsm-listen-darwin.c
new file mode 100644
index 0000000..0741fe8
--- /dev/null
+++ b/compat/fsmonitor/fsm-listen-darwin.c
@@ -0,0 +1,427 @@
+#ifndef __clang__
+#include "fsm-darwin-gcc.h"
+#else
+#include <CoreFoundation/CoreFoundation.h>
+#include <CoreServices/CoreServices.h>
+
+#ifndef AVAILABLE_MAC_OS_X_VERSION_10_13_AND_LATER
+/*
+ * This enum value was added in 10.13 to:
+ *
+ * /Applications/Xcode.app/Contents/Developer/Platforms/ \
+ *    MacOSX.platform/Developer/SDKs/MacOSX.sdk/System/ \
+ *    Library/Frameworks/CoreServices.framework/Frameworks/ \
+ *    FSEvents.framework/Versions/Current/Headers/FSEvents.h
+ *
+ * If we're compiling against an older SDK, this symbol won't be
+ * present.  Silently define it here so that we don't have to ifdef
+ * the logging or masking below.  This should be harmless since older
+ * versions of macOS won't ever emit this FS event anyway.
+ */
+#define kFSEventStreamEventFlagItemCloned         0x00400000
+#endif
+#endif
+
+#include "cache.h"
+#include "fsmonitor.h"
+#include "fsm-listen.h"
+#include "fsmonitor--daemon.h"
+
+struct fsmonitor_daemon_backend_data
+{
+	CFStringRef cfsr_worktree_path;
+	CFStringRef cfsr_gitdir_path;
+
+	CFArrayRef cfar_paths_to_watch;
+	int nr_paths_watching;
+
+	FSEventStreamRef stream;
+
+	CFRunLoopRef rl;
+
+	enum shutdown_style {
+		SHUTDOWN_EVENT = 0,
+		FORCE_SHUTDOWN,
+		FORCE_ERROR_STOP,
+	} shutdown_style;
+
+	unsigned int stream_scheduled:1;
+	unsigned int stream_started:1;
+};
+
+static void log_flags_set(const char *path, const FSEventStreamEventFlags flag)
+{
+	struct strbuf msg = STRBUF_INIT;
+
+	if (flag & kFSEventStreamEventFlagMustScanSubDirs)
+		strbuf_addstr(&msg, "MustScanSubDirs|");
+	if (flag & kFSEventStreamEventFlagUserDropped)
+		strbuf_addstr(&msg, "UserDropped|");
+	if (flag & kFSEventStreamEventFlagKernelDropped)
+		strbuf_addstr(&msg, "KernelDropped|");
+	if (flag & kFSEventStreamEventFlagEventIdsWrapped)
+		strbuf_addstr(&msg, "EventIdsWrapped|");
+	if (flag & kFSEventStreamEventFlagHistoryDone)
+		strbuf_addstr(&msg, "HistoryDone|");
+	if (flag & kFSEventStreamEventFlagRootChanged)
+		strbuf_addstr(&msg, "RootChanged|");
+	if (flag & kFSEventStreamEventFlagMount)
+		strbuf_addstr(&msg, "Mount|");
+	if (flag & kFSEventStreamEventFlagUnmount)
+		strbuf_addstr(&msg, "Unmount|");
+	if (flag & kFSEventStreamEventFlagItemChangeOwner)
+		strbuf_addstr(&msg, "ItemChangeOwner|");
+	if (flag & kFSEventStreamEventFlagItemCreated)
+		strbuf_addstr(&msg, "ItemCreated|");
+	if (flag & kFSEventStreamEventFlagItemFinderInfoMod)
+		strbuf_addstr(&msg, "ItemFinderInfoMod|");
+	if (flag & kFSEventStreamEventFlagItemInodeMetaMod)
+		strbuf_addstr(&msg, "ItemInodeMetaMod|");
+	if (flag & kFSEventStreamEventFlagItemIsDir)
+		strbuf_addstr(&msg, "ItemIsDir|");
+	if (flag & kFSEventStreamEventFlagItemIsFile)
+		strbuf_addstr(&msg, "ItemIsFile|");
+	if (flag & kFSEventStreamEventFlagItemIsHardlink)
+		strbuf_addstr(&msg, "ItemIsHardlink|");
+	if (flag & kFSEventStreamEventFlagItemIsLastHardlink)
+		strbuf_addstr(&msg, "ItemIsLastHardlink|");
+	if (flag & kFSEventStreamEventFlagItemIsSymlink)
+		strbuf_addstr(&msg, "ItemIsSymlink|");
+	if (flag & kFSEventStreamEventFlagItemModified)
+		strbuf_addstr(&msg, "ItemModified|");
+	if (flag & kFSEventStreamEventFlagItemRemoved)
+		strbuf_addstr(&msg, "ItemRemoved|");
+	if (flag & kFSEventStreamEventFlagItemRenamed)
+		strbuf_addstr(&msg, "ItemRenamed|");
+	if (flag & kFSEventStreamEventFlagItemXattrMod)
+		strbuf_addstr(&msg, "ItemXattrMod|");
+	if (flag & kFSEventStreamEventFlagOwnEvent)
+		strbuf_addstr(&msg, "OwnEvent|");
+	if (flag & kFSEventStreamEventFlagItemCloned)
+		strbuf_addstr(&msg, "ItemCloned|");
+
+	trace_printf_key(&trace_fsmonitor, "fsevent: '%s', flags=%u %s",
+			 path, flag, msg.buf);
+
+	strbuf_release(&msg);
+}
+
+static int ef_is_root_delete(const FSEventStreamEventFlags ef)
+{
+	return (ef & kFSEventStreamEventFlagItemIsDir &&
+		ef & kFSEventStreamEventFlagItemRemoved);
+}
+
+static int ef_is_root_renamed(const FSEventStreamEventFlags ef)
+{
+	return (ef & kFSEventStreamEventFlagItemIsDir &&
+		ef & kFSEventStreamEventFlagItemRenamed);
+}
+
+static int ef_is_dropped(const FSEventStreamEventFlags ef)
+{
+	return (ef & kFSEventStreamEventFlagMustScanSubDirs ||
+		ef & kFSEventStreamEventFlagKernelDropped ||
+		ef & kFSEventStreamEventFlagUserDropped);
+}
+
+static void fsevent_callback(ConstFSEventStreamRef streamRef,
+			     void *ctx,
+			     size_t num_of_events,
+			     void *event_paths,
+			     const FSEventStreamEventFlags event_flags[],
+			     const FSEventStreamEventId event_ids[])
+{
+	struct fsmonitor_daemon_state *state = ctx;
+	struct fsmonitor_daemon_backend_data *data = state->backend_data;
+	char **paths = (char **)event_paths;
+	struct fsmonitor_batch *batch = NULL;
+	struct string_list cookie_list = STRING_LIST_INIT_DUP;
+	const char *path_k;
+	const char *slash;
+	int k;
+	struct strbuf tmp = STRBUF_INIT;
+
+	/*
+	 * Build a list of all filesystem changes into a private/local
+	 * list and without holding any locks.
+	 */
+	for (k = 0; k < num_of_events; k++) {
+		/*
+		 * On Mac, we receive an array of absolute paths.
+		 */
+		path_k = paths[k];
+
+		/*
+		 * If you want to debug FSEvents, log them to GIT_TRACE_FSMONITOR.
+		 * Please don't log them to Trace2.
+		 *
+		 * trace_printf_key(&trace_fsmonitor, "Path: '%s'", path_k);
+		 */
+
+		/*
+		 * If event[k] is marked as dropped, we assume that we have
+		 * lost sync with the filesystem and should flush our cached
+		 * data.  We need to:
+		 *
+		 * [1] Abort/wake any client threads waiting for a cookie and
+		 *     flush the cached state data (the current token), and
+		 *     create a new token.
+		 *
+		 * [2] Discard the batch that we were locally building (since
+		 *     they are conceptually relative to the just flushed
+		 *     token).
+		 */
+		if (ef_is_dropped(event_flags[k])) {
+			if (trace_pass_fl(&trace_fsmonitor))
+				log_flags_set(path_k, event_flags[k]);
+
+			fsmonitor_force_resync(state);
+			fsmonitor_batch__free_list(batch);
+			string_list_clear(&cookie_list, 0);
+
+			/*
+			 * We assume that any events that we received
+			 * in this callback after this dropped event
+			 * may still be valid, so we continue rather
+			 * than break.  (And just in case there is a
+			 * delete of ".git" hiding in there.)
+			 */
+			continue;
+		}
+
+		switch (fsmonitor_classify_path_absolute(state, path_k)) {
+
+		case IS_INSIDE_DOT_GIT_WITH_COOKIE_PREFIX:
+		case IS_INSIDE_GITDIR_WITH_COOKIE_PREFIX:
+			/* special case cookie files within .git or gitdir */
+
+			/* Use just the filename of the cookie file. */
+			slash = find_last_dir_sep(path_k);
+			string_list_append(&cookie_list,
+					   slash ? slash + 1 : path_k);
+			break;
+
+		case IS_INSIDE_DOT_GIT:
+		case IS_INSIDE_GITDIR:
+			/* ignore all other paths inside of .git or gitdir */
+			break;
+
+		case IS_DOT_GIT:
+		case IS_GITDIR:
+			/*
+			 * If .git directory is deleted or renamed away,
+			 * we have to quit.
+			 */
+			if (ef_is_root_delete(event_flags[k])) {
+				trace_printf_key(&trace_fsmonitor,
+						 "event: gitdir removed");
+				goto force_shutdown;
+			}
+			if (ef_is_root_renamed(event_flags[k])) {
+				trace_printf_key(&trace_fsmonitor,
+						 "event: gitdir renamed");
+				goto force_shutdown;
+			}
+			break;
+
+		case IS_WORKDIR_PATH:
+			/* try to queue normal pathnames */
+
+			if (trace_pass_fl(&trace_fsmonitor))
+				log_flags_set(path_k, event_flags[k]);
+
+			/*
+			 * Because of the implicit "binning" (the
+			 * kernel calls us at a given frequency) and
+			 * de-duping (the kernel is free to combine
+			 * multiple events for a given pathname), an
+			 * individual fsevent could be marked as both
+			 * a file and directory.  Add it to the queue
+			 * with both spellings so that the client will
+			 * know how much to invalidate/refresh.
+			 */
+
+			if (event_flags[k] & kFSEventStreamEventFlagItemIsFile) {
+				const char *rel = path_k +
+					state->path_worktree_watch.len + 1;
+
+				if (!batch)
+					batch = fsmonitor_batch__new();
+				fsmonitor_batch__add_path(batch, rel);
+			}
+
+			if (event_flags[k] & kFSEventStreamEventFlagItemIsDir) {
+				const char *rel = path_k +
+					state->path_worktree_watch.len + 1;
+
+				strbuf_reset(&tmp);
+				strbuf_addstr(&tmp, rel);
+				strbuf_addch(&tmp, '/');
+
+				if (!batch)
+					batch = fsmonitor_batch__new();
+				fsmonitor_batch__add_path(batch, tmp.buf);
+			}
+
+			break;
+
+		case IS_OUTSIDE_CONE:
+		default:
+			trace_printf_key(&trace_fsmonitor,
+					 "ignoring '%s'", path_k);
+			break;
+		}
+	}
+
+	fsmonitor_publish(state, batch, &cookie_list);
+	string_list_clear(&cookie_list, 0);
+	strbuf_release(&tmp);
+	return;
+
+force_shutdown:
+	fsmonitor_batch__free_list(batch);
+	string_list_clear(&cookie_list, 0);
+
+	data->shutdown_style = FORCE_SHUTDOWN;
+	CFRunLoopStop(data->rl);
+	strbuf_release(&tmp);
+	return;
+}
+
+/*
+ * In the call to `FSEventStreamCreate()` to setup our watch, the
+ * `latency` argument determines the frequency of calls to our callback
+ * with new FS events.  Too slow and events get dropped; too fast and
+ * we burn CPU unnecessarily.  Since it is rather obscure, I don't
+ * think this needs to be a config setting.  I've done extensive
+ * testing on my systems and chosen the value below.  It gives good
+ * results and I've not seen any dropped events.
+ *
+ * With a latency of 0.1, I was seeing lots of dropped events during
+ * the "touch 100000" files test within t/perf/p7519, but with a
+ * latency of 0.001 I did not see any dropped events.  So I'm going
+ * to assume that this is the "correct" value.
+ *
+ * https://developer.apple.com/documentation/coreservices/1443980-fseventstreamcreate
+ */
+
+int fsm_listen__ctor(struct fsmonitor_daemon_state *state)
+{
+	FSEventStreamCreateFlags flags = kFSEventStreamCreateFlagNoDefer |
+		kFSEventStreamCreateFlagWatchRoot |
+		kFSEventStreamCreateFlagFileEvents;
+	FSEventStreamContext ctx = {
+		0,
+		state,
+		NULL,
+		NULL,
+		NULL
+	};
+	struct fsmonitor_daemon_backend_data *data;
+	const void *dir_array[2];
+
+	CALLOC_ARRAY(data, 1);
+	state->backend_data = data;
+
+	data->cfsr_worktree_path = CFStringCreateWithCString(
+		NULL, state->path_worktree_watch.buf, kCFStringEncodingUTF8);
+	dir_array[data->nr_paths_watching++] = data->cfsr_worktree_path;
+
+	if (state->nr_paths_watching > 1) {
+		data->cfsr_gitdir_path = CFStringCreateWithCString(
+			NULL, state->path_gitdir_watch.buf,
+			kCFStringEncodingUTF8);
+		dir_array[data->nr_paths_watching++] = data->cfsr_gitdir_path;
+	}
+
+	data->cfar_paths_to_watch = CFArrayCreate(NULL, dir_array,
+						  data->nr_paths_watching,
+						  NULL);
+	data->stream = FSEventStreamCreate(NULL, fsevent_callback, &ctx,
+					   data->cfar_paths_to_watch,
+					   kFSEventStreamEventIdSinceNow,
+					   0.001, flags);
+	if (data->stream == NULL)
+		goto failed;
+
+	/*
+	 * `data->rl` needs to be set inside the listener thread.
+	 */
+
+	return 0;
+
+failed:
+	error(_("Unable to create FSEventStream."));
+
+	FREE_AND_NULL(state->backend_data);
+	return -1;
+}
+
+void fsm_listen__dtor(struct fsmonitor_daemon_state *state)
+{
+	struct fsmonitor_daemon_backend_data *data;
+
+	if (!state || !state->backend_data)
+		return;
+
+	data = state->backend_data;
+
+	if (data->stream) {
+		if (data->stream_started)
+			FSEventStreamStop(data->stream);
+		if (data->stream_scheduled)
+			FSEventStreamInvalidate(data->stream);
+		FSEventStreamRelease(data->stream);
+	}
+
+	FREE_AND_NULL(state->backend_data);
+}
+
+void fsm_listen__stop_async(struct fsmonitor_daemon_state *state)
+{
+	struct fsmonitor_daemon_backend_data *data;
+
+	data = state->backend_data;
+	data->shutdown_style = SHUTDOWN_EVENT;
+
+	CFRunLoopStop(data->rl);
+}
+
+void fsm_listen__loop(struct fsmonitor_daemon_state *state)
+{
+	struct fsmonitor_daemon_backend_data *data;
+
+	data = state->backend_data;
+
+	data->rl = CFRunLoopGetCurrent();
+
+	FSEventStreamScheduleWithRunLoop(data->stream, data->rl, kCFRunLoopDefaultMode);
+	data->stream_scheduled = 1;
+
+	if (!FSEventStreamStart(data->stream)) {
+		error(_("Failed to start the FSEventStream"));
+		goto force_error_stop_without_loop;
+	}
+	data->stream_started = 1;
+
+	CFRunLoopRun();
+
+	switch (data->shutdown_style) {
+	case FORCE_ERROR_STOP:
+		state->error_code = -1;
+		/* fall thru */
+	case FORCE_SHUTDOWN:
+		ipc_server_stop_async(state->ipc_server_data);
+		/* fall thru */
+	case SHUTDOWN_EVENT:
+	default:
+		break;
+	}
+	return;
+
+force_error_stop_without_loop:
+	state->error_code = -1;
+	ipc_server_stop_async(state->ipc_server_data);
+	return;
+}
diff --git a/compat/fsmonitor/fsm-listen-win32.c b/compat/fsmonitor/fsm-listen-win32.c
new file mode 100644
index 0000000..5b928ab
--- /dev/null
+++ b/compat/fsmonitor/fsm-listen-win32.c
@@ -0,0 +1,586 @@
+#include "cache.h"
+#include "config.h"
+#include "fsmonitor.h"
+#include "fsm-listen.h"
+#include "fsmonitor--daemon.h"
+
+/*
+ * The documentation of ReadDirectoryChangesW() states that the maximum
+ * buffer size is 64K when the monitored directory is remote.
+ *
+ * Larger buffers may be used when the monitored directory is local and
+ * will help us receive events faster from the kernel and avoid dropped
+ * events.
+ *
+ * So we try to use a very large buffer and silently fallback to 64K if
+ * we get an error.
+ */
+#define MAX_RDCW_BUF_FALLBACK (65536)
+#define MAX_RDCW_BUF          (65536 * 8)
+
+struct one_watch
+{
+	char buffer[MAX_RDCW_BUF];
+	DWORD buf_len;
+	DWORD count;
+
+	struct strbuf path;
+	HANDLE hDir;
+	HANDLE hEvent;
+	OVERLAPPED overlapped;
+
+	/*
+	 * Is there an active ReadDirectoryChangesW() call pending.  If so, we
+	 * need to later call GetOverlappedResult() and possibly CancelIoEx().
+	 */
+	BOOL is_active;
+};
+
+struct fsmonitor_daemon_backend_data
+{
+	struct one_watch *watch_worktree;
+	struct one_watch *watch_gitdir;
+
+	HANDLE hEventShutdown;
+
+	HANDLE hListener[3]; /* we don't own these handles */
+#define LISTENER_SHUTDOWN 0
+#define LISTENER_HAVE_DATA_WORKTREE 1
+#define LISTENER_HAVE_DATA_GITDIR 2
+	int nr_listener_handles;
+};
+
+/*
+ * Convert the WCHAR path from the notification into UTF8 and
+ * then normalize it.
+ */
+static int normalize_path_in_utf8(FILE_NOTIFY_INFORMATION *info,
+				  struct strbuf *normalized_path)
+{
+	int reserve;
+	int len = 0;
+
+	strbuf_reset(normalized_path);
+	if (!info->FileNameLength)
+		goto normalize;
+
+	/*
+	 * Pre-reserve enough space in the UTF8 buffer for
+	 * each Unicode WCHAR character to be mapped into a
+	 * sequence of 2 UTF8 characters.  That should let us
+	 * avoid ERROR_INSUFFICIENT_BUFFER 99.9+% of the time.
+	 */
+	reserve = info->FileNameLength + 1;
+	strbuf_grow(normalized_path, reserve);
+
+	for (;;) {
+		len = WideCharToMultiByte(CP_UTF8, 0, info->FileName,
+					  info->FileNameLength / sizeof(WCHAR),
+					  normalized_path->buf,
+					  strbuf_avail(normalized_path) - 1,
+					  NULL, NULL);
+		if (len > 0)
+			goto normalize;
+		if (GetLastError() != ERROR_INSUFFICIENT_BUFFER) {
+			error(_("[GLE %ld] could not convert path to UTF-8: '%.*ls'"),
+			      GetLastError(),
+			      (int)(info->FileNameLength / sizeof(WCHAR)),
+			      info->FileName);
+			return -1;
+		}
+
+		strbuf_grow(normalized_path,
+			    strbuf_avail(normalized_path) + reserve);
+	}
+
+normalize:
+	strbuf_setlen(normalized_path, len);
+	return strbuf_normalize_path(normalized_path);
+}
+
+void fsm_listen__stop_async(struct fsmonitor_daemon_state *state)
+{
+	SetEvent(state->backend_data->hListener[LISTENER_SHUTDOWN]);
+}
+
+static struct one_watch *create_watch(struct fsmonitor_daemon_state *state,
+				      const char *path)
+{
+	struct one_watch *watch = NULL;
+	DWORD desired_access = FILE_LIST_DIRECTORY;
+	DWORD share_mode =
+		FILE_SHARE_WRITE | FILE_SHARE_READ | FILE_SHARE_DELETE;
+	HANDLE hDir;
+	wchar_t wpath[MAX_PATH];
+
+	if (xutftowcs_path(wpath, path) < 0) {
+		error(_("could not convert to wide characters: '%s'"), path);
+		return NULL;
+	}
+
+	hDir = CreateFileW(wpath,
+			   desired_access, share_mode, NULL, OPEN_EXISTING,
+			   FILE_FLAG_BACKUP_SEMANTICS | FILE_FLAG_OVERLAPPED,
+			   NULL);
+	if (hDir == INVALID_HANDLE_VALUE) {
+		error(_("[GLE %ld] could not watch '%s'"),
+		      GetLastError(), path);
+		return NULL;
+	}
+
+	CALLOC_ARRAY(watch, 1);
+
+	watch->buf_len = sizeof(watch->buffer); /* assume full MAX_RDCW_BUF */
+
+	strbuf_init(&watch->path, 0);
+	strbuf_addstr(&watch->path, path);
+
+	watch->hDir = hDir;
+	watch->hEvent = CreateEvent(NULL, TRUE, FALSE, NULL);
+
+	return watch;
+}
+
+static void destroy_watch(struct one_watch *watch)
+{
+	if (!watch)
+		return;
+
+	strbuf_release(&watch->path);
+	if (watch->hDir != INVALID_HANDLE_VALUE)
+		CloseHandle(watch->hDir);
+	if (watch->hEvent != INVALID_HANDLE_VALUE)
+		CloseHandle(watch->hEvent);
+
+	free(watch);
+}
+
+static int start_rdcw_watch(struct fsmonitor_daemon_backend_data *data,
+			    struct one_watch *watch)
+{
+	DWORD dwNotifyFilter =
+		FILE_NOTIFY_CHANGE_FILE_NAME |
+		FILE_NOTIFY_CHANGE_DIR_NAME |
+		FILE_NOTIFY_CHANGE_ATTRIBUTES |
+		FILE_NOTIFY_CHANGE_SIZE |
+		FILE_NOTIFY_CHANGE_LAST_WRITE |
+		FILE_NOTIFY_CHANGE_CREATION;
+
+	ResetEvent(watch->hEvent);
+
+	memset(&watch->overlapped, 0, sizeof(watch->overlapped));
+	watch->overlapped.hEvent = watch->hEvent;
+
+	/*
+	 * Queue an async call using Overlapped IO.  This returns immediately.
+	 * Our event handle will be signalled when the real result is available.
+	 *
+	 * The return value here just means that we successfully queued it.
+	 * We won't know if the Read...() actually produces data until later.
+	 */
+	watch->is_active = ReadDirectoryChangesW(
+		watch->hDir, watch->buffer, watch->buf_len, TRUE,
+		dwNotifyFilter, &watch->count, &watch->overlapped, NULL);
+
+	if (watch->is_active)
+		return 0;
+
+	error(_("ReadDirectoryChangedW failed on '%s' [GLE %ld]"),
+	      watch->path.buf, GetLastError());
+	return -1;
+}
+
+static int recv_rdcw_watch(struct one_watch *watch)
+{
+	DWORD gle;
+
+	watch->is_active = FALSE;
+
+	/*
+	 * The overlapped result is ready.  If the Read...() was successful
+	 * we finally receive the actual result into our buffer.
+	 */
+	if (GetOverlappedResult(watch->hDir, &watch->overlapped, &watch->count,
+				TRUE))
+		return 0;
+
+	gle = GetLastError();
+	if (gle == ERROR_INVALID_PARAMETER &&
+	    /*
+	     * The kernel throws an invalid parameter error when our
+	     * buffer is too big and we are pointed at a remote
+	     * directory (and possibly for other reasons).  Quietly
+	     * set it down and try again.
+	     *
+	     * See note about MAX_RDCW_BUF at the top.
+	     */
+	    watch->buf_len > MAX_RDCW_BUF_FALLBACK) {
+		watch->buf_len = MAX_RDCW_BUF_FALLBACK;
+		return -2;
+	}
+
+	/*
+	 * NEEDSWORK: If an external <gitdir> is deleted, the above
+	 * returns an error.  I'm not sure that there's anything that
+	 * we can do here other than failing -- the <worktree>/.git
+	 * link file would be broken anyway.  We might try to check
+	 * for that and return a better error message, but I'm not
+	 * sure it is worth it.
+	 */
+
+	error(_("GetOverlappedResult failed on '%s' [GLE %ld]"),
+	      watch->path.buf, gle);
+	return -1;
+}
+
+static void cancel_rdcw_watch(struct one_watch *watch)
+{
+	DWORD count;
+
+	if (!watch || !watch->is_active)
+		return;
+
+	/*
+	 * The calls to ReadDirectoryChangesW() and GetOverlappedResult()
+	 * form a "pair" (my term) where we queue an IO and promise to
+	 * hang around and wait for the kernel to give us the result.
+	 *
+	 * If for some reason after we queue the IO, we have to quit
+	 * or otherwise not stick around for the second half, we must
+	 * tell the kernel to abort the IO.  This prevents the kernel
+	 * from writing to our buffer and/or signalling our event
+	 * after we free them.
+	 *
+	 * (Ask me how much fun it was to track that one down).
+	 */
+	CancelIoEx(watch->hDir, &watch->overlapped);
+	GetOverlappedResult(watch->hDir, &watch->overlapped, &count, TRUE);
+	watch->is_active = FALSE;
+}
+
+/*
+ * Process filesystem events that happen anywhere (recursively) under the
+ * <worktree> root directory.  For a normal working directory, this includes
+ * both version controlled files and the contents of the .git/ directory.
+ *
+ * If <worktree>/.git is a file, then we only see events for the file
+ * itself.
+ */
+static int process_worktree_events(struct fsmonitor_daemon_state *state)
+{
+	struct fsmonitor_daemon_backend_data *data = state->backend_data;
+	struct one_watch *watch = data->watch_worktree;
+	struct strbuf path = STRBUF_INIT;
+	struct string_list cookie_list = STRING_LIST_INIT_DUP;
+	struct fsmonitor_batch *batch = NULL;
+	const char *p = watch->buffer;
+
+	/*
+	 * If the kernel gets more events than will fit in the kernel
+	 * buffer associated with our RDCW handle, it drops them and
+	 * returns a count of zero.
+	 *
+	 * Yes, the call returns WITHOUT error and with length zero.
+	 * This is the documented behavior.  (My testing has confirmed
+	 * that it also sets the last error to ERROR_NOTIFY_ENUM_DIR,
+	 * but we do not rely on that since the function did not
+	 * return an error and it is not documented.)
+	 *
+	 * (The "overflow" case is not ambiguous with the "no data" case
+	 * because we did an INFINITE wait.)
+	 *
+	 * This means we have a gap in coverage.  Tell the daemon layer
+	 * to resync.
+	 */
+	if (!watch->count) {
+		trace2_data_string("fsmonitor", NULL, "fsm-listen/kernel",
+				   "overflow");
+		fsmonitor_force_resync(state);
+		return LISTENER_HAVE_DATA_WORKTREE;
+	}
+
+	/*
+	 * On Windows, `info` contains an "array" of paths that are
+	 * relative to the root of whichever directory handle received
+	 * the event.
+	 */
+	for (;;) {
+		FILE_NOTIFY_INFORMATION *info = (void *)p;
+		const char *slash;
+		enum fsmonitor_path_type t;
+
+		strbuf_reset(&path);
+		if (normalize_path_in_utf8(info, &path) == -1)
+			goto skip_this_path;
+
+		t = fsmonitor_classify_path_workdir_relative(path.buf);
+
+		switch (t) {
+		case IS_INSIDE_DOT_GIT_WITH_COOKIE_PREFIX:
+			/* special case cookie files within .git */
+
+			/* Use just the filename of the cookie file. */
+			slash = find_last_dir_sep(path.buf);
+			string_list_append(&cookie_list,
+					   slash ? slash + 1 : path.buf);
+			break;
+
+		case IS_INSIDE_DOT_GIT:
+			/* ignore everything inside of "<worktree>/.git/" */
+			break;
+
+		case IS_DOT_GIT:
+			/* "<worktree>/.git" was deleted (or renamed away) */
+			if ((info->Action == FILE_ACTION_REMOVED) ||
+			    (info->Action == FILE_ACTION_RENAMED_OLD_NAME)) {
+				trace2_data_string("fsmonitor", NULL,
+						   "fsm-listen/dotgit",
+						   "removed");
+				goto force_shutdown;
+			}
+			break;
+
+		case IS_WORKDIR_PATH:
+			/* queue normal pathname */
+			if (!batch)
+				batch = fsmonitor_batch__new();
+			fsmonitor_batch__add_path(batch, path.buf);
+			break;
+
+		case IS_GITDIR:
+		case IS_INSIDE_GITDIR:
+		case IS_INSIDE_GITDIR_WITH_COOKIE_PREFIX:
+		default:
+			BUG("unexpected path classification '%d' for '%s'",
+			    t, path.buf);
+		}
+
+skip_this_path:
+		if (!info->NextEntryOffset)
+			break;
+		p += info->NextEntryOffset;
+	}
+
+	fsmonitor_publish(state, batch, &cookie_list);
+	batch = NULL;
+	string_list_clear(&cookie_list, 0);
+	strbuf_release(&path);
+	return LISTENER_HAVE_DATA_WORKTREE;
+
+force_shutdown:
+	fsmonitor_batch__free_list(batch);
+	string_list_clear(&cookie_list, 0);
+	strbuf_release(&path);
+	return LISTENER_SHUTDOWN;
+}
+
+/*
+ * Process filesystem events that happened anywhere (recursively) under the
+ * external <gitdir> (such as non-primary worktrees or submodules).
+ * We only care about cookie files that our client threads created here.
+ *
+ * Note that we DO NOT get filesystem events on the external <gitdir>
+ * itself (it is not inside something that we are watching).  In particular,
+ * we do not get an event if the external <gitdir> is deleted.
+ */
+static int process_gitdir_events(struct fsmonitor_daemon_state *state)
+{
+	struct fsmonitor_daemon_backend_data *data = state->backend_data;
+	struct one_watch *watch = data->watch_gitdir;
+	struct strbuf path = STRBUF_INIT;
+	struct string_list cookie_list = STRING_LIST_INIT_DUP;
+	const char *p = watch->buffer;
+
+	if (!watch->count) {
+		trace2_data_string("fsmonitor", NULL, "fsm-listen/kernel",
+				   "overflow");
+		fsmonitor_force_resync(state);
+		return LISTENER_HAVE_DATA_GITDIR;
+	}
+
+	for (;;) {
+		FILE_NOTIFY_INFORMATION *info = (void *)p;
+		const char *slash;
+		enum fsmonitor_path_type t;
+
+		strbuf_reset(&path);
+		if (normalize_path_in_utf8(info, &path) == -1)
+			goto skip_this_path;
+
+		t = fsmonitor_classify_path_gitdir_relative(path.buf);
+
+		switch (t) {
+		case IS_INSIDE_GITDIR_WITH_COOKIE_PREFIX:
+			/* special case cookie files within gitdir */
+
+			/* Use just the filename of the cookie file. */
+			slash = find_last_dir_sep(path.buf);
+			string_list_append(&cookie_list,
+					   slash ? slash + 1 : path.buf);
+			break;
+
+		case IS_INSIDE_GITDIR:
+			goto skip_this_path;
+
+		default:
+			BUG("unexpected path classification '%d' for '%s'",
+			    t, path.buf);
+		}
+
+skip_this_path:
+		if (!info->NextEntryOffset)
+			break;
+		p += info->NextEntryOffset;
+	}
+
+	fsmonitor_publish(state, NULL, &cookie_list);
+	string_list_clear(&cookie_list, 0);
+	strbuf_release(&path);
+	return LISTENER_HAVE_DATA_GITDIR;
+}
+
+void fsm_listen__loop(struct fsmonitor_daemon_state *state)
+{
+	struct fsmonitor_daemon_backend_data *data = state->backend_data;
+	DWORD dwWait;
+	int result;
+
+	state->error_code = 0;
+
+	if (start_rdcw_watch(data, data->watch_worktree) == -1)
+		goto force_error_stop;
+
+	if (data->watch_gitdir &&
+	    start_rdcw_watch(data, data->watch_gitdir) == -1)
+		goto force_error_stop;
+
+	for (;;) {
+		dwWait = WaitForMultipleObjects(data->nr_listener_handles,
+						data->hListener,
+						FALSE, INFINITE);
+
+		if (dwWait == WAIT_OBJECT_0 + LISTENER_HAVE_DATA_WORKTREE) {
+			result = recv_rdcw_watch(data->watch_worktree);
+			if (result == -1) {
+				/* hard error */
+				goto force_error_stop;
+			}
+			if (result == -2) {
+				/* retryable error */
+				if (start_rdcw_watch(data, data->watch_worktree) == -1)
+					goto force_error_stop;
+				continue;
+			}
+
+			/* have data */
+			if (process_worktree_events(state) == LISTENER_SHUTDOWN)
+				goto force_shutdown;
+			if (start_rdcw_watch(data, data->watch_worktree) == -1)
+				goto force_error_stop;
+			continue;
+		}
+
+		if (dwWait == WAIT_OBJECT_0 + LISTENER_HAVE_DATA_GITDIR) {
+			result = recv_rdcw_watch(data->watch_gitdir);
+			if (result == -1) {
+				/* hard error */
+				goto force_error_stop;
+			}
+			if (result == -2) {
+				/* retryable error */
+				if (start_rdcw_watch(data, data->watch_gitdir) == -1)
+					goto force_error_stop;
+				continue;
+			}
+
+			/* have data */
+			if (process_gitdir_events(state) == LISTENER_SHUTDOWN)
+				goto force_shutdown;
+			if (start_rdcw_watch(data, data->watch_gitdir) == -1)
+				goto force_error_stop;
+			continue;
+		}
+
+		if (dwWait == WAIT_OBJECT_0 + LISTENER_SHUTDOWN)
+			goto clean_shutdown;
+
+		error(_("could not read directory changes [GLE %ld]"),
+		      GetLastError());
+		goto force_error_stop;
+	}
+
+force_error_stop:
+	state->error_code = -1;
+
+force_shutdown:
+	/*
+	 * Tell the IPC thead pool to stop (which completes the await
+	 * in the main thread (which will also signal this thread (if
+	 * we are still alive))).
+	 */
+	ipc_server_stop_async(state->ipc_server_data);
+
+clean_shutdown:
+	cancel_rdcw_watch(data->watch_worktree);
+	cancel_rdcw_watch(data->watch_gitdir);
+}
+
+int fsm_listen__ctor(struct fsmonitor_daemon_state *state)
+{
+	struct fsmonitor_daemon_backend_data *data;
+
+	CALLOC_ARRAY(data, 1);
+
+	data->hEventShutdown = CreateEvent(NULL, TRUE, FALSE, NULL);
+
+	data->watch_worktree = create_watch(state,
+					    state->path_worktree_watch.buf);
+	if (!data->watch_worktree)
+		goto failed;
+
+	if (state->nr_paths_watching > 1) {
+		data->watch_gitdir = create_watch(state,
+						  state->path_gitdir_watch.buf);
+		if (!data->watch_gitdir)
+			goto failed;
+	}
+
+	data->hListener[LISTENER_SHUTDOWN] = data->hEventShutdown;
+	data->nr_listener_handles++;
+
+	data->hListener[LISTENER_HAVE_DATA_WORKTREE] =
+		data->watch_worktree->hEvent;
+	data->nr_listener_handles++;
+
+	if (data->watch_gitdir) {
+		data->hListener[LISTENER_HAVE_DATA_GITDIR] =
+			data->watch_gitdir->hEvent;
+		data->nr_listener_handles++;
+	}
+
+	state->backend_data = data;
+	return 0;
+
+failed:
+	CloseHandle(data->hEventShutdown);
+	destroy_watch(data->watch_worktree);
+	destroy_watch(data->watch_gitdir);
+
+	return -1;
+}
+
+void fsm_listen__dtor(struct fsmonitor_daemon_state *state)
+{
+	struct fsmonitor_daemon_backend_data *data;
+
+	if (!state || !state->backend_data)
+		return;
+
+	data = state->backend_data;
+
+	CloseHandle(data->hEventShutdown);
+	destroy_watch(data->watch_worktree);
+	destroy_watch(data->watch_gitdir);
+
+	FREE_AND_NULL(state->backend_data);
+}
diff --git a/compat/fsmonitor/fsm-listen.h b/compat/fsmonitor/fsm-listen.h
new file mode 100644
index 0000000..f053934
--- /dev/null
+++ b/compat/fsmonitor/fsm-listen.h
@@ -0,0 +1,49 @@
+#ifndef FSM_LISTEN_H
+#define FSM_LISTEN_H
+
+/* This needs to be implemented by each backend */
+
+#ifdef HAVE_FSMONITOR_DAEMON_BACKEND
+
+struct fsmonitor_daemon_state;
+
+/*
+ * Initialize platform-specific data for the fsmonitor listener thread.
+ * This will be called from the main thread PRIOR to staring the
+ * fsmonitor_fs_listener thread.
+ *
+ * Returns 0 if successful.
+ * Returns -1 otherwise.
+ */
+int fsm_listen__ctor(struct fsmonitor_daemon_state *state);
+
+/*
+ * Cleanup platform-specific data for the fsmonitor listener thread.
+ * This will be called from the main thread AFTER joining the listener.
+ */
+void fsm_listen__dtor(struct fsmonitor_daemon_state *state);
+
+/*
+ * The main body of the platform-specific event loop to watch for
+ * filesystem events.  This will run in the fsmonitor_fs_listen thread.
+ *
+ * It should call `ipc_server_stop_async()` if the listener thread
+ * prematurely terminates (because of a filesystem error or if it
+ * detects that the .git directory has been deleted).  (It should NOT
+ * do so if the listener thread receives a normal shutdown signal from
+ * the IPC layer.)
+ *
+ * It should set `state->error_code` to -1 if the daemon should exit
+ * with an error.
+ */
+void fsm_listen__loop(struct fsmonitor_daemon_state *state);
+
+/*
+ * Gently request that the fsmonitor listener thread shutdown.
+ * It does not wait for it to stop.  The caller should do a JOIN
+ * to wait for it.
+ */
+void fsm_listen__stop_async(struct fsmonitor_daemon_state *state);
+
+#endif /* HAVE_FSMONITOR_DAEMON_BACKEND */
+#endif /* FSM_LISTEN_H */
diff --git a/compat/mingw.c b/compat/mingw.c
index 5e42191..6fe80fd 100644
--- a/compat/mingw.c
+++ b/compat/mingw.c
@@ -9,6 +9,8 @@
 #include "win32/lazyload.h"
 #include "../config.h"
 #include "dir.h"
+#define SECURITY_WIN32
+#include <sspi.h>
 
 #define HCAST(type, handle) ((type)(intptr_t)handle)
 
@@ -960,9 +962,11 @@ static inline void time_t_to_filetime(time_t t, FILETIME *ft)
 int mingw_utime (const char *file_name, const struct utimbuf *times)
 {
 	FILETIME mft, aft;
-	int fh, rc;
+	int rc;
 	DWORD attrs;
 	wchar_t wfilename[MAX_PATH];
+	HANDLE osfilehandle;
+
 	if (xutftowcs_path(wfilename, file_name) < 0)
 		return -1;
 
@@ -974,7 +978,17 @@ int mingw_utime (const char *file_name, const struct utimbuf *times)
 		SetFileAttributesW(wfilename, attrs & ~FILE_ATTRIBUTE_READONLY);
 	}
 
-	if ((fh = _wopen(wfilename, O_RDWR | O_BINARY)) < 0) {
+	osfilehandle = CreateFileW(wfilename,
+				   FILE_WRITE_ATTRIBUTES,
+				   0 /*FileShare.None*/,
+				   NULL,
+				   OPEN_EXISTING,
+				   (attrs != INVALID_FILE_ATTRIBUTES &&
+					(attrs & FILE_ATTRIBUTE_DIRECTORY)) ?
+					FILE_FLAG_BACKUP_SEMANTICS : 0,
+				   NULL);
+	if (osfilehandle == INVALID_HANDLE_VALUE) {
+		errno = err_win_to_posix(GetLastError());
 		rc = -1;
 		goto revert_attrs;
 	}
@@ -986,12 +1000,15 @@ int mingw_utime (const char *file_name, const struct utimbuf *times)
 		GetSystemTimeAsFileTime(&mft);
 		aft = mft;
 	}
-	if (!SetFileTime((HANDLE)_get_osfhandle(fh), NULL, &aft, &mft)) {
+
+	if (!SetFileTime(osfilehandle, NULL, &aft, &mft)) {
 		errno = EINVAL;
 		rc = -1;
 	} else
 		rc = 0;
-	close(fh);
+
+	if (osfilehandle != INVALID_HANDLE_VALUE)
+		CloseHandle(osfilehandle);
 
 revert_attrs:
 	if (attrs != INVALID_FILE_ATTRIBUTES &&
@@ -1009,7 +1026,7 @@ size_t mingw_strftime(char *s, size_t max,
 	/* a pointer to the original strftime in case we can't find the UCRT version */
 	static size_t (*fallback)(char *, size_t, const char *, const struct tm *) = strftime;
 	size_t ret;
-	DECLARE_PROC_ADDR(ucrtbase.dll, size_t, strftime, char *, size_t,
+	DECLARE_PROC_ADDR(ucrtbase.dll, size_t, __cdecl, strftime, char *, size_t,
 		const char *, const struct tm *);
 
 	if (INIT_PROC_ADDR(strftime))
@@ -1126,6 +1143,10 @@ char *mingw_getcwd(char *pointer, int len)
 	}
 	if (!ret || ret >= ARRAY_SIZE(wpointer))
 		return NULL;
+	if (GetFileAttributesW(wpointer) == INVALID_FILE_ATTRIBUTES) {
+		errno = ENOENT;
+		return NULL;
+	}
 	if (xwcstoutf(pointer, wpointer, len) < 0)
 		return NULL;
 	convert_slashes(pointer);
@@ -2186,7 +2207,7 @@ enum EXTENDED_NAME_FORMAT {
 
 static char *get_extended_user_info(enum EXTENDED_NAME_FORMAT type)
 {
-	DECLARE_PROC_ADDR(secur32.dll, BOOL, GetUserNameExW,
+	DECLARE_PROC_ADDR(secur32.dll, BOOL, SEC_ENTRY, GetUserNameExW,
 		enum EXTENDED_NAME_FORMAT, LPCWSTR, PULONG);
 	static wchar_t wbuffer[1024];
 	DWORD len;
diff --git a/compat/mingw.h b/compat/mingw.h
index ffa53a4..494cc8d 100644
--- a/compat/mingw.h
+++ b/compat/mingw.h
@@ -329,6 +329,9 @@ int mingw_getpagesize(void);
 #define getpagesize mingw_getpagesize
 #endif
 
+int win32_fsync_no_flush(int fd);
+#define fsync_no_flush win32_fsync_no_flush
+
 struct rlimit {
 	unsigned int rlim_cur;
 };
diff --git a/compat/qsort_s.c b/compat/qsort_s.c
index 52d1f0a..0f7ff30 100644
--- a/compat/qsort_s.c
+++ b/compat/qsort_s.c
@@ -49,21 +49,15 @@ int git_qsort_s(void *b, size_t n, size_t s,
 		int (*cmp)(const void *, const void *, void *), void *ctx)
 {
 	const size_t size = st_mult(n, s);
-	char buf[1024];
+	char *tmp;
 
 	if (!n)
 		return 0;
 	if (!b || !cmp)
 		return -1;
 
-	if (size < sizeof(buf)) {
-		/* The temporary array fits on the small on-stack buffer. */
-		msort_with_tmp(b, n, s, cmp, buf, ctx);
-	} else {
-		/* It's somewhat large, so malloc it.  */
-		char *tmp = xmalloc(size);
-		msort_with_tmp(b, n, s, cmp, tmp, ctx);
-		free(tmp);
-	}
+	tmp = xmalloc(size);
+	msort_with_tmp(b, n, s, cmp, tmp, ctx);
+	free(tmp);
 	return 0;
 }
diff --git a/compat/terminal.c b/compat/terminal.c
index 5b903e7..7db330c 100644
--- a/compat/terminal.c
+++ b/compat/terminal.c
@@ -1,4 +1,4 @@
-#include "git-compat-util.h"
+#include "cache.h"
 #include "compat/terminal.h"
 #include "sigchain.h"
 #include "strbuf.h"
@@ -11,7 +11,7 @@
 static void restore_term_on_signal(int sig)
 {
 	restore_term();
-	sigchain_pop(sig);
+	/* restore_term calls sigchain_pop_common */
 	raise(sig);
 }
 
@@ -20,55 +20,227 @@ static void restore_term_on_signal(int sig)
 #define INPUT_PATH "/dev/tty"
 #define OUTPUT_PATH "/dev/tty"
 
+static volatile sig_atomic_t term_fd_needs_closing;
 static int term_fd = -1;
 static struct termios old_term;
 
+static const char *background_resume_msg;
+static const char *restore_error_msg;
+static volatile sig_atomic_t ttou_received;
+
+/* async safe error function for use by signal handlers. */
+static void write_err(const char *msg)
+{
+	write_in_full(2, "error: ", strlen("error: "));
+	write_in_full(2, msg, strlen(msg));
+	write_in_full(2, "\n", 1);
+}
+
+static void print_background_resume_msg(int signo)
+{
+	int saved_errno = errno;
+	sigset_t mask;
+	struct sigaction old_sa;
+	struct sigaction sa = { .sa_handler = SIG_DFL };
+
+	ttou_received = 1;
+	write_err(background_resume_msg);
+	sigaction(signo, &sa, &old_sa);
+	raise(signo);
+	sigemptyset(&mask);
+	sigaddset(&mask, signo);
+	sigprocmask(SIG_UNBLOCK, &mask, NULL);
+	/* Stopped here */
+	sigprocmask(SIG_BLOCK, &mask, NULL);
+	sigaction(signo, &old_sa, NULL);
+	errno = saved_errno;
+}
+
+static void restore_terminal_on_suspend(int signo)
+{
+	int saved_errno = errno;
+	int res;
+	struct termios t;
+	sigset_t mask;
+	struct sigaction old_sa;
+	struct sigaction sa = { .sa_handler = SIG_DFL };
+	int can_restore = 1;
+
+	if (tcgetattr(term_fd, &t) < 0)
+		can_restore = 0;
+
+	if (tcsetattr(term_fd, TCSAFLUSH, &old_term) < 0)
+		write_err(restore_error_msg);
+
+	sigaction(signo, &sa, &old_sa);
+	raise(signo);
+	sigemptyset(&mask);
+	sigaddset(&mask, signo);
+	sigprocmask(SIG_UNBLOCK, &mask, NULL);
+	/* Stopped here */
+	sigprocmask(SIG_BLOCK, &mask, NULL);
+	sigaction(signo, &old_sa, NULL);
+	if (!can_restore) {
+		write_err(restore_error_msg);
+		goto out;
+	}
+	/*
+	 * If we resume in the background then we receive SIGTTOU when calling
+	 * tcsetattr() below. Set up a handler to print an error message in that
+	 * case.
+	 */
+	sigemptyset(&mask);
+	sigaddset(&mask, SIGTTOU);
+	sa.sa_mask = old_sa.sa_mask;
+	sa.sa_handler = print_background_resume_msg;
+	sa.sa_flags = SA_RESTART;
+	sigaction(SIGTTOU, &sa, &old_sa);
+ again:
+	ttou_received = 0;
+	sigprocmask(SIG_UNBLOCK, &mask, NULL);
+	res = tcsetattr(term_fd, TCSAFLUSH, &t);
+	sigprocmask(SIG_BLOCK, &mask, NULL);
+	if (ttou_received)
+		goto again;
+	else if (res < 0)
+		write_err(restore_error_msg);
+	sigaction(SIGTTOU, &old_sa, NULL);
+ out:
+	errno = saved_errno;
+}
+
+static void reset_job_signals(void)
+{
+	if (restore_error_msg) {
+		signal(SIGTTIN, SIG_DFL);
+		signal(SIGTTOU, SIG_DFL);
+		signal(SIGTSTP, SIG_DFL);
+		restore_error_msg = NULL;
+		background_resume_msg = NULL;
+	}
+}
+
+static void close_term_fd(void)
+{
+	if (term_fd_needs_closing)
+		close(term_fd);
+	term_fd_needs_closing = 0;
+	term_fd = -1;
+}
+
 void restore_term(void)
 {
 	if (term_fd < 0)
 		return;
 
 	tcsetattr(term_fd, TCSAFLUSH, &old_term);
-	close(term_fd);
-	term_fd = -1;
+	close_term_fd();
+	sigchain_pop_common();
+	reset_job_signals();
 }
 
-int save_term(int full_duplex)
+int save_term(enum save_term_flags flags)
 {
+	struct sigaction sa;
+
 	if (term_fd < 0)
-		term_fd = open("/dev/tty", O_RDWR);
+		term_fd = ((flags & SAVE_TERM_STDIN)
+			   ? 0
+			   : open("/dev/tty", O_RDWR));
+	if (term_fd < 0)
+		return -1;
+	term_fd_needs_closing = !(flags & SAVE_TERM_STDIN);
+	if (tcgetattr(term_fd, &old_term) < 0) {
+		close_term_fd();
+		return -1;
+	}
+	sigchain_push_common(restore_term_on_signal);
+	/*
+	 * If job control is disabled then the shell will have set the
+	 * disposition of SIGTSTP to SIG_IGN.
+	 */
+	sigaction(SIGTSTP, NULL, &sa);
+	if (sa.sa_handler == SIG_IGN)
+		return 0;
+
+	/* avoid calling gettext() from signal handler */
+	background_resume_msg = _("cannot resume in the background, please use 'fg' to resume");
+	restore_error_msg = _("cannot restore terminal settings");
+	sa.sa_handler = restore_terminal_on_suspend;
+	sa.sa_flags = SA_RESTART;
+	sigemptyset(&sa.sa_mask);
+	sigaddset(&sa.sa_mask, SIGTSTP);
+	sigaddset(&sa.sa_mask, SIGTTIN);
+	sigaddset(&sa.sa_mask, SIGTTOU);
+	sigaction(SIGTSTP, &sa, NULL);
+	sigaction(SIGTTIN, &sa, NULL);
+	sigaction(SIGTTOU, &sa, NULL);
 
-	return (term_fd < 0) ? -1 : tcgetattr(term_fd, &old_term);
+	return 0;
 }
 
-static int disable_bits(tcflag_t bits)
+static int disable_bits(enum save_term_flags flags, tcflag_t bits)
 {
 	struct termios t;
 
-	if (save_term(0) < 0)
-		goto error;
+	if (save_term(flags) < 0)
+		return -1;
 
 	t = old_term;
-	sigchain_push_common(restore_term_on_signal);
 
 	t.c_lflag &= ~bits;
+	if (bits & ICANON) {
+		t.c_cc[VMIN] = 1;
+		t.c_cc[VTIME] = 0;
+	}
 	if (!tcsetattr(term_fd, TCSAFLUSH, &t))
 		return 0;
 
-error:
-	close(term_fd);
-	term_fd = -1;
+	sigchain_pop_common();
+	reset_job_signals();
+	close_term_fd();
 	return -1;
 }
 
-static int disable_echo(void)
+static int disable_echo(enum save_term_flags flags)
 {
-	return disable_bits(ECHO);
+	return disable_bits(flags, ECHO);
 }
 
-static int enable_non_canonical(void)
+static int enable_non_canonical(enum save_term_flags flags)
 {
-	return disable_bits(ICANON | ECHO);
+	return disable_bits(flags, ICANON | ECHO);
+}
+
+/*
+ * On macos it is not possible to use poll() with a terminal so use select
+ * instead.
+ */
+static int getchar_with_timeout(int timeout)
+{
+	struct timeval tv, *tvp = NULL;
+	fd_set readfds;
+	int res;
+
+ again:
+	if (timeout >= 0) {
+		tv.tv_sec = timeout / 1000;
+		tv.tv_usec = (timeout % 1000) * 1000;
+		tvp = &tv;
+	}
+
+	FD_ZERO(&readfds);
+	FD_SET(0, &readfds);
+	res = select(1, &readfds, NULL, NULL, tvp);
+	if (!res)
+		return EOF;
+	if (res < 0) {
+		if (errno == EINTR)
+			goto again;
+		else
+			return EOF;
+	}
+	return getchar();
 }
 
 #elif defined(GIT_WINDOWS_NATIVE)
@@ -100,6 +272,8 @@ void restore_term(void)
 		return;
 	}
 
+	sigchain_pop_common();
+
 	if (hconin == INVALID_HANDLE_VALUE)
 		return;
 
@@ -114,7 +288,7 @@ void restore_term(void)
 	hconin = hconout = INVALID_HANDLE_VALUE;
 }
 
-int save_term(int full_duplex)
+int save_term(enum save_term_flags flags)
 {
 	hconin = CreateFileA("CONIN$", GENERIC_READ | GENERIC_WRITE,
 	    FILE_SHARE_READ, NULL, OPEN_EXISTING,
@@ -122,7 +296,7 @@ int save_term(int full_duplex)
 	if (hconin == INVALID_HANDLE_VALUE)
 		return -1;
 
-	if (full_duplex) {
+	if (flags & SAVE_TERM_DUPLEX) {
 		hconout = CreateFileA("CONOUT$", GENERIC_READ | GENERIC_WRITE,
 			FILE_SHARE_WRITE, NULL, OPEN_EXISTING,
 			FILE_ATTRIBUTE_NORMAL, NULL);
@@ -134,6 +308,7 @@ int save_term(int full_duplex)
 
 	GetConsoleMode(hconin, &cmode_in);
 	use_stty = 0;
+	sigchain_push_common(restore_term_on_signal);
 	return 0;
 error:
 	CloseHandle(hconin);
@@ -141,7 +316,7 @@ error:
 	return -1;
 }
 
-static int disable_bits(DWORD bits)
+static int disable_bits(enum save_term_flags flags, DWORD bits)
 {
 	if (use_stty) {
 		struct child_process cp = CHILD_PROCESS_INIT;
@@ -150,7 +325,11 @@ static int disable_bits(DWORD bits)
 
 		if (bits & ENABLE_LINE_INPUT) {
 			string_list_append(&stty_restore, "icanon");
-			strvec_push(&cp.args, "-icanon");
+			/*
+			 * POSIX allows VMIN and VTIME to overlap with VEOF and
+			 * VEOL - let's hope that is not the case on windows.
+			 */
+			strvec_pushl(&cp.args, "-icanon", "min", "1", "time", "0", NULL);
 		}
 
 		if (bits & ENABLE_ECHO_INPUT) {
@@ -174,27 +353,28 @@ static int disable_bits(DWORD bits)
 		use_stty = 0;
 	}
 
-	if (save_term(0) < 0)
+	if (save_term(flags) < 0)
 		return -1;
 
-	sigchain_push_common(restore_term_on_signal);
 	if (!SetConsoleMode(hconin, cmode_in & ~bits)) {
 		CloseHandle(hconin);
 		hconin = INVALID_HANDLE_VALUE;
+		sigchain_pop_common();
 		return -1;
 	}
 
 	return 0;
 }
 
-static int disable_echo(void)
+static int disable_echo(enum save_term_flags flags)
 {
-	return disable_bits(ENABLE_ECHO_INPUT);
+	return disable_bits(flags, ENABLE_ECHO_INPUT);
 }
 
-static int enable_non_canonical(void)
+static int enable_non_canonical(enum save_term_flags flags)
 {
-	return disable_bits(ENABLE_ECHO_INPUT | ENABLE_LINE_INPUT | ENABLE_PROCESSED_INPUT);
+	return disable_bits(flags,
+			    ENABLE_ECHO_INPUT | ENABLE_LINE_INPUT | ENABLE_PROCESSED_INPUT);
 }
 
 /*
@@ -228,6 +408,16 @@ static int mingw_getchar(void)
 }
 #define getchar mingw_getchar
 
+static int getchar_with_timeout(int timeout)
+{
+	struct pollfd pfd = { .fd = 0, .events = POLLIN };
+
+	if (poll(&pfd, 1, timeout) < 1)
+		return EOF;
+
+	return getchar();
+}
+
 #endif
 
 #ifndef FORCE_TEXT
@@ -250,7 +440,7 @@ char *git_terminal_prompt(const char *prompt, int echo)
 		return NULL;
 	}
 
-	if (!echo && disable_echo()) {
+	if (!echo && disable_echo(0)) {
 		fclose(input_fh);
 		fclose(output_fh);
 		return NULL;
@@ -344,7 +534,7 @@ int read_key_without_echo(struct strbuf *buf)
 	static int warning_displayed;
 	int ch;
 
-	if (warning_displayed || enable_non_canonical() < 0) {
+	if (warning_displayed || enable_non_canonical(SAVE_TERM_STDIN) < 0) {
 		if (!warning_displayed) {
 			warning("reading single keystrokes not supported on "
 				"this platform; reading line instead");
@@ -378,14 +568,9 @@ int read_key_without_echo(struct strbuf *buf)
 		 * half a second when we know that the sequence is complete.
 		 */
 		while (!is_known_escape_sequence(buf->buf)) {
-			struct pollfd pfd = { .fd = 0, .events = POLLIN };
-
-			if (poll(&pfd, 1, 500) < 1)
-				break;
-
-			ch = getchar();
+			ch = getchar_with_timeout(500);
 			if (ch == EOF)
-				return 0;
+				break;
 			strbuf_addch(buf, ch);
 		}
 	}
@@ -396,10 +581,10 @@ int read_key_without_echo(struct strbuf *buf)
 
 #else
 
-int save_term(int full_duplex)
+int save_term(enum save_term_flags flags)
 {
-	/* full_duplex == 1, but no support available */
-	return -full_duplex;
+	/* no duplex support available */
+	return -!!(flags & SAVE_TERM_DUPLEX);
 }
 
 void restore_term(void)
diff --git a/compat/terminal.h b/compat/terminal.h
index e1770c5..79ed00c 100644
--- a/compat/terminal.h
+++ b/compat/terminal.h
@@ -1,7 +1,22 @@
 #ifndef COMPAT_TERMINAL_H
 #define COMPAT_TERMINAL_H
 
-int save_term(int full_duplex);
+enum save_term_flags {
+	/* Save input and output settings */
+	SAVE_TERM_DUPLEX = 1 << 0,
+	/* Save stdin rather than /dev/tty (fails if stdin is not a terminal) */
+	SAVE_TERM_STDIN  = 1 << 1,
+};
+
+/*
+ * Save the terminal attributes so they can be restored later by a
+ * call to restore_term(). Note that every successful call to
+ * save_term() must be matched by a call to restore_term() even if the
+ * attributes have not been changed. Returns 0 on success, -1 on
+ * failure.
+ */
+int save_term(enum save_term_flags flags);
+/* Restore the terminal attributes that were saved with save_term() */
 void restore_term(void);
 
 char *git_terminal_prompt(const char *prompt, int echo);
diff --git a/compat/unsetenv.c b/compat/unsetenv.c
index bf5fd70..b9d34af 100644
--- a/compat/unsetenv.c
+++ b/compat/unsetenv.c
@@ -1,6 +1,6 @@
 #include "../git-compat-util.h"
 
-void gitunsetenv (const char *name)
+int gitunsetenv(const char *name)
 {
 #if !defined(__MINGW32__)
      extern char **environ;
@@ -24,4 +24,6 @@ void gitunsetenv (const char *name)
 	  ++dst;
      }
      environ[dst] = NULL;
+
+     return 0;
 }
diff --git a/compat/win32/flush.c b/compat/win32/flush.c
new file mode 100644
index 0000000..291f90e
--- /dev/null
+++ b/compat/win32/flush.c
@@ -0,0 +1,28 @@
+#include "git-compat-util.h"
+#include <winternl.h>
+#include "lazyload.h"
+
+int win32_fsync_no_flush(int fd)
+{
+       IO_STATUS_BLOCK io_status;
+
+#define FLUSH_FLAGS_FILE_DATA_ONLY 1
+
+       DECLARE_PROC_ADDR(ntdll.dll, NTSTATUS, NTAPI, NtFlushBuffersFileEx,
+			 HANDLE FileHandle, ULONG Flags, PVOID Parameters, ULONG ParameterSize,
+			 PIO_STATUS_BLOCK IoStatusBlock);
+
+       if (!INIT_PROC_ADDR(NtFlushBuffersFileEx)) {
+		errno = ENOSYS;
+		return -1;
+       }
+
+       memset(&io_status, 0, sizeof(io_status));
+       if (NtFlushBuffersFileEx((HANDLE)_get_osfhandle(fd), FLUSH_FLAGS_FILE_DATA_ONLY,
+				NULL, 0, &io_status)) {
+		errno = EINVAL;
+		return -1;
+       }
+
+       return 0;
+}
diff --git a/compat/win32/lazyload.h b/compat/win32/lazyload.h
index 2b36371..f2bb96c 100644
--- a/compat/win32/lazyload.h
+++ b/compat/win32/lazyload.h
@@ -4,7 +4,7 @@
 /*
  * A pair of macros to simplify loading of DLL functions. Example:
  *
- *   DECLARE_PROC_ADDR(kernel32.dll, BOOL, CreateHardLinkW,
+ *   DECLARE_PROC_ADDR(kernel32.dll, BOOL, WINAPI, CreateHardLinkW,
  *                     LPCWSTR, LPCWSTR, LPSECURITY_ATTRIBUTES);
  *
  *   if (!INIT_PROC_ADDR(CreateHardLinkW))
@@ -25,10 +25,10 @@ struct proc_addr {
 };
 
 /* Declares a function to be loaded dynamically from a DLL. */
-#define DECLARE_PROC_ADDR(dll, rettype, function, ...) \
+#define DECLARE_PROC_ADDR(dll, rettype, convention, function, ...) \
 	static struct proc_addr proc_addr_##function = \
 	{ #dll, #function, NULL, 0 }; \
-	typedef rettype (WINAPI *proc_type_##function)(__VA_ARGS__); \
+	typedef rettype (convention *proc_type_##function)(__VA_ARGS__); \
 	static proc_type_##function function
 
 /*
diff --git a/compat/win32/trace2_win32_process_info.c b/compat/win32/trace2_win32_process_info.c
index 8ccbd1c..a53fd92 100644
--- a/compat/win32/trace2_win32_process_info.c
+++ b/compat/win32/trace2_win32_process_info.c
@@ -143,8 +143,8 @@ static void get_is_being_debugged(void)
  */
 static void get_peak_memory_info(void)
 {
-	DECLARE_PROC_ADDR(psapi.dll, BOOL, GetProcessMemoryInfo, HANDLE,
-			  PPROCESS_MEMORY_COUNTERS, DWORD);
+	DECLARE_PROC_ADDR(psapi.dll, BOOL, WINAPI, GetProcessMemoryInfo,
+			  HANDLE, PPROCESS_MEMORY_COUNTERS, DWORD);
 
 	if (INIT_PROC_ADDR(GetProcessMemoryInfo)) {
 		PROCESS_MEMORY_COUNTERS pmc;
diff --git a/compat/winansi.c b/compat/winansi.c
index c27b20a..3abe8dd 100644
--- a/compat/winansi.c
+++ b/compat/winansi.c
@@ -3,6 +3,7 @@
  */
 
 #undef NOGDI
+
 #include "../git-compat-util.h"
 #include <wingdi.h>
 #include <winreg.h>
@@ -45,8 +46,9 @@ typedef struct _CONSOLE_FONT_INFOEX {
 static void warn_if_raster_font(void)
 {
 	DWORD fontFamily = 0;
-	DECLARE_PROC_ADDR(kernel32.dll, BOOL, GetCurrentConsoleFontEx,
-			HANDLE, BOOL, PCONSOLE_FONT_INFOEX);
+	DECLARE_PROC_ADDR(kernel32.dll, BOOL, WINAPI,
+			GetCurrentConsoleFontEx, HANDLE, BOOL,
+			PCONSOLE_FONT_INFOEX);
 
 	/* don't bother if output was ascii only */
 	if (!non_ascii_used)
diff --git a/compat/zlib-uncompress2.c b/compat/zlib-uncompress2.c
new file mode 100644
index 0000000..77a1b08
--- /dev/null
+++ b/compat/zlib-uncompress2.c
@@ -0,0 +1,96 @@
+#include "git-compat-util.h"
+
+#if ZLIB_VERNUM < 0x1290
+/* taken from zlib's uncompr.c
+
+   commit cacf7f1d4e3d44d871b605da3b647f07d718623f
+   Author: Mark Adler <madler@alumni.caltech.edu>
+   Date:   Sun Jan 15 09:18:46 2017 -0800
+
+       zlib 1.2.11
+
+*/
+
+/*
+ * Copyright (C) 1995-2003, 2010, 2014, 2016 Jean-loup Gailly, Mark Adler
+ * For conditions of distribution and use, see copyright notice in zlib.h
+ */
+
+/* clang-format off */
+
+/* ===========================================================================
+     Decompresses the source buffer into the destination buffer.  *sourceLen is
+   the byte length of the source buffer. Upon entry, *destLen is the total size
+   of the destination buffer, which must be large enough to hold the entire
+   uncompressed data. (The size of the uncompressed data must have been saved
+   previously by the compressor and transmitted to the decompressor by some
+   mechanism outside the scope of this compression library.) Upon exit,
+   *destLen is the size of the decompressed data and *sourceLen is the number
+   of source bytes consumed. Upon return, source + *sourceLen points to the
+   first unused input byte.
+
+     uncompress returns Z_OK if success, Z_MEM_ERROR if there was not enough
+   memory, Z_BUF_ERROR if there was not enough room in the output buffer, or
+   Z_DATA_ERROR if the input data was corrupted, including if the input data is
+   an incomplete zlib stream.
+*/
+int ZEXPORT uncompress2 (
+    Bytef *dest,
+    uLongf *destLen,
+    const Bytef *source,
+    uLong *sourceLen) {
+    z_stream stream;
+    int err;
+    const uInt max = (uInt)-1;
+    uLong len, left;
+    Byte buf[1];    /* for detection of incomplete stream when *destLen == 0 */
+
+    len = *sourceLen;
+    if (*destLen) {
+	left = *destLen;
+	*destLen = 0;
+    }
+    else {
+	left = 1;
+	dest = buf;
+    }
+
+    stream.next_in = (z_const Bytef *)source;
+    stream.avail_in = 0;
+    stream.zalloc = (alloc_func)0;
+    stream.zfree = (free_func)0;
+    stream.opaque = (voidpf)0;
+
+    err = inflateInit(&stream);
+    if (err != Z_OK) return err;
+
+    stream.next_out = dest;
+    stream.avail_out = 0;
+
+    do {
+	if (stream.avail_out == 0) {
+	    stream.avail_out = left > (uLong)max ? max : (uInt)left;
+	    left -= stream.avail_out;
+	}
+	if (stream.avail_in == 0) {
+	    stream.avail_in = len > (uLong)max ? max : (uInt)len;
+	    len -= stream.avail_in;
+	}
+	err = inflate(&stream, Z_NO_FLUSH);
+    } while (err == Z_OK);
+
+    *sourceLen -= len + stream.avail_in;
+    if (dest != buf)
+	*destLen = stream.total_out;
+    else if (stream.total_out && err == Z_BUF_ERROR)
+	left = 1;
+
+    inflateEnd(&stream);
+    return err == Z_STREAM_END ? Z_OK :
+	   err == Z_NEED_DICT ? Z_DATA_ERROR  :
+	   err == Z_BUF_ERROR && left + stream.avail_out ? Z_DATA_ERROR :
+	   err;
+}
+#else
+static void *dummy_variable = &dummy_variable;
+#endif