mingw: add a cache below mingw's lstat and dirent implementations

Checking the work tree status is quite slow on Windows, due to slow
`lstat()` emulation (git calls `lstat()` once for each file in the
index). Windows operating system APIs seem to be much better at scanning
the status of entire directories than checking single files.

Add an `lstat()` implementation that uses a cache for lstat data. Cache
misses read the entire parent directory and add it to the cache.
Subsequent `lstat()` calls for the same directory are served directly
from the cache.

Also implement `opendir()`/`readdir()`/`closedir()` so that they create
and use directory listings in the cache.

The cache doesn't track file system changes and doesn't plug into any
modifying file APIs, so it has to be explicitly enabled for git functions
that don't modify the working copy.

Note: in an earlier version of this patch, the cache was always active and
tracked file system changes via ReadDirectoryChangesW. However, this was
much more complex and had negative impact on the performance of modifying
git commands such as 'git checkout'.

Signed-off-by: Karsten Blees <blees@dcon.de>
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>

compat/win32/fscache.c

@@ -0,0 +1,462 @@
+#include "../../cache.h"
+#include "../../hashmap.h"
+#include "../win32.h"
+#include "fscache.h"
+#include "../../dir.h"
+
+static int initialized;
+static volatile long enabled;
+static struct hashmap map;
+static CRITICAL_SECTION mutex;
+
+/*
+ * An entry in the file system cache. Used for both entire directory listings
+ * and file entries.
+ */
+#pragma GCC diagnostic push
+#pragma GCC diagnostic ignored "-Wpedantic"
+struct fsentry {
+	struct hashmap_entry ent;
+	mode_t st_mode;
+	/* Pointer to the directory listing, or NULL for the listing itself. */
+	struct fsentry *list;
+	/* Pointer to the next file entry of the list. */
+	struct fsentry *next;
+
+	union {
+		/* Reference count of the directory listing. */
+		volatile long refcnt;
+		struct {
+			/* More stat members (only used for file entries). */
+			off64_t st_size;
+			struct timespec st_atim;
+			struct timespec st_mtim;
+			struct timespec st_ctim;
+		} s;
+	} u;
+
+	/* Length of name. */
+	unsigned short len;
+	/*
+	 * Name of the entry. For directory listings: relative path of the
+	 * directory, without trailing '/' (empty for cwd()). For file entries:
+	 * name of the file. Typically points to the end of the structure if
+	 * the fsentry is allocated on the heap (see fsentry_alloc), or to a
+	 * local variable if on the stack (see fsentry_init).
+	 */
+	struct dirent dirent;
+};
+#pragma GCC diagnostic pop
+
+struct heap_fsentry {
+	union {
+		struct fsentry ent;
+		char dummy[sizeof(struct fsentry) + MAX_PATH];
+	} u;
+};
+
+/*
+ * Compares the paths of two fsentry structures for equality.
+ */
+static int fsentry_cmp(void *unused_cmp_data,
+		       const struct fsentry *fse1, const struct fsentry *fse2,
+		       void *unused_keydata)
+{
+	int res;
+	if (fse1 == fse2)
+		return 0;
+
+	/* compare the list parts first */
+	if (fse1->list != fse2->list &&
+	    (res = fsentry_cmp(NULL, fse1->list ? fse1->list : fse1,
+			       fse2->list ? fse2->list	: fse2, NULL)))
+		return res;
+
+	/* if list parts are equal, compare len and name */
+	if (fse1->len != fse2->len)
+		return fse1->len - fse2->len;
+	return fspathncmp(fse1->dirent.d_name, fse2->dirent.d_name, fse1->len);
+}
+
+/*
+ * Calculates the hash code of an fsentry structure's path.
+ */
+static unsigned int fsentry_hash(const struct fsentry *fse)
+{
+	unsigned int hash = fse->list ? fse->list->ent.hash : 0;
+	return hash ^ memihash(fse->dirent.d_name, fse->len);
+}
+
+/*
+ * Initialize an fsentry structure for use by fsentry_hash and fsentry_cmp.
+ */
+static void fsentry_init(struct fsentry *fse, struct fsentry *list,
+			 const char *name, size_t len)
+{
+	fse->list = list;
+	if (len > MAX_PATH)
+		BUG("Trying to allocate fsentry for long path '%.*s'",
+		    (int)len, name);
+	memcpy(fse->dirent.d_name, name, len);
+	fse->dirent.d_name[len] = 0;
+	fse->len = len;
+	hashmap_entry_init(&fse->ent, fsentry_hash(fse));
+}
+
+/*
+ * Allocate an fsentry structure on the heap.
+ */
+static struct fsentry *fsentry_alloc(struct fsentry *list, const char *name,
+		size_t len)
+{
+	/* overallocate fsentry and copy the name to the end */
+	struct fsentry *fse = xmalloc(sizeof(struct fsentry) + len + 1);
+	/* init the rest of the structure */
+	fsentry_init(fse, list, name, len);
+	fse->next = NULL;
+	fse->u.refcnt = 1;
+	return fse;
+}
+
+/*
+ * Add a reference to an fsentry.
+ */
+inline static void fsentry_addref(struct fsentry *fse)
+{
+	if (fse->list)
+		fse = fse->list;
+
+	InterlockedIncrement(&(fse->u.refcnt));
+}
+
+/*
+ * Release the reference to an fsentry, frees the memory if its the last ref.
+ */
+static void fsentry_release(struct fsentry *fse)
+{
+	if (fse->list)
+		fse = fse->list;
+
+	if (InterlockedDecrement(&(fse->u.refcnt)))
+		return;
+
+	while (fse) {
+		struct fsentry *next = fse->next;
+		free(fse);
+		fse = next;
+	}
+}
+
+/*
+ * Allocate and initialize an fsentry from a WIN32_FIND_DATA structure.
+ */
+static struct fsentry *fseentry_create_entry(struct fsentry *list,
+					     const WIN32_FIND_DATAW *fdata)
+{
+	char buf[MAX_PATH * 3];
+	int len;
+	struct fsentry *fse;
+	len = xwcstoutf(buf, fdata->cFileName, ARRAY_SIZE(buf));
+
+	fse = fsentry_alloc(list, buf, len);
+
+	fse->st_mode = file_attr_to_st_mode(fdata->dwFileAttributes);
+	fse->dirent.d_type = S_ISDIR(fse->st_mode) ? DT_DIR : DT_REG;
+	fse->u.s.st_size = (((off64_t) (fdata->nFileSizeHigh)) << 32)
+			| fdata->nFileSizeLow;
+	filetime_to_timespec(&(fdata->ftLastAccessTime), &(fse->u.s.st_atim));
+	filetime_to_timespec(&(fdata->ftLastWriteTime), &(fse->u.s.st_mtim));
+	filetime_to_timespec(&(fdata->ftCreationTime), &(fse->u.s.st_ctim));
+
+	return fse;
+}
+
+/*
+ * Create an fsentry-based directory listing (similar to opendir / readdir).
+ * Dir should not contain trailing '/'. Use an empty string for the current
+ * directory (not "."!).
+ */
+static struct fsentry *fsentry_create_list(const struct fsentry *dir)
+{
+	wchar_t pattern[MAX_PATH + 2]; /* + 2 for '/' '*' */
+	WIN32_FIND_DATAW fdata;
+	HANDLE h;
+	int wlen;
+	struct fsentry *list, **phead;
+	DWORD err;
+
+	/* convert name to UTF-16 and check length < MAX_PATH */
+	if ((wlen = xutftowcsn(pattern, dir->dirent.d_name, MAX_PATH,
+			       dir->len)) < 0) {
+		if (errno == ERANGE)
+			errno = ENAMETOOLONG;
+		return NULL;
+	}
+
+	/* append optional '/' and wildcard '*' */
+	if (wlen)
+		pattern[wlen++] = '/';
+	pattern[wlen++] = '*';
+	pattern[wlen] = 0;
+
+	/* open find handle */
+	h = FindFirstFileW(pattern, &fdata);
+	if (h == INVALID_HANDLE_VALUE) {
+		err = GetLastError();
+		errno = (err == ERROR_DIRECTORY) ? ENOTDIR : err_win_to_posix(err);
+		return NULL;
+	}
+
+	/* allocate object to hold directory listing */
+	list = fsentry_alloc(NULL, dir->dirent.d_name, dir->len);
+
+	/* walk directory and build linked list of fsentry structures */
+	phead = &list->next;
+	do {
+		*phead = fseentry_create_entry(list, &fdata);
+		phead = &(*phead)->next;
+	} while (FindNextFileW(h, &fdata));
+
+	/* remember result of last FindNextFile, then close find handle */
+	err = GetLastError();
+	FindClose(h);
+
+	/* return the list if we've got all the files */
+	if (err == ERROR_NO_MORE_FILES)
+		return list;
+
+	/* otherwise free the list and return error */
+	fsentry_release(list);
+	errno = err_win_to_posix(err);
+	return NULL;
+}
+
+/*
+ * Adds a directory listing to the cache.
+ */
+static void fscache_add(struct fsentry *fse)
+{
+	if (fse->list)
+		fse = fse->list;
+
+	for (; fse; fse = fse->next)
+		hashmap_add(&map, &fse->ent);
+}
+
+/*
+ * Clears the cache.
+ */
+static void fscache_clear(void)
+{
+	hashmap_clear_and_free(&map, struct fsentry, ent);
+	hashmap_init(&map, (hashmap_cmp_fn)fsentry_cmp, NULL, 0);
+}
+
+/*
+ * Checks if the cache is enabled for the given path.
+ */
+static inline int fscache_enabled(const char *path)
+{
+	return enabled > 0 && !is_absolute_path(path);
+}
+
+/*
+ * Looks up or creates a cache entry for the specified key.
+ */
+static struct fsentry *fscache_get(struct fsentry *key)
+{
+	struct fsentry *fse;
+
+	EnterCriticalSection(&mutex);
+	/* check if entry is in cache */
+	fse = hashmap_get_entry(&map, key, ent, NULL);
+	if (fse) {
+		fsentry_addref(fse);
+		LeaveCriticalSection(&mutex);
+		return fse;
+	}
+	/* if looking for a file, check if directory listing is in cache */
+	if (!fse && key->list) {
+		fse = hashmap_get_entry(&map, key->list, ent, NULL);
+		if (fse) {
+			LeaveCriticalSection(&mutex);
+			/* dir entry without file entry -> file doesn't exist */
+			errno = ENOENT;
+			return NULL;
+		}
+	}
+
+	/* create the directory listing (outside mutex!) */
+	LeaveCriticalSection(&mutex);
+	fse = fsentry_create_list(key->list ? key->list : key);
+	if (!fse)
+		return NULL;
+
+	EnterCriticalSection(&mutex);
+	/* add directory listing if it hasn't been added by some other thread */
+	if (!hashmap_get_entry(&map, key, ent, NULL))
+		fscache_add(fse);
+
+	/* lookup file entry if requested (fse already points to directory) */
+	if (key->list)
+		fse = hashmap_get_entry(&map, key, ent, NULL);
+
+	/* return entry or ENOENT */
+	if (fse)
+		fsentry_addref(fse);
+	else
+		errno = ENOENT;
+
+	LeaveCriticalSection(&mutex);
+	return fse;
+}
+
+/*
+ * Enables or disables the cache. Note that the cache is read-only, changes to
+ * the working directory are NOT reflected in the cache while enabled.
+ */
+int fscache_enable(int enable)
+{
+	int result;
+
+	if (!initialized) {
+		/* allow the cache to be disabled entirely */
+		if (!core_fscache)
+			return 0;
+
+		InitializeCriticalSection(&mutex);
+		hashmap_init(&map, (hashmap_cmp_fn) fsentry_cmp, NULL, 0);
+		initialized = 1;
+	}
+
+	result = enable ? InterlockedIncrement(&enabled)
+			: InterlockedDecrement(&enabled);
+
+	if (enable && result == 1) {
+		/* redirect opendir and lstat to the fscache implementations */
+		opendir = fscache_opendir;
+		lstat = fscache_lstat;
+	} else if (!enable && !result) {
+		/* reset opendir and lstat to the original implementations */
+		opendir = dirent_opendir;
+		lstat = mingw_lstat;
+		EnterCriticalSection(&mutex);
+		fscache_clear();
+		LeaveCriticalSection(&mutex);
+	}
+	return result;
+}
+
+/*
+ * Lstat replacement, uses the cache if enabled, otherwise redirects to
+ * mingw_lstat.
+ */
+int fscache_lstat(const char *filename, struct stat *st)
+{
+	int dirlen, base, len;
+	struct heap_fsentry key[2];
+	struct fsentry *fse;
+
+	if (!fscache_enabled(filename))
+		return mingw_lstat(filename, st);
+
+	/* split filename into path + name */
+	len = strlen(filename);
+	if (len && is_dir_sep(filename[len - 1]))
+		len--;
+	base = len;
+	while (base && !is_dir_sep(filename[base - 1]))
+		base--;
+	dirlen = base ? base - 1 : 0;
+
+	/* lookup entry for path + name in cache */
+	fsentry_init(&key[0].u.ent, NULL, filename, dirlen);
+	fsentry_init(&key[1].u.ent, &key[0].u.ent, filename + base, len - base);
+	fse = fscache_get(&key[1].u.ent);
+	if (!fse) {
+		errno = ENOENT;
+		return -1;
+	}
+
+	/* copy stat data */
+	st->st_ino = 0;
+	st->st_gid = 0;
+	st->st_uid = 0;
+	st->st_dev = 0;
+	st->st_rdev = 0;
+	st->st_nlink = 1;
+	st->st_mode = fse->st_mode;
+	st->st_size = fse->u.s.st_size;
+	st->st_atim = fse->u.s.st_atim;
+	st->st_mtim = fse->u.s.st_mtim;
+	st->st_ctim = fse->u.s.st_ctim;
+
+	/* don't forget to release fsentry */
+	fsentry_release(fse);
+	return 0;
+}
+
+typedef struct fscache_DIR {
+	struct DIR base_dir; /* extend base struct DIR */
+	struct fsentry *pfsentry;
+	struct dirent *dirent;
+} fscache_DIR;
+
+/*
+ * Readdir replacement.
+ */
+static struct dirent *fscache_readdir(DIR *base_dir)
+{
+	fscache_DIR *dir = (fscache_DIR*) base_dir;
+	struct fsentry *next = dir->pfsentry->next;
+	if (!next)
+		return NULL;
+	dir->pfsentry = next;
+	dir->dirent = &next->dirent;
+	return dir->dirent;
+}
+
+/*
+ * Closedir replacement.
+ */
+static int fscache_closedir(DIR *base_dir)
+{
+	fscache_DIR *dir = (fscache_DIR*) base_dir;
+	fsentry_release(dir->pfsentry);
+	free(dir);
+	return 0;
+}
+
+/*
+ * Opendir replacement, uses a directory listing from the cache if enabled,
+ * otherwise calls original dirent implementation.
+ */
+DIR *fscache_opendir(const char *dirname)
+{
+	struct heap_fsentry key;
+	struct fsentry *list;
+	fscache_DIR *dir;
+	int len;
+
+	if (!fscache_enabled(dirname))
+		return dirent_opendir(dirname);
+
+	/* prepare name (strip trailing '/', replace '.') */
+	len = strlen(dirname);
+	if ((len == 1 && dirname[0] == '.') ||
+	    (len && is_dir_sep(dirname[len - 1])))
+		len--;
+
+	/* get directory listing from cache */
+	fsentry_init(&key.u.ent, NULL, dirname, len);
+	list = fscache_get(&key.u.ent);
+	if (!list)
+		return NULL;
+
+	/* alloc and return DIR structure */
+	dir = (fscache_DIR*) xmalloc(sizeof(fscache_DIR));
+	dir->base_dir.preaddir = fscache_readdir;
+	dir->base_dir.pclosedir = fscache_closedir;
+	dir->pfsentry = list;
+	return (DIR*) dir;
+}

compat/win32/fscache.h

@@ -0,0 +1,10 @@
+#ifndef FSCACHE_H
+#define FSCACHE_H
+
+int fscache_enable(int enable);
+#define enable_fscache(x) fscache_enable(x)
+
+DIR *fscache_opendir(const char *dir);
+int fscache_lstat(const char *file_name, struct stat *buf);
+
+#endif

config.mak.uname

@@ -486,7 +486,7 @@ endif
 		compat/win32/path-utils.o \
 		compat/win32/pthread.o compat/win32/syslog.o \
 		compat/win32/trace2_win32_process_info.o \
-		compat/win32/dirent.o
+		compat/win32/dirent.o compat/win32/fscache.o
 	COMPAT_CFLAGS = -D__USE_MINGW_ACCESS -DDETECT_MSYS_TTY -DENSURE_MSYSTEM_IS_SET -DNOGDI -DHAVE_STRING_H -Icompat -Icompat/regex -Icompat/win32 -DSTRIP_EXTENSION=\".exe\"
 	BASIC_LDFLAGS = -IGNORE:4217 -IGNORE:4049 -NOLOGO
 	# invalidcontinue.obj allows Git's source code to close the same file
@@ -674,7 +674,7 @@ ifeq ($(uname_S),MINGW)
 		compat/win32/flush.o \
 		compat/win32/path-utils.o \
 		compat/win32/pthread.o compat/win32/syslog.o \
-		compat/win32/dirent.o
+		compat/win32/dirent.o compat/win32/fscache.o
 	BASIC_CFLAGS += -DWIN32
 	EXTLIBS += -lws2_32
 	GITLIBS += git.res

contrib/buildsystems/CMakeLists.txt

@@ -307,7 +307,8 @@ if(CMAKE_SYSTEM_NAME STREQUAL "Windows")
 		compat/win32/trace2_win32_process_info.c
 		compat/win32/dirent.c
 		compat/nedmalloc/nedmalloc.c
-		compat/strdup.c)
+		compat/strdup.c
+		compat/win32/fscache.c)
 	set(NO_UNIX_SOCKETS 1)
 
 elseif(CMAKE_SYSTEM_NAME STREQUAL "Linux")

git-compat-util.h

@@ -249,9 +249,11 @@ static inline int is_xplatform_dir_sep(int c)
 /* pull in Windows compatibility stuff */
 #include "compat/win32/path-utils.h"
 #include "compat/mingw.h"
+#include "compat/win32/fscache.h"
 #elif defined(_MSC_VER)
 #include "compat/win32/path-utils.h"
 #include "compat/msvc.h"
+#include "compat/win32/fscache.h"
 #else
 #include <sys/utsname.h>
 #include <sys/wait.h>