Teach Git to handle huge files in smudge/clean

convert.c

@@ -613,7 +613,7 @@ static int crlf_to_worktree(const char *src, size_t len, struct strbuf *buf,
 
 struct filter_params {
 	const char *src;
-	unsigned long size;
+	size_t size;
 	int fd;
 	const char *cmd;
 	const char *path;

delta.h

@@ -90,15 +90,15 @@ static inline unsigned long get_delta_hdr_size(const unsigned char **datap,
 					       const unsigned char *top)
 {
 	const unsigned char *data = *datap;
-	unsigned long cmd, size = 0;
+	size_t cmd, size = 0;
 	int i = 0;
 	do {
 		cmd = *data++;
-		size |= (cmd & 0x7f) << i;
+		size |= st_left_shift(cmd & 0x7f, i);
 		i += 7;
 	} while (cmd & 0x80 && data < top);
 	*datap = data;
-	return size;
+	return cast_size_t_to_ulong(size);
 }
 
 #endif

entry.c

@@ -82,11 +82,13 @@ static int create_file(const char *path, unsigned int mode)
 	return open(path, O_WRONLY | O_CREAT | O_EXCL, mode);
 }
 
-void *read_blob_entry(const struct cache_entry *ce, unsigned long *size)
+void *read_blob_entry(const struct cache_entry *ce, size_t *size)
 {
 	enum object_type type;
-	void *blob_data = read_object_file(&ce->oid, &type, size);
+	unsigned long ul;
+	void *blob_data = read_object_file(&ce->oid, &type, &ul);
 
+	*size = ul;
 	if (blob_data) {
 		if (type == OBJ_BLOB)
 			return blob_data;
@@ -270,7 +272,7 @@ static int write_entry(struct cache_entry *ce, char *path, struct conv_attrs *ca
 	int fd, ret, fstat_done = 0;
 	char *new_blob;
 	struct strbuf buf = STRBUF_INIT;
-	unsigned long size;
+	size_t size;
 	ssize_t wrote;
 	size_t newsize = 0;
 	struct stat st;

entry.h

@@ -51,7 +51,7 @@ int finish_delayed_checkout(struct checkout *state, int *nr_checkouts);
  */
 void unlink_entry(const struct cache_entry *ce);
 
-void *read_blob_entry(const struct cache_entry *ce, unsigned long *size);
+void *read_blob_entry(const struct cache_entry *ce, size_t *size);
 int fstat_checkout_output(int fd, const struct checkout *state, struct stat *st);
 void update_ce_after_write(const struct checkout *state, struct cache_entry *ce,
 			   struct stat *st);

git-compat-util.h

@@ -113,6 +113,14 @@
 #define unsigned_mult_overflows(a, b) \
     ((a) && (b) > maximum_unsigned_value_of_type(a) / (a))
 
+/*
+ * Returns true if the left shift of "a" by "shift" bits will
+ * overflow. The type of "a" must be unsigned.
+ */
+#define unsigned_left_shift_overflows(a, shift) \
+    ((shift) < bitsizeof(a) && \
+     (a) > maximum_unsigned_value_of_type(a) >> (shift))
+
 #ifdef __GNUC__
 #define TYPEOF(x) (__typeof__(x))
 #else
@@ -859,6 +867,23 @@ static inline size_t st_sub(size_t a, size_t b)
 	return a - b;
 }
 
+static inline size_t st_left_shift(size_t a, unsigned shift)
+{
+	if (unsigned_left_shift_overflows(a, shift))
+		die("size_t overflow: %"PRIuMAX" << %u",
+		    (uintmax_t)a, shift);
+	return a << shift;
+}
+
+static inline unsigned long cast_size_t_to_ulong(size_t a)
+{
+	if (a != (unsigned long)a)
+		die("object too large to read on this platform: %"
+		    PRIuMAX" is cut off to %lu",
+		    (uintmax_t)a, (unsigned long)a);
+	return (unsigned long)a;
+}
+
 #ifdef HAVE_ALLOCA_H
 # include <alloca.h>
 # define xalloca(size)      (alloca(size))

object-file.c

@@ -1344,7 +1344,7 @@ static int parse_loose_header_extended(const char *hdr, struct object_info *oi,
 				       unsigned int flags)
 {
 	const char *type_buf = hdr;
-	unsigned long size;
+	size_t size;
 	int type, type_len = 0;
 
 	/*
@@ -1388,12 +1388,12 @@ static int parse_loose_header_extended(const char *hdr, struct object_info *oi,
 			if (c > 9)
 				break;
 			hdr++;
-			size = size * 10 + c;
+			size = st_add(st_mult(size, 10), c);
 		}
 	}
 
 	if (oi->sizep)
-		*oi->sizep = size;
+		*oi->sizep = cast_size_t_to_ulong(size);
 
 	/*
 	 * The length must be followed by a zero byte

packfile.c

@@ -1059,7 +1059,7 @@ unsigned long unpack_object_header_buffer(const unsigned char *buf,
 		unsigned long len, enum object_type *type, unsigned long *sizep)
 {
 	unsigned shift;
-	unsigned long size, c;
+	size_t size, c;
 	unsigned long used = 0;
 
 	c = buf[used++];
@@ -1073,10 +1073,10 @@ unsigned long unpack_object_header_buffer(const unsigned char *buf,
 			break;
 		}
 		c = buf[used++];
-		size += (c & 0x7f) << shift;
+		size = st_add(size, st_left_shift(c & 0x7f, shift));
 		shift += 7;
 	}
-	*sizep = size;
+	*sizep = cast_size_t_to_ulong(size);
 	return used;
 }
 

parallel-checkout.c

@@ -261,7 +261,7 @@ static int write_pc_item_to_fd(struct parallel_checkout_item *pc_item, int fd,
 	struct stream_filter *filter;
 	struct strbuf buf = STRBUF_INIT;
 	char *blob;
-	unsigned long size;
+	size_t size;
 	ssize_t wrote;
 
 	/* Sanity check */

t/helper/test-genzeros.c

@@ -3,18 +3,31 @@
 
 int cmd__genzeros(int argc, const char **argv)
 {
-	long count;
+	/* static, so that it is NUL-initialized */
+	static const char zeros[256 * 1024];
+	intmax_t count;
+	ssize_t n;
 
 	if (argc > 2) {
 		fprintf(stderr, "usage: %s [<count>]\n", argv[0]);
 		return 1;
 	}
 
-	count = argc > 1 ? strtol(argv[1], NULL, 0) : -1L;
+	count = argc > 1 ? strtoimax(argv[1], NULL, 0) : -1;
 
-	while (count < 0 || count--) {
-		if (putchar(0) == EOF)
+	/* Writing out individual NUL bytes is slow... */
+	while (count < 0)
+		if (write(1, zeros, ARRAY_SIZE(zeros)) < 0)
 			return -1;
+
+	while (count > 0) {
+		n = write(1, zeros, count < ARRAY_SIZE(zeros) ?
+			  count : ARRAY_SIZE(zeros));
+
+		if (n < 0)
+			return -1;
+
+		count -= n;
 	}
 
 	return 0;

t/t1051-large-conversion.sh

@@ -83,4 +83,29 @@ test_expect_success 'ident converts on output' '
 	test_cmp small.clean large.clean
 '
 
+# This smudge filter prepends 5GB of zeros to the file it checks out. This
+# ensures that smudging doesn't mangle large files on 64-bit Windows.
+test_expect_success EXPENSIVE,SIZE_T_IS_64BIT,!LONG_IS_64BIT \
+		'files over 4GB convert on output' '
+	test_commit test small "a small file" &&
+	test_config filter.makelarge.smudge \
+		"test-tool genzeros $((5*1024*1024*1024)) && cat" &&
+	echo "small filter=makelarge" >.gitattributes &&
+	rm small &&
+	git checkout -- small &&
+	size=$(test_file_size small) &&
+	test "$size" -ge $((5 * 1024 * 1024 * 1024))
+'
+
+# This clean filter writes down the size of input it receives. By checking against
+# the actual size, we ensure that cleaning doesn't mangle large files on 64-bit Windows.
+test_expect_success EXPENSIVE,SIZE_T_IS_64BIT,!LONG_IS_64BIT \
+		'files over 4GB convert on input' '
+	test-tool genzeros $((5*1024*1024*1024)) >big &&
+	test_config filter.checklarge.clean "wc -c >big.size" &&
+	echo "big filter=checklarge" >.gitattributes &&
+	git add big &&
+	test $(test_file_size big) -eq $(cat big.size)
+'
+
 test_done

t/test-lib.sh

@@ -1642,6 +1642,10 @@ build_option () {
 	sed -ne "s/^$1: //p"
 }
 
+test_lazy_prereq SIZE_T_IS_64BIT '
+	test 8 -eq "$(build_option sizeof-size_t)"
+'
+
 test_lazy_prereq LONG_IS_64BIT '
 	test 8 -le "$(build_option sizeof-long)"
 '