zlib: zlib can only process 4GB at a time

The size of objects we read from the repository and data we try to put
into the repository are represented in "unsigned long", so that on larger
architectures we can handle objects that weigh more than 4GB.

But the interface defined in zlib.h to communicate with inflate/deflate
limits avail_in (how many bytes of input are we calling zlib with) and
avail_out (how many bytes of output from zlib are we ready to accept)
fields effectively to 4GB by defining their type to be uInt.

In many places in our code, we allocate a large buffer (e.g. mmap'ing a
large loose object file) and tell zlib its size by assigning the size to
avail_in field of the stream, but that will truncate the high octets of
the real size. The worst part of this story is that we often pass around
z_stream (the state object used by zlib) to keep track of the number of
used bytes in input/output buffer by inspecting these two fields, which
practically limits our callchain to the same 4GB limit.

Wrap z_stream in another structure git_zstream that can express avail_in
and avail_out in unsigned long. For now, just die() when the caller gives
a size that cannot be given to a single zlib call. In later patches in the
series, we would make git_inflate() and git_deflate() internally loop to
give callers an illusion that our "improved" version of zlib interface can
operate on a buffer larger than 4GB in one go.

Signed-off-by: Junio C Hamano <[email protected]>

Author: gitster

archive-zip.c

@@ -90,7 +90,7 @@ static void copy_le32(unsigned char *dest, unsigned int n)
 static void *zlib_deflate(void *data, unsigned long size,
 		int compression_level, unsigned long *compressed_size)
 {
-	z_stream stream;
+	git_zstream stream;
 	unsigned long maxsize;
 	void *buffer;
 	int result;

builtin/apply.c

@@ -1634,7 +1634,7 @@ static inline int metadata_changes(struct patch *patch)
 static char *inflate_it(const void *data, unsigned long size,
 			unsigned long inflated_size)
 {
-	z_stream stream;
+	git_zstream stream;
 	void *out;
 	int st;
 

builtin/index-pack.c

@@ -265,7 +265,7 @@ static void unlink_base_data(struct base_data *c)
 static void *unpack_entry_data(unsigned long offset, unsigned long size)
 {
 	int status;
-	z_stream stream;
+	git_zstream stream;
 	void *buf = xmalloc(size);
 
 	memset(&stream, 0, sizeof(stream));
@@ -355,7 +355,7 @@ static void *get_data_from_pack(struct object_entry *obj)
 	off_t from = obj[0].idx.offset + obj[0].hdr_size;
 	unsigned long len = obj[1].idx.offset - from;
 	unsigned char *data, *inbuf;
-	z_stream stream;
+	git_zstream stream;
 	int status;
 
 	data = xmalloc(obj->size);
@@ -666,7 +666,7 @@ static void parse_pack_objects(unsigned char *sha1)
 
 static int write_compressed(struct sha1file *f, void *in, unsigned int size)
 {
-	z_stream stream;
+	git_zstream stream;
 	int status;
 	unsigned char outbuf[4096];
 

builtin/pack-objects.c

@@ -126,7 +126,7 @@ static void *get_delta(struct object_entry *entry)
 
 static unsigned long do_compress(void **pptr, unsigned long size)
 {
-	z_stream stream;
+	git_zstream stream;
 	void *in, *out;
 	unsigned long maxsize;
 
@@ -160,7 +160,7 @@ static int check_pack_inflate(struct packed_git *p,
 		off_t len,
 		unsigned long expect)
 {
-	z_stream stream;
+	git_zstream stream;
 	unsigned char fakebuf[4096], *in;
 	int st;
 
@@ -187,12 +187,12 @@ static void copy_pack_data(struct sha1file *f,
 		off_t len)
 {
 	unsigned char *in;
-	unsigned int avail;
+	unsigned long avail;
 
 	while (len) {
 		in = use_pack(p, w_curs, offset, &avail);
 		if (avail > len)
-			avail = (unsigned int)len;
+			avail = (unsigned long)len;
 		sha1write(f, in, avail);
 		offset += avail;
 		len -= avail;
@@ -994,7 +994,7 @@ static void check_object(struct object_entry *entry)
 		const unsigned char *base_ref = NULL;
 		struct object_entry *base_entry;
 		unsigned long used, used_0;
-		unsigned int avail;
+		unsigned long avail;
 		off_t ofs;
 		unsigned char *buf, c;
 

builtin/unpack-objects.c

@@ -90,7 +90,7 @@ static void use(int bytes)
 
 static void *get_data(unsigned long size)
 {
-	z_stream stream;
+	git_zstream stream;
 	void *buf = xmalloc(size);
 
 	memset(&stream, 0, sizeof(stream));

cache.h

@@ -16,18 +16,27 @@
 #endif
 
 #include <zlib.h>
-
-void git_inflate_init(z_streamp strm);
-void git_inflate_init_gzip_only(z_streamp strm);
-void git_inflate_end(z_streamp strm);
-int git_inflate(z_streamp strm, int flush);
-
-void git_deflate_init(z_streamp strm, int level);
-void git_deflate_init_gzip(z_streamp strm, int level);
-void git_deflate_end(z_streamp strm);
-int git_deflate_end_gently(z_streamp strm);
-int git_deflate(z_streamp strm, int flush);
-unsigned long git_deflate_bound(z_streamp, unsigned long);
+typedef struct git_zstream {
+	z_stream z;
+	unsigned long avail_in;
+	unsigned long avail_out;
+	unsigned long total_in;
+	unsigned long total_out;
+	unsigned char *next_in;
+	unsigned char *next_out;
+} git_zstream;
+
+void git_inflate_init(git_zstream *);
+void git_inflate_init_gzip_only(git_zstream *);
+void git_inflate_end(git_zstream *);
+int git_inflate(git_zstream *, int flush);
+
+void git_deflate_init(git_zstream *, int level);
+void git_deflate_init_gzip(git_zstream *, int level);
+void git_deflate_end(git_zstream *);
+int git_deflate_end_gently(git_zstream *);
+int git_deflate(git_zstream *, int flush);
+unsigned long git_deflate_bound(git_zstream *, unsigned long);
 
 #if defined(DT_UNKNOWN) && !defined(NO_D_TYPE_IN_DIRENT)
 #define DTYPE(de)	((de)->d_type)
@@ -991,7 +1000,7 @@ extern struct packed_git *find_sha1_pack(const unsigned char *sha1,
 extern void pack_report(void);
 extern int open_pack_index(struct packed_git *);
 extern void close_pack_index(struct packed_git *);
-extern unsigned char *use_pack(struct packed_git *, struct pack_window **, off_t, unsigned int *);
+extern unsigned char *use_pack(struct packed_git *, struct pack_window **, off_t, unsigned long *);
 extern void close_pack_windows(struct packed_git *);
 extern void unuse_pack(struct pack_window **);
 extern void free_pack_by_name(const char *);

diff.c

@@ -1729,7 +1729,7 @@ static unsigned char *deflate_it(char *data,
 {
 	int bound;
 	unsigned char *deflated;
-	z_stream stream;
+	git_zstream stream;
 
 	memset(&stream, 0, sizeof(stream));
 	git_deflate_init(&stream, zlib_compression_level);

fast-import.c

@@ -1017,7 +1017,7 @@ static int store_object(
 	unsigned char sha1[20];
 	unsigned long hdrlen, deltalen;
 	git_SHA_CTX c;
-	z_stream s;
+	git_zstream s;
 
 	hdrlen = sprintf((char *)hdr,"%s %lu", typename(type),
 		(unsigned long)dat->len) + 1;
@@ -1163,7 +1163,7 @@ static void stream_blob(uintmax_t len, unsigned char *sha1out, uintmax_t mark)
 	off_t offset;
 	git_SHA_CTX c;
 	git_SHA_CTX pack_file_ctx;
-	z_stream s;
+	git_zstream s;
 	int status = Z_OK;
 
 	/* Determine if we should auto-checkpoint. */

http-backend.c

@@ -271,7 +271,7 @@ static struct rpc_service *select_service(const char *name)
 
 static void inflate_request(const char *prog_name, int out)
 {
-	z_stream stream;
+	git_zstream stream;
 	unsigned char in_buf[8192];
 	unsigned char out_buf[8192];
 	unsigned long cnt = 0;

http-push.c

@@ -352,7 +352,7 @@ static void start_put(struct transfer_request *request)
 	unsigned long len;
 	int hdrlen;
 	ssize_t size;
-	z_stream stream;
+	git_zstream stream;
 
 	unpacked = read_sha1_file(request->obj->sha1, &type, &len);
 	hdrlen = sprintf(hdr, "%s %lu", typename(type), len) + 1;