1 files changed, 406 insertions, 12 deletions

diff --git a/object-file.c b/object-file.c
index 2bc36ab3ee..4675c8ed6b 100644
--- a/object-file.c
+++ b/object-file.c
@@ -10,7 +10,6 @@
 #define USE_THE_REPOSITORY_VARIABLE
 
 #include "git-compat-util.h"
-#include "bulk-checkin.h"
 #include "convert.h"
 #include "dir.h"
 #include "environment.h"
@@ -28,6 +27,8 @@
 #include "read-cache-ll.h"
 #include "setup.h"
 #include "streaming.h"
+#include "tempfile.h"
+#include "tmp-objdir.h"
 
 /* The maximum size for an object header. */
 #define MAX_HEADER_LEN 32
@@ -666,6 +667,93 @@ void hash_object_file(const struct git_hash_algo *algo, const void *buf,
 	write_object_file_prepare(algo, buf, len, type, oid, hdr, &hdrlen);
 }
 
+struct transaction_packfile {
+	char *pack_tmp_name;
+	struct hashfile *f;
+	off_t offset;
+	struct pack_idx_option pack_idx_opts;
+
+	struct pack_idx_entry **written;
+	uint32_t alloc_written;
+	uint32_t nr_written;
+};
+
+struct odb_transaction {
+	struct object_database *odb;
+
+	struct tmp_objdir *objdir;
+	struct transaction_packfile packfile;
+};
+
+static void prepare_loose_object_transaction(struct odb_transaction *transaction)
+{
+	/*
+	 * We lazily create the temporary object directory
+	 * the first time an object might be added, since
+	 * callers may not know whether any objects will be
+	 * added at the time they call object_file_transaction_begin.
+	 */
+	if (!transaction || transaction->objdir)
+		return;
+
+	transaction->objdir = tmp_objdir_create(transaction->odb->repo, "bulk-fsync");
+	if (transaction->objdir)
+		tmp_objdir_replace_primary_odb(transaction->objdir, 0);
+}
+
+static void fsync_loose_object_transaction(struct odb_transaction *transaction,
+					   int fd, const char *filename)
+{
+	/*
+	 * If we have an active ODB transaction, we issue a call that
+	 * cleans the filesystem page cache but avoids a hardware flush
+	 * command. Later on we will issue a single hardware flush
+	 * before renaming the objects to their final names as part of
+	 * flush_batch_fsync.
+	 */
+	if (!transaction || !transaction->objdir ||
+	    git_fsync(fd, FSYNC_WRITEOUT_ONLY) < 0) {
+		if (errno == ENOSYS)
+			warning(_("core.fsyncMethod = batch is unsupported on this platform"));
+		fsync_or_die(fd, filename);
+	}
+}
+
+/*
+ * Cleanup after batch-mode fsync_object_files.
+ */
+static void flush_loose_object_transaction(struct odb_transaction *transaction)
+{
+	struct strbuf temp_path = STRBUF_INIT;
+	struct tempfile *temp;
+
+	if (!transaction->objdir)
+		return;
+
+	/*
+	 * Issue a full hardware flush against a temporary file to ensure
+	 * that all objects are durable before any renames occur. The code in
+	 * fsync_loose_object_transaction has already issued a writeout
+	 * request, but it has not flushed any writeback cache in the storage
+	 * hardware or any filesystem logs. This fsync call acts as a barrier
+	 * to ensure that the data in each new object file is durable before
+	 * the final name is visible.
+	 */
+	strbuf_addf(&temp_path, "%s/bulk_fsync_XXXXXX",
+		    repo_get_object_directory(transaction->odb->repo));
+	temp = xmks_tempfile(temp_path.buf);
+	fsync_or_die(get_tempfile_fd(temp), get_tempfile_path(temp));
+	delete_tempfile(&temp);
+	strbuf_release(&temp_path);
+
+	/*
+	 * Make the object files visible in the primary ODB after their data is
+	 * fully durable.
+	 */
+	tmp_objdir_migrate(transaction->objdir);
+	transaction->objdir = NULL;
+}
+
 /* Finalize a file on disk, and close it. */
 static void close_loose_object(struct odb_source *source,
 			       int fd, const char *filename)
@@ -674,7 +762,7 @@ static void close_loose_object(struct odb_source *source,
 		goto out;
 
 	if (batch_fsync_enabled(FSYNC_COMPONENT_LOOSE_OBJECT))
-		fsync_loose_object_bulk_checkin(fd, filename);
+		fsync_loose_object_transaction(source->odb->transaction, fd, filename);
 	else if (fsync_object_files > 0)
 		fsync_or_die(fd, filename);
 	else
@@ -852,7 +940,7 @@ static int write_loose_object(struct odb_source *source,
 	static struct strbuf filename = STRBUF_INIT;
 
 	if (batch_fsync_enabled(FSYNC_COMPONENT_LOOSE_OBJECT))
-		prepare_loose_object_bulk_checkin();
+		prepare_loose_object_transaction(source->odb->transaction);
 
 	odb_loose_path(source, &filename, oid);
 
@@ -941,7 +1029,7 @@ int stream_loose_object(struct odb_source *source,
 	int hdrlen;
 
 	if (batch_fsync_enabled(FSYNC_COMPONENT_LOOSE_OBJECT))
-		prepare_loose_object_bulk_checkin();
+		prepare_loose_object_transaction(source->odb->transaction);
 
 	/* Since oid is not determined, save tmp file to odb path. */
 	strbuf_addf(&filename, "%s/", source->path);
@@ -1243,6 +1331,274 @@ static int index_core(struct index_state *istate,
 	return ret;
 }
 
+static int already_written(struct odb_transaction *transaction,
+			   struct object_id *oid)
+{
+	/* The object may already exist in the repository */
+	if (odb_has_object(transaction->odb, oid,
+			   HAS_OBJECT_RECHECK_PACKED | HAS_OBJECT_FETCH_PROMISOR))
+		return 1;
+
+	/* Might want to keep the list sorted */
+	for (uint32_t i = 0; i < transaction->packfile.nr_written; i++)
+		if (oideq(&transaction->packfile.written[i]->oid, oid))
+			return 1;
+
+	/* This is a new object we need to keep */
+	return 0;
+}
+
+/* Lazily create backing packfile for the state */
+static void prepare_packfile_transaction(struct odb_transaction *transaction,
+					 unsigned flags)
+{
+	struct transaction_packfile *state = &transaction->packfile;
+	if (!(flags & INDEX_WRITE_OBJECT) || state->f)
+		return;
+
+	state->f = create_tmp_packfile(transaction->odb->repo,
+				       &state->pack_tmp_name);
+	reset_pack_idx_option(&state->pack_idx_opts);
+
+	/* Pretend we are going to write only one object */
+	state->offset = write_pack_header(state->f, 1);
+	if (!state->offset)
+		die_errno("unable to write pack header");
+}
+
+/*
+ * Read the contents from fd for size bytes, streaming it to the
+ * packfile in state while updating the hash in ctx. Signal a failure
+ * by returning a negative value when the resulting pack would exceed
+ * the pack size limit and this is not the first object in the pack,
+ * so that the caller can discard what we wrote from the current pack
+ * by truncating it and opening a new one. The caller will then call
+ * us again after rewinding the input fd.
+ *
+ * The already_hashed_to pointer is kept untouched by the caller to
+ * make sure we do not hash the same byte when we are called
+ * again. This way, the caller does not have to checkpoint its hash
+ * status before calling us just in case we ask it to call us again
+ * with a new pack.
+ */
+static int stream_blob_to_pack(struct transaction_packfile *state,
+			       struct git_hash_ctx *ctx, off_t *already_hashed_to,
+			       int fd, size_t size, const char *path,
+			       unsigned flags)
+{
+	git_zstream s;
+	unsigned char ibuf[16384];
+	unsigned char obuf[16384];
+	unsigned hdrlen;
+	int status = Z_OK;
+	int write_object = (flags & INDEX_WRITE_OBJECT);
+	off_t offset = 0;
+
+	git_deflate_init(&s, pack_compression_level);
+
+	hdrlen = encode_in_pack_object_header(obuf, sizeof(obuf), OBJ_BLOB, size);
+	s.next_out = obuf + hdrlen;
+	s.avail_out = sizeof(obuf) - hdrlen;
+
+	while (status != Z_STREAM_END) {
+		if (size && !s.avail_in) {
+			size_t rsize = size < sizeof(ibuf) ? size : sizeof(ibuf);
+			ssize_t read_result = read_in_full(fd, ibuf, rsize);
+			if (read_result < 0)
+				die_errno("failed to read from '%s'", path);
+			if ((size_t)read_result != rsize)
+				die("failed to read %u bytes from '%s'",
+				    (unsigned)rsize, path);
+			offset += rsize;
+			if (*already_hashed_to < offset) {
+				size_t hsize = offset - *already_hashed_to;
+				if (rsize < hsize)
+					hsize = rsize;
+				if (hsize)
+					git_hash_update(ctx, ibuf, hsize);
+				*already_hashed_to = offset;
+			}
+			s.next_in = ibuf;
+			s.avail_in = rsize;
+			size -= rsize;
+		}
+
+		status = git_deflate(&s, size ? 0 : Z_FINISH);
+
+		if (!s.avail_out || status == Z_STREAM_END) {
+			if (write_object) {
+				size_t written = s.next_out - obuf;
+
+				/* would we bust the size limit? */
+				if (state->nr_written &&
+				    pack_size_limit_cfg &&
+				    pack_size_limit_cfg < state->offset + written) {
+					git_deflate_abort(&s);
+					return -1;
+				}
+
+				hashwrite(state->f, obuf, written);
+				state->offset += written;
+			}
+			s.next_out = obuf;
+			s.avail_out = sizeof(obuf);
+		}
+
+		switch (status) {
+		case Z_OK:
+		case Z_BUF_ERROR:
+		case Z_STREAM_END:
+			continue;
+		default:
+			die("unexpected deflate failure: %d", status);
+		}
+	}
+	git_deflate_end(&s);
+	return 0;
+}
+
+static void flush_packfile_transaction(struct odb_transaction *transaction)
+{
+	struct transaction_packfile *state = &transaction->packfile;
+	struct repository *repo = transaction->odb->repo;
+	unsigned char hash[GIT_MAX_RAWSZ];
+	struct strbuf packname = STRBUF_INIT;
+	char *idx_tmp_name = NULL;
+
+	if (!state->f)
+		return;
+
+	if (state->nr_written == 0) {
+		close(state->f->fd);
+		free_hashfile(state->f);
+		unlink(state->pack_tmp_name);
+		goto clear_exit;
+	} else if (state->nr_written == 1) {
+		finalize_hashfile(state->f, hash, FSYNC_COMPONENT_PACK,
+				  CSUM_HASH_IN_STREAM | CSUM_FSYNC | CSUM_CLOSE);
+	} else {
+		int fd = finalize_hashfile(state->f, hash, FSYNC_COMPONENT_PACK, 0);
+		fixup_pack_header_footer(repo->hash_algo, fd, hash, state->pack_tmp_name,
+					 state->nr_written, hash,
+					 state->offset);
+		close(fd);
+	}
+
+	strbuf_addf(&packname, "%s/pack/pack-%s.",
+		    repo_get_object_directory(transaction->odb->repo),
+		    hash_to_hex_algop(hash, repo->hash_algo));
+
+	stage_tmp_packfiles(repo, &packname, state->pack_tmp_name,
+			    state->written, state->nr_written, NULL,
+			    &state->pack_idx_opts, hash, &idx_tmp_name);
+	rename_tmp_packfile_idx(repo, &packname, &idx_tmp_name);
+
+	for (uint32_t i = 0; i < state->nr_written; i++)
+		free(state->written[i]);
+
+clear_exit:
+	free(idx_tmp_name);
+	free(state->pack_tmp_name);
+	free(state->written);
+	memset(state, 0, sizeof(*state));
+
+	strbuf_release(&packname);
+	/* Make objects we just wrote available to ourselves */
+	odb_reprepare(repo->objects);
+}
+
+/*
+ * This writes the specified object to a packfile. Objects written here
+ * during the same transaction are written to the same packfile. The
+ * packfile is not flushed until the transaction is flushed. The caller
+ * is expected to ensure a valid transaction is setup for objects to be
+ * recorded to.
+ *
+ * This also bypasses the usual "convert-to-git" dance, and that is on
+ * purpose. We could write a streaming version of the converting
+ * functions and insert that before feeding the data to fast-import
+ * (or equivalent in-core API described above). However, that is
+ * somewhat complicated, as we do not know the size of the filter
+ * result, which we need to know beforehand when writing a git object.
+ * Since the primary motivation for trying to stream from the working
+ * tree file and to avoid mmaping it in core is to deal with large
+ * binary blobs, they generally do not want to get any conversion, and
+ * callers should avoid this code path when filters are requested.
+ */
+static int index_blob_packfile_transaction(struct odb_transaction *transaction,
+					   struct object_id *result_oid, int fd,
+					   size_t size, const char *path,
+					   unsigned flags)
+{
+	struct transaction_packfile *state = &transaction->packfile;
+	off_t seekback, already_hashed_to;
+	struct git_hash_ctx ctx;
+	unsigned char obuf[16384];
+	unsigned header_len;
+	struct hashfile_checkpoint checkpoint;
+	struct pack_idx_entry *idx = NULL;
+
+	seekback = lseek(fd, 0, SEEK_CUR);
+	if (seekback == (off_t)-1)
+		return error("cannot find the current offset");
+
+	header_len = format_object_header((char *)obuf, sizeof(obuf),
+					  OBJ_BLOB, size);
+	transaction->odb->repo->hash_algo->init_fn(&ctx);
+	git_hash_update(&ctx, obuf, header_len);
+
+	/* Note: idx is non-NULL when we are writing */
+	if ((flags & INDEX_WRITE_OBJECT) != 0) {
+		CALLOC_ARRAY(idx, 1);
+
+		prepare_packfile_transaction(transaction, flags);
+		hashfile_checkpoint_init(state->f, &checkpoint);
+	}
+
+	already_hashed_to = 0;
+
+	while (1) {
+		prepare_packfile_transaction(transaction, flags);
+		if (idx) {
+			hashfile_checkpoint(state->f, &checkpoint);
+			idx->offset = state->offset;
+			crc32_begin(state->f);
+		}
+		if (!stream_blob_to_pack(state, &ctx, &already_hashed_to,
+					 fd, size, path, flags))
+			break;
+		/*
+		 * Writing this object to the current pack will make
+		 * it too big; we need to truncate it, start a new
+		 * pack, and write into it.
+		 */
+		if (!idx)
+			BUG("should not happen");
+		hashfile_truncate(state->f, &checkpoint);
+		state->offset = checkpoint.offset;
+		flush_packfile_transaction(transaction);
+		if (lseek(fd, seekback, SEEK_SET) == (off_t)-1)
+			return error("cannot seek back");
+	}
+	git_hash_final_oid(result_oid, &ctx);
+	if (!idx)
+		return 0;
+
+	idx->crc32 = crc32_end(state->f);
+	if (already_written(transaction, result_oid)) {
+		hashfile_truncate(state->f, &checkpoint);
+		state->offset = checkpoint.offset;
+		free(idx);
+	} else {
+		oidcpy(&idx->oid, result_oid);
+		ALLOC_GROW(state->written,
+			   state->nr_written + 1,
+			   state->alloc_written);
+		state->written[state->nr_written++] = idx;
+	}
+	return 0;
+}
+
 int index_fd(struct index_state *istate, struct object_id *oid,
 	     int fd, struct stat *st,
 	     enum object_type type, const char *path, unsigned flags)
@@ -1253,18 +1609,27 @@ int index_fd(struct index_state *istate, struct object_id *oid,
 	 * Call xsize_t() only when needed to avoid potentially unnecessary
 	 * die() for large files.
 	 */
-	if (type == OBJ_BLOB && path && would_convert_to_git_filter_fd(istate, path))
+	if (type == OBJ_BLOB && path && would_convert_to_git_filter_fd(istate, path)) {
 		ret = index_stream_convert_blob(istate, oid, fd, path, flags);
-	else if (!S_ISREG(st->st_mode))
+	} else if (!S_ISREG(st->st_mode)) {
 		ret = index_pipe(istate, oid, fd, type, path, flags);
-	else if ((st->st_size >= 0 && (size_t) st->st_size <= repo_settings_get_big_file_threshold(istate->repo)) ||
-		 type != OBJ_BLOB ||
-		 (path && would_convert_to_git(istate, path)))
+	} else if ((st->st_size >= 0 &&
+		    (size_t)st->st_size <= repo_settings_get_big_file_threshold(istate->repo)) ||
+		   type != OBJ_BLOB ||
+		   (path && would_convert_to_git(istate, path))) {
 		ret = index_core(istate, oid, fd, xsize_t(st->st_size),
 				 type, path, flags);
-	else
-		ret = index_blob_bulk_checkin(oid, fd, xsize_t(st->st_size), path,
-					     flags);
+	} else {
+		struct odb_transaction *transaction;
+
+		transaction = odb_transaction_begin(the_repository->objects);
+		ret = index_blob_packfile_transaction(the_repository->objects->transaction,
+						      oid, fd,
+						      xsize_t(st->st_size),
+						      path, flags);
+		odb_transaction_commit(transaction);
+	}
+
 	close(fd);
 	return ret;
 }
@@ -1601,3 +1966,32 @@ out:
 		munmap(map, mapsize);
 	return ret;
 }
+
+struct odb_transaction *object_file_transaction_begin(struct odb_source *source)
+{
+	struct object_database *odb = source->odb;
+
+	if (odb->transaction)
+		return NULL;
+
+	CALLOC_ARRAY(odb->transaction, 1);
+	odb->transaction->odb = odb;
+
+	return odb->transaction;
+}
+
+void object_file_transaction_commit(struct odb_transaction *transaction)
+{
+	if (!transaction)
+		return;
+
+	/*
+	 * Ensure the transaction ending matches the pending transaction.
+	 */
+	ASSERT(transaction == transaction->odb->transaction);
+
+	flush_loose_object_transaction(transaction);
+	flush_packfile_transaction(transaction);
+	transaction->odb->transaction = NULL;
+	free(transaction);
+}