libceph: switch to calc_target(), part 2

The crux of this is getting rid of ceph_osdc_build_request(), so that
MOSDOp can be encoded not before but after calc_target() calculates the
actual target.  Encoding now happens within ceph_osdc_start_request().

Also nuked is the accompanying bunch of pointers into the encoded
buffer that was used to update fields on each send - instead, the
entire front is re-encoded.  If we want to support target->name_len !=
base->name_len in the future, there is no other way, because oid is
surrounded by other fields in the encoded buffer.

Encoding OSD ops and adding data items to the request message were
mixed together in osd_req_encode_op().  While we want to re-encode OSD
ops, we don't want to add duplicate data items to the message when
resending, so all call to ceph_osdc_msg_data_add() are factored out
into a new setup_request_data().
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>

drivers/block/rbd.c

@@ -1896,27 +1896,17 @@ static void rbd_osd_req_format_read(struct rbd_obj_request *obj_request)
 {
 	struct rbd_img_request *img_request = obj_request->img_request;
 	struct ceph_osd_request *osd_req = obj_request->osd_req;
-	u64 snap_id;
-
-	rbd_assert(osd_req != NULL);
 
-	snap_id = img_request ? img_request->snap_id : CEPH_NOSNAP;
-	ceph_osdc_build_request(osd_req, obj_request->offset,
-			NULL, snap_id, NULL);
+	if (img_request)
+		osd_req->r_snapid = img_request->snap_id;
 }
 
 static void rbd_osd_req_format_write(struct rbd_obj_request *obj_request)
 {
-	struct rbd_img_request *img_request = obj_request->img_request;
 	struct ceph_osd_request *osd_req = obj_request->osd_req;
-	struct ceph_snap_context *snapc;
-	struct timespec mtime = CURRENT_TIME;
-
-	rbd_assert(osd_req != NULL);
 
-	snapc = img_request ? img_request->snapc : NULL;
-	ceph_osdc_build_request(osd_req, obj_request->offset,
-			snapc, CEPH_NOSNAP, &mtime);
+	osd_req->r_mtime = CURRENT_TIME;
+	osd_req->r_data_offset = obj_request->offset;
 }
 
 /*

fs/ceph/addr.c

@@ -376,8 +376,6 @@ static int start_read(struct inode *inode, struct list_head *page_list, int max)
 	req->r_callback = finish_read;
 	req->r_inode = inode;
 
-	ceph_osdc_build_request(req, off, NULL, vino.snap, NULL);
-
 	dout("start_read %p starting %p %lld~%lld\n", inode, req, off, len);
 	ret = ceph_osdc_start_request(osdc, req, false);
 	if (ret < 0)
@@ -1063,10 +1061,7 @@ static int ceph_writepages_start(struct address_space *mapping,
 			pages = NULL;
 		}
 
-		vino = ceph_vino(inode);
-		ceph_osdc_build_request(req, offset, snapc, vino.snap,
-					&inode->i_mtime);
-
+		req->r_mtime = inode->i_mtime;
 		rc = ceph_osdc_start_request(&fsc->client->osdc, req, true);
 		BUG_ON(rc);
 		req = NULL;
@@ -1614,7 +1609,7 @@ int ceph_uninline_data(struct file *filp, struct page *locked_page)
 		goto out;
 	}
 
-	ceph_osdc_build_request(req, 0, NULL, CEPH_NOSNAP, &inode->i_mtime);
+	req->r_mtime = inode->i_mtime;
 	err = ceph_osdc_start_request(&fsc->client->osdc, req, false);
 	if (!err)
 		err = ceph_osdc_wait_request(&fsc->client->osdc, req);
@@ -1657,7 +1652,7 @@ int ceph_uninline_data(struct file *filp, struct page *locked_page)
 			goto out_put;
 	}
 
-	ceph_osdc_build_request(req, 0, NULL, CEPH_NOSNAP, &inode->i_mtime);
+	req->r_mtime = inode->i_mtime;
 	err = ceph_osdc_start_request(&fsc->client->osdc, req, false);
 	if (!err)
 		err = ceph_osdc_wait_request(&fsc->client->osdc, req);
@@ -1790,12 +1785,9 @@ static int __ceph_pool_perm_get(struct ceph_inode_info *ci, u32 pool)
 
 	osd_req_op_raw_data_in_pages(rd_req, 0, pages, PAGE_SIZE,
 				     0, false, true);
-	ceph_osdc_build_request(rd_req, 0, NULL, CEPH_NOSNAP,
-				&ci->vfs_inode.i_mtime);
 	err = ceph_osdc_start_request(&fsc->client->osdc, rd_req, false);
 
-	ceph_osdc_build_request(wr_req, 0, NULL, CEPH_NOSNAP,
-				&ci->vfs_inode.i_mtime);
+	wr_req->r_mtime = ci->vfs_inode.i_mtime;
 	err2 = ceph_osdc_start_request(&fsc->client->osdc, wr_req, false);
 
 	if (!err)

fs/ceph/file.c

@@ -727,8 +727,8 @@ static void ceph_aio_retry_work(struct work_struct *work)
 	req->r_ops[0] = orig_req->r_ops[0];
 	osd_req_op_init(req, 1, CEPH_OSD_OP_STARTSYNC, 0);
 
-	ceph_osdc_build_request(req, req->r_ops[0].extent.offset,
-				snapc, CEPH_NOSNAP, &aio_req->mtime);
+	req->r_mtime = aio_req->mtime;
+	req->r_data_offset = req->r_ops[0].extent.offset;
 
 	ceph_osdc_put_request(orig_req);
 
@@ -882,14 +882,12 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter,
 					(pos+len) | (PAGE_SIZE - 1));
 
 			osd_req_op_init(req, 1, CEPH_OSD_OP_STARTSYNC, 0);
+			req->r_mtime = mtime;
 		}
 
-
 		osd_req_op_extent_osd_data_pages(req, 0, pages, len, start,
 						 false, false);
 
-		ceph_osdc_build_request(req, pos, snapc, vino.snap, &mtime);
-
 		if (aio_req) {
 			aio_req->total_len += len;
 			aio_req->num_reqs++;
@@ -1074,9 +1072,7 @@ ceph_sync_write(struct kiocb *iocb, struct iov_iter *from, loff_t pos,
 		osd_req_op_extent_osd_data_pages(req, 0, pages, len, 0,
 						false, true);
 
-		/* BUG_ON(vino.snap != CEPH_NOSNAP); */
-		ceph_osdc_build_request(req, pos, snapc, vino.snap, &mtime);
-
+		req->r_mtime = mtime;
 		ret = ceph_osdc_start_request(&fsc->client->osdc, req, false);
 		if (!ret)
 			ret = ceph_osdc_wait_request(&fsc->client->osdc, req);
@@ -1532,9 +1528,7 @@ static int ceph_zero_partial_object(struct inode *inode,
 		goto out;
 	}
 
-	ceph_osdc_build_request(req, offset, NULL, ceph_vino(inode).snap,
-				&inode->i_mtime);
-
+	req->r_mtime = inode->i_mtime;
 	ret = ceph_osdc_start_request(&fsc->client->osdc, req, false);
 	if (!ret) {
 		ret = ceph_osdc_wait_request(&fsc->client->osdc, req);

include/linux/ceph/osd_client.h

@@ -104,7 +104,7 @@ struct ceph_osd_req_op {
 			struct ceph_osd_data response_data;
 			__u8 class_len;
 			__u8 method_len;
-			__u8 argc;
+			u32 indata_len;
 		} cls;
 		struct {
 			u64 cookie;
@@ -162,14 +162,6 @@ struct ceph_osd_request {
 	/* request osd ops array  */
 	unsigned int		r_num_ops;
 
-	/* these are updated on each send */
-	__le32           *r_request_osdmap_epoch;
-	__le32           *r_request_flags;
-	__le64           *r_request_pool;
-	void             *r_request_pgid;
-	__le32           *r_request_attempts;
-	struct ceph_eversion *r_request_reassert_version;
-
 	int               r_result;
 	int               r_got_reply;
 	int		  r_linger;
@@ -180,16 +172,22 @@ struct ceph_osd_request {
 	struct completion r_completion, r_safe_completion;
 	ceph_osdc_callback_t r_callback;
 	ceph_osdc_unsafe_callback_t r_unsafe_callback;
-	struct ceph_eversion r_reassert_version;
 	struct list_head  r_unsafe_item;
 
 	struct inode *r_inode;         	      /* for use by callbacks */
 	void *r_priv;			      /* ditto */
 
-	u64               r_snapid;
-	unsigned long     r_stamp;            /* send OR check time */
+	/* set by submitter */
+	u64 r_snapid;                         /* for reads, CEPH_NOSNAP o/w */
+	struct ceph_snap_context *r_snapc;    /* for writes */
+	struct timespec r_mtime;              /* ditto */
+	u64 r_data_offset;                    /* ditto */
 
-	struct ceph_snap_context *r_snapc;    /* snap context for writes */
+	/* internal */
+	unsigned long r_stamp;                /* jiffies, send or check time */
+	int r_attempts;
+	struct ceph_eversion r_replay_version; /* aka reassert_version */
+	u32 r_last_force_resend;
 
 	struct ceph_osd_req_op r_ops[];
 };
@@ -334,11 +332,6 @@ extern struct ceph_osd_request *ceph_osdc_alloc_request(struct ceph_osd_client *
 					       gfp_t gfp_flags);
 int ceph_osdc_alloc_messages(struct ceph_osd_request *req, gfp_t gfp);
 
-extern void ceph_osdc_build_request(struct ceph_osd_request *req, u64 off,
-				    struct ceph_snap_context *snapc,
-				    u64 snap_id,
-				    struct timespec *mtime);
-
 extern struct ceph_osd_request *ceph_osdc_new_request(struct ceph_osd_client *,
 				      struct ceph_file_layout *layout,
 				      struct ceph_vino vino,

include/linux/ceph/rados.h

@@ -394,6 +394,13 @@ enum {
 	CEPH_OSD_FLAG_SKIPRWLOCKS =   0x10000,  /* skip rw locks */
 	CEPH_OSD_FLAG_IGNORE_OVERLAY = 0x20000, /* ignore pool overlay */
 	CEPH_OSD_FLAG_FLUSH =         0x40000,  /* this is part of flush */
+	CEPH_OSD_FLAG_MAP_SNAP_CLONE = 0x80000,  /* map snap direct to clone id */
+	CEPH_OSD_FLAG_ENFORCE_SNAPC   = 0x100000,  /* use snapc provided even if
+						      pool uses pool snaps */
+	CEPH_OSD_FLAG_REDIRECTED   = 0x200000,  /* op has been redirected */
+	CEPH_OSD_FLAG_KNOWN_REDIR = 0x400000,  /* redirect bit is authoritative */
+	CEPH_OSD_FLAG_FULL_TRY =    0x800000,  /* try op despite full flag */
+	CEPH_OSD_FLAG_FULL_FORCE = 0x1000000,  /* force op despite full flag */
 };
 
 enum {

net/ceph/debugfs.c

@@ -145,6 +145,43 @@ static int monc_show(struct seq_file *s, void *p)
 	return 0;
 }
 
+static void dump_target(struct seq_file *s, struct ceph_osd_request_target *t)
+{
+	int i;
+
+	seq_printf(s, "osd%d\t%llu.%x\t[", t->osd, t->pgid.pool, t->pgid.seed);
+	for (i = 0; i < t->up.size; i++)
+		seq_printf(s, "%s%d", (!i ? "" : ","), t->up.osds[i]);
+	seq_printf(s, "]/%d\t[", t->up.primary);
+	for (i = 0; i < t->acting.size; i++)
+		seq_printf(s, "%s%d", (!i ? "" : ","), t->acting.osds[i]);
+	seq_printf(s, "]/%d\t%*pE\t0x%x", t->acting.primary,
+		   t->target_oid.name_len, t->target_oid.name, t->flags);
+	if (t->paused)
+		seq_puts(s, "\tP");
+}
+
+static void dump_request(struct seq_file *s, struct ceph_osd_request *req)
+{
+	int i;
+
+	seq_printf(s, "%llu\t", req->r_tid);
+	dump_target(s, &req->r_t);
+
+	seq_printf(s, "\t%d\t%u'%llu", req->r_attempts,
+		   le32_to_cpu(req->r_replay_version.epoch),
+		   le64_to_cpu(req->r_replay_version.version));
+
+	for (i = 0; i < req->r_num_ops; i++) {
+		struct ceph_osd_req_op *op = &req->r_ops[i];
+
+		seq_printf(s, "%s%s", (i == 0 ? "\t" : ","),
+			   ceph_osd_op_name(op->op));
+	}
+
+	seq_putc(s, '\n');
+}
+
 static int osdc_show(struct seq_file *s, void *pp)
 {
 	struct ceph_client *client = s->private;
@@ -154,32 +191,10 @@ static int osdc_show(struct seq_file *s, void *pp)
 	mutex_lock(&osdc->request_mutex);
 	for (p = rb_first(&osdc->requests); p; p = rb_next(p)) {
 		struct ceph_osd_request *req;
-		unsigned int i;
-		int opcode;
 
 		req = rb_entry(p, struct ceph_osd_request, r_node);
 
-		seq_printf(s, "%lld\tosd%d\t%lld.%x\t", req->r_tid,
-			   req->r_osd ? req->r_osd->o_osd : -1,
-			   req->r_t.pgid.pool, req->r_t.pgid.seed);
-
-		seq_printf(s, "%*pE", req->r_base_oid.name_len,
-			   req->r_base_oid.name);
-
-		if (req->r_reassert_version.epoch)
-			seq_printf(s, "\t%u'%llu",
-			   (unsigned int)le32_to_cpu(req->r_reassert_version.epoch),
-			   le64_to_cpu(req->r_reassert_version.version));
-		else
-			seq_printf(s, "\t");
-
-		for (i = 0; i < req->r_num_ops; i++) {
-			opcode = req->r_ops[i].op;
-			seq_printf(s, "%s%s", (i == 0 ? "\t" : ","),
-				   ceph_osd_op_name(opcode));
-		}
-
-		seq_printf(s, "\n");
+		dump_request(s, req);
 	}
 	mutex_unlock(&osdc->request_mutex);
 	return 0;