[RFC PATCH 3/5] libceph: add sparse read support to OSD client

[Jeff Layton <jlayton@kernel.org>]

Add a new sparse_read operation for the OSD client, driven by its own
state machine. The messenger can repeatedly call the sparse_read
operation, and it will pass back the necessary info to set up to read
the next extent of data, while zeroing in the sparse regions.

Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
 include/linux/ceph/osd_client.h |  37 ++++++++
 net/ceph/osd_client.c           | 161 +++++++++++++++++++++++++++++++-
 2 files changed, 193 insertions(+), 5 deletions(-)

diff --git a/include/linux/ceph/osd_client.h b/include/linux/ceph/osd_client.h
index 3431011f364d..9405cf3f5b45 100644
--- a/include/linux/ceph/osd_client.h
+++ b/include/linux/ceph/osd_client.h
@@ -29,6 +29,42 @@ typedef void (*ceph_osdc_callback_t)(struct ceph_osd_request *);
 
 #define CEPH_HOMELESS_OSD	-1
 
+enum ceph_sparse_read_state {
+	CEPH_SPARSE_READ_COUNT	= 0,
+	CEPH_SPARSE_READ_EXTENTS,
+	CEPH_SPARSE_READ_DATA_LEN,
+	CEPH_SPARSE_READ_DATA,
+};
+
+/* A single extent in a SPARSE_READ reply */
+struct ceph_sparse_extent {
+	__le64	off;
+	__le64	len;
+} __attribute__((packed));
+
+/*
+ * A SPARSE_READ reply is a 32-bit count of extents, followed by an array of
+ * 64-bit offset/length pairs, and then all of the actual file data
+ * concatenated after it (sans holes).
+ *
+ * Unfortunately, we don't know how long the extent array is until we've
+ * started reading the data section of the reply, so for a real sparse read, we
+ * have to allocate the array after alloc_msg returns.
+ *
+ * For the common case of a single extent, we keep an embedded extent here so
+ * we can avoid the extra allocation.
+ */
+struct ceph_sparse_read {
+	enum ceph_sparse_read_state	sr_state;	/* state machine state */
+	u64				sr_offset;	/* request obj offset */
+	u64				sr_pos;		/* current pos in buffer */
+	int				sr_index;	/* current extent index */
+	__le32				sr_count;	/* number of extents */
+	__le32				sr_datalen;	/* length of actual data */
+	struct ceph_sparse_extent	*sr_extent;	/* extent array */
+	struct ceph_sparse_extent	sr_emb_ext[1];	/* extent (for trivial cases */
+};
+
 /* a given osd we're communicating with */
 struct ceph_osd {
 	refcount_t o_ref;
@@ -46,6 +82,7 @@ struct ceph_osd {
 	unsigned long lru_ttl;
 	struct list_head o_keepalive_item;
 	struct mutex lock;
+	struct ceph_sparse_read	o_sparse_read;
 };
 
 #define CEPH_OSD_SLAB_OPS	2
diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c
index 1c5815530e0d..602e2315728a 100644
--- a/net/ceph/osd_client.c
+++ b/net/ceph/osd_client.c
@@ -376,6 +376,7 @@ static void osd_req_op_data_release(struct ceph_osd_request *osd_req,
 
 	switch (op->op) {
 	case CEPH_OSD_OP_READ:
+	case CEPH_OSD_OP_SPARSE_READ:
 	case CEPH_OSD_OP_WRITE:
 	case CEPH_OSD_OP_WRITEFULL:
 		ceph_osd_data_release(&op->extent.osd_data);
@@ -706,6 +707,7 @@ static void get_num_data_items(struct ceph_osd_request *req,
 		/* reply */
 		case CEPH_OSD_OP_STAT:
 		case CEPH_OSD_OP_READ:
+		case CEPH_OSD_OP_SPARSE_READ:
 		case CEPH_OSD_OP_LIST_WATCHERS:
 			*num_reply_data_items += 1;
 			break;
@@ -775,7 +777,7 @@ void osd_req_op_extent_init(struct ceph_osd_request *osd_req,
 
 	BUG_ON(opcode != CEPH_OSD_OP_READ && opcode != CEPH_OSD_OP_WRITE &&
 	       opcode != CEPH_OSD_OP_WRITEFULL && opcode != CEPH_OSD_OP_ZERO &&
-	       opcode != CEPH_OSD_OP_TRUNCATE);
+	       opcode != CEPH_OSD_OP_TRUNCATE && opcode != CEPH_OSD_OP_SPARSE_READ);
 
 	op->extent.offset = offset;
 	op->extent.length = length;
@@ -984,6 +986,7 @@ static u32 osd_req_encode_op(struct ceph_osd_op *dst,
 	case CEPH_OSD_OP_STAT:
 		break;
 	case CEPH_OSD_OP_READ:
+	case CEPH_OSD_OP_SPARSE_READ:
 	case CEPH_OSD_OP_WRITE:
 	case CEPH_OSD_OP_WRITEFULL:
 	case CEPH_OSD_OP_ZERO:
@@ -1080,7 +1083,12 @@ struct ceph_osd_request *ceph_osdc_new_request(struct ceph_osd_client *osdc,
 
 	BUG_ON(opcode != CEPH_OSD_OP_READ && opcode != CEPH_OSD_OP_WRITE &&
 	       opcode != CEPH_OSD_OP_ZERO && opcode != CEPH_OSD_OP_TRUNCATE &&
-	       opcode != CEPH_OSD_OP_CREATE && opcode != CEPH_OSD_OP_DELETE);
+	       opcode != CEPH_OSD_OP_CREATE && opcode != CEPH_OSD_OP_DELETE &&
+	       opcode != CEPH_OSD_OP_SPARSE_READ);
+
+	/* can't handle more than one sparse read data item at a time yet */
+	if (WARN_ON_ONCE(opcode == CEPH_OSD_OP_SPARSE_READ && num_ops > 1))
+		return ERR_PTR(-EINVAL);
 
 	req = ceph_osdc_alloc_request(osdc, snapc, num_ops, use_mempool,
 					GFP_NOFS);
@@ -2037,6 +2045,7 @@ static void setup_request_data(struct ceph_osd_request *req)
 					       &op->raw_data_in);
 			break;
 		case CEPH_OSD_OP_READ:
+		case CEPH_OSD_OP_SPARSE_READ:
 			ceph_osdc_msg_data_add(reply_msg,
 					       &op->extent.osd_data);
 			break;
@@ -2443,6 +2452,20 @@ static void submit_request(struct ceph_osd_request *req, bool wrlocked)
 	__submit_request(req, wrlocked);
 }
 
+static void ceph_init_sparse_read(struct ceph_sparse_read *sr, u64 offset)
+{
+	if (sr->sr_extent != sr->sr_emb_ext)
+		kfree(sr->sr_extent);
+	sr->sr_state = CEPH_SPARSE_READ_COUNT;
+	sr->sr_offset = offset;
+	sr->sr_pos = 0;
+	sr->sr_count = 0;
+	sr->sr_index = 0;
+	sr->sr_extent = sr->sr_emb_ext;
+	sr->sr_extent[0].off = 0;
+	sr->sr_extent[0].len = 0;
+}
+
 static void finish_request(struct ceph_osd_request *req)
 {
 	struct ceph_osd_client *osdc = req->r_osdc;
@@ -2452,8 +2475,10 @@ static void finish_request(struct ceph_osd_request *req)
 
 	req->r_end_latency = ktime_get();
 
-	if (req->r_osd)
+	if (req->r_osd) {
+		ceph_init_sparse_read(&req->r_osd->o_sparse_read, 0);
 		unlink_request(req->r_osd, req);
+	}
 	atomic_dec(&osdc->num_requests);
 
 	/*
@@ -3655,6 +3680,7 @@ static void handle_reply(struct ceph_osd *osd, struct ceph_msg *msg)
 	struct MOSDOpReply m;
 	u64 tid = le64_to_cpu(msg->hdr.tid);
 	u32 data_len = 0;
+	u32 result_len = 0;
 	int ret;
 	int i;
 
@@ -3749,6 +3775,13 @@ static void handle_reply(struct ceph_osd *osd, struct ceph_msg *msg)
 		req->r_ops[i].rval = m.rval[i];
 		req->r_ops[i].outdata_len = m.outdata_len[i];
 		data_len += m.outdata_len[i];
+		if (req->r_ops[i].op == CEPH_OSD_OP_SPARSE_READ) {
+			struct ceph_sparse_read *sr = &osd->o_sparse_read;
+
+			result_len += sr->sr_pos - sr->sr_offset;
+		} else {
+			result_len += m.outdata_len[i];
+		}
 	}
 	if (data_len != le32_to_cpu(msg->hdr.data_len)) {
 		pr_err("sum of lens %u != %u for tid %llu\n", data_len,
@@ -3763,7 +3796,7 @@ static void handle_reply(struct ceph_osd *osd, struct ceph_msg *msg)
 	 * one (type of) reply back.
 	 */
 	WARN_ON(!(m.flags & CEPH_OSD_FLAG_ONDISK));
-	req->r_result = m.result ?: data_len;
+	req->r_result = m.result ?: result_len;
 	finish_request(req);
 	mutex_unlock(&osd->lock);
 	up_read(&osdc->lock);
@@ -5398,6 +5431,22 @@ static void osd_dispatch(struct ceph_connection *con, struct ceph_msg *msg)
 	ceph_msg_put(msg);
 }
 
+static bool is_sparse_read(struct ceph_osd_request *req, u64 *off)
+{
+	int i;
+
+	if (!(req->r_flags & CEPH_OSD_FLAG_READ))
+		return false;
+
+	for (i = 0; i < req->r_num_ops; ++i) {
+		if (req->r_ops[i].op == CEPH_OSD_OP_SPARSE_READ) {
+			*off = req->r_ops[i].extent.offset;
+			return true;
+		}
+	}
+	return false;
+}
+
 /*
  * Lookup and return message for incoming reply.  Don't try to do
  * anything about a larger than preallocated data portion of the
@@ -5414,6 +5463,8 @@ static struct ceph_msg *get_reply(struct ceph_connection *con,
 	int front_len = le32_to_cpu(hdr->front_len);
 	int data_len = le32_to_cpu(hdr->data_len);
 	u64 tid = le64_to_cpu(hdr->tid);
+	u64 sroff;
+	bool sparse;
 
 	down_read(&osdc->lock);
 	if (!osd_registered(osd)) {
@@ -5446,7 +5497,9 @@ static struct ceph_msg *get_reply(struct ceph_connection *con,
 		req->r_reply = m;
 	}
 
-	if (data_len > req->r_reply->data_length) {
+	sparse = is_sparse_read(req, &sroff);
+
+	if (!sparse && (data_len > req->r_reply->data_length)) {
 		pr_warn("%s osd%d tid %llu data %d > preallocated %zu, skipping\n",
 			__func__, osd->o_osd, req->r_tid, data_len,
 			req->r_reply->data_length);
@@ -5456,6 +5509,10 @@ static struct ceph_msg *get_reply(struct ceph_connection *con,
 	}
 
 	m = ceph_msg_get(req->r_reply);
+	m->sparse_read = sparse;
+	if (sparse)
+		ceph_init_sparse_read(&osd->o_sparse_read, sroff);
+
 	dout("get_reply tid %lld %p\n", tid, m);
 
 out_unlock_session:
@@ -5688,9 +5745,103 @@ static int osd_check_message_signature(struct ceph_msg *msg)
 	return ceph_auth_check_message_signature(auth, msg);
 }
 
+static void zero_range(struct ceph_msg *msg, u64 off, u64 len)
+{
+	struct ceph_msg_data_cursor cursor;
+
+	ceph_msg_data_cursor_init(&cursor, msg, off + len);
+
+	while (len) {
+		struct page *page;
+		size_t poff, plen;
+		bool last = false;
+
+		page = ceph_msg_data_next(&cursor, &poff, &plen, &last);
+		if (WARN_ON_ONCE(!page))
+			break;
+		if (plen > len)
+			plen = len;
+		zero_user_segment(page, poff, plen);
+		len -= plen;
+	}
+}
+
+static int osd_sparse_read(struct ceph_connection *con, u64 *poff, u64 *plen, char **pbuf)
+{
+	struct ceph_osd *o = con->private;
+	struct ceph_sparse_read *sr = &o->o_sparse_read;
+	u32 count = __le32_to_cpu(sr->sr_count);
+	u64 eoff, elen;
+
+	switch (sr->sr_state) {
+	case CEPH_SPARSE_READ_COUNT:
+		/* number of extents */
+		*plen = sizeof(sr->sr_count);
+		*pbuf = (char *)&sr->sr_count;
+		sr->sr_state = CEPH_SPARSE_READ_EXTENTS;
+		break;
+	case CEPH_SPARSE_READ_EXTENTS:
+		dout("got %u extents\n", count);
+
+		if (count > 0) {
+			if (count > 1) {
+				/* can't use the embedded extent array */
+				sr->sr_extent = kmalloc_array(count, sizeof(*sr->sr_extent),
+							   GFP_NOIO);
+				if (!sr->sr_extent)
+					return -ENOMEM;
+			}
+			*plen = count * sizeof(*sr->sr_extent);
+			*pbuf = (char *)sr->sr_extent;
+			sr->sr_state = CEPH_SPARSE_READ_DATA_LEN;
+			break;
+		}
+		/* No extents? Fall through to reading data len */
+		fallthrough;
+	case CEPH_SPARSE_READ_DATA_LEN:
+		*plen = sizeof(sr->sr_datalen);
+		*pbuf = (char *)&sr->sr_datalen;
+		sr->sr_state = CEPH_SPARSE_READ_DATA;
+		break;
+	case CEPH_SPARSE_READ_DATA:
+		/* on first extent, set last offset to starting pos */
+		if (sr->sr_index == 0)
+			sr->sr_pos = sr->sr_offset;
+
+		if (sr->sr_index >= count) {
+			dout("sr_index %d count %d\n", sr->sr_index, count);
+			return 0;
+		}
+
+		eoff = le64_to_cpu(sr->sr_extent[sr->sr_index].off);
+		elen = le64_to_cpu(sr->sr_extent[sr->sr_index].len);
+
+		dout("ext %d off 0x%llx len 0x%llx\n", sr->sr_index, eoff, elen);
+
+		/* zero out anything from sr_pos to start of extent */
+		if (sr->sr_pos < eoff)
+			zero_range(con->in_msg,
+				   sr->sr_pos - sr->sr_offset,
+				   eoff - sr->sr_pos);
+
+		/* Pass back extent info for msgr to set up buffer */
+		*poff = eoff - sr->sr_offset;
+		*plen = elen;
+
+		/* Set position for next extent */
+		sr->sr_pos = *poff + elen;
+
+		/* Bump the array index */
+		++sr->sr_index;
+		break;
+	}
+	return 1;
+}
+
 static const struct ceph_connection_operations osd_con_ops = {
 	.get = osd_get_con,
 	.put = osd_put_con,
+	.sparse_read = osd_sparse_read,
 	.alloc_msg = osd_alloc_msg,
 	.dispatch = osd_dispatch,
 	.fault = osd_fault,
-- 
2.34.1