md/r5cache: State machine for raid5-cache write back mode

This patch adds state machine for raid5-cache. With log device, the
raid456 array could operate in two different modes (r5c_journal_mode):
  - write-back (R5C_MODE_WRITE_BACK)
  - write-through (R5C_MODE_WRITE_THROUGH)

Existing code of raid5-cache only has write-through mode. For write-back
cache, it is necessary to extend the state machine.

With write-back cache, every stripe could operate in two different
phases:
  - caching
  - writing-out

In caching phase, the stripe handles writes as:
  - write to journal
  - return IO

In writing-out phase, the stripe behaviors as a stripe in write through
mode R5C_MODE_WRITE_THROUGH.

STRIPE_R5C_CACHING is added to sh->state to differentiate caching and
writing-out phase.

Please note: this is a "no-op" patch for raid5-cache write-through
mode.

The following detailed explanation is copied from the raid5-cache.c:

/*
 * raid5 cache state machine
 *
 * With rhe RAID cache, each stripe works in two phases:
 *      - caching phase
 *      - writing-out phase
 *
 * These two phases are controlled by bit STRIPE_R5C_CACHING:
 *   if STRIPE_R5C_CACHING == 0, the stripe is in writing-out phase
 *   if STRIPE_R5C_CACHING == 1, the stripe is in caching phase
 *
 * When there is no journal, or the journal is in write-through mode,
 * the stripe is always in writing-out phase.
 *
 * For write-back journal, the stripe is sent to caching phase on write
 * (r5c_handle_stripe_dirtying). r5c_make_stripe_write_out() kicks off
 * the write-out phase by clearing STRIPE_R5C_CACHING.
 *
 * Stripes in caching phase do not write the raid disks. Instead, all
 * writes are committed from the log device. Therefore, a stripe in
 * caching phase handles writes as:
 *      - write to log device
 *      - return IO
 *
 * Stripes in writing-out phase handle writes as:
 *      - calculate parity
 *      - write pending data and parity to journal
 *      - write data and parity to raid disks
 *      - return IO for pending writes
 */

Signed-off-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Shaohua Li <shli@fb.com>

3 files changed, 211 insertions, 8 deletions

diff --git a/drivers/md/raid5-cache.c b/drivers/md/raid5-cache.c
index 33fc85015147..02a554434747 100644
--- a/drivers/md/raid5-cache.c
+++ b/drivers/md/raid5-cache.c
@@ -40,6 +40,47 @@
  */
 #define R5L_POOL_SIZE	4
 
+/*
+ * r5c journal modes of the array: write-back or write-through.
+ * write-through mode has identical behavior as existing log only
+ * implementation.
+ */
+enum r5c_journal_mode {
+	R5C_JOURNAL_MODE_WRITE_THROUGH = 0,
+	R5C_JOURNAL_MODE_WRITE_BACK = 1,
+};
+
+/*
+ * raid5 cache state machine
+ *
+ * With rhe RAID cache, each stripe works in two phases:
+ *	- caching phase
+ *	- writing-out phase
+ *
+ * These two phases are controlled by bit STRIPE_R5C_CACHING:
+ *   if STRIPE_R5C_CACHING == 0, the stripe is in writing-out phase
+ *   if STRIPE_R5C_CACHING == 1, the stripe is in caching phase
+ *
+ * When there is no journal, or the journal is in write-through mode,
+ * the stripe is always in writing-out phase.
+ *
+ * For write-back journal, the stripe is sent to caching phase on write
+ * (r5c_try_caching_write). r5c_make_stripe_write_out() kicks off
+ * the write-out phase by clearing STRIPE_R5C_CACHING.
+ *
+ * Stripes in caching phase do not write the raid disks. Instead, all
+ * writes are committed from the log device. Therefore, a stripe in
+ * caching phase handles writes as:
+ *	- write to log device
+ *	- return IO
+ *
+ * Stripes in writing-out phase handle writes as:
+ *	- calculate parity
+ *	- write pending data and parity to journal
+ *	- write data and parity to raid disks
+ *	- return IO for pending writes
+ */
+
 struct r5l_log {
 	struct md_rdev *rdev;
 
@@ -96,6 +137,9 @@ struct r5l_log {
 	spinlock_t no_space_stripes_lock;
 
 	bool need_cache_flush;
+
+	/* for r5c_cache */
+	enum r5c_journal_mode r5c_journal_mode;
 };
 
 /*
@@ -133,6 +177,12 @@ enum r5l_io_unit_state {
 	IO_UNIT_STRIPE_END = 3,	/* stripes data finished writing to raid */
 };
 
+bool r5c_is_writeback(struct r5l_log *log)
+{
+	return (log != NULL &&
+		log->r5c_journal_mode == R5C_JOURNAL_MODE_WRITE_BACK);
+}
+
 static sector_t r5l_ring_add(struct r5l_log *log, sector_t start, sector_t inc)
 {
 	start += inc;
@@ -168,12 +218,51 @@ static void __r5l_set_io_unit_state(struct r5l_io_unit *io,
 	io->state = state;
 }
 
+/*
+ * Put the stripe into writing-out phase by clearing STRIPE_R5C_CACHING.
+ * This function should only be called in write-back mode.
+ */
+static void r5c_make_stripe_write_out(struct stripe_head *sh)
+{
+	struct r5conf *conf = sh->raid_conf;
+	struct r5l_log *log = conf->log;
+
+	BUG_ON(!r5c_is_writeback(log));
+
+	WARN_ON(!test_bit(STRIPE_R5C_CACHING, &sh->state));
+	clear_bit(STRIPE_R5C_CACHING, &sh->state);
+}
+
+/*
+ * Setting proper flags after writing (or flushing) data and/or parity to the
+ * log device. This is called from r5l_log_endio() or r5l_log_flush_endio().
+ */
+static void r5c_finish_cache_stripe(struct stripe_head *sh)
+{
+	struct r5l_log *log = sh->raid_conf->log;
+
+	if (log->r5c_journal_mode == R5C_JOURNAL_MODE_WRITE_THROUGH) {
+		BUG_ON(test_bit(STRIPE_R5C_CACHING, &sh->state));
+		/*
+		 * Set R5_InJournal for parity dev[pd_idx]. This means
+		 * all data AND parity in the journal. For RAID 6, it is
+		 * NOT necessary to set the flag for dev[qd_idx], as the
+		 * two parities are written out together.
+		 */
+		set_bit(R5_InJournal, &sh->dev[sh->pd_idx].flags);
+	} else
+		BUG(); /* write-back logic in next patch */
+}
+
 static void r5l_io_run_stripes(struct r5l_io_unit *io)
 {
 	struct stripe_head *sh, *next;
 
 	list_for_each_entry_safe(sh, next, &io->stripe_list, log_list) {
 		list_del_init(&sh->log_list);
+
+		r5c_finish_cache_stripe(sh);
+
 		set_bit(STRIPE_HANDLE, &sh->state);
 		raid5_release_stripe(sh);
 	}
@@ -412,18 +501,19 @@ static int r5l_log_stripe(struct r5l_log *log, struct stripe_head *sh,
 		r5l_append_payload_page(log, sh->dev[i].page);
 	}
 
-	if (sh->qd_idx >= 0) {
+	if (parity_pages == 2) {
 		r5l_append_payload_meta(log, R5LOG_PAYLOAD_PARITY,
 					sh->sector, sh->dev[sh->pd_idx].log_checksum,
 					sh->dev[sh->qd_idx].log_checksum, true);
 		r5l_append_payload_page(log, sh->dev[sh->pd_idx].page);
 		r5l_append_payload_page(log, sh->dev[sh->qd_idx].page);
-	} else {
+	} else if (parity_pages == 1) {
 		r5l_append_payload_meta(log, R5LOG_PAYLOAD_PARITY,
 					sh->sector, sh->dev[sh->pd_idx].log_checksum,
 					0, false);
 		r5l_append_payload_page(log, sh->dev[sh->pd_idx].page);
-	}
+	} else  /* Just writing data, not parity, in caching phase */
+		BUG_ON(parity_pages != 0);
 
 	list_add_tail(&sh->log_list, &io->stripe_list);
 	atomic_inc(&io->pending_stripe);
@@ -455,6 +545,8 @@ int r5l_write_stripe(struct r5l_log *log, struct stripe_head *sh)
 		return -EAGAIN;
 	}
 
+	WARN_ON(test_bit(STRIPE_R5C_CACHING, &sh->state));
+
 	for (i = 0; i < sh->disks; i++) {
 		void *addr;
 
@@ -1112,6 +1204,49 @@ static void r5l_write_super(struct r5l_log *log, sector_t cp)
 	set_bit(MD_CHANGE_DEVS, &mddev->flags);
 }
 
+/*
+ * Try handle write operation in caching phase. This function should only
+ * be called in write-back mode.
+ *
+ * If all outstanding writes can be handled in caching phase, returns 0
+ * If writes requires write-out phase, call r5c_make_stripe_write_out()
+ * and returns -EAGAIN
+ */
+int r5c_try_caching_write(struct r5conf *conf,
+			  struct stripe_head *sh,
+			  struct stripe_head_state *s,
+			  int disks)
+{
+	struct r5l_log *log = conf->log;
+
+	BUG_ON(!r5c_is_writeback(log));
+
+	/* more write-back logic in next patches */
+	r5c_make_stripe_write_out(sh);
+	return -EAGAIN;
+}
+
+/*
+ * clean up the stripe (clear R5_InJournal for dev[pd_idx] etc.) after the
+ * stripe is committed to RAID disks.
+ */
+void r5c_finish_stripe_write_out(struct r5conf *conf,
+				 struct stripe_head *sh,
+				 struct stripe_head_state *s)
+{
+	if (!conf->log ||
+	    !test_bit(R5_InJournal, &sh->dev[sh->pd_idx].flags))
+		return;
+
+	WARN_ON(test_bit(STRIPE_R5C_CACHING, &sh->state));
+	clear_bit(R5_InJournal, &sh->dev[sh->pd_idx].flags);
+
+	if (conf->log->r5c_journal_mode == R5C_JOURNAL_MODE_WRITE_THROUGH)
+		return;
+	BUG();  /* write-back logic in following patches */
+}
+
+
 static int r5l_load_log(struct r5l_log *log)
 {
 	struct md_rdev *rdev = log->rdev;
@@ -1249,6 +1384,8 @@ int r5l_init_log(struct r5conf *conf, struct md_rdev *rdev)
 	INIT_LIST_HEAD(&log->no_space_stripes);
 	spin_lock_init(&log->no_space_stripes_lock);
 
+	log->r5c_journal_mode = R5C_JOURNAL_MODE_WRITE_THROUGH;
+
 	if (r5l_load_log(log))
 		goto error;
 
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 34895f3218d9..7c98eb06d1b2 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -4107,6 +4107,9 @@ static void analyse_stripe(struct stripe_head *sh, struct stripe_head_state *s)
 			if (rdev && !test_bit(Faulty, &rdev->flags))
 				do_recovery = 1;
 		}
+
+		if (test_bit(R5_InJournal, &dev->flags))
+			s->injournal++;
 	}
 	if (test_bit(STRIPE_SYNCING, &sh->state)) {
 		/* If there is a failed device being replaced,
@@ -4386,14 +4389,47 @@ static void handle_stripe(struct stripe_head *sh)
 	    || s.expanding)
 		handle_stripe_fill(sh, &s, disks);
 
-	/* Now to consider new write requests and what else, if anything
-	 * should be read.  We do not handle new writes when:
+	/*
+	 * When the stripe finishes full journal write cycle (write to journal
+	 * and raid disk), this is the clean up procedure so it is ready for
+	 * next operation.
+	 */
+	r5c_finish_stripe_write_out(conf, sh, &s);
+
+	/*
+	 * Now to consider new write requests, cache write back and what else,
+	 * if anything should be read.  We do not handle new writes when:
 	 * 1/ A 'write' operation (copy+xor) is already in flight.
 	 * 2/ A 'check' operation is in flight, as it may clobber the parity
 	 *    block.
+	 * 3/ A r5c cache log write is in flight.
 	 */
-	if (s.to_write && !sh->reconstruct_state && !sh->check_state)
-		handle_stripe_dirtying(conf, sh, &s, disks);
+
+	if (!sh->reconstruct_state && !sh->check_state && !sh->log_io) {
+		if (!r5c_is_writeback(conf->log)) {
+			if (s.to_write)
+				handle_stripe_dirtying(conf, sh, &s, disks);
+		} else { /* write back cache */
+			int ret = 0;
+
+			/* First, try handle writes in caching phase */
+			if (s.to_write)
+				ret = r5c_try_caching_write(conf, sh, &s,
+							    disks);
+			/*
+			 * If caching phase failed: ret == -EAGAIN
+			 *    OR
+			 * stripe under reclaim: !caching && injournal
+			 *
+			 * fall back to handle_stripe_dirtying()
+			 */
+			if (ret == -EAGAIN ||
+			    /* stripe under reclaim: !caching && injournal */
+			    (!test_bit(STRIPE_R5C_CACHING, &sh->state) &&
+			     s.injournal > 0))
+				handle_stripe_dirtying(conf, sh, &s, disks);
+		}
+	}
 
 	/* maybe we need to check and possibly fix the parity for this stripe
 	 * Any reads will already have been scheduled, so we just see if enough
@@ -5110,6 +5146,7 @@ static void raid5_make_request(struct mddev *mddev, struct bio * bi)
 	 * data on failed drives.
 	 */
 	if (rw == READ && mddev->degraded == 0 &&
+	    !r5c_is_writeback(conf->log) &&
 	    mddev->reshape_position == MaxSector) {
 		bi = chunk_aligned_read(mddev, bi);
 		if (!bi)
diff --git a/drivers/md/raid5.h b/drivers/md/raid5.h
index ffc13c4d7e63..c9590a8e1425 100644
--- a/drivers/md/raid5.h
+++ b/drivers/md/raid5.h
@@ -264,6 +264,7 @@ struct stripe_head_state {
 	int syncing, expanding, expanded, replacing;
 	int locked, uptodate, to_read, to_write, failed, written;
 	int to_fill, compute, req_compute, non_overwrite;
+	int injournal;
 	int failed_num[2];
 	int p_failed, q_failed;
 	int dec_preread_active;
@@ -313,6 +314,11 @@ enum r5dev_flags {
 			 */
 	R5_Discard,	/* Discard the stripe */
 	R5_SkipCopy,	/* Don't copy data from bio to stripe cache */
+	R5_InJournal,	/* data being written is in the journal device.
+			 * if R5_InJournal is set for parity pd_idx, all the
+			 * data and parity being written are in the journal
+			 * device
+			 */
 };
 
 /*
@@ -345,7 +351,23 @@ enum {
 	STRIPE_BITMAP_PENDING,	/* Being added to bitmap, don't add
 				 * to batch yet.
 				 */
-	STRIPE_LOG_TRAPPED, /* trapped into log */
+	STRIPE_LOG_TRAPPED,	/* trapped into log (see raid5-cache.c)
+				 * this bit is used in two scenarios:
+				 *
+				 * 1. write-out phase
+				 *  set in first entry of r5l_write_stripe
+				 *  clear in second entry of r5l_write_stripe
+				 *  used to bypass logic in handle_stripe
+				 *
+				 * 2. caching phase
+				 *  set in r5c_try_caching_write()
+				 *  clear when journal write is done
+				 *  used to initiate r5c_cache_data()
+				 *  also used to bypass logic in handle_stripe
+				 */
+	STRIPE_R5C_CACHING,	/* the stripe is in caching phase
+				 * see more detail in the raid5-cache.c
+				 */
 };
 
 #define STRIPE_EXPAND_SYNC_FLAGS \
@@ -710,4 +732,11 @@ extern void r5l_stripe_write_finished(struct stripe_head *sh);
 extern int r5l_handle_flush_request(struct r5l_log *log, struct bio *bio);
 extern void r5l_quiesce(struct r5l_log *log, int state);
 extern bool r5l_log_disk_error(struct r5conf *conf);
+extern bool r5c_is_writeback(struct r5l_log *log);
+extern int
+r5c_try_caching_write(struct r5conf *conf, struct stripe_head *sh,
+		      struct stripe_head_state *s, int disks);
+extern void
+r5c_finish_stripe_write_out(struct r5conf *conf, struct stripe_head *sh,
+			    struct stripe_head_state *s);
 #endif