Merge tag 'md/4.14-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/shli/md

Pull MD updates from Shaohua Li:
 "This update mainly fixes bugs:

   - Make raid5 ppl support several ppl from Pawel

   - Several raid5-cache bug fixes from Song

   - Bitmap fixes from Neil and Me

   - One raid1/10 regression fix since 4.12 from Me

   - Other small fixes and cleanup"

* tag 'md/4.14-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/shli/md:
  md/bitmap: disable bitmap_resize for file-backed bitmaps.
  raid5-ppl: Recovery support for multiple partial parity logs
  md: Runtime support for multiple ppls
  md/raid0: attach correct cgroup info in bio
  lib/raid6: align AVX512 constants to 512 bits, not bytes
  raid5: remove raid5_build_block
  md/r5cache: call mddev_lock/unlock() in r5c_journal_mode_show
  md: replace seq_release_private with seq_release
  md: notify about new spare disk in the container
  md/raid1/10: reset bio allocated from mempool
  md/raid5: release/flush io in raid5_do_work()
  md/bitmap: copy correct data for bitmap super

1 files changed, 130 insertions, 41 deletions

diff --git a/drivers/md/raid5-ppl.c b/drivers/md/raid5-ppl.c
index 1e237c40d6fa..cd026c88f7ef 100644
--- a/drivers/md/raid5-ppl.c
+++ b/drivers/md/raid5-ppl.c
@@ -87,6 +87,8 @@
  * The current io_unit accepting new stripes is always at the end of the list.
  */
 
+#define PPL_SPACE_SIZE (128 * 1024)
+
 struct ppl_conf {
 	struct mddev *mddev;
 
@@ -122,6 +124,10 @@ struct ppl_log {
 					 * always at the end of io_list */
 	spinlock_t io_list_lock;
 	struct list_head io_list;	/* all io_units of this log */
+
+	sector_t next_io_sector;
+	unsigned int entry_space;
+	bool use_multippl;
 };
 
 #define PPL_IO_INLINE_BVECS 32
@@ -264,13 +270,12 @@ static int ppl_log_stripe(struct ppl_log *log, struct stripe_head *sh)
 	int i;
 	sector_t data_sector = 0;
 	int data_disks = 0;
-	unsigned int entry_space = (log->rdev->ppl.size << 9) - PPL_HEADER_SIZE;
 	struct r5conf *conf = sh->raid_conf;
 
 	pr_debug("%s: stripe: %llu\n", __func__, (unsigned long long)sh->sector);
 
 	/* check if current io_unit is full */
-	if (io && (io->pp_size == entry_space ||
+	if (io && (io->pp_size == log->entry_space ||
 		   io->entries_count == PPL_HDR_MAX_ENTRIES)) {
 		pr_debug("%s: add io_unit blocked by seq: %llu\n",
 			 __func__, io->seq);
@@ -451,12 +456,25 @@ static void ppl_submit_iounit(struct ppl_io_unit *io)
 	pplhdr->entries_count = cpu_to_le32(io->entries_count);
 	pplhdr->checksum = cpu_to_le32(~crc32c_le(~0, pplhdr, PPL_HEADER_SIZE));
 
+	/* Rewind the buffer if current PPL is larger then remaining space */
+	if (log->use_multippl &&
+	    log->rdev->ppl.sector + log->rdev->ppl.size - log->next_io_sector <
+	    (PPL_HEADER_SIZE + io->pp_size) >> 9)
+		log->next_io_sector = log->rdev->ppl.sector;
+
+
 	bio->bi_end_io = ppl_log_endio;
 	bio->bi_opf = REQ_OP_WRITE | REQ_FUA;
 	bio_set_dev(bio, log->rdev->bdev);
-	bio->bi_iter.bi_sector = log->rdev->ppl.sector;
+	bio->bi_iter.bi_sector = log->next_io_sector;
 	bio_add_page(bio, io->header_page, PAGE_SIZE, 0);
 
+	pr_debug("%s: log->current_io_sector: %llu\n", __func__,
+	    (unsigned long long)log->next_io_sector);
+
+	if (log->use_multippl)
+		log->next_io_sector += (PPL_HEADER_SIZE + io->pp_size) >> 9;
+
 	list_for_each_entry(sh, &io->stripe_list, log_list) {
 		/* entries for full stripe writes have no partial parity */
 		if (test_bit(STRIPE_FULL_WRITE, &sh->state))
@@ -813,12 +831,14 @@ out:
 	return ret;
 }
 
-static int ppl_recover(struct ppl_log *log, struct ppl_header *pplhdr)
+static int ppl_recover(struct ppl_log *log, struct ppl_header *pplhdr,
+		       sector_t offset)
 {
 	struct ppl_conf *ppl_conf = log->ppl_conf;
 	struct md_rdev *rdev = log->rdev;
 	struct mddev *mddev = rdev->mddev;
-	sector_t ppl_sector = rdev->ppl.sector + (PPL_HEADER_SIZE >> 9);
+	sector_t ppl_sector = rdev->ppl.sector + offset +
+			      (PPL_HEADER_SIZE >> 9);
 	struct page *page;
 	int i;
 	int ret = 0;
@@ -902,6 +922,9 @@ static int ppl_write_empty_header(struct ppl_log *log)
 		return -ENOMEM;
 
 	pplhdr = page_address(page);
+	/* zero out PPL space to avoid collision with old PPLs */
+	blkdev_issue_zeroout(rdev->bdev, rdev->ppl.sector,
+			    log->rdev->ppl.size, GFP_NOIO, 0);
 	memset(pplhdr->reserved, 0xff, PPL_HDR_RESERVED);
 	pplhdr->signature = cpu_to_le32(log->ppl_conf->signature);
 	pplhdr->checksum = cpu_to_le32(~crc32c_le(~0, pplhdr, PAGE_SIZE));
@@ -922,63 +945,110 @@ static int ppl_load_distributed(struct ppl_log *log)
 	struct ppl_conf *ppl_conf = log->ppl_conf;
 	struct md_rdev *rdev = log->rdev;
 	struct mddev *mddev = rdev->mddev;
-	struct page *page;
-	struct ppl_header *pplhdr;
+	struct page *page, *page2, *tmp;
+	struct ppl_header *pplhdr = NULL, *prev_pplhdr = NULL;
 	u32 crc, crc_stored;
 	u32 signature;
-	int ret = 0;
+	int ret = 0, i;
+	sector_t pplhdr_offset = 0, prev_pplhdr_offset = 0;
 
 	pr_debug("%s: disk: %d\n", __func__, rdev->raid_disk);
-
-	/* read PPL header */
+	/* read PPL headers, find the recent one */
 	page = alloc_page(GFP_KERNEL);
 	if (!page)
 		return -ENOMEM;
 
-	if (!sync_page_io(rdev, rdev->ppl.sector - rdev->data_offset,
-			  PAGE_SIZE, page, REQ_OP_READ, 0, false)) {
-		md_error(mddev, rdev);
-		ret = -EIO;
-		goto out;
+	page2 = alloc_page(GFP_KERNEL);
+	if (!page2) {
+		__free_page(page);
+		return -ENOMEM;
 	}
-	pplhdr = page_address(page);
 
-	/* check header validity */
-	crc_stored = le32_to_cpu(pplhdr->checksum);
-	pplhdr->checksum = 0;
-	crc = ~crc32c_le(~0, pplhdr, PAGE_SIZE);
+	/* searching ppl area for latest ppl */
+	while (pplhdr_offset < rdev->ppl.size - (PPL_HEADER_SIZE >> 9)) {
+		if (!sync_page_io(rdev,
+				  rdev->ppl.sector - rdev->data_offset +
+				  pplhdr_offset, PAGE_SIZE, page, REQ_OP_READ,
+				  0, false)) {
+			md_error(mddev, rdev);
+			ret = -EIO;
+			/* if not able to read - don't recover any PPL */
+			pplhdr = NULL;
+			break;
+		}
+		pplhdr = page_address(page);
+
+		/* check header validity */
+		crc_stored = le32_to_cpu(pplhdr->checksum);
+		pplhdr->checksum = 0;
+		crc = ~crc32c_le(~0, pplhdr, PAGE_SIZE);
+
+		if (crc_stored != crc) {
+			pr_debug("%s: ppl header crc does not match: stored: 0x%x calculated: 0x%x (offset: %llu)\n",
+				 __func__, crc_stored, crc,
+				 (unsigned long long)pplhdr_offset);
+			pplhdr = prev_pplhdr;
+			pplhdr_offset = prev_pplhdr_offset;
+			break;
+		}
 
-	if (crc_stored != crc) {
-		pr_debug("%s: ppl header crc does not match: stored: 0x%x calculated: 0x%x\n",
-			 __func__, crc_stored, crc);
-		ppl_conf->mismatch_count++;
-		goto out;
-	}
+		signature = le32_to_cpu(pplhdr->signature);
 
-	signature = le32_to_cpu(pplhdr->signature);
+		if (mddev->external) {
+			/*
+			 * For external metadata the header signature is set and
+			 * validated in userspace.
+			 */
+			ppl_conf->signature = signature;
+		} else if (ppl_conf->signature != signature) {
+			pr_debug("%s: ppl header signature does not match: stored: 0x%x configured: 0x%x (offset: %llu)\n",
+				 __func__, signature, ppl_conf->signature,
+				 (unsigned long long)pplhdr_offset);
+			pplhdr = prev_pplhdr;
+			pplhdr_offset = prev_pplhdr_offset;
+			break;
+		}
 
-	if (mddev->external) {
-		/*
-		 * For external metadata the header signature is set and
-		 * validated in userspace.
-		 */
-		ppl_conf->signature = signature;
-	} else if (ppl_conf->signature != signature) {
-		pr_debug("%s: ppl header signature does not match: stored: 0x%x configured: 0x%x\n",
-			 __func__, signature, ppl_conf->signature);
-		ppl_conf->mismatch_count++;
-		goto out;
+		if (prev_pplhdr && le64_to_cpu(prev_pplhdr->generation) >
+		    le64_to_cpu(pplhdr->generation)) {
+			/* previous was newest */
+			pplhdr = prev_pplhdr;
+			pplhdr_offset = prev_pplhdr_offset;
+			break;
+		}
+
+		prev_pplhdr_offset = pplhdr_offset;
+		prev_pplhdr = pplhdr;
+
+		tmp = page;
+		page = page2;
+		page2 = tmp;
+
+		/* calculate next potential ppl offset */
+		for (i = 0; i < le32_to_cpu(pplhdr->entries_count); i++)
+			pplhdr_offset +=
+			    le32_to_cpu(pplhdr->entries[i].pp_size) >> 9;
+		pplhdr_offset += PPL_HEADER_SIZE >> 9;
 	}
 
+	/* no valid ppl found */
+	if (!pplhdr)
+		ppl_conf->mismatch_count++;
+	else
+		pr_debug("%s: latest PPL found at offset: %llu, with generation: %llu\n",
+		    __func__, (unsigned long long)pplhdr_offset,
+		    le64_to_cpu(pplhdr->generation));
+
 	/* attempt to recover from log if we are starting a dirty array */
-	if (!mddev->pers && mddev->recovery_cp != MaxSector)
-		ret = ppl_recover(log, pplhdr);
-out:
+	if (pplhdr && !mddev->pers && mddev->recovery_cp != MaxSector)
+		ret = ppl_recover(log, pplhdr, pplhdr_offset);
+
 	/* write empty header if we are starting the array */
 	if (!ret && !mddev->pers)
 		ret = ppl_write_empty_header(log);
 
 	__free_page(page);
+	__free_page(page2);
 
 	pr_debug("%s: return: %d mismatch_count: %d recovered_entries: %d\n",
 		 __func__, ret, ppl_conf->mismatch_count,
@@ -1031,6 +1101,7 @@ static int ppl_load(struct ppl_conf *ppl_conf)
 static void __ppl_exit_log(struct ppl_conf *ppl_conf)
 {
 	clear_bit(MD_HAS_PPL, &ppl_conf->mddev->flags);
+	clear_bit(MD_HAS_MULTIPLE_PPLS, &ppl_conf->mddev->flags);
 
 	kfree(ppl_conf->child_logs);
 
@@ -1099,6 +1170,22 @@ static int ppl_validate_rdev(struct md_rdev *rdev)
 	return 0;
 }
 
+static void ppl_init_child_log(struct ppl_log *log, struct md_rdev *rdev)
+{
+	if ((rdev->ppl.size << 9) >= (PPL_SPACE_SIZE +
+				      PPL_HEADER_SIZE) * 2) {
+		log->use_multippl = true;
+		set_bit(MD_HAS_MULTIPLE_PPLS,
+			&log->ppl_conf->mddev->flags);
+		log->entry_space = PPL_SPACE_SIZE;
+	} else {
+		log->use_multippl = false;
+		log->entry_space = (log->rdev->ppl.size << 9) -
+				   PPL_HEADER_SIZE;
+	}
+	log->next_io_sector = rdev->ppl.sector;
+}
+
 int ppl_init_log(struct r5conf *conf)
 {
 	struct ppl_conf *ppl_conf;
@@ -1196,6 +1283,7 @@ int ppl_init_log(struct r5conf *conf)
 			q = bdev_get_queue(rdev->bdev);
 			if (test_bit(QUEUE_FLAG_WC, &q->queue_flags))
 				need_cache_flush = true;
+			ppl_init_child_log(log, rdev);
 		}
 	}
 
@@ -1261,6 +1349,7 @@ int ppl_modify_log(struct r5conf *conf, struct md_rdev *rdev, bool add)
 		if (!ret) {
 			log->rdev = rdev;
 			ret = ppl_write_empty_header(log);
+			ppl_init_child_log(log, rdev);
 		}
 	} else {
 		log->rdev = NULL;