Merge branch 'for-linus' of git://neil.brown.name/md

* 'for-linus' of git://neil.brown.name/md: (75 commits)
  md/raid10: handle further errors during fix_read_error better.
  md/raid10: Handle read errors during recovery better.
  md/raid10: simplify read error handling during recovery.
  md/raid10: record bad blocks due to write errors during resync/recovery.
  md/raid10:  attempt to fix read errors during resync/check
  md/raid10:  Handle write errors by updating badblock log.
  md/raid10: clear bad-block record when write succeeds.
  md/raid10: avoid writing to known bad blocks on known bad drives.
  md/raid10 record bad blocks as needed during recovery.
  md/raid10: avoid reading known bad blocks during resync/recovery.
  md/raid10 - avoid reading from known bad blocks - part 3
  md/raid10: avoid reading from known bad blocks - part 2
  md/raid10: avoid reading from known bad blocks - part 1
  md/raid10: Split handle_read_error out from raid10d.
  md/raid10: simplify/reindent some loops.
  md/raid5: Clear bad blocks on successful write.
  md/raid5.  Don't write to known bad block on doubtful devices.
  md/raid5: write errors should be recorded as bad blocks if possible.
  md/raid5: use bad-block log to improve handling of uncorrectable read errors.
  md/raid5: avoid reading from known bad blocks.
  ...

Documentation/md.txt

@@ -360,18 +360,20 @@ Each directory contains:
         A file recording the current state of the device in the array
 	which can be a comma separated list of
 	      faulty   - device has been kicked from active use due to
-                         a detected fault
+                         a detected fault or it has unacknowledged bad
+                         blocks
 	      in_sync  - device is a fully in-sync member of the array
 	      writemostly - device will only be subject to read
 		         requests if there are no other options.
 			 This applies only to raid1 arrays.
-	      blocked  - device has failed, metadata is "external",
-	                 and the failure hasn't been acknowledged yet.
+	      blocked  - device has failed, and the failure hasn't been
+			 acknowledged yet by the metadata handler.
 			 Writes that would write to this device if
 			 it were not faulty are blocked.
 	      spare    - device is working, but not a full member.
 			 This includes spares that are in the process
 			 of being recovered to
+	      write_error - device has ever seen a write error.
 	This list may grow in future.
 	This can be written to.
 	Writing "faulty"  simulates a failure on the device.
@@ -379,9 +381,11 @@ Each directory contains:
 	Writing "writemostly" sets the writemostly flag.
 	Writing "-writemostly" clears the writemostly flag.
 	Writing "blocked" sets the "blocked" flag.
-	Writing "-blocked" clears the "blocked" flag and allows writes
-		to complete.
+	Writing "-blocked" clears the "blocked" flags and allows writes
+		to complete and possibly simulates an error.
 	Writing "in_sync" sets the in_sync flag.
+	Writing "write_error" sets writeerrorseen flag.
+	Writing "-write_error" clears writeerrorseen flag.
 
 	This file responds to select/poll. Any change to 'faulty'
 	or 'blocked' causes an event.
@@ -419,7 +423,6 @@ Each directory contains:
         written, it will be rejected.
 
       recovery_start
-
         When the device is not 'in_sync', this records the number of
 	sectors from the start of the device which are known to be
 	correct.  This is normally zero, but during a recovery
@@ -435,6 +438,20 @@ Each directory contains:
 	Setting this to 'none' is equivalent to setting 'in_sync'.
 	Setting to any other value also clears the 'in_sync' flag.
 	
+      bad_blocks
+	This gives the list of all known bad blocks in the form of
+	start address and length (in sectors respectively). If output
+	is too big to fit in a page, it will be truncated. Writing
+	"sector length" to this file adds new acknowledged (i.e.
+	recorded to disk safely) bad blocks.
+
+      unacknowledged_bad_blocks
+	This gives the list of known-but-not-yet-saved-to-disk bad
+	blocks in the same form of 'bad_blocks'. If output is too big
+	to fit in a page, it will be truncated. Writing to this file
+	adds bad blocks without acknowledging them. This is largely
+	for testing.
+
 
 
 An active md device will also contain and entry for each active device

drivers/md/bitmap.c

@@ -29,7 +29,6 @@
 #include "md.h"
 #include "bitmap.h"
 
-#include <linux/dm-dirty-log.h>
 /* debug macros */
 
 #define DEBUG 0
@@ -775,10 +774,8 @@ static inline unsigned long file_page_offset(struct bitmap *bitmap, unsigned lon
  * 0 or page 1
  */
 static inline struct page *filemap_get_page(struct bitmap *bitmap,
-					unsigned long chunk)
+					    unsigned long chunk)
 {
-	if (bitmap->filemap == NULL)
-		return NULL;
 	if (file_page_index(bitmap, chunk) >= bitmap->file_pages)
 		return NULL;
 	return bitmap->filemap[file_page_index(bitmap, chunk)
@@ -878,28 +875,19 @@ enum bitmap_page_attr {
 static inline void set_page_attr(struct bitmap *bitmap, struct page *page,
 				enum bitmap_page_attr attr)
 {
-	if (page)
-		__set_bit((page->index<<2) + attr, bitmap->filemap_attr);
-	else
-		__set_bit(attr, &bitmap->logattrs);
+	__set_bit((page->index<<2) + attr, bitmap->filemap_attr);
 }
 
 static inline void clear_page_attr(struct bitmap *bitmap, struct page *page,
 				enum bitmap_page_attr attr)
 {
-	if (page)
-		__clear_bit((page->index<<2) + attr, bitmap->filemap_attr);
-	else
-		__clear_bit(attr, &bitmap->logattrs);
+	__clear_bit((page->index<<2) + attr, bitmap->filemap_attr);
 }
 
 static inline unsigned long test_page_attr(struct bitmap *bitmap, struct page *page,
 					   enum bitmap_page_attr attr)
 {
-	if (page)
-		return test_bit((page->index<<2) + attr, bitmap->filemap_attr);
-	else
-		return test_bit(attr, &bitmap->logattrs);
+	return test_bit((page->index<<2) + attr, bitmap->filemap_attr);
 }
 
 /*
@@ -912,30 +900,26 @@ static inline unsigned long test_page_attr(struct bitmap *bitmap, struct page *p
 static void bitmap_file_set_bit(struct bitmap *bitmap, sector_t block)
 {
 	unsigned long bit;
-	struct page *page = NULL;
+	struct page *page;
 	void *kaddr;
 	unsigned long chunk = block >> CHUNK_BLOCK_SHIFT(bitmap);
 
-	if (!bitmap->filemap) {
-		struct dm_dirty_log *log = bitmap->mddev->bitmap_info.log;
-		if (log)
-			log->type->mark_region(log, chunk);
-	} else {
+	if (!bitmap->filemap)
+		return;
 
-		page = filemap_get_page(bitmap, chunk);
-		if (!page)
-			return;
-		bit = file_page_offset(bitmap, chunk);
+	page = filemap_get_page(bitmap, chunk);
+	if (!page)
+		return;
+	bit = file_page_offset(bitmap, chunk);
 
-		/* set the bit */
-		kaddr = kmap_atomic(page, KM_USER0);
-		if (bitmap->flags & BITMAP_HOSTENDIAN)
-			set_bit(bit, kaddr);
-		else
-			__test_and_set_bit_le(bit, kaddr);
-		kunmap_atomic(kaddr, KM_USER0);
-		PRINTK("set file bit %lu page %lu\n", bit, page->index);
-	}
+	/* set the bit */
+	kaddr = kmap_atomic(page, KM_USER0);
+	if (bitmap->flags & BITMAP_HOSTENDIAN)
+		set_bit(bit, kaddr);
+	else
+		__set_bit_le(bit, kaddr);
+	kunmap_atomic(kaddr, KM_USER0);
+	PRINTK("set file bit %lu page %lu\n", bit, page->index);
 	/* record page number so it gets flushed to disk when unplug occurs */
 	set_page_attr(bitmap, page, BITMAP_PAGE_DIRTY);
 }
@@ -952,16 +936,6 @@ void bitmap_unplug(struct bitmap *bitmap)
 
 	if (!bitmap)
 		return;
-	if (!bitmap->filemap) {
-		/* Must be using a dirty_log */
-		struct dm_dirty_log *log = bitmap->mddev->bitmap_info.log;
-		dirty = test_and_clear_bit(BITMAP_PAGE_DIRTY, &bitmap->logattrs);
-		need_write = test_and_clear_bit(BITMAP_PAGE_NEEDWRITE, &bitmap->logattrs);
-		if (dirty || need_write)
-			if (log->type->flush(log))
-				bitmap->flags |= BITMAP_WRITE_ERROR;
-		goto out;
-	}
 
 	/* look at each page to see if there are any set bits that need to be
 	 * flushed out to disk */
@@ -990,7 +964,6 @@ void bitmap_unplug(struct bitmap *bitmap)
 		else
 			md_super_wait(bitmap->mddev);
 	}
-out:
 	if (bitmap->flags & BITMAP_WRITE_ERROR)
 		bitmap_file_kick(bitmap);
 }
@@ -1199,7 +1172,6 @@ void bitmap_daemon_work(mddev_t *mddev)
 	struct page *page = NULL, *lastpage = NULL;
 	sector_t blocks;
 	void *paddr;
-	struct dm_dirty_log *log = mddev->bitmap_info.log;
 
 	/* Use a mutex to guard daemon_work against
 	 * bitmap_destroy.
@@ -1224,12 +1196,11 @@ void bitmap_daemon_work(mddev_t *mddev)
 	spin_lock_irqsave(&bitmap->lock, flags);
 	for (j = 0; j < bitmap->chunks; j++) {
 		bitmap_counter_t *bmc;
-		if (!bitmap->filemap) {
-			if (!log)
-				/* error or shutdown */
-				break;
-		} else
-			page = filemap_get_page(bitmap, j);
+		if (!bitmap->filemap)
+			/* error or shutdown */
+			break;
+
+		page = filemap_get_page(bitmap, j);
 
 		if (page != lastpage) {
 			/* skip this page unless it's marked as needing cleaning */
@@ -1298,17 +1269,16 @@ void bitmap_daemon_work(mddev_t *mddev)
 						  -1);
 
 				/* clear the bit */
-				if (page) {
-					paddr = kmap_atomic(page, KM_USER0);
-					if (bitmap->flags & BITMAP_HOSTENDIAN)
-						clear_bit(file_page_offset(bitmap, j),
-							  paddr);
-					else
-						__test_and_clear_bit_le(file_page_offset(bitmap, j),
-							       paddr);
-					kunmap_atomic(paddr, KM_USER0);
-				} else
-					log->type->clear_region(log, j);
+				paddr = kmap_atomic(page, KM_USER0);
+				if (bitmap->flags & BITMAP_HOSTENDIAN)
+					clear_bit(file_page_offset(bitmap, j),
+						  paddr);
+				else
+					__clear_bit_le(
+							file_page_offset(bitmap,
+									 j),
+							paddr);
+				kunmap_atomic(paddr, KM_USER0);
 			}
 		} else
 			j |= PAGE_COUNTER_MASK;
@@ -1316,16 +1286,12 @@ void bitmap_daemon_work(mddev_t *mddev)
 	spin_unlock_irqrestore(&bitmap->lock, flags);
 
 	/* now sync the final page */
-	if (lastpage != NULL || log != NULL) {
+	if (lastpage != NULL) {
 		spin_lock_irqsave(&bitmap->lock, flags);
 		if (test_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE)) {
 			clear_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE);
 			spin_unlock_irqrestore(&bitmap->lock, flags);
-			if (lastpage)
-				write_page(bitmap, lastpage, 0);
-			else
-				if (log->type->flush(log))
-					bitmap->flags |= BITMAP_WRITE_ERROR;
+			write_page(bitmap, lastpage, 0);
 		} else {
 			set_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE);
 			spin_unlock_irqrestore(&bitmap->lock, flags);
@@ -1767,12 +1733,10 @@ int bitmap_create(mddev_t *mddev)
 	BUILD_BUG_ON(sizeof(bitmap_super_t) != 256);
 
 	if (!file
-	    && !mddev->bitmap_info.offset
-	    && !mddev->bitmap_info.log) /* bitmap disabled, nothing to do */
+	    && !mddev->bitmap_info.offset) /* bitmap disabled, nothing to do */
 		return 0;
 
 	BUG_ON(file && mddev->bitmap_info.offset);
-	BUG_ON(mddev->bitmap_info.offset && mddev->bitmap_info.log);
 
 	bitmap = kzalloc(sizeof(*bitmap), GFP_KERNEL);
 	if (!bitmap)
@@ -1863,6 +1827,7 @@ int bitmap_create(mddev_t *mddev)
 int bitmap_load(mddev_t *mddev)
 {
 	int err = 0;
+	sector_t start = 0;
 	sector_t sector = 0;
 	struct bitmap *bitmap = mddev->bitmap;
 
@@ -1881,24 +1846,14 @@ int bitmap_load(mddev_t *mddev)
 	}
 	bitmap_close_sync(bitmap);
 
-	if (mddev->bitmap_info.log) {
-		unsigned long i;
-		struct dm_dirty_log *log = mddev->bitmap_info.log;
-		for (i = 0; i < bitmap->chunks; i++)
-			if (!log->type->in_sync(log, i, 1))
-				bitmap_set_memory_bits(bitmap,
-						       (sector_t)i << CHUNK_BLOCK_SHIFT(bitmap),
-						       1);
-	} else {
-		sector_t start = 0;
-		if (mddev->degraded == 0
-		    || bitmap->events_cleared == mddev->events)
-			/* no need to keep dirty bits to optimise a
-			 * re-add of a missing device */
-			start = mddev->recovery_cp;
-
-		err = bitmap_init_from_disk(bitmap, start);
-	}
+	if (mddev->degraded == 0
+	    || bitmap->events_cleared == mddev->events)
+		/* no need to keep dirty bits to optimise a
+		 * re-add of a missing device */
+		start = mddev->recovery_cp;
+
+	err = bitmap_init_from_disk(bitmap, start);
+
 	if (err)
 		goto out;
 

drivers/md/bitmap.h

@@ -212,10 +212,6 @@ struct bitmap {
 	unsigned long file_pages; /* number of pages in the file */
 	int last_page_size; /* bytes in the last page */
 
-	unsigned long logattrs; /* used when filemap_attr doesn't exist
-				 * because we are working with a dirty_log
-				 */
-
 	unsigned long flags;
 
 	int allclean;
@@ -237,7 +233,6 @@ struct bitmap {
 	wait_queue_head_t behind_wait;
 
 	struct sysfs_dirent *sysfs_can_clear;
-
 };
 
 /* the bitmap API */

drivers/md/md.c

@@ -215,6 +215,55 @@ struct bio *bio_clone_mddev(struct bio *bio, gfp_t gfp_mask,
 }
 EXPORT_SYMBOL_GPL(bio_clone_mddev);
 
+void md_trim_bio(struct bio *bio, int offset, int size)
+{
+	/* 'bio' is a cloned bio which we need to trim to match
+	 * the given offset and size.
+	 * This requires adjusting bi_sector, bi_size, and bi_io_vec
+	 */
+	int i;
+	struct bio_vec *bvec;
+	int sofar = 0;
+
+	size <<= 9;
+	if (offset == 0 && size == bio->bi_size)
+		return;
+
+	bio->bi_sector += offset;
+	bio->bi_size = size;
+	offset <<= 9;
+	clear_bit(BIO_SEG_VALID, &bio->bi_flags);
+
+	while (bio->bi_idx < bio->bi_vcnt &&
+	       bio->bi_io_vec[bio->bi_idx].bv_len <= offset) {
+		/* remove this whole bio_vec */
+		offset -= bio->bi_io_vec[bio->bi_idx].bv_len;
+		bio->bi_idx++;
+	}
+	if (bio->bi_idx < bio->bi_vcnt) {
+		bio->bi_io_vec[bio->bi_idx].bv_offset += offset;
+		bio->bi_io_vec[bio->bi_idx].bv_len -= offset;
+	}
+	/* avoid any complications with bi_idx being non-zero*/
+	if (bio->bi_idx) {
+		memmove(bio->bi_io_vec, bio->bi_io_vec+bio->bi_idx,
+			(bio->bi_vcnt - bio->bi_idx) * sizeof(struct bio_vec));
+		bio->bi_vcnt -= bio->bi_idx;
+		bio->bi_idx = 0;
+	}
+	/* Make sure vcnt and last bv are not too big */
+	bio_for_each_segment(bvec, bio, i) {
+		if (sofar + bvec->bv_len > size)
+			bvec->bv_len = size - sofar;
+		if (bvec->bv_len == 0) {
+			bio->bi_vcnt = i;
+			break;
+		}
+		sofar += bvec->bv_len;
+	}
+}
+EXPORT_SYMBOL_GPL(md_trim_bio);
+
 /*
  * We have a system wide 'event count' that is incremented
  * on any 'interesting' event, and readers of /proc/mdstat
@@ -757,6 +806,10 @@ static void free_disk_sb(mdk_rdev_t * rdev)
 		rdev->sb_start = 0;
 		rdev->sectors = 0;
 	}
+	if (rdev->bb_page) {
+		put_page(rdev->bb_page);
+		rdev->bb_page = NULL;
+	}
 }
 
 
@@ -1025,7 +1078,7 @@ static int super_90_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version
 	ret = -EINVAL;
 
 	bdevname(rdev->bdev, b);
-	sb = (mdp_super_t*)page_address(rdev->sb_page);
+	sb = page_address(rdev->sb_page);
 
 	if (sb->md_magic != MD_SB_MAGIC) {
 		printk(KERN_ERR "md: invalid raid superblock magic on %s\n",
@@ -1054,6 +1107,7 @@ static int super_90_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version
 	rdev->preferred_minor = sb->md_minor;
 	rdev->data_offset = 0;
 	rdev->sb_size = MD_SB_BYTES;
+	rdev->badblocks.shift = -1;
 
 	if (sb->level == LEVEL_MULTIPATH)
 		rdev->desc_nr = -1;
@@ -1064,7 +1118,7 @@ static int super_90_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version
 		ret = 1;
 	} else {
 		__u64 ev1, ev2;
-		mdp_super_t *refsb = (mdp_super_t*)page_address(refdev->sb_page);
+		mdp_super_t *refsb = page_address(refdev->sb_page);
 		if (!uuid_equal(refsb, sb)) {
 			printk(KERN_WARNING "md: %s has different UUID to %s\n",
 				b, bdevname(refdev->bdev,b2));
@@ -1099,7 +1153,7 @@ static int super_90_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version
 static int super_90_validate(mddev_t *mddev, mdk_rdev_t *rdev)
 {
 	mdp_disk_t *desc;
-	mdp_super_t *sb = (mdp_super_t *)page_address(rdev->sb_page);
+	mdp_super_t *sb = page_address(rdev->sb_page);
 	__u64 ev1 = md_event(sb);
 
 	rdev->raid_disk = -1;
@@ -1230,7 +1284,7 @@ static void super_90_sync(mddev_t *mddev, mdk_rdev_t *rdev)
 
 	rdev->sb_size = MD_SB_BYTES;
 
-	sb = (mdp_super_t*)page_address(rdev->sb_page);
+	sb = page_address(rdev->sb_page);
 
 	memset(sb, 0, sizeof(*sb));
 
@@ -1395,6 +1449,8 @@ static __le32 calc_sb_1_csum(struct mdp_superblock_1 * sb)
 	return cpu_to_le32(csum);
 }
 
+static int md_set_badblocks(struct badblocks *bb, sector_t s, int sectors,
+			    int acknowledged);
 static int super_1_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version)
 {
 	struct mdp_superblock_1 *sb;
@@ -1435,7 +1491,7 @@ static int super_1_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version)
 	if (ret) return ret;
 
 
-	sb = (struct mdp_superblock_1*)page_address(rdev->sb_page);
+	sb = page_address(rdev->sb_page);
 
 	if (sb->magic != cpu_to_le32(MD_SB_MAGIC) ||
 	    sb->major_version != cpu_to_le32(1) ||
@@ -1473,12 +1529,52 @@ static int super_1_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version)
 	else
 		rdev->desc_nr = le32_to_cpu(sb->dev_number);
 
+	if (!rdev->bb_page) {
+		rdev->bb_page = alloc_page(GFP_KERNEL);
+		if (!rdev->bb_page)
+			return -ENOMEM;
+	}
+	if ((le32_to_cpu(sb->feature_map) & MD_FEATURE_BAD_BLOCKS) &&
+	    rdev->badblocks.count == 0) {
+		/* need to load the bad block list.
+		 * Currently we limit it to one page.
+		 */
+		s32 offset;
+		sector_t bb_sector;
+		u64 *bbp;
+		int i;
+		int sectors = le16_to_cpu(sb->bblog_size);
+		if (sectors > (PAGE_SIZE / 512))
+			return -EINVAL;
+		offset = le32_to_cpu(sb->bblog_offset);
+		if (offset == 0)
+			return -EINVAL;
+		bb_sector = (long long)offset;
+		if (!sync_page_io(rdev, bb_sector, sectors << 9,
+				  rdev->bb_page, READ, true))
+			return -EIO;
+		bbp = (u64 *)page_address(rdev->bb_page);
+		rdev->badblocks.shift = sb->bblog_shift;
+		for (i = 0 ; i < (sectors << (9-3)) ; i++, bbp++) {
+			u64 bb = le64_to_cpu(*bbp);
+			int count = bb & (0x3ff);
+			u64 sector = bb >> 10;
+			sector <<= sb->bblog_shift;
+			count <<= sb->bblog_shift;
+			if (bb + 1 == 0)
+				break;
+			if (md_set_badblocks(&rdev->badblocks,
+					     sector, count, 1) == 0)
+				return -EINVAL;
+		}
+	} else if (sb->bblog_offset == 0)
+		rdev->badblocks.shift = -1;
+
 	if (!refdev) {
 		ret = 1;
 	} else {
 		__u64 ev1, ev2;
-		struct mdp_superblock_1 *refsb = 
-			(struct mdp_superblock_1*)page_address(refdev->sb_page);
+		struct mdp_superblock_1 *refsb = page_address(refdev->sb_page);
 
 		if (memcmp(sb->set_uuid, refsb->set_uuid, 16) != 0 ||
 		    sb->level != refsb->level ||
@@ -1513,7 +1609,7 @@ static int super_1_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version)
 
 static int super_1_validate(mddev_t *mddev, mdk_rdev_t *rdev)
 {
-	struct mdp_superblock_1 *sb = (struct mdp_superblock_1*)page_address(rdev->sb_page);
+	struct mdp_superblock_1 *sb = page_address(rdev->sb_page);
 	__u64 ev1 = le64_to_cpu(sb->events);
 
 	rdev->raid_disk = -1;
@@ -1619,13 +1715,12 @@ static void super_1_sync(mddev_t *mddev, mdk_rdev_t *rdev)
 	int max_dev, i;
 	/* make rdev->sb match mddev and rdev data. */
 
-	sb = (struct mdp_superblock_1*)page_address(rdev->sb_page);
+	sb = page_address(rdev->sb_page);
 
 	sb->feature_map = 0;
 	sb->pad0 = 0;
 	sb->recovery_offset = cpu_to_le64(0);
 	memset(sb->pad1, 0, sizeof(sb->pad1));
-	memset(sb->pad2, 0, sizeof(sb->pad2));
 	memset(sb->pad3, 0, sizeof(sb->pad3));
 
 	sb->utime = cpu_to_le64((__u64)mddev->utime);
@@ -1665,6 +1760,40 @@ static void super_1_sync(mddev_t *mddev, mdk_rdev_t *rdev)
 		sb->new_chunk = cpu_to_le32(mddev->new_chunk_sectors);
 	}
 
+	if (rdev->badblocks.count == 0)
+		/* Nothing to do for bad blocks*/ ;
+	else if (sb->bblog_offset == 0)
+		/* Cannot record bad blocks on this device */
+		md_error(mddev, rdev);
+	else {
+		struct badblocks *bb = &rdev->badblocks;
+		u64 *bbp = (u64 *)page_address(rdev->bb_page);
+		u64 *p = bb->page;
+		sb->feature_map |= cpu_to_le32(MD_FEATURE_BAD_BLOCKS);
+		if (bb->changed) {
+			unsigned seq;
+
+retry:
+			seq = read_seqbegin(&bb->lock);
+
+			memset(bbp, 0xff, PAGE_SIZE);
+
+			for (i = 0 ; i < bb->count ; i++) {
+				u64 internal_bb = *p++;
+				u64 store_bb = ((BB_OFFSET(internal_bb) << 10)
+						| BB_LEN(internal_bb));
+				*bbp++ = cpu_to_le64(store_bb);
+			}
+			if (read_seqretry(&bb->lock, seq))
+				goto retry;
+
+			bb->sector = (rdev->sb_start +
+				      (int)le32_to_cpu(sb->bblog_offset));
+			bb->size = le16_to_cpu(sb->bblog_size);
+			bb->changed = 0;
+		}
+	}
+
 	max_dev = 0;
 	list_for_each_entry(rdev2, &mddev->disks, same_set)
 		if (rdev2->desc_nr+1 > max_dev)
@@ -1724,7 +1853,7 @@ super_1_rdev_size_change(mdk_rdev_t *rdev, sector_t num_sectors)
 			num_sectors = max_sectors;
 		rdev->sb_start = sb_start;
 	}
-	sb = (struct mdp_superblock_1 *) page_address(rdev->sb_page);
+	sb = page_address(rdev->sb_page);
 	sb->data_size = cpu_to_le64(num_sectors);
 	sb->super_offset = rdev->sb_start;
 	sb->sb_csum = calc_sb_1_csum(sb);
@@ -1922,7 +2051,7 @@ static int bind_rdev_to_array(mdk_rdev_t * rdev, mddev_t * mddev)
 	bd_link_disk_holder(rdev->bdev, mddev->gendisk);
 
 	/* May as well allow recovery to be retried once */
-	mddev->recovery_disabled = 0;
+	mddev->recovery_disabled++;
 
 	return 0;
 
@@ -1953,6 +2082,9 @@ static void unbind_rdev_from_array(mdk_rdev_t * rdev)
 	sysfs_remove_link(&rdev->kobj, "block");
 	sysfs_put(rdev->sysfs_state);
 	rdev->sysfs_state = NULL;
+	kfree(rdev->badblocks.page);
+	rdev->badblocks.count = 0;
+	rdev->badblocks.page = NULL;
 	/* We need to delay this, otherwise we can deadlock when
 	 * writing to 'remove' to "dev/state".  We also need
 	 * to delay it due to rcu usage.
@@ -2127,10 +2259,10 @@ static void print_rdev(mdk_rdev_t *rdev, int major_version)
 		printk(KERN_INFO "md: rdev superblock (MJ:%d):\n", major_version);
 		switch (major_version) {
 		case 0:
-			print_sb_90((mdp_super_t*)page_address(rdev->sb_page));
+			print_sb_90(page_address(rdev->sb_page));
 			break;
 		case 1:
-			print_sb_1((struct mdp_superblock_1 *)page_address(rdev->sb_page));
+			print_sb_1(page_address(rdev->sb_page));
 			break;
 		}
 	} else
@@ -2194,6 +2326,7 @@ static void md_update_sb(mddev_t * mddev, int force_change)
 	mdk_rdev_t *rdev;
 	int sync_req;
 	int nospares = 0;
+	int any_badblocks_changed = 0;
 
 repeat:
 	/* First make sure individual recovery_offsets are correct */
@@ -2208,8 +2341,18 @@ repeat:
 	if (!mddev->persistent) {
 		clear_bit(MD_CHANGE_CLEAN, &mddev->flags);
 		clear_bit(MD_CHANGE_DEVS, &mddev->flags);
-		if (!mddev->external)
+		if (!mddev->external) {
 			clear_bit(MD_CHANGE_PENDING, &mddev->flags);
+			list_for_each_entry(rdev, &mddev->disks, same_set) {
+				if (rdev->badblocks.changed) {
+					md_ack_all_badblocks(&rdev->badblocks);
+					md_error(mddev, rdev);
+				}
+				clear_bit(Blocked, &rdev->flags);
+				clear_bit(BlockedBadBlocks, &rdev->flags);
+				wake_up(&rdev->blocked_wait);
+			}
+		}
 		wake_up(&mddev->sb_wait);
 		return;
 	}
@@ -2265,6 +2408,14 @@ repeat:
 		MD_BUG();
 		mddev->events --;
 	}
+
+	list_for_each_entry(rdev, &mddev->disks, same_set) {
+		if (rdev->badblocks.changed)
+			any_badblocks_changed++;
+		if (test_bit(Faulty, &rdev->flags))
+			set_bit(FaultRecorded, &rdev->flags);
+	}
+
 	sync_sbs(mddev, nospares);
 	spin_unlock_irq(&mddev->write_lock);
 
@@ -2290,6 +2441,13 @@ repeat:
 				bdevname(rdev->bdev,b),
 				(unsigned long long)rdev->sb_start);
 			rdev->sb_events = mddev->events;
+			if (rdev->badblocks.size) {
+				md_super_write(mddev, rdev,
+					       rdev->badblocks.sector,
+					       rdev->badblocks.size << 9,
+					       rdev->bb_page);
+				rdev->badblocks.size = 0;
+			}
 
 		} else
 			dprintk(")\n");
@@ -2313,6 +2471,15 @@ repeat:
 	if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery))
 		sysfs_notify(&mddev->kobj, NULL, "sync_completed");
 
+	list_for_each_entry(rdev, &mddev->disks, same_set) {
+		if (test_and_clear_bit(FaultRecorded, &rdev->flags))
+			clear_bit(Blocked, &rdev->flags);
+
+		if (any_badblocks_changed)
+			md_ack_all_badblocks(&rdev->badblocks);
+		clear_bit(BlockedBadBlocks, &rdev->flags);
+		wake_up(&rdev->blocked_wait);
+	}
 }
 
 /* words written to sysfs files may, or may not, be \n terminated.
@@ -2347,7 +2514,8 @@ state_show(mdk_rdev_t *rdev, char *page)
 	char *sep = "";
 	size_t len = 0;
 
-	if (test_bit(Faulty, &rdev->flags)) {
+	if (test_bit(Faulty, &rdev->flags) ||
+	    rdev->badblocks.unacked_exist) {
 		len+= sprintf(page+len, "%sfaulty",sep);
 		sep = ",";
 	}
@@ -2359,7 +2527,8 @@ state_show(mdk_rdev_t *rdev, char *page)
 		len += sprintf(page+len, "%swrite_mostly",sep);
 		sep = ",";
 	}
-	if (test_bit(Blocked, &rdev->flags)) {
+	if (test_bit(Blocked, &rdev->flags) ||
+	    rdev->badblocks.unacked_exist) {
 		len += sprintf(page+len, "%sblocked", sep);
 		sep = ",";
 	}
@@ -2368,6 +2537,10 @@ state_show(mdk_rdev_t *rdev, char *page)
 		len += sprintf(page+len, "%sspare", sep);
 		sep = ",";
 	}
+	if (test_bit(WriteErrorSeen, &rdev->flags)) {
+		len += sprintf(page+len, "%swrite_error", sep);
+		sep = ",";
+	}
 	return len+sprintf(page+len, "\n");
 }
 
@@ -2375,13 +2548,15 @@ static ssize_t
 state_store(mdk_rdev_t *rdev, const char *buf, size_t len)
 {
 	/* can write
-	 *  faulty  - simulates and error
+	 *  faulty  - simulates an error
 	 *  remove  - disconnects the device
 	 *  writemostly - sets write_mostly
 	 *  -writemostly - clears write_mostly
-	 *  blocked - sets the Blocked flag
-	 *  -blocked - clears the Blocked flag
+	 *  blocked - sets the Blocked flags
+	 *  -blocked - clears the Blocked and possibly simulates an error
 	 *  insync - sets Insync providing device isn't active
+	 *  write_error - sets WriteErrorSeen
+	 *  -write_error - clears WriteErrorSeen
 	 */
 	int err = -EINVAL;
 	if (cmd_match(buf, "faulty") && rdev->mddev->pers) {
@@ -2408,7 +2583,15 @@ state_store(mdk_rdev_t *rdev, const char *buf, size_t len)
 		set_bit(Blocked, &rdev->flags);
 		err = 0;
 	} else if (cmd_match(buf, "-blocked")) {
+		if (!test_bit(Faulty, &rdev->flags) &&
+		    test_bit(BlockedBadBlocks, &rdev->flags)) {
+			/* metadata handler doesn't understand badblocks,
+			 * so we need to fail the device
+			 */
+			md_error(rdev->mddev, rdev);
+		}
 		clear_bit(Blocked, &rdev->flags);
+		clear_bit(BlockedBadBlocks, &rdev->flags);
 		wake_up(&rdev->blocked_wait);
 		set_bit(MD_RECOVERY_NEEDED, &rdev->mddev->recovery);
 		md_wakeup_thread(rdev->mddev->thread);
@@ -2417,6 +2600,12 @@ state_store(mdk_rdev_t *rdev, const char *buf, size_t len)
 	} else if (cmd_match(buf, "insync") && rdev->raid_disk == -1) {
 		set_bit(In_sync, &rdev->flags);
 		err = 0;
+	} else if (cmd_match(buf, "write_error")) {
+		set_bit(WriteErrorSeen, &rdev->flags);
+		err = 0;
+	} else if (cmd_match(buf, "-write_error")) {
+		clear_bit(WriteErrorSeen, &rdev->flags);
+		err = 0;
 	}
 	if (!err)
 		sysfs_notify_dirent_safe(rdev->sysfs_state);
@@ -2459,7 +2648,6 @@ slot_store(mdk_rdev_t *rdev, const char *buf, size_t len)
 {
 	char *e;
 	int err;
-	char nm[20];
 	int slot = simple_strtoul(buf, &e, 10);
 	if (strncmp(buf, "none", 4)==0)
 		slot = -1;
@@ -2482,8 +2670,7 @@ slot_store(mdk_rdev_t *rdev, const char *buf, size_t len)
 			hot_remove_disk(rdev->mddev, rdev->raid_disk);
 		if (err)
 			return err;
-		sprintf(nm, "rd%d", rdev->raid_disk);
-		sysfs_remove_link(&rdev->mddev->kobj, nm);
+		sysfs_unlink_rdev(rdev->mddev, rdev);
 		rdev->raid_disk = -1;
 		set_bit(MD_RECOVERY_NEEDED, &rdev->mddev->recovery);
 		md_wakeup_thread(rdev->mddev->thread);
@@ -2522,8 +2709,7 @@ slot_store(mdk_rdev_t *rdev, const char *buf, size_t len)
 			return err;
 		} else
 			sysfs_notify_dirent_safe(rdev->sysfs_state);
-		sprintf(nm, "rd%d", rdev->raid_disk);
-		if (sysfs_create_link(&rdev->mddev->kobj, &rdev->kobj, nm))
+		if (sysfs_link_rdev(rdev->mddev, rdev))
 			/* failure here is OK */;
 		/* don't wakeup anyone, leave that to userspace. */
 	} else {
@@ -2712,6 +2898,39 @@ static ssize_t recovery_start_store(mdk_rdev_t *rdev, const char *buf, size_t le
 static struct rdev_sysfs_entry rdev_recovery_start =
 __ATTR(recovery_start, S_IRUGO|S_IWUSR, recovery_start_show, recovery_start_store);
 
+
+static ssize_t
+badblocks_show(struct badblocks *bb, char *page, int unack);
+static ssize_t
+badblocks_store(struct badblocks *bb, const char *page, size_t len, int unack);
+
+static ssize_t bb_show(mdk_rdev_t *rdev, char *page)
+{
+	return badblocks_show(&rdev->badblocks, page, 0);
+}
+static ssize_t bb_store(mdk_rdev_t *rdev, const char *page, size_t len)
+{
+	int rv = badblocks_store(&rdev->badblocks, page, len, 0);
+	/* Maybe that ack was all we needed */
+	if (test_and_clear_bit(BlockedBadBlocks, &rdev->flags))
+		wake_up(&rdev->blocked_wait);
+	return rv;
+}
+static struct rdev_sysfs_entry rdev_bad_blocks =
+__ATTR(bad_blocks, S_IRUGO|S_IWUSR, bb_show, bb_store);
+
+
+static ssize_t ubb_show(mdk_rdev_t *rdev, char *page)
+{
+	return badblocks_show(&rdev->badblocks, page, 1);
+}
+static ssize_t ubb_store(mdk_rdev_t *rdev, const char *page, size_t len)
+{
+	return badblocks_store(&rdev->badblocks, page, len, 1);
+}
+static struct rdev_sysfs_entry rdev_unack_bad_blocks =
+__ATTR(unacknowledged_bad_blocks, S_IRUGO|S_IWUSR, ubb_show, ubb_store);
+
 static struct attribute *rdev_default_attrs[] = {
 	&rdev_state.attr,
 	&rdev_errors.attr,
@@ -2719,6 +2938,8 @@ static struct attribute *rdev_default_attrs[] = {
 	&rdev_offset.attr,
 	&rdev_size.attr,
 	&rdev_recovery_start.attr,
+	&rdev_bad_blocks.attr,
+	&rdev_unack_bad_blocks.attr,
 	NULL,
 };
 static ssize_t
@@ -2782,7 +3003,7 @@ static struct kobj_type rdev_ktype = {
 	.default_attrs	= rdev_default_attrs,
 };
 
-void md_rdev_init(mdk_rdev_t *rdev)
+int md_rdev_init(mdk_rdev_t *rdev)
 {
 	rdev->desc_nr = -1;
 	rdev->saved_raid_disk = -1;
@@ -2792,12 +3013,27 @@ void md_rdev_init(mdk_rdev_t *rdev)
 	rdev->sb_events = 0;
 	rdev->last_read_error.tv_sec  = 0;
 	rdev->last_read_error.tv_nsec = 0;
+	rdev->sb_loaded = 0;
+	rdev->bb_page = NULL;
 	atomic_set(&rdev->nr_pending, 0);
 	atomic_set(&rdev->read_errors, 0);
 	atomic_set(&rdev->corrected_errors, 0);
 
 	INIT_LIST_HEAD(&rdev->same_set);
 	init_waitqueue_head(&rdev->blocked_wait);
+
+	/* Add space to store bad block list.
+	 * This reserves the space even on arrays where it cannot
+	 * be used - I wonder if that matters
+	 */
+	rdev->badblocks.count = 0;
+	rdev->badblocks.shift = 0;
+	rdev->badblocks.page = kmalloc(PAGE_SIZE, GFP_KERNEL);
+	seqlock_init(&rdev->badblocks.lock);
+	if (rdev->badblocks.page == NULL)
+		return -ENOMEM;
+
+	return 0;
 }
 EXPORT_SYMBOL_GPL(md_rdev_init);
 /*
@@ -2823,8 +3059,11 @@ static mdk_rdev_t *md_import_device(dev_t newdev, int super_format, int super_mi
 		return ERR_PTR(-ENOMEM);
 	}
 
-	md_rdev_init(rdev);
-	if ((err = alloc_disk_sb(rdev)))
+	err = md_rdev_init(rdev);
+	if (err)
+		goto abort_free;
+	err = alloc_disk_sb(rdev);
+	if (err)
 		goto abort_free;
 
 	err = lock_rdev(rdev, newdev, super_format == -2);
@@ -2860,15 +3099,17 @@ static mdk_rdev_t *md_import_device(dev_t newdev, int super_format, int super_mi
 			goto abort_free;
 		}
 	}
+	if (super_format == -1)
+		/* hot-add for 0.90, or non-persistent: so no badblocks */
+		rdev->badblocks.shift = -1;
 
 	return rdev;
 
 abort_free:
-	if (rdev->sb_page) {
-		if (rdev->bdev)
-			unlock_rdev(rdev);
-		free_disk_sb(rdev);
-	}
+	if (rdev->bdev)
+		unlock_rdev(rdev);
+	free_disk_sb(rdev);
+	kfree(rdev->badblocks.page);
 	kfree(rdev);
 	return ERR_PTR(err);
 }
@@ -3149,15 +3390,13 @@ level_store(mddev_t *mddev, const char *buf, size_t len)
 	}
 
 	list_for_each_entry(rdev, &mddev->disks, same_set) {
-		char nm[20];
 		if (rdev->raid_disk < 0)
 			continue;
 		if (rdev->new_raid_disk >= mddev->raid_disks)
 			rdev->new_raid_disk = -1;
 		if (rdev->new_raid_disk == rdev->raid_disk)
 			continue;
-		sprintf(nm, "rd%d", rdev->raid_disk);
-		sysfs_remove_link(&mddev->kobj, nm);
+		sysfs_unlink_rdev(mddev, rdev);
 	}
 	list_for_each_entry(rdev, &mddev->disks, same_set) {
 		if (rdev->raid_disk < 0)
@@ -3168,11 +3407,10 @@ level_store(mddev_t *mddev, const char *buf, size_t len)
 		if (rdev->raid_disk < 0)
 			clear_bit(In_sync, &rdev->flags);
 		else {
-			char nm[20];
-			sprintf(nm, "rd%d", rdev->raid_disk);
-			if(sysfs_create_link(&mddev->kobj, &rdev->kobj, nm))
-				printk("md: cannot register %s for %s after level change\n",
-				       nm, mdname(mddev));
+			if (sysfs_link_rdev(mddev, rdev))
+				printk(KERN_WARNING "md: cannot register rd%d"
+				       " for %s after level change\n",
+				       rdev->raid_disk, mdname(mddev));
 		}
 	}
 
@@ -4504,7 +4742,8 @@ int md_run(mddev_t *mddev)
 	}
 
 	if (mddev->bio_set == NULL)
-		mddev->bio_set = bioset_create(BIO_POOL_SIZE, sizeof(mddev));
+		mddev->bio_set = bioset_create(BIO_POOL_SIZE,
+					       sizeof(mddev_t *));
 
 	spin_lock(&pers_lock);
 	pers = find_pers(mddev->level, mddev->clevel);
@@ -4621,12 +4860,9 @@ int md_run(mddev_t *mddev)
 	smp_wmb();
 	mddev->ready = 1;
 	list_for_each_entry(rdev, &mddev->disks, same_set)
-		if (rdev->raid_disk >= 0) {
-			char nm[20];
-			sprintf(nm, "rd%d", rdev->raid_disk);
-			if (sysfs_create_link(&mddev->kobj, &rdev->kobj, nm))
+		if (rdev->raid_disk >= 0)
+			if (sysfs_link_rdev(mddev, rdev))
 				/* failure here is OK */;
-		}
 	
 	set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 	
@@ -4854,11 +5090,8 @@ static int do_md_stop(mddev_t * mddev, int mode, int is_open)
 		sysfs_notify_dirent_safe(mddev->sysfs_state);
 
 		list_for_each_entry(rdev, &mddev->disks, same_set)
-			if (rdev->raid_disk >= 0) {
-				char nm[20];
-				sprintf(nm, "rd%d", rdev->raid_disk);
-				sysfs_remove_link(&mddev->kobj, nm);
-			}
+			if (rdev->raid_disk >= 0)
+				sysfs_unlink_rdev(mddev, rdev);
 
 		set_capacity(disk, 0);
 		mutex_unlock(&mddev->open_mutex);
@@ -6198,18 +6431,7 @@ void md_error(mddev_t *mddev, mdk_rdev_t *rdev)
 	if (!rdev || test_bit(Faulty, &rdev->flags))
 		return;
 
-	if (mddev->external)
-		set_bit(Blocked, &rdev->flags);
-/*
-	dprintk("md_error dev:%s, rdev:(%d:%d), (caller: %p,%p,%p,%p).\n",
-		mdname(mddev),
-		MAJOR(rdev->bdev->bd_dev), MINOR(rdev->bdev->bd_dev),
-		__builtin_return_address(0),__builtin_return_address(1),
-		__builtin_return_address(2),__builtin_return_address(3));
-*/
-	if (!mddev->pers)
-		return;
-	if (!mddev->pers->error_handler)
+	if (!mddev->pers || !mddev->pers->error_handler)
 		return;
 	mddev->pers->error_handler(mddev,rdev);
 	if (mddev->degraded)
@@ -6933,11 +7155,14 @@ void md_do_sync(mddev_t *mddev)
 			atomic_add(sectors, &mddev->recovery_active);
 		}
 
+		if (test_bit(MD_RECOVERY_INTR, &mddev->recovery))
+			break;
+
 		j += sectors;
 		if (j>1) mddev->curr_resync = j;
 		mddev->curr_mark_cnt = io_sectors;
 		if (last_check == 0)
-			/* this is the earliers that rebuilt will be
+			/* this is the earliest that rebuild will be
 			 * visible in /proc/mdstat
 			 */
 			md_new_event(mddev);
@@ -6946,10 +7171,6 @@ void md_do_sync(mddev_t *mddev)
 			continue;
 
 		last_check = io_sectors;
-
-		if (test_bit(MD_RECOVERY_INTR, &mddev->recovery))
-			break;
-
 	repeat:
 		if (time_after_eq(jiffies, mark[last_mark] + SYNC_MARK_STEP )) {
 			/* step marks */
@@ -7067,29 +7288,23 @@ static int remove_and_add_spares(mddev_t *mddev)
 		    atomic_read(&rdev->nr_pending)==0) {
 			if (mddev->pers->hot_remove_disk(
 				    mddev, rdev->raid_disk)==0) {
-				char nm[20];
-				sprintf(nm,"rd%d", rdev->raid_disk);
-				sysfs_remove_link(&mddev->kobj, nm);
+				sysfs_unlink_rdev(mddev, rdev);
 				rdev->raid_disk = -1;
 			}
 		}
 
-	if (mddev->degraded && !mddev->recovery_disabled) {
+	if (mddev->degraded) {
 		list_for_each_entry(rdev, &mddev->disks, same_set) {
 			if (rdev->raid_disk >= 0 &&
 			    !test_bit(In_sync, &rdev->flags) &&
-			    !test_bit(Faulty, &rdev->flags) &&
-			    !test_bit(Blocked, &rdev->flags))
+			    !test_bit(Faulty, &rdev->flags))
 				spares++;
 			if (rdev->raid_disk < 0
 			    && !test_bit(Faulty, &rdev->flags)) {
 				rdev->recovery_offset = 0;
 				if (mddev->pers->
 				    hot_add_disk(mddev, rdev) == 0) {
-					char nm[20];
-					sprintf(nm, "rd%d", rdev->raid_disk);
-					if (sysfs_create_link(&mddev->kobj,
-							      &rdev->kobj, nm))
+					if (sysfs_link_rdev(mddev, rdev))
 						/* failure here is OK */;
 					spares++;
 					md_new_event(mddev);
@@ -7138,6 +7353,8 @@ static void reap_sync_thread(mddev_t *mddev)
 	set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 	sysfs_notify_dirent_safe(mddev->sysfs_action);
 	md_new_event(mddev);
+	if (mddev->event_work.func)
+		queue_work(md_misc_wq, &mddev->event_work);
 }
 
 /*
@@ -7170,9 +7387,6 @@ void md_check_recovery(mddev_t *mddev)
 	if (mddev->bitmap)
 		bitmap_daemon_work(mddev);
 
-	if (mddev->ro)
-		return;
-
 	if (signal_pending(current)) {
 		if (mddev->pers->sync_request && !mddev->external) {
 			printk(KERN_INFO "md: %s in immediate safe mode\n",
@@ -7209,9 +7423,7 @@ void md_check_recovery(mddev_t *mddev)
 				    atomic_read(&rdev->nr_pending)==0) {
 					if (mddev->pers->hot_remove_disk(
 						    mddev, rdev->raid_disk)==0) {
-						char nm[20];
-						sprintf(nm,"rd%d", rdev->raid_disk);
-						sysfs_remove_link(&mddev->kobj, nm);
+						sysfs_unlink_rdev(mddev, rdev);
 						rdev->raid_disk = -1;
 					}
 				}
@@ -7331,12 +7543,499 @@ void md_wait_for_blocked_rdev(mdk_rdev_t *rdev, mddev_t *mddev)
 {
 	sysfs_notify_dirent_safe(rdev->sysfs_state);
 	wait_event_timeout(rdev->blocked_wait,
-			   !test_bit(Blocked, &rdev->flags),
+			   !test_bit(Blocked, &rdev->flags) &&
+			   !test_bit(BlockedBadBlocks, &rdev->flags),
 			   msecs_to_jiffies(5000));
 	rdev_dec_pending(rdev, mddev);
 }
 EXPORT_SYMBOL(md_wait_for_blocked_rdev);
 
+
+/* Bad block management.
+ * We can record which blocks on each device are 'bad' and so just
+ * fail those blocks, or that stripe, rather than the whole device.
+ * Entries in the bad-block table are 64bits wide.  This comprises:
+ * Length of bad-range, in sectors: 0-511 for lengths 1-512
+ * Start of bad-range, sector offset, 54 bits (allows 8 exbibytes)
+ *  A 'shift' can be set so that larger blocks are tracked and
+ *  consequently larger devices can be covered.
+ * 'Acknowledged' flag - 1 bit. - the most significant bit.
+ *
+ * Locking of the bad-block table uses a seqlock so md_is_badblock
+ * might need to retry if it is very unlucky.
+ * We will sometimes want to check for bad blocks in a bi_end_io function,
+ * so we use the write_seqlock_irq variant.
+ *
+ * When looking for a bad block we specify a range and want to
+ * know if any block in the range is bad.  So we binary-search
+ * to the last range that starts at-or-before the given endpoint,
+ * (or "before the sector after the target range")
+ * then see if it ends after the given start.
+ * We return
+ *  0 if there are no known bad blocks in the range
+ *  1 if there are known bad block which are all acknowledged
+ * -1 if there are bad blocks which have not yet been acknowledged in metadata.
+ * plus the start/length of the first bad section we overlap.
+ */
+int md_is_badblock(struct badblocks *bb, sector_t s, int sectors,
+		   sector_t *first_bad, int *bad_sectors)
+{
+	int hi;
+	int lo = 0;
+	u64 *p = bb->page;
+	int rv = 0;
+	sector_t target = s + sectors;
+	unsigned seq;
+
+	if (bb->shift > 0) {
+		/* round the start down, and the end up */
+		s >>= bb->shift;
+		target += (1<<bb->shift) - 1;
+		target >>= bb->shift;
+		sectors = target - s;
+	}
+	/* 'target' is now the first block after the bad range */
+
+retry:
+	seq = read_seqbegin(&bb->lock);
+
+	hi = bb->count;
+
+	/* Binary search between lo and hi for 'target'
+	 * i.e. for the last range that starts before 'target'
+	 */
+	/* INVARIANT: ranges before 'lo' and at-or-after 'hi'
+	 * are known not to be the last range before target.
+	 * VARIANT: hi-lo is the number of possible
+	 * ranges, and decreases until it reaches 1
+	 */
+	while (hi - lo > 1) {
+		int mid = (lo + hi) / 2;
+		sector_t a = BB_OFFSET(p[mid]);
+		if (a < target)
+			/* This could still be the one, earlier ranges
+			 * could not. */
+			lo = mid;
+		else
+			/* This and later ranges are definitely out. */
+			hi = mid;
+	}
+	/* 'lo' might be the last that started before target, but 'hi' isn't */
+	if (hi > lo) {
+		/* need to check all range that end after 's' to see if
+		 * any are unacknowledged.
+		 */
+		while (lo >= 0 &&
+		       BB_OFFSET(p[lo]) + BB_LEN(p[lo]) > s) {
+			if (BB_OFFSET(p[lo]) < target) {
+				/* starts before the end, and finishes after
+				 * the start, so they must overlap
+				 */
+				if (rv != -1 && BB_ACK(p[lo]))
+					rv = 1;
+				else
+					rv = -1;
+				*first_bad = BB_OFFSET(p[lo]);
+				*bad_sectors = BB_LEN(p[lo]);
+			}
+			lo--;
+		}
+	}
+
+	if (read_seqretry(&bb->lock, seq))
+		goto retry;
+
+	return rv;
+}
+EXPORT_SYMBOL_GPL(md_is_badblock);
+
+/*
+ * Add a range of bad blocks to the table.
+ * This might extend the table, or might contract it
+ * if two adjacent ranges can be merged.
+ * We binary-search to find the 'insertion' point, then
+ * decide how best to handle it.
+ */
+static int md_set_badblocks(struct badblocks *bb, sector_t s, int sectors,
+			    int acknowledged)
+{
+	u64 *p;
+	int lo, hi;
+	int rv = 1;
+
+	if (bb->shift < 0)
+		/* badblocks are disabled */
+		return 0;
+
+	if (bb->shift) {
+		/* round the start down, and the end up */
+		sector_t next = s + sectors;
+		s >>= bb->shift;
+		next += (1<<bb->shift) - 1;
+		next >>= bb->shift;
+		sectors = next - s;
+	}
+
+	write_seqlock_irq(&bb->lock);
+
+	p = bb->page;
+	lo = 0;
+	hi = bb->count;
+	/* Find the last range that starts at-or-before 's' */
+	while (hi - lo > 1) {
+		int mid = (lo + hi) / 2;
+		sector_t a = BB_OFFSET(p[mid]);
+		if (a <= s)
+			lo = mid;
+		else
+			hi = mid;
+	}
+	if (hi > lo && BB_OFFSET(p[lo]) > s)
+		hi = lo;
+
+	if (hi > lo) {
+		/* we found a range that might merge with the start
+		 * of our new range
+		 */
+		sector_t a = BB_OFFSET(p[lo]);
+		sector_t e = a + BB_LEN(p[lo]);
+		int ack = BB_ACK(p[lo]);
+		if (e >= s) {
+			/* Yes, we can merge with a previous range */
+			if (s == a && s + sectors >= e)
+				/* new range covers old */
+				ack = acknowledged;
+			else
+				ack = ack && acknowledged;
+
+			if (e < s + sectors)
+				e = s + sectors;
+			if (e - a <= BB_MAX_LEN) {
+				p[lo] = BB_MAKE(a, e-a, ack);
+				s = e;
+			} else {
+				/* does not all fit in one range,
+				 * make p[lo] maximal
+				 */
+				if (BB_LEN(p[lo]) != BB_MAX_LEN)
+					p[lo] = BB_MAKE(a, BB_MAX_LEN, ack);
+				s = a + BB_MAX_LEN;
+			}
+			sectors = e - s;
+		}
+	}
+	if (sectors && hi < bb->count) {
+		/* 'hi' points to the first range that starts after 's'.
+		 * Maybe we can merge with the start of that range */
+		sector_t a = BB_OFFSET(p[hi]);
+		sector_t e = a + BB_LEN(p[hi]);
+		int ack = BB_ACK(p[hi]);
+		if (a <= s + sectors) {
+			/* merging is possible */
+			if (e <= s + sectors) {
+				/* full overlap */
+				e = s + sectors;
+				ack = acknowledged;
+			} else
+				ack = ack && acknowledged;
+
+			a = s;
+			if (e - a <= BB_MAX_LEN) {
+				p[hi] = BB_MAKE(a, e-a, ack);
+				s = e;
+			} else {
+				p[hi] = BB_MAKE(a, BB_MAX_LEN, ack);
+				s = a + BB_MAX_LEN;
+			}
+			sectors = e - s;
+			lo = hi;
+			hi++;
+		}
+	}
+	if (sectors == 0 && hi < bb->count) {
+		/* we might be able to combine lo and hi */
+		/* Note: 's' is at the end of 'lo' */
+		sector_t a = BB_OFFSET(p[hi]);
+		int lolen = BB_LEN(p[lo]);
+		int hilen = BB_LEN(p[hi]);
+		int newlen = lolen + hilen - (s - a);
+		if (s >= a && newlen < BB_MAX_LEN) {
+			/* yes, we can combine them */
+			int ack = BB_ACK(p[lo]) && BB_ACK(p[hi]);
+			p[lo] = BB_MAKE(BB_OFFSET(p[lo]), newlen, ack);
+			memmove(p + hi, p + hi + 1,
+				(bb->count - hi - 1) * 8);
+			bb->count--;
+		}
+	}
+	while (sectors) {
+		/* didn't merge (it all).
+		 * Need to add a range just before 'hi' */
+		if (bb->count >= MD_MAX_BADBLOCKS) {
+			/* No room for more */
+			rv = 0;
+			break;
+		} else {
+			int this_sectors = sectors;
+			memmove(p + hi + 1, p + hi,
+				(bb->count - hi) * 8);
+			bb->count++;
+
+			if (this_sectors > BB_MAX_LEN)
+				this_sectors = BB_MAX_LEN;
+			p[hi] = BB_MAKE(s, this_sectors, acknowledged);
+			sectors -= this_sectors;
+			s += this_sectors;
+		}
+	}
+
+	bb->changed = 1;
+	if (!acknowledged)
+		bb->unacked_exist = 1;
+	write_sequnlock_irq(&bb->lock);
+
+	return rv;
+}
+
+int rdev_set_badblocks(mdk_rdev_t *rdev, sector_t s, int sectors,
+		       int acknowledged)
+{
+	int rv = md_set_badblocks(&rdev->badblocks,
+				  s + rdev->data_offset, sectors, acknowledged);
+	if (rv) {
+		/* Make sure they get written out promptly */
+		set_bit(MD_CHANGE_CLEAN, &rdev->mddev->flags);
+		md_wakeup_thread(rdev->mddev->thread);
+	}
+	return rv;
+}
+EXPORT_SYMBOL_GPL(rdev_set_badblocks);
+
+/*
+ * Remove a range of bad blocks from the table.
+ * This may involve extending the table if we spilt a region,
+ * but it must not fail.  So if the table becomes full, we just
+ * drop the remove request.
+ */
+static int md_clear_badblocks(struct badblocks *bb, sector_t s, int sectors)
+{
+	u64 *p;
+	int lo, hi;
+	sector_t target = s + sectors;
+	int rv = 0;
+
+	if (bb->shift > 0) {
+		/* When clearing we round the start up and the end down.
+		 * This should not matter as the shift should align with
+		 * the block size and no rounding should ever be needed.
+		 * However it is better the think a block is bad when it
+		 * isn't than to think a block is not bad when it is.
+		 */
+		s += (1<<bb->shift) - 1;
+		s >>= bb->shift;
+		target >>= bb->shift;
+		sectors = target - s;
+	}
+
+	write_seqlock_irq(&bb->lock);
+
+	p = bb->page;
+	lo = 0;
+	hi = bb->count;
+	/* Find the last range that starts before 'target' */
+	while (hi - lo > 1) {
+		int mid = (lo + hi) / 2;
+		sector_t a = BB_OFFSET(p[mid]);
+		if (a < target)
+			lo = mid;
+		else
+			hi = mid;
+	}
+	if (hi > lo) {
+		/* p[lo] is the last range that could overlap the
+		 * current range.  Earlier ranges could also overlap,
+		 * but only this one can overlap the end of the range.
+		 */
+		if (BB_OFFSET(p[lo]) + BB_LEN(p[lo]) > target) {
+			/* Partial overlap, leave the tail of this range */
+			int ack = BB_ACK(p[lo]);
+			sector_t a = BB_OFFSET(p[lo]);
+			sector_t end = a + BB_LEN(p[lo]);
+
+			if (a < s) {
+				/* we need to split this range */
+				if (bb->count >= MD_MAX_BADBLOCKS) {
+					rv = 0;
+					goto out;
+				}
+				memmove(p+lo+1, p+lo, (bb->count - lo) * 8);
+				bb->count++;
+				p[lo] = BB_MAKE(a, s-a, ack);
+				lo++;
+			}
+			p[lo] = BB_MAKE(target, end - target, ack);
+			/* there is no longer an overlap */
+			hi = lo;
+			lo--;
+		}
+		while (lo >= 0 &&
+		       BB_OFFSET(p[lo]) + BB_LEN(p[lo]) > s) {
+			/* This range does overlap */
+			if (BB_OFFSET(p[lo]) < s) {
+				/* Keep the early parts of this range. */
+				int ack = BB_ACK(p[lo]);
+				sector_t start = BB_OFFSET(p[lo]);
+				p[lo] = BB_MAKE(start, s - start, ack);
+				/* now low doesn't overlap, so.. */
+				break;
+			}
+			lo--;
+		}
+		/* 'lo' is strictly before, 'hi' is strictly after,
+		 * anything between needs to be discarded
+		 */
+		if (hi - lo > 1) {
+			memmove(p+lo+1, p+hi, (bb->count - hi) * 8);
+			bb->count -= (hi - lo - 1);
+		}
+	}
+
+	bb->changed = 1;
+out:
+	write_sequnlock_irq(&bb->lock);
+	return rv;
+}
+
+int rdev_clear_badblocks(mdk_rdev_t *rdev, sector_t s, int sectors)
+{
+	return md_clear_badblocks(&rdev->badblocks,
+				  s + rdev->data_offset,
+				  sectors);
+}
+EXPORT_SYMBOL_GPL(rdev_clear_badblocks);
+
+/*
+ * Acknowledge all bad blocks in a list.
+ * This only succeeds if ->changed is clear.  It is used by
+ * in-kernel metadata updates
+ */
+void md_ack_all_badblocks(struct badblocks *bb)
+{
+	if (bb->page == NULL || bb->changed)
+		/* no point even trying */
+		return;
+	write_seqlock_irq(&bb->lock);
+
+	if (bb->changed == 0) {
+		u64 *p = bb->page;
+		int i;
+		for (i = 0; i < bb->count ; i++) {
+			if (!BB_ACK(p[i])) {
+				sector_t start = BB_OFFSET(p[i]);
+				int len = BB_LEN(p[i]);
+				p[i] = BB_MAKE(start, len, 1);
+			}
+		}
+		bb->unacked_exist = 0;
+	}
+	write_sequnlock_irq(&bb->lock);
+}
+EXPORT_SYMBOL_GPL(md_ack_all_badblocks);
+
+/* sysfs access to bad-blocks list.
+ * We present two files.
+ * 'bad-blocks' lists sector numbers and lengths of ranges that
+ *    are recorded as bad.  The list is truncated to fit within
+ *    the one-page limit of sysfs.
+ *    Writing "sector length" to this file adds an acknowledged
+ *    bad block list.
+ * 'unacknowledged-bad-blocks' lists bad blocks that have not yet
+ *    been acknowledged.  Writing to this file adds bad blocks
+ *    without acknowledging them.  This is largely for testing.
+ */
+
+static ssize_t
+badblocks_show(struct badblocks *bb, char *page, int unack)
+{
+	size_t len;
+	int i;
+	u64 *p = bb->page;
+	unsigned seq;
+
+	if (bb->shift < 0)
+		return 0;
+
+retry:
+	seq = read_seqbegin(&bb->lock);
+
+	len = 0;
+	i = 0;
+
+	while (len < PAGE_SIZE && i < bb->count) {
+		sector_t s = BB_OFFSET(p[i]);
+		unsigned int length = BB_LEN(p[i]);
+		int ack = BB_ACK(p[i]);
+		i++;
+
+		if (unack && ack)
+			continue;
+
+		len += snprintf(page+len, PAGE_SIZE-len, "%llu %u\n",
+				(unsigned long long)s << bb->shift,
+				length << bb->shift);
+	}
+	if (unack && len == 0)
+		bb->unacked_exist = 0;
+
+	if (read_seqretry(&bb->lock, seq))
+		goto retry;
+
+	return len;
+}
+
+#define DO_DEBUG 1
+
+static ssize_t
+badblocks_store(struct badblocks *bb, const char *page, size_t len, int unack)
+{
+	unsigned long long sector;
+	int length;
+	char newline;
+#ifdef DO_DEBUG
+	/* Allow clearing via sysfs *only* for testing/debugging.
+	 * Normally only a successful write may clear a badblock
+	 */
+	int clear = 0;
+	if (page[0] == '-') {
+		clear = 1;
+		page++;
+	}
+#endif /* DO_DEBUG */
+
+	switch (sscanf(page, "%llu %d%c", &sector, &length, &newline)) {
+	case 3:
+		if (newline != '\n')
+			return -EINVAL;
+	case 2:
+		if (length <= 0)
+			return -EINVAL;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+#ifdef DO_DEBUG
+	if (clear) {
+		md_clear_badblocks(bb, sector, length);
+		return len;
+	}
+#endif /* DO_DEBUG */
+	if (md_set_badblocks(bb, sector, length, !unack))
+		return len;
+	else
+		return -ENOSPC;
+}
+
 static int md_notify_reboot(struct notifier_block *this,
 			    unsigned long code, void *x)
 {

drivers/md/md.h

@@ -29,6 +29,13 @@
 typedef struct mddev_s mddev_t;
 typedef struct mdk_rdev_s mdk_rdev_t;
 
+/* Bad block numbers are stored sorted in a single page.
+ * 64bits is used for each block or extent.
+ * 54 bits are sector number, 9 bits are extent size,
+ * 1 bit is an 'acknowledged' flag.
+ */
+#define MD_MAX_BADBLOCKS	(PAGE_SIZE/8)
+
 /*
  * MD's 'extended' device
  */
@@ -48,7 +55,7 @@ struct mdk_rdev_s
 	struct block_device *meta_bdev;
 	struct block_device *bdev;	/* block device handle */
 
-	struct page	*sb_page;
+	struct page	*sb_page, *bb_page;
 	int		sb_loaded;
 	__u64		sb_events;
 	sector_t	data_offset;	/* start of data in array */
@@ -74,9 +81,29 @@ struct mdk_rdev_s
 #define	In_sync		2		/* device is in_sync with rest of array */
 #define	WriteMostly	4		/* Avoid reading if at all possible */
 #define	AutoDetected	7		/* added by auto-detect */
-#define Blocked		8		/* An error occurred on an externally
-					 * managed array, don't allow writes
+#define Blocked		8		/* An error occurred but has not yet
+					 * been acknowledged by the metadata
+					 * handler, so don't allow writes
 					 * until it is cleared */
+#define WriteErrorSeen	9		/* A write error has been seen on this
+					 * device
+					 */
+#define FaultRecorded	10		/* Intermediate state for clearing
+					 * Blocked.  The Fault is/will-be
+					 * recorded in the metadata, but that
+					 * metadata hasn't been stored safely
+					 * on disk yet.
+					 */
+#define BlockedBadBlocks 11		/* A writer is blocked because they
+					 * found an unacknowledged bad-block.
+					 * This can safely be cleared at any
+					 * time, and the writer will re-check.
+					 * It may be set at any time, and at
+					 * worst the writer will timeout and
+					 * re-check.  So setting it as
+					 * accurately as possible is good, but
+					 * not absolutely critical.
+					 */
 	wait_queue_head_t blocked_wait;
 
 	int desc_nr;			/* descriptor index in the superblock */
@@ -111,8 +138,54 @@ struct mdk_rdev_s
 
 	struct sysfs_dirent *sysfs_state; /* handle for 'state'
 					   * sysfs entry */
+
+	struct badblocks {
+		int	count;		/* count of bad blocks */
+		int	unacked_exist;	/* there probably are unacknowledged
+					 * bad blocks.  This is only cleared
+					 * when a read discovers none
+					 */
+		int	shift;		/* shift from sectors to block size
+					 * a -ve shift means badblocks are
+					 * disabled.*/
+		u64	*page;		/* badblock list */
+		int	changed;
+		seqlock_t lock;
+
+		sector_t sector;
+		sector_t size;		/* in sectors */
+	} badblocks;
 };
 
+#define BB_LEN_MASK	(0x00000000000001FFULL)
+#define BB_OFFSET_MASK	(0x7FFFFFFFFFFFFE00ULL)
+#define BB_ACK_MASK	(0x8000000000000000ULL)
+#define BB_MAX_LEN	512
+#define BB_OFFSET(x)	(((x) & BB_OFFSET_MASK) >> 9)
+#define BB_LEN(x)	(((x) & BB_LEN_MASK) + 1)
+#define BB_ACK(x)	(!!((x) & BB_ACK_MASK))
+#define BB_MAKE(a, l, ack) (((a)<<9) | ((l)-1) | ((u64)(!!(ack)) << 63))
+
+extern int md_is_badblock(struct badblocks *bb, sector_t s, int sectors,
+			  sector_t *first_bad, int *bad_sectors);
+static inline int is_badblock(mdk_rdev_t *rdev, sector_t s, int sectors,
+			      sector_t *first_bad, int *bad_sectors)
+{
+	if (unlikely(rdev->badblocks.count)) {
+		int rv = md_is_badblock(&rdev->badblocks, rdev->data_offset + s,
+					sectors,
+					first_bad, bad_sectors);
+		if (rv)
+			*first_bad -= rdev->data_offset;
+		return rv;
+	}
+	return 0;
+}
+extern int rdev_set_badblocks(mdk_rdev_t *rdev, sector_t s, int sectors,
+			      int acknowledged);
+extern int rdev_clear_badblocks(mdk_rdev_t *rdev, sector_t s, int sectors);
+extern void md_ack_all_badblocks(struct badblocks *bb);
+
 struct mddev_s
 {
 	void				*private;
@@ -239,9 +312,12 @@ struct mddev_s
 #define	MD_RECOVERY_FROZEN	9
 
 	unsigned long			recovery;
-	int				recovery_disabled; /* if we detect that recovery
-							    * will always fail, set this
-							    * so we don't loop trying */
+	/* If a RAID personality determines that recovery (of a particular
+	 * device) will fail due to a read error on the source device, it
+	 * takes a copy of this number and does not attempt recovery again
+	 * until this number changes.
+	 */
+	int				recovery_disabled;
 
 	int				in_sync;	/* know to not need resync */
 	/* 'open_mutex' avoids races between 'md_open' and 'do_md_stop', so
@@ -304,11 +380,6 @@ struct mddev_s
 							 * hot-adding a bitmap.  It should
 							 * eventually be settable by sysfs.
 							 */
-		/* When md is serving under dm, it might use a
-		 * dirty_log to store the bits.
-		 */
-		struct dm_dirty_log *log;
-
 		struct mutex		mutex;
 		unsigned long		chunksize;
 		unsigned long		daemon_sleep; /* how many jiffies between updates? */
@@ -413,6 +484,20 @@ static inline char * mdname (mddev_t * mddev)
 	return mddev->gendisk ? mddev->gendisk->disk_name : "mdX";
 }
 
+static inline int sysfs_link_rdev(mddev_t *mddev, mdk_rdev_t *rdev)
+{
+	char nm[20];
+	sprintf(nm, "rd%d", rdev->raid_disk);
+	return sysfs_create_link(&mddev->kobj, &rdev->kobj, nm);
+}
+
+static inline void sysfs_unlink_rdev(mddev_t *mddev, mdk_rdev_t *rdev)
+{
+	char nm[20];
+	sprintf(nm, "rd%d", rdev->raid_disk);
+	sysfs_remove_link(&mddev->kobj, nm);
+}
+
 /*
  * iterates through some rdev ringlist. It's safe to remove the
  * current 'rdev'. Dont touch 'tmp' though.
@@ -505,7 +590,7 @@ extern void mddev_init(mddev_t *mddev);
 extern int md_run(mddev_t *mddev);
 extern void md_stop(mddev_t *mddev);
 extern void md_stop_writes(mddev_t *mddev);
-extern void md_rdev_init(mdk_rdev_t *rdev);
+extern int md_rdev_init(mdk_rdev_t *rdev);
 
 extern void mddev_suspend(mddev_t *mddev);
 extern void mddev_resume(mddev_t *mddev);
@@ -514,4 +599,5 @@ extern struct bio *bio_clone_mddev(struct bio *bio, gfp_t gfp_mask,
 extern struct bio *bio_alloc_mddev(gfp_t gfp_mask, int nr_iovecs,
 				   mddev_t *mddev);
 extern int mddev_check_plugged(mddev_t *mddev);
+extern void md_trim_bio(struct bio *bio, int offset, int size);
 #endif /* _MD_MD_H */

drivers/md/raid1.c

@@ -35,16 +35,13 @@
 #include <linux/delay.h>
 #include <linux/blkdev.h>
 #include <linux/seq_file.h>
+#include <linux/ratelimit.h>
 #include "md.h"
 #include "raid1.h"
 #include "bitmap.h"
 
 #define DEBUG 0
-#if DEBUG
-#define PRINTK(x...) printk(x)
-#else
-#define PRINTK(x...)
-#endif
+#define PRINTK(x...) do { if (DEBUG) printk(x); } while (0)
 
 /*
  * Number of guaranteed r1bios in case of extreme VM load:
@@ -166,7 +163,7 @@ static void put_all_bios(conf_t *conf, r1bio_t *r1_bio)
 
 	for (i = 0; i < conf->raid_disks; i++) {
 		struct bio **bio = r1_bio->bios + i;
-		if (*bio && *bio != IO_BLOCKED)
+		if (!BIO_SPECIAL(*bio))
 			bio_put(*bio);
 		*bio = NULL;
 	}
@@ -176,12 +173,6 @@ static void free_r1bio(r1bio_t *r1_bio)
 {
 	conf_t *conf = r1_bio->mddev->private;
 
-	/*
-	 * Wake up any possible resync thread that waits for the device
-	 * to go idle.
-	 */
-	allow_barrier(conf);
-
 	put_all_bios(conf, r1_bio);
 	mempool_free(r1_bio, conf->r1bio_pool);
 }
@@ -222,6 +213,33 @@ static void reschedule_retry(r1bio_t *r1_bio)
  * operation and are ready to return a success/failure code to the buffer
  * cache layer.
  */
+static void call_bio_endio(r1bio_t *r1_bio)
+{
+	struct bio *bio = r1_bio->master_bio;
+	int done;
+	conf_t *conf = r1_bio->mddev->private;
+
+	if (bio->bi_phys_segments) {
+		unsigned long flags;
+		spin_lock_irqsave(&conf->device_lock, flags);
+		bio->bi_phys_segments--;
+		done = (bio->bi_phys_segments == 0);
+		spin_unlock_irqrestore(&conf->device_lock, flags);
+	} else
+		done = 1;
+
+	if (!test_bit(R1BIO_Uptodate, &r1_bio->state))
+		clear_bit(BIO_UPTODATE, &bio->bi_flags);
+	if (done) {
+		bio_endio(bio, 0);
+		/*
+		 * Wake up any possible resync thread that waits for the device
+		 * to go idle.
+		 */
+		allow_barrier(conf);
+	}
+}
+
 static void raid_end_bio_io(r1bio_t *r1_bio)
 {
 	struct bio *bio = r1_bio->master_bio;
@@ -234,8 +252,7 @@ static void raid_end_bio_io(r1bio_t *r1_bio)
 			(unsigned long long) bio->bi_sector +
 				(bio->bi_size >> 9) - 1);
 
-		bio_endio(bio,
-			test_bit(R1BIO_Uptodate, &r1_bio->state) ? 0 : -EIO);
+		call_bio_endio(r1_bio);
 	}
 	free_r1bio(r1_bio);
 }
@@ -287,36 +304,52 @@ static void raid1_end_read_request(struct bio *bio, int error)
 		 * oops, read error:
 		 */
 		char b[BDEVNAME_SIZE];
-		if (printk_ratelimit())
-			printk(KERN_ERR "md/raid1:%s: %s: rescheduling sector %llu\n",
-			       mdname(conf->mddev),
-			       bdevname(conf->mirrors[mirror].rdev->bdev,b), (unsigned long long)r1_bio->sector);
+		printk_ratelimited(
+			KERN_ERR "md/raid1:%s: %s: "
+			"rescheduling sector %llu\n",
+			mdname(conf->mddev),
+			bdevname(conf->mirrors[mirror].rdev->bdev,
+				 b),
+			(unsigned long long)r1_bio->sector);
+		set_bit(R1BIO_ReadError, &r1_bio->state);
 		reschedule_retry(r1_bio);
 	}
 
 	rdev_dec_pending(conf->mirrors[mirror].rdev, conf->mddev);
 }
 
+static void close_write(r1bio_t *r1_bio)
+{
+	/* it really is the end of this request */
+	if (test_bit(R1BIO_BehindIO, &r1_bio->state)) {
+		/* free extra copy of the data pages */
+		int i = r1_bio->behind_page_count;
+		while (i--)
+			safe_put_page(r1_bio->behind_bvecs[i].bv_page);
+		kfree(r1_bio->behind_bvecs);
+		r1_bio->behind_bvecs = NULL;
+	}
+	/* clear the bitmap if all writes complete successfully */
+	bitmap_endwrite(r1_bio->mddev->bitmap, r1_bio->sector,
+			r1_bio->sectors,
+			!test_bit(R1BIO_Degraded, &r1_bio->state),
+			test_bit(R1BIO_BehindIO, &r1_bio->state));
+	md_write_end(r1_bio->mddev);
+}
+
 static void r1_bio_write_done(r1bio_t *r1_bio)
 {
-	if (atomic_dec_and_test(&r1_bio->remaining))
-	{
-		/* it really is the end of this request */
-		if (test_bit(R1BIO_BehindIO, &r1_bio->state)) {
-			/* free extra copy of the data pages */
-			int i = r1_bio->behind_page_count;
-			while (i--)
-				safe_put_page(r1_bio->behind_pages[i]);
-			kfree(r1_bio->behind_pages);
-			r1_bio->behind_pages = NULL;
-		}
-		/* clear the bitmap if all writes complete successfully */
-		bitmap_endwrite(r1_bio->mddev->bitmap, r1_bio->sector,
-				r1_bio->sectors,
-				!test_bit(R1BIO_Degraded, &r1_bio->state),
-				test_bit(R1BIO_BehindIO, &r1_bio->state));
-		md_write_end(r1_bio->mddev);
-		raid_end_bio_io(r1_bio);
+	if (!atomic_dec_and_test(&r1_bio->remaining))
+		return;
+
+	if (test_bit(R1BIO_WriteError, &r1_bio->state))
+		reschedule_retry(r1_bio);
+	else {
+		close_write(r1_bio);
+		if (test_bit(R1BIO_MadeGood, &r1_bio->state))
+			reschedule_retry(r1_bio);
+		else
+			raid_end_bio_io(r1_bio);
 	}
 }
 
@@ -336,13 +369,11 @@ static void raid1_end_write_request(struct bio *bio, int error)
 	/*
 	 * 'one mirror IO has finished' event handler:
 	 */
-	r1_bio->bios[mirror] = NULL;
-	to_put = bio;
 	if (!uptodate) {
-		md_error(r1_bio->mddev, conf->mirrors[mirror].rdev);
-		/* an I/O failed, we can't clear the bitmap */
-		set_bit(R1BIO_Degraded, &r1_bio->state);
-	} else
+		set_bit(WriteErrorSeen,
+			&conf->mirrors[mirror].rdev->flags);
+		set_bit(R1BIO_WriteError, &r1_bio->state);
+	} else {
 		/*
 		 * Set R1BIO_Uptodate in our master bio, so that we
 		 * will return a good error code for to the higher
@@ -353,8 +384,22 @@ static void raid1_end_write_request(struct bio *bio, int error)
 		 * to user-side. So if something waits for IO, then it
 		 * will wait for the 'master' bio.
 		 */
+		sector_t first_bad;
+		int bad_sectors;
+
+		r1_bio->bios[mirror] = NULL;
+		to_put = bio;
 		set_bit(R1BIO_Uptodate, &r1_bio->state);
 
+		/* Maybe we can clear some bad blocks. */
+		if (is_badblock(conf->mirrors[mirror].rdev,
+				r1_bio->sector, r1_bio->sectors,
+				&first_bad, &bad_sectors)) {
+			r1_bio->bios[mirror] = IO_MADE_GOOD;
+			set_bit(R1BIO_MadeGood, &r1_bio->state);
+		}
+	}
+
 	update_head_pos(mirror, r1_bio);
 
 	if (behind) {
@@ -377,11 +422,13 @@ static void raid1_end_write_request(struct bio *bio, int error)
 				       (unsigned long long) mbio->bi_sector,
 				       (unsigned long long) mbio->bi_sector +
 				       (mbio->bi_size >> 9) - 1);
-				bio_endio(mbio, 0);
+				call_bio_endio(r1_bio);
 			}
 		}
 	}
-	rdev_dec_pending(conf->mirrors[mirror].rdev, conf->mddev);
+	if (r1_bio->bios[mirror] == NULL)
+		rdev_dec_pending(conf->mirrors[mirror].rdev,
+				 conf->mddev);
 
 	/*
 	 * Let's see if all mirrored write operations have finished
@@ -408,10 +455,11 @@ static void raid1_end_write_request(struct bio *bio, int error)
  *
  * The rdev for the device selected will have nr_pending incremented.
  */
-static int read_balance(conf_t *conf, r1bio_t *r1_bio)
+static int read_balance(conf_t *conf, r1bio_t *r1_bio, int *max_sectors)
 {
 	const sector_t this_sector = r1_bio->sector;
-	const int sectors = r1_bio->sectors;
+	int sectors;
+	int best_good_sectors;
 	int start_disk;
 	int best_disk;
 	int i;
@@ -426,8 +474,11 @@ static int read_balance(conf_t *conf, r1bio_t *r1_bio)
 	 * We take the first readable disk when above the resync window.
 	 */
  retry:
+	sectors = r1_bio->sectors;
 	best_disk = -1;
 	best_dist = MaxSector;
+	best_good_sectors = 0;
+
 	if (conf->mddev->recovery_cp < MaxSector &&
 	    (this_sector + sectors >= conf->next_resync)) {
 		choose_first = 1;
@@ -439,6 +490,9 @@ static int read_balance(conf_t *conf, r1bio_t *r1_bio)
 
 	for (i = 0 ; i < conf->raid_disks ; i++) {
 		sector_t dist;
+		sector_t first_bad;
+		int bad_sectors;
+
 		int disk = start_disk + i;
 		if (disk >= conf->raid_disks)
 			disk -= conf->raid_disks;
@@ -461,6 +515,35 @@ static int read_balance(conf_t *conf, r1bio_t *r1_bio)
 		/* This is a reasonable device to use.  It might
 		 * even be best.
 		 */
+		if (is_badblock(rdev, this_sector, sectors,
+				&first_bad, &bad_sectors)) {
+			if (best_dist < MaxSector)
+				/* already have a better device */
+				continue;
+			if (first_bad <= this_sector) {
+				/* cannot read here. If this is the 'primary'
+				 * device, then we must not read beyond
+				 * bad_sectors from another device..
+				 */
+				bad_sectors -= (this_sector - first_bad);
+				if (choose_first && sectors > bad_sectors)
+					sectors = bad_sectors;
+				if (best_good_sectors > sectors)
+					best_good_sectors = sectors;
+
+			} else {
+				sector_t good_sectors = first_bad - this_sector;
+				if (good_sectors > best_good_sectors) {
+					best_good_sectors = good_sectors;
+					best_disk = disk;
+				}
+				if (choose_first)
+					break;
+			}
+			continue;
+		} else
+			best_good_sectors = sectors;
+
 		dist = abs(this_sector - conf->mirrors[disk].head_position);
 		if (choose_first
 		    /* Don't change to another disk for sequential reads */
@@ -489,10 +572,12 @@ static int read_balance(conf_t *conf, r1bio_t *r1_bio)
 			rdev_dec_pending(rdev, conf->mddev);
 			goto retry;
 		}
+		sectors = best_good_sectors;
 		conf->next_seq_sect = this_sector + sectors;
 		conf->last_used = best_disk;
 	}
 	rcu_read_unlock();
+	*max_sectors = sectors;
 
 	return best_disk;
 }
@@ -672,30 +757,31 @@ static void alloc_behind_pages(struct bio *bio, r1bio_t *r1_bio)
 {
 	int i;
 	struct bio_vec *bvec;
-	struct page **pages = kzalloc(bio->bi_vcnt * sizeof(struct page*),
+	struct bio_vec *bvecs = kzalloc(bio->bi_vcnt * sizeof(struct bio_vec),
 					GFP_NOIO);
-	if (unlikely(!pages))
+	if (unlikely(!bvecs))
 		return;
 
 	bio_for_each_segment(bvec, bio, i) {
-		pages[i] = alloc_page(GFP_NOIO);
-		if (unlikely(!pages[i]))
+		bvecs[i] = *bvec;
+		bvecs[i].bv_page = alloc_page(GFP_NOIO);
+		if (unlikely(!bvecs[i].bv_page))
 			goto do_sync_io;
-		memcpy(kmap(pages[i]) + bvec->bv_offset,
-			kmap(bvec->bv_page) + bvec->bv_offset, bvec->bv_len);
-		kunmap(pages[i]);
+		memcpy(kmap(bvecs[i].bv_page) + bvec->bv_offset,
+		       kmap(bvec->bv_page) + bvec->bv_offset, bvec->bv_len);
+		kunmap(bvecs[i].bv_page);
 		kunmap(bvec->bv_page);
 	}
-	r1_bio->behind_pages = pages;
+	r1_bio->behind_bvecs = bvecs;
 	r1_bio->behind_page_count = bio->bi_vcnt;
 	set_bit(R1BIO_BehindIO, &r1_bio->state);
 	return;
 
 do_sync_io:
 	for (i = 0; i < bio->bi_vcnt; i++)
-		if (pages[i])
-			put_page(pages[i]);
-	kfree(pages);
+		if (bvecs[i].bv_page)
+			put_page(bvecs[i].bv_page);
+	kfree(bvecs);
 	PRINTK("%dB behind alloc failed, doing sync I/O\n", bio->bi_size);
 }
 
@@ -705,7 +791,7 @@ static int make_request(mddev_t *mddev, struct bio * bio)
 	mirror_info_t *mirror;
 	r1bio_t *r1_bio;
 	struct bio *read_bio;
-	int i, targets = 0, disks;
+	int i, disks;
 	struct bitmap *bitmap;
 	unsigned long flags;
 	const int rw = bio_data_dir(bio);
@@ -713,6 +799,9 @@ static int make_request(mddev_t *mddev, struct bio * bio)
 	const unsigned long do_flush_fua = (bio->bi_rw & (REQ_FLUSH | REQ_FUA));
 	mdk_rdev_t *blocked_rdev;
 	int plugged;
+	int first_clone;
+	int sectors_handled;
+	int max_sectors;
 
 	/*
 	 * Register the new request and wait if the reconstruction
@@ -759,11 +848,24 @@ static int make_request(mddev_t *mddev, struct bio * bio)
 	r1_bio->mddev = mddev;
 	r1_bio->sector = bio->bi_sector;
 
+	/* We might need to issue multiple reads to different
+	 * devices if there are bad blocks around, so we keep
+	 * track of the number of reads in bio->bi_phys_segments.
+	 * If this is 0, there is only one r1_bio and no locking
+	 * will be needed when requests complete.  If it is
+	 * non-zero, then it is the number of not-completed requests.
+	 */
+	bio->bi_phys_segments = 0;
+	clear_bit(BIO_SEG_VALID, &bio->bi_flags);
+
 	if (rw == READ) {
 		/*
 		 * read balancing logic:
 		 */
-		int rdisk = read_balance(conf, r1_bio);
+		int rdisk;
+
+read_again:
+		rdisk = read_balance(conf, r1_bio, &max_sectors);
 
 		if (rdisk < 0) {
 			/* couldn't find anywhere to read from */
@@ -784,6 +886,8 @@ static int make_request(mddev_t *mddev, struct bio * bio)
 		r1_bio->read_disk = rdisk;
 
 		read_bio = bio_clone_mddev(bio, GFP_NOIO, mddev);
+		md_trim_bio(read_bio, r1_bio->sector - bio->bi_sector,
+			    max_sectors);
 
 		r1_bio->bios[rdisk] = read_bio;
 
@@ -793,16 +897,52 @@ static int make_request(mddev_t *mddev, struct bio * bio)
 		read_bio->bi_rw = READ | do_sync;
 		read_bio->bi_private = r1_bio;
 
-		generic_make_request(read_bio);
+		if (max_sectors < r1_bio->sectors) {
+			/* could not read all from this device, so we will
+			 * need another r1_bio.
+			 */
+
+			sectors_handled = (r1_bio->sector + max_sectors
+					   - bio->bi_sector);
+			r1_bio->sectors = max_sectors;
+			spin_lock_irq(&conf->device_lock);
+			if (bio->bi_phys_segments == 0)
+				bio->bi_phys_segments = 2;
+			else
+				bio->bi_phys_segments++;
+			spin_unlock_irq(&conf->device_lock);
+			/* Cannot call generic_make_request directly
+			 * as that will be queued in __make_request
+			 * and subsequent mempool_alloc might block waiting
+			 * for it.  So hand bio over to raid1d.
+			 */
+			reschedule_retry(r1_bio);
+
+			r1_bio = mempool_alloc(conf->r1bio_pool, GFP_NOIO);
+
+			r1_bio->master_bio = bio;
+			r1_bio->sectors = (bio->bi_size >> 9) - sectors_handled;
+			r1_bio->state = 0;
+			r1_bio->mddev = mddev;
+			r1_bio->sector = bio->bi_sector + sectors_handled;
+			goto read_again;
+		} else
+			generic_make_request(read_bio);
 		return 0;
 	}
 
 	/*
 	 * WRITE:
 	 */
-	/* first select target devices under spinlock and
+	/* first select target devices under rcu_lock and
 	 * inc refcount on their rdev.  Record them by setting
 	 * bios[x] to bio
+	 * If there are known/acknowledged bad blocks on any device on
+	 * which we have seen a write error, we want to avoid writing those
+	 * blocks.
+	 * This potentially requires several writes to write around
+	 * the bad blocks.  Each set of writes gets it's own r1bio
+	 * with a set of bios attached.
 	 */
 	plugged = mddev_check_plugged(mddev);
 
@@ -810,6 +950,7 @@ static int make_request(mddev_t *mddev, struct bio * bio)
  retry_write:
 	blocked_rdev = NULL;
 	rcu_read_lock();
+	max_sectors = r1_bio->sectors;
 	for (i = 0;  i < disks; i++) {
 		mdk_rdev_t *rdev = rcu_dereference(conf->mirrors[i].rdev);
 		if (rdev && unlikely(test_bit(Blocked, &rdev->flags))) {
@@ -817,17 +958,56 @@ static int make_request(mddev_t *mddev, struct bio * bio)
 			blocked_rdev = rdev;
 			break;
 		}
-		if (rdev && !test_bit(Faulty, &rdev->flags)) {
-			atomic_inc(&rdev->nr_pending);
-			if (test_bit(Faulty, &rdev->flags)) {
+		r1_bio->bios[i] = NULL;
+		if (!rdev || test_bit(Faulty, &rdev->flags)) {
+			set_bit(R1BIO_Degraded, &r1_bio->state);
+			continue;
+		}
+
+		atomic_inc(&rdev->nr_pending);
+		if (test_bit(WriteErrorSeen, &rdev->flags)) {
+			sector_t first_bad;
+			int bad_sectors;
+			int is_bad;
+
+			is_bad = is_badblock(rdev, r1_bio->sector,
+					     max_sectors,
+					     &first_bad, &bad_sectors);
+			if (is_bad < 0) {
+				/* mustn't write here until the bad block is
+				 * acknowledged*/
+				set_bit(BlockedBadBlocks, &rdev->flags);
+				blocked_rdev = rdev;
+				break;
+			}
+			if (is_bad && first_bad <= r1_bio->sector) {
+				/* Cannot write here at all */
+				bad_sectors -= (r1_bio->sector - first_bad);
+				if (bad_sectors < max_sectors)
+					/* mustn't write more than bad_sectors
+					 * to other devices yet
+					 */
+					max_sectors = bad_sectors;
 				rdev_dec_pending(rdev, mddev);
-				r1_bio->bios[i] = NULL;
-			} else {
-				r1_bio->bios[i] = bio;
-				targets++;
+				/* We don't set R1BIO_Degraded as that
+				 * only applies if the disk is
+				 * missing, so it might be re-added,
+				 * and we want to know to recover this
+				 * chunk.
+				 * In this case the device is here,
+				 * and the fact that this chunk is not
+				 * in-sync is recorded in the bad
+				 * block log
+				 */
+				continue;
 			}
-		} else
-			r1_bio->bios[i] = NULL;
+			if (is_bad) {
+				int good_sectors = first_bad - r1_bio->sector;
+				if (good_sectors < max_sectors)
+					max_sectors = good_sectors;
+			}
+		}
+		r1_bio->bios[i] = bio;
 	}
 	rcu_read_unlock();
 
@@ -838,51 +1018,57 @@ static int make_request(mddev_t *mddev, struct bio * bio)
 		for (j = 0; j < i; j++)
 			if (r1_bio->bios[j])
 				rdev_dec_pending(conf->mirrors[j].rdev, mddev);
-
+		r1_bio->state = 0;
 		allow_barrier(conf);
 		md_wait_for_blocked_rdev(blocked_rdev, mddev);
 		wait_barrier(conf);
 		goto retry_write;
 	}
 
-	BUG_ON(targets == 0); /* we never fail the last device */
-
-	if (targets < conf->raid_disks) {
-		/* array is degraded, we will not clear the bitmap
-		 * on I/O completion (see raid1_end_write_request) */
-		set_bit(R1BIO_Degraded, &r1_bio->state);
+	if (max_sectors < r1_bio->sectors) {
+		/* We are splitting this write into multiple parts, so
+		 * we need to prepare for allocating another r1_bio.
+		 */
+		r1_bio->sectors = max_sectors;
+		spin_lock_irq(&conf->device_lock);
+		if (bio->bi_phys_segments == 0)
+			bio->bi_phys_segments = 2;
+		else
+			bio->bi_phys_segments++;
+		spin_unlock_irq(&conf->device_lock);
 	}
-
-	/* do behind I/O ?
-	 * Not if there are too many, or cannot allocate memory,
-	 * or a reader on WriteMostly is waiting for behind writes 
-	 * to flush */
-	if (bitmap &&
-	    (atomic_read(&bitmap->behind_writes)
-	     < mddev->bitmap_info.max_write_behind) &&
-	    !waitqueue_active(&bitmap->behind_wait))
-		alloc_behind_pages(bio, r1_bio);
+	sectors_handled = r1_bio->sector + max_sectors - bio->bi_sector;
 
 	atomic_set(&r1_bio->remaining, 1);
 	atomic_set(&r1_bio->behind_remaining, 0);
 
-	bitmap_startwrite(bitmap, bio->bi_sector, r1_bio->sectors,
-				test_bit(R1BIO_BehindIO, &r1_bio->state));
+	first_clone = 1;
 	for (i = 0; i < disks; i++) {
 		struct bio *mbio;
 		if (!r1_bio->bios[i])
 			continue;
 
 		mbio = bio_clone_mddev(bio, GFP_NOIO, mddev);
-		r1_bio->bios[i] = mbio;
-
-		mbio->bi_sector	= r1_bio->sector + conf->mirrors[i].rdev->data_offset;
-		mbio->bi_bdev = conf->mirrors[i].rdev->bdev;
-		mbio->bi_end_io	= raid1_end_write_request;
-		mbio->bi_rw = WRITE | do_flush_fua | do_sync;
-		mbio->bi_private = r1_bio;
-
-		if (r1_bio->behind_pages) {
+		md_trim_bio(mbio, r1_bio->sector - bio->bi_sector, max_sectors);
+
+		if (first_clone) {
+			/* do behind I/O ?
+			 * Not if there are too many, or cannot
+			 * allocate memory, or a reader on WriteMostly
+			 * is waiting for behind writes to flush */
+			if (bitmap &&
+			    (atomic_read(&bitmap->behind_writes)
+			     < mddev->bitmap_info.max_write_behind) &&
+			    !waitqueue_active(&bitmap->behind_wait))
+				alloc_behind_pages(mbio, r1_bio);
+
+			bitmap_startwrite(bitmap, r1_bio->sector,
+					  r1_bio->sectors,
+					  test_bit(R1BIO_BehindIO,
+						   &r1_bio->state));
+			first_clone = 0;
+		}
+		if (r1_bio->behind_bvecs) {
 			struct bio_vec *bvec;
 			int j;
 
@@ -894,11 +1080,20 @@ static int make_request(mddev_t *mddev, struct bio * bio)
 			 * them all
 			 */
 			__bio_for_each_segment(bvec, mbio, j, 0)
-				bvec->bv_page = r1_bio->behind_pages[j];
+				bvec->bv_page = r1_bio->behind_bvecs[j].bv_page;
 			if (test_bit(WriteMostly, &conf->mirrors[i].rdev->flags))
 				atomic_inc(&r1_bio->behind_remaining);
 		}
 
+		r1_bio->bios[i] = mbio;
+
+		mbio->bi_sector	= (r1_bio->sector +
+				   conf->mirrors[i].rdev->data_offset);
+		mbio->bi_bdev = conf->mirrors[i].rdev->bdev;
+		mbio->bi_end_io	= raid1_end_write_request;
+		mbio->bi_rw = WRITE | do_flush_fua | do_sync;
+		mbio->bi_private = r1_bio;
+
 		atomic_inc(&r1_bio->remaining);
 		spin_lock_irqsave(&conf->device_lock, flags);
 		bio_list_add(&conf->pending_bio_list, mbio);
@@ -909,6 +1104,19 @@ static int make_request(mddev_t *mddev, struct bio * bio)
 	/* In case raid1d snuck in to freeze_array */
 	wake_up(&conf->wait_barrier);
 
+	if (sectors_handled < (bio->bi_size >> 9)) {
+		/* We need another r1_bio.  It has already been counted
+		 * in bio->bi_phys_segments
+		 */
+		r1_bio = mempool_alloc(conf->r1bio_pool, GFP_NOIO);
+		r1_bio->master_bio = bio;
+		r1_bio->sectors = (bio->bi_size >> 9) - sectors_handled;
+		r1_bio->state = 0;
+		r1_bio->mddev = mddev;
+		r1_bio->sector = bio->bi_sector + sectors_handled;
+		goto retry_write;
+	}
+
 	if (do_sync || !bitmap || !plugged)
 		md_wakeup_thread(mddev->thread);
 
@@ -952,9 +1160,10 @@ static void error(mddev_t *mddev, mdk_rdev_t *rdev)
 		 * However don't try a recovery from this drive as
 		 * it is very likely to fail.
 		 */
-		mddev->recovery_disabled = 1;
+		conf->recovery_disabled = mddev->recovery_disabled;
 		return;
 	}
+	set_bit(Blocked, &rdev->flags);
 	if (test_and_clear_bit(In_sync, &rdev->flags)) {
 		unsigned long flags;
 		spin_lock_irqsave(&conf->device_lock, flags);
@@ -1027,7 +1236,7 @@ static int raid1_spare_active(mddev_t *mddev)
 		    && !test_bit(Faulty, &rdev->flags)
 		    && !test_and_set_bit(In_sync, &rdev->flags)) {
 			count++;
-			sysfs_notify_dirent(rdev->sysfs_state);
+			sysfs_notify_dirent_safe(rdev->sysfs_state);
 		}
 	}
 	spin_lock_irqsave(&conf->device_lock, flags);
@@ -1048,6 +1257,9 @@ static int raid1_add_disk(mddev_t *mddev, mdk_rdev_t *rdev)
 	int first = 0;
 	int last = mddev->raid_disks - 1;
 
+	if (mddev->recovery_disabled == conf->recovery_disabled)
+		return -EBUSY;
+
 	if (rdev->raid_disk >= 0)
 		first = last = rdev->raid_disk;
 
@@ -1103,7 +1315,7 @@ static int raid1_remove_disk(mddev_t *mddev, int number)
 		 * is not possible.
 		 */
 		if (!test_bit(Faulty, &rdev->flags) &&
-		    !mddev->recovery_disabled &&
+		    mddev->recovery_disabled != conf->recovery_disabled &&
 		    mddev->degraded < conf->raid_disks) {
 			err = -EBUSY;
 			goto abort;
@@ -1155,6 +1367,8 @@ static void end_sync_write(struct bio *bio, int error)
 	conf_t *conf = mddev->private;
 	int i;
 	int mirror=0;
+	sector_t first_bad;
+	int bad_sectors;
 
 	for (i = 0; i < conf->raid_disks; i++)
 		if (r1_bio->bios[i] == bio) {
@@ -1172,18 +1386,48 @@ static void end_sync_write(struct bio *bio, int error)
 			s += sync_blocks;
 			sectors_to_go -= sync_blocks;
 		} while (sectors_to_go > 0);
-		md_error(mddev, conf->mirrors[mirror].rdev);
-	}
+		set_bit(WriteErrorSeen,
+			&conf->mirrors[mirror].rdev->flags);
+		set_bit(R1BIO_WriteError, &r1_bio->state);
+	} else if (is_badblock(conf->mirrors[mirror].rdev,
+			       r1_bio->sector,
+			       r1_bio->sectors,
+			       &first_bad, &bad_sectors) &&
+		   !is_badblock(conf->mirrors[r1_bio->read_disk].rdev,
+				r1_bio->sector,
+				r1_bio->sectors,
+				&first_bad, &bad_sectors)
+		)
+		set_bit(R1BIO_MadeGood, &r1_bio->state);
 
 	update_head_pos(mirror, r1_bio);
 
 	if (atomic_dec_and_test(&r1_bio->remaining)) {
-		sector_t s = r1_bio->sectors;
-		put_buf(r1_bio);
-		md_done_sync(mddev, s, uptodate);
+		int s = r1_bio->sectors;
+		if (test_bit(R1BIO_MadeGood, &r1_bio->state) ||
+		    test_bit(R1BIO_WriteError, &r1_bio->state))
+			reschedule_retry(r1_bio);
+		else {
+			put_buf(r1_bio);
+			md_done_sync(mddev, s, uptodate);
+		}
 	}
 }
 
+static int r1_sync_page_io(mdk_rdev_t *rdev, sector_t sector,
+			    int sectors, struct page *page, int rw)
+{
+	if (sync_page_io(rdev, sector, sectors << 9, page, rw, false))
+		/* success */
+		return 1;
+	if (rw == WRITE)
+		set_bit(WriteErrorSeen, &rdev->flags);
+	/* need to record an error - either for the block or the device */
+	if (!rdev_set_badblocks(rdev, sector, sectors, 0))
+		md_error(rdev->mddev, rdev);
+	return 0;
+}
+
 static int fix_sync_read_error(r1bio_t *r1_bio)
 {
 	/* Try some synchronous reads of other devices to get
@@ -1193,6 +1437,9 @@ static int fix_sync_read_error(r1bio_t *r1_bio)
 	 * We don't need to freeze the array, because being in an
 	 * active sync request, there is no normal IO, and
 	 * no overlapping syncs.
+	 * We don't need to check is_badblock() again as we
+	 * made sure that anything with a bad block in range
+	 * will have bi_end_io clear.
 	 */
 	mddev_t *mddev = r1_bio->mddev;
 	conf_t *conf = mddev->private;
@@ -1217,9 +1464,7 @@ static int fix_sync_read_error(r1bio_t *r1_bio)
 				 * active, and resync is currently active
 				 */
 				rdev = conf->mirrors[d].rdev;
-				if (sync_page_io(rdev,
-						 sect,
-						 s<<9,
+				if (sync_page_io(rdev, sect, s<<9,
 						 bio->bi_io_vec[idx].bv_page,
 						 READ, false)) {
 					success = 1;
@@ -1233,16 +1478,36 @@ static int fix_sync_read_error(r1bio_t *r1_bio)
 
 		if (!success) {
 			char b[BDEVNAME_SIZE];
-			/* Cannot read from anywhere, array is toast */
-			md_error(mddev, conf->mirrors[r1_bio->read_disk].rdev);
+			int abort = 0;
+			/* Cannot read from anywhere, this block is lost.
+			 * Record a bad block on each device.  If that doesn't
+			 * work just disable and interrupt the recovery.
+			 * Don't fail devices as that won't really help.
+			 */
 			printk(KERN_ALERT "md/raid1:%s: %s: unrecoverable I/O read error"
 			       " for block %llu\n",
 			       mdname(mddev),
 			       bdevname(bio->bi_bdev, b),
 			       (unsigned long long)r1_bio->sector);
-			md_done_sync(mddev, r1_bio->sectors, 0);
-			put_buf(r1_bio);
-			return 0;
+			for (d = 0; d < conf->raid_disks; d++) {
+				rdev = conf->mirrors[d].rdev;
+				if (!rdev || test_bit(Faulty, &rdev->flags))
+					continue;
+				if (!rdev_set_badblocks(rdev, sect, s, 0))
+					abort = 1;
+			}
+			if (abort) {
+				mddev->recovery_disabled = 1;
+				set_bit(MD_RECOVERY_INTR, &mddev->recovery);
+				md_done_sync(mddev, r1_bio->sectors, 0);
+				put_buf(r1_bio);
+				return 0;
+			}
+			/* Try next page */
+			sectors -= s;
+			sect += s;
+			idx++;
+			continue;
 		}
 
 		start = d;
@@ -1254,16 +1519,12 @@ static int fix_sync_read_error(r1bio_t *r1_bio)
 			if (r1_bio->bios[d]->bi_end_io != end_sync_read)
 				continue;
 			rdev = conf->mirrors[d].rdev;
-			if (sync_page_io(rdev,
-					 sect,
-					 s<<9,
-					 bio->bi_io_vec[idx].bv_page,
-					 WRITE, false) == 0) {
+			if (r1_sync_page_io(rdev, sect, s,
+					    bio->bi_io_vec[idx].bv_page,
+					    WRITE) == 0) {
 				r1_bio->bios[d]->bi_end_io = NULL;
 				rdev_dec_pending(rdev, mddev);
-				md_error(mddev, rdev);
-			} else
-				atomic_add(s, &rdev->corrected_errors);
+			}
 		}
 		d = start;
 		while (d != r1_bio->read_disk) {
@@ -1273,12 +1534,10 @@ static int fix_sync_read_error(r1bio_t *r1_bio)
 			if (r1_bio->bios[d]->bi_end_io != end_sync_read)
 				continue;
 			rdev = conf->mirrors[d].rdev;
-			if (sync_page_io(rdev,
-					 sect,
-					 s<<9,
-					 bio->bi_io_vec[idx].bv_page,
-					 READ, false) == 0)
-				md_error(mddev, rdev);
+			if (r1_sync_page_io(rdev, sect, s,
+					    bio->bi_io_vec[idx].bv_page,
+					    READ) != 0)
+				atomic_add(s, &rdev->corrected_errors);
 		}
 		sectors -= s;
 		sect += s;
@@ -1420,7 +1679,7 @@ static void sync_request_write(mddev_t *mddev, r1bio_t *r1_bio)
  *
  *	1.	Retries failed read operations on working mirrors.
  *	2.	Updates the raid superblock when problems encounter.
- *	3.	Performs writes following reads for array syncronising.
+ *	3.	Performs writes following reads for array synchronising.
  */
 
 static void fix_read_error(conf_t *conf, int read_disk,
@@ -1443,9 +1702,14 @@ static void fix_read_error(conf_t *conf, int read_disk,
 			 * which is the thread that might remove
 			 * a device.  If raid1d ever becomes multi-threaded....
 			 */
+			sector_t first_bad;
+			int bad_sectors;
+
 			rdev = conf->mirrors[d].rdev;
 			if (rdev &&
 			    test_bit(In_sync, &rdev->flags) &&
+			    is_badblock(rdev, sect, s,
+					&first_bad, &bad_sectors) == 0 &&
 			    sync_page_io(rdev, sect, s<<9,
 					 conf->tmppage, READ, false))
 				success = 1;
@@ -1457,8 +1721,10 @@ static void fix_read_error(conf_t *conf, int read_disk,
 		} while (!success && d != read_disk);
 
 		if (!success) {
-			/* Cannot read from anywhere -- bye bye array */
-			md_error(mddev, conf->mirrors[read_disk].rdev);
+			/* Cannot read from anywhere - mark it bad */
+			mdk_rdev_t *rdev = conf->mirrors[read_disk].rdev;
+			if (!rdev_set_badblocks(rdev, sect, s, 0))
+				md_error(mddev, rdev);
 			break;
 		}
 		/* write it back and re-read */
@@ -1469,13 +1735,9 @@ static void fix_read_error(conf_t *conf, int read_disk,
 			d--;
 			rdev = conf->mirrors[d].rdev;
 			if (rdev &&
-			    test_bit(In_sync, &rdev->flags)) {
-				if (sync_page_io(rdev, sect, s<<9,
-						 conf->tmppage, WRITE, false)
-				    == 0)
-					/* Well, this device is dead */
-					md_error(mddev, rdev);
-			}
+			    test_bit(In_sync, &rdev->flags))
+				r1_sync_page_io(rdev, sect, s,
+						conf->tmppage, WRITE);
 		}
 		d = start;
 		while (d != read_disk) {
@@ -1486,12 +1748,8 @@ static void fix_read_error(conf_t *conf, int read_disk,
 			rdev = conf->mirrors[d].rdev;
 			if (rdev &&
 			    test_bit(In_sync, &rdev->flags)) {
-				if (sync_page_io(rdev, sect, s<<9,
-						 conf->tmppage, READ, false)
-				    == 0)
-					/* Well, this device is dead */
-					md_error(mddev, rdev);
-				else {
+				if (r1_sync_page_io(rdev, sect, s,
+						    conf->tmppage, READ)) {
 					atomic_add(s, &rdev->corrected_errors);
 					printk(KERN_INFO
 					       "md/raid1:%s: read error corrected "
@@ -1508,21 +1766,255 @@ static void fix_read_error(conf_t *conf, int read_disk,
 	}
 }
 
+static void bi_complete(struct bio *bio, int error)
+{
+	complete((struct completion *)bio->bi_private);
+}
+
+static int submit_bio_wait(int rw, struct bio *bio)
+{
+	struct completion event;
+	rw |= REQ_SYNC;
+
+	init_completion(&event);
+	bio->bi_private = &event;
+	bio->bi_end_io = bi_complete;
+	submit_bio(rw, bio);
+	wait_for_completion(&event);
+
+	return test_bit(BIO_UPTODATE, &bio->bi_flags);
+}
+
+static int narrow_write_error(r1bio_t *r1_bio, int i)
+{
+	mddev_t *mddev = r1_bio->mddev;
+	conf_t *conf = mddev->private;
+	mdk_rdev_t *rdev = conf->mirrors[i].rdev;
+	int vcnt, idx;
+	struct bio_vec *vec;
+
+	/* bio has the data to be written to device 'i' where
+	 * we just recently had a write error.
+	 * We repeatedly clone the bio and trim down to one block,
+	 * then try the write.  Where the write fails we record
+	 * a bad block.
+	 * It is conceivable that the bio doesn't exactly align with
+	 * blocks.  We must handle this somehow.
+	 *
+	 * We currently own a reference on the rdev.
+	 */
+
+	int block_sectors;
+	sector_t sector;
+	int sectors;
+	int sect_to_write = r1_bio->sectors;
+	int ok = 1;
+
+	if (rdev->badblocks.shift < 0)
+		return 0;
+
+	block_sectors = 1 << rdev->badblocks.shift;
+	sector = r1_bio->sector;
+	sectors = ((sector + block_sectors)
+		   & ~(sector_t)(block_sectors - 1))
+		- sector;
+
+	if (test_bit(R1BIO_BehindIO, &r1_bio->state)) {
+		vcnt = r1_bio->behind_page_count;
+		vec = r1_bio->behind_bvecs;
+		idx = 0;
+		while (vec[idx].bv_page == NULL)
+			idx++;
+	} else {
+		vcnt = r1_bio->master_bio->bi_vcnt;
+		vec = r1_bio->master_bio->bi_io_vec;
+		idx = r1_bio->master_bio->bi_idx;
+	}
+	while (sect_to_write) {
+		struct bio *wbio;
+		if (sectors > sect_to_write)
+			sectors = sect_to_write;
+		/* Write at 'sector' for 'sectors'*/
+
+		wbio = bio_alloc_mddev(GFP_NOIO, vcnt, mddev);
+		memcpy(wbio->bi_io_vec, vec, vcnt * sizeof(struct bio_vec));
+		wbio->bi_sector = r1_bio->sector;
+		wbio->bi_rw = WRITE;
+		wbio->bi_vcnt = vcnt;
+		wbio->bi_size = r1_bio->sectors << 9;
+		wbio->bi_idx = idx;
+
+		md_trim_bio(wbio, sector - r1_bio->sector, sectors);
+		wbio->bi_sector += rdev->data_offset;
+		wbio->bi_bdev = rdev->bdev;
+		if (submit_bio_wait(WRITE, wbio) == 0)
+			/* failure! */
+			ok = rdev_set_badblocks(rdev, sector,
+						sectors, 0)
+				&& ok;
+
+		bio_put(wbio);
+		sect_to_write -= sectors;
+		sector += sectors;
+		sectors = block_sectors;
+	}
+	return ok;
+}
+
+static void handle_sync_write_finished(conf_t *conf, r1bio_t *r1_bio)
+{
+	int m;
+	int s = r1_bio->sectors;
+	for (m = 0; m < conf->raid_disks ; m++) {
+		mdk_rdev_t *rdev = conf->mirrors[m].rdev;
+		struct bio *bio = r1_bio->bios[m];
+		if (bio->bi_end_io == NULL)
+			continue;
+		if (test_bit(BIO_UPTODATE, &bio->bi_flags) &&
+		    test_bit(R1BIO_MadeGood, &r1_bio->state)) {
+			rdev_clear_badblocks(rdev, r1_bio->sector, s);
+		}
+		if (!test_bit(BIO_UPTODATE, &bio->bi_flags) &&
+		    test_bit(R1BIO_WriteError, &r1_bio->state)) {
+			if (!rdev_set_badblocks(rdev, r1_bio->sector, s, 0))
+				md_error(conf->mddev, rdev);
+		}
+	}
+	put_buf(r1_bio);
+	md_done_sync(conf->mddev, s, 1);
+}
+
+static void handle_write_finished(conf_t *conf, r1bio_t *r1_bio)
+{
+	int m;
+	for (m = 0; m < conf->raid_disks ; m++)
+		if (r1_bio->bios[m] == IO_MADE_GOOD) {
+			mdk_rdev_t *rdev = conf->mirrors[m].rdev;
+			rdev_clear_badblocks(rdev,
+					     r1_bio->sector,
+					     r1_bio->sectors);
+			rdev_dec_pending(rdev, conf->mddev);
+		} else if (r1_bio->bios[m] != NULL) {
+			/* This drive got a write error.  We need to
+			 * narrow down and record precise write
+			 * errors.
+			 */
+			if (!narrow_write_error(r1_bio, m)) {
+				md_error(conf->mddev,
+					 conf->mirrors[m].rdev);
+				/* an I/O failed, we can't clear the bitmap */
+				set_bit(R1BIO_Degraded, &r1_bio->state);
+			}
+			rdev_dec_pending(conf->mirrors[m].rdev,
+					 conf->mddev);
+		}
+	if (test_bit(R1BIO_WriteError, &r1_bio->state))
+		close_write(r1_bio);
+	raid_end_bio_io(r1_bio);
+}
+
+static void handle_read_error(conf_t *conf, r1bio_t *r1_bio)
+{
+	int disk;
+	int max_sectors;
+	mddev_t *mddev = conf->mddev;
+	struct bio *bio;
+	char b[BDEVNAME_SIZE];
+	mdk_rdev_t *rdev;
+
+	clear_bit(R1BIO_ReadError, &r1_bio->state);
+	/* we got a read error. Maybe the drive is bad.  Maybe just
+	 * the block and we can fix it.
+	 * We freeze all other IO, and try reading the block from
+	 * other devices.  When we find one, we re-write
+	 * and check it that fixes the read error.
+	 * This is all done synchronously while the array is
+	 * frozen
+	 */
+	if (mddev->ro == 0) {
+		freeze_array(conf);
+		fix_read_error(conf, r1_bio->read_disk,
+			       r1_bio->sector, r1_bio->sectors);
+		unfreeze_array(conf);
+	} else
+		md_error(mddev, conf->mirrors[r1_bio->read_disk].rdev);
+
+	bio = r1_bio->bios[r1_bio->read_disk];
+	bdevname(bio->bi_bdev, b);
+read_more:
+	disk = read_balance(conf, r1_bio, &max_sectors);
+	if (disk == -1) {
+		printk(KERN_ALERT "md/raid1:%s: %s: unrecoverable I/O"
+		       " read error for block %llu\n",
+		       mdname(mddev), b, (unsigned long long)r1_bio->sector);
+		raid_end_bio_io(r1_bio);
+	} else {
+		const unsigned long do_sync
+			= r1_bio->master_bio->bi_rw & REQ_SYNC;
+		if (bio) {
+			r1_bio->bios[r1_bio->read_disk] =
+				mddev->ro ? IO_BLOCKED : NULL;
+			bio_put(bio);
+		}
+		r1_bio->read_disk = disk;
+		bio = bio_clone_mddev(r1_bio->master_bio, GFP_NOIO, mddev);
+		md_trim_bio(bio, r1_bio->sector - bio->bi_sector, max_sectors);
+		r1_bio->bios[r1_bio->read_disk] = bio;
+		rdev = conf->mirrors[disk].rdev;
+		printk_ratelimited(KERN_ERR
+				   "md/raid1:%s: redirecting sector %llu"
+				   " to other mirror: %s\n",
+				   mdname(mddev),
+				   (unsigned long long)r1_bio->sector,
+				   bdevname(rdev->bdev, b));
+		bio->bi_sector = r1_bio->sector + rdev->data_offset;
+		bio->bi_bdev = rdev->bdev;
+		bio->bi_end_io = raid1_end_read_request;
+		bio->bi_rw = READ | do_sync;
+		bio->bi_private = r1_bio;
+		if (max_sectors < r1_bio->sectors) {
+			/* Drat - have to split this up more */
+			struct bio *mbio = r1_bio->master_bio;
+			int sectors_handled = (r1_bio->sector + max_sectors
+					       - mbio->bi_sector);
+			r1_bio->sectors = max_sectors;
+			spin_lock_irq(&conf->device_lock);
+			if (mbio->bi_phys_segments == 0)
+				mbio->bi_phys_segments = 2;
+			else
+				mbio->bi_phys_segments++;
+			spin_unlock_irq(&conf->device_lock);
+			generic_make_request(bio);
+			bio = NULL;
+
+			r1_bio = mempool_alloc(conf->r1bio_pool, GFP_NOIO);
+
+			r1_bio->master_bio = mbio;
+			r1_bio->sectors = (mbio->bi_size >> 9)
+					  - sectors_handled;
+			r1_bio->state = 0;
+			set_bit(R1BIO_ReadError, &r1_bio->state);
+			r1_bio->mddev = mddev;
+			r1_bio->sector = mbio->bi_sector + sectors_handled;
+
+			goto read_more;
+		} else
+			generic_make_request(bio);
+	}
+}
+
 static void raid1d(mddev_t *mddev)
 {
 	r1bio_t *r1_bio;
-	struct bio *bio;
 	unsigned long flags;
 	conf_t *conf = mddev->private;
 	struct list_head *head = &conf->retry_list;
-	mdk_rdev_t *rdev;
 	struct blk_plug plug;
 
 	md_check_recovery(mddev);
 
 	blk_start_plug(&plug);
 	for (;;) {
-		char b[BDEVNAME_SIZE];
 
 		if (atomic_read(&mddev->plug_cnt) == 0)
 			flush_pending_writes(conf);
@@ -1539,62 +2031,26 @@ static void raid1d(mddev_t *mddev)
 
 		mddev = r1_bio->mddev;
 		conf = mddev->private;
-		if (test_bit(R1BIO_IsSync, &r1_bio->state))
-			sync_request_write(mddev, r1_bio);
-		else {
-			int disk;
-
-			/* we got a read error. Maybe the drive is bad.  Maybe just
-			 * the block and we can fix it.
-			 * We freeze all other IO, and try reading the block from
-			 * other devices.  When we find one, we re-write
-			 * and check it that fixes the read error.
-			 * This is all done synchronously while the array is
-			 * frozen
+		if (test_bit(R1BIO_IsSync, &r1_bio->state)) {
+			if (test_bit(R1BIO_MadeGood, &r1_bio->state) ||
+			    test_bit(R1BIO_WriteError, &r1_bio->state))
+				handle_sync_write_finished(conf, r1_bio);
+			else
+				sync_request_write(mddev, r1_bio);
+		} else if (test_bit(R1BIO_MadeGood, &r1_bio->state) ||
+			   test_bit(R1BIO_WriteError, &r1_bio->state))
+			handle_write_finished(conf, r1_bio);
+		else if (test_bit(R1BIO_ReadError, &r1_bio->state))
+			handle_read_error(conf, r1_bio);
+		else
+			/* just a partial read to be scheduled from separate
+			 * context
 			 */
-			if (mddev->ro == 0) {
-				freeze_array(conf);
-				fix_read_error(conf, r1_bio->read_disk,
-					       r1_bio->sector,
-					       r1_bio->sectors);
-				unfreeze_array(conf);
-			} else
-				md_error(mddev,
-					 conf->mirrors[r1_bio->read_disk].rdev);
-
-			bio = r1_bio->bios[r1_bio->read_disk];
-			if ((disk=read_balance(conf, r1_bio)) == -1) {
-				printk(KERN_ALERT "md/raid1:%s: %s: unrecoverable I/O"
-				       " read error for block %llu\n",
-				       mdname(mddev),
-				       bdevname(bio->bi_bdev,b),
-				       (unsigned long long)r1_bio->sector);
-				raid_end_bio_io(r1_bio);
-			} else {
-				const unsigned long do_sync = r1_bio->master_bio->bi_rw & REQ_SYNC;
-				r1_bio->bios[r1_bio->read_disk] =
-					mddev->ro ? IO_BLOCKED : NULL;
-				r1_bio->read_disk = disk;
-				bio_put(bio);
-				bio = bio_clone_mddev(r1_bio->master_bio,
-						      GFP_NOIO, mddev);
-				r1_bio->bios[r1_bio->read_disk] = bio;
-				rdev = conf->mirrors[disk].rdev;
-				if (printk_ratelimit())
-					printk(KERN_ERR "md/raid1:%s: redirecting sector %llu to"
-					       " other mirror: %s\n",
-					       mdname(mddev),
-					       (unsigned long long)r1_bio->sector,
-					       bdevname(rdev->bdev,b));
-				bio->bi_sector = r1_bio->sector + rdev->data_offset;
-				bio->bi_bdev = rdev->bdev;
-				bio->bi_end_io = raid1_end_read_request;
-				bio->bi_rw = READ | do_sync;
-				bio->bi_private = r1_bio;
-				generic_make_request(bio);
-			}
-		}
+			generic_make_request(r1_bio->bios[r1_bio->read_disk]);
+
 		cond_resched();
+		if (mddev->flags & ~(1<<MD_CHANGE_PENDING))
+			md_check_recovery(mddev);
 	}
 	blk_finish_plug(&plug);
 }
@@ -1636,6 +2092,8 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i
 	int write_targets = 0, read_targets = 0;
 	sector_t sync_blocks;
 	int still_degraded = 0;
+	int good_sectors = RESYNC_SECTORS;
+	int min_bad = 0; /* number of sectors that are bad in all devices */
 
 	if (!conf->r1buf_pool)
 		if (init_resync(conf))
@@ -1723,36 +2181,89 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i
 
 		rdev = rcu_dereference(conf->mirrors[i].rdev);
 		if (rdev == NULL ||
-			   test_bit(Faulty, &rdev->flags)) {
+		    test_bit(Faulty, &rdev->flags)) {
 			still_degraded = 1;
-			continue;
 		} else if (!test_bit(In_sync, &rdev->flags)) {
 			bio->bi_rw = WRITE;
 			bio->bi_end_io = end_sync_write;
 			write_targets ++;
 		} else {
 			/* may need to read from here */
-			bio->bi_rw = READ;
-			bio->bi_end_io = end_sync_read;
-			if (test_bit(WriteMostly, &rdev->flags)) {
-				if (wonly < 0)
-					wonly = i;
-			} else {
-				if (disk < 0)
-					disk = i;
+			sector_t first_bad = MaxSector;
+			int bad_sectors;
+
+			if (is_badblock(rdev, sector_nr, good_sectors,
+					&first_bad, &bad_sectors)) {
+				if (first_bad > sector_nr)
+					good_sectors = first_bad - sector_nr;
+				else {
+					bad_sectors -= (sector_nr - first_bad);
+					if (min_bad == 0 ||
+					    min_bad > bad_sectors)
+						min_bad = bad_sectors;
+				}
+			}
+			if (sector_nr < first_bad) {
+				if (test_bit(WriteMostly, &rdev->flags)) {
+					if (wonly < 0)
+						wonly = i;
+				} else {
+					if (disk < 0)
+						disk = i;
+				}
+				bio->bi_rw = READ;
+				bio->bi_end_io = end_sync_read;
+				read_targets++;
 			}
-			read_targets++;
 		}
-		atomic_inc(&rdev->nr_pending);
-		bio->bi_sector = sector_nr + rdev->data_offset;
-		bio->bi_bdev = rdev->bdev;
-		bio->bi_private = r1_bio;
+		if (bio->bi_end_io) {
+			atomic_inc(&rdev->nr_pending);
+			bio->bi_sector = sector_nr + rdev->data_offset;
+			bio->bi_bdev = rdev->bdev;
+			bio->bi_private = r1_bio;
+		}
 	}
 	rcu_read_unlock();
 	if (disk < 0)
 		disk = wonly;
 	r1_bio->read_disk = disk;
 
+	if (read_targets == 0 && min_bad > 0) {
+		/* These sectors are bad on all InSync devices, so we
+		 * need to mark them bad on all write targets
+		 */
+		int ok = 1;
+		for (i = 0 ; i < conf->raid_disks ; i++)
+			if (r1_bio->bios[i]->bi_end_io == end_sync_write) {
+				mdk_rdev_t *rdev =
+					rcu_dereference(conf->mirrors[i].rdev);
+				ok = rdev_set_badblocks(rdev, sector_nr,
+							min_bad, 0
+					) && ok;
+			}
+		set_bit(MD_CHANGE_DEVS, &mddev->flags);
+		*skipped = 1;
+		put_buf(r1_bio);
+
+		if (!ok) {
+			/* Cannot record the badblocks, so need to
+			 * abort the resync.
+			 * If there are multiple read targets, could just
+			 * fail the really bad ones ???
+			 */
+			conf->recovery_disabled = mddev->recovery_disabled;
+			set_bit(MD_RECOVERY_INTR, &mddev->recovery);
+			return 0;
+		} else
+			return min_bad;
+
+	}
+	if (min_bad > 0 && min_bad < good_sectors) {
+		/* only resync enough to reach the next bad->good
+		 * transition */
+		good_sectors = min_bad;
+	}
+
 	if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery) && read_targets > 0)
 		/* extra read targets are also write targets */
 		write_targets += read_targets-1;
@@ -1769,6 +2280,8 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i
 
 	if (max_sector > mddev->resync_max)
 		max_sector = mddev->resync_max; /* Don't do IO beyond here */
+	if (max_sector > sector_nr + good_sectors)
+		max_sector = sector_nr + good_sectors;
 	nr_sectors = 0;
 	sync_blocks = 0;
 	do {
@@ -2154,18 +2667,13 @@ static int raid1_reshape(mddev_t *mddev)
 	for (d = d2 = 0; d < conf->raid_disks; d++) {
 		mdk_rdev_t *rdev = conf->mirrors[d].rdev;
 		if (rdev && rdev->raid_disk != d2) {
-			char nm[20];
-			sprintf(nm, "rd%d", rdev->raid_disk);
-			sysfs_remove_link(&mddev->kobj, nm);
+			sysfs_unlink_rdev(mddev, rdev);
 			rdev->raid_disk = d2;
-			sprintf(nm, "rd%d", rdev->raid_disk);
-			sysfs_remove_link(&mddev->kobj, nm);
-			if (sysfs_create_link(&mddev->kobj,
-					      &rdev->kobj, nm))
+			sysfs_unlink_rdev(mddev, rdev);
+			if (sysfs_link_rdev(mddev, rdev))
 				printk(KERN_WARNING
-				       "md/raid1:%s: cannot register "
-				       "%s\n",
-				       mdname(mddev), nm);
+				       "md/raid1:%s: cannot register rd%d\n",
+				       mdname(mddev), rdev->raid_disk);
 		}
 		if (rdev)
 			newmirrors[d2++].rdev = rdev;

drivers/md/raid1.h

@@ -48,6 +48,12 @@ struct r1_private_data_s {
 					    * (fresh device added).
 					    * Cleared when a sync completes.
 					    */
+	int			recovery_disabled; /* when the same as
+						    * mddev->recovery_disabled
+						    * we don't allow recovery
+						    * to be attempted as we
+						    * expect a read error
+						    */
 
 	wait_queue_head_t	wait_barrier;
 
@@ -95,7 +101,7 @@ struct r1bio_s {
 
 	struct list_head	retry_list;
 	/* Next two are only valid when R1BIO_BehindIO is set */
-	struct page		**behind_pages;
+	struct bio_vec		*behind_bvecs;
 	int			behind_page_count;
 	/*
 	 * if the IO is in WRITE direction, then multiple bios are used.
@@ -110,13 +116,24 @@ struct r1bio_s {
  * correct the read error.  To keep track of bad blocks on a per-bio
  * level, we store IO_BLOCKED in the appropriate 'bios' pointer
  */
-#define IO_BLOCKED ((struct bio*)1)
+#define IO_BLOCKED ((struct bio *)1)
+/* When we successfully write to a known bad-block, we need to remove the
+ * bad-block marking which must be done from process context.  So we record
+ * the success by setting bios[n] to IO_MADE_GOOD
+ */
+#define IO_MADE_GOOD ((struct bio *)2)
+
+#define BIO_SPECIAL(bio) ((unsigned long)bio <= 2)
 
 /* bits for r1bio.state */
 #define	R1BIO_Uptodate	0
 #define	R1BIO_IsSync	1
 #define	R1BIO_Degraded	2
 #define	R1BIO_BehindIO	3
+/* Set ReadError on bios that experience a readerror so that
+ * raid1d knows what to do with them.
+ */
+#define R1BIO_ReadError 4
 /* For write-behind requests, we call bi_end_io when
  * the last non-write-behind device completes, providing
  * any write was successful.  Otherwise we call when
@@ -125,6 +142,11 @@ struct r1bio_s {
  * Record that bi_end_io was called with this flag...
  */
 #define	R1BIO_Returned 6
+/* If a write for this request means we can clear some
+ * known-bad-block records, we set this flag
+ */
+#define	R1BIO_MadeGood 7
+#define	R1BIO_WriteError 8
 
 extern int md_raid1_congested(mddev_t *mddev, int bits);
 

drivers/md/raid10.c

@@ -22,6 +22,7 @@
 #include <linux/delay.h>
 #include <linux/blkdev.h>
 #include <linux/seq_file.h>
+#include <linux/ratelimit.h>
 #include "md.h"
 #include "raid10.h"
 #include "raid0.h"
@@ -123,7 +124,14 @@ static void * r10buf_pool_alloc(gfp_t gfp_flags, void *data)
 	for (j = 0 ; j < nalloc; j++) {
 		bio = r10_bio->devs[j].bio;
 		for (i = 0; i < RESYNC_PAGES; i++) {
-			page = alloc_page(gfp_flags);
+			if (j == 1 && !test_bit(MD_RECOVERY_SYNC,
+						&conf->mddev->recovery)) {
+				/* we can share bv_page's during recovery */
+				struct bio *rbio = r10_bio->devs[0].bio;
+				page = rbio->bi_io_vec[i].bv_page;
+				get_page(page);
+			} else
+				page = alloc_page(gfp_flags);
 			if (unlikely(!page))
 				goto out_free_pages;
 
@@ -173,7 +181,7 @@ static void put_all_bios(conf_t *conf, r10bio_t *r10_bio)
 
 	for (i = 0; i < conf->copies; i++) {
 		struct bio **bio = & r10_bio->devs[i].bio;
-		if (*bio && *bio != IO_BLOCKED)
+		if (!BIO_SPECIAL(*bio))
 			bio_put(*bio);
 		*bio = NULL;
 	}
@@ -183,12 +191,6 @@ static void free_r10bio(r10bio_t *r10_bio)
 {
 	conf_t *conf = r10_bio->mddev->private;
 
-	/*
-	 * Wake up any possible resync thread that waits for the device
-	 * to go idle.
-	 */
-	allow_barrier(conf);
-
 	put_all_bios(conf, r10_bio);
 	mempool_free(r10_bio, conf->r10bio_pool);
 }
@@ -227,9 +229,27 @@ static void reschedule_retry(r10bio_t *r10_bio)
 static void raid_end_bio_io(r10bio_t *r10_bio)
 {
 	struct bio *bio = r10_bio->master_bio;
+	int done;
+	conf_t *conf = r10_bio->mddev->private;
 
-	bio_endio(bio,
-		test_bit(R10BIO_Uptodate, &r10_bio->state) ? 0 : -EIO);
+	if (bio->bi_phys_segments) {
+		unsigned long flags;
+		spin_lock_irqsave(&conf->device_lock, flags);
+		bio->bi_phys_segments--;
+		done = (bio->bi_phys_segments == 0);
+		spin_unlock_irqrestore(&conf->device_lock, flags);
+	} else
+		done = 1;
+	if (!test_bit(R10BIO_Uptodate, &r10_bio->state))
+		clear_bit(BIO_UPTODATE, &bio->bi_flags);
+	if (done) {
+		bio_endio(bio, 0);
+		/*
+		 * Wake up any possible resync thread that waits for the device
+		 * to go idle.
+		 */
+		allow_barrier(conf);
+	}
 	free_r10bio(r10_bio);
 }
 
@@ -244,6 +264,26 @@ static inline void update_head_pos(int slot, r10bio_t *r10_bio)
 		r10_bio->devs[slot].addr + (r10_bio->sectors);
 }
 
+/*
+ * Find the disk number which triggered given bio
+ */
+static int find_bio_disk(conf_t *conf, r10bio_t *r10_bio,
+			 struct bio *bio, int *slotp)
+{
+	int slot;
+
+	for (slot = 0; slot < conf->copies; slot++)
+		if (r10_bio->devs[slot].bio == bio)
+			break;
+
+	BUG_ON(slot == conf->copies);
+	update_head_pos(slot, r10_bio);
+
+	if (slotp)
+		*slotp = slot;
+	return r10_bio->devs[slot].devnum;
+}
+
 static void raid10_end_read_request(struct bio *bio, int error)
 {
 	int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
@@ -277,34 +317,45 @@ static void raid10_end_read_request(struct bio *bio, int error)
 		 * oops, read error - keep the refcount on the rdev
 		 */
 		char b[BDEVNAME_SIZE];
-		if (printk_ratelimit())
-			printk(KERN_ERR "md/raid10:%s: %s: rescheduling sector %llu\n",
-			       mdname(conf->mddev),
-			       bdevname(conf->mirrors[dev].rdev->bdev,b), (unsigned long long)r10_bio->sector);
+		printk_ratelimited(KERN_ERR
+				   "md/raid10:%s: %s: rescheduling sector %llu\n",
+				   mdname(conf->mddev),
+				   bdevname(conf->mirrors[dev].rdev->bdev, b),
+				   (unsigned long long)r10_bio->sector);
+		set_bit(R10BIO_ReadError, &r10_bio->state);
 		reschedule_retry(r10_bio);
 	}
 }
 
+static void close_write(r10bio_t *r10_bio)
+{
+	/* clear the bitmap if all writes complete successfully */
+	bitmap_endwrite(r10_bio->mddev->bitmap, r10_bio->sector,
+			r10_bio->sectors,
+			!test_bit(R10BIO_Degraded, &r10_bio->state),
+			0);
+	md_write_end(r10_bio->mddev);
+}
+
 static void raid10_end_write_request(struct bio *bio, int error)
 {
 	int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
 	r10bio_t *r10_bio = bio->bi_private;
-	int slot, dev;
+	int dev;
+	int dec_rdev = 1;
 	conf_t *conf = r10_bio->mddev->private;
+	int slot;
 
-	for (slot = 0; slot < conf->copies; slot++)
-		if (r10_bio->devs[slot].bio == bio)
-			break;
-	dev = r10_bio->devs[slot].devnum;
+	dev = find_bio_disk(conf, r10_bio, bio, &slot);
 
 	/*
 	 * this branch is our 'one mirror IO has finished' event handler:
 	 */
 	if (!uptodate) {
-		md_error(r10_bio->mddev, conf->mirrors[dev].rdev);
-		/* an I/O failed, we can't clear the bitmap */
-		set_bit(R10BIO_Degraded, &r10_bio->state);
-	} else
+		set_bit(WriteErrorSeen,	&conf->mirrors[dev].rdev->flags);
+		set_bit(R10BIO_WriteError, &r10_bio->state);
+		dec_rdev = 0;
+	} else {
 		/*
 		 * Set R10BIO_Uptodate in our master bio, so that
 		 * we will return a good error code for to the higher
@@ -314,9 +365,22 @@ static void raid10_end_write_request(struct bio *bio, int error)
 		 * user-side. So if something waits for IO, then it will
 		 * wait for the 'master' bio.
 		 */
+		sector_t first_bad;
+		int bad_sectors;
+
 		set_bit(R10BIO_Uptodate, &r10_bio->state);
 
-	update_head_pos(slot, r10_bio);
+		/* Maybe we can clear some bad blocks. */
+		if (is_badblock(conf->mirrors[dev].rdev,
+				r10_bio->devs[slot].addr,
+				r10_bio->sectors,
+				&first_bad, &bad_sectors)) {
+			bio_put(bio);
+			r10_bio->devs[slot].bio = IO_MADE_GOOD;
+			dec_rdev = 0;
+			set_bit(R10BIO_MadeGood, &r10_bio->state);
+		}
+	}
 
 	/*
 	 *
@@ -324,16 +388,18 @@ static void raid10_end_write_request(struct bio *bio, int error)
 	 * already.
 	 */
 	if (atomic_dec_and_test(&r10_bio->remaining)) {
-		/* clear the bitmap if all writes complete successfully */
-		bitmap_endwrite(r10_bio->mddev->bitmap, r10_bio->sector,
-				r10_bio->sectors,
-				!test_bit(R10BIO_Degraded, &r10_bio->state),
-				0);
-		md_write_end(r10_bio->mddev);
-		raid_end_bio_io(r10_bio);
+		if (test_bit(R10BIO_WriteError, &r10_bio->state))
+			reschedule_retry(r10_bio);
+		else {
+			close_write(r10_bio);
+			if (test_bit(R10BIO_MadeGood, &r10_bio->state))
+				reschedule_retry(r10_bio);
+			else
+				raid_end_bio_io(r10_bio);
+		}
 	}
-
-	rdev_dec_pending(conf->mirrors[dev].rdev, conf->mddev);
+	if (dec_rdev)
+		rdev_dec_pending(conf->mirrors[dev].rdev, conf->mddev);
 }
 
 
@@ -484,11 +550,12 @@ static int raid10_mergeable_bvec(struct request_queue *q,
  * FIXME: possibly should rethink readbalancing and do it differently
  * depending on near_copies / far_copies geometry.
  */
-static int read_balance(conf_t *conf, r10bio_t *r10_bio)
+static int read_balance(conf_t *conf, r10bio_t *r10_bio, int *max_sectors)
 {
 	const sector_t this_sector = r10_bio->sector;
 	int disk, slot;
-	const int sectors = r10_bio->sectors;
+	int sectors = r10_bio->sectors;
+	int best_good_sectors;
 	sector_t new_distance, best_dist;
 	mdk_rdev_t *rdev;
 	int do_balance;
@@ -497,8 +564,10 @@ static int read_balance(conf_t *conf, r10bio_t *r10_bio)
 	raid10_find_phys(conf, r10_bio);
 	rcu_read_lock();
 retry:
+	sectors = r10_bio->sectors;
 	best_slot = -1;
 	best_dist = MaxSector;
+	best_good_sectors = 0;
 	do_balance = 1;
 	/*
 	 * Check if we can balance. We can balance on the whole
@@ -511,6 +580,10 @@ retry:
 		do_balance = 0;
 
 	for (slot = 0; slot < conf->copies ; slot++) {
+		sector_t first_bad;
+		int bad_sectors;
+		sector_t dev_sector;
+
 		if (r10_bio->devs[slot].bio == IO_BLOCKED)
 			continue;
 		disk = r10_bio->devs[slot].devnum;
@@ -520,6 +593,37 @@ retry:
 		if (!test_bit(In_sync, &rdev->flags))
 			continue;
 
+		dev_sector = r10_bio->devs[slot].addr;
+		if (is_badblock(rdev, dev_sector, sectors,
+				&first_bad, &bad_sectors)) {
+			if (best_dist < MaxSector)
+				/* Already have a better slot */
+				continue;
+			if (first_bad <= dev_sector) {
+				/* Cannot read here.  If this is the
+				 * 'primary' device, then we must not read
+				 * beyond 'bad_sectors' from another device.
+				 */
+				bad_sectors -= (dev_sector - first_bad);
+				if (!do_balance && sectors > bad_sectors)
+					sectors = bad_sectors;
+				if (best_good_sectors > sectors)
+					best_good_sectors = sectors;
+			} else {
+				sector_t good_sectors =
+					first_bad - dev_sector;
+				if (good_sectors > best_good_sectors) {
+					best_good_sectors = good_sectors;
+					best_slot = slot;
+				}
+				if (!do_balance)
+					/* Must read from here */
+					break;
+			}
+			continue;
+		} else
+			best_good_sectors = sectors;
+
 		if (!do_balance)
 			break;
 
@@ -561,6 +665,7 @@ retry:
 	} else
 		disk = -1;
 	rcu_read_unlock();
+	*max_sectors = best_good_sectors;
 
 	return disk;
 }
@@ -734,6 +839,8 @@ static int make_request(mddev_t *mddev, struct bio * bio)
 	unsigned long flags;
 	mdk_rdev_t *blocked_rdev;
 	int plugged;
+	int sectors_handled;
+	int max_sectors;
 
 	if (unlikely(bio->bi_rw & REQ_FLUSH)) {
 		md_flush_request(mddev, bio);
@@ -808,12 +915,26 @@ static int make_request(mddev_t *mddev, struct bio * bio)
 	r10_bio->sector = bio->bi_sector;
 	r10_bio->state = 0;
 
+	/* We might need to issue multiple reads to different
+	 * devices if there are bad blocks around, so we keep
+	 * track of the number of reads in bio->bi_phys_segments.
+	 * If this is 0, there is only one r10_bio and no locking
+	 * will be needed when the request completes.  If it is
+	 * non-zero, then it is the number of not-completed requests.
+	 */
+	bio->bi_phys_segments = 0;
+	clear_bit(BIO_SEG_VALID, &bio->bi_flags);
+
 	if (rw == READ) {
 		/*
 		 * read balancing logic:
 		 */
-		int disk = read_balance(conf, r10_bio);
-		int slot = r10_bio->read_slot;
+		int disk;
+		int slot;
+
+read_again:
+		disk = read_balance(conf, r10_bio, &max_sectors);
+		slot = r10_bio->read_slot;
 		if (disk < 0) {
 			raid_end_bio_io(r10_bio);
 			return 0;
@@ -821,6 +942,8 @@ static int make_request(mddev_t *mddev, struct bio * bio)
 		mirror = conf->mirrors + disk;
 
 		read_bio = bio_clone_mddev(bio, GFP_NOIO, mddev);
+		md_trim_bio(read_bio, r10_bio->sector - bio->bi_sector,
+			    max_sectors);
 
 		r10_bio->devs[slot].bio = read_bio;
 
@@ -831,7 +954,37 @@ static int make_request(mddev_t *mddev, struct bio * bio)
 		read_bio->bi_rw = READ | do_sync;
 		read_bio->bi_private = r10_bio;
 
-		generic_make_request(read_bio);
+		if (max_sectors < r10_bio->sectors) {
+			/* Could not read all from this device, so we will
+			 * need another r10_bio.
+			 */
+			sectors_handled = (r10_bio->sectors + max_sectors
+					   - bio->bi_sector);
+			r10_bio->sectors = max_sectors;
+			spin_lock_irq(&conf->device_lock);
+			if (bio->bi_phys_segments == 0)
+				bio->bi_phys_segments = 2;
+			else
+				bio->bi_phys_segments++;
+			spin_unlock(&conf->device_lock);
+			/* Cannot call generic_make_request directly
+			 * as that will be queued in __generic_make_request
+			 * and subsequent mempool_alloc might block
+			 * waiting for it.  so hand bio over to raid10d.
+			 */
+			reschedule_retry(r10_bio);
+
+			r10_bio = mempool_alloc(conf->r10bio_pool, GFP_NOIO);
+
+			r10_bio->master_bio = bio;
+			r10_bio->sectors = ((bio->bi_size >> 9)
+					    - sectors_handled);
+			r10_bio->state = 0;
+			r10_bio->mddev = mddev;
+			r10_bio->sector = bio->bi_sector + sectors_handled;
+			goto read_again;
+		} else
+			generic_make_request(read_bio);
 		return 0;
 	}
 
@@ -841,13 +994,22 @@ static int make_request(mddev_t *mddev, struct bio * bio)
 	/* first select target devices under rcu_lock and
 	 * inc refcount on their rdev.  Record them by setting
 	 * bios[x] to bio
+	 * If there are known/acknowledged bad blocks on any device
+	 * on which we have seen a write error, we want to avoid
+	 * writing to those blocks.  This potentially requires several
+	 * writes to write around the bad blocks.  Each set of writes
+	 * gets its own r10_bio with a set of bios attached.  The number
+	 * of r10_bios is recored in bio->bi_phys_segments just as with
+	 * the read case.
 	 */
 	plugged = mddev_check_plugged(mddev);
 
 	raid10_find_phys(conf, r10_bio);
- retry_write:
+retry_write:
 	blocked_rdev = NULL;
 	rcu_read_lock();
+	max_sectors = r10_bio->sectors;
+
 	for (i = 0;  i < conf->copies; i++) {
 		int d = r10_bio->devs[i].devnum;
 		mdk_rdev_t *rdev = rcu_dereference(conf->mirrors[d].rdev);
@@ -856,13 +1018,55 @@ static int make_request(mddev_t *mddev, struct bio * bio)
 			blocked_rdev = rdev;
 			break;
 		}
-		if (rdev && !test_bit(Faulty, &rdev->flags)) {
-			atomic_inc(&rdev->nr_pending);
-			r10_bio->devs[i].bio = bio;
-		} else {
-			r10_bio->devs[i].bio = NULL;
+		r10_bio->devs[i].bio = NULL;
+		if (!rdev || test_bit(Faulty, &rdev->flags)) {
 			set_bit(R10BIO_Degraded, &r10_bio->state);
+			continue;
 		}
+		if (test_bit(WriteErrorSeen, &rdev->flags)) {
+			sector_t first_bad;
+			sector_t dev_sector = r10_bio->devs[i].addr;
+			int bad_sectors;
+			int is_bad;
+
+			is_bad = is_badblock(rdev, dev_sector,
+					     max_sectors,
+					     &first_bad, &bad_sectors);
+			if (is_bad < 0) {
+				/* Mustn't write here until the bad block
+				 * is acknowledged
+				 */
+				atomic_inc(&rdev->nr_pending);
+				set_bit(BlockedBadBlocks, &rdev->flags);
+				blocked_rdev = rdev;
+				break;
+			}
+			if (is_bad && first_bad <= dev_sector) {
+				/* Cannot write here at all */
+				bad_sectors -= (dev_sector - first_bad);
+				if (bad_sectors < max_sectors)
+					/* Mustn't write more than bad_sectors
+					 * to other devices yet
+					 */
+					max_sectors = bad_sectors;
+				/* We don't set R10BIO_Degraded as that
+				 * only applies if the disk is missing,
+				 * so it might be re-added, and we want to
+				 * know to recover this chunk.
+				 * In this case the device is here, and the
+				 * fact that this chunk is not in-sync is
+				 * recorded in the bad block log.
+				 */
+				continue;
+			}
+			if (is_bad) {
+				int good_sectors = first_bad - dev_sector;
+				if (good_sectors < max_sectors)
+					max_sectors = good_sectors;
+			}
+		}
+		r10_bio->devs[i].bio = bio;
+		atomic_inc(&rdev->nr_pending);
 	}
 	rcu_read_unlock();
 
@@ -882,8 +1086,22 @@ static int make_request(mddev_t *mddev, struct bio * bio)
 		goto retry_write;
 	}
 
+	if (max_sectors < r10_bio->sectors) {
+		/* We are splitting this into multiple parts, so
+		 * we need to prepare for allocating another r10_bio.
+		 */
+		r10_bio->sectors = max_sectors;
+		spin_lock_irq(&conf->device_lock);
+		if (bio->bi_phys_segments == 0)
+			bio->bi_phys_segments = 2;
+		else
+			bio->bi_phys_segments++;
+		spin_unlock_irq(&conf->device_lock);
+	}
+	sectors_handled = r10_bio->sector + max_sectors - bio->bi_sector;
+
 	atomic_set(&r10_bio->remaining, 1);
-	bitmap_startwrite(mddev->bitmap, bio->bi_sector, r10_bio->sectors, 0);
+	bitmap_startwrite(mddev->bitmap, r10_bio->sector, r10_bio->sectors, 0);
 
 	for (i = 0; i < conf->copies; i++) {
 		struct bio *mbio;
@@ -892,10 +1110,12 @@ static int make_request(mddev_t *mddev, struct bio * bio)
 			continue;
 
 		mbio = bio_clone_mddev(bio, GFP_NOIO, mddev);
+		md_trim_bio(mbio, r10_bio->sector - bio->bi_sector,
+			    max_sectors);
 		r10_bio->devs[i].bio = mbio;
 
-		mbio->bi_sector	= r10_bio->devs[i].addr+
-			conf->mirrors[d].rdev->data_offset;
+		mbio->bi_sector	= (r10_bio->devs[i].addr+
+				   conf->mirrors[d].rdev->data_offset);
 		mbio->bi_bdev = conf->mirrors[d].rdev->bdev;
 		mbio->bi_end_io	= raid10_end_write_request;
 		mbio->bi_rw = WRITE | do_sync | do_fua;
@@ -920,6 +1140,21 @@ static int make_request(mddev_t *mddev, struct bio * bio)
 	/* In case raid10d snuck in to freeze_array */
 	wake_up(&conf->wait_barrier);
 
+	if (sectors_handled < (bio->bi_size >> 9)) {
+		/* We need another r10_bio.  It has already been counted
+		 * in bio->bi_phys_segments.
+		 */
+		r10_bio = mempool_alloc(conf->r10bio_pool, GFP_NOIO);
+
+		r10_bio->master_bio = bio;
+		r10_bio->sectors = (bio->bi_size >> 9) - sectors_handled;
+
+		r10_bio->mddev = mddev;
+		r10_bio->sector = bio->bi_sector + sectors_handled;
+		r10_bio->state = 0;
+		goto retry_write;
+	}
+
 	if (do_sync || !mddev->bitmap || !plugged)
 		md_wakeup_thread(mddev->thread);
 	return 0;
@@ -949,6 +1184,30 @@ static void status(struct seq_file *seq, mddev_t *mddev)
 	seq_printf(seq, "]");
 }
 
+/* check if there are enough drives for
+ * every block to appear on atleast one.
+ * Don't consider the device numbered 'ignore'
+ * as we might be about to remove it.
+ */
+static int enough(conf_t *conf, int ignore)
+{
+	int first = 0;
+
+	do {
+		int n = conf->copies;
+		int cnt = 0;
+		while (n--) {
+			if (conf->mirrors[first].rdev &&
+			    first != ignore)
+				cnt++;
+			first = (first+1) % conf->raid_disks;
+		}
+		if (cnt == 0)
+			return 0;
+	} while (first != 0);
+	return 1;
+}
+
 static void error(mddev_t *mddev, mdk_rdev_t *rdev)
 {
 	char b[BDEVNAME_SIZE];
@@ -961,13 +1220,9 @@ static void error(mddev_t *mddev, mdk_rdev_t *rdev)
 	 * else mark the drive as failed
 	 */
 	if (test_bit(In_sync, &rdev->flags)
-	    && conf->raid_disks-mddev->degraded == 1)
+	    && !enough(conf, rdev->raid_disk))
 		/*
 		 * Don't fail the drive, just return an IO error.
-		 * The test should really be more sophisticated than
-		 * "working_disks == 1", but it isn't critical, and
-		 * can wait until we do more sophisticated "is the drive
-		 * really dead" tests...
 		 */
 		return;
 	if (test_and_clear_bit(In_sync, &rdev->flags)) {
@@ -980,6 +1235,7 @@ static void error(mddev_t *mddev, mdk_rdev_t *rdev)
 		 */
 		set_bit(MD_RECOVERY_INTR, &mddev->recovery);
 	}
+	set_bit(Blocked, &rdev->flags);
 	set_bit(Faulty, &rdev->flags);
 	set_bit(MD_CHANGE_DEVS, &mddev->flags);
 	printk(KERN_ALERT
@@ -1022,27 +1278,6 @@ static void close_sync(conf_t *conf)
 	conf->r10buf_pool = NULL;
 }
 
-/* check if there are enough drives for
- * every block to appear on atleast one
- */
-static int enough(conf_t *conf)
-{
-	int first = 0;
-
-	do {
-		int n = conf->copies;
-		int cnt = 0;
-		while (n--) {
-			if (conf->mirrors[first].rdev)
-				cnt++;
-			first = (first+1) % conf->raid_disks;
-		}
-		if (cnt == 0)
-			return 0;
-	} while (first != 0);
-	return 1;
-}
-
 static int raid10_spare_active(mddev_t *mddev)
 {
 	int i;
@@ -1078,7 +1313,6 @@ static int raid10_add_disk(mddev_t *mddev, mdk_rdev_t *rdev)
 	conf_t *conf = mddev->private;
 	int err = -EEXIST;
 	int mirror;
-	mirror_info_t *p;
 	int first = 0;
 	int last = conf->raid_disks - 1;
 
@@ -1087,44 +1321,47 @@ static int raid10_add_disk(mddev_t *mddev, mdk_rdev_t *rdev)
 		 * very different from resync
 		 */
 		return -EBUSY;
-	if (!enough(conf))
+	if (!enough(conf, -1))
 		return -EINVAL;
 
 	if (rdev->raid_disk >= 0)
 		first = last = rdev->raid_disk;
 
-	if (rdev->saved_raid_disk >= 0 &&
-	    rdev->saved_raid_disk >= first &&
+	if (rdev->saved_raid_disk >= first &&
 	    conf->mirrors[rdev->saved_raid_disk].rdev == NULL)
 		mirror = rdev->saved_raid_disk;
 	else
 		mirror = first;
-	for ( ; mirror <= last ; mirror++)
-		if ( !(p=conf->mirrors+mirror)->rdev) {
-
-			disk_stack_limits(mddev->gendisk, rdev->bdev,
-					  rdev->data_offset << 9);
-			/* as we don't honour merge_bvec_fn, we must
-			 * never risk violating it, so limit
-			 * ->max_segments to one lying with a single
-			 * page, as a one page request is never in
-			 * violation.
-			 */
-			if (rdev->bdev->bd_disk->queue->merge_bvec_fn) {
-				blk_queue_max_segments(mddev->queue, 1);
-				blk_queue_segment_boundary(mddev->queue,
-							   PAGE_CACHE_SIZE - 1);
-			}
+	for ( ; mirror <= last ; mirror++) {
+		mirror_info_t *p = &conf->mirrors[mirror];
+		if (p->recovery_disabled == mddev->recovery_disabled)
+			continue;
+		if (!p->rdev)
+			continue;
 
-			p->head_position = 0;
-			rdev->raid_disk = mirror;
-			err = 0;
-			if (rdev->saved_raid_disk != mirror)
-				conf->fullsync = 1;
-			rcu_assign_pointer(p->rdev, rdev);
-			break;
+		disk_stack_limits(mddev->gendisk, rdev->bdev,
+				  rdev->data_offset << 9);
+		/* as we don't honour merge_bvec_fn, we must
+		 * never risk violating it, so limit
+		 * ->max_segments to one lying with a single
+		 * page, as a one page request is never in
+		 * violation.
+		 */
+		if (rdev->bdev->bd_disk->queue->merge_bvec_fn) {
+			blk_queue_max_segments(mddev->queue, 1);
+			blk_queue_segment_boundary(mddev->queue,
+						   PAGE_CACHE_SIZE - 1);
 		}
 
+		p->head_position = 0;
+		rdev->raid_disk = mirror;
+		err = 0;
+		if (rdev->saved_raid_disk != mirror)
+			conf->fullsync = 1;
+		rcu_assign_pointer(p->rdev, rdev);
+		break;
+	}
+
 	md_integrity_add_rdev(rdev, mddev);
 	print_conf(conf);
 	return err;
@@ -1149,7 +1386,8 @@ static int raid10_remove_disk(mddev_t *mddev, int number)
 		 * is not possible.
 		 */
 		if (!test_bit(Faulty, &rdev->flags) &&
-		    enough(conf)) {
+		    mddev->recovery_disabled != p->recovery_disabled &&
+		    enough(conf, -1)) {
 			err = -EBUSY;
 			goto abort;
 		}
@@ -1174,24 +1412,18 @@ static void end_sync_read(struct bio *bio, int error)
 {
 	r10bio_t *r10_bio = bio->bi_private;
 	conf_t *conf = r10_bio->mddev->private;
-	int i,d;
+	int d;
 
-	for (i=0; i<conf->copies; i++)
-		if (r10_bio->devs[i].bio == bio)
-			break;
-	BUG_ON(i == conf->copies);
-	update_head_pos(i, r10_bio);
-	d = r10_bio->devs[i].devnum;
+	d = find_bio_disk(conf, r10_bio, bio, NULL);
 
 	if (test_bit(BIO_UPTODATE, &bio->bi_flags))
 		set_bit(R10BIO_Uptodate, &r10_bio->state);
-	else {
+	else
+		/* The write handler will notice the lack of
+		 * R10BIO_Uptodate and record any errors etc
+		 */
 		atomic_add(r10_bio->sectors,
 			   &conf->mirrors[d].rdev->corrected_errors);
-		if (!test_bit(MD_RECOVERY_SYNC, &conf->mddev->recovery))
-			md_error(r10_bio->mddev,
-				 conf->mirrors[d].rdev);
-	}
 
 	/* for reconstruct, we always reschedule after a read.
 	 * for resync, only after all reads
@@ -1206,40 +1438,60 @@ static void end_sync_read(struct bio *bio, int error)
 	}
 }
 
-static void end_sync_write(struct bio *bio, int error)
+static void end_sync_request(r10bio_t *r10_bio)
 {
-	int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
-	r10bio_t *r10_bio = bio->bi_private;
 	mddev_t *mddev = r10_bio->mddev;
-	conf_t *conf = mddev->private;
-	int i,d;
-
-	for (i = 0; i < conf->copies; i++)
-		if (r10_bio->devs[i].bio == bio)
-			break;
-	d = r10_bio->devs[i].devnum;
 
-	if (!uptodate)
-		md_error(mddev, conf->mirrors[d].rdev);
-
-	update_head_pos(i, r10_bio);
-
-	rdev_dec_pending(conf->mirrors[d].rdev, mddev);
 	while (atomic_dec_and_test(&r10_bio->remaining)) {
 		if (r10_bio->master_bio == NULL) {
 			/* the primary of several recovery bios */
 			sector_t s = r10_bio->sectors;
-			put_buf(r10_bio);
+			if (test_bit(R10BIO_MadeGood, &r10_bio->state) ||
+			    test_bit(R10BIO_WriteError, &r10_bio->state))
+				reschedule_retry(r10_bio);
+			else
+				put_buf(r10_bio);
 			md_done_sync(mddev, s, 1);
 			break;
 		} else {
 			r10bio_t *r10_bio2 = (r10bio_t *)r10_bio->master_bio;
-			put_buf(r10_bio);
+			if (test_bit(R10BIO_MadeGood, &r10_bio->state) ||
+			    test_bit(R10BIO_WriteError, &r10_bio->state))
+				reschedule_retry(r10_bio);
+			else
+				put_buf(r10_bio);
 			r10_bio = r10_bio2;
 		}
 	}
 }
 
+static void end_sync_write(struct bio *bio, int error)
+{
+	int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
+	r10bio_t *r10_bio = bio->bi_private;
+	mddev_t *mddev = r10_bio->mddev;
+	conf_t *conf = mddev->private;
+	int d;
+	sector_t first_bad;
+	int bad_sectors;
+	int slot;
+
+	d = find_bio_disk(conf, r10_bio, bio, &slot);
+
+	if (!uptodate) {
+		set_bit(WriteErrorSeen, &conf->mirrors[d].rdev->flags);
+		set_bit(R10BIO_WriteError, &r10_bio->state);
+	} else if (is_badblock(conf->mirrors[d].rdev,
+			     r10_bio->devs[slot].addr,
+			     r10_bio->sectors,
+			     &first_bad, &bad_sectors))
+		set_bit(R10BIO_MadeGood, &r10_bio->state);
+
+	rdev_dec_pending(conf->mirrors[d].rdev, mddev);
+
+	end_sync_request(r10_bio);
+}
+
 /*
  * Note: sync and recover and handled very differently for raid10
  * This code is for resync.
@@ -1299,11 +1551,12 @@ static void sync_request_write(mddev_t *mddev, r10bio_t *r10_bio)
 			if (j == vcnt)
 				continue;
 			mddev->resync_mismatches += r10_bio->sectors;
+			if (test_bit(MD_RECOVERY_CHECK, &mddev->recovery))
+				/* Don't fix anything. */
+				continue;
 		}
-		if (test_bit(MD_RECOVERY_CHECK, &mddev->recovery))
-			/* Don't fix anything. */
-			continue;
-		/* Ok, we need to write this bio
+		/* Ok, we need to write this bio, either to correct an
+		 * inconsistency or to correct an unreadable block.
 		 * First we need to fixup bv_offset, bv_len and
 		 * bi_vecs, as the read request might have corrupted these
 		 */
@@ -1355,32 +1608,107 @@ done:
  * The second for writing.
  *
  */
+static void fix_recovery_read_error(r10bio_t *r10_bio)
+{
+	/* We got a read error during recovery.
+	 * We repeat the read in smaller page-sized sections.
+	 * If a read succeeds, write it to the new device or record
+	 * a bad block if we cannot.
+	 * If a read fails, record a bad block on both old and
+	 * new devices.
+	 */
+	mddev_t *mddev = r10_bio->mddev;
+	conf_t *conf = mddev->private;
+	struct bio *bio = r10_bio->devs[0].bio;
+	sector_t sect = 0;
+	int sectors = r10_bio->sectors;
+	int idx = 0;
+	int dr = r10_bio->devs[0].devnum;
+	int dw = r10_bio->devs[1].devnum;
+
+	while (sectors) {
+		int s = sectors;
+		mdk_rdev_t *rdev;
+		sector_t addr;
+		int ok;
+
+		if (s > (PAGE_SIZE>>9))
+			s = PAGE_SIZE >> 9;
+
+		rdev = conf->mirrors[dr].rdev;
+		addr = r10_bio->devs[0].addr + sect,
+		ok = sync_page_io(rdev,
+				  addr,
+				  s << 9,
+				  bio->bi_io_vec[idx].bv_page,
+				  READ, false);
+		if (ok) {
+			rdev = conf->mirrors[dw].rdev;
+			addr = r10_bio->devs[1].addr + sect;
+			ok = sync_page_io(rdev,
+					  addr,
+					  s << 9,
+					  bio->bi_io_vec[idx].bv_page,
+					  WRITE, false);
+			if (!ok)
+				set_bit(WriteErrorSeen, &rdev->flags);
+		}
+		if (!ok) {
+			/* We don't worry if we cannot set a bad block -
+			 * it really is bad so there is no loss in not
+			 * recording it yet
+			 */
+			rdev_set_badblocks(rdev, addr, s, 0);
+
+			if (rdev != conf->mirrors[dw].rdev) {
+				/* need bad block on destination too */
+				mdk_rdev_t *rdev2 = conf->mirrors[dw].rdev;
+				addr = r10_bio->devs[1].addr + sect;
+				ok = rdev_set_badblocks(rdev2, addr, s, 0);
+				if (!ok) {
+					/* just abort the recovery */
+					printk(KERN_NOTICE
+					       "md/raid10:%s: recovery aborted"
+					       " due to read error\n",
+					       mdname(mddev));
+
+					conf->mirrors[dw].recovery_disabled
+						= mddev->recovery_disabled;
+					set_bit(MD_RECOVERY_INTR,
+						&mddev->recovery);
+					break;
+				}
+			}
+		}
+
+		sectors -= s;
+		sect += s;
+		idx++;
+	}
+}
 
 static void recovery_request_write(mddev_t *mddev, r10bio_t *r10_bio)
 {
 	conf_t *conf = mddev->private;
-	int i, d;
-	struct bio *bio, *wbio;
+	int d;
+	struct bio *wbio;
 
+	if (!test_bit(R10BIO_Uptodate, &r10_bio->state)) {
+		fix_recovery_read_error(r10_bio);
+		end_sync_request(r10_bio);
+		return;
+	}
 
-	/* move the pages across to the second bio
+	/*
+	 * share the pages with the first bio
 	 * and submit the write request
 	 */
-	bio = r10_bio->devs[0].bio;
 	wbio = r10_bio->devs[1].bio;
-	for (i=0; i < wbio->bi_vcnt; i++) {
-		struct page *p = bio->bi_io_vec[i].bv_page;
-		bio->bi_io_vec[i].bv_page = wbio->bi_io_vec[i].bv_page;
-		wbio->bi_io_vec[i].bv_page = p;
-	}
 	d = r10_bio->devs[1].devnum;
 
 	atomic_inc(&conf->mirrors[d].rdev->nr_pending);
 	md_sync_acct(conf->mirrors[d].rdev->bdev, wbio->bi_size >> 9);
-	if (test_bit(R10BIO_Uptodate, &r10_bio->state))
-		generic_make_request(wbio);
-	else
-		bio_endio(wbio, -EIO);
+	generic_make_request(wbio);
 }
 
 
@@ -1421,6 +1749,26 @@ static void check_decay_read_errors(mddev_t *mddev, mdk_rdev_t *rdev)
 		atomic_set(&rdev->read_errors, read_errors >> hours_since_last);
 }
 
+static int r10_sync_page_io(mdk_rdev_t *rdev, sector_t sector,
+			    int sectors, struct page *page, int rw)
+{
+	sector_t first_bad;
+	int bad_sectors;
+
+	if (is_badblock(rdev, sector, sectors, &first_bad, &bad_sectors)
+	    && (rw == READ || test_bit(WriteErrorSeen, &rdev->flags)))
+		return -1;
+	if (sync_page_io(rdev, sector, sectors << 9, page, rw, false))
+		/* success */
+		return 1;
+	if (rw == WRITE)
+		set_bit(WriteErrorSeen, &rdev->flags);
+	/* need to record an error - either for the block or the device */
+	if (!rdev_set_badblocks(rdev, sector, sectors, 0))
+		md_error(rdev->mddev, rdev);
+	return 0;
+}
+
 /*
  * This is a kernel thread which:
  *
@@ -1476,10 +1824,15 @@ static void fix_read_error(conf_t *conf, mddev_t *mddev, r10bio_t *r10_bio)
 
 		rcu_read_lock();
 		do {
+			sector_t first_bad;
+			int bad_sectors;
+
 			d = r10_bio->devs[sl].devnum;
 			rdev = rcu_dereference(conf->mirrors[d].rdev);
 			if (rdev &&
-			    test_bit(In_sync, &rdev->flags)) {
+			    test_bit(In_sync, &rdev->flags) &&
+			    is_badblock(rdev, r10_bio->devs[sl].addr + sect, s,
+					&first_bad, &bad_sectors) == 0) {
 				atomic_inc(&rdev->nr_pending);
 				rcu_read_unlock();
 				success = sync_page_io(rdev,
@@ -1499,9 +1852,19 @@ static void fix_read_error(conf_t *conf, mddev_t *mddev, r10bio_t *r10_bio)
 		rcu_read_unlock();
 
 		if (!success) {
-			/* Cannot read from anywhere -- bye bye array */
+			/* Cannot read from anywhere, just mark the block
+			 * as bad on the first device to discourage future
+			 * reads.
+			 */
 			int dn = r10_bio->devs[r10_bio->read_slot].devnum;
-			md_error(mddev, conf->mirrors[dn].rdev);
+			rdev = conf->mirrors[dn].rdev;
+
+			if (!rdev_set_badblocks(
+				    rdev,
+				    r10_bio->devs[r10_bio->read_slot].addr
+				    + sect,
+				    s, 0))
+				md_error(mddev, rdev);
 			break;
 		}
 
@@ -1516,80 +1879,82 @@ static void fix_read_error(conf_t *conf, mddev_t *mddev, r10bio_t *r10_bio)
 			sl--;
 			d = r10_bio->devs[sl].devnum;
 			rdev = rcu_dereference(conf->mirrors[d].rdev);
-			if (rdev &&
-			    test_bit(In_sync, &rdev->flags)) {
-				atomic_inc(&rdev->nr_pending);
-				rcu_read_unlock();
-				atomic_add(s, &rdev->corrected_errors);
-				if (sync_page_io(rdev,
-						 r10_bio->devs[sl].addr +
-						 sect,
-						 s<<9, conf->tmppage, WRITE, false)
-				    == 0) {
-					/* Well, this device is dead */
-					printk(KERN_NOTICE
-					       "md/raid10:%s: read correction "
-					       "write failed"
-					       " (%d sectors at %llu on %s)\n",
-					       mdname(mddev), s,
-					       (unsigned long long)(
-						       sect + rdev->data_offset),
-					       bdevname(rdev->bdev, b));
-					printk(KERN_NOTICE "md/raid10:%s: %s: failing "
-					       "drive\n",
-					       mdname(mddev),
-					       bdevname(rdev->bdev, b));
-					md_error(mddev, rdev);
-				}
-				rdev_dec_pending(rdev, mddev);
-				rcu_read_lock();
+			if (!rdev ||
+			    !test_bit(In_sync, &rdev->flags))
+				continue;
+
+			atomic_inc(&rdev->nr_pending);
+			rcu_read_unlock();
+			if (r10_sync_page_io(rdev,
+					     r10_bio->devs[sl].addr +
+					     sect,
+					     s<<9, conf->tmppage, WRITE)
+			    == 0) {
+				/* Well, this device is dead */
+				printk(KERN_NOTICE
+				       "md/raid10:%s: read correction "
+				       "write failed"
+				       " (%d sectors at %llu on %s)\n",
+				       mdname(mddev), s,
+				       (unsigned long long)(
+					       sect + rdev->data_offset),
+				       bdevname(rdev->bdev, b));
+				printk(KERN_NOTICE "md/raid10:%s: %s: failing "
+				       "drive\n",
+				       mdname(mddev),
+				       bdevname(rdev->bdev, b));
 			}
+			rdev_dec_pending(rdev, mddev);
+			rcu_read_lock();
 		}
 		sl = start;
 		while (sl != r10_bio->read_slot) {
+			char b[BDEVNAME_SIZE];
 
 			if (sl==0)
 				sl = conf->copies;
 			sl--;
 			d = r10_bio->devs[sl].devnum;
 			rdev = rcu_dereference(conf->mirrors[d].rdev);
-			if (rdev &&
-			    test_bit(In_sync, &rdev->flags)) {
-				char b[BDEVNAME_SIZE];
-				atomic_inc(&rdev->nr_pending);
-				rcu_read_unlock();
-				if (sync_page_io(rdev,
-						 r10_bio->devs[sl].addr +
-						 sect,
-						 s<<9, conf->tmppage,
-						 READ, false) == 0) {
-					/* Well, this device is dead */
-					printk(KERN_NOTICE
-					       "md/raid10:%s: unable to read back "
-					       "corrected sectors"
-					       " (%d sectors at %llu on %s)\n",
-					       mdname(mddev), s,
-					       (unsigned long long)(
-						       sect + rdev->data_offset),
-					       bdevname(rdev->bdev, b));
-					printk(KERN_NOTICE "md/raid10:%s: %s: failing drive\n",
-					       mdname(mddev),
-					       bdevname(rdev->bdev, b));
-
-					md_error(mddev, rdev);
-				} else {
-					printk(KERN_INFO
-					       "md/raid10:%s: read error corrected"
-					       " (%d sectors at %llu on %s)\n",
-					       mdname(mddev), s,
-					       (unsigned long long)(
-						       sect + rdev->data_offset),
-					       bdevname(rdev->bdev, b));
-				}
+			if (!rdev ||
+			    !test_bit(In_sync, &rdev->flags))
+				continue;
 
-				rdev_dec_pending(rdev, mddev);
-				rcu_read_lock();
+			atomic_inc(&rdev->nr_pending);
+			rcu_read_unlock();
+			switch (r10_sync_page_io(rdev,
+					     r10_bio->devs[sl].addr +
+					     sect,
+					     s<<9, conf->tmppage,
+						 READ)) {
+			case 0:
+				/* Well, this device is dead */
+				printk(KERN_NOTICE
+				       "md/raid10:%s: unable to read back "
+				       "corrected sectors"
+				       " (%d sectors at %llu on %s)\n",
+				       mdname(mddev), s,
+				       (unsigned long long)(
+					       sect + rdev->data_offset),
+				       bdevname(rdev->bdev, b));
+				printk(KERN_NOTICE "md/raid10:%s: %s: failing "
+				       "drive\n",
+				       mdname(mddev),
+				       bdevname(rdev->bdev, b));
+				break;
+			case 1:
+				printk(KERN_INFO
+				       "md/raid10:%s: read error corrected"
+				       " (%d sectors at %llu on %s)\n",
+				       mdname(mddev), s,
+				       (unsigned long long)(
+					       sect + rdev->data_offset),
+				       bdevname(rdev->bdev, b));
+				atomic_add(s, &rdev->corrected_errors);
 			}
+
+			rdev_dec_pending(rdev, mddev);
+			rcu_read_lock();
 		}
 		rcu_read_unlock();
 
@@ -1598,21 +1963,254 @@ static void fix_read_error(conf_t *conf, mddev_t *mddev, r10bio_t *r10_bio)
 	}
 }
 
+static void bi_complete(struct bio *bio, int error)
+{
+	complete((struct completion *)bio->bi_private);
+}
+
+static int submit_bio_wait(int rw, struct bio *bio)
+{
+	struct completion event;
+	rw |= REQ_SYNC;
+
+	init_completion(&event);
+	bio->bi_private = &event;
+	bio->bi_end_io = bi_complete;
+	submit_bio(rw, bio);
+	wait_for_completion(&event);
+
+	return test_bit(BIO_UPTODATE, &bio->bi_flags);
+}
+
+static int narrow_write_error(r10bio_t *r10_bio, int i)
+{
+	struct bio *bio = r10_bio->master_bio;
+	mddev_t *mddev = r10_bio->mddev;
+	conf_t *conf = mddev->private;
+	mdk_rdev_t *rdev = conf->mirrors[r10_bio->devs[i].devnum].rdev;
+	/* bio has the data to be written to slot 'i' where
+	 * we just recently had a write error.
+	 * We repeatedly clone the bio and trim down to one block,
+	 * then try the write.  Where the write fails we record
+	 * a bad block.
+	 * It is conceivable that the bio doesn't exactly align with
+	 * blocks.  We must handle this.
+	 *
+	 * We currently own a reference to the rdev.
+	 */
+
+	int block_sectors;
+	sector_t sector;
+	int sectors;
+	int sect_to_write = r10_bio->sectors;
+	int ok = 1;
+
+	if (rdev->badblocks.shift < 0)
+		return 0;
+
+	block_sectors = 1 << rdev->badblocks.shift;
+	sector = r10_bio->sector;
+	sectors = ((r10_bio->sector + block_sectors)
+		   & ~(sector_t)(block_sectors - 1))
+		- sector;
+
+	while (sect_to_write) {
+		struct bio *wbio;
+		if (sectors > sect_to_write)
+			sectors = sect_to_write;
+		/* Write at 'sector' for 'sectors' */
+		wbio = bio_clone_mddev(bio, GFP_NOIO, mddev);
+		md_trim_bio(wbio, sector - bio->bi_sector, sectors);
+		wbio->bi_sector = (r10_bio->devs[i].addr+
+				   rdev->data_offset+
+				   (sector - r10_bio->sector));
+		wbio->bi_bdev = rdev->bdev;
+		if (submit_bio_wait(WRITE, wbio) == 0)
+			/* Failure! */
+			ok = rdev_set_badblocks(rdev, sector,
+						sectors, 0)
+				&& ok;
+
+		bio_put(wbio);
+		sect_to_write -= sectors;
+		sector += sectors;
+		sectors = block_sectors;
+	}
+	return ok;
+}
+
+static void handle_read_error(mddev_t *mddev, r10bio_t *r10_bio)
+{
+	int slot = r10_bio->read_slot;
+	int mirror = r10_bio->devs[slot].devnum;
+	struct bio *bio;
+	conf_t *conf = mddev->private;
+	mdk_rdev_t *rdev;
+	char b[BDEVNAME_SIZE];
+	unsigned long do_sync;
+	int max_sectors;
+
+	/* we got a read error. Maybe the drive is bad.  Maybe just
+	 * the block and we can fix it.
+	 * We freeze all other IO, and try reading the block from
+	 * other devices.  When we find one, we re-write
+	 * and check it that fixes the read error.
+	 * This is all done synchronously while the array is
+	 * frozen.
+	 */
+	if (mddev->ro == 0) {
+		freeze_array(conf);
+		fix_read_error(conf, mddev, r10_bio);
+		unfreeze_array(conf);
+	}
+	rdev_dec_pending(conf->mirrors[mirror].rdev, mddev);
+
+	bio = r10_bio->devs[slot].bio;
+	bdevname(bio->bi_bdev, b);
+	r10_bio->devs[slot].bio =
+		mddev->ro ? IO_BLOCKED : NULL;
+read_more:
+	mirror = read_balance(conf, r10_bio, &max_sectors);
+	if (mirror == -1) {
+		printk(KERN_ALERT "md/raid10:%s: %s: unrecoverable I/O"
+		       " read error for block %llu\n",
+		       mdname(mddev), b,
+		       (unsigned long long)r10_bio->sector);
+		raid_end_bio_io(r10_bio);
+		bio_put(bio);
+		return;
+	}
+
+	do_sync = (r10_bio->master_bio->bi_rw & REQ_SYNC);
+	if (bio)
+		bio_put(bio);
+	slot = r10_bio->read_slot;
+	rdev = conf->mirrors[mirror].rdev;
+	printk_ratelimited(
+		KERN_ERR
+		"md/raid10:%s: %s: redirecting"
+		"sector %llu to another mirror\n",
+		mdname(mddev),
+		bdevname(rdev->bdev, b),
+		(unsigned long long)r10_bio->sector);
+	bio = bio_clone_mddev(r10_bio->master_bio,
+			      GFP_NOIO, mddev);
+	md_trim_bio(bio,
+		    r10_bio->sector - bio->bi_sector,
+		    max_sectors);
+	r10_bio->devs[slot].bio = bio;
+	bio->bi_sector = r10_bio->devs[slot].addr
+		+ rdev->data_offset;
+	bio->bi_bdev = rdev->bdev;
+	bio->bi_rw = READ | do_sync;
+	bio->bi_private = r10_bio;
+	bio->bi_end_io = raid10_end_read_request;
+	if (max_sectors < r10_bio->sectors) {
+		/* Drat - have to split this up more */
+		struct bio *mbio = r10_bio->master_bio;
+		int sectors_handled =
+			r10_bio->sector + max_sectors
+			- mbio->bi_sector;
+		r10_bio->sectors = max_sectors;
+		spin_lock_irq(&conf->device_lock);
+		if (mbio->bi_phys_segments == 0)
+			mbio->bi_phys_segments = 2;
+		else
+			mbio->bi_phys_segments++;
+		spin_unlock_irq(&conf->device_lock);
+		generic_make_request(bio);
+		bio = NULL;
+
+		r10_bio = mempool_alloc(conf->r10bio_pool,
+					GFP_NOIO);
+		r10_bio->master_bio = mbio;
+		r10_bio->sectors = (mbio->bi_size >> 9)
+			- sectors_handled;
+		r10_bio->state = 0;
+		set_bit(R10BIO_ReadError,
+			&r10_bio->state);
+		r10_bio->mddev = mddev;
+		r10_bio->sector = mbio->bi_sector
+			+ sectors_handled;
+
+		goto read_more;
+	} else
+		generic_make_request(bio);
+}
+
+static void handle_write_completed(conf_t *conf, r10bio_t *r10_bio)
+{
+	/* Some sort of write request has finished and it
+	 * succeeded in writing where we thought there was a
+	 * bad block.  So forget the bad block.
+	 * Or possibly if failed and we need to record
+	 * a bad block.
+	 */
+	int m;
+	mdk_rdev_t *rdev;
+
+	if (test_bit(R10BIO_IsSync, &r10_bio->state) ||
+	    test_bit(R10BIO_IsRecover, &r10_bio->state)) {
+		for (m = 0; m < conf->copies; m++) {
+			int dev = r10_bio->devs[m].devnum;
+			rdev = conf->mirrors[dev].rdev;
+			if (r10_bio->devs[m].bio == NULL)
+				continue;
+			if (test_bit(BIO_UPTODATE,
+				     &r10_bio->devs[m].bio->bi_flags)) {
+				rdev_clear_badblocks(
+					rdev,
+					r10_bio->devs[m].addr,
+					r10_bio->sectors);
+			} else {
+				if (!rdev_set_badblocks(
+					    rdev,
+					    r10_bio->devs[m].addr,
+					    r10_bio->sectors, 0))
+					md_error(conf->mddev, rdev);
+			}
+		}
+		put_buf(r10_bio);
+	} else {
+		for (m = 0; m < conf->copies; m++) {
+			int dev = r10_bio->devs[m].devnum;
+			struct bio *bio = r10_bio->devs[m].bio;
+			rdev = conf->mirrors[dev].rdev;
+			if (bio == IO_MADE_GOOD) {
+				rdev_clear_badblocks(
+					rdev,
+					r10_bio->devs[m].addr,
+					r10_bio->sectors);
+				rdev_dec_pending(rdev, conf->mddev);
+			} else if (bio != NULL &&
+				   !test_bit(BIO_UPTODATE, &bio->bi_flags)) {
+				if (!narrow_write_error(r10_bio, m)) {
+					md_error(conf->mddev, rdev);
+					set_bit(R10BIO_Degraded,
+						&r10_bio->state);
+				}
+				rdev_dec_pending(rdev, conf->mddev);
+			}
+		}
+		if (test_bit(R10BIO_WriteError,
+			     &r10_bio->state))
+			close_write(r10_bio);
+		raid_end_bio_io(r10_bio);
+	}
+}
+
 static void raid10d(mddev_t *mddev)
 {
 	r10bio_t *r10_bio;
-	struct bio *bio;
 	unsigned long flags;
 	conf_t *conf = mddev->private;
 	struct list_head *head = &conf->retry_list;
-	mdk_rdev_t *rdev;
 	struct blk_plug plug;
 
 	md_check_recovery(mddev);
 
 	blk_start_plug(&plug);
 	for (;;) {
-		char b[BDEVNAME_SIZE];
 
 		flush_pending_writes(conf);
 
@@ -1628,64 +2226,26 @@ static void raid10d(mddev_t *mddev)
 
 		mddev = r10_bio->mddev;
 		conf = mddev->private;
-		if (test_bit(R10BIO_IsSync, &r10_bio->state))
+		if (test_bit(R10BIO_MadeGood, &r10_bio->state) ||
+		    test_bit(R10BIO_WriteError, &r10_bio->state))
+			handle_write_completed(conf, r10_bio);
+		else if (test_bit(R10BIO_IsSync, &r10_bio->state))
 			sync_request_write(mddev, r10_bio);
 		else if (test_bit(R10BIO_IsRecover, &r10_bio->state))
 			recovery_request_write(mddev, r10_bio);
+		else if (test_bit(R10BIO_ReadError, &r10_bio->state))
+			handle_read_error(mddev, r10_bio);
 		else {
-			int slot = r10_bio->read_slot;
-			int mirror = r10_bio->devs[slot].devnum;
-			/* we got a read error. Maybe the drive is bad.  Maybe just
-			 * the block and we can fix it.
-			 * We freeze all other IO, and try reading the block from
-			 * other devices.  When we find one, we re-write
-			 * and check it that fixes the read error.
-			 * This is all done synchronously while the array is
-			 * frozen.
+			/* just a partial read to be scheduled from a
+			 * separate context
 			 */
-			if (mddev->ro == 0) {
-				freeze_array(conf);
-				fix_read_error(conf, mddev, r10_bio);
-				unfreeze_array(conf);
-			}
-			rdev_dec_pending(conf->mirrors[mirror].rdev, mddev);
-
-			bio = r10_bio->devs[slot].bio;
-			r10_bio->devs[slot].bio =
-				mddev->ro ? IO_BLOCKED : NULL;
-			mirror = read_balance(conf, r10_bio);
-			if (mirror == -1) {
-				printk(KERN_ALERT "md/raid10:%s: %s: unrecoverable I/O"
-				       " read error for block %llu\n",
-				       mdname(mddev),
-				       bdevname(bio->bi_bdev,b),
-				       (unsigned long long)r10_bio->sector);
-				raid_end_bio_io(r10_bio);
-				bio_put(bio);
-			} else {
-				const unsigned long do_sync = (r10_bio->master_bio->bi_rw & REQ_SYNC);
-				bio_put(bio);
-				slot = r10_bio->read_slot;
-				rdev = conf->mirrors[mirror].rdev;
-				if (printk_ratelimit())
-					printk(KERN_ERR "md/raid10:%s: %s: redirecting sector %llu to"
-					       " another mirror\n",
-					       mdname(mddev),
-					       bdevname(rdev->bdev,b),
-					       (unsigned long long)r10_bio->sector);
-				bio = bio_clone_mddev(r10_bio->master_bio,
-						      GFP_NOIO, mddev);
-				r10_bio->devs[slot].bio = bio;
-				bio->bi_sector = r10_bio->devs[slot].addr
-					+ rdev->data_offset;
-				bio->bi_bdev = rdev->bdev;
-				bio->bi_rw = READ | do_sync;
-				bio->bi_private = r10_bio;
-				bio->bi_end_io = raid10_end_read_request;
-				generic_make_request(bio);
-			}
+			int slot = r10_bio->read_slot;
+			generic_make_request(r10_bio->devs[slot].bio);
 		}
+
 		cond_resched();
+		if (mddev->flags & ~(1<<MD_CHANGE_PENDING))
+			md_check_recovery(mddev);
 	}
 	blk_finish_plug(&plug);
 }
@@ -1746,7 +2306,6 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr,
 	int i;
 	int max_sync;
 	sector_t sync_blocks;
-
 	sector_t sectors_skipped = 0;
 	int chunks_skipped = 0;
 
@@ -1828,7 +2387,7 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr,
 	max_sync = RESYNC_PAGES << (PAGE_SHIFT-9);
 	if (!test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) {
 		/* recovery... the complicated one */
-		int j, k;
+		int j;
 		r10_bio = NULL;
 
 		for (i=0 ; i<conf->raid_disks; i++) {
@@ -1836,6 +2395,7 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr,
 			r10bio_t *rb2;
 			sector_t sect;
 			int must_sync;
+			int any_working;
 
 			if (conf->mirrors[i].rdev == NULL ||
 			    test_bit(In_sync, &conf->mirrors[i].rdev->flags)) 
@@ -1887,19 +2447,42 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr,
 			must_sync = bitmap_start_sync(mddev->bitmap, sect,
 						      &sync_blocks, still_degraded);
 
+			any_working = 0;
 			for (j=0; j<conf->copies;j++) {
+				int k;
 				int d = r10_bio->devs[j].devnum;
+				sector_t from_addr, to_addr;
+				mdk_rdev_t *rdev;
+				sector_t sector, first_bad;
+				int bad_sectors;
 				if (!conf->mirrors[d].rdev ||
 				    !test_bit(In_sync, &conf->mirrors[d].rdev->flags))
 					continue;
 				/* This is where we read from */
+				any_working = 1;
+				rdev = conf->mirrors[d].rdev;
+				sector = r10_bio->devs[j].addr;
+
+				if (is_badblock(rdev, sector, max_sync,
+						&first_bad, &bad_sectors)) {
+					if (first_bad > sector)
+						max_sync = first_bad - sector;
+					else {
+						bad_sectors -= (sector
+								- first_bad);
+						if (max_sync > bad_sectors)
+							max_sync = bad_sectors;
+						continue;
+					}
+				}
 				bio = r10_bio->devs[0].bio;
 				bio->bi_next = biolist;
 				biolist = bio;
 				bio->bi_private = r10_bio;
 				bio->bi_end_io = end_sync_read;
 				bio->bi_rw = READ;
-				bio->bi_sector = r10_bio->devs[j].addr +
+				from_addr = r10_bio->devs[j].addr;
+				bio->bi_sector = from_addr +
 					conf->mirrors[d].rdev->data_offset;
 				bio->bi_bdev = conf->mirrors[d].rdev->bdev;
 				atomic_inc(&conf->mirrors[d].rdev->nr_pending);
@@ -1916,26 +2499,48 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr,
 				bio->bi_private = r10_bio;
 				bio->bi_end_io = end_sync_write;
 				bio->bi_rw = WRITE;
-				bio->bi_sector = r10_bio->devs[k].addr +
+				to_addr = r10_bio->devs[k].addr;
+				bio->bi_sector = to_addr +
 					conf->mirrors[i].rdev->data_offset;
 				bio->bi_bdev = conf->mirrors[i].rdev->bdev;
 
 				r10_bio->devs[0].devnum = d;
+				r10_bio->devs[0].addr = from_addr;
 				r10_bio->devs[1].devnum = i;
+				r10_bio->devs[1].addr = to_addr;
 
 				break;
 			}
 			if (j == conf->copies) {
-				/* Cannot recover, so abort the recovery */
+				/* Cannot recover, so abort the recovery or
+				 * record a bad block */
 				put_buf(r10_bio);
 				if (rb2)
 					atomic_dec(&rb2->remaining);
 				r10_bio = rb2;
-				if (!test_and_set_bit(MD_RECOVERY_INTR,
-						      &mddev->recovery))
-					printk(KERN_INFO "md/raid10:%s: insufficient "
-					       "working devices for recovery.\n",
-					       mdname(mddev));
+				if (any_working) {
+					/* problem is that there are bad blocks
+					 * on other device(s)
+					 */
+					int k;
+					for (k = 0; k < conf->copies; k++)
+						if (r10_bio->devs[k].devnum == i)
+							break;
+					if (!rdev_set_badblocks(
+						    conf->mirrors[i].rdev,
+						    r10_bio->devs[k].addr,
+						    max_sync, 0))
+						any_working = 0;
+				}
+				if (!any_working)  {
+					if (!test_and_set_bit(MD_RECOVERY_INTR,
+							      &mddev->recovery))
+						printk(KERN_INFO "md/raid10:%s: insufficient "
+						       "working devices for recovery.\n",
+						       mdname(mddev));
+					conf->mirrors[i].recovery_disabled
+						= mddev->recovery_disabled;
+				}
 				break;
 			}
 		}
@@ -1979,12 +2584,28 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr,
 
 		for (i=0; i<conf->copies; i++) {
 			int d = r10_bio->devs[i].devnum;
+			sector_t first_bad, sector;
+			int bad_sectors;
+
 			bio = r10_bio->devs[i].bio;
 			bio->bi_end_io = NULL;
 			clear_bit(BIO_UPTODATE, &bio->bi_flags);
 			if (conf->mirrors[d].rdev == NULL ||
 			    test_bit(Faulty, &conf->mirrors[d].rdev->flags))
 				continue;
+			sector = r10_bio->devs[i].addr;
+			if (is_badblock(conf->mirrors[d].rdev,
+					sector, max_sync,
+					&first_bad, &bad_sectors)) {
+				if (first_bad > sector)
+					max_sync = first_bad - sector;
+				else {
+					bad_sectors -= (sector - first_bad);
+					if (max_sync > bad_sectors)
+						max_sync = max_sync;
+					continue;
+				}
+			}
 			atomic_inc(&conf->mirrors[d].rdev->nr_pending);
 			atomic_inc(&r10_bio->remaining);
 			bio->bi_next = biolist;
@@ -1992,7 +2613,7 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr,
 			bio->bi_private = r10_bio;
 			bio->bi_end_io = end_sync_read;
 			bio->bi_rw = READ;
-			bio->bi_sector = r10_bio->devs[i].addr +
+			bio->bi_sector = sector +
 				conf->mirrors[d].rdev->data_offset;
 			bio->bi_bdev = conf->mirrors[d].rdev->bdev;
 			count++;
@@ -2079,7 +2700,8 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr,
 	return sectors_skipped + nr_sectors;
  giveup:
 	/* There is nowhere to write, so all non-sync
-	 * drives must be failed, so try the next chunk...
+	 * drives must be failed or in resync, all drives
+	 * have a bad block, so try the next chunk...
 	 */
 	if (sector_nr + max_sync < max_sector)
 		max_sector = sector_nr + max_sync;
@@ -2249,6 +2871,7 @@ static int run(mddev_t *mddev)
 				 (conf->raid_disks / conf->near_copies));
 
 	list_for_each_entry(rdev, &mddev->disks, same_set) {
+
 		disk_idx = rdev->raid_disk;
 		if (disk_idx >= conf->raid_disks
 		    || disk_idx < 0)
@@ -2271,7 +2894,7 @@ static int run(mddev_t *mddev)
 		disk->head_position = 0;
 	}
 	/* need to check that every block has at least one working mirror */
-	if (!enough(conf)) {
+	if (!enough(conf, -1)) {
 		printk(KERN_ERR "md/raid10:%s: not enough operational mirrors.\n",
 		       mdname(mddev));
 		goto out_free_conf;

drivers/md/raid10.h

@@ -6,6 +6,11 @@ typedef struct mirror_info mirror_info_t;
 struct mirror_info {
 	mdk_rdev_t	*rdev;
 	sector_t	head_position;
+	int		recovery_disabled;	/* matches
+						 * mddev->recovery_disabled
+						 * when we shouldn't try
+						 * recovering this device.
+						 */
 };
 
 typedef struct r10bio_s r10bio_t;
@@ -113,10 +118,26 @@ struct r10bio_s {
  * level, we store IO_BLOCKED in the appropriate 'bios' pointer
  */
 #define IO_BLOCKED ((struct bio*)1)
+/* When we successfully write to a known bad-block, we need to remove the
+ * bad-block marking which must be done from process context.  So we record
+ * the success by setting devs[n].bio to IO_MADE_GOOD
+ */
+#define IO_MADE_GOOD ((struct bio *)2)
+
+#define BIO_SPECIAL(bio) ((unsigned long)bio <= 2)
 
 /* bits for r10bio.state */
 #define	R10BIO_Uptodate	0
 #define	R10BIO_IsSync	1
 #define	R10BIO_IsRecover 2
 #define	R10BIO_Degraded 3
+/* Set ReadError on bios that experience a read error
+ * so that raid10d knows what to do with them.
+ */
+#define	R10BIO_ReadError 4
+/* If a write for this request means we can clear some
+ * known-bad-block records, we set this flag.
+ */
+#define	R10BIO_MadeGood 5
+#define	R10BIO_WriteError 6
 #endif

drivers/md/raid5.c

@@ -51,6 +51,7 @@
 #include <linux/seq_file.h>
 #include <linux/cpu.h>
 #include <linux/slab.h>
+#include <linux/ratelimit.h>
 #include "md.h"
 #include "raid5.h"
 #include "raid0.h"
@@ -96,8 +97,6 @@
 #define __inline__
 #endif
 
-#define printk_rl(args...) ((void) (printk_ratelimit() && printk(args)))
-
 /*
  * We maintain a biased count of active stripes in the bottom 16 bits of
  * bi_phys_segments, and a count of processed stripes in the upper 16 bits
@@ -341,7 +340,7 @@ static void init_stripe(struct stripe_head *sh, sector_t sector, int previous)
 			       (unsigned long long)sh->sector, i, dev->toread,
 			       dev->read, dev->towrite, dev->written,
 			       test_bit(R5_LOCKED, &dev->flags));
-			BUG();
+			WARN_ON(1);
 		}
 		dev->flags = 0;
 		raid5_build_block(sh, i, previous);
@@ -527,6 +526,36 @@ static void ops_run_io(struct stripe_head *sh, struct stripe_head_state *s)
 			atomic_inc(&rdev->nr_pending);
 		rcu_read_unlock();
 
+		/* We have already checked bad blocks for reads.  Now
+		 * need to check for writes.
+		 */
+		while ((rw & WRITE) && rdev &&
+		       test_bit(WriteErrorSeen, &rdev->flags)) {
+			sector_t first_bad;
+			int bad_sectors;
+			int bad = is_badblock(rdev, sh->sector, STRIPE_SECTORS,
+					      &first_bad, &bad_sectors);
+			if (!bad)
+				break;
+
+			if (bad < 0) {
+				set_bit(BlockedBadBlocks, &rdev->flags);
+				if (!conf->mddev->external &&
+				    conf->mddev->flags) {
+					/* It is very unlikely, but we might
+					 * still need to write out the
+					 * bad block log - better give it
+					 * a chance*/
+					md_check_recovery(conf->mddev);
+				}
+				md_wait_for_blocked_rdev(rdev, conf->mddev);
+			} else {
+				/* Acknowledged bad block - skip the write */
+				rdev_dec_pending(rdev, conf->mddev);
+				rdev = NULL;
+			}
+		}
+
 		if (rdev) {
 			if (s->syncing || s->expanding || s->expanded)
 				md_sync_acct(rdev->bdev, STRIPE_SECTORS);
@@ -548,10 +577,6 @@ static void ops_run_io(struct stripe_head *sh, struct stripe_head_state *s)
 			bi->bi_io_vec[0].bv_offset = 0;
 			bi->bi_size = STRIPE_SIZE;
 			bi->bi_next = NULL;
-			if ((rw & WRITE) &&
-			    test_bit(R5_ReWrite, &sh->dev[i].flags))
-				atomic_add(STRIPE_SECTORS,
-					&rdev->corrected_errors);
 			generic_make_request(bi);
 		} else {
 			if (rw & WRITE)
@@ -1020,12 +1045,12 @@ ops_run_biodrain(struct stripe_head *sh, struct dma_async_tx_descriptor *tx)
 		if (test_and_clear_bit(R5_Wantdrain, &dev->flags)) {
 			struct bio *wbi;
 
-			spin_lock(&sh->lock);
+			spin_lock_irq(&sh->raid_conf->device_lock);
 			chosen = dev->towrite;
 			dev->towrite = NULL;
 			BUG_ON(dev->written);
 			wbi = dev->written = chosen;
-			spin_unlock(&sh->lock);
+			spin_unlock_irq(&sh->raid_conf->device_lock);
 
 			while (wbi && wbi->bi_sector <
 				dev->sector + STRIPE_SECTORS) {
@@ -1315,12 +1340,11 @@ static void raid_run_ops(struct stripe_head *sh, unsigned long ops_request)
 static int grow_one_stripe(raid5_conf_t *conf)
 {
 	struct stripe_head *sh;
-	sh = kmem_cache_alloc(conf->slab_cache, GFP_KERNEL);
+	sh = kmem_cache_zalloc(conf->slab_cache, GFP_KERNEL);
 	if (!sh)
 		return 0;
-	memset(sh, 0, sizeof(*sh) + (conf->pool_size-1)*sizeof(struct r5dev));
+
 	sh->raid_conf = conf;
-	spin_lock_init(&sh->lock);
 	#ifdef CONFIG_MULTICORE_RAID456
 	init_waitqueue_head(&sh->ops.wait_for_ops);
 	#endif
@@ -1435,14 +1459,11 @@ static int resize_stripes(raid5_conf_t *conf, int newsize)
 		return -ENOMEM;
 
 	for (i = conf->max_nr_stripes; i; i--) {
-		nsh = kmem_cache_alloc(sc, GFP_KERNEL);
+		nsh = kmem_cache_zalloc(sc, GFP_KERNEL);
 		if (!nsh)
 			break;
 
-		memset(nsh, 0, sizeof(*nsh) + (newsize-1)*sizeof(struct r5dev));
-
 		nsh->raid_conf = conf;
-		spin_lock_init(&nsh->lock);
 		#ifdef CONFIG_MULTICORE_RAID456
 		init_waitqueue_head(&nsh->ops.wait_for_ops);
 		#endif
@@ -1587,12 +1608,15 @@ static void raid5_end_read_request(struct bio * bi, int error)
 		set_bit(R5_UPTODATE, &sh->dev[i].flags);
 		if (test_bit(R5_ReadError, &sh->dev[i].flags)) {
 			rdev = conf->disks[i].rdev;
-			printk_rl(KERN_INFO "md/raid:%s: read error corrected"
-				  " (%lu sectors at %llu on %s)\n",
-				  mdname(conf->mddev), STRIPE_SECTORS,
-				  (unsigned long long)(sh->sector
-						       + rdev->data_offset),
-				  bdevname(rdev->bdev, b));
+			printk_ratelimited(
+				KERN_INFO
+				"md/raid:%s: read error corrected"
+				" (%lu sectors at %llu on %s)\n",
+				mdname(conf->mddev), STRIPE_SECTORS,
+				(unsigned long long)(sh->sector
+						     + rdev->data_offset),
+				bdevname(rdev->bdev, b));
+			atomic_add(STRIPE_SECTORS, &rdev->corrected_errors);
 			clear_bit(R5_ReadError, &sh->dev[i].flags);
 			clear_bit(R5_ReWrite, &sh->dev[i].flags);
 		}
@@ -1606,22 +1630,24 @@ static void raid5_end_read_request(struct bio * bi, int error)
 		clear_bit(R5_UPTODATE, &sh->dev[i].flags);
 		atomic_inc(&rdev->read_errors);
 		if (conf->mddev->degraded >= conf->max_degraded)
-			printk_rl(KERN_WARNING
-				  "md/raid:%s: read error not correctable "
-				  "(sector %llu on %s).\n",
-				  mdname(conf->mddev),
-				  (unsigned long long)(sh->sector
-						       + rdev->data_offset),
-				  bdn);
+			printk_ratelimited(
+				KERN_WARNING
+				"md/raid:%s: read error not correctable "
+				"(sector %llu on %s).\n",
+				mdname(conf->mddev),
+				(unsigned long long)(sh->sector
+						     + rdev->data_offset),
+				bdn);
 		else if (test_bit(R5_ReWrite, &sh->dev[i].flags))
 			/* Oh, no!!! */
-			printk_rl(KERN_WARNING
-				  "md/raid:%s: read error NOT corrected!! "
-				  "(sector %llu on %s).\n",
-				  mdname(conf->mddev),
-				  (unsigned long long)(sh->sector
-						       + rdev->data_offset),
-				  bdn);
+			printk_ratelimited(
+				KERN_WARNING
+				"md/raid:%s: read error NOT corrected!! "
+				"(sector %llu on %s).\n",
+				mdname(conf->mddev),
+				(unsigned long long)(sh->sector
+						     + rdev->data_offset),
+				bdn);
 		else if (atomic_read(&rdev->read_errors)
 			 > conf->max_nr_stripes)
 			printk(KERN_WARNING
@@ -1649,6 +1675,8 @@ static void raid5_end_write_request(struct bio *bi, int error)
 	raid5_conf_t *conf = sh->raid_conf;
 	int disks = sh->disks, i;
 	int uptodate = test_bit(BIO_UPTODATE, &bi->bi_flags);
+	sector_t first_bad;
+	int bad_sectors;
 
 	for (i=0 ; i<disks; i++)
 		if (bi == &sh->dev[i].req)
@@ -1662,8 +1690,12 @@ static void raid5_end_write_request(struct bio *bi, int error)
 		return;
 	}
 
-	if (!uptodate)
-		md_error(conf->mddev, conf->disks[i].rdev);
+	if (!uptodate) {
+		set_bit(WriteErrorSeen, &conf->disks[i].rdev->flags);
+		set_bit(R5_WriteError, &sh->dev[i].flags);
+	} else if (is_badblock(conf->disks[i].rdev, sh->sector, STRIPE_SECTORS,
+			       &first_bad, &bad_sectors))
+		set_bit(R5_MadeGood, &sh->dev[i].flags);
 
 	rdev_dec_pending(conf->disks[i].rdev, conf->mddev);
 	
@@ -1710,6 +1742,7 @@ static void error(mddev_t *mddev, mdk_rdev_t *rdev)
 		 */
 		set_bit(MD_RECOVERY_INTR, &mddev->recovery);
 	}
+	set_bit(Blocked, &rdev->flags);
 	set_bit(Faulty, &rdev->flags);
 	set_bit(MD_CHANGE_DEVS, &mddev->flags);
 	printk(KERN_ALERT
@@ -1760,7 +1793,7 @@ static sector_t raid5_compute_sector(raid5_conf_t *conf, sector_t r_sector,
 	/*
 	 * Select the parity disk based on the user selected algorithm.
 	 */
-	pd_idx = qd_idx = ~0;
+	pd_idx = qd_idx = -1;
 	switch(conf->level) {
 	case 4:
 		pd_idx = data_disks;
@@ -2143,12 +2176,11 @@ static int add_stripe_bio(struct stripe_head *sh, struct bio *bi, int dd_idx, in
 	raid5_conf_t *conf = sh->raid_conf;
 	int firstwrite=0;
 
-	pr_debug("adding bh b#%llu to stripe s#%llu\n",
+	pr_debug("adding bi b#%llu to stripe s#%llu\n",
 		(unsigned long long)bi->bi_sector,
 		(unsigned long long)sh->sector);
 
 
-	spin_lock(&sh->lock);
 	spin_lock_irq(&conf->device_lock);
 	if (forwrite) {
 		bip = &sh->dev[dd_idx].towrite;
@@ -2169,19 +2201,6 @@ static int add_stripe_bio(struct stripe_head *sh, struct bio *bi, int dd_idx, in
 		bi->bi_next = *bip;
 	*bip = bi;
 	bi->bi_phys_segments++;
-	spin_unlock_irq(&conf->device_lock);
-	spin_unlock(&sh->lock);
-
-	pr_debug("added bi b#%llu to stripe s#%llu, disk %d.\n",
-		(unsigned long long)bi->bi_sector,
-		(unsigned long long)sh->sector, dd_idx);
-
-	if (conf->mddev->bitmap && firstwrite) {
-		bitmap_startwrite(conf->mddev->bitmap, sh->sector,
-				  STRIPE_SECTORS, 0);
-		sh->bm_seq = conf->seq_flush+1;
-		set_bit(STRIPE_BIT_DELAY, &sh->state);
-	}
 
 	if (forwrite) {
 		/* check if page is covered */
@@ -2196,12 +2215,23 @@ static int add_stripe_bio(struct stripe_head *sh, struct bio *bi, int dd_idx, in
 		if (sector >= sh->dev[dd_idx].sector + STRIPE_SECTORS)
 			set_bit(R5_OVERWRITE, &sh->dev[dd_idx].flags);
 	}
+	spin_unlock_irq(&conf->device_lock);
+
+	pr_debug("added bi b#%llu to stripe s#%llu, disk %d.\n",
+		(unsigned long long)(*bip)->bi_sector,
+		(unsigned long long)sh->sector, dd_idx);
+
+	if (conf->mddev->bitmap && firstwrite) {
+		bitmap_startwrite(conf->mddev->bitmap, sh->sector,
+				  STRIPE_SECTORS, 0);
+		sh->bm_seq = conf->seq_flush+1;
+		set_bit(STRIPE_BIT_DELAY, &sh->state);
+	}
 	return 1;
 
  overlap:
 	set_bit(R5_Overlap, &sh->dev[dd_idx].flags);
 	spin_unlock_irq(&conf->device_lock);
-	spin_unlock(&sh->lock);
 	return 0;
 }
 
@@ -2238,9 +2268,18 @@ handle_failed_stripe(raid5_conf_t *conf, struct stripe_head *sh,
 			rcu_read_lock();
 			rdev = rcu_dereference(conf->disks[i].rdev);
 			if (rdev && test_bit(In_sync, &rdev->flags))
-				/* multiple read failures in one stripe */
-				md_error(conf->mddev, rdev);
+				atomic_inc(&rdev->nr_pending);
+			else
+				rdev = NULL;
 			rcu_read_unlock();
+			if (rdev) {
+				if (!rdev_set_badblocks(
+					    rdev,
+					    sh->sector,
+					    STRIPE_SECTORS, 0))
+					md_error(conf->mddev, rdev);
+				rdev_dec_pending(rdev, conf->mddev);
+			}
 		}
 		spin_lock_irq(&conf->device_lock);
 		/* fail all writes first */
@@ -2308,6 +2347,10 @@ handle_failed_stripe(raid5_conf_t *conf, struct stripe_head *sh,
 		if (bitmap_end)
 			bitmap_endwrite(conf->mddev->bitmap, sh->sector,
 					STRIPE_SECTORS, 0, 0);
+		/* If we were in the middle of a write the parity block might
+		 * still be locked - so just clear all R5_LOCKED flags
+		 */
+		clear_bit(R5_LOCKED, &sh->dev[i].flags);
 	}
 
 	if (test_and_clear_bit(STRIPE_FULL_WRITE, &sh->state))
@@ -2315,109 +2358,73 @@ handle_failed_stripe(raid5_conf_t *conf, struct stripe_head *sh,
 			md_wakeup_thread(conf->mddev->thread);
 }
 
-/* fetch_block5 - checks the given member device to see if its data needs
- * to be read or computed to satisfy a request.
- *
- * Returns 1 when no more member devices need to be checked, otherwise returns
- * 0 to tell the loop in handle_stripe_fill5 to continue
- */
-static int fetch_block5(struct stripe_head *sh, struct stripe_head_state *s,
-			int disk_idx, int disks)
-{
-	struct r5dev *dev = &sh->dev[disk_idx];
-	struct r5dev *failed_dev = &sh->dev[s->failed_num];
-
-	/* is the data in this block needed, and can we get it? */
-	if (!test_bit(R5_LOCKED, &dev->flags) &&
-	    !test_bit(R5_UPTODATE, &dev->flags) &&
-	    (dev->toread ||
-	     (dev->towrite && !test_bit(R5_OVERWRITE, &dev->flags)) ||
-	     s->syncing || s->expanding ||
-	     (s->failed &&
-	      (failed_dev->toread ||
-	       (failed_dev->towrite &&
-		!test_bit(R5_OVERWRITE, &failed_dev->flags)))))) {
-		/* We would like to get this block, possibly by computing it,
-		 * otherwise read it if the backing disk is insync
-		 */
-		if ((s->uptodate == disks - 1) &&
-		    (s->failed && disk_idx == s->failed_num)) {
-			set_bit(STRIPE_COMPUTE_RUN, &sh->state);
-			set_bit(STRIPE_OP_COMPUTE_BLK, &s->ops_request);
-			set_bit(R5_Wantcompute, &dev->flags);
-			sh->ops.target = disk_idx;
-			sh->ops.target2 = -1;
-			s->req_compute = 1;
-			/* Careful: from this point on 'uptodate' is in the eye
-			 * of raid_run_ops which services 'compute' operations
-			 * before writes. R5_Wantcompute flags a block that will
-			 * be R5_UPTODATE by the time it is needed for a
-			 * subsequent operation.
-			 */
-			s->uptodate++;
-			return 1; /* uptodate + compute == disks */
-		} else if (test_bit(R5_Insync, &dev->flags)) {
-			set_bit(R5_LOCKED, &dev->flags);
-			set_bit(R5_Wantread, &dev->flags);
-			s->locked++;
-			pr_debug("Reading block %d (sync=%d)\n", disk_idx,
-				s->syncing);
-		}
-	}
-
-	return 0;
-}
-
-/**
- * handle_stripe_fill5 - read or compute data to satisfy pending requests.
- */
-static void handle_stripe_fill5(struct stripe_head *sh,
-			struct stripe_head_state *s, int disks)
+static void
+handle_failed_sync(raid5_conf_t *conf, struct stripe_head *sh,
+		   struct stripe_head_state *s)
 {
+	int abort = 0;
 	int i;
 
-	/* look for blocks to read/compute, skip this if a compute
-	 * is already in flight, or if the stripe contents are in the
-	 * midst of changing due to a write
+	md_done_sync(conf->mddev, STRIPE_SECTORS, 0);
+	clear_bit(STRIPE_SYNCING, &sh->state);
+	s->syncing = 0;
+	/* There is nothing more to do for sync/check/repair.
+	 * For recover we need to record a bad block on all
+	 * non-sync devices, or abort the recovery
 	 */
-	if (!test_bit(STRIPE_COMPUTE_RUN, &sh->state) && !sh->check_state &&
-	    !sh->reconstruct_state)
-		for (i = disks; i--; )
-			if (fetch_block5(sh, s, i, disks))
-				break;
-	set_bit(STRIPE_HANDLE, &sh->state);
+	if (!test_bit(MD_RECOVERY_RECOVER, &conf->mddev->recovery))
+		return;
+	/* During recovery devices cannot be removed, so locking and
+	 * refcounting of rdevs is not needed
+	 */
+	for (i = 0; i < conf->raid_disks; i++) {
+		mdk_rdev_t *rdev = conf->disks[i].rdev;
+		if (!rdev
+		    || test_bit(Faulty, &rdev->flags)
+		    || test_bit(In_sync, &rdev->flags))
+			continue;
+		if (!rdev_set_badblocks(rdev, sh->sector,
+					STRIPE_SECTORS, 0))
+			abort = 1;
+	}
+	if (abort) {
+		conf->recovery_disabled = conf->mddev->recovery_disabled;
+		set_bit(MD_RECOVERY_INTR, &conf->mddev->recovery);
+	}
 }
 
-/* fetch_block6 - checks the given member device to see if its data needs
+/* fetch_block - checks the given member device to see if its data needs
  * to be read or computed to satisfy a request.
  *
  * Returns 1 when no more member devices need to be checked, otherwise returns
- * 0 to tell the loop in handle_stripe_fill6 to continue
+ * 0 to tell the loop in handle_stripe_fill to continue
  */
-static int fetch_block6(struct stripe_head *sh, struct stripe_head_state *s,
-			 struct r6_state *r6s, int disk_idx, int disks)
+static int fetch_block(struct stripe_head *sh, struct stripe_head_state *s,
+		       int disk_idx, int disks)
 {
 	struct r5dev *dev = &sh->dev[disk_idx];
-	struct r5dev *fdev[2] = { &sh->dev[r6s->failed_num[0]],
-				  &sh->dev[r6s->failed_num[1]] };
+	struct r5dev *fdev[2] = { &sh->dev[s->failed_num[0]],
+				  &sh->dev[s->failed_num[1]] };
 
+	/* is the data in this block needed, and can we get it? */
 	if (!test_bit(R5_LOCKED, &dev->flags) &&
 	    !test_bit(R5_UPTODATE, &dev->flags) &&
 	    (dev->toread ||
 	     (dev->towrite && !test_bit(R5_OVERWRITE, &dev->flags)) ||
 	     s->syncing || s->expanding ||
-	     (s->failed >= 1 &&
-	      (fdev[0]->toread || s->to_write)) ||
-	     (s->failed >= 2 &&
-	      (fdev[1]->toread || s->to_write)))) {
+	     (s->failed >= 1 && fdev[0]->toread) ||
+	     (s->failed >= 2 && fdev[1]->toread) ||
+	     (sh->raid_conf->level <= 5 && s->failed && fdev[0]->towrite &&
+	      !test_bit(R5_OVERWRITE, &fdev[0]->flags)) ||
+	     (sh->raid_conf->level == 6 && s->failed && s->to_write))) {
 		/* we would like to get this block, possibly by computing it,
 		 * otherwise read it if the backing disk is insync
 		 */
 		BUG_ON(test_bit(R5_Wantcompute, &dev->flags));
 		BUG_ON(test_bit(R5_Wantread, &dev->flags));
 		if ((s->uptodate == disks - 1) &&
-		    (s->failed && (disk_idx == r6s->failed_num[0] ||
-				   disk_idx == r6s->failed_num[1]))) {
+		    (s->failed && (disk_idx == s->failed_num[0] ||
+				   disk_idx == s->failed_num[1]))) {
 			/* have disk failed, and we're requested to fetch it;
 			 * do compute it
 			 */
@@ -2429,6 +2436,12 @@ static int fetch_block6(struct stripe_head *sh, struct stripe_head_state *s,
 			sh->ops.target = disk_idx;
 			sh->ops.target2 = -1; /* no 2nd target */
 			s->req_compute = 1;
+			/* Careful: from this point on 'uptodate' is in the eye
+			 * of raid_run_ops which services 'compute' operations
+			 * before writes. R5_Wantcompute flags a block that will
+			 * be R5_UPTODATE by the time it is needed for a
+			 * subsequent operation.
+			 */
 			s->uptodate++;
 			return 1;
 		} else if (s->uptodate == disks-2 && s->failed >= 2) {
@@ -2469,11 +2482,11 @@ static int fetch_block6(struct stripe_head *sh, struct stripe_head_state *s,
 }
 
 /**
- * handle_stripe_fill6 - read or compute data to satisfy pending requests.
+ * handle_stripe_fill - read or compute data to satisfy pending requests.
  */
-static void handle_stripe_fill6(struct stripe_head *sh,
-			struct stripe_head_state *s, struct r6_state *r6s,
-			int disks)
+static void handle_stripe_fill(struct stripe_head *sh,
+			       struct stripe_head_state *s,
+			       int disks)
 {
 	int i;
 
@@ -2484,7 +2497,7 @@ static void handle_stripe_fill6(struct stripe_head *sh,
 	if (!test_bit(STRIPE_COMPUTE_RUN, &sh->state) && !sh->check_state &&
 	    !sh->reconstruct_state)
 		for (i = disks; i--; )
-			if (fetch_block6(sh, s, r6s, i, disks))
+			if (fetch_block(sh, s, i, disks))
 				break;
 	set_bit(STRIPE_HANDLE, &sh->state);
 }
@@ -2540,11 +2553,19 @@ static void handle_stripe_clean_event(raid5_conf_t *conf,
 			md_wakeup_thread(conf->mddev->thread);
 }
 
-static void handle_stripe_dirtying5(raid5_conf_t *conf,
-		struct stripe_head *sh,	struct stripe_head_state *s, int disks)
+static void handle_stripe_dirtying(raid5_conf_t *conf,
+				   struct stripe_head *sh,
+				   struct stripe_head_state *s,
+				   int disks)
 {
 	int rmw = 0, rcw = 0, i;
-	for (i = disks; i--; ) {
+	if (conf->max_degraded == 2) {
+		/* RAID6 requires 'rcw' in current implementation
+		 * Calculate the real rcw later - for now fake it
+		 * look like rcw is cheaper
+		 */
+		rcw = 1; rmw = 2;
+	} else for (i = disks; i--; ) {
 		/* would I have to read this buffer for read_modify_write */
 		struct r5dev *dev = &sh->dev[i];
 		if ((dev->towrite || i == sh->pd_idx) &&
@@ -2591,16 +2612,19 @@ static void handle_stripe_dirtying5(raid5_conf_t *conf,
 				}
 			}
 		}
-	if (rcw <= rmw && rcw > 0)
+	if (rcw <= rmw && rcw > 0) {
 		/* want reconstruct write, but need to get some data */
+		rcw = 0;
 		for (i = disks; i--; ) {
 			struct r5dev *dev = &sh->dev[i];
 			if (!test_bit(R5_OVERWRITE, &dev->flags) &&
-			    i != sh->pd_idx &&
+			    i != sh->pd_idx && i != sh->qd_idx &&
 			    !test_bit(R5_LOCKED, &dev->flags) &&
 			    !(test_bit(R5_UPTODATE, &dev->flags) ||
-			    test_bit(R5_Wantcompute, &dev->flags)) &&
-			    test_bit(R5_Insync, &dev->flags)) {
+			      test_bit(R5_Wantcompute, &dev->flags))) {
+				rcw++;
+				if (!test_bit(R5_Insync, &dev->flags))
+					continue; /* it's a failed drive */
 				if (
 				  test_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) {
 					pr_debug("Read_old block "
@@ -2614,6 +2638,7 @@ static void handle_stripe_dirtying5(raid5_conf_t *conf,
 				}
 			}
 		}
+	}
 	/* now if nothing is locked, and if we have enough data,
 	 * we can start a write request
 	 */
@@ -2630,53 +2655,6 @@ static void handle_stripe_dirtying5(raid5_conf_t *conf,
 		schedule_reconstruction(sh, s, rcw == 0, 0);
 }
 
-static void handle_stripe_dirtying6(raid5_conf_t *conf,
-		struct stripe_head *sh,	struct stripe_head_state *s,
-		struct r6_state *r6s, int disks)
-{
-	int rcw = 0, pd_idx = sh->pd_idx, i;
-	int qd_idx = sh->qd_idx;
-
-	set_bit(STRIPE_HANDLE, &sh->state);
-	for (i = disks; i--; ) {
-		struct r5dev *dev = &sh->dev[i];
-		/* check if we haven't enough data */
-		if (!test_bit(R5_OVERWRITE, &dev->flags) &&
-		    i != pd_idx && i != qd_idx &&
-		    !test_bit(R5_LOCKED, &dev->flags) &&
-		    !(test_bit(R5_UPTODATE, &dev->flags) ||
-		      test_bit(R5_Wantcompute, &dev->flags))) {
-			rcw++;
-			if (!test_bit(R5_Insync, &dev->flags))
-				continue; /* it's a failed drive */
-
-			if (
-			  test_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) {
-				pr_debug("Read_old stripe %llu "
-					"block %d for Reconstruct\n",
-				     (unsigned long long)sh->sector, i);
-				set_bit(R5_LOCKED, &dev->flags);
-				set_bit(R5_Wantread, &dev->flags);
-				s->locked++;
-			} else {
-				pr_debug("Request delayed stripe %llu "
-					"block %d for Reconstruct\n",
-				     (unsigned long long)sh->sector, i);
-				set_bit(STRIPE_DELAYED, &sh->state);
-				set_bit(STRIPE_HANDLE, &sh->state);
-			}
-		}
-	}
-	/* now if nothing is locked, and if we have enough data, we can start a
-	 * write request
-	 */
-	if ((s->req_compute || !test_bit(STRIPE_COMPUTE_RUN, &sh->state)) &&
-	    s->locked == 0 && rcw == 0 &&
-	    !test_bit(STRIPE_BIT_DELAY, &sh->state)) {
-		schedule_reconstruction(sh, s, 1, 0);
-	}
-}
-
 static void handle_parity_checks5(raid5_conf_t *conf, struct stripe_head *sh,
 				struct stripe_head_state *s, int disks)
 {
@@ -2695,7 +2673,7 @@ static void handle_parity_checks5(raid5_conf_t *conf, struct stripe_head *sh,
 			s->uptodate--;
 			break;
 		}
-		dev = &sh->dev[s->failed_num];
+		dev = &sh->dev[s->failed_num[0]];
 		/* fall through */
 	case check_state_compute_result:
 		sh->check_state = check_state_idle;
@@ -2767,7 +2745,7 @@ static void handle_parity_checks5(raid5_conf_t *conf, struct stripe_head *sh,
 
 static void handle_parity_checks6(raid5_conf_t *conf, struct stripe_head *sh,
 				  struct stripe_head_state *s,
-				  struct r6_state *r6s, int disks)
+				  int disks)
 {
 	int pd_idx = sh->pd_idx;
 	int qd_idx = sh->qd_idx;
@@ -2786,14 +2764,14 @@ static void handle_parity_checks6(raid5_conf_t *conf, struct stripe_head *sh,
 	switch (sh->check_state) {
 	case check_state_idle:
 		/* start a new check operation if there are < 2 failures */
-		if (s->failed == r6s->q_failed) {
+		if (s->failed == s->q_failed) {
 			/* The only possible failed device holds Q, so it
 			 * makes sense to check P (If anything else were failed,
 			 * we would have used P to recreate it).
 			 */
 			sh->check_state = check_state_run;
 		}
-		if (!r6s->q_failed && s->failed < 2) {
+		if (!s->q_failed && s->failed < 2) {
 			/* Q is not failed, and we didn't use it to generate
 			 * anything, so it makes sense to check it
 			 */
@@ -2835,13 +2813,13 @@ static void handle_parity_checks6(raid5_conf_t *conf, struct stripe_head *sh,
 		 */
 		BUG_ON(s->uptodate < disks - 1); /* We don't need Q to recover */
 		if (s->failed == 2) {
-			dev = &sh->dev[r6s->failed_num[1]];
+			dev = &sh->dev[s->failed_num[1]];
 			s->locked++;
 			set_bit(R5_LOCKED, &dev->flags);
 			set_bit(R5_Wantwrite, &dev->flags);
 		}
 		if (s->failed >= 1) {
-			dev = &sh->dev[r6s->failed_num[0]];
+			dev = &sh->dev[s->failed_num[0]];
 			s->locked++;
 			set_bit(R5_LOCKED, &dev->flags);
 			set_bit(R5_Wantwrite, &dev->flags);
@@ -2928,8 +2906,7 @@ static void handle_parity_checks6(raid5_conf_t *conf, struct stripe_head *sh,
 	}
 }
 
-static void handle_stripe_expansion(raid5_conf_t *conf, struct stripe_head *sh,
-				struct r6_state *r6s)
+static void handle_stripe_expansion(raid5_conf_t *conf, struct stripe_head *sh)
 {
 	int i;
 
@@ -2971,7 +2948,7 @@ static void handle_stripe_expansion(raid5_conf_t *conf, struct stripe_head *sh,
 			set_bit(R5_UPTODATE, &sh2->dev[dd_idx].flags);
 			for (j = 0; j < conf->raid_disks; j++)
 				if (j != sh2->pd_idx &&
-				    (!r6s || j != sh2->qd_idx) &&
+				    j != sh2->qd_idx &&
 				    !test_bit(R5_Expanded, &sh2->dev[j].flags))
 					break;
 			if (j == conf->raid_disks) {
@@ -3006,43 +2983,35 @@ static void handle_stripe_expansion(raid5_conf_t *conf, struct stripe_head *sh,
  *
  */
 
-static void handle_stripe5(struct stripe_head *sh)
+static void analyse_stripe(struct stripe_head *sh, struct stripe_head_state *s)
 {
 	raid5_conf_t *conf = sh->raid_conf;
-	int disks = sh->disks, i;
-	struct bio *return_bi = NULL;
-	struct stripe_head_state s;
+	int disks = sh->disks;
 	struct r5dev *dev;
-	mdk_rdev_t *blocked_rdev = NULL;
-	int prexor;
-	int dec_preread_active = 0;
+	int i;
 
-	memset(&s, 0, sizeof(s));
-	pr_debug("handling stripe %llu, state=%#lx cnt=%d, pd_idx=%d check:%d "
-		 "reconstruct:%d\n", (unsigned long long)sh->sector, sh->state,
-		 atomic_read(&sh->count), sh->pd_idx, sh->check_state,
-		 sh->reconstruct_state);
+	memset(s, 0, sizeof(*s));
 
-	spin_lock(&sh->lock);
-	clear_bit(STRIPE_HANDLE, &sh->state);
-	clear_bit(STRIPE_DELAYED, &sh->state);
-
-	s.syncing = test_bit(STRIPE_SYNCING, &sh->state);
-	s.expanding = test_bit(STRIPE_EXPAND_SOURCE, &sh->state);
-	s.expanded = test_bit(STRIPE_EXPAND_READY, &sh->state);
+	s->syncing = test_bit(STRIPE_SYNCING, &sh->state);
+	s->expanding = test_bit(STRIPE_EXPAND_SOURCE, &sh->state);
+	s->expanded = test_bit(STRIPE_EXPAND_READY, &sh->state);
+	s->failed_num[0] = -1;
+	s->failed_num[1] = -1;
 
 	/* Now to look around and see what can be done */
 	rcu_read_lock();
+	spin_lock_irq(&conf->device_lock);
 	for (i=disks; i--; ) {
 		mdk_rdev_t *rdev;
+		sector_t first_bad;
+		int bad_sectors;
+		int is_bad = 0;
 
 		dev = &sh->dev[i];
 
-		pr_debug("check %d: state 0x%lx toread %p read %p write %p "
-			"written %p\n",	i, dev->flags, dev->toread, dev->read,
-			dev->towrite, dev->written);
-
-		/* maybe we can request a biofill operation
+		pr_debug("check %d: state 0x%lx read %p write %p written %p\n",
+			i, dev->flags, dev->toread, dev->towrite, dev->written);
+		/* maybe we can reply to a read
 		 *
 		 * new wantfill requests are only permitted while
 		 * ops_complete_biofill is guaranteed to be inactive
@@ -3052,37 +3021,74 @@ static void handle_stripe5(struct stripe_head *sh)
 			set_bit(R5_Wantfill, &dev->flags);
 
 		/* now count some things */
-		if (test_bit(R5_LOCKED, &dev->flags)) s.locked++;
-		if (test_bit(R5_UPTODATE, &dev->flags)) s.uptodate++;
-		if (test_bit(R5_Wantcompute, &dev->flags)) s.compute++;
+		if (test_bit(R5_LOCKED, &dev->flags))
+			s->locked++;
+		if (test_bit(R5_UPTODATE, &dev->flags))
+			s->uptodate++;
+		if (test_bit(R5_Wantcompute, &dev->flags)) {
+			s->compute++;
+			BUG_ON(s->compute > 2);
+		}
 
 		if (test_bit(R5_Wantfill, &dev->flags))
-			s.to_fill++;
+			s->to_fill++;
 		else if (dev->toread)
-			s.to_read++;
+			s->to_read++;
 		if (dev->towrite) {
-			s.to_write++;
+			s->to_write++;
 			if (!test_bit(R5_OVERWRITE, &dev->flags))
-				s.non_overwrite++;
+				s->non_overwrite++;
 		}
 		if (dev->written)
-			s.written++;
+			s->written++;
 		rdev = rcu_dereference(conf->disks[i].rdev);
-		if (blocked_rdev == NULL &&
-		    rdev && unlikely(test_bit(Blocked, &rdev->flags))) {
-			blocked_rdev = rdev;
-			atomic_inc(&rdev->nr_pending);
+		if (rdev) {
+			is_bad = is_badblock(rdev, sh->sector, STRIPE_SECTORS,
+					     &first_bad, &bad_sectors);
+			if (s->blocked_rdev == NULL
+			    && (test_bit(Blocked, &rdev->flags)
+				|| is_bad < 0)) {
+				if (is_bad < 0)
+					set_bit(BlockedBadBlocks,
+						&rdev->flags);
+				s->blocked_rdev = rdev;
+				atomic_inc(&rdev->nr_pending);
+			}
 		}
 		clear_bit(R5_Insync, &dev->flags);
 		if (!rdev)
 			/* Not in-sync */;
-		else if (test_bit(In_sync, &rdev->flags))
+		else if (is_bad) {
+			/* also not in-sync */
+			if (!test_bit(WriteErrorSeen, &rdev->flags)) {
+				/* treat as in-sync, but with a read error
+				 * which we can now try to correct
+				 */
+				set_bit(R5_Insync, &dev->flags);
+				set_bit(R5_ReadError, &dev->flags);
+			}
+		} else if (test_bit(In_sync, &rdev->flags))
 			set_bit(R5_Insync, &dev->flags);
 		else {
-			/* could be in-sync depending on recovery/reshape status */
+			/* in sync if before recovery_offset */
 			if (sh->sector + STRIPE_SECTORS <= rdev->recovery_offset)
 				set_bit(R5_Insync, &dev->flags);
 		}
+		if (test_bit(R5_WriteError, &dev->flags)) {
+			clear_bit(R5_Insync, &dev->flags);
+			if (!test_bit(Faulty, &rdev->flags)) {
+				s->handle_bad_blocks = 1;
+				atomic_inc(&rdev->nr_pending);
+			} else
+				clear_bit(R5_WriteError, &dev->flags);
+		}
+		if (test_bit(R5_MadeGood, &dev->flags)) {
+			if (!test_bit(Faulty, &rdev->flags)) {
+				s->handle_bad_blocks = 1;
+				atomic_inc(&rdev->nr_pending);
+			} else
+				clear_bit(R5_MadeGood, &dev->flags);
+		}
 		if (!test_bit(R5_Insync, &dev->flags)) {
 			/* The ReadError flag will just be confusing now */
 			clear_bit(R5_ReadError, &dev->flags);
@@ -3091,313 +3097,60 @@ static void handle_stripe5(struct stripe_head *sh)
 		if (test_bit(R5_ReadError, &dev->flags))
 			clear_bit(R5_Insync, &dev->flags);
 		if (!test_bit(R5_Insync, &dev->flags)) {
-			s.failed++;
-			s.failed_num = i;
+			if (s->failed < 2)
+				s->failed_num[s->failed] = i;
+			s->failed++;
 		}
 	}
+	spin_unlock_irq(&conf->device_lock);
 	rcu_read_unlock();
-
-	if (unlikely(blocked_rdev)) {
-		if (s.syncing || s.expanding || s.expanded ||
-		    s.to_write || s.written) {
-			set_bit(STRIPE_HANDLE, &sh->state);
-			goto unlock;
-		}
-		/* There is nothing for the blocked_rdev to block */
-		rdev_dec_pending(blocked_rdev, conf->mddev);
-		blocked_rdev = NULL;
-	}
-
-	if (s.to_fill && !test_bit(STRIPE_BIOFILL_RUN, &sh->state)) {
-		set_bit(STRIPE_OP_BIOFILL, &s.ops_request);
-		set_bit(STRIPE_BIOFILL_RUN, &sh->state);
-	}
-
-	pr_debug("locked=%d uptodate=%d to_read=%d"
-		" to_write=%d failed=%d failed_num=%d\n",
-		s.locked, s.uptodate, s.to_read, s.to_write,
-		s.failed, s.failed_num);
-	/* check if the array has lost two devices and, if so, some requests might
-	 * need to be failed
-	 */
-	if (s.failed > 1 && s.to_read+s.to_write+s.written)
-		handle_failed_stripe(conf, sh, &s, disks, &return_bi);
-	if (s.failed > 1 && s.syncing) {
-		md_done_sync(conf->mddev, STRIPE_SECTORS,0);
-		clear_bit(STRIPE_SYNCING, &sh->state);
-		s.syncing = 0;
-	}
-
-	/* might be able to return some write requests if the parity block
-	 * is safe, or on a failed drive
-	 */
-	dev = &sh->dev[sh->pd_idx];
-	if ( s.written &&
-	     ((test_bit(R5_Insync, &dev->flags) &&
-	       !test_bit(R5_LOCKED, &dev->flags) &&
-	       test_bit(R5_UPTODATE, &dev->flags)) ||
-	       (s.failed == 1 && s.failed_num == sh->pd_idx)))
-		handle_stripe_clean_event(conf, sh, disks, &return_bi);
-
-	/* Now we might consider reading some blocks, either to check/generate
-	 * parity, or to satisfy requests
-	 * or to load a block that is being partially written.
-	 */
-	if (s.to_read || s.non_overwrite ||
-	    (s.syncing && (s.uptodate + s.compute < disks)) || s.expanding)
-		handle_stripe_fill5(sh, &s, disks);
-
-	/* Now we check to see if any write operations have recently
-	 * completed
-	 */
-	prexor = 0;
-	if (sh->reconstruct_state == reconstruct_state_prexor_drain_result)
-		prexor = 1;
-	if (sh->reconstruct_state == reconstruct_state_drain_result ||
-	    sh->reconstruct_state == reconstruct_state_prexor_drain_result) {
-		sh->reconstruct_state = reconstruct_state_idle;
-
-		/* All the 'written' buffers and the parity block are ready to
-		 * be written back to disk
-		 */
-		BUG_ON(!test_bit(R5_UPTODATE, &sh->dev[sh->pd_idx].flags));
-		for (i = disks; i--; ) {
-			dev = &sh->dev[i];
-			if (test_bit(R5_LOCKED, &dev->flags) &&
-				(i == sh->pd_idx || dev->written)) {
-				pr_debug("Writing block %d\n", i);
-				set_bit(R5_Wantwrite, &dev->flags);
-				if (prexor)
-					continue;
-				if (!test_bit(R5_Insync, &dev->flags) ||
-				    (i == sh->pd_idx && s.failed == 0))
-					set_bit(STRIPE_INSYNC, &sh->state);
-			}
-		}
-		if (test_and_clear_bit(STRIPE_PREREAD_ACTIVE, &sh->state))
-			dec_preread_active = 1;
-	}
-
-	/* Now to consider new write requests 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.
-	 */
-	if (s.to_write && !sh->reconstruct_state && !sh->check_state)
-		handle_stripe_dirtying5(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
-	 * data is available.  The parity check is held off while parity
-	 * dependent operations are in flight.
-	 */
-	if (sh->check_state ||
-	    (s.syncing && s.locked == 0 &&
-	     !test_bit(STRIPE_COMPUTE_RUN, &sh->state) &&
-	     !test_bit(STRIPE_INSYNC, &sh->state)))
-		handle_parity_checks5(conf, sh, &s, disks);
-
-	if (s.syncing && s.locked == 0 && test_bit(STRIPE_INSYNC, &sh->state)) {
-		md_done_sync(conf->mddev, STRIPE_SECTORS,1);
-		clear_bit(STRIPE_SYNCING, &sh->state);
-	}
-
-	/* If the failed drive is just a ReadError, then we might need to progress
-	 * the repair/check process
-	 */
-	if (s.failed == 1 && !conf->mddev->ro &&
-	    test_bit(R5_ReadError, &sh->dev[s.failed_num].flags)
-	    && !test_bit(R5_LOCKED, &sh->dev[s.failed_num].flags)
-	    && test_bit(R5_UPTODATE, &sh->dev[s.failed_num].flags)
-		) {
-		dev = &sh->dev[s.failed_num];
-		if (!test_bit(R5_ReWrite, &dev->flags)) {
-			set_bit(R5_Wantwrite, &dev->flags);
-			set_bit(R5_ReWrite, &dev->flags);
-			set_bit(R5_LOCKED, &dev->flags);
-			s.locked++;
-		} else {
-			/* let's read it back */
-			set_bit(R5_Wantread, &dev->flags);
-			set_bit(R5_LOCKED, &dev->flags);
-			s.locked++;
-		}
-	}
-
-	/* Finish reconstruct operations initiated by the expansion process */
-	if (sh->reconstruct_state == reconstruct_state_result) {
-		struct stripe_head *sh2
-			= get_active_stripe(conf, sh->sector, 1, 1, 1);
-		if (sh2 && test_bit(STRIPE_EXPAND_SOURCE, &sh2->state)) {
-			/* sh cannot be written until sh2 has been read.
-			 * so arrange for sh to be delayed a little
-			 */
-			set_bit(STRIPE_DELAYED, &sh->state);
-			set_bit(STRIPE_HANDLE, &sh->state);
-			if (!test_and_set_bit(STRIPE_PREREAD_ACTIVE,
-					      &sh2->state))
-				atomic_inc(&conf->preread_active_stripes);
-			release_stripe(sh2);
-			goto unlock;
-		}
-		if (sh2)
-			release_stripe(sh2);
-
-		sh->reconstruct_state = reconstruct_state_idle;
-		clear_bit(STRIPE_EXPANDING, &sh->state);
-		for (i = conf->raid_disks; i--; ) {
-			set_bit(R5_Wantwrite, &sh->dev[i].flags);
-			set_bit(R5_LOCKED, &sh->dev[i].flags);
-			s.locked++;
-		}
-	}
-
-	if (s.expanded && test_bit(STRIPE_EXPANDING, &sh->state) &&
-	    !sh->reconstruct_state) {
-		/* Need to write out all blocks after computing parity */
-		sh->disks = conf->raid_disks;
-		stripe_set_idx(sh->sector, conf, 0, sh);
-		schedule_reconstruction(sh, &s, 1, 1);
-	} else if (s.expanded && !sh->reconstruct_state && s.locked == 0) {
-		clear_bit(STRIPE_EXPAND_READY, &sh->state);
-		atomic_dec(&conf->reshape_stripes);
-		wake_up(&conf->wait_for_overlap);
-		md_done_sync(conf->mddev, STRIPE_SECTORS, 1);
-	}
-
-	if (s.expanding && s.locked == 0 &&
-	    !test_bit(STRIPE_COMPUTE_RUN, &sh->state))
-		handle_stripe_expansion(conf, sh, NULL);
-
- unlock:
-	spin_unlock(&sh->lock);
-
-	/* wait for this device to become unblocked */
-	if (unlikely(blocked_rdev))
-		md_wait_for_blocked_rdev(blocked_rdev, conf->mddev);
-
-	if (s.ops_request)
-		raid_run_ops(sh, s.ops_request);
-
-	ops_run_io(sh, &s);
-
-	if (dec_preread_active) {
-		/* We delay this until after ops_run_io so that if make_request
-		 * is waiting on a flush, it won't continue until the writes
-		 * have actually been submitted.
-		 */
-		atomic_dec(&conf->preread_active_stripes);
-		if (atomic_read(&conf->preread_active_stripes) <
-		    IO_THRESHOLD)
-			md_wakeup_thread(conf->mddev->thread);
-	}
-	return_io(return_bi);
 }
 
-static void handle_stripe6(struct stripe_head *sh)
+static void handle_stripe(struct stripe_head *sh)
 {
+	struct stripe_head_state s;
 	raid5_conf_t *conf = sh->raid_conf;
+	int i;
+	int prexor;
 	int disks = sh->disks;
-	struct bio *return_bi = NULL;
-	int i, pd_idx = sh->pd_idx, qd_idx = sh->qd_idx;
-	struct stripe_head_state s;
-	struct r6_state r6s;
-	struct r5dev *dev, *pdev, *qdev;
-	mdk_rdev_t *blocked_rdev = NULL;
-	int dec_preread_active = 0;
+	struct r5dev *pdev, *qdev;
+
+	clear_bit(STRIPE_HANDLE, &sh->state);
+	if (test_and_set_bit(STRIPE_ACTIVE, &sh->state)) {
+		/* already being handled, ensure it gets handled
+		 * again when current action finishes */
+		set_bit(STRIPE_HANDLE, &sh->state);
+		return;
+	}
+
+	if (test_and_clear_bit(STRIPE_SYNC_REQUESTED, &sh->state)) {
+		set_bit(STRIPE_SYNCING, &sh->state);
+		clear_bit(STRIPE_INSYNC, &sh->state);
+	}
+	clear_bit(STRIPE_DELAYED, &sh->state);
 
 	pr_debug("handling stripe %llu, state=%#lx cnt=%d, "
 		"pd_idx=%d, qd_idx=%d\n, check:%d, reconstruct:%d\n",
 	       (unsigned long long)sh->sector, sh->state,
-	       atomic_read(&sh->count), pd_idx, qd_idx,
+	       atomic_read(&sh->count), sh->pd_idx, sh->qd_idx,
 	       sh->check_state, sh->reconstruct_state);
-	memset(&s, 0, sizeof(s));
-
-	spin_lock(&sh->lock);
-	clear_bit(STRIPE_HANDLE, &sh->state);
-	clear_bit(STRIPE_DELAYED, &sh->state);
-
-	s.syncing = test_bit(STRIPE_SYNCING, &sh->state);
-	s.expanding = test_bit(STRIPE_EXPAND_SOURCE, &sh->state);
-	s.expanded = test_bit(STRIPE_EXPAND_READY, &sh->state);
-	/* Now to look around and see what can be done */
-
-	rcu_read_lock();
-	for (i=disks; i--; ) {
-		mdk_rdev_t *rdev;
-		dev = &sh->dev[i];
 
-		pr_debug("check %d: state 0x%lx read %p write %p written %p\n",
-			i, dev->flags, dev->toread, dev->towrite, dev->written);
-		/* maybe we can reply to a read
-		 *
-		 * new wantfill requests are only permitted while
-		 * ops_complete_biofill is guaranteed to be inactive
-		 */
-		if (test_bit(R5_UPTODATE, &dev->flags) && dev->toread &&
-		    !test_bit(STRIPE_BIOFILL_RUN, &sh->state))
-			set_bit(R5_Wantfill, &dev->flags);
+	analyse_stripe(sh, &s);
 
-		/* now count some things */
-		if (test_bit(R5_LOCKED, &dev->flags)) s.locked++;
-		if (test_bit(R5_UPTODATE, &dev->flags)) s.uptodate++;
-		if (test_bit(R5_Wantcompute, &dev->flags)) {
-			s.compute++;
-			BUG_ON(s.compute > 2);
-		}
-
-		if (test_bit(R5_Wantfill, &dev->flags)) {
-			s.to_fill++;
-		} else if (dev->toread)
-			s.to_read++;
-		if (dev->towrite) {
-			s.to_write++;
-			if (!test_bit(R5_OVERWRITE, &dev->flags))
-				s.non_overwrite++;
-		}
-		if (dev->written)
-			s.written++;
-		rdev = rcu_dereference(conf->disks[i].rdev);
-		if (blocked_rdev == NULL &&
-		    rdev && unlikely(test_bit(Blocked, &rdev->flags))) {
-			blocked_rdev = rdev;
-			atomic_inc(&rdev->nr_pending);
-		}
-		clear_bit(R5_Insync, &dev->flags);
-		if (!rdev)
-			/* Not in-sync */;
-		else if (test_bit(In_sync, &rdev->flags))
-			set_bit(R5_Insync, &dev->flags);
-		else {
-			/* in sync if before recovery_offset */
-			if (sh->sector + STRIPE_SECTORS <= rdev->recovery_offset)
-				set_bit(R5_Insync, &dev->flags);
-		}
-		if (!test_bit(R5_Insync, &dev->flags)) {
-			/* The ReadError flag will just be confusing now */
-			clear_bit(R5_ReadError, &dev->flags);
-			clear_bit(R5_ReWrite, &dev->flags);
-		}
-		if (test_bit(R5_ReadError, &dev->flags))
-			clear_bit(R5_Insync, &dev->flags);
-		if (!test_bit(R5_Insync, &dev->flags)) {
-			if (s.failed < 2)
-				r6s.failed_num[s.failed] = i;
-			s.failed++;
-		}
+	if (s.handle_bad_blocks) {
+		set_bit(STRIPE_HANDLE, &sh->state);
+		goto finish;
 	}
-	rcu_read_unlock();
 
-	if (unlikely(blocked_rdev)) {
+	if (unlikely(s.blocked_rdev)) {
 		if (s.syncing || s.expanding || s.expanded ||
 		    s.to_write || s.written) {
 			set_bit(STRIPE_HANDLE, &sh->state);
-			goto unlock;
+			goto finish;
 		}
 		/* There is nothing for the blocked_rdev to block */
-		rdev_dec_pending(blocked_rdev, conf->mddev);
-		blocked_rdev = NULL;
+		rdev_dec_pending(s.blocked_rdev, conf->mddev);
+		s.blocked_rdev = NULL;
 	}
 
 	if (s.to_fill && !test_bit(STRIPE_BIOFILL_RUN, &sh->state)) {
@@ -3408,83 +3161,88 @@ static void handle_stripe6(struct stripe_head *sh)
 	pr_debug("locked=%d uptodate=%d to_read=%d"
 	       " to_write=%d failed=%d failed_num=%d,%d\n",
 	       s.locked, s.uptodate, s.to_read, s.to_write, s.failed,
-	       r6s.failed_num[0], r6s.failed_num[1]);
-	/* check if the array has lost >2 devices and, if so, some requests
-	 * might need to be failed
+	       s.failed_num[0], s.failed_num[1]);
+	/* check if the array has lost more than max_degraded devices and,
+	 * if so, some requests might need to be failed.
 	 */
-	if (s.failed > 2 && s.to_read+s.to_write+s.written)
-		handle_failed_stripe(conf, sh, &s, disks, &return_bi);
-	if (s.failed > 2 && s.syncing) {
-		md_done_sync(conf->mddev, STRIPE_SECTORS,0);
-		clear_bit(STRIPE_SYNCING, &sh->state);
-		s.syncing = 0;
-	}
+	if (s.failed > conf->max_degraded && s.to_read+s.to_write+s.written)
+		handle_failed_stripe(conf, sh, &s, disks, &s.return_bi);
+	if (s.failed > conf->max_degraded && s.syncing)
+		handle_failed_sync(conf, sh, &s);
 
 	/*
 	 * might be able to return some write requests if the parity blocks
 	 * are safe, or on a failed drive
 	 */
-	pdev = &sh->dev[pd_idx];
-	r6s.p_failed = (s.failed >= 1 && r6s.failed_num[0] == pd_idx)
-		|| (s.failed >= 2 && r6s.failed_num[1] == pd_idx);
-	qdev = &sh->dev[qd_idx];
-	r6s.q_failed = (s.failed >= 1 && r6s.failed_num[0] == qd_idx)
-		|| (s.failed >= 2 && r6s.failed_num[1] == qd_idx);
-
-	if ( s.written &&
-	     ( r6s.p_failed || ((test_bit(R5_Insync, &pdev->flags)
+	pdev = &sh->dev[sh->pd_idx];
+	s.p_failed = (s.failed >= 1 && s.failed_num[0] == sh->pd_idx)
+		|| (s.failed >= 2 && s.failed_num[1] == sh->pd_idx);
+	qdev = &sh->dev[sh->qd_idx];
+	s.q_failed = (s.failed >= 1 && s.failed_num[0] == sh->qd_idx)
+		|| (s.failed >= 2 && s.failed_num[1] == sh->qd_idx)
+		|| conf->level < 6;
+
+	if (s.written &&
+	    (s.p_failed || ((test_bit(R5_Insync, &pdev->flags)
 			     && !test_bit(R5_LOCKED, &pdev->flags)
 			     && test_bit(R5_UPTODATE, &pdev->flags)))) &&
-	     ( r6s.q_failed || ((test_bit(R5_Insync, &qdev->flags)
+	    (s.q_failed || ((test_bit(R5_Insync, &qdev->flags)
 			     && !test_bit(R5_LOCKED, &qdev->flags)
 			     && test_bit(R5_UPTODATE, &qdev->flags)))))
-		handle_stripe_clean_event(conf, sh, disks, &return_bi);
+		handle_stripe_clean_event(conf, sh, disks, &s.return_bi);
 
 	/* Now we might consider reading some blocks, either to check/generate
 	 * parity, or to satisfy requests
 	 * or to load a block that is being partially written.
 	 */
-	if (s.to_read || s.non_overwrite || (s.to_write && s.failed) ||
-	    (s.syncing && (s.uptodate + s.compute < disks)) || s.expanding)
-		handle_stripe_fill6(sh, &s, &r6s, disks);
+	if (s.to_read || s.non_overwrite
+	    || (conf->level == 6 && s.to_write && s.failed)
+	    || (s.syncing && (s.uptodate + s.compute < disks)) || s.expanding)
+		handle_stripe_fill(sh, &s, disks);
 
 	/* Now we check to see if any write operations have recently
 	 * completed
 	 */
-	if (sh->reconstruct_state == reconstruct_state_drain_result) {
-
+	prexor = 0;
+	if (sh->reconstruct_state == reconstruct_state_prexor_drain_result)
+		prexor = 1;
+	if (sh->reconstruct_state == reconstruct_state_drain_result ||
+	    sh->reconstruct_state == reconstruct_state_prexor_drain_result) {
 		sh->reconstruct_state = reconstruct_state_idle;
-		/* All the 'written' buffers and the parity blocks are ready to
+
+		/* All the 'written' buffers and the parity block are ready to
 		 * be written back to disk
 		 */
 		BUG_ON(!test_bit(R5_UPTODATE, &sh->dev[sh->pd_idx].flags));
-		BUG_ON(!test_bit(R5_UPTODATE, &sh->dev[qd_idx].flags));
+		BUG_ON(sh->qd_idx >= 0 &&
+		       !test_bit(R5_UPTODATE, &sh->dev[sh->qd_idx].flags));
 		for (i = disks; i--; ) {
-			dev = &sh->dev[i];
+			struct r5dev *dev = &sh->dev[i];
 			if (test_bit(R5_LOCKED, &dev->flags) &&
-			    (i == sh->pd_idx || i == qd_idx ||
-			     dev->written)) {
+				(i == sh->pd_idx || i == sh->qd_idx ||
+				 dev->written)) {
 				pr_debug("Writing block %d\n", i);
-				BUG_ON(!test_bit(R5_UPTODATE, &dev->flags));
 				set_bit(R5_Wantwrite, &dev->flags);
+				if (prexor)
+					continue;
 				if (!test_bit(R5_Insync, &dev->flags) ||
-				    ((i == sh->pd_idx || i == qd_idx) &&
-				      s.failed == 0))
+				    ((i == sh->pd_idx || i == sh->qd_idx)  &&
+				     s.failed == 0))
 					set_bit(STRIPE_INSYNC, &sh->state);
 			}
 		}
 		if (test_and_clear_bit(STRIPE_PREREAD_ACTIVE, &sh->state))
-			dec_preread_active = 1;
+			s.dec_preread_active = 1;
 	}
 
 	/* Now to consider new write requests and what else, if anything
 	 * should be read.  We do not handle new writes when:
-	 * 1/ A 'write' operation (copy+gen_syndrome) is already in flight.
+	 * 1/ A 'write' operation (copy+xor) is already in flight.
 	 * 2/ A 'check' operation is in flight, as it may clobber the parity
 	 *    block.
 	 */
 	if (s.to_write && !sh->reconstruct_state && !sh->check_state)
-		handle_stripe_dirtying6(conf, sh, &s, &r6s, disks);
+		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
@@ -3494,20 +3252,24 @@ static void handle_stripe6(struct stripe_head *sh)
 	if (sh->check_state ||
 	    (s.syncing && s.locked == 0 &&
 	     !test_bit(STRIPE_COMPUTE_RUN, &sh->state) &&
-	     !test_bit(STRIPE_INSYNC, &sh->state)))
-		handle_parity_checks6(conf, sh, &s, &r6s, disks);
+	     !test_bit(STRIPE_INSYNC, &sh->state))) {
+		if (conf->level == 6)
+			handle_parity_checks6(conf, sh, &s, disks);
+		else
+			handle_parity_checks5(conf, sh, &s, disks);
+	}
 
 	if (s.syncing && s.locked == 0 && test_bit(STRIPE_INSYNC, &sh->state)) {
-		md_done_sync(conf->mddev, STRIPE_SECTORS,1);
+		md_done_sync(conf->mddev, STRIPE_SECTORS, 1);
 		clear_bit(STRIPE_SYNCING, &sh->state);
 	}
 
 	/* If the failed drives are just a ReadError, then we might need
 	 * to progress the repair/check process
 	 */
-	if (s.failed <= 2 && !conf->mddev->ro)
+	if (s.failed <= conf->max_degraded && !conf->mddev->ro)
 		for (i = 0; i < s.failed; i++) {
-			dev = &sh->dev[r6s.failed_num[i]];
+			struct r5dev *dev = &sh->dev[s.failed_num[i]];
 			if (test_bit(R5_ReadError, &dev->flags)
 			    && !test_bit(R5_LOCKED, &dev->flags)
 			    && test_bit(R5_UPTODATE, &dev->flags)
@@ -3526,8 +3288,26 @@ static void handle_stripe6(struct stripe_head *sh)
 			}
 		}
 
+
 	/* Finish reconstruct operations initiated by the expansion process */
 	if (sh->reconstruct_state == reconstruct_state_result) {
+		struct stripe_head *sh_src
+			= get_active_stripe(conf, sh->sector, 1, 1, 1);
+		if (sh_src && test_bit(STRIPE_EXPAND_SOURCE, &sh_src->state)) {
+			/* sh cannot be written until sh_src has been read.
+			 * so arrange for sh to be delayed a little
+			 */
+			set_bit(STRIPE_DELAYED, &sh->state);
+			set_bit(STRIPE_HANDLE, &sh->state);
+			if (!test_and_set_bit(STRIPE_PREREAD_ACTIVE,
+					      &sh_src->state))
+				atomic_inc(&conf->preread_active_stripes);
+			release_stripe(sh_src);
+			goto finish;
+		}
+		if (sh_src)
+			release_stripe(sh_src);
+
 		sh->reconstruct_state = reconstruct_state_idle;
 		clear_bit(STRIPE_EXPANDING, &sh->state);
 		for (i = conf->raid_disks; i--; ) {
@@ -3539,24 +3319,7 @@ static void handle_stripe6(struct stripe_head *sh)
 
 	if (s.expanded && test_bit(STRIPE_EXPANDING, &sh->state) &&
 	    !sh->reconstruct_state) {
-		struct stripe_head *sh2
-			= get_active_stripe(conf, sh->sector, 1, 1, 1);
-		if (sh2 && test_bit(STRIPE_EXPAND_SOURCE, &sh2->state)) {
-			/* sh cannot be written until sh2 has been read.
-			 * so arrange for sh to be delayed a little
-			 */
-			set_bit(STRIPE_DELAYED, &sh->state);
-			set_bit(STRIPE_HANDLE, &sh->state);
-			if (!test_and_set_bit(STRIPE_PREREAD_ACTIVE,
-					      &sh2->state))
-				atomic_inc(&conf->preread_active_stripes);
-			release_stripe(sh2);
-			goto unlock;
-		}
-		if (sh2)
-			release_stripe(sh2);
-
-		/* Need to write out all blocks after computing P&Q */
+		/* Need to write out all blocks after computing parity */
 		sh->disks = conf->raid_disks;
 		stripe_set_idx(sh->sector, conf, 0, sh);
 		schedule_reconstruction(sh, &s, 1, 1);
@@ -3569,22 +3332,39 @@ static void handle_stripe6(struct stripe_head *sh)
 
 	if (s.expanding && s.locked == 0 &&
 	    !test_bit(STRIPE_COMPUTE_RUN, &sh->state))
-		handle_stripe_expansion(conf, sh, &r6s);
-
- unlock:
-	spin_unlock(&sh->lock);
+		handle_stripe_expansion(conf, sh);
 
+finish:
 	/* wait for this device to become unblocked */
-	if (unlikely(blocked_rdev))
-		md_wait_for_blocked_rdev(blocked_rdev, conf->mddev);
+	if (unlikely(s.blocked_rdev))
+		md_wait_for_blocked_rdev(s.blocked_rdev, conf->mddev);
+
+	if (s.handle_bad_blocks)
+		for (i = disks; i--; ) {
+			mdk_rdev_t *rdev;
+			struct r5dev *dev = &sh->dev[i];
+			if (test_and_clear_bit(R5_WriteError, &dev->flags)) {
+				/* We own a safe reference to the rdev */
+				rdev = conf->disks[i].rdev;
+				if (!rdev_set_badblocks(rdev, sh->sector,
+							STRIPE_SECTORS, 0))
+					md_error(conf->mddev, rdev);
+				rdev_dec_pending(rdev, conf->mddev);
+			}
+			if (test_and_clear_bit(R5_MadeGood, &dev->flags)) {
+				rdev = conf->disks[i].rdev;
+				rdev_clear_badblocks(rdev, sh->sector,
+						     STRIPE_SECTORS);
+				rdev_dec_pending(rdev, conf->mddev);
+			}
+		}
 
 	if (s.ops_request)
 		raid_run_ops(sh, s.ops_request);
 
 	ops_run_io(sh, &s);
 
-
-	if (dec_preread_active) {
+	if (s.dec_preread_active) {
 		/* We delay this until after ops_run_io so that if make_request
 		 * is waiting on a flush, it won't continue until the writes
 		 * have actually been submitted.
@@ -3595,15 +3375,9 @@ static void handle_stripe6(struct stripe_head *sh)
 			md_wakeup_thread(conf->mddev->thread);
 	}
 
-	return_io(return_bi);
-}
+	return_io(s.return_bi);
 
-static void handle_stripe(struct stripe_head *sh)
-{
-	if (sh->raid_conf->level == 6)
-		handle_stripe6(sh);
-	else
-		handle_stripe5(sh);
+	clear_bit(STRIPE_ACTIVE, &sh->state);
 }
 
 static void raid5_activate_delayed(raid5_conf_t *conf)
@@ -3833,6 +3607,9 @@ static int chunk_aligned_read(mddev_t *mddev, struct bio * raid_bio)
 	rcu_read_lock();
 	rdev = rcu_dereference(conf->disks[dd_idx].rdev);
 	if (rdev && test_bit(In_sync, &rdev->flags)) {
+		sector_t first_bad;
+		int bad_sectors;
+
 		atomic_inc(&rdev->nr_pending);
 		rcu_read_unlock();
 		raid_bio->bi_next = (void*)rdev;
@@ -3840,8 +3617,10 @@ static int chunk_aligned_read(mddev_t *mddev, struct bio * raid_bio)
 		align_bi->bi_flags &= ~(1 << BIO_SEG_VALID);
 		align_bi->bi_sector += rdev->data_offset;
 
-		if (!bio_fits_rdev(align_bi)) {
-			/* too big in some way */
+		if (!bio_fits_rdev(align_bi) ||
+		    is_badblock(rdev, align_bi->bi_sector, align_bi->bi_size>>9,
+				&first_bad, &bad_sectors)) {
+			/* too big in some way, or has a known bad block */
 			bio_put(align_bi);
 			rdev_dec_pending(rdev, mddev);
 			return 0;
@@ -4016,7 +3795,7 @@ static int make_request(mddev_t *mddev, struct bio * bi)
 				}
 			}
 
-			if (bio_data_dir(bi) == WRITE &&
+			if (rw == WRITE &&
 			    logical_sector >= mddev->suspend_lo &&
 			    logical_sector < mddev->suspend_hi) {
 				release_stripe(sh);
@@ -4034,7 +3813,7 @@ static int make_request(mddev_t *mddev, struct bio * bi)
 			}
 
 			if (test_bit(STRIPE_EXPANDING, &sh->state) ||
-			    !add_stripe_bio(sh, bi, dd_idx, (bi->bi_rw&RW_MASK))) {
+			    !add_stripe_bio(sh, bi, dd_idx, rw)) {
 				/* Stripe is busy expanding or
 				 * add failed due to overlap.  Flush everything
 				 * and wait a while
@@ -4375,10 +4154,7 @@ static inline sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *ski
 
 	bitmap_start_sync(mddev->bitmap, sector_nr, &sync_blocks, still_degraded);
 
-	spin_lock(&sh->lock);
-	set_bit(STRIPE_SYNCING, &sh->state);
-	clear_bit(STRIPE_INSYNC, &sh->state);
-	spin_unlock(&sh->lock);
+	set_bit(STRIPE_SYNC_REQUESTED, &sh->state);
 
 	handle_stripe(sh);
 	release_stripe(sh);
@@ -4509,6 +4285,9 @@ static void raid5d(mddev_t *mddev)
 		release_stripe(sh);
 		cond_resched();
 
+		if (mddev->flags & ~(1<<MD_CHANGE_PENDING))
+			md_check_recovery(mddev);
+
 		spin_lock_irq(&conf->device_lock);
 	}
 	pr_debug("%d stripes handled\n", handled);
@@ -5313,6 +5092,7 @@ static int raid5_remove_disk(mddev_t *mddev, int number)
 		 * isn't possible.
 		 */
 		if (!test_bit(Faulty, &rdev->flags) &&
+		    mddev->recovery_disabled != conf->recovery_disabled &&
 		    !has_failed(conf) &&
 		    number < conf->raid_disks) {
 			err = -EBUSY;
@@ -5341,6 +5121,9 @@ static int raid5_add_disk(mddev_t *mddev, mdk_rdev_t *rdev)
 	int first = 0;
 	int last = conf->raid_disks - 1;
 
+	if (mddev->recovery_disabled == conf->recovery_disabled)
+		return -EBUSY;
+
 	if (has_failed(conf))
 		/* no point adding a device */
 		return -EINVAL;
@@ -5519,16 +5302,14 @@ static int raid5_start_reshape(mddev_t *mddev)
 			if (rdev->raid_disk < 0 &&
 			    !test_bit(Faulty, &rdev->flags)) {
 				if (raid5_add_disk(mddev, rdev) == 0) {
-					char nm[20];
 					if (rdev->raid_disk
 					    >= conf->previous_raid_disks) {
 						set_bit(In_sync, &rdev->flags);
 						added_devices++;
 					} else
 						rdev->recovery_offset = 0;
-					sprintf(nm, "rd%d", rdev->raid_disk);
-					if (sysfs_create_link(&mddev->kobj,
-							      &rdev->kobj, nm))
+
+					if (sysfs_link_rdev(mddev, rdev))
 						/* Failure here is OK */;
 				}
 			} else if (rdev->raid_disk >= conf->previous_raid_disks
@@ -5624,9 +5405,7 @@ static void raid5_finish_reshape(mddev_t *mddev)
 			     d++) {
 				mdk_rdev_t *rdev = conf->disks[d].rdev;
 				if (rdev && raid5_remove_disk(mddev, d) == 0) {
-					char nm[20];
-					sprintf(nm, "rd%d", rdev->raid_disk);
-					sysfs_remove_link(&mddev->kobj, nm);
+					sysfs_unlink_rdev(mddev, rdev);
 					rdev->raid_disk = -1;
 				}
 			}

drivers/md/raid5.h

@@ -6,11 +6,11 @@
 
 /*
  *
- * Each stripe contains one buffer per disc.  Each buffer can be in
+ * Each stripe contains one buffer per device.  Each buffer can be in
  * one of a number of states stored in "flags".  Changes between
- * these states happen *almost* exclusively under a per-stripe
- * spinlock.  Some very specific changes can happen in bi_end_io, and
- * these are not protected by the spin lock.
+ * these states happen *almost* exclusively under the protection of the
+ * STRIPE_ACTIVE flag.  Some very specific changes can happen in bi_end_io, and
+ * these are not protected by STRIPE_ACTIVE.
  *
  * The flag bits that are used to represent these states are:
  *   R5_UPTODATE and R5_LOCKED
@@ -76,12 +76,10 @@
  * block and the cached buffer are successfully written, any buffer on
  * a written list can be returned with b_end_io.
  *
- * The write list and read list both act as fifos.  The read list is
- * protected by the device_lock.  The write and written lists are
- * protected by the stripe lock.  The device_lock, which can be
- * claimed while the stipe lock is held, is only for list
- * manipulations and will only be held for a very short time.  It can
- * be claimed from interrupts.
+ * The write list and read list both act as fifos.  The read list,
+ * write list and written list are protected by the device_lock.
+ * The device_lock is only for list manipulations and will only be
+ * held for a very short time.  It can be claimed from interrupts.
  *
  *
  * Stripes in the stripe cache can be on one of two lists (or on
@@ -96,7 +94,6 @@
  *
  * The inactive_list, handle_list and hash bucket lists are all protected by the
  * device_lock.
- *  - stripes on the inactive_list never have their stripe_lock held.
  *  - stripes have a reference counter. If count==0, they are on a list.
  *  - If a stripe might need handling, STRIPE_HANDLE is set.
  *  - When refcount reaches zero, then if STRIPE_HANDLE it is put on
@@ -116,10 +113,10 @@
  *  attach a request to an active stripe (add_stripe_bh())
  *     lockdev attach-buffer unlockdev
  *  handle a stripe (handle_stripe())
- *     lockstripe clrSTRIPE_HANDLE ...
+ *     setSTRIPE_ACTIVE,  clrSTRIPE_HANDLE ...
  *		(lockdev check-buffers unlockdev) ..
  *		change-state ..
- *		record io/ops needed unlockstripe schedule io/ops
+ *		record io/ops needed clearSTRIPE_ACTIVE schedule io/ops
  *  release an active stripe (release_stripe())
  *     lockdev if (!--cnt) { if  STRIPE_HANDLE, add to handle_list else add to inactive-list } unlockdev
  *
@@ -128,8 +125,7 @@
  * on a cached buffer, and plus one if the stripe is undergoing stripe
  * operations.
  *
- * Stripe operations are performed outside the stripe lock,
- * the stripe operations are:
+ * The stripe operations are:
  * -copying data between the stripe cache and user application buffers
  * -computing blocks to save a disk access, or to recover a missing block
  * -updating the parity on a write operation (reconstruct write and
@@ -159,7 +155,8 @@
  */
 
 /*
- * Operations state - intermediate states that are visible outside of sh->lock
+ * Operations state - intermediate states that are visible outside of 
+ *   STRIPE_ACTIVE.
  * In general _idle indicates nothing is running, _run indicates a data
  * processing operation is active, and _result means the data processing result
  * is stable and can be acted upon.  For simple operations like biofill and
@@ -209,7 +206,6 @@ struct stripe_head {
 	short			ddf_layout;/* use DDF ordering to calculate Q */
 	unsigned long		state;		/* state flags */
 	atomic_t		count;	      /* nr of active thread/requests */
-	spinlock_t		lock;
 	int			bm_seq;	/* sequence number for bitmap flushes */
 	int			disks;		/* disks in stripe */
 	enum check_states	check_state;
@@ -240,19 +236,20 @@ struct stripe_head {
 };
 
 /* stripe_head_state - collects and tracks the dynamic state of a stripe_head
- *     for handle_stripe.  It is only valid under spin_lock(sh->lock);
+ *     for handle_stripe.
  */
 struct stripe_head_state {
 	int syncing, expanding, expanded;
 	int locked, uptodate, to_read, to_write, failed, written;
 	int to_fill, compute, req_compute, non_overwrite;
-	int failed_num;
+	int failed_num[2];
+	int p_failed, q_failed;
+	int dec_preread_active;
 	unsigned long ops_request;
-};
 
-/* r6_state - extra state data only relevant to r6 */
-struct r6_state {
-	int p_failed, q_failed, failed_num[2];
+	struct bio *return_bi;
+	mdk_rdev_t *blocked_rdev;
+	int handle_bad_blocks;
 };
 
 /* Flags */
@@ -268,14 +265,16 @@ struct r6_state {
 #define	R5_ReWrite	9	/* have tried to over-write the readerror */
 
 #define	R5_Expanded	10	/* This block now has post-expand data */
-#define	R5_Wantcompute	11 /* compute_block in progress treat as
-				    * uptodate
-				    */
-#define	R5_Wantfill	12 /* dev->toread contains a bio that needs
-				    * filling
-				    */
-#define R5_Wantdrain	13 /* dev->towrite needs to be drained */
-#define R5_WantFUA	14	/* Write should be FUA */
+#define	R5_Wantcompute	11	/* compute_block in progress treat as
+				 * uptodate
+				 */
+#define	R5_Wantfill	12	/* dev->toread contains a bio that needs
+				 * filling
+				 */
+#define	R5_Wantdrain	13	/* dev->towrite needs to be drained */
+#define	R5_WantFUA	14	/* Write should be FUA */
+#define	R5_WriteError	15	/* got a write error - need to record it */
+#define	R5_MadeGood	16	/* A bad block has been fixed by writing to it*/
 /*
  * Write method
  */
@@ -289,21 +288,25 @@ struct r6_state {
 /*
  * Stripe state
  */
-#define STRIPE_HANDLE		2
-#define	STRIPE_SYNCING		3
-#define	STRIPE_INSYNC		4
-#define	STRIPE_PREREAD_ACTIVE	5
-#define	STRIPE_DELAYED		6
-#define	STRIPE_DEGRADED		7
-#define	STRIPE_BIT_DELAY	8
-#define	STRIPE_EXPANDING	9
-#define	STRIPE_EXPAND_SOURCE	10
-#define	STRIPE_EXPAND_READY	11
-#define	STRIPE_IO_STARTED	12 /* do not count towards 'bypass_count' */
-#define	STRIPE_FULL_WRITE	13 /* all blocks are set to be overwritten */
-#define	STRIPE_BIOFILL_RUN	14
-#define	STRIPE_COMPUTE_RUN	15
-#define	STRIPE_OPS_REQ_PENDING	16
+enum {
+	STRIPE_ACTIVE,
+	STRIPE_HANDLE,
+	STRIPE_SYNC_REQUESTED,
+	STRIPE_SYNCING,
+	STRIPE_INSYNC,
+	STRIPE_PREREAD_ACTIVE,
+	STRIPE_DELAYED,
+	STRIPE_DEGRADED,
+	STRIPE_BIT_DELAY,
+	STRIPE_EXPANDING,
+	STRIPE_EXPAND_SOURCE,
+	STRIPE_EXPAND_READY,
+	STRIPE_IO_STARTED,	/* do not count towards 'bypass_count' */
+	STRIPE_FULL_WRITE,	/* all blocks are set to be overwritten */
+	STRIPE_BIOFILL_RUN,
+	STRIPE_COMPUTE_RUN,
+	STRIPE_OPS_REQ_PENDING,
+};
 
 /*
  * Operation request flags
@@ -336,7 +339,7 @@ struct r6_state {
  * PREREAD_ACTIVE.
  * In stripe_handle, if we find pre-reading is necessary, we do it if
  * PREREAD_ACTIVE is set, else we set DELAYED which will send it to the delayed queue.
- * HANDLE gets cleared if stripe_handle leave nothing locked.
+ * HANDLE gets cleared if stripe_handle leaves nothing locked.
  */
 
 
@@ -399,7 +402,7 @@ struct raid5_private_data {
 					    * (fresh device added).
 					    * Cleared when a sync completes.
 					    */
-
+	int			recovery_disabled;
 	/* per cpu variables */
 	struct raid5_percpu {
 		struct page	*spare_page; /* Used when checking P/Q in raid6 */

include/linux/raid/md_p.h

@@ -245,10 +245,16 @@ struct mdp_superblock_1 {
 	__u8	device_uuid[16]; /* user-space setable, ignored by kernel */
 	__u8	devflags;	/* per-device flags.  Only one defined...*/
 #define	WriteMostly1	1	/* mask for writemostly flag in above */
-	__u8	pad2[64-57];	/* set to 0 when writing */
+	/* Bad block log.  If there are any bad blocks the feature flag is set.
+	 * If offset and size are non-zero, that space is reserved and available
+	 */
+	__u8	bblog_shift;	/* shift from sectors to block size */
+	__le16	bblog_size;	/* number of sectors reserved for list */
+	__le32	bblog_offset;	/* sector offset from superblock to bblog,
+				 * signed - not unsigned */
 
 	/* array state information - 64 bytes */
-	__le64	utime;		/* 40 bits second, 24 btes microseconds */
+	__le64	utime;		/* 40 bits second, 24 bits microseconds */
 	__le64	events;		/* incremented when superblock updated */
 	__le64	resync_offset;	/* data before this offset (from data_offset) known to be in sync */
 	__le32	sb_csum;	/* checksum up to devs[max_dev] */
@@ -270,8 +276,8 @@ struct mdp_superblock_1 {
 					   * must be honoured
 					   */
 #define	MD_FEATURE_RESHAPE_ACTIVE	4
+#define	MD_FEATURE_BAD_BLOCKS		8 /* badblock list is not empty */
 
-#define	MD_FEATURE_ALL			(1|2|4)
+#define	MD_FEATURE_ALL			(1|2|4|8)
 
 #endif 
-