[MTD] NAND Signal that a bitflip was corrected by ECC

Return -EUCLEAN on read when a bitflip was detected and corrected, so the
clients can react and eventually copy the affected block to a spare one.
Make all in kernel users aware of the change.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

6 files changed, 46 insertions, 24 deletions

diff --git a/drivers/mtd/inftlcore.c b/drivers/mtd/inftlcore.c
index efb1a95aa0a0..1e21a2c3dd29 100644
--- a/drivers/mtd/inftlcore.c
+++ b/drivers/mtd/inftlcore.c
@@ -355,7 +355,7 @@ static u16 INFTL_foldchain(struct INFTLrecord *inftl, unsigned thisVUC, unsigned
                 ret = mtd->read(mtd, (inftl->EraseSize * BlockMap[block]) +
                                 (block * SECTORSIZE), SECTORSIZE, &retlen,
                                 movebuf);
-                if (ret < 0) {
+                if (ret < 0 && ret != -EUCLEAN) {
                         ret = mtd->read(mtd,
                                         (inftl->EraseSize * BlockMap[block]) +
                                         (block * SECTORSIZE), SECTORSIZE,
@@ -922,7 +922,10 @@ foundit:
         } else {
                 size_t retlen;
                 loff_t ptr = (thisEUN * inftl->EraseSize) + blockofs;
-                if (mtd->read(mtd, ptr, SECTORSIZE, &retlen, buffer))
+                int ret = mtd->read(mtd, ptr, SECTORSIZE, &retlen, buffer);
+
+                /* Handle corrected bit flips gracefully */
+                if (ret < 0 && ret != -EUCLEAN)
                         return -EIO;
         }
         return 0;
diff --git a/drivers/mtd/mtdchar.c b/drivers/mtd/mtdchar.c
index 7522fc3a2827..a48210d58b92 100644
--- a/drivers/mtd/mtdchar.c
+++ b/drivers/mtd/mtdchar.c
@@ -199,10 +199,13 @@ static ssize_t mtd_read(struct file *file, char __user *buf, size_t count,loff_t
                 /* Nand returns -EBADMSG on ecc errors, but it returns
                  * the data. For our userspace tools it is important
                  * to dump areas with ecc errors !
+                 * For kernel internal usage it also might return -EUCLEAN
+                 * to signal the caller that a bitflip has occured and has
+                 * been corrected by the ECC algorithm.
                  * Userspace software which accesses NAND this way
                  * must be aware of the fact that it deals with NAND
                  */
-                if (!ret || (ret == -EBADMSG)) {
+                if (!ret || (ret == -EUCLEAN) || (ret == -EBADMSG)) {
                         *ppos += retlen;
                         if (copy_to_user(buf, kbuf, retlen)) {
                                 kfree(kbuf);
diff --git a/drivers/mtd/mtdconcat.c b/drivers/mtd/mtdconcat.c
index 38151b8e6631..3c8d5e6fa010 100644
--- a/drivers/mtd/mtdconcat.c
+++ b/drivers/mtd/mtdconcat.c
@@ -56,7 +56,7 @@ concat_read(struct mtd_info *mtd, loff_t from, size_t len,
             size_t * retlen, u_char * buf)
 {
         struct mtd_concat *concat = CONCAT(mtd);
-        int err = -EINVAL;
+        int ret = 0, err = -EINVAL;
         int i;
 
         *retlen = 0;
@@ -80,9 +80,18 @@ concat_read(struct mtd_info *mtd, loff_t from, size_t len,
 
                 err = subdev->read(subdev, from, size, &retsize, buf);
 
-                if (err)
+                if (err && (err != -EBADMSG) && (err != -EUCLEAN))
                         break;
 
+                /* Save information about bitflips! */
+                if (err) {
+                        if (err == -EBADMSG)
+                                ret = err;
+                        else if (!ret)
+                                ret = err;
+                        err = 0;
+                }
+
                 *retlen += retsize;
                 len -= size;
                 if (len == 0)
@@ -92,7 +101,7 @@ concat_read(struct mtd_info *mtd, loff_t from, size_t len,
                 buf += size;
                 from = 0;
         }
-        return err;
+        return err ? err : ret;
 }
 
 static int
diff --git a/drivers/mtd/nand/nand_base.c b/drivers/mtd/nand/nand_base.c
index b8e6e1579cf1..7a3a44907715 100644
--- a/drivers/mtd/nand/nand_base.c
+++ b/drivers/mtd/nand/nand_base.c
@@ -1035,7 +1035,10 @@ static int nand_do_read_ops(struct mtd_info *mtd, loff_t from,
         if (ret)
                 return ret;
 
-        return mtd->ecc_stats.failed - stats.failed ? -EBADMSG : 0;
+        if (mtd->ecc_stats.failed - stats.failed)
+                return -EBADMSG;
+
+        return  mtd->ecc_stats.corrected - stats.corrected ? -EUCLEAN : 0;
 }
 
 /**
diff --git a/drivers/mtd/nftlcore.c b/drivers/mtd/nftlcore.c
index f6ffe7949b26..dc7573501d8c 100644
--- a/drivers/mtd/nftlcore.c
+++ b/drivers/mtd/nftlcore.c
@@ -422,7 +422,7 @@ static u16 NFTL_foldchain (struct NFTLrecord *nftl, unsigned thisVUC, unsigned p
 
                 ret = mtd->read(mtd, (nftl->EraseSize * BlockMap[block]) + (block * 512),
                                 512, &retlen, movebuf);
-                if (ret < 0) {
+                if (ret < 0 && ret != -EUCLEAN) {
                         ret = mtd->read(mtd, (nftl->EraseSize * BlockMap[block])
                                         + (block * 512), 512, &retlen,
                                         movebuf);
@@ -768,7 +768,9 @@ static int nftl_readblock(struct mtd_blktrans_dev *mbd, unsigned long block,
         } else {
                 loff_t ptr = (lastgoodEUN * nftl->EraseSize) + blockofs;
                 size_t retlen;
-                if (mtd->read(mtd, ptr, 512, &retlen, buffer))
+                int res = mtd->read(mtd, ptr, 512, &retlen, buffer);
+
+                if (res < 0 && res != -EUCLEAN)
                         return -EIO;
         }
         return 0;
diff --git a/fs/jffs2/wbuf.c b/fs/jffs2/wbuf.c
index 1195d06d4373..a7f153f79ecb 100644
--- a/fs/jffs2/wbuf.c
+++ b/fs/jffs2/wbuf.c
@@ -296,10 +296,11 @@ static void jffs2_wbuf_recover(struct jffs2_sb_info *c)
                 /* Do the read... */
                 ret = c->mtd->read(c->mtd, start, c->wbuf_ofs - start, &retlen, buf);
 
-                if (ret == -EBADMSG && retlen == c->wbuf_ofs - start) {
-                        /* ECC recovered */
+                /* ECC recovered ? */
+                if ((ret == -EUCLEAN || ret == -EBADMSG) &&
+                    (retlen == c->wbuf_ofs - start))
                         ret = 0;
-                }
+
                 if (ret || retlen != c->wbuf_ofs - start) {
                         printk(KERN_CRIT "Old data are already lost in wbuf recovery. Data loss ensues.\n");
 
@@ -908,20 +909,21 @@ int jffs2_flash_read(struct jffs2_sb_info *c, loff_t ofs, size_t len, size_t *re
         down_read(&c->wbuf_sem);
         ret = c->mtd->read(c->mtd, ofs, len, retlen, buf);
 
-        if ( (ret == -EBADMSG) && (*retlen == len) ) {
-                printk(KERN_WARNING "mtd->read(0x%zx bytes from 0x%llx) returned ECC error\n",
-                       len, ofs);
+        if ( (ret == -EBADMSG || ret == -EUCLEAN) && (*retlen == len) ) {
+                if (ret == -EBADMSG)
+                        printk(KERN_WARNING "mtd->read(0x%zx bytes from 0x%llx)"
+                               " returned ECC error\n", len, ofs);
                 /*
-                 * We have the raw data without ECC correction in the buffer, maybe
-                 * we are lucky and all data or parts are correct. We check the node.
-                 * If data are corrupted node check will sort it out.
-                 * We keep this block, it will fail on write or erase and the we
-                 * mark it bad. Or should we do that now? But we should give him a chance.
-                 * Maybe we had a system crash or power loss before the ecc write or
-                 * a erase was completed.
+                 * We have the raw data without ECC correction in the buffer,
+                 * maybe we are lucky and all data or parts are correct. We
+                 * check the node.  If data are corrupted node check will sort
+                 * it out.  We keep this block, it will fail on write or erase
+                 * and the we mark it bad. Or should we do that now? But we
+                 * should give him a chance.  Maybe we had a system crash or
+                 * power loss before the ecc write or a erase was completed.
                  * So we return success. :)
                  */
-                ret = 0;
+                ret = 0;
         }
 
         /* if no writebuffer available or write buffer empty, return */
@@ -943,7 +945,7 @@ int jffs2_flash_read(struct jffs2_sb_info *c, loff_t ofs, size_t len, size_t *re
                 orbf = (c->wbuf_ofs - ofs);     /* offset in read buffer */
                 if (orbf > len)                 /* is write beyond write buffer ? */
                         goto exit;
-                lwbf = len - orbf;              /* number of bytes to copy */
+                lwbf = len - orbf;              /* number of bytes to copy */
                 if (lwbf > c->wbuf_len)
                         lwbf = c->wbuf_len;
         }