[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 <[e-mail]>

drivers/mtd/inftlcore.c

···
 		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,
···
 	} 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;

drivers/mtd/mtdchar.c

···
 		/* 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);

drivers/mtd/mtdconcat.c

···
 	    size_t * retlen, u_char * buf)
 {
 	struct mtd_concat *concat = CONCAT(mtd);
-	int err = -EINVAL;
+	int ret = 0, err = -EINVAL;
 	int i;
 
 	*retlen = 0;
···
 
 		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;
···
 		buf += size;
 		from = 0;
 	}
-	return err;
+	return err ? err : ret;
 }
 
 static int

drivers/mtd/nand/nand_base.c

···
 	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;
 }
 
 /**

drivers/mtd/nftlcore.c

···
 
 		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);
···
 	} 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;

fs/jffs2/wbuf.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");
 
···
 	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 */
···
 		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;
 	}