Merge tag 'exfat-for-6.15-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/linkinjeon/exfat

Pull exfat updates from Namjae Jeon:

 - Fix random stack corruption and incorrect error returns in
   exfat_get_block()

 - Optimize exfat_get_block() by improving checking corner cases

 - Fix an endless loop by self-linked chain in exfat_find_last_cluster

 - Remove dead EXFAT_CLUSTERS_UNTRACKED codes

 - Add missing shutdown check

 - Improve the delete performance with discard mount option

* tag 'exfat-for-6.15-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/linkinjeon/exfat:
  exfat: call bh_read in get_block only when necessary
  exfat: fix potential wrong error return from get_block
  exfat: fix missing shutdown check
  exfat: fix the infinite loop in exfat_find_last_cluster()
  exfat: fix random stack corruption after get_block
  exfat: remove count used cluster from exfat_statfs()
  exfat: support batch discard of clusters when freeing clusters

Author: torvalds

fs/exfat/balloc.c

@@ -147,7 +147,6 @@ int exfat_clear_bitmap(struct inode *inode, unsigned int clu, bool sync)
 	unsigned int ent_idx;
 	struct super_block *sb = inode->i_sb;
 	struct exfat_sb_info *sbi = EXFAT_SB(sb);
-	struct exfat_mount_options *opts = &sbi->options;
 
 	if (!is_valid_cluster(sbi, clu))
 		return -EIO;
@@ -163,19 +162,6 @@ int exfat_clear_bitmap(struct inode *inode, unsigned int clu, bool sync)
 
 	exfat_update_bh(sbi->vol_amap[i], sync);
 
-	if (opts->discard) {
-		int ret_discard;
-
-		ret_discard = sb_issue_discard(sb,
-			exfat_cluster_to_sector(sbi, clu),
-			(1 << sbi->sect_per_clus_bits), GFP_NOFS, 0);
-
-		if (ret_discard == -EOPNOTSUPP) {
-			exfat_err(sb, "discard not supported by device, disabling");
-			opts->discard = 0;
-		}
-	}
-
 	return 0;
 }
 

fs/exfat/exfat_fs.h

@@ -14,8 +14,6 @@
 
 #define EXFAT_ROOT_INO		1
 
-#define EXFAT_CLUSTERS_UNTRACKED (~0u)
-
 /*
  * exfat error flags
  */

fs/exfat/fatent.c

@@ -144,6 +144,20 @@ int exfat_chain_cont_cluster(struct super_block *sb, unsigned int chain,
 	return 0;
 }
 
+static inline void exfat_discard_cluster(struct super_block *sb,
+		unsigned int clu, unsigned int num_clusters)
+{
+	int ret;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+
+	ret = sb_issue_discard(sb, exfat_cluster_to_sector(sbi, clu),
+			sbi->sect_per_clus * num_clusters, GFP_NOFS, 0);
+	if (ret == -EOPNOTSUPP) {
+		exfat_err(sb, "discard not supported by device, disabling");
+		sbi->options.discard = 0;
+	}
+}
+
 /* This function must be called with bitmap_lock held */
 static int __exfat_free_cluster(struct inode *inode, struct exfat_chain *p_chain)
 {
@@ -196,7 +210,12 @@ static int __exfat_free_cluster(struct inode *inode, struct exfat_chain *p_chain
 			clu++;
 			num_clusters++;
 		} while (num_clusters < p_chain->size);
+
+		if (sbi->options.discard)
+			exfat_discard_cluster(sb, p_chain->dir, p_chain->size);
 	} else {
+		unsigned int nr_clu = 1;
+
 		do {
 			bool sync = false;
 			unsigned int n_clu = clu;
@@ -215,6 +234,16 @@ static int __exfat_free_cluster(struct inode *inode, struct exfat_chain *p_chain
 
 			if (exfat_clear_bitmap(inode, clu, (sync && IS_DIRSYNC(inode))))
 				break;
+
+			if (sbi->options.discard) {
+				if (n_clu == clu + 1)
+					nr_clu++;
+				else {
+					exfat_discard_cluster(sb, clu - nr_clu + 1, nr_clu);
+					nr_clu = 1;
+				}
+			}
+
 			clu = n_clu;
 			num_clusters++;
 
@@ -265,7 +294,7 @@ int exfat_find_last_cluster(struct super_block *sb, struct exfat_chain *p_chain,
 		clu = next;
 		if (exfat_ent_get(sb, clu, &next))
 			return -EIO;
-	} while (next != EXFAT_EOF_CLUSTER);
+	} while (next != EXFAT_EOF_CLUSTER && count <= p_chain->size);
 
 	if (p_chain->size != count) {
 		exfat_fs_error(sb,

fs/exfat/file.c

@@ -582,6 +582,9 @@ static ssize_t exfat_file_write_iter(struct kiocb *iocb, struct iov_iter *iter)
 	loff_t pos = iocb->ki_pos;
 	loff_t valid_size;
 
+	if (unlikely(exfat_forced_shutdown(inode->i_sb)))
+		return -EIO;
+
 	inode_lock(inode);
 
 	valid_size = ei->valid_size;
@@ -635,6 +638,16 @@ static ssize_t exfat_file_write_iter(struct kiocb *iocb, struct iov_iter *iter)
 	return ret;
 }
 
+static ssize_t exfat_file_read_iter(struct kiocb *iocb, struct iov_iter *iter)
+{
+	struct inode *inode = file_inode(iocb->ki_filp);
+
+	if (unlikely(exfat_forced_shutdown(inode->i_sb)))
+		return -EIO;
+
+	return generic_file_read_iter(iocb, iter);
+}
+
 static vm_fault_t exfat_page_mkwrite(struct vm_fault *vmf)
 {
 	int err;
@@ -672,22 +685,34 @@ static const struct vm_operations_struct exfat_file_vm_ops = {
 
 static int exfat_file_mmap(struct file *file, struct vm_area_struct *vma)
 {
+	if (unlikely(exfat_forced_shutdown(file_inode(file)->i_sb)))
+		return -EIO;
+
 	file_accessed(file);
 	vma->vm_ops = &exfat_file_vm_ops;
 	return 0;
 }
 
+static ssize_t exfat_splice_read(struct file *in, loff_t *ppos,
+		struct pipe_inode_info *pipe, size_t len, unsigned int flags)
+{
+	if (unlikely(exfat_forced_shutdown(file_inode(in)->i_sb)))
+		return -EIO;
+
+	return filemap_splice_read(in, ppos, pipe, len, flags);
+}
+
 const struct file_operations exfat_file_operations = {
 	.llseek		= generic_file_llseek,
-	.read_iter	= generic_file_read_iter,
+	.read_iter	= exfat_file_read_iter,
 	.write_iter	= exfat_file_write_iter,
 	.unlocked_ioctl = exfat_ioctl,
 #ifdef CONFIG_COMPAT
 	.compat_ioctl = exfat_compat_ioctl,
 #endif
 	.mmap		= exfat_file_mmap,
 	.fsync		= exfat_file_fsync,
-	.splice_read	= filemap_splice_read,
+	.splice_read	= exfat_splice_read,
 	.splice_write	= iter_file_splice_write,
 };
 

fs/exfat/inode.c

@@ -274,9 +274,11 @@ static int exfat_get_block(struct inode *inode, sector_t iblock,
 	sector_t last_block;
 	sector_t phys = 0;
 	sector_t valid_blks;
+	loff_t i_size;
 
 	mutex_lock(&sbi->s_lock);
-	last_block = EXFAT_B_TO_BLK_ROUND_UP(i_size_read(inode), sb);
+	i_size = i_size_read(inode);
+	last_block = EXFAT_B_TO_BLK_ROUND_UP(i_size, sb);
 	if (iblock >= last_block && !create)
 		goto done;
 
@@ -305,77 +307,99 @@ static int exfat_get_block(struct inode *inode, sector_t iblock,
 	if (buffer_delay(bh_result))
 		clear_buffer_delay(bh_result);
 
-	if (create) {
+	/*
+	 * In most cases, we just need to set bh_result to mapped, unmapped
+	 * or new status as follows:
+	 *  1. i_size == valid_size
+	 *  2. write case (create == 1)
+	 *  3. direct_read (!bh_result->b_folio)
+	 *     -> the unwritten part will be zeroed in exfat_direct_IO()
+	 *
+	 * Otherwise, in the case of buffered read, it is necessary to take
+	 * care the last nested block if valid_size is not equal to i_size.
+	 */
+	if (i_size == ei->valid_size || create || !bh_result->b_folio)
 		valid_blks = EXFAT_B_TO_BLK_ROUND_UP(ei->valid_size, sb);
+	else
+		valid_blks = EXFAT_B_TO_BLK(ei->valid_size, sb);
 
-		if (iblock + max_blocks < valid_blks) {
-			/* The range has been written, map it */
-			goto done;
-		} else if (iblock < valid_blks) {
-			/*
-			 * The range has been partially written,
-			 * map the written part.
-			 */
-			max_blocks = valid_blks - iblock;
-			goto done;
-		}
+	/* The range has been fully written, map it */
+	if (iblock + max_blocks < valid_blks)
+		goto done;
 
-		/* The area has not been written, map and mark as new. */
-		set_buffer_new(bh_result);
+	/* The range has been partially written, map the written part */
+	if (iblock < valid_blks) {
+		max_blocks = valid_blks - iblock;
+		goto done;
+	}
 
+	/* The area has not been written, map and mark as new for create case */
+	if (create) {
+		set_buffer_new(bh_result);
 		ei->valid_size = EXFAT_BLK_TO_B(iblock + max_blocks, sb);
 		mark_inode_dirty(inode);
-	} else {
-		valid_blks = EXFAT_B_TO_BLK(ei->valid_size, sb);
+		goto done;
+	}
 
-		if (iblock + max_blocks < valid_blks) {
-			/* The range has been written, map it */
+	/*
+	 * The area has just one block partially written.
+	 * In that case, we should read and fill the unwritten part of
+	 * a block with zero.
+	 */
+	if (bh_result->b_folio && iblock == valid_blks &&
+	    (ei->valid_size & (sb->s_blocksize - 1))) {
+		loff_t size, pos;
+		void *addr;
+
+		max_blocks = 1;
+
+		/*
+		 * No buffer_head is allocated.
+		 * (1) bmap: It's enough to set blocknr without I/O.
+		 * (2) read: The unwritten part should be filled with zero.
+		 *           If a folio does not have any buffers,
+		 *           let's returns -EAGAIN to fallback to
+		 *           block_read_full_folio() for per-bh IO.
+		 */
+		if (!folio_buffers(bh_result->b_folio)) {
+			err = -EAGAIN;
 			goto done;
-		} else if (iblock < valid_blks) {
-			/*
-			 * The area has been partially written,
-			 * map the written part.
-			 */
-			max_blocks = valid_blks - iblock;
+		}
+
+		pos = EXFAT_BLK_TO_B(iblock, sb);
+		size = ei->valid_size - pos;
+		addr = folio_address(bh_result->b_folio) +
+			offset_in_folio(bh_result->b_folio, pos);
+
+		/* Check if bh->b_data points to proper addr in folio */
+		if (bh_result->b_data != addr) {
+			exfat_fs_error_ratelimit(sb,
+					"b_data(%p) != folio_addr(%p)",
+					bh_result->b_data, addr);
+			err = -EINVAL;
 			goto done;
-		} else if (iblock == valid_blks &&
-			   (ei->valid_size & (sb->s_blocksize - 1))) {
-			/*
-			 * The block has been partially written,
-			 * zero the unwritten part and map the block.
-			 */
-			loff_t size, off, pos;
-
-			max_blocks = 1;
-
-			/*
-			 * For direct read, the unwritten part will be zeroed in
-			 * exfat_direct_IO()
-			 */
-			if (!bh_result->b_folio)
-				goto done;
-
-			pos = EXFAT_BLK_TO_B(iblock, sb);
-			size = ei->valid_size - pos;
-			off = pos & (PAGE_SIZE - 1);
-
-			folio_set_bh(bh_result, bh_result->b_folio, off);
-			err = bh_read(bh_result, 0);
-			if (err < 0)
-				goto unlock_ret;
-
-			folio_zero_segment(bh_result->b_folio, off + size,
-					off + sb->s_blocksize);
-		} else {
-			/*
-			 * The range has not been written, clear the mapped flag
-			 * to only zero the cache and do not read from disk.
-			 */
-			clear_buffer_mapped(bh_result);
 		}
+
+		/* Read a block */
+		err = bh_read(bh_result, 0);
+		if (err < 0)
+			goto done;
+
+		/* Zero unwritten part of a block */
+		memset(bh_result->b_data + size, 0, bh_result->b_size - size);
+		err = 0;
+		goto done;
 	}
+
+	/*
+	 * The area has not been written, clear mapped for read/bmap cases.
+	 * If so, it will be filled with zero without reading from disk.
+	 */
+	clear_buffer_mapped(bh_result);
 done:
 	bh_result->b_size = EXFAT_BLK_TO_B(max_blocks, sb);
+	if (err < 0)
+		clear_buffer_mapped(bh_result);
 unlock_ret:
 	mutex_unlock(&sbi->s_lock);
 	return err;

fs/exfat/super.c

@@ -67,15 +67,6 @@ static int exfat_statfs(struct dentry *dentry, struct kstatfs *buf)
 	struct exfat_sb_info *sbi = EXFAT_SB(sb);
 	unsigned long long id = huge_encode_dev(sb->s_bdev->bd_dev);
 
-	if (sbi->used_clusters == EXFAT_CLUSTERS_UNTRACKED) {
-		mutex_lock(&sbi->s_lock);
-		if (exfat_count_used_clusters(sb, &sbi->used_clusters)) {
-			mutex_unlock(&sbi->s_lock);
-			return -EIO;
-		}
-		mutex_unlock(&sbi->s_lock);
-	}
-
 	buf->f_type = sb->s_magic;
 	buf->f_bsize = sbi->cluster_size;
 	buf->f_blocks = sbi->num_clusters - 2; /* clu 0 & 1 */
@@ -531,7 +522,6 @@ static int exfat_read_boot_sector(struct super_block *sb)
 	sbi->vol_flags = le16_to_cpu(p_boot->vol_flags);
 	sbi->vol_flags_persistent = sbi->vol_flags & (VOLUME_DIRTY | MEDIA_FAILURE);
 	sbi->clu_srch_ptr = EXFAT_FIRST_CLUSTER;
-	sbi->used_clusters = EXFAT_CLUSTERS_UNTRACKED;
 
 	/* check consistencies */
 	if ((u64)sbi->num_FAT_sectors << p_boot->sect_size_bits <