linux-kernel - [PATCH -mm] nilfs2: fix improper return values of nilfs_get

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Date:	Sun, 22 Mar 2009 21:11:03 +0900
From:	Ryusuke Konishi <konishi.ryusuke@....ntt.co.jp>
To:	Andrew Morton <akpm@...ux-foundation.org>
Cc:	linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org,
	Ryusuke Konishi <konishi.ryusuke@....ntt.co.jp>
Subject: [PATCH -mm] nilfs2: fix improper return values of nilfs_get_cpinfo ioctl

A few tool developers gave me requests for fixing inconvenient return
value of nilfs_get_cpinfo() ioctl; if the requested mode is
NILFS_SNAPSHOT and the specified start entry is not a snapshot, the
ioctl unnaturally returns one as the number of acquired snapshot item.

In addition, the ioctl function returns an ENOENT error for
checkpoints within blocks deleted by garbage collection.

These behaviors require corrections for programs which enumerate
snapshots.  This resolves the inconvenience by changing the return
values to zero for the above cases.

Signed-off-by: Ryusuke Konishi <konishi.ryusuke@....ntt.co.jp>
---
 fs/nilfs2/cpfile.c |   30 ++++++++++++++++++------------
 1 files changed, 18 insertions(+), 12 deletions(-)

diff --git a/fs/nilfs2/cpfile.c b/fs/nilfs2/cpfile.c
index 2e3b066..e90b60d 100644
--- a/fs/nilfs2/cpfile.c
+++ b/fs/nilfs2/cpfile.c
@@ -434,7 +434,7 @@ static ssize_t nilfs_cpfile_do_get_ssinfo(struct inode *cpfile, __u64 *cnop,
 	__u64 curr = *cnop, next;
 	unsigned long curr_blkoff, next_blkoff;
 	void *kaddr;
-	int n, ret;
+	int n = 0, ret;
 
 	down_read(&NILFS_MDT(cpfile)->mi_sem);
 
@@ -458,27 +458,33 @@ static ssize_t nilfs_cpfile_do_get_ssinfo(struct inode *cpfile, __u64 *cnop,
 
 	curr_blkoff = nilfs_cpfile_get_blkoff(cpfile, curr);
 	ret = nilfs_cpfile_get_checkpoint_block(cpfile, curr, 0, &bh);
-	if (ret < 0)
+	if (unlikely(ret < 0)) {
+		if (ret == -ENOENT)
+			ret = 0; /* No snapshots (started from a hole block) */
 		goto out;
+	}
 	kaddr = kmap_atomic(bh->b_page, KM_USER0);
-	for (n = 0; n < nci; n++) {
-		cp = nilfs_cpfile_block_get_checkpoint(
-			cpfile, curr, bh, kaddr);
-		nilfs_cpfile_checkpoint_to_cpinfo(cpfile, cp, &ci[n]);
-		next = le64_to_cpu(cp->cp_snapshot_list.ssl_next);
-		if (next == 0) {
-			curr = ~(__u64)0; /* Terminator */
-			n++;
+	while (n < nci) {
+		cp = nilfs_cpfile_block_get_checkpoint(cpfile, curr, bh, kaddr);
+		curr = ~(__u64)0; /* Terminator */
+		if (unlikely(nilfs_checkpoint_invalid(cp) ||
+			     !nilfs_checkpoint_snapshot(cp)))
 			break;
-		}
+		nilfs_cpfile_checkpoint_to_cpinfo(cpfile, cp, &ci[n++]);
+		next = le64_to_cpu(cp->cp_snapshot_list.ssl_next);
+		if (next == 0)
+			break; /* reach end of the snapshot list */
+
 		next_blkoff = nilfs_cpfile_get_blkoff(cpfile, next);
 		if (curr_blkoff != next_blkoff) {
 			kunmap_atomic(kaddr, KM_USER0);
 			brelse(bh);
 			ret = nilfs_cpfile_get_checkpoint_block(cpfile, next,
 								0, &bh);
-			if (ret < 0)
+			if (unlikely(ret < 0)) {
+				WARN_ON(ret == -ENOENT);
 				goto out;
+			}
 			kaddr = kmap_atomic(bh->b_page, KM_USER0);
 		}
 		curr = next;
-- 
1.5.6.5

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/