[PATCH RESEND] vfs: Return EINVAL for default SEEK_HOLE, SEEK_DATA implementation

Jan Kara <[email protected]>
Newsgroups gmane.comp.file-systems.jfs.general,gmane.linux.file-systems,gmane.comp.file-systems.reiserfs.general,gmane.linux.cluster.redhat.cluster.devel,gmane.comp.file-systems.ext4,gmane.comp.file-systems.ocfs2.devel
Message-ID <[email protected]>
Generic implementation of SEEK_HOLE & SEEK_DATA in
generic_file_llseek_size() and default_llseek() behaved as if everything
within i_size is data and everything beyond i_size is a hole. That makes
sense at the first sight (and definitely is a valid implementation of
the spec) but at the second sight it isn't very useful. If anyone
bothers with looking for holes / data, he should be better told we don't
really know so that he can fall back to his chosen backup strategy
(think of e.g. cp(1)).

This is a userspace API change however the kernel interface is there
only for two years so hopefully userspace is still prepared to handle
EINVAL return value.

Signed-off-by: Jan Kara <[email protected]>
---
 fs/read_write.c | 38 ++------------------------------------
 1 file changed, 2 insertions(+), 36 deletions(-)

Al, what do you think about this change?

diff --git a/fs/read_write.c b/fs/read_write.c
index edc5746a902a..71525d75742f 100644
--- a/fs/read_write.c
+++ b/fs/read_write.c
@@ -111,22 +111,8 @@ generic_file_llseek_size(struct file *file, loff_t offset, int whence,
 		spin_unlock(&file->f_lock);
 		return offset;
 	case SEEK_DATA:
-		/*
-		 * In the generic case the entire file is data, so as long as
-		 * offset isn't at the end of the file then the offset is data.
-		 */
-		if (offset >= eof)
-			return -ENXIO;
-		break;
 	case SEEK_HOLE:
-		/*
-		 * There is a virtual hole at the end of the file, so as long as
-		 * offset isn't i_size or larger, return i_size.
-		 */
-		if (offset >= eof)
-			return -ENXIO;
-		offset = eof;
-		break;
+		return -EINVAL;
 	}
 
 	return vfs_setpos(file, offset, maxsize);
@@ -214,28 +200,8 @@ loff_t default_llseek(struct file *file, loff_t offset, int whence)
 			offset += file->f_pos;
 			break;
 		case SEEK_DATA:
-			/*
-			 * In the generic case the entire file is data, so as
-			 * long as offset isn't at the end of the file then the
-			 * offset is data.
-			 */
-			if (offset >= inode->i_size) {
-				retval = -ENXIO;
-				goto out;
-			}
-			break;
 		case SEEK_HOLE:
-			/*
-			 * There is a virtual hole at the end of the file, so
-			 * as long as offset isn't i_size or larger, return
-			 * i_size.
-			 */
-			if (offset >= inode->i_size) {
-				retval = -ENXIO;
-				goto out;
-			}
-			offset = inode->i_size;
-			break;
+			return -EINVAL;
 	}
 	retval = -EINVAL;
 	if (offset >= 0 || unsigned_offsets(file)) {
-- 
1.8.1.4


------------------------------------------------------------------------------
Meet PCI DSS 3.0 Compliance Requirements with EventLog Analyzer
Achieve PCI DSS 3.0 Compliant Status with Out-of-the-box PCI DSS Reports
Are you Audit-Ready for PCI DSS 3.0 Compliance? Download White paper
Comply to PCI DSS 3.0 Requirement 10 and 11.5 with EventLog Analyzer
http://pubads.g.doubleclick.net/gampad/clk?id=154622311&iu=/4140/ostg.clktrk
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.