All of lore.kernel.org
 help / color / mirror / Atom feed
* request for inclusion of 2 xfs patches
@ 2017-05-29  9:27 Nikolay Borisov
  2017-05-29  9:28 ` Nikolay Borisov
  0 siblings, 1 reply; 3+ messages in thread
From: Nikolay Borisov @ 2017-05-29  9:27 UTC (permalink / raw)
  To: stable; +Cc: sandeen, linux-xfs, Brian Foster

[-- Attachment #1: Type: text/plain, Size: 472 bytes --]

Hello, 

I'd like to request the following two patches to be 
included in 4.4 stable: 

0facef7fb053 ("xfs: in _attrlist_by_handle, copy the cursor back to userspace")

2a6fba6d2311 ("xfs: only return -errno or success from attr ->put_listent")

The fix a failure in xfstest/269 and also enable the xfs' XFS_IOC_ATTRLIST_BY_HANDLE
ioctl to function properly. The first commit applies cleanly, whereas the second has 
a minor conflict due t spacing. Backport is attached. 

[-- Attachment #2: 0002-xfs-only-return-errno-or-success-from-attr-put_liste.patch --]
[-- Type: text/x-patch, Size: 3845 bytes --]

>From 42432ed1a893868c62df93005afab8a92de95986 Mon Sep 17 00:00:00 2001
From: Eric Sandeen <sandeen@sandeen.net>
Date: Wed, 6 Apr 2016 07:57:18 +1000
Subject: [PATCH 2/2] xfs: only return -errno or success from attr
 ->put_listent

Today, the put_listent formatters return either 1 or 0; if
they return 1, some callers treat this as an error and return
it up the stack, despite "1" not being a valid (negative)
error code.

The intent seems to be that if the input buffer is full,
we set seen_enough or set count = -1, and return 1;
but some callers check the return before checking the
seen_enough or count fields of the context.

Fix this by only returning non-zero for actual errors
encountered, and rely on the caller to first check the
return value, then check the values in the context to
decide what to do.

Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Acked-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
---
 fs/xfs/xfs_attr.h      |  1 +
 fs/xfs/xfs_attr_list.c |  8 +++-----
 fs/xfs/xfs_xattr.c     | 15 ++++++++++-----
 3 files changed, 14 insertions(+), 10 deletions(-)

diff --git a/fs/xfs/xfs_attr.h b/fs/xfs/xfs_attr.h
index dd4824589470..234331227c0c 100644
--- a/fs/xfs/xfs_attr.h
+++ b/fs/xfs/xfs_attr.h
@@ -112,6 +112,7 @@ typedef struct attrlist_cursor_kern {
  *========================================================================*/
 
 
+/* Return 0 on success, or -errno; other state communicated via *context */
 typedef int (*put_listent_func_t)(struct xfs_attr_list_context *, int,
 			      unsigned char *, int, int, unsigned char *);
 
diff --git a/fs/xfs/xfs_attr_list.c b/fs/xfs/xfs_attr_list.c
index 4fa14820e2e2..c8be331a3196 100644
--- a/fs/xfs/xfs_attr_list.c
+++ b/fs/xfs/xfs_attr_list.c
@@ -108,16 +108,14 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 					   (int)sfe->namelen,
 					   (int)sfe->valuelen,
 					   &sfe->nameval[sfe->namelen]);
-
+			if (error)
+				return error;
 			/*
 			 * Either search callback finished early or
 			 * didn't fit it all in the buffer after all.
 			 */
 			if (context->seen_enough)
 				break;
-
-			if (error)
-				return error;
 			sfe = XFS_ATTR_SF_NEXTENTRY(sfe);
 		}
 		trace_xfs_attr_list_sf_all(context);
@@ -581,7 +579,7 @@ xfs_attr_put_listent(
 		trace_xfs_attr_list_full(context);
 		alist->al_more = 1;
 		context->seen_enough = 1;
-		return 1;
+		return 0;
 	}
 
 	aep = (attrlist_ent_t *)&context->alist[context->firstu];
diff --git a/fs/xfs/xfs_xattr.c b/fs/xfs/xfs_xattr.c
index 839b35ca21c6..e6dae28dfa1a 100644
--- a/fs/xfs/xfs_xattr.c
+++ b/fs/xfs/xfs_xattr.c
@@ -180,7 +180,7 @@ xfs_xattr_put_listent(
 	arraytop = context->count + prefix_len + namelen + 1;
 	if (arraytop > context->firstu) {
 		context->count = -1;	/* insufficient space */
-		return 1;
+		return 0;
 	}
 	offset = (char *)context->alist + context->count;
 	strncpy(offset, xfs_xattr_prefix(flags), prefix_len);
@@ -222,12 +222,15 @@ list_one_attr(const char *name, const size_t len, void *data,
 }
 
 ssize_t
-xfs_vn_listxattr(struct dentry *dentry, char *data, size_t size)
+xfs_vn_listxattr(
+	struct dentry	*dentry,
+	char		*data,
+	size_t		size)
 {
 	struct xfs_attr_list_context context;
 	struct attrlist_cursor_kern cursor = { 0 };
-	struct inode		*inode = d_inode(dentry);
-	int			error;
+	struct inode	*inode = d_inode(dentry);
+	int		error;
 
 	/*
 	 * First read the regular on-disk attributes.
@@ -245,7 +248,9 @@ xfs_vn_listxattr(struct dentry *dentry, char *data, size_t size)
 	else
 		context.put_listent = xfs_xattr_put_listent_sizes;
 
-	xfs_attr_list_int(&context);
+	error = xfs_attr_list_int(&context);
+	if (error)
+		return error;
 	if (context.count < 0)
 		return -ERANGE;
 
-- 
2.7.4


^ permalink raw reply related	[flat|nested] 3+ messages in thread

* Re: request for inclusion of 2 xfs patches
  2017-05-29  9:27 request for inclusion of 2 xfs patches Nikolay Borisov
@ 2017-05-29  9:28 ` Nikolay Borisov
  2017-06-05 15:09   ` Greg KH
  0 siblings, 1 reply; 3+ messages in thread
From: Nikolay Borisov @ 2017-05-29  9:28 UTC (permalink / raw)
  To: stable; +Cc: sandeen, linux-xfs, Brian Foster



On 29.05.2017 12:27, Nikolay Borisov wrote:
> Hello, 
> 
> I'd like to request the following two patches to be 
> included in 4.4 stable: 
> 
> 0facef7fb053 ("xfs: in _attrlist_by_handle, copy the cursor back to userspace")
> 
> 2a6fba6d2311 ("xfs: only return -errno or success from attr ->put_listent")
> 
> The fix a failure in xfstest/269 and also enable the xfs' XFS_IOC_ATTRLIST_BY_HANDLE
> ioctl to function properly. The first commit applies cleanly, whereas the second has 
> a minor conflict due t spacing. Backport is attached. 
> 

Please omit my Acked-by tagged when applying the backport

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: request for inclusion of 2 xfs patches
  2017-05-29  9:28 ` Nikolay Borisov
@ 2017-06-05 15:09   ` Greg KH
  0 siblings, 0 replies; 3+ messages in thread
From: Greg KH @ 2017-06-05 15:09 UTC (permalink / raw)
  To: Nikolay Borisov; +Cc: stable, sandeen, linux-xfs, Brian Foster

On Mon, May 29, 2017 at 12:28:19PM +0300, Nikolay Borisov wrote:
> 
> 
> On 29.05.2017 12:27, Nikolay Borisov wrote:
> > Hello, 
> > 
> > I'd like to request the following two patches to be 
> > included in 4.4 stable: 
> > 
> > 0facef7fb053 ("xfs: in _attrlist_by_handle, copy the cursor back to userspace")
> > 
> > 2a6fba6d2311 ("xfs: only return -errno or success from attr ->put_listent")
> > 
> > The fix a failure in xfstest/269 and also enable the xfs' XFS_IOC_ATTRLIST_BY_HANDLE
> > ioctl to function properly. The first commit applies cleanly, whereas the second has 
> > a minor conflict due t spacing. Backport is attached. 
> > 
> 
> Please omit my Acked-by tagged when applying the backport

Heh, now done, both queue up, thanks.

greg k-h

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2017-06-05 15:10 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-05-29  9:27 request for inclusion of 2 xfs patches Nikolay Borisov
2017-05-29  9:28 ` Nikolay Borisov
2017-06-05 15:09   ` Greg KH

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.