public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@caldera.de>
To: torvalds@transmeta.com
Cc: alan@redhat.com, linux-kernel@vger.kernel.org,
	linux-privs-discuss@sourceforge.net
Subject: Re: [Linux-privs-discuss] [PATCH] fix permission checks for executables
Date: Wed, 8 Aug 2001 19:32:28 +0200	[thread overview]
Message-ID: <20010808193228.A22007@caldera.de> (raw)
In-Reply-To: <20010808182219.A12652@caldera.de>
In-Reply-To: <20010808182219.A12652@caldera.de>; from hch@caldera.de on Wed, Aug 08, 2001 at 06:22:19PM +0200

On Wed, Aug 08, 2001 at 06:22:19PM +0200, Christoph Hellwig wrote:
> Hi Linux,
> 
> vfs_permission in the Linux 2.4 series tries to check for
> CAP_DAC_OVERRIDE if the modes didn't match.  This means even
> for an file without executable bits set at all, root will
> be reported that it is.   I've actually found one apllication
> (scomail under linux-abi) that fails because of this, besides
> not matching my reading of Posix 1003.1e.
> 
> Of the operating systems with capabilty-like features at least
> OpenUNIX gets it right, of the others at least OpenServer and
> 4.4BSD, but these semantics seem natural to me anyway..

Andreas Gruenbacher pointed out that it is much leaner to use
MAY_READ¸MAY_EXEC and MAY_WRITE instead of abusing the stat-macros.

Here is a patch that changes the complete function to use these
and also cleans up and clarifies the comments.

	Christoph


--- linux.really_plain/fs/namei.c	Wed Aug  8 17:56:58 2001
+++ linux.plain/fs/namei.c	Wed Aug  8 19:27:19 2001
@@ -140,7 +140,7 @@
 }
 
 /*
- *	permission()
+ *	vfs_permission()
  *
  * is used to check for read/write/execute permissions on a file.
  * We use "fsuid" for this, letting us set arbitrary permissions
@@ -151,24 +151,40 @@
 {
 	int mode = inode->i_mode;
 
-	if ((mask & S_IWOTH) && IS_RDONLY(inode) &&
-		 (S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode)))
-		return -EROFS; /* Nobody gets write access to a read-only fs */
-
-	if ((mask & S_IWOTH) && IS_IMMUTABLE(inode))
-		return -EACCES; /* Nobody gets write access to an immutable file */
+	if (mask & MAY_WRITE) {
+		/*
+		 * Nobody gets write access to a read-only fs.
+		 */
+		if (IS_RDONLY(inode) &&
+		    (S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode)))
+			return -EROFS;
+
+		/*
+		 * Nobody gets write access to an immutable file.
+		 */
+		if (IS_IMMUTABLE(inode))
+			return -EACCES;
+	}
 
 	if (current->fsuid == inode->i_uid)
 		mode >>= 6;
 	else if (in_group_p(inode->i_gid))
 		mode >>= 3;
 
-	if (((mode & mask & S_IRWXO) == mask) || capable(CAP_DAC_OVERRIDE))
+	if (((mode & mask & (MAY_READ|MAY_WRITE|MAY_EXEC)) == mask))
 		return 0;
 
-	/* read and search access */
-	if ((mask == S_IROTH) ||
-	    (S_ISDIR(inode->i_mode)  && !(mask & ~(S_IROTH | S_IXOTH))))
+	/*
+	 * Only read/write DACs are overridable.
+	 */
+	if ((mask & (MAY_READ|MAY_WRITE)) || S_ISDIR(inode->i_mode))
+		if (capable(CAP_DAC_OVERRIDE))
+			return 0;
+
+	/*
+	 * Searching includes executable on directories, else just read.
+	 */
+	if (mask == MAY_READ || (S_ISDIR(inode->i_mode) && !(mask & MAY_WRITE)))
 		if (capable(CAP_DAC_READ_SEARCH))
 			return 0;
 

      reply	other threads:[~2001-08-08 17:33 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2001-08-08 16:22 [PATCH] fix permission checks for executables Christoph Hellwig
2001-08-08 17:32 ` Christoph Hellwig [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20010808193228.A22007@caldera.de \
    --to=hch@caldera.de \
    --cc=alan@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-privs-discuss@sourceforge.net \
    --cc=torvalds@transmeta.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox