All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andreas Gruenbacher <andreas.gruenbacher@gmail.com>
To: linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
	linux-nfs@vger.kernel.org
Cc: Andreas Gruenbacher <agruen@kernel.org>
Subject: [RFC v3 27/45] richacl: Propagate everyone@ permissions to other aces
Date: Fri, 24 Apr 2015 13:04:24 +0200	[thread overview]
Message-ID: <93925a334bfd554616ab0ff9986dd8ddf88d1e9a.1429868795.git.agruenba@redhat.com> (raw)
In-Reply-To: <cover.1429868795.git.agruenba@redhat.com>
In-Reply-To: <cover.1429868795.git.agruenba@redhat.com>

The trailing everyone@ allow ace can grant permissions to all file
classes including the owner and group class.  Before we can apply the
other mask to this entry to turn it into an "other class" entry, we need
to ensure that members of the owner or group class will not lose any
permissions from that ace.

Conceptually, we do this by inserting additional <who>:<allow>::allow
entries before the trailing everyone@ allow ace with the same
permissions as the trailing everyone@ allow ace for owner@, group@, and
all explicitly mentioned users and groups.  (In practice, we will rarely
need to insert any additional aces in this step.)

Signed-off-by: Andreas Gruenbacher <agruen@kernel.org>
---
 fs/richacl_compat.c | 190 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 190 insertions(+)

diff --git a/fs/richacl_compat.c b/fs/richacl_compat.c
index 63f4373..f43d007 100644
--- a/fs/richacl_compat.c
+++ b/fs/richacl_compat.c
@@ -224,3 +224,193 @@ richacl_move_everyone_aces_down(struct richacl_alloc *x)
 	}
 	return 0;
 }
+
+/**
+ * __richacl_propagate_everyone  -  propagate everyone@ permissions up for @who
+ * @x:		acl and number of allocated entries
+ * @who:	identifier to propagate permissions for
+ * @allow:	permissions to propagate up
+ *
+ * Propagate the permissions in @allow up from the end of the acl to the start
+ * for the specified principal @who.
+ *
+ * The simplest possible approach to achieve this would be to insert a
+ * "<who>:<allow>::allow" ace before the final everyone@ allow ace.  Since this
+ * would often result in aces which are not needed or which could be merged
+ * with an existing ace, we make the following optimizations:
+ *
+ *   - We go through the acl and determine which permissions are already
+ *     allowed or denied to @who, and we remove those permissions from
+ *     @allow.
+ *
+ *   - If the acl contains an allow ace for @who and no aces after this entry
+ *     deny permissions in @allow, we add the permissions in @allow to this
+ *     ace.  (Propagating permissions across a deny ace which can match the
+ *     process can elevate permissions.)
+ *
+ * This transformation does not alter the permissions that the acl grants.
+ */
+static int
+__richacl_propagate_everyone(struct richacl_alloc *x, struct richace *who,
+			  unsigned int allow)
+{
+	struct richace *allow_last = NULL, *ace;
+
+	/*
+	 * Remove the permissions from allow that are already determined for
+	 * this who value, and figure out if there is an allow entry for
+	 * this who value that is "reachable" from the trailing everyone@
+	 * allow ace.
+	 */
+	richacl_for_each_entry(ace, x->acl) {
+		if (richace_is_inherit_only(ace))
+			continue;
+		if (richace_is_allow(ace)) {
+			if (richace_is_same_identifier(ace, who)) {
+				allow &= ~ace->e_mask;
+				allow_last = ace;
+			}
+		} else if (richace_is_deny(ace)) {
+			if (richace_is_same_identifier(ace, who))
+				allow &= ~ace->e_mask;
+			else if (allow & ace->e_mask)
+				allow_last = NULL;
+		}
+	}
+	if (allow) {
+		if (allow_last)
+			return richace_change_mask(x, &allow_last,
+						   allow_last->e_mask | allow);
+		else {
+			struct richace who_copy;
+
+			ace = x->acl->a_entries + x->acl->a_count - 1;
+			memcpy(&who_copy, who, sizeof(struct richace));
+			if (richacl_insert_entry(x, &ace))
+				return -1;
+			memcpy(ace, &who_copy, sizeof(struct richace));
+			ace->e_type = RICHACE_ACCESS_ALLOWED_ACE_TYPE;
+			richace_clear_inheritance_flags(ace);
+			ace->e_mask = allow;
+		}
+	}
+	return 0;
+}
+
+/**
+ * richacl_propagate_everyone  -  propagate everyone@ permissions up the acl
+ * @x:		acl and number of allocated entries
+ *
+ * Make sure that owner@, group@, and all other users and groups mentioned in
+ * the acl will not lose any permissions when finally applying the other mask
+ * to the everyone@ allow ace at the end of the acl.  If one of those
+ * principals is not granted some of those permissions, insert an additional
+ * allow ace for that principal into the acl before the final everyone@ allow
+ * ace.
+ *
+ * For example, the following acl implicitly grants everyone rwx access:
+ *
+ *    joe:r::allow
+ *    everyone@:rwx::allow
+ *
+ * When applying mode 0660 to this acl, owner@ and group@ would lose rw access,
+ * and joe would lose w access even though the mode does not exclude those
+ * permissions.  To fix this problem, we insert additional allow aces into the
+ * acl.  The result after applying mode 0660 is:
+ *
+ *    joe:rw::allow
+ *    owner@:rw::allow
+ *    group@:rw::allow
+ *
+ * Deny aces complicate the matter.  For example, the following acl grants
+ * everyone but joe write access:
+ *
+ *    joe:w::deny
+ *    everyone@:rwx::allow
+ *
+ * When applying mode 0660 to this acl, owner@ and group@ would lose rw access,
+ * and joe would lose r access.  The result after inserting additional allow
+ * aces and applying mode 0660 is:
+ *
+ *    joe:w::deny
+ *    owner@:rw::allow
+ *    group@:rw::allow
+ *    joe:r::allow
+ *
+ * Inserting the additional aces does not alter the permissions that the acl
+ * grants.
+ */
+static int
+richacl_propagate_everyone(struct richacl_alloc *x)
+{
+	struct richace who = { .e_flags = RICHACE_SPECIAL_WHO };
+	struct richacl *acl = x->acl;
+	struct richace *ace;
+	unsigned int owner_allow, group_allow;
+
+	/*
+	 * If the owner mask contains permissions which are not in
+	 * the group mask, the group mask contains permissions which
+	 * are not in the other mask, or the owner class contains
+	 * permissions which are not in the other mask,	we may need
+	 * to propagate permissions up from the everyone@ allow ace.
+	 * The third condition is implied by the first two.
+	 */
+	if (!((acl->a_owner_mask & ~acl->a_group_mask) ||
+	      (acl->a_group_mask & ~acl->a_other_mask)))
+		return 0;
+	if (!acl->a_count)
+		return 0;
+	ace = acl->a_entries + acl->a_count - 1;
+	if (richace_is_inherit_only(ace) || !richace_is_everyone(ace))
+		return 0;
+	if (!(ace->e_mask & ~(acl->a_group_mask & acl->a_other_mask)))
+		/*
+		 * None of the allowed permissions will get masked.
+		 */
+		return 0;
+	owner_allow = (ace->e_mask & acl->a_owner_mask) |
+		      RICHACE_POSIX_OWNER_ALLOWED;
+	group_allow = ace->e_mask & acl->a_group_mask;
+
+	/* Propagate everyone@ permissions through to owner@. */
+	if (owner_allow & ~(acl->a_group_mask & acl->a_other_mask)) {
+		who.e_id.special = RICHACE_OWNER_SPECIAL_ID;
+		if (__richacl_propagate_everyone(x, &who, owner_allow))
+			return -1;
+		acl = x->acl;
+	}
+
+	if (group_allow & ~acl->a_other_mask) {
+		int n;
+
+		/* Propagate everyone@ permissions through to group@. */
+		who.e_id.special = RICHACE_GROUP_SPECIAL_ID;
+		if (__richacl_propagate_everyone(x, &who, group_allow))
+			return -1;
+		acl = x->acl;
+
+		/*
+		 * Start from the entry before the trailing everyone@ allow
+		 * entry. We will not hit everyone@ entries in the loop.
+		 */
+		for (n = acl->a_count - 2; n != -1; n--) {
+			ace = acl->a_entries + n;
+
+			if (richace_is_inherit_only(ace) ||
+			    richace_is_owner(ace) ||
+			    richace_is_group(ace))
+				continue;
+			if (richace_is_allow(ace) || richace_is_deny(ace)) {
+				/*
+				 * Any inserted entry will end up below the
+				 * current entry
+				 */
+				if (__richacl_propagate_everyone(x, ace,
+								 group_allow))
+					return -1;
+			}
+		}
+	}
+	return 0;
+}
-- 
2.1.0


  parent reply	other threads:[~2015-04-24 11:12 UTC|newest]

Thread overview: 154+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-04-24 11:03 [RFC v3 00/45] Richacls Andreas Gruenbacher
2015-04-24 11:03 ` Andreas Gruenbacher
2015-04-24 11:03 ` [RFC v3 01/45] vfs: Minor documentation fix Andreas Gruenbacher
2015-04-24 11:03 ` [RFC v3 02/45] uapi: Remove kernel internal declaration Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 03/45] nfsd: Checking for acl support does not require fetching any acls Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 04/45] vfs: Shrink struct posix_acl Andreas Gruenbacher
2015-04-25  7:56   ` Rasmus Villemoes
2015-04-27 10:27     ` Andreas Grünbacher
2015-04-24 11:04 ` [RFC v3 05/45] vfs: Add IS_ACL() and IS_RICHACL() tests Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 06/45] vfs: Add MAY_CREATE_FILE and MAY_CREATE_DIR permission flags Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 07/45] vfs: Add MAY_DELETE_SELF and MAY_DELETE_CHILD " Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 08/45] vfs: Make the inode passed to inode_change_ok non-const Andreas Gruenbacher
2015-04-24 11:04   ` Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 09/45] vfs: Add permission flags for setting file attributes Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 10/45] richacl: In-memory representation and helper functions Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 11/45] richacl: Permission mapping functions Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 12/45] richacl: Compute maximum file masks from an acl Andreas Gruenbacher
2015-05-12 21:44   ` J. Bruce Fields
2015-04-24 11:04 ` [RFC v3 13/45] richacl: Update the file masks in chmod() Andreas Gruenbacher
2015-04-24 11:04   ` Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 14/45] richacl: Permission check algorithm Andreas Gruenbacher
2015-04-24 11:04   ` Andreas Gruenbacher
2015-05-22 21:08   ` J. Bruce Fields
2015-05-27 11:25     ` Andreas Grünbacher
2015-05-27 15:05       ` J. Bruce Fields
2015-04-24 11:04 ` [RFC v3 15/45] vfs: Cache base_acl objects in inodes Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 16/45] vfs: Cache richacl in struct inode Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 17/45] richacl: Create-time inheritance Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 18/45] richacl: Check if an acl is equivalent to a file mode Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 19/45] richacl: Also recognize nontrivial mode-equivalent acls Andreas Gruenbacher
2015-05-14 19:28   ` J. Bruce Fields
2015-05-15 20:04     ` J. Bruce Fields
2015-05-15 20:04       ` J. Bruce Fields
2015-05-15 20:51   ` J. Bruce Fields
2015-05-15 20:51     ` J. Bruce Fields
2015-05-27  9:24     ` Andreas Grünbacher
2015-05-27 15:06       ` J. Bruce Fields
2015-05-27 15:06         ` J. Bruce Fields
2015-04-24 11:04 ` [RFC v3 20/45] richacl: Automatic Inheritance Andreas Gruenbacher
2015-05-13 18:01   ` Frank Filz
2015-05-13 18:01     ` Frank Filz
2015-05-13 20:22     ` Andreas Grünbacher
2015-05-13 20:22       ` Andreas Grünbacher
2015-05-13 20:28       ` Jeremy Allison
2015-05-13 20:28         ` Jeremy Allison
2015-05-13 20:47         ` Andreas Grünbacher
2015-05-13 20:47           ` Andreas Grünbacher
2015-05-13 20:55           ` Jeremy Allison
2015-05-13 20:55             ` Jeremy Allison
2015-05-13 21:15             ` Andreas Grünbacher
2015-05-13 21:15               ` Andreas Grünbacher
2015-05-13 20:38       ` Frank Filz
2015-05-13 20:38         ` Frank Filz
2015-05-13 21:05         ` Andreas Grünbacher
2015-05-13 22:56           ` Frank Filz
2015-05-13 22:56             ` Frank Filz
2015-05-13 23:52             ` Andreas Grünbacher
2015-05-14 15:09   ` J. Bruce Fields
2015-05-14 15:09     ` J. Bruce Fields
2015-05-18 20:39     ` Andreas Grünbacher
2015-05-18 20:39       ` Andreas Grünbacher
2015-05-18 20:43       ` J. Bruce Fields
2015-05-18 20:43         ` J. Bruce Fields
2015-04-24 11:04 ` [RFC v3 21/45] richacl: xattr mapping functions Andreas Gruenbacher
2015-04-24 11:04   ` Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 22/45] vfs: Add richacl permission checking Andreas Gruenbacher
2015-04-24 11:04   ` Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 23/45] ext4: Add richacl support Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 24/45] ext4: Add richacl feature flag Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 25/45] richacl: acl editing helper functions Andreas Gruenbacher
2015-04-24 11:04   ` Andreas Gruenbacher
2015-05-15 21:33   ` J. Bruce Fields
2015-05-15 21:33     ` J. Bruce Fields
2015-04-24 11:04 ` [RFC v3 26/45] richacl: Move everyone@ aces down the acl Andreas Gruenbacher
2015-05-18 21:20   ` J. Bruce Fields
2015-04-24 11:04 ` Andreas Gruenbacher [this message]
2015-04-24 11:04 ` [RFC v3 28/45] richacl: Isolate the owner and group classes Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 29/45] richacl: Apply the file masks to a richacl Andreas Gruenbacher
2015-05-26 21:56   ` J. Bruce Fields
2015-05-26 21:56     ` J. Bruce Fields
2015-04-24 11:04 ` [RFC v3 30/45] richacl: Create richacl from mode values Andreas Gruenbacher
2015-05-28 19:24   ` J. Bruce Fields
2015-05-28 19:24     ` J. Bruce Fields
2015-05-28 19:25     ` J. Bruce Fields
2015-05-28 19:58       ` Andreas Grünbacher
2015-05-28 19:58         ` Andreas Grünbacher
2015-04-24 11:04 ` [RFC v3 31/45] richacl: Create acl with masks applied in richacl_from_mode() Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 32/45] nfsd: Remove dead declarations Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 33/45] nfsd: Keep list of acls to dispose of in compoundargs Andreas Gruenbacher
2015-05-28 19:52   ` J. Bruce Fields
2015-05-28 19:52     ` J. Bruce Fields
2015-05-28 20:28     ` Andreas Grünbacher
2015-05-28 20:28       ` Andreas Grünbacher
2015-04-24 11:04 ` [RFC v3 34/45] nfsd: Use richacls as internal acl representation Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 35/45] nfsd: Add richacl support Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 36/45] NFSv4: Fix GETATTR bitmap verification Andreas Gruenbacher
2015-05-28 20:33   ` J. Bruce Fields
2015-05-28 20:33     ` J. Bruce Fields
2015-05-28 20:50     ` Trond Myklebust
2015-05-28 20:50       ` Trond Myklebust
2015-05-28 21:05       ` Andreas Grünbacher
2015-05-28 21:05         ` Andreas Grünbacher
2015-05-28 21:40         ` Trond Myklebust
2015-05-28 21:40           ` Trond Myklebust
2015-05-28 21:55           ` Trond Myklebust
2015-05-28 21:55             ` Trond Myklebust
2015-05-28 22:09             ` Andreas Grünbacher
2015-05-28 22:09               ` Andreas Grünbacher
2015-05-28 22:24               ` Trond Myklebust
2015-05-28 22:24                 ` Trond Myklebust
2015-05-28 22:28                 ` Andreas Grünbacher
2015-05-28 21:15     ` Andreas Grünbacher
2015-05-28 21:28     ` Andreas Grünbacher
2015-05-28 21:28       ` Andreas Grünbacher
2015-04-24 11:04 ` [RFC v3 37/45] nfs/sunrpc: No more encode and decode function pointer casting Andreas Gruenbacher
2015-05-28 23:11   ` Trond Myklebust
2015-05-28 23:40     ` Andreas Grünbacher
2015-05-29  0:37       ` Trond Myklebust
2015-05-29  0:37         ` Trond Myklebust
2015-05-29 15:30         ` Andreas Grünbacher
2015-05-29 15:30           ` Andreas Grünbacher
2015-04-24 11:04 ` [RFC v3 38/45] nfs/sunrpc: Return status code from encode functions Andreas Gruenbacher
2015-05-28 23:39   ` Trond Myklebust
2015-05-28 23:39     ` Trond Myklebust
2015-04-24 11:04 ` [RFC v3 39/45] nfs3: Return posix acl encode errors Andreas Gruenbacher
2015-04-24 11:04   ` Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 40/45] nfs: Remove unused xdr page offsets in getacl/setacl arguments Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 41/45] rpc: Allow to demand-allocate pages to encode into Andreas Gruenbacher
2015-05-28 23:24   ` Trond Myklebust
2015-05-28 23:24     ` Trond Myklebust
2015-05-29 15:46     ` Andreas Grünbacher
2015-04-24 11:04 ` [RFC v3 42/45] nfs: Add richacl support Andreas Gruenbacher
2015-05-28 23:06   ` Trond Myklebust
2015-05-28 23:29     ` Andreas Grünbacher
2015-05-29 13:15     ` Trond Myklebust
2015-05-29 15:00       ` Andreas Grünbacher
2015-05-29 15:24         ` Trond Myklebust
2015-05-29 15:45           ` Andreas Grünbacher
2015-05-29 15:54             ` Trond Myklebust
2015-05-29 15:54               ` Trond Myklebust
2015-05-29 15:54           ` Frank Filz
2015-05-29 15:54             ` Frank Filz
2016-06-23 16:37   ` Weston Andros Adamson
2016-06-23 17:15     ` Frank Filz
2016-07-15  8:00     ` Andreas Grünbacher
2015-04-24 11:04 ` [RFC v3 43/45] uapi/nfs: Add NFSv4.1 ACL definitions Andreas Gruenbacher
2015-05-29 14:57   ` J. Bruce Fields
2015-04-24 11:04 ` [RFC v3 44/45] nfsd: Add support for the v4.1 dacl attribute Andreas Gruenbacher
2015-04-24 11:04 ` [RFC v3 45/45] nfs: " Andreas Gruenbacher
2015-05-23 15:50 ` [RFC v3 00/45] Richacls Jeremy Allison
2015-05-23 15:50   ` Jeremy Allison
2015-05-23 15:56   ` Andreas Grünbacher
2015-05-23 15:56     ` Andreas Grünbacher
2015-06-25 20:03   ` Stefan (metze) Metzmacher

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=93925a334bfd554616ab0ff9986dd8ddf88d1e9a.1429868795.git.agruenba@redhat.com \
    --to=andreas.gruenbacher@gmail.com \
    --cc=agruen@kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-nfs@vger.kernel.org \
    /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 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.