linux-nfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "J. Bruce Fields" <bfields@fieldses.org>
To: Andreas Gruenbacher <agruenba@redhat.com>
Cc: linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
	linux-nfs@vger.kernel.org, linux-api@vger.kernel.org,
	linux-cifs@vger.kernel.org,
	linux-security-module@vger.kernel.org,
	Andreas Gruenbacher <agruen@kernel.org>
Subject: Re: [RFC v7 26/41] richacl: Apply the file masks to a richacl
Date: Wed, 23 Sep 2015 15:18:32 -0400	[thread overview]
Message-ID: <20150923191832.GA29577@fieldses.org> (raw)
In-Reply-To: <20150922191108.GC19127@fieldses.org>

On Tue, Sep 22, 2015 at 03:11:08PM -0400, bfields wrote:
> On Sat, Sep 05, 2015 at 12:27:21PM +0200, Andreas Gruenbacher wrote:
> > Put all the pieces of the acl transformation puzzle together for
> > computing a richacl which has the file masks "applied" so that the
> > standard nfsv4 access check algorithm can be used on the richacl.
> > 
> > Signed-off-by: Andreas Gruenbacher <agruen@kernel.org>
> > ---
> >  fs/richacl_compat.c     | 110 ++++++++++++++++++++++++++++++++++++++++++++++++
> >  include/linux/richacl.h |   3 ++
> >  2 files changed, 113 insertions(+)
> > 
> > diff --git a/fs/richacl_compat.c b/fs/richacl_compat.c
> > index 412844c..9681efe 100644
> > --- a/fs/richacl_compat.c
> > +++ b/fs/richacl_compat.c
> > @@ -730,3 +730,113 @@ richacl_isolate_group_class(struct richacl_alloc *alloc)
> >  	}
> >  	return 0;
> >  }
> > +
> > +/**
> > + * __richacl_apply_masks  -  apply the file masks to all aces
> > + * @alloc:	acl and number of allocated entries
> > + *
> > + * Apply the owner mask to owner@ aces, the other mask to
> > + * everyone@ aces, and the group mask to all other aces.
> > + *
> > + * The previous transformations have brought the acl into a
> > + * form in which applying the masks will not lead to the
> > + * accidental loss of permissions anymore.
> > + */
> > +static int
> > +__richacl_apply_masks(struct richacl_alloc *alloc, kuid_t owner)
> > +{
> > +	struct richace *ace;
> > +
> > +	richacl_for_each_entry(ace, alloc->acl) {
> > +		unsigned int mask;
> > +
> > +		if (richace_is_inherit_only(ace) || !richace_is_allow(ace))
> > +			continue;
> > +		if (richace_is_owner(ace) ||
> > +		    (richace_is_unix_user(ace) && uid_eq(owner, ace->e_id.uid)))
> > +			mask = alloc->acl->a_owner_mask;
> 
> Is treating matching user aces like owner aces what you intended to do,
> and if so, why?

That does look wrong to me; in an example like:

	file owner bfields
	mask 0700, not WRITE_THROUGH
	bfields:rwx::allow

The permission algorithm grants nothing to anyone, but it looks to me
like richacl_apply_masks just leaves this as

	bfields:rwx::allow

but it would give the right result (an empty/deny-all ACL) if it weren't
for this odd case here.

--b.

> 
> --b.
> 
> > +		else if (richace_is_everyone(ace))
> > +			mask = alloc->acl->a_other_mask;
> > +		else
> > +			mask = alloc->acl->a_group_mask;
> > +		if (richace_change_mask(alloc, &ace, ace->e_mask & mask))
> > +			return -1;
> > +	}
> > +	return 0;
> > +}
> > +
> > +/**
> > + * richacl_apply_masks  -  apply the masks to the acl
> > + *
> > + * Transform @acl so that the standard NFSv4 permission check algorithm (which
> > + * is not aware of file masks) will compute the same access decisions as the
> > + * richacl permission check algorithm (which looks at the acl and the file
> > + * masks).
> > + *
> > + * This algorithm is split into several steps:
> > + *
> > + *   - Move everyone@ aces to the end of the acl.  This simplifies the other
> > + *     transformations, and allows the everyone@ allow ace at the end of the
> > + *     acl to eventually allow permissions to the other class only.
> > + *
> > + *   - Propagate everyone@ permissions up the acl.  This transformation makes
> > + *     sure that the owner and group class aces won't lose any permissions when
> > + *     we apply the other mask to the everyone@ allow ace at the end of the acl.
> > + *
> > + *   - Apply the file masks to all aces.
> > + *
> > + *   - Make sure the owner is granted the owner mask permissions.
> > + *
> > + *   - Make sure everyone is granted the other mask permissions.
> > + *
> > + *   - Make sure that the owner is not granted any permissions beyond the owner
> > + *     mask from group class aces or from everyone@.
> > + *
> > + *   - Make sure that the group class is not granted any permissions from
> > + *     everyone@.
> > + *
> > + * The algorithm is exact except for richacls which cannot be represented as an
> > + * acl alone: for example, given this acl:
> > + *
> > + *    group@:rw::allow
> > + *
> > + * when file masks corresponding to mode 0600 are applied, the owner would only
> > + * get rw access if he is a member of the owning group.  This algorithm would
> > + * produce an empty acl in this case.  We fix this case by modifying
> > + * richacl_permission() so that the group mask is always applied to group class
> > + * aces.  With this fix, the owner would not have any access (beyond the
> > + * implicit permissions always granted to owners).
> > + *
> > + * NOTE: Depending on the acl and file masks, this algorithm can increase the
> > + * number of aces by almost a factor of three in the worst case. This may make
> > + * the acl too large for some purposes.
> > + */
> > +int
> > +richacl_apply_masks(struct richacl **acl, kuid_t owner)
> > +{
> > +	if ((*acl)->a_flags & RICHACL_MASKED) {
> > +		struct richacl_alloc alloc = {
> > +			.acl = richacl_clone(*acl, GFP_KERNEL),
> > +			.count = (*acl)->a_count,
> > +		};
> > +		if (!alloc.acl)
> > +			return -ENOMEM;
> > +
> > +		if (richacl_move_everyone_aces_down(&alloc) ||
> > +		    richacl_propagate_everyone(&alloc) ||
> > +		    __richacl_apply_masks(&alloc, owner) ||
> > +		    richacl_set_owner_permissions(&alloc) ||
> > +		    richacl_set_other_permissions(&alloc) ||
> > +		    richacl_isolate_owner_class(&alloc) ||
> > +		    richacl_isolate_group_class(&alloc)) {
> > +			richacl_put(alloc.acl);
> > +			return -ENOMEM;
> > +		}
> > +
> > +		alloc.acl->a_flags &= ~(RICHACL_WRITE_THROUGH | RICHACL_MASKED);
> > +		richacl_put(*acl);
> > +		*acl = alloc.acl;
> > +	}
> > +	return 0;
> > +}
> > +EXPORT_SYMBOL_GPL(richacl_apply_masks);
> > diff --git a/include/linux/richacl.h b/include/linux/richacl.h
> > index 832b06c..a945f3c 100644
> > --- a/include/linux/richacl.h
> > +++ b/include/linux/richacl.h
> > @@ -332,4 +332,7 @@ extern struct richacl *richacl_inherit(const struct richacl *, int);
> >  extern int richacl_permission(struct inode *, const struct richacl *, int);
> >  extern struct richacl *richacl_create(struct inode *, struct inode *);
> >  
> > +/* richacl_compat.c */
> > +extern int richacl_apply_masks(struct richacl **, kuid_t);
> > +
> >  #endif /* __RICHACL_H */
> > -- 
> > 2.4.3
> > 
> > --
> > To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
> > the body of a message to majordomo@vger.kernel.org
> > More majordomo info at  http://vger.kernel.org/majordomo-info.html

  reply	other threads:[~2015-09-23 19:18 UTC|newest]

Thread overview: 109+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-09-05 10:26 [RFC v7 00/41] Richacls Andreas Gruenbacher
2015-09-05 10:26 ` [RFC v7 01/41] vfs: Add IS_ACL() and IS_RICHACL() tests Andreas Gruenbacher
2015-09-05 10:26 ` [RFC v7 02/41] vfs: Add MAY_CREATE_FILE and MAY_CREATE_DIR permission flags Andreas Gruenbacher
2015-09-05 10:26 ` [RFC v7 03/41] vfs: Add MAY_DELETE_SELF and MAY_DELETE_CHILD " Andreas Gruenbacher
2015-09-06  8:14   ` [PATCH] " Andreas Gruenbacher
2015-09-11 20:30     ` J. Bruce Fields
2015-09-05 10:26 ` [RFC v7 04/41] vfs: Make the inode passed to inode_change_ok non-const Andreas Gruenbacher
2015-09-05 10:27 ` [RFC v7 05/41] vfs: Add permission flags for setting file attributes Andreas Gruenbacher
2015-09-05 10:27 ` [RFC v7 06/41] richacl: In-memory representation and helper functions Andreas Gruenbacher
2015-09-05 10:27 ` [RFC v7 07/41] richacl: Permission mapping functions Andreas Gruenbacher
2015-09-05 10:27 ` [RFC v7 08/41] richacl: Compute maximum file masks from an acl Andreas Gruenbacher
2015-09-05 10:27 ` [RFC v7 09/41] richacl: Update the file masks in chmod() Andreas Gruenbacher
2015-09-11 20:35   ` J. Bruce Fields
2015-09-05 10:27 ` [RFC v7 10/41] richacl: Permission check algorithm Andreas Gruenbacher
2015-09-11 21:16   ` J. Bruce Fields
2015-09-11 22:12     ` Andreas Grünbacher
2015-09-17 17:30       ` J. Bruce Fields
2015-09-05 10:27 ` [RFC v7 11/41] vfs: Cache base_acl objects in inodes Andreas Gruenbacher
2015-09-05 10:27 ` [RFC v7 12/41] vfs: Cache richacl in struct inode Andreas Gruenbacher
2015-09-05 10:27 ` [RFC v7 13/41] richacl: Check if an acl is equivalent to a file mode Andreas Gruenbacher
2015-09-17 18:22   ` J. Bruce Fields
2015-09-18  0:56     ` J. Bruce Fields
2015-09-21 13:59       ` Austin S Hemmelgarn
2015-09-21 14:38         ` J. Bruce Fields
2015-09-21 17:00           ` Austin S Hemmelgarn
2015-09-21 17:48             ` J. Bruce Fields
2015-09-21 15:31         ` J. Bruce Fields
2015-09-21 23:26       ` Andreas Gruenbacher
2015-09-21 23:20     ` Andreas Gruenbacher
2015-09-17 18:37   ` J. Bruce Fields
2015-09-05 10:27 ` [RFC v7 14/41] richacl: Create-time inheritance Andreas Gruenbacher
2015-09-18 17:58   ` J. Bruce Fields
2015-09-21 20:37     ` Andreas Gruenbacher
2015-09-05 10:27 ` [RFC v7 15/41] richacl: Automatic Inheritance Andreas Gruenbacher
2015-09-18 18:40   ` J. Bruce Fields
2015-09-21 21:19     ` Andreas Gruenbacher
2015-09-22  1:51       ` J. Bruce Fields
2015-09-23 13:55         ` J. Bruce Fields
2015-09-05 10:27 ` [RFC v7 16/41] richacl: xattr mapping functions Andreas Gruenbacher
2015-09-05 10:27 ` [RFC v7 17/41] vfs: Add richacl permission checking Andreas Gruenbacher
2015-09-05 10:27 ` [RFC v7 18/41] ext4: Add richacl support Andreas Gruenbacher
2015-09-23  2:30   ` Aneesh Kumar K.V
2015-09-05 10:27 ` [RFC v7 19/41] ext4: Add richacl feature flag Andreas Gruenbacher
2015-09-23  2:31   ` Aneesh Kumar K.V
2015-09-05 10:27 ` [RFC v7 20/41] richacl: acl editing helper functions Andreas Gruenbacher
2015-09-18 18:54   ` J. Bruce Fields
2015-09-05 10:27 ` [RFC v7 21/41] richacl: Move everyone@ aces down the acl Andreas Gruenbacher
2015-09-18 19:35   ` J. Bruce Fields
2015-09-21 21:43     ` Andreas Gruenbacher
2015-09-22  1:52       ` J. Bruce Fields
2015-09-05 10:27 ` [RFC v7 22/41] richacl: Propagate everyone@ permissions to other aces Andreas Gruenbacher
2015-09-18 21:36   ` J. Bruce Fields
2015-09-21 23:44     ` Andreas Gruenbacher
2015-09-18 21:56   ` J. Bruce Fields
2015-09-21 19:24     ` J. Bruce Fields
2015-09-23  1:24       ` Andreas Gruenbacher
2015-09-23  1:39       ` Andreas Gruenbacher
2015-09-23  1:46         ` J. Bruce Fields
2015-09-05 10:27 ` [RFC v7 23/41] richacl: Set the owner permissions to the owner mask Andreas Gruenbacher
2015-09-21 21:00   ` J. Bruce Fields
2015-09-05 10:27 ` [RFC v7 24/41] richacl: Set the other permissions to the other mask Andreas Gruenbacher
2015-09-23 14:03   ` J. Bruce Fields
2015-09-23 14:12     ` Andreas Grünbacher
2015-09-05 10:27 ` [RFC v7 25/41] richacl: Isolate the owner and group classes Andreas Gruenbacher
2015-09-22 16:06   ` J. Bruce Fields
2015-09-23 13:11     ` Andreas Gruenbacher
2015-09-23 13:15       ` J. Bruce Fields
2015-09-22 19:02   ` J. Bruce Fields
2015-09-23 13:33     ` Andreas Gruenbacher
2015-09-25 11:25       ` Andreas Gruenbacher
2015-09-25 20:17         ` J. Bruce Fields
2015-09-22 19:02   ` J. Bruce Fields
2015-09-05 10:27 ` [RFC v7 26/41] richacl: Apply the file masks to a richacl Andreas Gruenbacher
2015-09-22 19:11   ` J. Bruce Fields
2015-09-23 19:18     ` J. Bruce Fields [this message]
2015-09-23 20:29       ` Andreas Gruenbacher
2015-09-23 20:33         ` J. Bruce Fields
2015-09-23 20:40           ` Andreas Gruenbacher
2015-09-23 21:05             ` J. Bruce Fields
2015-09-23 22:14               ` Andreas Gruenbacher
2015-09-24 15:28                 ` J. Bruce Fields
2015-09-24 15:48                   ` Andreas Gruenbacher
2015-09-22 20:50   ` J. Bruce Fields
2015-09-24 18:33   ` J. Bruce Fields
2015-09-25 16:21     ` [PATCH] richacl: Possible other write-through fix Andreas Gruenbacher
2015-09-25 16:45     ` Andreas Gruenbacher
2015-09-25 18:36       ` J. Bruce Fields
2015-09-05 10:27 ` [RFC v7 27/41] richacl: Create richacl from mode values Andreas Gruenbacher
2015-09-23 20:11   ` J. Bruce Fields
2015-09-05 10:27 ` [RFC v7 28/41] nfsd: Keep list of acls to dispose of in compoundargs Andreas Gruenbacher
2015-09-23 20:28   ` J. Bruce Fields
2015-09-05 10:27 ` [RFC v7 29/41] nfsd: Use richacls as internal acl representation Andreas Gruenbacher
2015-09-24 19:29   ` J. Bruce Fields
2015-09-05 10:27 ` [RFC v7 30/41] nfsd: Add richacl support Andreas Gruenbacher
2015-09-24 19:38   ` J. Bruce Fields
2015-09-05 10:27 ` [RFC v7 31/41] nfsd: Add support for the v4.1 dacl attribute Andreas Gruenbacher
2015-09-24 19:59   ` J. Bruce Fields
2015-09-25 16:37     ` Andreas Gruenbacher
2015-09-05 10:27 ` [RFC v7 32/41] nfsd: Add support for the MAY_CREATE_{FILE,DIR} permissions Andreas Gruenbacher
2015-09-24 20:01   ` J. Bruce Fields
2015-09-05 10:27 ` [RFC v7 33/41] richacl: Add support for unmapped identifiers Andreas Gruenbacher
2015-09-05 10:27 ` [RFC v7 34/41] ext4: Don't allow unmapped identifiers in richacls Andreas Gruenbacher
2015-09-05 10:27 ` [RFC v7 35/41] sunrpc: Allow to demand-allocate pages to encode into Andreas Gruenbacher
2015-09-05 10:27 ` [RFC v7 36/41] sunrpc: Add xdr_init_encode_pages Andreas Gruenbacher
2015-09-05 10:27 ` [RFC v7 37/41] nfs: Fix GETATTR bitmap verification Andreas Gruenbacher
2015-09-05 10:27 ` [RFC v7 38/41] nfs: Remove unused xdr page offsets in getacl/setacl arguments Andreas Gruenbacher
2015-09-05 10:27 ` [RFC v7 39/41] nfs: Add richacl support Andreas Gruenbacher
2015-09-05 10:27 ` [RFC v7 40/41] nfs: Add support for the v4.1 dacl attribute Andreas Gruenbacher
2015-09-05 10:27 ` [RFC v7 41/41] richacl: uapi header split Andreas Gruenbacher

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=20150923191832.GA29577@fieldses.org \
    --to=bfields@fieldses.org \
    --cc=agruen@kernel.org \
    --cc=agruenba@redhat.com \
    --cc=linux-api@vger.kernel.org \
    --cc=linux-cifs@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-nfs@vger.kernel.org \
    --cc=linux-security-module@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).