Linux-Security-Module Archive on lore.kernel.org
 help / color / Atom feed
From: Roberto Sassu <roberto.sassu@huawei.com>
To: Mimi Zohar <zohar@linux.ibm.com>, "mjg59@google.com" <mjg59@google.com>
Cc: "linux-integrity@vger.kernel.org"
	<linux-integrity@vger.kernel.org>,
	"linux-security-module@vger.kernel.org" 
	<linux-security-module@vger.kernel.org>,
	"linux-fsdevel@vger.kernel.org" <linux-fsdevel@vger.kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: RE: [PATCH v5 06/12] evm: Ignore INTEGRITY_NOLABEL/INTEGRITY_NOXATTRS if conditions are safe
Date: Mon, 3 May 2021 07:55:29 +0000
Message-ID: <c12f18094cc0479faa3f0f152b4964de@huawei.com> (raw)
In-Reply-To: <b8790b57e289980d4fe1133d15203ce016d2319d.camel@linux.ibm.com>

> From: Mimi Zohar [mailto:zohar@linux.ibm.com]
> Sent: Monday, May 3, 2021 2:13 AM
> Hi Roberto,
> 
> On Wed, 2021-04-07 at 12:52 +0200, Roberto Sassu wrote:
> > When a file is being created, LSMs can set the initial label with the
> > inode_init_security hook. If no HMAC key is loaded, the new file will have
> > LSM xattrs but not the HMAC. It is also possible that the file remains
> > without protected xattrs after creation if no active LSM provided it.
> >
> > Unfortunately, EVM will deny any further metadata operation on new files,
> > as evm_protect_xattr() will always return the INTEGRITY_NOLABEL error, or
> > INTEGRITY_NOXATTRS if no protected xattrs exist. This would limit the
> > usability of EVM when only a public key is loaded, as commands such as cp
> > or tar with the option to preserve xattrs won't work.
> >
> > This patch ignores these errors when they won't be an issue, if no HMAC
> key
> > is loaded and cannot be loaded in the future (which can be enforced by
> > setting the EVM_SETUP_COMPLETE initialization flag).
> >
> > Signed-off-by: Roberto Sassu <roberto.sassu@huawei.com>
> > ---
> >  security/integrity/evm/evm_main.c | 23 ++++++++++++++++++++++-
> >  1 file changed, 22 insertions(+), 1 deletion(-)
> >
> > diff --git a/security/integrity/evm/evm_main.c
> b/security/integrity/evm/evm_main.c
> > index 998818283fda..6556e8c22da9 100644
> > --- a/security/integrity/evm/evm_main.c
> > +++ b/security/integrity/evm/evm_main.c
> > @@ -90,6 +90,24 @@ static bool evm_key_loaded(void)
> >  	return (bool)(evm_initialized & EVM_KEY_MASK);
> >  }
> >
> > +/*
> > + * Ignoring INTEGRITY_NOLABEL/INTEGRITY_NOXATTRS is safe if no HMAC
> key
> > + * is loaded and the EVM_SETUP_COMPLETE initialization flag is set.
> > + */
> > +static bool evm_ignore_error_safe(enum integrity_status evm_status)
> > +{
> > +	if (evm_initialized & EVM_INIT_HMAC)
> > +		return false;
> > +
> > +	if (!(evm_initialized & EVM_SETUP_COMPLETE))
> > +		return false;
> > +
> > +	if (evm_status != INTEGRITY_NOLABEL && evm_status !=
> INTEGRITY_NOXATTRS)
> > +		return false;
> > +
> > +	return true;
> > +}
> > +
> >  static int evm_find_protected_xattrs(struct dentry *dentry)
> >  {
> >  	struct inode *inode = d_backing_inode(dentry);
> > @@ -354,6 +372,8 @@ static int evm_protect_xattr(struct dentry *dentry,
> const char *xattr_name,
> >  				    -EPERM, 0);
> >  	}
> >  out:
> > +	if (evm_ignore_error_safe(evm_status))
> > +		return 0;
> 
> I agree with the concept, but the function name doesn't provide enough
> context.  Perhaps defining a function more along the lines of
> "evm_hmac_disabled()" would be more appropriate and at the same time
> self documenting.

Since the function checks if the passed error can be ignored,
would evm_ignore_error_hmac_disabled() also be ok?

> >  	if (evm_status != INTEGRITY_PASS)
> >  		integrity_audit_msg(AUDIT_INTEGRITY_METADATA,
> d_backing_inode(dentry),
> >  				    dentry->d_name.name,
> "appraise_metadata",
> > @@ -515,7 +535,8 @@ int evm_inode_setattr(struct dentry *dentry, struct
> iattr *attr)
> >  		return 0;
> >  	evm_status = evm_verify_current_integrity(dentry);
> >  	if ((evm_status == INTEGRITY_PASS) ||
> > -	    (evm_status == INTEGRITY_NOXATTRS))
> > +	    (evm_status == INTEGRITY_NOXATTRS) ||
> > +	    (evm_ignore_error_safe(evm_status)))
> 
> It would also remove the INTEGRITY_NOXATTRS test duplication here.

Ok.

Thanks

Roberto

HUAWEI TECHNOLOGIES Duesseldorf GmbH, HRB 56063
Managing Director: Li Peng, Li Jian, Shi Yanli

> thanks,
> 
> Mimi
> 
> >  		return 0;
> >  	integrity_audit_msg(AUDIT_INTEGRITY_METADATA,
> d_backing_inode(dentry),
> >  			    dentry->d_name.name, "appraise_metadata",


  reply index

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-04-07 10:52 [PATCH v5 00/12] evm: Improve usability of portable signatures Roberto Sassu
2021-04-07 10:52 ` [PATCH v5 01/12] evm: Execute evm_inode_init_security() only when an HMAC key is loaded Roberto Sassu
2021-04-07 10:52 ` [PATCH v5 02/12] evm: Load EVM key in ima_load_x509() to avoid appraisal Roberto Sassu
2021-04-07 10:52 ` [PATCH v5 03/12] evm: Refuse EVM_ALLOW_METADATA_WRITES only if an HMAC key is loaded Roberto Sassu
2021-04-30 20:28   ` Mimi Zohar
2021-04-07 10:52 ` [PATCH v5 04/12] ima: Move ima_reset_appraise_flags() call to post hooks Roberto Sassu
2021-04-07 16:17   ` Casey Schaufler
2021-04-07 16:31     ` Roberto Sassu
2021-04-26 19:49     ` Mimi Zohar
2021-04-07 10:52 ` [PATCH v5 05/12] evm: Introduce evm_status_revalidate() Roberto Sassu
2021-04-07 10:52 ` [PATCH v5 06/12] evm: Ignore INTEGRITY_NOLABEL/INTEGRITY_NOXATTRS if conditions are safe Roberto Sassu
2021-05-03  0:12   ` Mimi Zohar
2021-05-03  7:55     ` Roberto Sassu [this message]
2021-05-03 12:07       ` Mimi Zohar
2021-05-03 14:15       ` Roberto Sassu
2021-05-03 14:34         ` Mimi Zohar
2021-05-04 13:16           ` Roberto Sassu
2021-05-04 13:45             ` Mimi Zohar
2021-04-07 10:52 ` [PATCH v5 07/12] evm: Allow xattr/attr operations for portable signatures Roberto Sassu
2021-05-03  0:12   ` Mimi Zohar
2021-05-04 14:28     ` Roberto Sassu
2021-05-04 14:49       ` Mimi Zohar
2021-04-07 10:52 ` [PATCH v5 08/12] evm: Pass user namespace to set/remove xattr hooks Roberto Sassu
2021-04-07 12:06   ` Christian Brauner
2021-04-07 10:52 ` [PATCH v5 09/12] evm: Allow setxattr() and setattr() for unmodified metadata Roberto Sassu
2021-04-07 12:05   ` Christian Brauner
2021-04-07 15:23   ` kernel test robot
2021-04-07 18:14   ` kernel test robot
2021-04-07 19:28     ` [RESEND][PATCH " Roberto Sassu
2021-05-03 13:00   ` [PATCH " Mimi Zohar
2021-05-03 14:48     ` Roberto Sassu
2021-05-03 15:13       ` Mimi Zohar
2021-05-03 15:30         ` Roberto Sassu
2021-05-03 15:11     ` Roberto Sassu
2021-05-03 15:26       ` Mimi Zohar
2021-05-03 15:32         ` Roberto Sassu
2021-05-03 15:48           ` Mimi Zohar
2021-04-07 10:52 ` [PATCH v5 10/12] ima: Allow imasig requirement to be satisfied by EVM portable signatures Roberto Sassu
2021-04-07 10:52 ` [PATCH v5 11/12] ima: Introduce template field evmsig and write to field sig as fallback Roberto Sassu
2021-04-07 10:52 ` [PATCH v5 12/12] ima: Don't remove security.ima if file must not be appraised Roberto Sassu

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=c12f18094cc0479faa3f0f152b4964de@huawei.com \
    --to=roberto.sassu@huawei.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-integrity@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-security-module@vger.kernel.org \
    --cc=mjg59@google.com \
    --cc=zohar@linux.ibm.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

Linux-Security-Module Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-security-module/0 linux-security-module/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-security-module linux-security-module/ https://lore.kernel.org/linux-security-module \
		linux-security-module@vger.kernel.org
	public-inbox-index linux-security-module

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-security-module


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git