From: "Mickaël Salaün" <mic@digikod.net>
To: Mimi Zohar <zohar@linux.ibm.com>
Cc: linux-integrity@vger.kernel.org, roberto.sassu@huawei.com,
linux-security-module@vger.kernel.org,
linux-kernel@vger.kernel.org, audit@vger.kernel.org,
Paul Moore <paul@paul-moore.com>
Subject: Re: [PATCH] ima: instantiate the bprm_creds_for_exec() hook
Date: Fri, 29 Nov 2024 12:06:11 +0100 [thread overview]
Message-ID: <20241129.keeDathoo3Oh@digikod.net> (raw)
In-Reply-To: <20241127210234.121546-1-zohar@linux.ibm.com>
For reference, here is the base patch series:
https://lore.kernel.org/all/20241112191858.162021-1-mic@digikod.net/
CCing audit@
On Wed, Nov 27, 2024 at 04:02:34PM -0500, Mimi Zohar wrote:
> Like direct file execution (e.g. ./script.sh), indirect file execution
> (e.g. sh script.sh) needs to be measured and appraised. Instantiate
> the new security_bprm_creds_for_exec() hook to measure and verify the
> indirect file's integrity. Unlike direct file execution, indirect file
> execution integrity is optionally enforced by the interpreter.
>
> Update the audit messages to differentiate between kernel and userspace
> enforced integrity.
I'm not sure to see the full picture. What is the difference between
execveat() calls and execveat() + AT_EXECVE_CHECK calls? Both are from
user space, the only difference is that the first can lead to a full
execution, but the intent is the same.
>
> Signed-off-by: Roberto Sassu <roberto.sassu@huawei.com>
> Signed-off-by: Mimi Zohar <zohar@linux.ibm.com>
> ---
> security/integrity/ima/ima_appraise.c | 84 ++++++++++++++++++++-------
> security/integrity/ima/ima_main.c | 22 +++++++
> 2 files changed, 86 insertions(+), 20 deletions(-)
>
> diff --git a/security/integrity/ima/ima_appraise.c b/security/integrity/ima/ima_appraise.c
> index 656c709b974f..b5f8e49cde9d 100644
> --- a/security/integrity/ima/ima_appraise.c
> +++ b/security/integrity/ima/ima_appraise.c
> @@ -8,6 +8,7 @@
> #include <linux/module.h>
> #include <linux/init.h>
> #include <linux/file.h>
> +#include <linux/binfmts.h>
> #include <linux/fs.h>
> #include <linux/xattr.h>
> #include <linux/magic.h>
> @@ -16,6 +17,7 @@
> #include <linux/fsverity.h>
> #include <keys/system_keyring.h>
> #include <uapi/linux/fsverity.h>
> +#include <linux/securebits.h>
>
> #include "ima.h"
>
> @@ -276,7 +278,8 @@ static int calc_file_id_hash(enum evm_ima_xattr_type type,
> */
> static int xattr_verify(enum ima_hooks func, struct ima_iint_cache *iint,
> struct evm_ima_xattr_data *xattr_value, int xattr_len,
> - enum integrity_status *status, const char **cause)
> + enum integrity_status *status, const char **cause,
> + bool is_check)
> {
> struct ima_max_digest_data hash;
> struct signature_v2_hdr *sig;
> @@ -292,9 +295,11 @@ static int xattr_verify(enum ima_hooks func, struct ima_iint_cache *iint,
> if (*status != INTEGRITY_PASS_IMMUTABLE) {
> if (iint->flags & IMA_DIGSIG_REQUIRED) {
> if (iint->flags & IMA_VERITY_REQUIRED)
> - *cause = "verity-signature-required";
> + *cause = !is_check ? "verity-signature-required" :
> + "verity-signature-required(userspace)";
This looks simpler (same for all following checks):
is_check ? "verity-signature-required(userspace)" : "verity-signature-required";
> else
> - *cause = "IMA-signature-required";
> + *cause = !is_check ? "IMA-signature-required" :
> + "IMA-signature-required(userspace)";
> *status = INTEGRITY_FAIL;
> break;
> }
> @@ -314,7 +319,8 @@ static int xattr_verify(enum ima_hooks func, struct ima_iint_cache *iint,
> else
> rc = -EINVAL;
> if (rc) {
> - *cause = "invalid-hash";
> + *cause = !is_check ? "invalid-hash" :
> + "invalid-hash(userspace)";
> *status = INTEGRITY_FAIL;
> break;
> }
> @@ -325,14 +331,16 @@ static int xattr_verify(enum ima_hooks func, struct ima_iint_cache *iint,
>
> mask = IMA_DIGSIG_REQUIRED | IMA_VERITY_REQUIRED;
> if ((iint->flags & mask) == mask) {
> - *cause = "verity-signature-required";
> + *cause = !is_check ? "verity-signature-required" :
> + "verity-signature-required(userspace)";
> *status = INTEGRITY_FAIL;
> break;
> }
>
> sig = (typeof(sig))xattr_value;
> if (sig->version >= 3) {
> - *cause = "invalid-signature-version";
> + *cause = !is_check ? "invalid-signature-version" :
> + "invalid-signature-version(userspace)";
> *status = INTEGRITY_FAIL;
> break;
> }
> @@ -353,7 +361,8 @@ static int xattr_verify(enum ima_hooks func, struct ima_iint_cache *iint,
> iint->ima_hash->digest,
> iint->ima_hash->length);
> if (rc) {
> - *cause = "invalid-signature";
> + *cause = !is_check ? "invalid-signature" :
> + "invalid-signature(userspace)";
> *status = INTEGRITY_FAIL;
> } else {
> *status = INTEGRITY_PASS;
> @@ -364,7 +373,8 @@ static int xattr_verify(enum ima_hooks func, struct ima_iint_cache *iint,
>
> if (iint->flags & IMA_DIGSIG_REQUIRED) {
> if (!(iint->flags & IMA_VERITY_REQUIRED)) {
> - *cause = "IMA-signature-required";
> + *cause = !is_check ? "IMA-signature-required" :
> + "IMA-signature-required(userspace)";
> *status = INTEGRITY_FAIL;
> break;
> }
> @@ -372,7 +382,8 @@ static int xattr_verify(enum ima_hooks func, struct ima_iint_cache *iint,
>
> sig = (typeof(sig))xattr_value;
> if (sig->version != 3) {
> - *cause = "invalid-signature-version";
> + *cause = !is_check ? "invalid-signature-version" :
> + "invalid-signature-version(userspace)";
> *status = INTEGRITY_FAIL;
> break;
> }
> @@ -382,7 +393,8 @@ static int xattr_verify(enum ima_hooks func, struct ima_iint_cache *iint,
> container_of(&hash.hdr,
> struct ima_digest_data, hdr));
> if (rc) {
> - *cause = "sigv3-hashing-error";
> + *cause = !is_check ? "sigv3-hashing-error" :
> + "sigv3-hashing-error(userspace)";
> *status = INTEGRITY_FAIL;
> break;
> }
> @@ -392,7 +404,8 @@ static int xattr_verify(enum ima_hooks func, struct ima_iint_cache *iint,
> xattr_len, hash.digest,
> hash.hdr.length);
> if (rc) {
> - *cause = "invalid-verity-signature";
> + *cause = !is_check ? "invalid-verity-signature" :
> + "invalid-verify-signature(userspace)";
> *status = INTEGRITY_FAIL;
> } else {
> *status = INTEGRITY_PASS;
> @@ -401,7 +414,8 @@ static int xattr_verify(enum ima_hooks func, struct ima_iint_cache *iint,
> break;
> default:
> *status = INTEGRITY_UNKNOWN;
> - *cause = "unknown-ima-data";
> + *cause = !is_check ? "unknown-ima-data" :
> + "unknown-ima-data(userspace)";
> break;
> }
>
> @@ -469,6 +483,18 @@ int ima_check_blacklist(struct ima_iint_cache *iint,
> return rc;
> }
>
> +static int is_bprm_creds_for_exec(enum ima_hooks func, struct file *file)
> +{
> + struct linux_binprm *bprm = NULL;
> +
> + if (func == BPRM_CHECK) {
> + bprm = container_of(&file, struct linux_binprm, file);
> + if (bprm->is_check)
> + return 1;
> + }
> + return 0;
> +}
> +
> /*
> * ima_appraise_measurement - appraise file measurement
> *
> @@ -489,11 +515,24 @@ int ima_appraise_measurement(enum ima_hooks func, struct ima_iint_cache *iint,
> enum integrity_status status = INTEGRITY_UNKNOWN;
> int rc = xattr_len;
> bool try_modsig = iint->flags & IMA_MODSIG_ALLOWED && modsig;
> + bool is_check = false;
>
> /* If not appraising a modsig, we need an xattr. */
> if (!(inode->i_opflags & IOP_XATTR) && !try_modsig)
> return INTEGRITY_UNKNOWN;
>
> + /*
> + * Unlike any of the other LSM hooks where the kernel enforces file
> + * integrity, enforcing file integrity for the bprm_creds_for_exec()
> + * LSM hook is left up to the discretion of the script interpreter
> + * (userspace).
> + *
> + * Since the SECBIT_EXEC_RESTRICT_FILE flag is just a hint as to
> + * userspace intentions, simply annotate the audit messages indicating
> + * a userspace based query.
> + */
> + is_check = is_bprm_creds_for_exec(func, file);
> +
> /* If reading the xattr failed and there's no modsig, error out. */
> if (rc <= 0 && !try_modsig) {
> if (rc && rc != -ENODATA)
> @@ -501,11 +540,14 @@ int ima_appraise_measurement(enum ima_hooks func, struct ima_iint_cache *iint,
>
> if (iint->flags & IMA_DIGSIG_REQUIRED) {
> if (iint->flags & IMA_VERITY_REQUIRED)
> - cause = "verity-signature-required";
> + cause = !is_check ? "verity-signature-required" :
> + "verity-signature-required(userspace)";
> else
> - cause = "IMA-signature-required";
> + cause = !is_check ? "IMA-signature-required" :
> + "IMA-signature-required(userspace)";
> } else {
> - cause = "missing-hash";
> + cause = !is_check ? "missing-hash" :
> + "missing-hash(userspace)";
> }
>
> status = INTEGRITY_NOLABEL;
> @@ -531,14 +573,15 @@ int ima_appraise_measurement(enum ima_hooks func, struct ima_iint_cache *iint,
> break;
> fallthrough;
> case INTEGRITY_NOLABEL: /* No security.evm xattr. */
> - cause = "missing-HMAC";
> + cause = !is_check ? "missing-HMAC" : "missing-HMAC(userspace)";
> goto out;
> case INTEGRITY_FAIL_IMMUTABLE:
> set_bit(IMA_DIGSIG, &iint->atomic_flags);
> - cause = "invalid-fail-immutable";
> + cause = !is_check ? "invalid-fail-immutable" :
> + "invalid-fail-immutable(userspace)";
> goto out;
> case INTEGRITY_FAIL: /* Invalid HMAC/signature. */
> - cause = "invalid-HMAC";
> + cause = !is_check ? "invalid-HMAC" : "invalid-HMAC(userspace)";
> goto out;
> default:
> WARN_ONCE(true, "Unexpected integrity status %d\n", status);
> @@ -546,7 +589,7 @@ int ima_appraise_measurement(enum ima_hooks func, struct ima_iint_cache *iint,
>
> if (xattr_value)
> rc = xattr_verify(func, iint, xattr_value, xattr_len, &status,
> - &cause);
> + &cause, is_check);
>
> /*
> * If we have a modsig and either no imasig or the imasig's key isn't
> @@ -568,7 +611,8 @@ int ima_appraise_measurement(enum ima_hooks func, struct ima_iint_cache *iint,
> ((inode->i_sb->s_iflags & SB_I_UNTRUSTED_MOUNTER) ||
> (iint->flags & IMA_FAIL_UNVERIFIABLE_SIGS))) {
> status = INTEGRITY_FAIL;
> - cause = "unverifiable-signature";
> + cause = !is_check ? "unverifiable-signature" :
> + "unverifiable-signature(userspace)";
> integrity_audit_msg(AUDIT_INTEGRITY_DATA, inode, filename,
> op, cause, rc, 0);
Instead of adding new causes, another option would be to add a new audit
record type (e.g. AUDIT_INTEGRITY_DATA_CHECK). This would help filter
these new kind of messages and I guess scale better.
Another alternative would be to extend the audit message with a new
field (e.g. "check=1"), but that would not help for efficient filtering.
> } else if (status != INTEGRITY_PASS) {
> diff --git a/security/integrity/ima/ima_main.c b/security/integrity/ima/ima_main.c
> index 06132cf47016..2b5d6bae77a4 100644
> --- a/security/integrity/ima/ima_main.c
> +++ b/security/integrity/ima/ima_main.c
> @@ -554,6 +554,27 @@ static int ima_bprm_check(struct linux_binprm *bprm)
> MAY_EXEC, CREDS_CHECK);
> }
>
> +/**
> + * ima_bprm_creds_for_exec - based on policy, collect/store/appraise measurement.
> + * @bprm: contains the linux_binprm structure
> + *
> + * Based on the IMA policy and the execvat(2) AT_CHECK flag, measure and
> + * appraise the integrity of a file to be executed by script interpreters.
> + * Unlike any of the other LSM hooks where the kernel enforces file integrity,
> + * enforcing file integrity is left up to the discretion of the script
> + * interpreter (userspace).
> + *
> + * On success return 0. On integrity appraisal error, assuming the file
> + * is in policy and IMA-appraisal is in enforcing mode, return -EACCES.
> + */
> +static int ima_bprm_creds_for_exec(struct linux_binprm *bprm)
> +{
> + if (!bprm->is_check)
> + return 0;
> +
> + return ima_bprm_check(bprm);
> +}
> +
> /**
> * ima_file_check - based on policy, collect/store measurement.
> * @file: pointer to the file to be measured
> @@ -1177,6 +1198,7 @@ static int __init init_ima(void)
>
> static struct security_hook_list ima_hooks[] __ro_after_init = {
> LSM_HOOK_INIT(bprm_check_security, ima_bprm_check),
> + LSM_HOOK_INIT(bprm_creds_for_exec, ima_bprm_creds_for_exec),
Why not replace bprm_check_security with bprm_creds_for_exec
implementation altogether?
> LSM_HOOK_INIT(file_post_open, ima_file_check),
> LSM_HOOK_INIT(inode_post_create_tmpfile, ima_post_create_tmpfile),
> LSM_HOOK_INIT(file_release, ima_file_free),
> --
> 2.47.0
>
>
next prev parent reply other threads:[~2024-11-29 11:06 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-11-27 21:02 [PATCH] ima: instantiate the bprm_creds_for_exec() hook Mimi Zohar
2024-11-28 15:57 ` kernel test robot
2024-11-28 21:07 ` kernel test robot
2024-11-29 11:06 ` Mickaël Salaün [this message]
2024-12-02 19:40 ` Mimi Zohar
2024-12-03 11:53 ` Mickaël Salaün
2024-12-03 16:03 ` Mimi Zohar
2024-12-02 21:25 ` Stefan Berger
2024-12-03 11:58 ` Mickaël Salaün
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=20241129.keeDathoo3Oh@digikod.net \
--to=mic@digikod.net \
--cc=audit@vger.kernel.org \
--cc=linux-integrity@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-security-module@vger.kernel.org \
--cc=paul@paul-moore.com \
--cc=roberto.sassu@huawei.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
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.