[PATCH v2] asymmetric_keys: check asymmetric_key_ids() for NULL before dereference

[PATCH v2] asymmetric_keys: check asymmetric_key_ids() for NULL before dereference

[Weiming Shi]

asymmetric_key_ids() returns key->payload.data[asym_key_ids], which can
be NULL for keys parsed by the PKCS#8 parser (pkcs8_parser.c explicitly
stores NULL in prep->payload.data[asym_key_ids]).

key_or_keyring_common() in restrict.c and find_asymmetric_key() in
asymmetric_type.c both dereference this return value without checking
for NULL. An unprivileged user can trigger a NULL pointer dereference
in key_or_keyring_common() by creating a PKCS#8 key, restricting a
keyring with key_or_keyring:<pkcs8_serial>, and adding an X.509 cert
to the restricted keyring. CONFIG_PKCS8_PRIVATE_KEY_PARSER=y is
required.

The following bash script can reproduce the issue:

  #!/bin/bash
  modprobe pkcs8_key_parser 2>/dev/null
  openssl genpkey -algorithm RSA -pkeyopt rsa_keygen_bits:1024 \
      -out /tmp/poc.pem 2>/dev/null
  openssl pkcs8 -topk8 -nocrypt -in /tmp/poc.pem \
      -outform DER -out /tmp/poc.p8
  openssl req -new -x509 -key /tmp/poc.pem -outform DER \
      -out /tmp/poc.der -days 365 -subj "/CN=Test" \
      -addext "subjectKeyIdentifier=hash" \
      -addext "authorityKeyIdentifier=keyid:always" 2>/dev/null
  PKCS8_ID=$(keyctl padd asymmetric pkcs8key @s < /tmp/poc.p8)
  KR=$(keyctl newring test_kr @s)
  keyctl restrict_keyring $KR asymmetric "key_or_keyring:$PKCS8_ID"
  keyctl padd asymmetric trigger $KR < /tmp/poc.der
  rm -f /tmp/poc.pem /tmp/poc.p8 /tmp/poc.der

 Oops: general protection fault, probably for non-canonical address 0xdffffc0000000000
 KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007]
 RIP: 0010:key_or_keyring_common (crypto/asymmetric_keys/restrict.c:205 crypto/asymmetric_keys/restrict.c:279)
 Call Trace:
  <TASK>
  __key_create_or_update (security/keys/key.c:884)
  key_create_or_update (security/keys/key.c:1021)
  __do_sys_add_key (security/keys/keyctl.c:134)
  do_syscall_64 (arch/x86/entry/common.c:52 arch/x86/entry/common.c:83)
  entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130)
  </TASK>
 Kernel panic - not syncing: Fatal exception

Add a NULL check in find_asymmetric_key(), mirroring the existing
pattern in asymmetric_match_key_ids() and asymmetric_key_describe().
In key_or_keyring_common(), skip the trusted key matching when it
has no key IDs and fall through to the check_dest path.

Fixes: 7d30198ee24f ("keys: X.509 public key issuer lookup without AKID")
Reported-by: Xiang Mei <xmei5@asu.edu>
Signed-off-by: Weiming Shi <bestswngs@gmail.com>
---
v2: add bash reproducer to commit message (Ignat)

 crypto/asymmetric_keys/asymmetric_type.c | 2 ++
 crypto/asymmetric_keys/restrict.c        | 9 +++++++--
 2 files changed, 9 insertions(+), 2 deletions(-)

diff --git a/crypto/asymmetric_keys/asymmetric_type.c b/crypto/asymmetric_keys/asymmetric_type.c
index 16a7ae16593c..22f04656d529 100644
--- a/crypto/asymmetric_keys/asymmetric_type.c
+++ b/crypto/asymmetric_keys/asymmetric_type.c
@@ -109,6 +109,8 @@ struct key *find_asymmetric_key(struct key *keyring,
 	if (id_0 && id_1) {
 		const struct asymmetric_key_ids *kids = asymmetric_key_ids(key);
 
+		if (!kids)
+			goto reject;
 		if (!kids->id[1]) {
 			pr_debug("First ID matches, but second is missing\n");
 			goto reject;
diff --git a/crypto/asymmetric_keys/restrict.c b/crypto/asymmetric_keys/restrict.c
index 86292965f493..ccf1084f720e 100644
--- a/crypto/asymmetric_keys/restrict.c
+++ b/crypto/asymmetric_keys/restrict.c
@@ -243,10 +243,14 @@ static int key_or_keyring_common(struct key *dest_keyring,
 			if (IS_ERR(key))
 				key = NULL;
 		} else if (trusted->type == &key_type_asymmetric) {
+			const struct asymmetric_key_ids *kids;
 			const struct asymmetric_key_id **signer_ids;
 
-			signer_ids = (const struct asymmetric_key_id **)
-				asymmetric_key_ids(trusted)->id;
+			kids = asymmetric_key_ids(trusted);
+			if (!kids)
+				goto skip_trusted;
+
+			signer_ids = (const struct asymmetric_key_id **)kids->id;
 
 			/*
 			 * The auth_ids come from the candidate key (the
@@ -290,6 +294,7 @@ static int key_or_keyring_common(struct key *dest_keyring,
 		}
 	}
 
+skip_trusted:
 	if (check_dest && !key) {
 		/* See if the destination has a key that signed this one. */
 		key = find_asymmetric_key(dest_keyring, sig->auth_ids[0],
-- 
2.43.0

Re: [PATCH v2] asymmetric_keys: check asymmetric_key_ids() for NULL before dereference

[Herbert Xu]

On Sat, May 02, 2026 at 09:33:29AM -0700, Weiming Shi wrote:
> asymmetric_key_ids() returns key->payload.data[asym_key_ids], which can
> be NULL for keys parsed by the PKCS#8 parser (pkcs8_parser.c explicitly
> stores NULL in prep->payload.data[asym_key_ids]).
> 
> key_or_keyring_common() in restrict.c and find_asymmetric_key() in
> asymmetric_type.c both dereference this return value without checking
> for NULL. An unprivileged user can trigger a NULL pointer dereference
> in key_or_keyring_common() by creating a PKCS#8 key, restricting a
> keyring with key_or_keyring:<pkcs8_serial>, and adding an X.509 cert
> to the restricted keyring. CONFIG_PKCS8_PRIVATE_KEY_PARSER=y is
> required.
> 
> The following bash script can reproduce the issue:
> 
>   #!/bin/bash
>   modprobe pkcs8_key_parser 2>/dev/null
>   openssl genpkey -algorithm RSA -pkeyopt rsa_keygen_bits:1024 \
>       -out /tmp/poc.pem 2>/dev/null
>   openssl pkcs8 -topk8 -nocrypt -in /tmp/poc.pem \
>       -outform DER -out /tmp/poc.p8
>   openssl req -new -x509 -key /tmp/poc.pem -outform DER \
>       -out /tmp/poc.der -days 365 -subj "/CN=Test" \
>       -addext "subjectKeyIdentifier=hash" \
>       -addext "authorityKeyIdentifier=keyid:always" 2>/dev/null
>   PKCS8_ID=$(keyctl padd asymmetric pkcs8key @s < /tmp/poc.p8)
>   KR=$(keyctl newring test_kr @s)
>   keyctl restrict_keyring $KR asymmetric "key_or_keyring:$PKCS8_ID"
>   keyctl padd asymmetric trigger $KR < /tmp/poc.der
>   rm -f /tmp/poc.pem /tmp/poc.p8 /tmp/poc.der
> 
>  Oops: general protection fault, probably for non-canonical address 0xdffffc0000000000
>  KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007]
>  RIP: 0010:key_or_keyring_common (crypto/asymmetric_keys/restrict.c:205 crypto/asymmetric_keys/restrict.c:279)
>  Call Trace:
>   <TASK>
>   __key_create_or_update (security/keys/key.c:884)
>   key_create_or_update (security/keys/key.c:1021)
>   __do_sys_add_key (security/keys/keyctl.c:134)
>   do_syscall_64 (arch/x86/entry/common.c:52 arch/x86/entry/common.c:83)
>   entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130)
>   </TASK>
>  Kernel panic - not syncing: Fatal exception
> 
> Add a NULL check in find_asymmetric_key(), mirroring the existing
> pattern in asymmetric_match_key_ids() and asymmetric_key_describe().
> In key_or_keyring_common(), skip the trusted key matching when it
> has no key IDs and fall through to the check_dest path.
> 
> Fixes: 7d30198ee24f ("keys: X.509 public key issuer lookup without AKID")
> Reported-by: Xiang Mei <xmei5@asu.edu>
> Signed-off-by: Weiming Shi <bestswngs@gmail.com>
> ---
> v2: add bash reproducer to commit message (Ignat)
> 
>  crypto/asymmetric_keys/asymmetric_type.c | 2 ++
>  crypto/asymmetric_keys/restrict.c        | 9 +++++++--
>  2 files changed, 9 insertions(+), 2 deletions(-)

Patch applied.  Thanks.
-- 
Email: Herbert Xu <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

Re: [PATCH v2] asymmetric_keys: check asymmetric_key_ids() for NULL before dereference

[Herbert Xu]

On Sat, May 02, 2026 at 09:33:29AM -0700, Weiming Shi wrote:
>
> diff --git a/crypto/asymmetric_keys/asymmetric_type.c b/crypto/asymmetric_keys/asymmetric_type.c
> index 16a7ae16593c..22f04656d529 100644
> --- a/crypto/asymmetric_keys/asymmetric_type.c
> +++ b/crypto/asymmetric_keys/asymmetric_type.c
> @@ -109,6 +109,8 @@ struct key *find_asymmetric_key(struct key *keyring,
>  	if (id_0 && id_1) {
>  		const struct asymmetric_key_ids *kids = asymmetric_key_ids(key);
>  
> +		if (!kids)
> +			goto reject;

This check is actually unnecessary because we've already matched
the key against the kid so it must be present.

I'd get rid of this check or perhaps add a comment instead.

>  		if (!kids->id[1]) {
>  			pr_debug("First ID matches, but second is missing\n");
>  			goto reject;
> diff --git a/crypto/asymmetric_keys/restrict.c b/crypto/asymmetric_keys/restrict.c
> index 86292965f493..ccf1084f720e 100644
> --- a/crypto/asymmetric_keys/restrict.c
> +++ b/crypto/asymmetric_keys/restrict.c
> @@ -243,10 +243,14 @@ static int key_or_keyring_common(struct key *dest_keyring,
>  			if (IS_ERR(key))
>  				key = NULL;
>  		} else if (trusted->type == &key_type_asymmetric) {
> +			const struct asymmetric_key_ids *kids;
>  			const struct asymmetric_key_id **signer_ids;
>  
> -			signer_ids = (const struct asymmetric_key_id **)
> -				asymmetric_key_ids(trusted)->id;
> +			kids = asymmetric_key_ids(trusted);
> +			if (!kids)
> +				goto skip_trusted;

Yes this is definitely buggy.

I think it was introduced by these two commits:

commit 3c58b2362ba828ee2970c66c6a6fd7b04fde4413
Author: David Howells <dhowells@redhat.com>
Date:   Tue Oct 9 17:47:46 2018 +0100

    KEYS: Implement PKCS#8 RSA Private Key parser [ver #2]

and

commit 7e3c4d22083f6e7316c5229b6197ca2d5335aa35
Author: Mat Martineau <martineau@kernel.org>
Date:   Mon Jun 27 16:45:16 2016 -0700

    KEYS: Restrict asymmetric key linkage using a specific keychain

So the Fixes header should point to them.

> @@ -290,6 +294,7 @@ static int key_or_keyring_common(struct key *dest_keyring,
>  		}
>  	}
>  
> +skip_trusted:
>  	if (check_dest && !key) {
>  		/* See if the destination has a key that signed this one. */
>  		key = find_asymmetric_key(dest_keyring, sig->auth_ids[0],

I'm not sure continuing here is a good idea.  Having a private key
here makes no sense whatsoever and we should just bail out right
away.

I would recommend returning an error of some sort if kids is NULL.

David/Lukas/Ignat, any opinions?

Thanks,
-- 
Email: Herbert Xu <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

Re: [PATCH v2] asymmetric_keys: check asymmetric_key_ids() for NULL before dereference

[Jarkko Sakkinen]

On Mon, Jun 22, 2026 at 12:16:07PM +0800, Herbert Xu wrote:
> On Sat, May 02, 2026 at 09:33:29AM -0700, Weiming Shi wrote:
> >
> > diff --git a/crypto/asymmetric_keys/asymmetric_type.c b/crypto/asymmetric_keys/asymmetric_type.c
> > index 16a7ae16593c..22f04656d529 100644
> > --- a/crypto/asymmetric_keys/asymmetric_type.c
> > +++ b/crypto/asymmetric_keys/asymmetric_type.c
> > @@ -109,6 +109,8 @@ struct key *find_asymmetric_key(struct key *keyring,
> >  	if (id_0 && id_1) {
> >  		const struct asymmetric_key_ids *kids = asymmetric_key_ids(key);
> >  
> > +		if (!kids)
> > +			goto reject;
> 
> This check is actually unnecessary because we've already matched
> the key against the kid so it must be present.
> 
> I'd get rid of this check or perhaps add a comment instead.

+1

> 
> >  		if (!kids->id[1]) {
> >  			pr_debug("First ID matches, but second is missing\n");
> >  			goto reject;
> > diff --git a/crypto/asymmetric_keys/restrict.c b/crypto/asymmetric_keys/restrict.c
> > index 86292965f493..ccf1084f720e 100644
> > --- a/crypto/asymmetric_keys/restrict.c
> > +++ b/crypto/asymmetric_keys/restrict.c
> > @@ -243,10 +243,14 @@ static int key_or_keyring_common(struct key *dest_keyring,
> >  			if (IS_ERR(key))
> >  				key = NULL;
> >  		} else if (trusted->type == &key_type_asymmetric) {
> > +			const struct asymmetric_key_ids *kids;
> >  			const struct asymmetric_key_id **signer_ids;
> >  
> > -			signer_ids = (const struct asymmetric_key_id **)
> > -				asymmetric_key_ids(trusted)->id;
> > +			kids = asymmetric_key_ids(trusted);
> > +			if (!kids)
> > +				goto skip_trusted;
> 
> Yes this is definitely buggy.
> 
> I think it was introduced by these two commits:
> 
> commit 3c58b2362ba828ee2970c66c6a6fd7b04fde4413
> Author: David Howells <dhowells@redhat.com>
> Date:   Tue Oct 9 17:47:46 2018 +0100
> 
>     KEYS: Implement PKCS#8 RSA Private Key parser [ver #2]
> 
> and
> 
> commit 7e3c4d22083f6e7316c5229b6197ca2d5335aa35
> Author: Mat Martineau <martineau@kernel.org>
> Date:   Mon Jun 27 16:45:16 2016 -0700
> 
>     KEYS: Restrict asymmetric key linkage using a specific keychain
> 
> So the Fixes header should point to them.

+1

> 
> > @@ -290,6 +294,7 @@ static int key_or_keyring_common(struct key *dest_keyring,
> >  		}
> >  	}
> >  
> > +skip_trusted:
> >  	if (check_dest && !key) {
> >  		/* See if the destination has a key that signed this one. */
> >  		key = find_asymmetric_key(dest_keyring, sig->auth_ids[0],
> 
> I'm not sure continuing here is a good idea.  Having a private key
> here makes no sense whatsoever and we should just bail out right
> away.
> 
> I would recommend returning an error of some sort if kids is NULL.
> 
> David/Lukas/Ignat, any opinions?

I think with a quick skim that you are right. I'll work on this area
for the next version.

> 
> Thanks,
> -- 
> Email: Herbert Xu <herbert@gondor.apana.org.au>
> Home Page: http://gondor.apana.org.au/~herbert/
> PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

Thanks for the review!

BR, Jarkko