From: Yang Xu xuyang2018.jy@cn.fujitsu.com
commit 2e356101e72ab1361821b3af024d64877d9a798d upstream.
Currently, when we add a new user key, the calltrace as below:
add_key() key_create_or_update() key_alloc() __key_instantiate_and_link generic_key_instantiate key_payload_reserve ......
Since commit a08bf91ce28e ("KEYS: allow reaching the keys quotas exactly"), we can reach max bytes/keys in key_alloc, but we forget to remove this limit when we reserver space for payload in key_payload_reserve. So we can only reach max keys but not max bytes when having delta between plen and type->def_datalen. Remove this limit when instantiating the key, so we can keep consistent with key_alloc.
Also, fix the similar problem in keyctl_chown_key().
Fixes: 0b77f5bfb45c ("keys: make the keyring quotas controllable through /proc/sys") Fixes: a08bf91ce28e ("KEYS: allow reaching the keys quotas exactly") Cc: stable@vger.kernel.org # 5.0.x Cc: Eric Biggers ebiggers@google.com Signed-off-by: Yang Xu xuyang2018.jy@cn.fujitsu.com Reviewed-by: Jarkko Sakkinen jarkko.sakkinen@linux.intel.com Reviewed-by: Eric Biggers ebiggers@google.com Signed-off-by: Jarkko Sakkinen jarkko.sakkinen@linux.intel.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org
--- security/keys/key.c | 2 +- security/keys/keyctl.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-)
--- a/security/keys/key.c +++ b/security/keys/key.c @@ -381,7 +381,7 @@ int key_payload_reserve(struct key *key, spin_lock(&key->user->lock);
if (delta > 0 && - (key->user->qnbytes + delta >= maxbytes || + (key->user->qnbytes + delta > maxbytes || key->user->qnbytes + delta < key->user->qnbytes)) { ret = -EDQUOT; } --- a/security/keys/keyctl.c +++ b/security/keys/keyctl.c @@ -937,8 +937,8 @@ long keyctl_chown_key(key_serial_t id, u key_quota_root_maxbytes : key_quota_maxbytes;
spin_lock(&newowner->lock); - if (newowner->qnkeys + 1 >= maxkeys || - newowner->qnbytes + key->quotalen >= maxbytes || + if (newowner->qnkeys + 1 > maxkeys || + newowner->qnbytes + key->quotalen > maxbytes || newowner->qnbytes + key->quotalen < newowner->qnbytes) goto quota_overrun;