* [PATCH net-next v2 1/5] net/tcp-ao: Drop support for most non-RFC-specified algorithms
2026-04-27 17:27 [PATCH net-next v2 0/5] Reimplement TCP-AO using crypto library Eric Biggers
@ 2026-04-27 17:27 ` Eric Biggers
2026-04-27 17:27 ` [PATCH net-next v2 2/5] net/tcp-ao: Use crypto library API instead of crypto_ahash Eric Biggers
` (4 subsequent siblings)
5 siblings, 0 replies; 8+ messages in thread
From: Eric Biggers @ 2026-04-27 17:27 UTC (permalink / raw)
To: netdev
Cc: linux-crypto, linux-kernel, Eric Dumazet, Neal Cardwell,
Kuniyuki Iwashima, David S . Miller, David Ahern, Jakub Kicinski,
Paolo Abeni, Simon Horman, Ard Biesheuvel, Jason A . Donenfeld,
Herbert Xu, Dmitry Safonov, Eric Biggers
RFC 5926 (https://datatracker.ietf.org/doc/html/rfc5926) specifies the
use of AES-128-CMAC and HMAC-SHA1 with TCP-AO. This includes a
specification for how traffic keys shall be derived for each algorithm.
Support for any other algorithms with TCP-AO isn't standardized, though
an expired Internet Draft (a work-in-progress document, not a standard)
from 2019 does propose adding HMAC-SHA256 support:
https://datatracker.ietf.org/doc/html/draft-nayak-tcp-sha2-03
Since both documents specify the KDF for each algorithm individually, it
isn't necessarily clear how any other algorithm should be integrated.
Nevertheless, the Linux implementation of TCP-AO allows userspace to
specify the MAC algorithm as a string tcp_ao_add::alg_name naming either
"cmac(aes128)" or an arbitrary algorithm in the crypto_ahash API. The
set of valid strings is undocumented. The implementation assumes that
"cmac(aes128)" is the only algorithm that requires an entropy extraction
step and that all algorithms accept keys with length equal to the
untruncated MAC; thus, arbitrary HMAC algorithms probably do work, but
some other MAC algorithms like AES-256-CMAC have never actually worked.
Unfortunately, this undocumented string allows many obsolete, insecure,
or redundant algorithms. For example, "hmac(md5)" and the
non-cryptographic "crc32" are accepted. It also ties the implementation
to crypto_ahash and requires that most memory be dynamically allocated,
making the implementation unnecessarily complex and inefficient. Still
furthermore, this implementation requires the crypto API to support
"transformation cloning", whose only user is this feature.
Fortunately, it's very likely that only a few algorithms are actually
used in practice. Let's restrict the set of allowed algorithms to
"cmac(aes128)" (or "cmac(aes)" with keylen=16), "hmac(sha1)", and
"hmac(sha256)". The first two are the actually standard ones, while
HMAC-SHA256 seems like a reasonable algorithm to continue supporting as
a Linux extension, considering the Internet Draft for it and the fact
that SHA-256 is the usual choice of upgrade from the outdated SHA-1.
If any other algorithm ever turns out to be needed, e.g. HMAC-SHA512, it
can of course be (re-)added in library form. However, note that the TCP
options space limits TCP-AO MACs to 20 bytes (160 bits) anyway, which
limits the potential benefit of any further upgrade to the algorithm.
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Eric Biggers <ebiggers@kernel.org>
---
net/ipv4/tcp_ao.c | 4 ++
tools/testing/selftests/net/tcp_ao/config | 1 -
.../selftests/net/tcp_ao/key-management.c | 41 ++-----------------
3 files changed, 7 insertions(+), 39 deletions(-)
diff --git a/net/ipv4/tcp_ao.c b/net/ipv4/tcp_ao.c
index a97cdf3e6af4..b21bd69b4e82 100644
--- a/net/ipv4/tcp_ao.c
+++ b/net/ipv4/tcp_ao.c
@@ -1561,10 +1561,14 @@ static struct tcp_ao_key *tcp_ao_key_alloc(struct sock *sk,
cmd->alg_name[ARRAY_SIZE(cmd->alg_name) - 1] = '\0';
/* RFC5926, 3.1.1.2. KDF_AES_128_CMAC */
if (!strcmp("cmac(aes128)", algo))
algo = "cmac(aes)";
+ else if (strcmp("hmac(sha1)", algo) &&
+ strcmp("hmac(sha256)", algo) &&
+ (strcmp("cmac(aes)", algo) || cmd->keylen != 16))
+ return ERR_PTR(-ENOENT);
/* Full TCP header (th->doff << 2) should fit into scratch area,
* see tcp_ao_hash_header().
*/
pool_id = tcp_sigpool_alloc_ahash(algo, 60);
diff --git a/tools/testing/selftests/net/tcp_ao/config b/tools/testing/selftests/net/tcp_ao/config
index f22148512365..47228a7d0b90 100644
--- a/tools/testing/selftests/net/tcp_ao/config
+++ b/tools/testing/selftests/net/tcp_ao/config
@@ -1,8 +1,7 @@
CONFIG_CRYPTO_CMAC=y
CONFIG_CRYPTO_HMAC=y
-CONFIG_CRYPTO_RMD160=y
CONFIG_CRYPTO_SHA1=y
CONFIG_IPV6=y
CONFIG_IPV6_MULTIPLE_TABLES=y
CONFIG_NET_L3_MASTER_DEV=y
CONFIG_NET_VRF=y
diff --git a/tools/testing/selftests/net/tcp_ao/key-management.c b/tools/testing/selftests/net/tcp_ao/key-management.c
index 69d9a7a05d5c..d86bb380b79f 100644
--- a/tools/testing/selftests/net/tcp_ao/key-management.c
+++ b/tools/testing/selftests/net/tcp_ao/key-management.c
@@ -378,35 +378,10 @@ static void check_listen_socket(void)
this_ip_dest, DEFAULT_TEST_PREFIX,
false, true, 20, 10, FAULT_CURRNEXT);
close(sk);
}
-static const char *fips_fpath = "/proc/sys/crypto/fips_enabled";
-static bool is_fips_enabled(void)
-{
- static int fips_checked = -1;
- FILE *fenabled;
- int enabled;
-
- if (fips_checked >= 0)
- return !!fips_checked;
- if (access(fips_fpath, R_OK)) {
- if (errno != ENOENT)
- test_error("Can't open %s", fips_fpath);
- fips_checked = 0;
- return false;
- }
- fenabled = fopen(fips_fpath, "r");
- if (!fenabled)
- test_error("Can't open %s", fips_fpath);
- if (fscanf(fenabled, "%d", &enabled) != 1)
- test_error("Can't read from %s", fips_fpath);
- fclose(fenabled);
- fips_checked = !!enabled;
- return !!fips_checked;
-}
-
struct test_key {
char password[TCP_AO_MAXKEYLEN];
const char *alg;
unsigned int len;
uint8_t client_keyid;
@@ -428,18 +403,11 @@ struct key_collection {
};
static struct key_collection collection;
#define TEST_MAX_MACLEN 16
-const char *test_algos[] = {
- "cmac(aes128)",
- "hmac(sha1)", "hmac(sha512)", "hmac(sha384)", "hmac(sha256)",
- "hmac(sha224)", "hmac(sha3-512)",
- /* only if !CONFIG_FIPS */
-#define TEST_NON_FIPS_ALGOS 2
- "hmac(rmd160)", "hmac(md5)"
-};
+const char *test_algos[] = { "cmac(aes128)", "hmac(sha1)", "hmac(sha256)" };
const unsigned int test_maclens[] = { 1, 4, 12, 16 };
#define MACLEN_SHIFT 2
#define ALGOS_SHIFT 4
static unsigned int make_mask(unsigned int shift, unsigned int prev_shift)
@@ -450,11 +418,11 @@ static unsigned int make_mask(unsigned int shift, unsigned int prev_shift)
}
static void init_key_in_collection(unsigned int index, bool randomized)
{
struct test_key *key = &collection.keys[index];
- unsigned int algos_nr, algos_index;
+ unsigned int algos_index;
/* Same for randomized and non-randomized test flows */
key->client_keyid = index;
key->server_keyid = 127 + index;
key->matches_client = 1;
@@ -472,14 +440,11 @@ static void init_key_in_collection(unsigned int index, bool randomized)
unsigned int shift = MACLEN_SHIFT;
key->maclen = test_maclens[index & make_mask(shift, 0)];
algos_index = index & make_mask(ALGOS_SHIFT, shift);
}
- algos_nr = ARRAY_SIZE(test_algos);
- if (is_fips_enabled())
- algos_nr -= TEST_NON_FIPS_ALGOS;
- key->alg = test_algos[algos_index % algos_nr];
+ key->alg = test_algos[algos_index % ARRAY_SIZE(test_algos)];
}
static int init_default_key_collection(unsigned int nr_keys, bool randomized)
{
size_t key_sz = sizeof(collection.keys[0]);
--
2.54.0
^ permalink raw reply related [flat|nested] 8+ messages in thread* [PATCH net-next v2 2/5] net/tcp-ao: Use crypto library API instead of crypto_ahash
2026-04-27 17:27 [PATCH net-next v2 0/5] Reimplement TCP-AO using crypto library Eric Biggers
2026-04-27 17:27 ` [PATCH net-next v2 1/5] net/tcp-ao: Drop support for most non-RFC-specified algorithms Eric Biggers
@ 2026-04-27 17:27 ` Eric Biggers
2026-04-27 17:27 ` [PATCH net-next v2 3/5] net/tcp-ao: Use stack-allocated MAC and traffic_key buffers Eric Biggers
` (3 subsequent siblings)
5 siblings, 0 replies; 8+ messages in thread
From: Eric Biggers @ 2026-04-27 17:27 UTC (permalink / raw)
To: netdev
Cc: linux-crypto, linux-kernel, Eric Dumazet, Neal Cardwell,
Kuniyuki Iwashima, David S . Miller, David Ahern, Jakub Kicinski,
Paolo Abeni, Simon Horman, Ard Biesheuvel, Jason A . Donenfeld,
Herbert Xu, Dmitry Safonov, Eric Biggers
Currently the kernel's TCP-AO implementation does the MAC and KDF
computations using the crypto_ahash API. This API is inefficient and
difficult to use, and it has required extensive workarounds in the form
of per-CPU preallocated objects (tcp_sigpool) to work at all.
Let's use lib/crypto/ instead. This means switching to straightforward
stack-allocated structures, virtually addressed buffers, and direct
function calls. It also means removing quite a bit of error handling.
This makes TCP-AO quite a bit faster.
This also enables many additional cleanups, which later commits will
handle: removing tcp-sigpool, removing support for crypto_tfm cloning,
removing more error handling, and replacing more dynamically-allocated
buffers with stack buffers based on the now-statically-known limits.
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Eric Biggers <ebiggers@kernel.org>
---
include/net/tcp_ao.h | 32 +-
net/ipv4/Kconfig | 5 +-
net/ipv4/tcp_ao.c | 523 +++++++++++-----------
net/ipv6/tcp_ao.c | 63 ++-
tools/testing/selftests/net/tcp_ao/config | 3 -
5 files changed, 309 insertions(+), 317 deletions(-)
diff --git a/include/net/tcp_ao.h b/include/net/tcp_ao.h
index 1e9e27d6e06b..20997aef3b0d 100644
--- a/include/net/tcp_ao.h
+++ b/include/net/tcp_ao.h
@@ -1,11 +1,10 @@
/* SPDX-License-Identifier: GPL-2.0-or-later */
#ifndef _TCP_AO_H
#define _TCP_AO_H
-#define TCP_AO_KEY_ALIGN 1
-#define __tcp_ao_key_align __aligned(TCP_AO_KEY_ALIGN)
+#include <crypto/sha2.h> /* for SHA256_DIGEST_SIZE */
union tcp_ao_addr {
struct in_addr a4;
#if IS_ENABLED(CONFIG_IPV6)
struct in6_addr a6;
@@ -30,15 +29,31 @@ struct tcp_ao_counters {
atomic64_t key_not_found;
atomic64_t ao_required;
atomic64_t dropped_icmp;
};
+enum tcp_ao_algo_id {
+ TCP_AO_ALGO_HMAC_SHA1 = 1, /* specified by RFC 5926 */
+ TCP_AO_ALGO_HMAC_SHA256, /* Linux extension */
+ TCP_AO_ALGO_AES_128_CMAC, /* specified by RFC 5926 */
+};
+
+/*
+ * This is the maximum untruncated MAC length, in bytes. Note that the MACs
+ * actually get truncated to 20 or fewer bytes to fit in the TCP options space.
+ */
+#define TCP_AO_MAX_MAC_LEN SHA256_DIGEST_SIZE
+
+#define TCP_AO_MAX_TRAFFIC_KEY_LEN SHA256_DIGEST_SIZE
+
+struct tcp_ao_mac_ctx;
+
struct tcp_ao_key {
struct hlist_node node;
union tcp_ao_addr addr;
- u8 key[TCP_AO_MAXKEYLEN] __tcp_ao_key_align;
- unsigned int tcp_sigpool_id;
+ u8 key[TCP_AO_MAXKEYLEN];
+ enum tcp_ao_algo_id algo;
unsigned int digest_size;
int l3index;
u8 prefixlen;
u8 family;
u8 keylen;
@@ -166,18 +181,19 @@ struct tcp6_ao_context {
__be16 dport;
__be32 sisn;
__be32 disn;
};
-struct tcp_sigpool;
/* Established states are fast-path and there always is current_key/rnext_key */
#define TCP_AO_ESTABLISHED (TCPF_ESTABLISHED | TCPF_FIN_WAIT1 | TCPF_FIN_WAIT2 | \
TCPF_CLOSE_WAIT | TCPF_LAST_ACK | TCPF_CLOSING)
int tcp_ao_transmit_skb(struct sock *sk, struct sk_buff *skb,
struct tcp_ao_key *key, struct tcphdr *th,
__u8 *hash_location);
+void tcp_ao_mac_update(struct tcp_ao_mac_ctx *mac_ctx, const void *data,
+ size_t data_len);
int tcp_ao_hash_skb(unsigned short int family,
char *ao_hash, struct tcp_ao_key *key,
const struct sock *sk, const struct sk_buff *skb,
const u8 *tkey, int hash_offset, u32 sne);
int tcp_parse_ao(struct sock *sk, int cmd, unsigned short int family,
@@ -186,12 +202,12 @@ struct tcp_ao_key *tcp_ao_established_key(const struct sock *sk,
struct tcp_ao_info *ao,
int sndid, int rcvid);
int tcp_ao_copy_all_matching(const struct sock *sk, struct sock *newsk,
struct request_sock *req, struct sk_buff *skb,
int family);
-int tcp_ao_calc_traffic_key(struct tcp_ao_key *mkt, u8 *key, void *ctx,
- unsigned int len, struct tcp_sigpool *hp);
+void tcp_ao_calc_traffic_key(const struct tcp_ao_key *mkt, u8 *traffic_key,
+ const void *input, unsigned int input_len);
void tcp_ao_destroy_sock(struct sock *sk, bool twsk);
void tcp_ao_time_wait(struct tcp_timewait_sock *tcptw, struct tcp_sock *tp);
bool tcp_ao_ignore_icmp(const struct sock *sk, int family, int type, int code);
int tcp_ao_get_mkts(struct sock *sk, sockptr_t optval, sockptr_t optlen);
int tcp_ao_get_sock_info(struct sock *sk, sockptr_t optval, sockptr_t optlen);
@@ -232,11 +248,11 @@ struct tcp_ao_key *tcp_v4_ao_lookup_rsk(const struct sock *sk,
int sndid, int rcvid);
int tcp_v4_ao_hash_skb(char *ao_hash, struct tcp_ao_key *key,
const struct sock *sk, const struct sk_buff *skb,
const u8 *tkey, int hash_offset, u32 sne);
/* ipv6 specific functions */
-int tcp_v6_ao_hash_pseudoheader(struct tcp_sigpool *hp,
+int tcp_v6_ao_hash_pseudoheader(struct tcp_ao_mac_ctx *mac_ctx,
const struct in6_addr *daddr,
const struct in6_addr *saddr, int nbytes);
int tcp_v6_ao_calc_key_skb(struct tcp_ao_key *mkt, u8 *key,
const struct sk_buff *skb, __be32 sisn, __be32 disn);
int tcp_v6_ao_calc_key_sk(struct tcp_ao_key *mkt, u8 *key,
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index 21e5164e30db..77b053b445a0 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -744,13 +744,14 @@ config DEFAULT_TCP_CONG
config TCP_SIGPOOL
tristate
config TCP_AO
bool "TCP: Authentication Option (RFC5925)"
- select CRYPTO
+ select CRYPTO_LIB_AES_CBC_MACS
+ select CRYPTO_LIB_SHA1
+ select CRYPTO_LIB_SHA256
select CRYPTO_LIB_UTILS
- select TCP_SIGPOOL
depends on 64BIT # seq-number extension needs WRITE_ONCE(u64)
help
TCP-AO specifies the use of stronger Message Authentication Codes (MACs),
protects against replays for long-lived TCP connections, and
provides more details on the association of security with TCP
diff --git a/net/ipv4/tcp_ao.c b/net/ipv4/tcp_ao.c
index b21bd69b4e82..0d24cbd66c9a 100644
--- a/net/ipv4/tcp_ao.c
+++ b/net/ipv4/tcp_ao.c
@@ -7,11 +7,13 @@
* Francesco Ruggeri <fruggeri@arista.com>
* Salam Noureddine <noureddine@arista.com>
*/
#define pr_fmt(fmt) "TCP: " fmt
-#include <crypto/hash.h>
+#include <crypto/aes-cbc-macs.h>
+#include <crypto/sha1.h>
+#include <crypto/sha2.h>
#include <crypto/utils.h>
#include <linux/inetdevice.h>
#include <linux/tcp.h>
#include <net/tcp.h>
@@ -19,36 +21,137 @@
#include <net/icmp.h>
#include <trace/events/tcp.h>
DEFINE_STATIC_KEY_DEFERRED_FALSE(tcp_ao_needed, HZ);
-int tcp_ao_calc_traffic_key(struct tcp_ao_key *mkt, u8 *key, void *ctx,
- unsigned int len, struct tcp_sigpool *hp)
-{
- struct scatterlist sg;
- int ret;
+static const struct tcp_ao_algo {
+ const char *name;
+ unsigned int digest_size;
+} tcp_ao_algos[] = {
+ [TCP_AO_ALGO_HMAC_SHA1] = {
+ .name = "hmac(sha1)",
+ .digest_size = SHA1_DIGEST_SIZE,
+ },
+ [TCP_AO_ALGO_HMAC_SHA256] = {
+ .name = "hmac(sha256)",
+ .digest_size = SHA256_DIGEST_SIZE,
+ },
+ [TCP_AO_ALGO_AES_128_CMAC] = {
+ .name = "cmac(aes128)",
+ .digest_size = AES_BLOCK_SIZE, /* same as AES_KEYSIZE_128 */
+ },
+};
+
+struct tcp_ao_mac_ctx {
+ enum tcp_ao_algo_id algo;
+ union {
+ struct hmac_sha1_ctx hmac_sha1;
+ struct hmac_sha256_ctx hmac_sha256;
+ struct {
+ struct aes_cmac_key key;
+ struct aes_cmac_ctx ctx;
+ } aes_cmac;
+ };
+};
+
+static const struct tcp_ao_algo *tcp_ao_find_algo(const char *name)
+{
+ for (size_t i = 0; i < ARRAY_SIZE(tcp_ao_algos); i++) {
+ const struct tcp_ao_algo *algo = &tcp_ao_algos[i];
+
+ if (!algo->name)
+ continue;
+ if (WARN_ON_ONCE(algo->digest_size > TCP_AO_MAX_MAC_LEN ||
+ algo->digest_size >
+ TCP_AO_MAX_TRAFFIC_KEY_LEN))
+ continue;
+ if (strcmp(name, algo->name) == 0)
+ return algo;
+ }
+ return NULL;
+}
- if (crypto_ahash_setkey(crypto_ahash_reqtfm(hp->req),
- mkt->key, mkt->keylen))
- goto clear_hash;
+static void tcp_ao_mac_init(struct tcp_ao_mac_ctx *mac_ctx,
+ enum tcp_ao_algo_id algo, const u8 *traffic_key)
+{
+ mac_ctx->algo = algo;
+ switch (mac_ctx->algo) {
+ case TCP_AO_ALGO_HMAC_SHA1:
+ hmac_sha1_init_usingrawkey(&mac_ctx->hmac_sha1, traffic_key,
+ SHA1_DIGEST_SIZE);
+ return;
+ case TCP_AO_ALGO_HMAC_SHA256:
+ hmac_sha256_init_usingrawkey(&mac_ctx->hmac_sha256, traffic_key,
+ SHA256_DIGEST_SIZE);
+ return;
+ case TCP_AO_ALGO_AES_128_CMAC:
+ aes_cmac_preparekey(&mac_ctx->aes_cmac.key, traffic_key,
+ AES_KEYSIZE_128);
+ aes_cmac_init(&mac_ctx->aes_cmac.ctx, &mac_ctx->aes_cmac.key);
+ return;
+ default:
+ WARN_ON_ONCE(1); /* algo was validated earlier. */
+ }
+}
- ret = crypto_ahash_init(hp->req);
- if (ret)
- goto clear_hash;
+void tcp_ao_mac_update(struct tcp_ao_mac_ctx *mac_ctx, const void *data,
+ size_t data_len)
+{
+ switch (mac_ctx->algo) {
+ case TCP_AO_ALGO_HMAC_SHA1:
+ hmac_sha1_update(&mac_ctx->hmac_sha1, data, data_len);
+ return;
+ case TCP_AO_ALGO_HMAC_SHA256:
+ hmac_sha256_update(&mac_ctx->hmac_sha256, data, data_len);
+ return;
+ case TCP_AO_ALGO_AES_128_CMAC:
+ aes_cmac_update(&mac_ctx->aes_cmac.ctx, data, data_len);
+ return;
+ default:
+ WARN_ON_ONCE(1); /* algo was validated earlier. */
+ }
+}
- sg_init_one(&sg, ctx, len);
- ahash_request_set_crypt(hp->req, &sg, key, len);
- crypto_ahash_update(hp->req);
+static void tcp_ao_mac_final(struct tcp_ao_mac_ctx *mac_ctx, u8 *out)
+{
+ switch (mac_ctx->algo) {
+ case TCP_AO_ALGO_HMAC_SHA1:
+ hmac_sha1_final(&mac_ctx->hmac_sha1, out);
+ return;
+ case TCP_AO_ALGO_HMAC_SHA256:
+ hmac_sha256_final(&mac_ctx->hmac_sha256, out);
+ return;
+ case TCP_AO_ALGO_AES_128_CMAC:
+ aes_cmac_final(&mac_ctx->aes_cmac.ctx, out);
+ return;
+ default:
+ WARN_ON_ONCE(1); /* algo was validated earlier. */
+ }
+}
- ret = crypto_ahash_final(hp->req);
- if (ret)
- goto clear_hash;
+void tcp_ao_calc_traffic_key(const struct tcp_ao_key *mkt, u8 *traffic_key,
+ const void *input, unsigned int input_len)
+{
+ switch (mkt->algo) {
+ case TCP_AO_ALGO_HMAC_SHA1:
+ hmac_sha1_usingrawkey(mkt->key, mkt->keylen, input, input_len,
+ traffic_key);
+ return;
+ case TCP_AO_ALGO_HMAC_SHA256:
+ hmac_sha256_usingrawkey(mkt->key, mkt->keylen, input, input_len,
+ traffic_key);
+ return;
+ case TCP_AO_ALGO_AES_128_CMAC: {
+ struct aes_cmac_key k;
- return 0;
-clear_hash:
- memset(key, 0, tcp_ao_digest_size(mkt));
- return 1;
+ aes_cmac_preparekey(&k, mkt->key, AES_KEYSIZE_128);
+ aes_cmac(&k, input, input_len, traffic_key);
+ return;
+ }
+ default:
+ WARN_ON_ONCE(1); /* algo was validated earlier. */
+ }
}
bool tcp_ao_ignore_icmp(const struct sock *sk, int family, int type, int code)
{
bool ignore_icmp = false;
@@ -252,33 +355,30 @@ static struct tcp_ao_key *tcp_ao_copy_key(struct sock *sk,
if (!new_key)
return NULL;
*new_key = *key;
INIT_HLIST_NODE(&new_key->node);
- tcp_sigpool_get(new_key->tcp_sigpool_id);
atomic64_set(&new_key->pkt_good, 0);
atomic64_set(&new_key->pkt_bad, 0);
return new_key;
}
static void tcp_ao_key_free_rcu(struct rcu_head *head)
{
struct tcp_ao_key *key = container_of(head, struct tcp_ao_key, rcu);
- tcp_sigpool_release(key->tcp_sigpool_id);
kfree_sensitive(key);
}
static void tcp_ao_info_free(struct tcp_ao_info *ao)
{
struct tcp_ao_key *key;
struct hlist_node *n;
hlist_for_each_entry_safe(key, n, &ao->head, node) {
hlist_del(&key->node);
- tcp_sigpool_release(key->tcp_sigpool_id);
kfree_sensitive(key);
}
kfree(ao);
static_branch_slow_dec_deferred(&tcp_ao_needed);
}
@@ -344,33 +444,26 @@ static int tcp_v4_ao_calc_key(struct tcp_ao_key *mkt, u8 *key,
struct kdf_input_block {
u8 counter;
u8 label[6];
struct tcp4_ao_context ctx;
__be16 outlen;
- } __packed * tmp;
- struct tcp_sigpool hp;
- int err;
-
- err = tcp_sigpool_start(mkt->tcp_sigpool_id, &hp);
- if (err)
- return err;
-
- tmp = hp.scratch;
- tmp->counter = 1;
- memcpy(tmp->label, "TCP-AO", 6);
- tmp->ctx.saddr = saddr;
- tmp->ctx.daddr = daddr;
- tmp->ctx.sport = sport;
- tmp->ctx.dport = dport;
- tmp->ctx.sisn = sisn;
- tmp->ctx.disn = disn;
- tmp->outlen = htons(tcp_ao_digest_size(mkt) * 8); /* in bits */
-
- err = tcp_ao_calc_traffic_key(mkt, key, tmp, sizeof(*tmp), &hp);
- tcp_sigpool_end(&hp);
-
- return err;
+ } __packed input = {
+ .counter = 1,
+ .label = "TCP-AO",
+ .ctx = {
+ .saddr = saddr,
+ .daddr = daddr,
+ .sport = sport,
+ .dport = dport,
+ .sisn = sisn,
+ .disn = disn,
+ },
+ .outlen = htons(tcp_ao_digest_size(mkt) * 8), /* in bits */
+ };
+
+ tcp_ao_calc_traffic_key(mkt, key, &input, sizeof(input));
+ return 0;
}
int tcp_v4_ao_calc_key_sk(struct tcp_ao_key *mkt, u8 *key,
const struct sock *sk,
__be32 sisn, __be32 disn, bool send)
@@ -433,60 +526,57 @@ static int tcp_ao_calc_key_skb(struct tcp_ao_key *mkt, u8 *key,
return tcp_v6_ao_calc_key_skb(mkt, key, skb, sisn, disn);
#endif
return -EAFNOSUPPORT;
}
-static int tcp_v4_ao_hash_pseudoheader(struct tcp_sigpool *hp,
+static int tcp_v4_ao_hash_pseudoheader(struct tcp_ao_mac_ctx *mac_ctx,
__be32 daddr, __be32 saddr,
int nbytes)
{
- struct tcp4_pseudohdr *bp;
- struct scatterlist sg;
+ struct tcp4_pseudohdr phdr = {
+ .saddr = saddr,
+ .daddr = daddr,
+ .pad = 0,
+ .protocol = IPPROTO_TCP,
+ .len = cpu_to_be16(nbytes),
+ };
- bp = hp->scratch;
- bp->saddr = saddr;
- bp->daddr = daddr;
- bp->pad = 0;
- bp->protocol = IPPROTO_TCP;
- bp->len = cpu_to_be16(nbytes);
-
- sg_init_one(&sg, bp, sizeof(*bp));
- ahash_request_set_crypt(hp->req, &sg, NULL, sizeof(*bp));
- return crypto_ahash_update(hp->req);
+ tcp_ao_mac_update(mac_ctx, &phdr, sizeof(phdr));
+ return 0;
}
static int tcp_ao_hash_pseudoheader(unsigned short int family,
const struct sock *sk,
const struct sk_buff *skb,
- struct tcp_sigpool *hp, int nbytes)
+ struct tcp_ao_mac_ctx *mac_ctx, int nbytes)
{
const struct tcphdr *th = tcp_hdr(skb);
/* TODO: Can we rely on checksum being zero to mean outbound pkt? */
if (!th->check) {
if (family == AF_INET)
- return tcp_v4_ao_hash_pseudoheader(hp, sk->sk_daddr,
+ return tcp_v4_ao_hash_pseudoheader(mac_ctx, sk->sk_daddr,
sk->sk_rcv_saddr, skb->len);
#if IS_ENABLED(CONFIG_IPV6)
else if (family == AF_INET6)
- return tcp_v6_ao_hash_pseudoheader(hp, &sk->sk_v6_daddr,
+ return tcp_v6_ao_hash_pseudoheader(mac_ctx, &sk->sk_v6_daddr,
&sk->sk_v6_rcv_saddr, skb->len);
#endif
else
return -EAFNOSUPPORT;
}
if (family == AF_INET) {
const struct iphdr *iph = ip_hdr(skb);
- return tcp_v4_ao_hash_pseudoheader(hp, iph->daddr,
+ return tcp_v4_ao_hash_pseudoheader(mac_ctx, iph->daddr,
iph->saddr, skb->len);
#if IS_ENABLED(CONFIG_IPV6)
} else if (family == AF_INET6) {
const struct ipv6hdr *iph = ipv6_hdr(skb);
- return tcp_v6_ao_hash_pseudoheader(hp, &iph->daddr,
+ return tcp_v6_ao_hash_pseudoheader(mac_ctx, &iph->daddr,
&iph->saddr, skb->len);
#endif
}
return -EAFNOSUPPORT;
}
@@ -504,35 +594,24 @@ u32 tcp_ao_compute_sne(u32 next_sne, u32 next_seq, u32 seq)
}
return sne;
}
-/* tcp_ao_hash_sne(struct tcp_sigpool *hp)
- * @hp - used for hashing
- * @sne - sne value
- */
-static int tcp_ao_hash_sne(struct tcp_sigpool *hp, u32 sne)
+static void tcp_ao_hash_sne(struct tcp_ao_mac_ctx *mac_ctx, u32 sne)
{
- struct scatterlist sg;
- __be32 *bp;
-
- bp = (__be32 *)hp->scratch;
- *bp = htonl(sne);
+ __be32 sne_be32 = htonl(sne);
- sg_init_one(&sg, bp, sizeof(*bp));
- ahash_request_set_crypt(hp->req, &sg, NULL, sizeof(*bp));
- return crypto_ahash_update(hp->req);
+ tcp_ao_mac_update(mac_ctx, &sne_be32, sizeof(sne_be32));
}
-static int tcp_ao_hash_header(struct tcp_sigpool *hp,
- const struct tcphdr *th,
- bool exclude_options, u8 *hash,
- int hash_offset, int hash_len)
+static void tcp_ao_hash_header(struct tcp_ao_mac_ctx *mac_ctx,
+ const struct tcphdr *th, bool exclude_options,
+ u8 *hash, int hash_offset, int hash_len)
{
- struct scatterlist sg;
- u8 *hdr = hp->scratch;
- int err, len;
+ /* Full TCP header (th->doff << 2) should fit into scratch area. */
+ u8 hdr[60];
+ int len;
/* We are not allowed to change tcphdr, make a local copy */
if (exclude_options) {
len = sizeof(*th) + sizeof(struct tcp_ao_hdr) + hash_len;
memcpy(hdr, th, sizeof(*th));
@@ -548,126 +627,105 @@ static int tcp_ao_hash_header(struct tcp_sigpool *hp,
/* zero out tcp-ao hash */
((struct tcphdr *)hdr)->check = 0;
memset(hdr + hash_offset, 0, hash_len);
}
- sg_init_one(&sg, hdr, len);
- ahash_request_set_crypt(hp->req, &sg, NULL, len);
- err = crypto_ahash_update(hp->req);
- WARN_ON_ONCE(err != 0);
- return err;
+ tcp_ao_mac_update(mac_ctx, hdr, len);
}
int tcp_ao_hash_hdr(unsigned short int family, char *ao_hash,
struct tcp_ao_key *key, const u8 *tkey,
const union tcp_ao_addr *daddr,
const union tcp_ao_addr *saddr,
const struct tcphdr *th, u32 sne)
{
- int tkey_len = tcp_ao_digest_size(key);
int hash_offset = ao_hash - (char *)th;
- struct tcp_sigpool hp;
- void *hash_buf = NULL;
-
- hash_buf = kmalloc(tkey_len, GFP_ATOMIC);
- if (!hash_buf)
- goto clear_hash_noput;
-
- if (tcp_sigpool_start(key->tcp_sigpool_id, &hp))
- goto clear_hash_noput;
-
- if (crypto_ahash_setkey(crypto_ahash_reqtfm(hp.req), tkey, tkey_len))
- goto clear_hash;
-
- if (crypto_ahash_init(hp.req))
- goto clear_hash;
+ struct tcp_ao_mac_ctx mac_ctx;
+ u8 hash_buf[TCP_AO_MAX_MAC_LEN];
- if (tcp_ao_hash_sne(&hp, sne))
- goto clear_hash;
+ tcp_ao_mac_init(&mac_ctx, key->algo, tkey);
+ tcp_ao_hash_sne(&mac_ctx, sne);
if (family == AF_INET) {
- if (tcp_v4_ao_hash_pseudoheader(&hp, daddr->a4.s_addr,
- saddr->a4.s_addr, th->doff * 4))
- goto clear_hash;
+ tcp_v4_ao_hash_pseudoheader(&mac_ctx, daddr->a4.s_addr,
+ saddr->a4.s_addr, th->doff * 4);
#if IS_ENABLED(CONFIG_IPV6)
} else if (family == AF_INET6) {
- if (tcp_v6_ao_hash_pseudoheader(&hp, &daddr->a6,
- &saddr->a6, th->doff * 4))
- goto clear_hash;
+ tcp_v6_ao_hash_pseudoheader(&mac_ctx, &daddr->a6,
+ &saddr->a6, th->doff * 4);
#endif
} else {
WARN_ON_ONCE(1);
goto clear_hash;
}
- if (tcp_ao_hash_header(&hp, th,
- !!(key->keyflags & TCP_AO_KEYF_EXCLUDE_OPT),
- ao_hash, hash_offset, tcp_ao_maclen(key)))
- goto clear_hash;
- ahash_request_set_crypt(hp.req, NULL, hash_buf, 0);
- if (crypto_ahash_final(hp.req))
- goto clear_hash;
+ tcp_ao_hash_header(&mac_ctx, th,
+ !!(key->keyflags & TCP_AO_KEYF_EXCLUDE_OPT),
+ ao_hash, hash_offset, tcp_ao_maclen(key));
+ tcp_ao_mac_final(&mac_ctx, hash_buf);
memcpy(ao_hash, hash_buf, tcp_ao_maclen(key));
- tcp_sigpool_end(&hp);
- kfree(hash_buf);
return 0;
clear_hash:
- tcp_sigpool_end(&hp);
-clear_hash_noput:
memset(ao_hash, 0, tcp_ao_maclen(key));
- kfree(hash_buf);
return 1;
}
+static void tcp_ao_hash_skb_data(struct tcp_ao_mac_ctx *mac_ctx,
+ const struct sk_buff *skb,
+ unsigned int header_len)
+{
+ const unsigned int head_data_len = skb_headlen(skb) > header_len ?
+ skb_headlen(skb) - header_len : 0;
+ const struct skb_shared_info *shi = skb_shinfo(skb);
+ struct sk_buff *frag_iter;
+ unsigned int i;
+
+ tcp_ao_mac_update(mac_ctx, (const u8 *)tcp_hdr(skb) + header_len,
+ head_data_len);
+
+ for (i = 0; i < shi->nr_frags; ++i) {
+ const skb_frag_t *f = &shi->frags[i];
+ u32 p_off, p_len, copied;
+ const void *vaddr;
+ struct page *p;
+
+ skb_frag_foreach_page(f, skb_frag_off(f), skb_frag_size(f),
+ p, p_off, p_len, copied) {
+ vaddr = kmap_local_page(p);
+ tcp_ao_mac_update(mac_ctx, vaddr + p_off, p_len);
+ kunmap_local(vaddr);
+ }
+ }
+
+ skb_walk_frags(skb, frag_iter)
+ tcp_ao_hash_skb_data(mac_ctx, frag_iter, 0);
+}
+
int tcp_ao_hash_skb(unsigned short int family,
char *ao_hash, struct tcp_ao_key *key,
const struct sock *sk, const struct sk_buff *skb,
const u8 *tkey, int hash_offset, u32 sne)
{
const struct tcphdr *th = tcp_hdr(skb);
- int tkey_len = tcp_ao_digest_size(key);
- struct tcp_sigpool hp;
- void *hash_buf = NULL;
-
- hash_buf = kmalloc(tkey_len, GFP_ATOMIC);
- if (!hash_buf)
- goto clear_hash_noput;
-
- if (tcp_sigpool_start(key->tcp_sigpool_id, &hp))
- goto clear_hash_noput;
-
- if (crypto_ahash_setkey(crypto_ahash_reqtfm(hp.req), tkey, tkey_len))
- goto clear_hash;
-
- /* For now use sha1 by default. Depends on alg in tcp_ao_key */
- if (crypto_ahash_init(hp.req))
- goto clear_hash;
+ struct tcp_ao_mac_ctx mac_ctx;
+ u8 hash_buf[TCP_AO_MAX_MAC_LEN];
- if (tcp_ao_hash_sne(&hp, sne))
- goto clear_hash;
- if (tcp_ao_hash_pseudoheader(family, sk, skb, &hp, skb->len))
- goto clear_hash;
- if (tcp_ao_hash_header(&hp, th,
- !!(key->keyflags & TCP_AO_KEYF_EXCLUDE_OPT),
- ao_hash, hash_offset, tcp_ao_maclen(key)))
- goto clear_hash;
- if (tcp_sigpool_hash_skb_data(&hp, skb, th->doff << 2))
- goto clear_hash;
- ahash_request_set_crypt(hp.req, NULL, hash_buf, 0);
- if (crypto_ahash_final(hp.req))
+ tcp_ao_mac_init(&mac_ctx, key->algo, tkey);
+ tcp_ao_hash_sne(&mac_ctx, sne);
+ if (tcp_ao_hash_pseudoheader(family, sk, skb, &mac_ctx, skb->len))
goto clear_hash;
+ tcp_ao_hash_header(&mac_ctx, th,
+ !!(key->keyflags & TCP_AO_KEYF_EXCLUDE_OPT),
+ ao_hash, hash_offset, tcp_ao_maclen(key));
+ tcp_ao_hash_skb_data(&mac_ctx, skb, th->doff << 2);
+ tcp_ao_mac_final(&mac_ctx, hash_buf);
memcpy(ao_hash, hash_buf, tcp_ao_maclen(key));
- tcp_sigpool_end(&hp);
- kfree(hash_buf);
return 0;
clear_hash:
- tcp_sigpool_end(&hp);
-clear_hash_noput:
memset(ao_hash, 0, tcp_ao_maclen(key));
- kfree(hash_buf);
return 1;
}
int tcp_v4_ao_hash_skb(char *ao_hash, struct tcp_ao_key *key,
const struct sock *sk, const struct sk_buff *skb,
@@ -1278,11 +1336,10 @@ int tcp_ao_copy_all_matching(const struct sock *sk, struct sock *newsk,
return 0;
free_and_exit:
hlist_for_each_entry_safe(key, key_head, &new_ao->head, node) {
hlist_del(&key->node);
- tcp_sigpool_release(key->tcp_sigpool_id);
atomic_sub(tcp_ao_sizeof_key(key), &newsk->sk_omem_alloc);
kfree_sensitive(key);
}
free_ao:
kfree(new_ao);
@@ -1334,27 +1391,14 @@ static int tcp_ao_verify_ipv4(struct sock *sk, struct tcp_ao_add *cmd,
*addr = (union tcp_ao_addr *)&sin->sin_addr;
return 0;
}
-static int tcp_ao_parse_crypto(struct tcp_ao_add *cmd, struct tcp_ao_key *key)
+static int tcp_ao_parse_crypto(const struct tcp_ao_add *cmd,
+ struct tcp_ao_key *key)
{
unsigned int syn_tcp_option_space;
- bool is_kdf_aes_128_cmac = false;
- struct crypto_ahash *tfm;
- struct tcp_sigpool hp;
- void *tmp_key = NULL;
- int err;
-
- /* RFC5926, 3.1.1.2. KDF_AES_128_CMAC */
- if (!strcmp("cmac(aes128)", cmd->alg_name)) {
- strscpy(cmd->alg_name, "cmac(aes)", sizeof(cmd->alg_name));
- is_kdf_aes_128_cmac = (cmd->keylen != 16);
- tmp_key = kmalloc(cmd->keylen, GFP_KERNEL);
- if (!tmp_key)
- return -ENOMEM;
- }
key->maclen = cmd->maclen ?: 12; /* 12 is the default in RFC5925 */
/* Check: maclen + tcp-ao header <= (MAX_TCP_OPTION_SPACE - mss
* - tstamp (including sackperm)
@@ -1386,68 +1430,31 @@ static int tcp_ao_parse_crypto(struct tcp_ao_add *cmd, struct tcp_ao_key *key)
*/
syn_tcp_option_space = MAX_TCP_OPTION_SPACE;
syn_tcp_option_space -= TCPOLEN_MSS_ALIGNED;
syn_tcp_option_space -= TCPOLEN_TSTAMP_ALIGNED;
syn_tcp_option_space -= TCPOLEN_WSCALE_ALIGNED;
- if (tcp_ao_len_aligned(key) > syn_tcp_option_space) {
- err = -EMSGSIZE;
- goto err_kfree;
- }
-
- key->keylen = cmd->keylen;
- memcpy(key->key, cmd->key, cmd->keylen);
-
- err = tcp_sigpool_start(key->tcp_sigpool_id, &hp);
- if (err)
- goto err_kfree;
-
- tfm = crypto_ahash_reqtfm(hp.req);
- if (is_kdf_aes_128_cmac) {
- void *scratch = hp.scratch;
- struct scatterlist sg;
-
- memcpy(tmp_key, cmd->key, cmd->keylen);
- sg_init_one(&sg, tmp_key, cmd->keylen);
-
- /* Using zero-key of 16 bytes as described in RFC5926 */
- memset(scratch, 0, 16);
- err = crypto_ahash_setkey(tfm, scratch, 16);
- if (err)
- goto err_pool_end;
-
- err = crypto_ahash_init(hp.req);
- if (err)
- goto err_pool_end;
-
- ahash_request_set_crypt(hp.req, &sg, key->key, cmd->keylen);
- err = crypto_ahash_update(hp.req);
- if (err)
- goto err_pool_end;
-
- err |= crypto_ahash_final(hp.req);
- if (err)
- goto err_pool_end;
- key->keylen = 16;
+ if (tcp_ao_len_aligned(key) > syn_tcp_option_space)
+ return -EMSGSIZE;
+
+ if (key->algo == TCP_AO_ALGO_AES_128_CMAC &&
+ cmd->keylen != AES_KEYSIZE_128) {
+ /* RFC5926, 3.1.1.2. KDF_AES_128_CMAC */
+ static const u8 zeroes[AES_KEYSIZE_128];
+ struct aes_cmac_key extractor;
+
+ aes_cmac_preparekey(&extractor, zeroes, AES_KEYSIZE_128);
+ aes_cmac(&extractor, cmd->key, cmd->keylen, key->key);
+ key->keylen = AES_KEYSIZE_128;
+ } else {
+ memcpy(key->key, cmd->key, cmd->keylen);
+ key->keylen = cmd->keylen;
}
- err = crypto_ahash_setkey(tfm, key->key, key->keylen);
- if (err)
- goto err_pool_end;
-
- tcp_sigpool_end(&hp);
- kfree_sensitive(tmp_key);
-
if (tcp_ao_maclen(key) > key->digest_size)
return -EINVAL;
return 0;
-
-err_pool_end:
- tcp_sigpool_end(&hp);
-err_kfree:
- kfree_sensitive(tmp_key);
- return err;
}
#if IS_ENABLED(CONFIG_IPV6)
static int tcp_ao_verify_ipv6(struct sock *sk, struct tcp_ao_add *cmd,
union tcp_ao_addr **paddr,
@@ -1547,58 +1554,37 @@ static struct tcp_ao_info *getsockopt_ao_info(struct sock *sk)
#define TCP_AO_GET_KEYF_VALID (TCP_AO_KEYF_IFINDEX)
static struct tcp_ao_key *tcp_ao_key_alloc(struct sock *sk,
struct tcp_ao_add *cmd)
{
- const char *algo = cmd->alg_name;
- unsigned int digest_size;
- struct crypto_ahash *tfm;
+ const struct tcp_ao_algo *algo;
struct tcp_ao_key *key;
- struct tcp_sigpool hp;
- int err, pool_id;
size_t size;
/* Force null-termination of alg_name */
cmd->alg_name[ARRAY_SIZE(cmd->alg_name) - 1] = '\0';
- /* RFC5926, 3.1.1.2. KDF_AES_128_CMAC */
- if (!strcmp("cmac(aes128)", algo))
- algo = "cmac(aes)";
- else if (strcmp("hmac(sha1)", algo) &&
- strcmp("hmac(sha256)", algo) &&
- (strcmp("cmac(aes)", algo) || cmd->keylen != 16))
- return ERR_PTR(-ENOENT);
-
- /* Full TCP header (th->doff << 2) should fit into scratch area,
- * see tcp_ao_hash_header().
+ /*
+ * For backwards compatibility, accept "cmac(aes)" as an alias for
+ * "cmac(aes128)", provided that the key length is exactly 128 bits.
*/
- pool_id = tcp_sigpool_alloc_ahash(algo, 60);
- if (pool_id < 0)
- return ERR_PTR(pool_id);
-
- err = tcp_sigpool_start(pool_id, &hp);
- if (err)
- goto err_free_pool;
+ if (strcmp(cmd->alg_name, "cmac(aes)") == 0 &&
+ cmd->keylen == AES_KEYSIZE_128)
+ strscpy(cmd->alg_name, "cmac(aes128)");
- tfm = crypto_ahash_reqtfm(hp.req);
- digest_size = crypto_ahash_digestsize(tfm);
- tcp_sigpool_end(&hp);
+ algo = tcp_ao_find_algo(cmd->alg_name);
+ if (!algo)
+ return ERR_PTR(-ENOENT);
- size = sizeof(struct tcp_ao_key) + (digest_size << 1);
+ size = sizeof(struct tcp_ao_key) + (algo->digest_size << 1);
key = sock_kmalloc(sk, size, GFP_KERNEL);
- if (!key) {
- err = -ENOMEM;
- goto err_free_pool;
- }
+ if (!key)
+ return ERR_PTR(-ENOMEM);
- key->tcp_sigpool_id = pool_id;
- key->digest_size = digest_size;
+ key->algo = algo - tcp_ao_algos;
+ key->digest_size = algo->digest_size;
return key;
-
-err_free_pool:
- tcp_sigpool_release(pool_id);
- return ERR_PTR(err);
}
static int tcp_ao_add_cmd(struct sock *sk, unsigned short int family,
sockptr_t optval, int optlen)
{
@@ -1755,11 +1741,10 @@ static int tcp_ao_add_cmd(struct sock *sk, unsigned short int family,
WRITE_ONCE(ao_info->rnext_key, key);
return 0;
err_free_sock:
atomic_sub(tcp_ao_sizeof_key(key), &sk->sk_omem_alloc);
- tcp_sigpool_release(key->tcp_sigpool_id);
kfree_sensitive(key);
err_free_ao:
if (first)
kfree(ao_info);
return ret;
@@ -2286,11 +2271,15 @@ static int tcp_ao_copy_mkts_to_user(const struct sock *sk,
opt_out.keylen = key->keylen;
opt_out.ifindex = key->l3index;
opt_out.pkt_good = atomic64_read(&key->pkt_good);
opt_out.pkt_bad = atomic64_read(&key->pkt_bad);
memcpy(&opt_out.key, key->key, key->keylen);
- tcp_sigpool_algo(key->tcp_sigpool_id, opt_out.alg_name, 64);
+ if (key->algo == TCP_AO_ALGO_AES_128_CMAC)
+ /* This is needed for backwards compatibility. */
+ strscpy(opt_out.alg_name, "cmac(aes)");
+ else
+ strscpy(opt_out.alg_name, tcp_ao_algos[key->algo].name);
/* Copy key to user */
if (copy_to_sockptr_offset(optval, out_offset,
&opt_out, bytes_to_write))
return -EFAULT;
diff --git a/net/ipv6/tcp_ao.c b/net/ipv6/tcp_ao.c
index 3c09ac26206e..2dcfe9dda7f4 100644
--- a/net/ipv6/tcp_ao.c
+++ b/net/ipv6/tcp_ao.c
@@ -5,11 +5,10 @@
*
* Authors: Dmitry Safonov <dima@arista.com>
* Francesco Ruggeri <fruggeri@arista.com>
* Salam Noureddine <noureddine@arista.com>
*/
-#include <crypto/hash.h>
#include <linux/tcp.h>
#include <net/tcp.h>
#include <net/ipv6.h>
@@ -22,33 +21,26 @@ static int tcp_v6_ao_calc_key(struct tcp_ao_key *mkt, u8 *key,
struct kdf_input_block {
u8 counter;
u8 label[6];
struct tcp6_ao_context ctx;
__be16 outlen;
- } __packed * tmp;
- struct tcp_sigpool hp;
- int err;
-
- err = tcp_sigpool_start(mkt->tcp_sigpool_id, &hp);
- if (err)
- return err;
-
- tmp = hp.scratch;
- tmp->counter = 1;
- memcpy(tmp->label, "TCP-AO", 6);
- tmp->ctx.saddr = *saddr;
- tmp->ctx.daddr = *daddr;
- tmp->ctx.sport = sport;
- tmp->ctx.dport = dport;
- tmp->ctx.sisn = sisn;
- tmp->ctx.disn = disn;
- tmp->outlen = htons(tcp_ao_digest_size(mkt) * 8); /* in bits */
-
- err = tcp_ao_calc_traffic_key(mkt, key, tmp, sizeof(*tmp), &hp);
- tcp_sigpool_end(&hp);
-
- return err;
+ } __packed input = {
+ .counter = 1,
+ .label = "TCP-AO",
+ .ctx = {
+ .saddr = *saddr,
+ .daddr = *daddr,
+ .sport = sport,
+ .dport = dport,
+ .sisn = sisn,
+ .disn = disn,
+ },
+ .outlen = htons(tcp_ao_digest_size(mkt) * 8), /* in bits */
+ };
+
+ tcp_ao_calc_traffic_key(mkt, key, &input, sizeof(input));
+ return 0;
}
int tcp_v6_ao_calc_key_skb(struct tcp_ao_key *mkt, u8 *key,
const struct sk_buff *skb,
__be32 sisn, __be32 disn)
@@ -110,27 +102,24 @@ struct tcp_ao_key *tcp_v6_ao_lookup_rsk(const struct sock *sk,
l3index = l3mdev_master_ifindex_by_index(sock_net(sk), ireq->ir_iif);
return tcp_ao_do_lookup(sk, l3index, (union tcp_ao_addr *)addr,
AF_INET6, sndid, rcvid);
}
-int tcp_v6_ao_hash_pseudoheader(struct tcp_sigpool *hp,
+int tcp_v6_ao_hash_pseudoheader(struct tcp_ao_mac_ctx *mac_ctx,
const struct in6_addr *daddr,
const struct in6_addr *saddr, int nbytes)
{
- struct tcp6_pseudohdr *bp;
- struct scatterlist sg;
-
- bp = hp->scratch;
/* 1. TCP pseudo-header (RFC2460) */
- bp->saddr = *saddr;
- bp->daddr = *daddr;
- bp->len = cpu_to_be32(nbytes);
- bp->protocol = cpu_to_be32(IPPROTO_TCP);
-
- sg_init_one(&sg, bp, sizeof(*bp));
- ahash_request_set_crypt(hp->req, &sg, NULL, sizeof(*bp));
- return crypto_ahash_update(hp->req);
+ struct tcp6_pseudohdr phdr = {
+ .saddr = *saddr,
+ .daddr = *daddr,
+ .len = cpu_to_be32(nbytes),
+ .protocol = cpu_to_be32(IPPROTO_TCP),
+ };
+
+ tcp_ao_mac_update(mac_ctx, &phdr, sizeof(phdr));
+ return 0;
}
int tcp_v6_ao_hash_skb(char *ao_hash, struct tcp_ao_key *key,
const struct sock *sk, const struct sk_buff *skb,
const u8 *tkey, int hash_offset, u32 sne)
diff --git a/tools/testing/selftests/net/tcp_ao/config b/tools/testing/selftests/net/tcp_ao/config
index 47228a7d0b90..1b120bfd89c4 100644
--- a/tools/testing/selftests/net/tcp_ao/config
+++ b/tools/testing/selftests/net/tcp_ao/config
@@ -1,8 +1,5 @@
-CONFIG_CRYPTO_CMAC=y
-CONFIG_CRYPTO_HMAC=y
-CONFIG_CRYPTO_SHA1=y
CONFIG_IPV6=y
CONFIG_IPV6_MULTIPLE_TABLES=y
CONFIG_NET_L3_MASTER_DEV=y
CONFIG_NET_VRF=y
CONFIG_TCP_AO=y
--
2.54.0
^ permalink raw reply related [flat|nested] 8+ messages in thread* [PATCH net-next v2 3/5] net/tcp-ao: Use stack-allocated MAC and traffic_key buffers
2026-04-27 17:27 [PATCH net-next v2 0/5] Reimplement TCP-AO using crypto library Eric Biggers
2026-04-27 17:27 ` [PATCH net-next v2 1/5] net/tcp-ao: Drop support for most non-RFC-specified algorithms Eric Biggers
2026-04-27 17:27 ` [PATCH net-next v2 2/5] net/tcp-ao: Use crypto library API instead of crypto_ahash Eric Biggers
@ 2026-04-27 17:27 ` Eric Biggers
2026-04-27 17:27 ` [PATCH net-next v2 4/5] net/tcp-ao: Return void from functions that can no longer fail Eric Biggers
` (2 subsequent siblings)
5 siblings, 0 replies; 8+ messages in thread
From: Eric Biggers @ 2026-04-27 17:27 UTC (permalink / raw)
To: netdev
Cc: linux-crypto, linux-kernel, Eric Dumazet, Neal Cardwell,
Kuniyuki Iwashima, David S . Miller, David Ahern, Jakub Kicinski,
Paolo Abeni, Simon Horman, Ard Biesheuvel, Jason A . Donenfeld,
Herbert Xu, Dmitry Safonov, Eric Biggers
Now that the maximum MAC and traffic key lengths are statically-known
small values, allocate MACs and traffic keys on the stack instead of
with kmalloc. This eliminates multiple failure-prone GFP_ATOMIC
allocations.
Note that some cases such as tcp_ao_prepare_reset() are left unchanged
for now since they would require slightly wider changes.
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Eric Biggers <ebiggers@kernel.org>
---
net/ipv4/tcp_ao.c | 44 +++++++++++---------------------------------
net/ipv6/tcp_ao.c | 17 +++++------------
2 files changed, 16 insertions(+), 45 deletions(-)
diff --git a/net/ipv4/tcp_ao.c b/net/ipv4/tcp_ao.c
index 0d24cbd66c9a..69f1d6d26562 100644
--- a/net/ipv4/tcp_ao.c
+++ b/net/ipv4/tcp_ao.c
@@ -737,26 +737,19 @@ int tcp_v4_ao_hash_skb(char *ao_hash, struct tcp_ao_key *key,
int tcp_v4_ao_synack_hash(char *ao_hash, struct tcp_ao_key *ao_key,
struct request_sock *req, const struct sk_buff *skb,
int hash_offset, u32 sne)
{
- void *hash_buf = NULL;
+ u8 tkey_buf[TCP_AO_MAX_TRAFFIC_KEY_LEN];
int err;
- hash_buf = kmalloc(tcp_ao_digest_size(ao_key), GFP_ATOMIC);
- if (!hash_buf)
- return -ENOMEM;
-
- err = tcp_v4_ao_calc_key_rsk(ao_key, hash_buf, req);
+ err = tcp_v4_ao_calc_key_rsk(ao_key, tkey_buf, req);
if (err)
- goto out;
+ return err;
- err = tcp_ao_hash_skb(AF_INET, ao_hash, ao_key, req_to_sk(req), skb,
- hash_buf, hash_offset, sne);
-out:
- kfree(hash_buf);
- return err;
+ return tcp_ao_hash_skb(AF_INET, ao_hash, ao_key, req_to_sk(req), skb,
+ tkey_buf, hash_offset, sne);
}
struct tcp_ao_key *tcp_v4_ao_lookup_rsk(const struct sock *sk,
struct request_sock *req,
int sndid, int rcvid)
@@ -867,13 +860,13 @@ int tcp_ao_prepare_reset(const struct sock *sk, struct sk_buff *skb,
int tcp_ao_transmit_skb(struct sock *sk, struct sk_buff *skb,
struct tcp_ao_key *key, struct tcphdr *th,
__u8 *hash_location)
{
struct tcp_skb_cb *tcb = TCP_SKB_CB(skb);
+ u8 tkey_buf[TCP_AO_MAX_TRAFFIC_KEY_LEN];
struct tcp_sock *tp = tcp_sk(sk);
struct tcp_ao_info *ao;
- void *tkey_buf = NULL;
u8 *traffic_key;
u32 sne;
ao = rcu_dereference_protected(tcp_sk(sk)->ao_info,
lockdep_sock_is_held(sk));
@@ -881,13 +874,10 @@ int tcp_ao_transmit_skb(struct sock *sk, struct sk_buff *skb,
if (unlikely(tcb->tcp_flags & TCPHDR_SYN)) {
__be32 disn;
if (!(tcb->tcp_flags & TCPHDR_ACK)) {
disn = 0;
- tkey_buf = kmalloc(tcp_ao_digest_size(key), GFP_ATOMIC);
- if (!tkey_buf)
- return -ENOMEM;
traffic_key = tkey_buf;
} else {
disn = ao->risn;
}
tp->af_specific->ao_calc_key_sk(key, traffic_key,
@@ -895,11 +885,10 @@ int tcp_ao_transmit_skb(struct sock *sk, struct sk_buff *skb,
}
sne = tcp_ao_compute_sne(READ_ONCE(ao->snd_sne), READ_ONCE(tp->snd_una),
ntohl(th->seq));
tp->af_specific->calc_ao_hash(hash_location, key, sk, skb, traffic_key,
hash_location - (u8 *)th, sne);
- kfree(tkey_buf);
return 0;
}
static struct tcp_ao_key *tcp_ao_inbound_lookup(unsigned short int family,
const struct sock *sk, const struct sk_buff *skb,
@@ -961,54 +950,48 @@ tcp_ao_verify_hash(const struct sock *sk, const struct sk_buff *skb,
const struct tcp_ao_hdr *aoh, struct tcp_ao_key *key,
u8 *traffic_key, u8 *phash, u32 sne, int l3index)
{
const struct tcphdr *th = tcp_hdr(skb);
u8 maclen = tcp_ao_hdr_maclen(aoh);
- void *hash_buf = NULL;
+ u8 hash_buf[TCP_AO_MAX_MAC_LEN];
if (maclen != tcp_ao_maclen(key)) {
NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPAOBAD);
atomic64_inc(&info->counters.pkt_bad);
atomic64_inc(&key->pkt_bad);
trace_tcp_ao_wrong_maclen(sk, skb, aoh->keyid,
aoh->rnext_keyid, maclen);
return SKB_DROP_REASON_TCP_AOFAILURE;
}
- hash_buf = kmalloc(tcp_ao_digest_size(key), GFP_ATOMIC);
- if (!hash_buf)
- return SKB_DROP_REASON_NOT_SPECIFIED;
-
/* XXX: make it per-AF callback? */
tcp_ao_hash_skb(family, hash_buf, key, sk, skb, traffic_key,
(phash - (u8 *)th), sne);
if (crypto_memneq(phash, hash_buf, maclen)) {
NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPAOBAD);
atomic64_inc(&info->counters.pkt_bad);
atomic64_inc(&key->pkt_bad);
trace_tcp_ao_mismatch(sk, skb, aoh->keyid,
aoh->rnext_keyid, maclen);
- kfree(hash_buf);
return SKB_DROP_REASON_TCP_AOFAILURE;
}
NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPAOGOOD);
atomic64_inc(&info->counters.pkt_good);
atomic64_inc(&key->pkt_good);
- kfree(hash_buf);
return SKB_NOT_DROPPED_YET;
}
enum skb_drop_reason
tcp_inbound_ao_hash(struct sock *sk, const struct sk_buff *skb,
unsigned short int family, const struct request_sock *req,
int l3index, const struct tcp_ao_hdr *aoh)
{
+ u8 tkey_buf[TCP_AO_MAX_TRAFFIC_KEY_LEN];
const struct tcphdr *th = tcp_hdr(skb);
u8 maclen = tcp_ao_hdr_maclen(aoh);
u8 *phash = (u8 *)(aoh + 1); /* hash goes just after the header */
struct tcp_ao_info *info;
- enum skb_drop_reason ret;
struct tcp_ao_key *key;
__be32 sisn, disn;
u8 *traffic_key;
int state;
u32 sne = 0;
@@ -1112,18 +1095,13 @@ tcp_inbound_ao_hash(struct sock *sk, const struct sk_buff *skb,
} else {
WARN_ONCE(1, "TCP-AO: Unexpected sk_state %d", state);
return SKB_DROP_REASON_TCP_AOFAILURE;
}
verify_hash:
- traffic_key = kmalloc(tcp_ao_digest_size(key), GFP_ATOMIC);
- if (!traffic_key)
- return SKB_DROP_REASON_NOT_SPECIFIED;
- tcp_ao_calc_key_skb(key, traffic_key, skb, sisn, disn, family);
- ret = tcp_ao_verify_hash(sk, skb, family, info, aoh, key,
- traffic_key, phash, sne, l3index);
- kfree(traffic_key);
- return ret;
+ tcp_ao_calc_key_skb(key, tkey_buf, skb, sisn, disn, family);
+ return tcp_ao_verify_hash(sk, skb, family, info, aoh, key,
+ tkey_buf, phash, sne, l3index);
key_not_found:
NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPAOKEYNOTFOUND);
atomic64_inc(&info->counters.key_not_found);
trace_tcp_ao_key_not_found(sk, skb, aoh->keyid,
diff --git a/net/ipv6/tcp_ao.c b/net/ipv6/tcp_ao.c
index 2dcfe9dda7f4..bf30b970181d 100644
--- a/net/ipv6/tcp_ao.c
+++ b/net/ipv6/tcp_ao.c
@@ -136,22 +136,15 @@ int tcp_v6_parse_ao(struct sock *sk, int cmd,
int tcp_v6_ao_synack_hash(char *ao_hash, struct tcp_ao_key *ao_key,
struct request_sock *req, const struct sk_buff *skb,
int hash_offset, u32 sne)
{
- void *hash_buf = NULL;
+ u8 tkey_buf[TCP_AO_MAX_TRAFFIC_KEY_LEN];
int err;
- hash_buf = kmalloc(tcp_ao_digest_size(ao_key), GFP_ATOMIC);
- if (!hash_buf)
- return -ENOMEM;
-
- err = tcp_v6_ao_calc_key_rsk(ao_key, hash_buf, req);
+ err = tcp_v6_ao_calc_key_rsk(ao_key, tkey_buf, req);
if (err)
- goto out;
+ return err;
- err = tcp_ao_hash_skb(AF_INET6, ao_hash, ao_key, req_to_sk(req), skb,
- hash_buf, hash_offset, sne);
-out:
- kfree(hash_buf);
- return err;
+ return tcp_ao_hash_skb(AF_INET6, ao_hash, ao_key, req_to_sk(req), skb,
+ tkey_buf, hash_offset, sne);
}
--
2.54.0
^ permalink raw reply related [flat|nested] 8+ messages in thread* [PATCH net-next v2 4/5] net/tcp-ao: Return void from functions that can no longer fail
2026-04-27 17:27 [PATCH net-next v2 0/5] Reimplement TCP-AO using crypto library Eric Biggers
` (2 preceding siblings ...)
2026-04-27 17:27 ` [PATCH net-next v2 3/5] net/tcp-ao: Use stack-allocated MAC and traffic_key buffers Eric Biggers
@ 2026-04-27 17:27 ` Eric Biggers
2026-04-27 17:27 ` [PATCH net-next v2 5/5] net/tcp: Remove tcp_sigpool Eric Biggers
2026-04-27 19:09 ` [PATCH net-next v2 0/5] Reimplement TCP-AO using crypto library Dmitry Safonov
5 siblings, 0 replies; 8+ messages in thread
From: Eric Biggers @ 2026-04-27 17:27 UTC (permalink / raw)
To: netdev
Cc: linux-crypto, linux-kernel, Eric Dumazet, Neal Cardwell,
Kuniyuki Iwashima, David S . Miller, David Ahern, Jakub Kicinski,
Paolo Abeni, Simon Horman, Ard Biesheuvel, Jason A . Donenfeld,
Herbert Xu, Dmitry Safonov, Eric Biggers
Since tcp-ao now uses the crypto library API instead of crypto_ahash,
and MACs and keys now have a statically-known maximum size, many tcp-ao
functions can no longer fail. Propagate this change up into the return
types of various functions.
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Eric Biggers <ebiggers@kernel.org>
---
include/net/tcp.h | 8 +--
include/net/tcp_ao.h | 44 +++++++--------
net/ipv4/tcp_ao.c | 128 ++++++++++++++++++++++--------------------
net/ipv4/tcp_output.c | 10 +---
net/ipv6/tcp_ao.c | 65 ++++++++++-----------
5 files changed, 123 insertions(+), 132 deletions(-)
diff --git a/include/net/tcp.h b/include/net/tcp.h
index ecbadcb3a744..8817e7b8cc67 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -2508,13 +2508,13 @@ struct tcp_sock_af_ops {
#ifdef CONFIG_TCP_AO
int (*ao_parse)(struct sock *sk, int optname, sockptr_t optval, int optlen);
struct tcp_ao_key *(*ao_lookup)(const struct sock *sk,
struct sock *addr_sk,
int sndid, int rcvid);
- int (*ao_calc_key_sk)(struct tcp_ao_key *mkt, u8 *key,
- const struct sock *sk,
- __be32 sisn, __be32 disn, bool send);
+ void (*ao_calc_key_sk)(struct tcp_ao_key *mkt, u8 *key,
+ const struct sock *sk,
+ __be32 sisn, __be32 disn, bool send);
int (*calc_ao_hash)(char *location, struct tcp_ao_key *ao,
const struct sock *sk, const struct sk_buff *skb,
const u8 *tkey, int hash_offset, u32 sne);
#endif
};
@@ -2531,11 +2531,11 @@ struct tcp_request_sock_ops {
#endif
#ifdef CONFIG_TCP_AO
struct tcp_ao_key *(*ao_lookup)(const struct sock *sk,
struct request_sock *req,
int sndid, int rcvid);
- int (*ao_calc_key)(struct tcp_ao_key *mkt, u8 *key, struct request_sock *sk);
+ void (*ao_calc_key)(struct tcp_ao_key *mkt, u8 *key, struct request_sock *sk);
int (*ao_synack_hash)(char *ao_hash, struct tcp_ao_key *mkt,
struct request_sock *req, const struct sk_buff *skb,
int hash_offset, u32 sne);
#endif
#ifdef CONFIG_SYN_COOKIES
diff --git a/include/net/tcp_ao.h b/include/net/tcp_ao.h
index 20997aef3b0d..29fd7b735afa 100644
--- a/include/net/tcp_ao.h
+++ b/include/net/tcp_ao.h
@@ -185,13 +185,13 @@ struct tcp6_ao_context {
/* Established states are fast-path and there always is current_key/rnext_key */
#define TCP_AO_ESTABLISHED (TCPF_ESTABLISHED | TCPF_FIN_WAIT1 | TCPF_FIN_WAIT2 | \
TCPF_CLOSE_WAIT | TCPF_LAST_ACK | TCPF_CLOSING)
-int tcp_ao_transmit_skb(struct sock *sk, struct sk_buff *skb,
- struct tcp_ao_key *key, struct tcphdr *th,
- __u8 *hash_location);
+void tcp_ao_transmit_skb(struct sock *sk, struct sk_buff *skb,
+ struct tcp_ao_key *key, struct tcphdr *th,
+ __u8 *hash_location);
void tcp_ao_mac_update(struct tcp_ao_mac_ctx *mac_ctx, const void *data,
size_t data_len);
int tcp_ao_hash_skb(unsigned short int family,
char *ao_hash, struct tcp_ao_key *key,
const struct sock *sk, const struct sk_buff *skb,
@@ -236,32 +236,33 @@ int tcp_v4_parse_ao(struct sock *sk, int cmd, sockptr_t optval, int optlen);
struct tcp_ao_key *tcp_v4_ao_lookup(const struct sock *sk, struct sock *addr_sk,
int sndid, int rcvid);
int tcp_v4_ao_synack_hash(char *ao_hash, struct tcp_ao_key *mkt,
struct request_sock *req, const struct sk_buff *skb,
int hash_offset, u32 sne);
-int tcp_v4_ao_calc_key_sk(struct tcp_ao_key *mkt, u8 *key,
- const struct sock *sk,
- __be32 sisn, __be32 disn, bool send);
-int tcp_v4_ao_calc_key_rsk(struct tcp_ao_key *mkt, u8 *key,
- struct request_sock *req);
+void tcp_v4_ao_calc_key_sk(struct tcp_ao_key *mkt, u8 *key,
+ const struct sock *sk,
+ __be32 sisn, __be32 disn, bool send);
+void tcp_v4_ao_calc_key_rsk(struct tcp_ao_key *mkt, u8 *key,
+ struct request_sock *req);
struct tcp_ao_key *tcp_v4_ao_lookup_rsk(const struct sock *sk,
struct request_sock *req,
int sndid, int rcvid);
int tcp_v4_ao_hash_skb(char *ao_hash, struct tcp_ao_key *key,
const struct sock *sk, const struct sk_buff *skb,
const u8 *tkey, int hash_offset, u32 sne);
/* ipv6 specific functions */
-int tcp_v6_ao_hash_pseudoheader(struct tcp_ao_mac_ctx *mac_ctx,
- const struct in6_addr *daddr,
- const struct in6_addr *saddr, int nbytes);
-int tcp_v6_ao_calc_key_skb(struct tcp_ao_key *mkt, u8 *key,
- const struct sk_buff *skb, __be32 sisn, __be32 disn);
-int tcp_v6_ao_calc_key_sk(struct tcp_ao_key *mkt, u8 *key,
- const struct sock *sk, __be32 sisn,
- __be32 disn, bool send);
-int tcp_v6_ao_calc_key_rsk(struct tcp_ao_key *mkt, u8 *key,
- struct request_sock *req);
+void tcp_v6_ao_hash_pseudoheader(struct tcp_ao_mac_ctx *mac_ctx,
+ const struct in6_addr *daddr,
+ const struct in6_addr *saddr, int nbytes);
+void tcp_v6_ao_calc_key_skb(struct tcp_ao_key *mkt, u8 *key,
+ const struct sk_buff *skb, __be32 sisn,
+ __be32 disn);
+void tcp_v6_ao_calc_key_sk(struct tcp_ao_key *mkt, u8 *key,
+ const struct sock *sk, __be32 sisn,
+ __be32 disn, bool send);
+void tcp_v6_ao_calc_key_rsk(struct tcp_ao_key *mkt, u8 *key,
+ struct request_sock *req);
struct tcp_ao_key *tcp_v6_ao_lookup(const struct sock *sk,
struct sock *addr_sk, int sndid, int rcvid);
struct tcp_ao_key *tcp_v6_ao_lookup_rsk(const struct sock *sk,
struct request_sock *req,
int sndid, int rcvid);
@@ -277,15 +278,14 @@ void tcp_ao_finish_connect(struct sock *sk, struct sk_buff *skb);
void tcp_ao_connect_init(struct sock *sk);
void tcp_ao_syncookie(struct sock *sk, const struct sk_buff *skb,
struct request_sock *req, unsigned short int family);
#else /* CONFIG_TCP_AO */
-static inline int tcp_ao_transmit_skb(struct sock *sk, struct sk_buff *skb,
- struct tcp_ao_key *key, struct tcphdr *th,
- __u8 *hash_location)
+static inline void tcp_ao_transmit_skb(struct sock *sk, struct sk_buff *skb,
+ struct tcp_ao_key *key,
+ struct tcphdr *th, __u8 *hash_location)
{
- return 0;
}
static inline void tcp_ao_syncookie(struct sock *sk, const struct sk_buff *skb,
struct request_sock *req, unsigned short int family)
{
diff --git a/net/ipv4/tcp_ao.c b/net/ipv4/tcp_ao.c
index 69f1d6d26562..36a64c1cd8c9 100644
--- a/net/ipv4/tcp_ao.c
+++ b/net/ipv4/tcp_ao.c
@@ -433,14 +433,14 @@ void tcp_ao_time_wait(struct tcp_timewait_sock *tcptw, struct tcp_sock *tp)
tcptw->ao_info = NULL;
}
}
/* 4 tuple and ISNs are expected in NBO */
-static int tcp_v4_ao_calc_key(struct tcp_ao_key *mkt, u8 *key,
- __be32 saddr, __be32 daddr,
- __be16 sport, __be16 dport,
- __be32 sisn, __be32 disn)
+static void tcp_v4_ao_calc_key(struct tcp_ao_key *mkt, u8 *key,
+ __be32 saddr, __be32 daddr,
+ __be16 sport, __be16 dport,
+ __be32 sisn, __be32 disn)
{
/* See RFC5926 3.1.1 */
struct kdf_input_block {
u8 counter;
u8 label[6];
@@ -459,91 +459,92 @@ static int tcp_v4_ao_calc_key(struct tcp_ao_key *mkt, u8 *key,
},
.outlen = htons(tcp_ao_digest_size(mkt) * 8), /* in bits */
};
tcp_ao_calc_traffic_key(mkt, key, &input, sizeof(input));
- return 0;
}
-int tcp_v4_ao_calc_key_sk(struct tcp_ao_key *mkt, u8 *key,
- const struct sock *sk,
- __be32 sisn, __be32 disn, bool send)
+void tcp_v4_ao_calc_key_sk(struct tcp_ao_key *mkt, u8 *key,
+ const struct sock *sk,
+ __be32 sisn, __be32 disn, bool send)
{
if (send)
- return tcp_v4_ao_calc_key(mkt, key, sk->sk_rcv_saddr,
- sk->sk_daddr, htons(sk->sk_num),
- sk->sk_dport, sisn, disn);
+ tcp_v4_ao_calc_key(mkt, key, sk->sk_rcv_saddr, sk->sk_daddr,
+ htons(sk->sk_num), sk->sk_dport, sisn, disn);
else
- return tcp_v4_ao_calc_key(mkt, key, sk->sk_daddr,
- sk->sk_rcv_saddr, sk->sk_dport,
- htons(sk->sk_num), disn, sisn);
+ tcp_v4_ao_calc_key(mkt, key, sk->sk_daddr, sk->sk_rcv_saddr,
+ sk->sk_dport, htons(sk->sk_num), disn, sisn);
}
static int tcp_ao_calc_key_sk(struct tcp_ao_key *mkt, u8 *key,
const struct sock *sk,
__be32 sisn, __be32 disn, bool send)
{
- if (mkt->family == AF_INET)
- return tcp_v4_ao_calc_key_sk(mkt, key, sk, sisn, disn, send);
+ if (mkt->family == AF_INET) {
+ tcp_v4_ao_calc_key_sk(mkt, key, sk, sisn, disn, send);
+ return 0;
+ }
#if IS_ENABLED(CONFIG_IPV6)
- else if (mkt->family == AF_INET6)
- return tcp_v6_ao_calc_key_sk(mkt, key, sk, sisn, disn, send);
+ if (mkt->family == AF_INET6) {
+ tcp_v6_ao_calc_key_sk(mkt, key, sk, sisn, disn, send);
+ return 0;
+ }
#endif
- else
- return -EOPNOTSUPP;
+ return -EOPNOTSUPP;
}
-int tcp_v4_ao_calc_key_rsk(struct tcp_ao_key *mkt, u8 *key,
- struct request_sock *req)
+void tcp_v4_ao_calc_key_rsk(struct tcp_ao_key *mkt, u8 *key,
+ struct request_sock *req)
{
struct inet_request_sock *ireq = inet_rsk(req);
- return tcp_v4_ao_calc_key(mkt, key,
- ireq->ir_loc_addr, ireq->ir_rmt_addr,
- htons(ireq->ir_num), ireq->ir_rmt_port,
- htonl(tcp_rsk(req)->snt_isn),
- htonl(tcp_rsk(req)->rcv_isn));
+ tcp_v4_ao_calc_key(mkt, key, ireq->ir_loc_addr, ireq->ir_rmt_addr,
+ htons(ireq->ir_num), ireq->ir_rmt_port,
+ htonl(tcp_rsk(req)->snt_isn),
+ htonl(tcp_rsk(req)->rcv_isn));
}
-static int tcp_v4_ao_calc_key_skb(struct tcp_ao_key *mkt, u8 *key,
- const struct sk_buff *skb,
- __be32 sisn, __be32 disn)
+static void tcp_v4_ao_calc_key_skb(struct tcp_ao_key *mkt, u8 *key,
+ const struct sk_buff *skb,
+ __be32 sisn, __be32 disn)
{
const struct iphdr *iph = ip_hdr(skb);
const struct tcphdr *th = tcp_hdr(skb);
- return tcp_v4_ao_calc_key(mkt, key, iph->saddr, iph->daddr,
- th->source, th->dest, sisn, disn);
+ tcp_v4_ao_calc_key(mkt, key, iph->saddr, iph->daddr, th->source,
+ th->dest, sisn, disn);
}
static int tcp_ao_calc_key_skb(struct tcp_ao_key *mkt, u8 *key,
const struct sk_buff *skb,
__be32 sisn, __be32 disn, int family)
{
- if (family == AF_INET)
- return tcp_v4_ao_calc_key_skb(mkt, key, skb, sisn, disn);
+ if (family == AF_INET) {
+ tcp_v4_ao_calc_key_skb(mkt, key, skb, sisn, disn);
+ return 0;
+ }
#if IS_ENABLED(CONFIG_IPV6)
- else if (family == AF_INET6)
- return tcp_v6_ao_calc_key_skb(mkt, key, skb, sisn, disn);
+ if (family == AF_INET6) {
+ tcp_v6_ao_calc_key_skb(mkt, key, skb, sisn, disn);
+ return 0;
+ }
#endif
return -EAFNOSUPPORT;
}
-static int tcp_v4_ao_hash_pseudoheader(struct tcp_ao_mac_ctx *mac_ctx,
- __be32 daddr, __be32 saddr,
- int nbytes)
+static void tcp_v4_ao_hash_pseudoheader(struct tcp_ao_mac_ctx *mac_ctx,
+ __be32 daddr, __be32 saddr, int nbytes)
{
struct tcp4_pseudohdr phdr = {
.saddr = saddr,
.daddr = daddr,
.pad = 0,
.protocol = IPPROTO_TCP,
.len = cpu_to_be16(nbytes),
};
tcp_ao_mac_update(mac_ctx, &phdr, sizeof(phdr));
- return 0;
}
static int tcp_ao_hash_pseudoheader(unsigned short int family,
const struct sock *sk,
const struct sk_buff *skb,
@@ -551,35 +552,42 @@ static int tcp_ao_hash_pseudoheader(unsigned short int family,
{
const struct tcphdr *th = tcp_hdr(skb);
/* TODO: Can we rely on checksum being zero to mean outbound pkt? */
if (!th->check) {
- if (family == AF_INET)
- return tcp_v4_ao_hash_pseudoheader(mac_ctx, sk->sk_daddr,
- sk->sk_rcv_saddr, skb->len);
+ if (family == AF_INET) {
+ tcp_v4_ao_hash_pseudoheader(mac_ctx, sk->sk_daddr,
+ sk->sk_rcv_saddr, skb->len);
+ return 0;
+ }
#if IS_ENABLED(CONFIG_IPV6)
- else if (family == AF_INET6)
- return tcp_v6_ao_hash_pseudoheader(mac_ctx, &sk->sk_v6_daddr,
- &sk->sk_v6_rcv_saddr, skb->len);
+ if (family == AF_INET6) {
+ tcp_v6_ao_hash_pseudoheader(mac_ctx, &sk->sk_v6_daddr,
+ &sk->sk_v6_rcv_saddr,
+ skb->len);
+ return 0;
+ }
#endif
- else
- return -EAFNOSUPPORT;
+ return -EAFNOSUPPORT;
}
if (family == AF_INET) {
const struct iphdr *iph = ip_hdr(skb);
- return tcp_v4_ao_hash_pseudoheader(mac_ctx, iph->daddr,
- iph->saddr, skb->len);
+ tcp_v4_ao_hash_pseudoheader(mac_ctx, iph->daddr, iph->saddr,
+ skb->len);
+ return 0;
+ }
#if IS_ENABLED(CONFIG_IPV6)
- } else if (family == AF_INET6) {
+ if (family == AF_INET6) {
const struct ipv6hdr *iph = ipv6_hdr(skb);
- return tcp_v6_ao_hash_pseudoheader(mac_ctx, &iph->daddr,
- &iph->saddr, skb->len);
-#endif
+ tcp_v6_ao_hash_pseudoheader(mac_ctx, &iph->daddr, &iph->saddr,
+ skb->len);
+ return 0;
}
+#endif
return -EAFNOSUPPORT;
}
u32 tcp_ao_compute_sne(u32 next_sne, u32 next_seq, u32 seq)
{
@@ -738,15 +746,12 @@ int tcp_v4_ao_hash_skb(char *ao_hash, struct tcp_ao_key *key,
int tcp_v4_ao_synack_hash(char *ao_hash, struct tcp_ao_key *ao_key,
struct request_sock *req, const struct sk_buff *skb,
int hash_offset, u32 sne)
{
u8 tkey_buf[TCP_AO_MAX_TRAFFIC_KEY_LEN];
- int err;
- err = tcp_v4_ao_calc_key_rsk(ao_key, tkey_buf, req);
- if (err)
- return err;
+ tcp_v4_ao_calc_key_rsk(ao_key, tkey_buf, req);
return tcp_ao_hash_skb(AF_INET, ao_hash, ao_key, req_to_sk(req), skb,
tkey_buf, hash_offset, sne);
}
@@ -855,13 +860,13 @@ int tcp_ao_prepare_reset(const struct sock *sk, struct sk_buff *skb,
snd_basis, seq);
}
return 0;
}
-int tcp_ao_transmit_skb(struct sock *sk, struct sk_buff *skb,
- struct tcp_ao_key *key, struct tcphdr *th,
- __u8 *hash_location)
+void tcp_ao_transmit_skb(struct sock *sk, struct sk_buff *skb,
+ struct tcp_ao_key *key, struct tcphdr *th,
+ __u8 *hash_location)
{
struct tcp_skb_cb *tcb = TCP_SKB_CB(skb);
u8 tkey_buf[TCP_AO_MAX_TRAFFIC_KEY_LEN];
struct tcp_sock *tp = tcp_sk(sk);
struct tcp_ao_info *ao;
@@ -885,11 +890,10 @@ int tcp_ao_transmit_skb(struct sock *sk, struct sk_buff *skb,
}
sne = tcp_ao_compute_sne(READ_ONCE(ao->snd_sne), READ_ONCE(tp->snd_una),
ntohl(th->seq));
tp->af_specific->calc_ao_hash(hash_location, key, sk, skb, traffic_key,
hash_location - (u8 *)th, sne);
- return 0;
}
static struct tcp_ao_key *tcp_ao_inbound_lookup(unsigned short int family,
const struct sock *sk, const struct sk_buff *skb,
int sndid, int rcvid, int l3index)
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index f9d8755705f7..2671355661e4 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -1661,18 +1661,12 @@ static int __tcp_transmit_skb(struct sock *sk, struct sk_buff *skb,
sk_gso_disable(sk);
tp->af_specific->calc_md5_hash(opts.hash_location,
key.md5_key, sk, skb);
#endif
} else if (tcp_key_is_ao(&key)) {
- int err;
-
- err = tcp_ao_transmit_skb(sk, skb, key.ao_key, th,
- opts.hash_location);
- if (err) {
- sk_skb_reason_drop(sk, skb, SKB_DROP_REASON_NOT_SPECIFIED);
- return -ENOMEM;
- }
+ tcp_ao_transmit_skb(sk, skb, key.ao_key, th,
+ opts.hash_location);
}
/* BPF prog is the last one writing header option */
bpf_skops_write_hdr_opt(sk, skb, NULL, NULL, 0, &opts);
diff --git a/net/ipv6/tcp_ao.c b/net/ipv6/tcp_ao.c
index bf30b970181d..01a8472805d1 100644
--- a/net/ipv6/tcp_ao.c
+++ b/net/ipv6/tcp_ao.c
@@ -10,15 +10,15 @@
#include <linux/tcp.h>
#include <net/tcp.h>
#include <net/ipv6.h>
-static int tcp_v6_ao_calc_key(struct tcp_ao_key *mkt, u8 *key,
- const struct in6_addr *saddr,
- const struct in6_addr *daddr,
- __be16 sport, __be16 dport,
- __be32 sisn, __be32 disn)
+static void tcp_v6_ao_calc_key(struct tcp_ao_key *mkt, u8 *key,
+ const struct in6_addr *saddr,
+ const struct in6_addr *daddr,
+ __be16 sport, __be16 dport,
+ __be32 sisn, __be32 disn)
{
struct kdf_input_block {
u8 counter;
u8 label[6];
struct tcp6_ao_context ctx;
@@ -36,49 +36,46 @@ static int tcp_v6_ao_calc_key(struct tcp_ao_key *mkt, u8 *key,
},
.outlen = htons(tcp_ao_digest_size(mkt) * 8), /* in bits */
};
tcp_ao_calc_traffic_key(mkt, key, &input, sizeof(input));
- return 0;
}
-int tcp_v6_ao_calc_key_skb(struct tcp_ao_key *mkt, u8 *key,
- const struct sk_buff *skb,
- __be32 sisn, __be32 disn)
+void tcp_v6_ao_calc_key_skb(struct tcp_ao_key *mkt, u8 *key,
+ const struct sk_buff *skb, __be32 sisn, __be32 disn)
{
const struct ipv6hdr *iph = ipv6_hdr(skb);
const struct tcphdr *th = tcp_hdr(skb);
- return tcp_v6_ao_calc_key(mkt, key, &iph->saddr,
- &iph->daddr, th->source,
- th->dest, sisn, disn);
+ tcp_v6_ao_calc_key(mkt, key, &iph->saddr, &iph->daddr, th->source,
+ th->dest, sisn, disn);
}
-int tcp_v6_ao_calc_key_sk(struct tcp_ao_key *mkt, u8 *key,
- const struct sock *sk, __be32 sisn,
- __be32 disn, bool send)
+void tcp_v6_ao_calc_key_sk(struct tcp_ao_key *mkt, u8 *key,
+ const struct sock *sk, __be32 sisn,
+ __be32 disn, bool send)
{
if (send)
- return tcp_v6_ao_calc_key(mkt, key, &sk->sk_v6_rcv_saddr,
- &sk->sk_v6_daddr, htons(sk->sk_num),
- sk->sk_dport, sisn, disn);
+ tcp_v6_ao_calc_key(mkt, key, &sk->sk_v6_rcv_saddr,
+ &sk->sk_v6_daddr, htons(sk->sk_num),
+ sk->sk_dport, sisn, disn);
else
- return tcp_v6_ao_calc_key(mkt, key, &sk->sk_v6_daddr,
- &sk->sk_v6_rcv_saddr, sk->sk_dport,
- htons(sk->sk_num), disn, sisn);
+ tcp_v6_ao_calc_key(mkt, key, &sk->sk_v6_daddr,
+ &sk->sk_v6_rcv_saddr, sk->sk_dport,
+ htons(sk->sk_num), disn, sisn);
}
-int tcp_v6_ao_calc_key_rsk(struct tcp_ao_key *mkt, u8 *key,
- struct request_sock *req)
+void tcp_v6_ao_calc_key_rsk(struct tcp_ao_key *mkt, u8 *key,
+ struct request_sock *req)
{
struct inet_request_sock *ireq = inet_rsk(req);
- return tcp_v6_ao_calc_key(mkt, key,
- &ireq->ir_v6_loc_addr, &ireq->ir_v6_rmt_addr,
- htons(ireq->ir_num), ireq->ir_rmt_port,
- htonl(tcp_rsk(req)->snt_isn),
- htonl(tcp_rsk(req)->rcv_isn));
+ tcp_v6_ao_calc_key(mkt, key,
+ &ireq->ir_v6_loc_addr, &ireq->ir_v6_rmt_addr,
+ htons(ireq->ir_num), ireq->ir_rmt_port,
+ htonl(tcp_rsk(req)->snt_isn),
+ htonl(tcp_rsk(req)->rcv_isn));
}
struct tcp_ao_key *tcp_v6_ao_lookup(const struct sock *sk,
struct sock *addr_sk,
int sndid, int rcvid)
@@ -102,24 +99,23 @@ struct tcp_ao_key *tcp_v6_ao_lookup_rsk(const struct sock *sk,
l3index = l3mdev_master_ifindex_by_index(sock_net(sk), ireq->ir_iif);
return tcp_ao_do_lookup(sk, l3index, (union tcp_ao_addr *)addr,
AF_INET6, sndid, rcvid);
}
-int tcp_v6_ao_hash_pseudoheader(struct tcp_ao_mac_ctx *mac_ctx,
- const struct in6_addr *daddr,
- const struct in6_addr *saddr, int nbytes)
+void tcp_v6_ao_hash_pseudoheader(struct tcp_ao_mac_ctx *mac_ctx,
+ const struct in6_addr *daddr,
+ const struct in6_addr *saddr, int nbytes)
{
/* 1. TCP pseudo-header (RFC2460) */
struct tcp6_pseudohdr phdr = {
.saddr = *saddr,
.daddr = *daddr,
.len = cpu_to_be32(nbytes),
.protocol = cpu_to_be32(IPPROTO_TCP),
};
tcp_ao_mac_update(mac_ctx, &phdr, sizeof(phdr));
- return 0;
}
int tcp_v6_ao_hash_skb(char *ao_hash, struct tcp_ao_key *key,
const struct sock *sk, const struct sk_buff *skb,
const u8 *tkey, int hash_offset, u32 sne)
@@ -137,14 +133,11 @@ int tcp_v6_parse_ao(struct sock *sk, int cmd,
int tcp_v6_ao_synack_hash(char *ao_hash, struct tcp_ao_key *ao_key,
struct request_sock *req, const struct sk_buff *skb,
int hash_offset, u32 sne)
{
u8 tkey_buf[TCP_AO_MAX_TRAFFIC_KEY_LEN];
- int err;
- err = tcp_v6_ao_calc_key_rsk(ao_key, tkey_buf, req);
- if (err)
- return err;
+ tcp_v6_ao_calc_key_rsk(ao_key, tkey_buf, req);
return tcp_ao_hash_skb(AF_INET6, ao_hash, ao_key, req_to_sk(req), skb,
tkey_buf, hash_offset, sne);
}
--
2.54.0
^ permalink raw reply related [flat|nested] 8+ messages in thread* [PATCH net-next v2 5/5] net/tcp: Remove tcp_sigpool
2026-04-27 17:27 [PATCH net-next v2 0/5] Reimplement TCP-AO using crypto library Eric Biggers
` (3 preceding siblings ...)
2026-04-27 17:27 ` [PATCH net-next v2 4/5] net/tcp-ao: Return void from functions that can no longer fail Eric Biggers
@ 2026-04-27 17:27 ` Eric Biggers
2026-04-27 19:09 ` [PATCH net-next v2 0/5] Reimplement TCP-AO using crypto library Dmitry Safonov
5 siblings, 0 replies; 8+ messages in thread
From: Eric Biggers @ 2026-04-27 17:27 UTC (permalink / raw)
To: netdev
Cc: linux-crypto, linux-kernel, Eric Dumazet, Neal Cardwell,
Kuniyuki Iwashima, David S . Miller, David Ahern, Jakub Kicinski,
Paolo Abeni, Simon Horman, Ard Biesheuvel, Jason A . Donenfeld,
Herbert Xu, Dmitry Safonov, Eric Biggers
tcp_sigpool is no longer used. It existed only as a workaround for
issues in the design of the crypto_ahash API, which have been avoided by
switching to the much easier-to-use library APIs instead. Remove it.
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Eric Biggers <ebiggers@kernel.org>
---
include/net/tcp.h | 34 ----
net/ipv4/Kconfig | 3 -
net/ipv4/Makefile | 1 -
net/ipv4/tcp_sigpool.c | 366 -----------------------------------------
4 files changed, 404 deletions(-)
delete mode 100644 net/ipv4/tcp_sigpool.c
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 8817e7b8cc67..660ad8c63395 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -2015,44 +2015,10 @@ struct tcp6_pseudohdr {
struct in6_addr daddr;
__be32 len;
__be32 protocol; /* including padding */
};
-/*
- * struct tcp_sigpool - per-CPU pool of ahash_requests
- * @scratch: per-CPU temporary area, that can be used between
- * tcp_sigpool_start() and tcp_sigpool_end() to perform
- * crypto request
- * @req: pre-allocated ahash request
- */
-struct tcp_sigpool {
- void *scratch;
- struct ahash_request *req;
-};
-
-int tcp_sigpool_alloc_ahash(const char *alg, size_t scratch_size);
-void tcp_sigpool_get(unsigned int id);
-void tcp_sigpool_release(unsigned int id);
-int tcp_sigpool_hash_skb_data(struct tcp_sigpool *hp,
- const struct sk_buff *skb,
- unsigned int header_len);
-
-/**
- * tcp_sigpool_start - disable bh and start using tcp_sigpool_ahash
- * @id: tcp_sigpool that was previously allocated by tcp_sigpool_alloc_ahash()
- * @c: returned tcp_sigpool for usage (uninitialized on failure)
- *
- * Returns: 0 on success, error otherwise.
- */
-int tcp_sigpool_start(unsigned int id, struct tcp_sigpool *c);
-/**
- * tcp_sigpool_end - enable bh and stop using tcp_sigpool
- * @c: tcp_sigpool context that was returned by tcp_sigpool_start()
- */
-void tcp_sigpool_end(struct tcp_sigpool *c);
-size_t tcp_sigpool_algo(unsigned int id, char *buf, size_t buf_len);
-/* - functions */
void tcp_v4_md5_hash_skb(char *md5_hash, const struct tcp_md5sig_key *key,
const struct sock *sk, const struct sk_buff *skb);
int tcp_md5_do_add(struct sock *sk, const union tcp_md5_addr *addr,
int family, u8 prefixlen, int l3index, u8 flags,
const u8 *newkey, u8 newkeylen);
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index 77b053b445a0..301b47660305 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -739,13 +739,10 @@ config DEFAULT_TCP_CONG
default "dctcp" if DEFAULT_DCTCP
default "cdg" if DEFAULT_CDG
default "bbr" if DEFAULT_BBR
default "cubic"
-config TCP_SIGPOOL
- tristate
-
config TCP_AO
bool "TCP: Authentication Option (RFC5925)"
select CRYPTO_LIB_AES_CBC_MACS
select CRYPTO_LIB_SHA1
select CRYPTO_LIB_SHA256
diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile
index 7f9f98813986..7964234f0d08 100644
--- a/net/ipv4/Makefile
+++ b/net/ipv4/Makefile
@@ -58,11 +58,10 @@ obj-$(CONFIG_TCP_CONG_NV) += tcp_nv.o
obj-$(CONFIG_TCP_CONG_VENO) += tcp_veno.o
obj-$(CONFIG_TCP_CONG_SCALABLE) += tcp_scalable.o
obj-$(CONFIG_TCP_CONG_LP) += tcp_lp.o
obj-$(CONFIG_TCP_CONG_YEAH) += tcp_yeah.o
obj-$(CONFIG_TCP_CONG_ILLINOIS) += tcp_illinois.o
-obj-$(CONFIG_TCP_SIGPOOL) += tcp_sigpool.o
obj-$(CONFIG_NET_SOCK_MSG) += tcp_bpf.o
obj-$(CONFIG_BPF_SYSCALL) += udp_bpf.o
obj-$(CONFIG_NETLABEL) += cipso_ipv4.o
obj-$(CONFIG_XFRM) += xfrm4_policy.o xfrm4_state.o xfrm4_input.o \
diff --git a/net/ipv4/tcp_sigpool.c b/net/ipv4/tcp_sigpool.c
deleted file mode 100644
index 10b2e5970c40..000000000000
--- a/net/ipv4/tcp_sigpool.c
+++ /dev/null
@@ -1,366 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0-or-later
-
-#include <crypto/hash.h>
-#include <linux/cpu.h>
-#include <linux/kref.h>
-#include <linux/module.h>
-#include <linux/mutex.h>
-#include <linux/percpu.h>
-#include <linux/workqueue.h>
-#include <net/tcp.h>
-
-static size_t __scratch_size;
-struct sigpool_scratch {
- local_lock_t bh_lock;
- void __rcu *pad;
-};
-
-static DEFINE_PER_CPU(struct sigpool_scratch, sigpool_scratch) = {
- .bh_lock = INIT_LOCAL_LOCK(bh_lock),
-};
-
-struct sigpool_entry {
- struct crypto_ahash *hash;
- const char *alg;
- struct kref kref;
- uint16_t needs_key:1,
- reserved:15;
-};
-
-#define CPOOL_SIZE (PAGE_SIZE / sizeof(struct sigpool_entry))
-static struct sigpool_entry cpool[CPOOL_SIZE];
-static unsigned int cpool_populated;
-static DEFINE_MUTEX(cpool_mutex);
-
-/* Slow-path */
-struct scratches_to_free {
- struct rcu_head rcu;
- unsigned int cnt;
- void *scratches[];
-};
-
-static void free_old_scratches(struct rcu_head *head)
-{
- struct scratches_to_free *stf;
-
- stf = container_of(head, struct scratches_to_free, rcu);
- while (stf->cnt--)
- kfree(stf->scratches[stf->cnt]);
- kfree(stf);
-}
-
-/**
- * sigpool_reserve_scratch - re-allocates scratch buffer, slow-path
- * @size: request size for the scratch/temp buffer
- */
-static int sigpool_reserve_scratch(size_t size)
-{
- struct scratches_to_free *stf;
- size_t stf_sz = struct_size(stf, scratches, num_possible_cpus());
- int cpu, err = 0;
-
- lockdep_assert_held(&cpool_mutex);
- if (__scratch_size >= size)
- return 0;
-
- stf = kmalloc(stf_sz, GFP_KERNEL);
- if (!stf)
- return -ENOMEM;
- stf->cnt = 0;
-
- size = max(size, __scratch_size);
- cpus_read_lock();
- for_each_possible_cpu(cpu) {
- void *scratch, *old_scratch;
-
- scratch = kmalloc_node(size, GFP_KERNEL, cpu_to_node(cpu));
- if (!scratch) {
- err = -ENOMEM;
- break;
- }
-
- old_scratch = rcu_replace_pointer(per_cpu(sigpool_scratch.pad, cpu),
- scratch, lockdep_is_held(&cpool_mutex));
- if (!cpu_online(cpu) || !old_scratch) {
- kfree(old_scratch);
- continue;
- }
- stf->scratches[stf->cnt++] = old_scratch;
- }
- cpus_read_unlock();
- if (!err)
- __scratch_size = size;
-
- call_rcu(&stf->rcu, free_old_scratches);
- return err;
-}
-
-static void sigpool_scratch_free(void)
-{
- int cpu;
-
- for_each_possible_cpu(cpu)
- kfree(rcu_replace_pointer(per_cpu(sigpool_scratch.pad, cpu),
- NULL, lockdep_is_held(&cpool_mutex)));
- __scratch_size = 0;
-}
-
-static int __cpool_try_clone(struct crypto_ahash *hash)
-{
- struct crypto_ahash *tmp;
-
- tmp = crypto_clone_ahash(hash);
- if (IS_ERR(tmp))
- return PTR_ERR(tmp);
-
- crypto_free_ahash(tmp);
- return 0;
-}
-
-static int __cpool_alloc_ahash(struct sigpool_entry *e, const char *alg)
-{
- struct crypto_ahash *cpu0_hash;
- int ret;
-
- e->alg = kstrdup(alg, GFP_KERNEL);
- if (!e->alg)
- return -ENOMEM;
-
- cpu0_hash = crypto_alloc_ahash(alg, 0, CRYPTO_ALG_ASYNC);
- if (IS_ERR(cpu0_hash)) {
- ret = PTR_ERR(cpu0_hash);
- goto out_free_alg;
- }
-
- e->needs_key = crypto_ahash_get_flags(cpu0_hash) & CRYPTO_TFM_NEED_KEY;
-
- ret = __cpool_try_clone(cpu0_hash);
- if (ret)
- goto out_free_cpu0_hash;
- e->hash = cpu0_hash;
- kref_init(&e->kref);
- return 0;
-
-out_free_cpu0_hash:
- crypto_free_ahash(cpu0_hash);
-out_free_alg:
- kfree(e->alg);
- e->alg = NULL;
- return ret;
-}
-
-/**
- * tcp_sigpool_alloc_ahash - allocates pool for ahash requests
- * @alg: name of async hash algorithm
- * @scratch_size: reserve a tcp_sigpool::scratch buffer of this size
- */
-int tcp_sigpool_alloc_ahash(const char *alg, size_t scratch_size)
-{
- int i, ret;
-
- /* slow-path */
- mutex_lock(&cpool_mutex);
- ret = sigpool_reserve_scratch(scratch_size);
- if (ret)
- goto out;
- for (i = 0; i < cpool_populated; i++) {
- if (!cpool[i].alg)
- continue;
- if (strcmp(cpool[i].alg, alg))
- continue;
-
- /* pairs with tcp_sigpool_release() */
- if (!kref_get_unless_zero(&cpool[i].kref))
- kref_init(&cpool[i].kref);
- ret = i;
- goto out;
- }
-
- for (i = 0; i < cpool_populated; i++) {
- if (!cpool[i].alg)
- break;
- }
- if (i >= CPOOL_SIZE) {
- ret = -ENOSPC;
- goto out;
- }
-
- ret = __cpool_alloc_ahash(&cpool[i], alg);
- if (!ret) {
- ret = i;
- if (i == cpool_populated)
- cpool_populated++;
- }
-out:
- mutex_unlock(&cpool_mutex);
- return ret;
-}
-EXPORT_SYMBOL_GPL(tcp_sigpool_alloc_ahash);
-
-static void __cpool_free_entry(struct sigpool_entry *e)
-{
- crypto_free_ahash(e->hash);
- kfree(e->alg);
- memset(e, 0, sizeof(*e));
-}
-
-static void cpool_cleanup_work_cb(struct work_struct *work)
-{
- bool free_scratch = true;
- unsigned int i;
-
- mutex_lock(&cpool_mutex);
- for (i = 0; i < cpool_populated; i++) {
- if (kref_read(&cpool[i].kref) > 0) {
- free_scratch = false;
- continue;
- }
- if (!cpool[i].alg)
- continue;
- __cpool_free_entry(&cpool[i]);
- }
- if (free_scratch)
- sigpool_scratch_free();
- mutex_unlock(&cpool_mutex);
-}
-
-static DECLARE_WORK(cpool_cleanup_work, cpool_cleanup_work_cb);
-static void cpool_schedule_cleanup(struct kref *kref)
-{
- schedule_work(&cpool_cleanup_work);
-}
-
-/**
- * tcp_sigpool_release - decreases number of users for a pool. If it was
- * the last user of the pool, releases any memory that was consumed.
- * @id: tcp_sigpool that was previously allocated by tcp_sigpool_alloc_ahash()
- */
-void tcp_sigpool_release(unsigned int id)
-{
- if (WARN_ON_ONCE(id >= cpool_populated || !cpool[id].alg))
- return;
-
- /* slow-path */
- kref_put(&cpool[id].kref, cpool_schedule_cleanup);
-}
-EXPORT_SYMBOL_GPL(tcp_sigpool_release);
-
-/**
- * tcp_sigpool_get - increases number of users (refcounter) for a pool
- * @id: tcp_sigpool that was previously allocated by tcp_sigpool_alloc_ahash()
- */
-void tcp_sigpool_get(unsigned int id)
-{
- if (WARN_ON_ONCE(id >= cpool_populated || !cpool[id].alg))
- return;
- kref_get(&cpool[id].kref);
-}
-EXPORT_SYMBOL_GPL(tcp_sigpool_get);
-
-int tcp_sigpool_start(unsigned int id, struct tcp_sigpool *c) __cond_acquires(0, RCU_BH)
-{
- struct crypto_ahash *hash;
-
- rcu_read_lock_bh();
- if (WARN_ON_ONCE(id >= cpool_populated || !cpool[id].alg)) {
- rcu_read_unlock_bh();
- return -EINVAL;
- }
-
- hash = crypto_clone_ahash(cpool[id].hash);
- if (IS_ERR(hash)) {
- rcu_read_unlock_bh();
- return PTR_ERR(hash);
- }
-
- c->req = ahash_request_alloc(hash, GFP_ATOMIC);
- if (!c->req) {
- crypto_free_ahash(hash);
- rcu_read_unlock_bh();
- return -ENOMEM;
- }
- ahash_request_set_callback(c->req, 0, NULL, NULL);
-
- /* Pairs with tcp_sigpool_reserve_scratch(), scratch area is
- * valid (allocated) until tcp_sigpool_end().
- */
- local_lock_nested_bh(&sigpool_scratch.bh_lock);
- c->scratch = rcu_dereference_bh(*this_cpu_ptr(&sigpool_scratch.pad));
- return 0;
-}
-EXPORT_SYMBOL_GPL(tcp_sigpool_start);
-
-void tcp_sigpool_end(struct tcp_sigpool *c) __releases(RCU_BH)
-{
- struct crypto_ahash *hash = crypto_ahash_reqtfm(c->req);
-
- local_unlock_nested_bh(&sigpool_scratch.bh_lock);
- rcu_read_unlock_bh();
- ahash_request_free(c->req);
- crypto_free_ahash(hash);
-}
-EXPORT_SYMBOL_GPL(tcp_sigpool_end);
-
-/**
- * tcp_sigpool_algo - return algorithm of tcp_sigpool
- * @id: tcp_sigpool that was previously allocated by tcp_sigpool_alloc_ahash()
- * @buf: buffer to return name of algorithm
- * @buf_len: size of @buf
- */
-size_t tcp_sigpool_algo(unsigned int id, char *buf, size_t buf_len)
-{
- if (WARN_ON_ONCE(id >= cpool_populated || !cpool[id].alg))
- return -EINVAL;
-
- return strscpy(buf, cpool[id].alg, buf_len);
-}
-EXPORT_SYMBOL_GPL(tcp_sigpool_algo);
-
-/**
- * tcp_sigpool_hash_skb_data - hash data in skb with initialized tcp_sigpool
- * @hp: tcp_sigpool pointer
- * @skb: buffer to add sign for
- * @header_len: TCP header length for this segment
- */
-int tcp_sigpool_hash_skb_data(struct tcp_sigpool *hp,
- const struct sk_buff *skb,
- unsigned int header_len)
-{
- const unsigned int head_data_len = skb_headlen(skb) > header_len ?
- skb_headlen(skb) - header_len : 0;
- const struct skb_shared_info *shi = skb_shinfo(skb);
- const struct tcphdr *tp = tcp_hdr(skb);
- struct ahash_request *req = hp->req;
- struct sk_buff *frag_iter;
- struct scatterlist sg;
- unsigned int i;
-
- sg_init_table(&sg, 1);
-
- sg_set_buf(&sg, ((u8 *)tp) + header_len, head_data_len);
- ahash_request_set_crypt(req, &sg, NULL, head_data_len);
- if (crypto_ahash_update(req))
- return 1;
-
- for (i = 0; i < shi->nr_frags; ++i) {
- const skb_frag_t *f = &shi->frags[i];
- unsigned int offset = skb_frag_off(f);
- struct page *page;
-
- page = skb_frag_page(f) + (offset >> PAGE_SHIFT);
- sg_set_page(&sg, page, skb_frag_size(f), offset_in_page(offset));
- ahash_request_set_crypt(req, &sg, NULL, skb_frag_size(f));
- if (crypto_ahash_update(req))
- return 1;
- }
-
- skb_walk_frags(skb, frag_iter)
- if (tcp_sigpool_hash_skb_data(hp, frag_iter, 0))
- return 1;
-
- return 0;
-}
-EXPORT_SYMBOL(tcp_sigpool_hash_skb_data);
-
-MODULE_LICENSE("GPL");
-MODULE_DESCRIPTION("Per-CPU pool of crypto requests");
--
2.54.0
^ permalink raw reply related [flat|nested] 8+ messages in thread* Re: [PATCH net-next v2 0/5] Reimplement TCP-AO using crypto library
2026-04-27 17:27 [PATCH net-next v2 0/5] Reimplement TCP-AO using crypto library Eric Biggers
` (4 preceding siblings ...)
2026-04-27 17:27 ` [PATCH net-next v2 5/5] net/tcp: Remove tcp_sigpool Eric Biggers
@ 2026-04-27 19:09 ` Dmitry Safonov
2026-04-27 20:01 ` Eric Biggers
5 siblings, 1 reply; 8+ messages in thread
From: Dmitry Safonov @ 2026-04-27 19:09 UTC (permalink / raw)
To: Eric Biggers
Cc: netdev, linux-crypto, linux-kernel, Eric Dumazet, Neal Cardwell,
Kuniyuki Iwashima, David S . Miller, David Ahern, Jakub Kicinski,
Paolo Abeni, Simon Horman, Ard Biesheuvel, Jason A . Donenfeld,
Herbert Xu, Dmitry Safonov
Hi Eric,
On Mon, 27 Apr 2026 at 18:27, Eric Biggers <ebiggers@kernel.org> wrote:
[..]
> To make this simplification and optimization possible, this series also
> updates the TCP-AO code to support a specific set of algorithms, rather
> than arbitrary algorithms that don't make sense and are very likely not
> being used, e.g. CRC-32 and HMAC-MD5.
>
> Specifically, this series retains the support for AES-128-CMAC,
> HMAC-SHA1, and HMAC-SHA256. AES-128-CMAC and HMAC-SHA1 are the only
> algorithms that are actually standardized for use in TCP-AO, while
> HMAC-SHA256 makes sense to continue supporting as a Linux extension. Of
> course, other algorithms can still be (re-)added later if ever needed.
> It's worth noting that TCP-AO MACs are limited to 20 bytes by the TCP
> options space, which limits the benefit of further algorithm upgrades.
>
> This series passes the tcp_ao selftests
> (sudo make -C tools/testing/selftests/net/tcp_ao/ run_tests).
>
> To get a sense for how much more efficient this makes the TCP-AO code,
> here's a microbenchmark for tcp_ao_hash_skb() with skb->len == 128:
>
> Algorithm Avg cycles (before) Avg cycles (after)
> --------- ------------------- ------------------
> HMAC-SHA1 3319 1256
> HMAC-SHA256 3311 1344
> AES-128-CMAC 2720 1107
I do like these numbers quite much! Yet, as I mentioned in version 1,
removing a fallback for other algorithms' support does not sound good
to me. There are two reasons:
- Ronald P. Bonica (the original RFC5925 author), together with Tony
Li do have an active RFC draft to support the additional algorithms
[1], potentially in addition to TCP Extended Options [2]
- There is at least one open-source BGP implementation (BIRD) that
allows using the algorithms that you are removing [3]. Without a
deprecation period and communication with at least known open source
users, it implies intentionally breaking them, which I can't agree
with.
I don't feel like Naking as we don't have any customers using anything
other than the 3 algorithms above (and BGP implementation is
[unfortunately] closed-source, so that would not feel appropriate even
if we had such customers), yet I do feel like it's worth and
appropriate to express my thoughts/concerns.
[1] https://www.ietf.org/archive/id/draft-bonica-tcpm-tcp-ao-algs-00.html
[2] https://www.ietf.org/archive/id/draft-bonica-tcpm-extended-options-00.html
[3] https://github.com/CZ-NIC/bird/blob/master/sysdep/linux/sysio.h#L246
Thanks,
Dmitry
^ permalink raw reply [flat|nested] 8+ messages in thread* Re: [PATCH net-next v2 0/5] Reimplement TCP-AO using crypto library
2026-04-27 19:09 ` [PATCH net-next v2 0/5] Reimplement TCP-AO using crypto library Dmitry Safonov
@ 2026-04-27 20:01 ` Eric Biggers
0 siblings, 0 replies; 8+ messages in thread
From: Eric Biggers @ 2026-04-27 20:01 UTC (permalink / raw)
To: Dmitry Safonov
Cc: netdev, linux-crypto, linux-kernel, Eric Dumazet, Neal Cardwell,
Kuniyuki Iwashima, David S . Miller, David Ahern, Jakub Kicinski,
Paolo Abeni, Simon Horman, Ard Biesheuvel, Jason A . Donenfeld,
Herbert Xu, Dmitry Safonov
On Mon, Apr 27, 2026 at 08:09:05PM +0100, Dmitry Safonov wrote:
> > To get a sense for how much more efficient this makes the TCP-AO code,
> > here's a microbenchmark for tcp_ao_hash_skb() with skb->len == 128:
> >
> > Algorithm Avg cycles (before) Avg cycles (after)
> > --------- ------------------- ------------------
> > HMAC-SHA1 3319 1256
> > HMAC-SHA256 3311 1344
> > AES-128-CMAC 2720 1107
>
>
> I do like these numbers quite much! Yet, as I mentioned in version 1,
> removing a fallback for other algorithms' support does not sound good
> to me. There are two reasons:
> - Ronald P. Bonica (the original RFC5925 author), together with Tony
> Li do have an active RFC draft to support the additional algorithms
> [1], potentially in addition to TCP Extended Options [2]
> - There is at least one open-source BGP implementation (BIRD) that
> allows using the algorithms that you are removing [3]. Without a
> deprecation period and communication with at least known open source
> users, it implies intentionally breaking them, which I can't agree
> with.
>
> I don't feel like Naking as we don't have any customers using anything
> other than the 3 algorithms above (and BGP implementation is
> [unfortunately] closed-source, so that would not feel appropriate even
> if we had such customers), yet I do feel like it's worth and
> appropriate to express my thoughts/concerns.
>
> [1] https://www.ietf.org/archive/id/draft-bonica-tcpm-tcp-ao-algs-00.html
> [2] https://www.ietf.org/archive/id/draft-bonica-tcpm-extended-options-00.html
> [3] https://github.com/CZ-NIC/bird/blob/master/sysdep/linux/sysio.h#L246
I think the usual "it's not really broken if no one notices" is likely
to apply here. Indeed, there have been many cases where algorithms have
been removed from the crypto API before, despite this theoretically
impacting UAPI. Just some of the removals from crypto_ahash over the
years that come to mind are the Tiger hash algorithms, multiple variants
of RIPEMD, multiple CRC variants, GHASH, Poly1305, NH, and POLYVAL.
The reality is that the crypto API's algorithm specification language
provides way more "flexibility" than anyone knows what to do with or
ever should have existed at all, let alone been exposed directly to
userspace.
We don't have a lot of choice but to clean up these old mistakes to keep
Linux maintainable going forwards, reduce the chance for user error, and
optimize for the things that actually matter.
And again, as I said, if there is another algorithm that someone
actually needs, we can add it back as a bug fix (or as a new feature,
considering that some never worked in the first place).
- Eric
^ permalink raw reply [flat|nested] 8+ messages in thread