p71924506
/
linux_old1
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
linux_old1/net/tls/tls_device_fallback.c

469 lines
13 KiB
C
Raw Normal View History

net/tls: Add generic NIC offload infrastructure This patch adds a generic infrastructure to offload TLS crypto to a network device. It enables the kernel TLS socket to skip encryption and authentication operations on the transmit side of the data path. Leaving those computationally expensive operations to the NIC. The NIC offload infrastructure builds TLS records and pushes them to the TCP layer just like the SW KTLS implementation and using the same API. TCP segmentation is mostly unaffected. Currently the only exception is that we prevent mixed SKBs where only part of the payload requires offload. In the future we are likely to add a similar restriction following a change cipher spec record. The notable differences between SW KTLS and NIC offloaded TLS implementations are as follows: 1. The offloaded implementation builds "plaintext TLS record", those records contain plaintext instead of ciphertext and place holder bytes instead of authentication tags. 2. The offloaded implementation maintains a mapping from TCP sequence number to TLS records. Thus given a TCP SKB sent from a NIC offloaded TLS socket, we can use the tls NIC offload infrastructure to obtain enough context to encrypt the payload of the SKB. A TLS record is released when the last byte of the record is ack'ed, this is done through the new icsk_clean_acked callback. The infrastructure should be extendable to support various NIC offload implementations. However it is currently written with the implementation below in mind: The NIC assumes that packets from each offloaded stream are sent as plaintext and in-order. It keeps track of the TLS records in the TCP stream. When a packet marked for offload is transmitted, the NIC encrypts the payload in-place and puts authentication tags in the relevant place holders. The responsibility for handling out-of-order packets (i.e. TCP retransmission, qdisc drops) falls on the netdev driver. The netdev driver keeps track of the expected TCP SN from the NIC's perspective. If the next packet to transmit matches the expected TCP SN, the driver advances the expected TCP SN, and transmits the packet with TLS offload indication. If the next packet to transmit does not match the expected TCP SN. The driver calls the TLS layer to obtain the TLS record that includes the TCP of the packet for transmission. Using this TLS record, the driver posts a work entry on the transmit queue to reconstruct the NIC TLS state required for the offload of the out-of-order packet. It updates the expected TCP SN accordingly and transmits the now in-order packet. The same queue is used for packet transmission and TLS context reconstruction to avoid the need for flushing the transmit queue before issuing the context reconstruction request. Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com> Signed-off-by: Boris Pismenny <borisp@mellanox.com> Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-30 15:16:16 +08:00
/* Copyright (c) 2018, Mellanox Technologies All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#include <net/tls.h>
#include <crypto/aead.h>
#include <crypto/scatterwalk.h>
#include <net/ip6_checksum.h>
static void chain_to_walk(struct scatterlist *sg, struct scatter_walk *walk)
{
struct scatterlist *src = walk->sg;
int diff = walk->offset - src->offset;
sg_set_page(sg, sg_page(src),
src->length - diff, walk->offset);
crypto: scatterwalk - remove 'chain' argument from scatterwalk_crypto_chain() All callers pass chain=0 to scatterwalk_crypto_chain(). Remove this unneeded parameter. Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-07-24 01:01:33 +08:00
scatterwalk_crypto_chain(sg, sg_next(src), 2);
net/tls: Add generic NIC offload infrastructure This patch adds a generic infrastructure to offload TLS crypto to a network device. It enables the kernel TLS socket to skip encryption and authentication operations on the transmit side of the data path. Leaving those computationally expensive operations to the NIC. The NIC offload infrastructure builds TLS records and pushes them to the TCP layer just like the SW KTLS implementation and using the same API. TCP segmentation is mostly unaffected. Currently the only exception is that we prevent mixed SKBs where only part of the payload requires offload. In the future we are likely to add a similar restriction following a change cipher spec record. The notable differences between SW KTLS and NIC offloaded TLS implementations are as follows: 1. The offloaded implementation builds "plaintext TLS record", those records contain plaintext instead of ciphertext and place holder bytes instead of authentication tags. 2. The offloaded implementation maintains a mapping from TCP sequence number to TLS records. Thus given a TCP SKB sent from a NIC offloaded TLS socket, we can use the tls NIC offload infrastructure to obtain enough context to encrypt the payload of the SKB. A TLS record is released when the last byte of the record is ack'ed, this is done through the new icsk_clean_acked callback. The infrastructure should be extendable to support various NIC offload implementations. However it is currently written with the implementation below in mind: The NIC assumes that packets from each offloaded stream are sent as plaintext and in-order. It keeps track of the TLS records in the TCP stream. When a packet marked for offload is transmitted, the NIC encrypts the payload in-place and puts authentication tags in the relevant place holders. The responsibility for handling out-of-order packets (i.e. TCP retransmission, qdisc drops) falls on the netdev driver. The netdev driver keeps track of the expected TCP SN from the NIC's perspective. If the next packet to transmit matches the expected TCP SN, the driver advances the expected TCP SN, and transmits the packet with TLS offload indication. If the next packet to transmit does not match the expected TCP SN. The driver calls the TLS layer to obtain the TLS record that includes the TCP of the packet for transmission. Using this TLS record, the driver posts a work entry on the transmit queue to reconstruct the NIC TLS state required for the offload of the out-of-order packet. It updates the expected TCP SN accordingly and transmits the now in-order packet. The same queue is used for packet transmission and TLS context reconstruction to avoid the need for flushing the transmit queue before issuing the context reconstruction request. Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com> Signed-off-by: Boris Pismenny <borisp@mellanox.com> Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-30 15:16:16 +08:00
}
static int tls_enc_record(struct aead_request *aead_req,
struct crypto_aead *aead, char *aad,
char *iv, __be64 rcd_sn,
struct scatter_walk *in,
struct scatter_walk *out, int *in_len)
{
unsigned char buf[TLS_HEADER_SIZE + TLS_CIPHER_AES_GCM_128_IV_SIZE];
struct scatterlist sg_in[3];
struct scatterlist sg_out[3];
u16 len;
int rc;
len = min_t(int, *in_len, ARRAY_SIZE(buf));
scatterwalk_copychunks(buf, in, len, 0);
scatterwalk_copychunks(buf, out, len, 1);
*in_len -= len;
if (!*in_len)
return 0;
scatterwalk_pagedone(in, 0, 1);
scatterwalk_pagedone(out, 1, 1);
len = buf[4] | (buf[3] << 8);
len -= TLS_CIPHER_AES_GCM_128_IV_SIZE;
tls_make_aad(aad, len - TLS_CIPHER_AES_GCM_128_TAG_SIZE,
net: tls: Add tls 1.3 support TLS 1.3 has minor changes from TLS 1.2 at the record layer. * Header now hardcodes the same version and application content type in the header. * The real content type is appended after the data, before encryption (or after decryption). * The IV is xored with the sequence number, instead of concatinating four bytes of IV with the explicit IV. * Zero-padding: No exlicit length is given, we search backwards from the end of the decrypted data for the first non-zero byte, which is the content type. Currently recv supports reading zero-padding, but there is no way for send to add zero padding. Signed-off-by: Dave Watson <davejwatson@fb.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-01-31 05:58:31 +08:00
(char *)&rcd_sn, sizeof(rcd_sn), buf[0],
TLS_1_2_VERSION);
net/tls: Add generic NIC offload infrastructure This patch adds a generic infrastructure to offload TLS crypto to a network device. It enables the kernel TLS socket to skip encryption and authentication operations on the transmit side of the data path. Leaving those computationally expensive operations to the NIC. The NIC offload infrastructure builds TLS records and pushes them to the TCP layer just like the SW KTLS implementation and using the same API. TCP segmentation is mostly unaffected. Currently the only exception is that we prevent mixed SKBs where only part of the payload requires offload. In the future we are likely to add a similar restriction following a change cipher spec record. The notable differences between SW KTLS and NIC offloaded TLS implementations are as follows: 1. The offloaded implementation builds "plaintext TLS record", those records contain plaintext instead of ciphertext and place holder bytes instead of authentication tags. 2. The offloaded implementation maintains a mapping from TCP sequence number to TLS records. Thus given a TCP SKB sent from a NIC offloaded TLS socket, we can use the tls NIC offload infrastructure to obtain enough context to encrypt the payload of the SKB. A TLS record is released when the last byte of the record is ack'ed, this is done through the new icsk_clean_acked callback. The infrastructure should be extendable to support various NIC offload implementations. However it is currently written with the implementation below in mind: The NIC assumes that packets from each offloaded stream are sent as plaintext and in-order. It keeps track of the TLS records in the TCP stream. When a packet marked for offload is transmitted, the NIC encrypts the payload in-place and puts authentication tags in the relevant place holders. The responsibility for handling out-of-order packets (i.e. TCP retransmission, qdisc drops) falls on the netdev driver. The netdev driver keeps track of the expected TCP SN from the NIC's perspective. If the next packet to transmit matches the expected TCP SN, the driver advances the expected TCP SN, and transmits the packet with TLS offload indication. If the next packet to transmit does not match the expected TCP SN. The driver calls the TLS layer to obtain the TLS record that includes the TCP of the packet for transmission. Using this TLS record, the driver posts a work entry on the transmit queue to reconstruct the NIC TLS state required for the offload of the out-of-order packet. It updates the expected TCP SN accordingly and transmits the now in-order packet. The same queue is used for packet transmission and TLS context reconstruction to avoid the need for flushing the transmit queue before issuing the context reconstruction request. Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com> Signed-off-by: Boris Pismenny <borisp@mellanox.com> Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-30 15:16:16 +08:00
memcpy(iv + TLS_CIPHER_AES_GCM_128_SALT_SIZE, buf + TLS_HEADER_SIZE,
TLS_CIPHER_AES_GCM_128_IV_SIZE);
sg_init_table(sg_in, ARRAY_SIZE(sg_in));
sg_init_table(sg_out, ARRAY_SIZE(sg_out));
sg_set_buf(sg_in, aad, TLS_AAD_SPACE_SIZE);
sg_set_buf(sg_out, aad, TLS_AAD_SPACE_SIZE);
chain_to_walk(sg_in + 1, in);
chain_to_walk(sg_out + 1, out);
*in_len -= len;
if (*in_len < 0) {
*in_len += TLS_CIPHER_AES_GCM_128_TAG_SIZE;
/* the input buffer doesn't contain the entire record.
* trim len accordingly. The resulting authentication tag
* will contain garbage, but we don't care, so we won't
* include any of it in the output skb
* Note that we assume the output buffer length
* is larger then input buffer length + tag size
*/
if (*in_len < 0)
len += *in_len;
*in_len = 0;
}
if (*in_len) {
scatterwalk_copychunks(NULL, in, len, 2);
scatterwalk_pagedone(in, 0, 1);
scatterwalk_copychunks(NULL, out, len, 2);
scatterwalk_pagedone(out, 1, 1);
}
len -= TLS_CIPHER_AES_GCM_128_TAG_SIZE;
aead_request_set_crypt(aead_req, sg_in, sg_out, len, iv);
rc = crypto_aead_encrypt(aead_req);
return rc;
}
static void tls_init_aead_request(struct aead_request *aead_req,
struct crypto_aead *aead)
{
aead_request_set_tfm(aead_req, aead);
aead_request_set_ad(aead_req, TLS_AAD_SPACE_SIZE);
}
static struct aead_request *tls_alloc_aead_request(struct crypto_aead *aead,
gfp_t flags)
{
unsigned int req_size = sizeof(struct aead_request) +
crypto_aead_reqsize(aead);
struct aead_request *aead_req;
aead_req = kzalloc(req_size, flags);
if (aead_req)
tls_init_aead_request(aead_req, aead);
return aead_req;
}
static int tls_enc_records(struct aead_request *aead_req,
struct crypto_aead *aead, struct scatterlist *sg_in,
struct scatterlist *sg_out, char *aad, char *iv,
u64 rcd_sn, int len)
{
struct scatter_walk out, in;
int rc;
scatterwalk_start(&in, sg_in);
scatterwalk_start(&out, sg_out);
do {
rc = tls_enc_record(aead_req, aead, aad, iv,
cpu_to_be64(rcd_sn), &in, &out, &len);
rcd_sn++;
} while (rc == 0 && len);
scatterwalk_done(&in, 0, 0);
scatterwalk_done(&out, 1, 0);
return rc;
}
/* Can't use icsk->icsk_af_ops->send_check here because the ip addresses
* might have been changed by NAT.
*/
static void update_chksum(struct sk_buff *skb, int headln)
{
struct tcphdr *th = tcp_hdr(skb);
int datalen = skb->len - headln;
const struct ipv6hdr *ipv6h;
const struct iphdr *iph;
/* We only changed the payload so if we are using partial we don't
* need to update anything.
*/
if (likely(skb->ip_summed == CHECKSUM_PARTIAL))
return;
skb->ip_summed = CHECKSUM_PARTIAL;
skb->csum_start = skb_transport_header(skb) - skb->head;
skb->csum_offset = offsetof(struct tcphdr, check);
if (skb->sk->sk_family == AF_INET6) {
ipv6h = ipv6_hdr(skb);
th->check = ~csum_ipv6_magic(&ipv6h->saddr, &ipv6h->daddr,
datalen, IPPROTO_TCP, 0);
} else {
iph = ip_hdr(skb);
th->check = ~csum_tcpudp_magic(iph->saddr, iph->daddr, datalen,
IPPROTO_TCP, 0);
}
}
static void complete_skb(struct sk_buff *nskb, struct sk_buff *skb, int headln)
{
net/tls: fix refcount adjustment in fallback Unlike atomic_add(), refcount_add() does not deal well with a negative argument. TLS fallback code reallocates the skb and is very likely to shrink the truesize, leading to: [ 189.513254] WARNING: CPU: 5 PID: 0 at lib/refcount.c:81 refcount_add_not_zero_checked+0x15c/0x180 Call Trace: refcount_add_checked+0x6/0x40 tls_enc_skb+0xb93/0x13e0 [tls] Once wmem_allocated count saturates the application can no longer send data on the socket. This is similar to Eric's fixes for GSO, TCP: commit 7ec318feeed1 ("tcp: gso: avoid refcount_t warning from tcp_gso_segment()") and UDP: commit 575b65bc5bff ("udp: avoid refcount_t saturation in __udp_gso_segment()"). Unlike the GSO case, for TLS fallback it's likely that the skb has shrunk, so the "likely" annotation is the other way around (likely branch being "sub"). Fixes: e8f69799810c ("net/tls: Add generic NIC offload infrastructure") Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: John Hurley <john.hurley@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-18 01:51:19 +08:00
struct sock *sk = skb->sk;
int delta;
net/tls: Add generic NIC offload infrastructure This patch adds a generic infrastructure to offload TLS crypto to a network device. It enables the kernel TLS socket to skip encryption and authentication operations on the transmit side of the data path. Leaving those computationally expensive operations to the NIC. The NIC offload infrastructure builds TLS records and pushes them to the TCP layer just like the SW KTLS implementation and using the same API. TCP segmentation is mostly unaffected. Currently the only exception is that we prevent mixed SKBs where only part of the payload requires offload. In the future we are likely to add a similar restriction following a change cipher spec record. The notable differences between SW KTLS and NIC offloaded TLS implementations are as follows: 1. The offloaded implementation builds "plaintext TLS record", those records contain plaintext instead of ciphertext and place holder bytes instead of authentication tags. 2. The offloaded implementation maintains a mapping from TCP sequence number to TLS records. Thus given a TCP SKB sent from a NIC offloaded TLS socket, we can use the tls NIC offload infrastructure to obtain enough context to encrypt the payload of the SKB. A TLS record is released when the last byte of the record is ack'ed, this is done through the new icsk_clean_acked callback. The infrastructure should be extendable to support various NIC offload implementations. However it is currently written with the implementation below in mind: The NIC assumes that packets from each offloaded stream are sent as plaintext and in-order. It keeps track of the TLS records in the TCP stream. When a packet marked for offload is transmitted, the NIC encrypts the payload in-place and puts authentication tags in the relevant place holders. The responsibility for handling out-of-order packets (i.e. TCP retransmission, qdisc drops) falls on the netdev driver. The netdev driver keeps track of the expected TCP SN from the NIC's perspective. If the next packet to transmit matches the expected TCP SN, the driver advances the expected TCP SN, and transmits the packet with TLS offload indication. If the next packet to transmit does not match the expected TCP SN. The driver calls the TLS layer to obtain the TLS record that includes the TCP of the packet for transmission. Using this TLS record, the driver posts a work entry on the transmit queue to reconstruct the NIC TLS state required for the offload of the out-of-order packet. It updates the expected TCP SN accordingly and transmits the now in-order packet. The same queue is used for packet transmission and TLS context reconstruction to avoid the need for flushing the transmit queue before issuing the context reconstruction request. Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com> Signed-off-by: Boris Pismenny <borisp@mellanox.com> Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-30 15:16:16 +08:00
skb_copy_header(nskb, skb);
skb_put(nskb, skb->len);
memcpy(nskb->data, skb->data, headln);
nskb->destructor = skb->destructor;
net/tls: fix refcount adjustment in fallback Unlike atomic_add(), refcount_add() does not deal well with a negative argument. TLS fallback code reallocates the skb and is very likely to shrink the truesize, leading to: [ 189.513254] WARNING: CPU: 5 PID: 0 at lib/refcount.c:81 refcount_add_not_zero_checked+0x15c/0x180 Call Trace: refcount_add_checked+0x6/0x40 tls_enc_skb+0xb93/0x13e0 [tls] Once wmem_allocated count saturates the application can no longer send data on the socket. This is similar to Eric's fixes for GSO, TCP: commit 7ec318feeed1 ("tcp: gso: avoid refcount_t warning from tcp_gso_segment()") and UDP: commit 575b65bc5bff ("udp: avoid refcount_t saturation in __udp_gso_segment()"). Unlike the GSO case, for TLS fallback it's likely that the skb has shrunk, so the "likely" annotation is the other way around (likely branch being "sub"). Fixes: e8f69799810c ("net/tls: Add generic NIC offload infrastructure") Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: John Hurley <john.hurley@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-18 01:51:19 +08:00
nskb->sk = sk;
net/tls: Add generic NIC offload infrastructure This patch adds a generic infrastructure to offload TLS crypto to a network device. It enables the kernel TLS socket to skip encryption and authentication operations on the transmit side of the data path. Leaving those computationally expensive operations to the NIC. The NIC offload infrastructure builds TLS records and pushes them to the TCP layer just like the SW KTLS implementation and using the same API. TCP segmentation is mostly unaffected. Currently the only exception is that we prevent mixed SKBs where only part of the payload requires offload. In the future we are likely to add a similar restriction following a change cipher spec record. The notable differences between SW KTLS and NIC offloaded TLS implementations are as follows: 1. The offloaded implementation builds "plaintext TLS record", those records contain plaintext instead of ciphertext and place holder bytes instead of authentication tags. 2. The offloaded implementation maintains a mapping from TCP sequence number to TLS records. Thus given a TCP SKB sent from a NIC offloaded TLS socket, we can use the tls NIC offload infrastructure to obtain enough context to encrypt the payload of the SKB. A TLS record is released when the last byte of the record is ack'ed, this is done through the new icsk_clean_acked callback. The infrastructure should be extendable to support various NIC offload implementations. However it is currently written with the implementation below in mind: The NIC assumes that packets from each offloaded stream are sent as plaintext and in-order. It keeps track of the TLS records in the TCP stream. When a packet marked for offload is transmitted, the NIC encrypts the payload in-place and puts authentication tags in the relevant place holders. The responsibility for handling out-of-order packets (i.e. TCP retransmission, qdisc drops) falls on the netdev driver. The netdev driver keeps track of the expected TCP SN from the NIC's perspective. If the next packet to transmit matches the expected TCP SN, the driver advances the expected TCP SN, and transmits the packet with TLS offload indication. If the next packet to transmit does not match the expected TCP SN. The driver calls the TLS layer to obtain the TLS record that includes the TCP of the packet for transmission. Using this TLS record, the driver posts a work entry on the transmit queue to reconstruct the NIC TLS state required for the offload of the out-of-order packet. It updates the expected TCP SN accordingly and transmits the now in-order packet. The same queue is used for packet transmission and TLS context reconstruction to avoid the need for flushing the transmit queue before issuing the context reconstruction request. Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com> Signed-off-by: Boris Pismenny <borisp@mellanox.com> Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-30 15:16:16 +08:00
skb->destructor = NULL;
skb->sk = NULL;
net/tls: fix refcount adjustment in fallback Unlike atomic_add(), refcount_add() does not deal well with a negative argument. TLS fallback code reallocates the skb and is very likely to shrink the truesize, leading to: [ 189.513254] WARNING: CPU: 5 PID: 0 at lib/refcount.c:81 refcount_add_not_zero_checked+0x15c/0x180 Call Trace: refcount_add_checked+0x6/0x40 tls_enc_skb+0xb93/0x13e0 [tls] Once wmem_allocated count saturates the application can no longer send data on the socket. This is similar to Eric's fixes for GSO, TCP: commit 7ec318feeed1 ("tcp: gso: avoid refcount_t warning from tcp_gso_segment()") and UDP: commit 575b65bc5bff ("udp: avoid refcount_t saturation in __udp_gso_segment()"). Unlike the GSO case, for TLS fallback it's likely that the skb has shrunk, so the "likely" annotation is the other way around (likely branch being "sub"). Fixes: e8f69799810c ("net/tls: Add generic NIC offload infrastructure") Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: John Hurley <john.hurley@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-18 01:51:19 +08:00
net/tls: avoid NULL pointer deref on nskb->sk in fallback update_chksum() accesses nskb->sk before it has been set by complete_skb(), move the init up. Fixes: e8f69799810c ("net/tls: Add generic NIC offload infrastructure") Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-30 03:19:12 +08:00
update_chksum(nskb, headln);
net/tls: fix refcount adjustment in fallback Unlike atomic_add(), refcount_add() does not deal well with a negative argument. TLS fallback code reallocates the skb and is very likely to shrink the truesize, leading to: [ 189.513254] WARNING: CPU: 5 PID: 0 at lib/refcount.c:81 refcount_add_not_zero_checked+0x15c/0x180 Call Trace: refcount_add_checked+0x6/0x40 tls_enc_skb+0xb93/0x13e0 [tls] Once wmem_allocated count saturates the application can no longer send data on the socket. This is similar to Eric's fixes for GSO, TCP: commit 7ec318feeed1 ("tcp: gso: avoid refcount_t warning from tcp_gso_segment()") and UDP: commit 575b65bc5bff ("udp: avoid refcount_t saturation in __udp_gso_segment()"). Unlike the GSO case, for TLS fallback it's likely that the skb has shrunk, so the "likely" annotation is the other way around (likely branch being "sub"). Fixes: e8f69799810c ("net/tls: Add generic NIC offload infrastructure") Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: John Hurley <john.hurley@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-18 01:51:19 +08:00
delta = nskb->truesize - skb->truesize;
if (likely(delta < 0))
WARN_ON_ONCE(refcount_sub_and_test(-delta, &sk->sk_wmem_alloc));
else if (delta)
refcount_add(delta, &sk->sk_wmem_alloc);
net/tls: Add generic NIC offload infrastructure This patch adds a generic infrastructure to offload TLS crypto to a network device. It enables the kernel TLS socket to skip encryption and authentication operations on the transmit side of the data path. Leaving those computationally expensive operations to the NIC. The NIC offload infrastructure builds TLS records and pushes them to the TCP layer just like the SW KTLS implementation and using the same API. TCP segmentation is mostly unaffected. Currently the only exception is that we prevent mixed SKBs where only part of the payload requires offload. In the future we are likely to add a similar restriction following a change cipher spec record. The notable differences between SW KTLS and NIC offloaded TLS implementations are as follows: 1. The offloaded implementation builds "plaintext TLS record", those records contain plaintext instead of ciphertext and place holder bytes instead of authentication tags. 2. The offloaded implementation maintains a mapping from TCP sequence number to TLS records. Thus given a TCP SKB sent from a NIC offloaded TLS socket, we can use the tls NIC offload infrastructure to obtain enough context to encrypt the payload of the SKB. A TLS record is released when the last byte of the record is ack'ed, this is done through the new icsk_clean_acked callback. The infrastructure should be extendable to support various NIC offload implementations. However it is currently written with the implementation below in mind: The NIC assumes that packets from each offloaded stream are sent as plaintext and in-order. It keeps track of the TLS records in the TCP stream. When a packet marked for offload is transmitted, the NIC encrypts the payload in-place and puts authentication tags in the relevant place holders. The responsibility for handling out-of-order packets (i.e. TCP retransmission, qdisc drops) falls on the netdev driver. The netdev driver keeps track of the expected TCP SN from the NIC's perspective. If the next packet to transmit matches the expected TCP SN, the driver advances the expected TCP SN, and transmits the packet with TLS offload indication. If the next packet to transmit does not match the expected TCP SN. The driver calls the TLS layer to obtain the TLS record that includes the TCP of the packet for transmission. Using this TLS record, the driver posts a work entry on the transmit queue to reconstruct the NIC TLS state required for the offload of the out-of-order packet. It updates the expected TCP SN accordingly and transmits the now in-order packet. The same queue is used for packet transmission and TLS context reconstruction to avoid the need for flushing the transmit queue before issuing the context reconstruction request. Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com> Signed-off-by: Boris Pismenny <borisp@mellanox.com> Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-30 15:16:16 +08:00
}
/* This function may be called after the user socket is already
* closed so make sure we don't use anything freed during
* tls_sk_proto_close here
*/
static int fill_sg_in(struct scatterlist *sg_in,
struct sk_buff *skb,
tls: Refactor tls_offload variable names For symmetry, we rename tls_offload_context to tls_offload_context_tx before we add tls_offload_context_rx. Signed-off-by: Boris Pismenny <borisp@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-13 19:33:39 +08:00
struct tls_offload_context_tx *ctx,
net/tls: Add generic NIC offload infrastructure This patch adds a generic infrastructure to offload TLS crypto to a network device. It enables the kernel TLS socket to skip encryption and authentication operations on the transmit side of the data path. Leaving those computationally expensive operations to the NIC. The NIC offload infrastructure builds TLS records and pushes them to the TCP layer just like the SW KTLS implementation and using the same API. TCP segmentation is mostly unaffected. Currently the only exception is that we prevent mixed SKBs where only part of the payload requires offload. In the future we are likely to add a similar restriction following a change cipher spec record. The notable differences between SW KTLS and NIC offloaded TLS implementations are as follows: 1. The offloaded implementation builds "plaintext TLS record", those records contain plaintext instead of ciphertext and place holder bytes instead of authentication tags. 2. The offloaded implementation maintains a mapping from TCP sequence number to TLS records. Thus given a TCP SKB sent from a NIC offloaded TLS socket, we can use the tls NIC offload infrastructure to obtain enough context to encrypt the payload of the SKB. A TLS record is released when the last byte of the record is ack'ed, this is done through the new icsk_clean_acked callback. The infrastructure should be extendable to support various NIC offload implementations. However it is currently written with the implementation below in mind: The NIC assumes that packets from each offloaded stream are sent as plaintext and in-order. It keeps track of the TLS records in the TCP stream. When a packet marked for offload is transmitted, the NIC encrypts the payload in-place and puts authentication tags in the relevant place holders. The responsibility for handling out-of-order packets (i.e. TCP retransmission, qdisc drops) falls on the netdev driver. The netdev driver keeps track of the expected TCP SN from the NIC's perspective. If the next packet to transmit matches the expected TCP SN, the driver advances the expected TCP SN, and transmits the packet with TLS offload indication. If the next packet to transmit does not match the expected TCP SN. The driver calls the TLS layer to obtain the TLS record that includes the TCP of the packet for transmission. Using this TLS record, the driver posts a work entry on the transmit queue to reconstruct the NIC TLS state required for the offload of the out-of-order packet. It updates the expected TCP SN accordingly and transmits the now in-order packet. The same queue is used for packet transmission and TLS context reconstruction to avoid the need for flushing the transmit queue before issuing the context reconstruction request. Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com> Signed-off-by: Boris Pismenny <borisp@mellanox.com> Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-30 15:16:16 +08:00
u64 *rcd_sn,
s32 *sync_size,
int *resync_sgs)
{
int tcp_payload_offset = skb_transport_offset(skb) + tcp_hdrlen(skb);
int payload_len = skb->len - tcp_payload_offset;
u32 tcp_seq = ntohl(tcp_hdr(skb)->seq);
struct tls_record_info *record;
unsigned long flags;
int remaining;
int i;
spin_lock_irqsave(&ctx->lock, flags);
record = tls_get_record(ctx, tcp_seq, rcd_sn);
if (!record) {
spin_unlock_irqrestore(&ctx->lock, flags);
return -EINVAL;
}
*sync_size = tcp_seq - tls_record_start_seq(record);
if (*sync_size < 0) {
int is_start_marker = tls_record_is_start_marker(record);
spin_unlock_irqrestore(&ctx->lock, flags);
/* This should only occur if the relevant record was
* already acked. In that case it should be ok
* to drop the packet and avoid retransmission.
*
* There is a corner case where the packet contains
* both an acked and a non-acked record.
* We currently don't handle that case and rely
* on TCP to retranmit a packet that doesn't contain
* already acked payload.
*/
if (!is_start_marker)
*sync_size = 0;
return -EINVAL;
}
remaining = *sync_size;
for (i = 0; remaining > 0; i++) {
skb_frag_t *frag = &record->frags[i];
__skb_frag_ref(frag);
sg_set_page(sg_in + i, skb_frag_page(frag),
skb_frag_size(frag), frag->page_offset);
remaining -= skb_frag_size(frag);
if (remaining < 0)
sg_in[i].length += remaining;
}
*resync_sgs = i;
spin_unlock_irqrestore(&ctx->lock, flags);
if (skb_to_sgvec(skb, &sg_in[i], tcp_payload_offset, payload_len) < 0)
return -EINVAL;
return 0;
}
static void fill_sg_out(struct scatterlist sg_out[3], void *buf,
struct tls_context *tls_ctx,
struct sk_buff *nskb,
int tcp_payload_offset,
int payload_len,
int sync_size,
void *dummy_buf)
{
sg_set_buf(&sg_out[0], dummy_buf, sync_size);
sg_set_buf(&sg_out[1], nskb->data + tcp_payload_offset, payload_len);
/* Add room for authentication tag produced by crypto */
dummy_buf += sync_size;
sg_set_buf(&sg_out[2], dummy_buf, TLS_CIPHER_AES_GCM_128_TAG_SIZE);
}
static struct sk_buff *tls_enc_skb(struct tls_context *tls_ctx,
struct scatterlist sg_out[3],
struct scatterlist *sg_in,
struct sk_buff *skb,
s32 sync_size, u64 rcd_sn)
{
int tcp_payload_offset = skb_transport_offset(skb) + tcp_hdrlen(skb);
tls: Refactor tls_offload variable names For symmetry, we rename tls_offload_context to tls_offload_context_tx before we add tls_offload_context_rx. Signed-off-by: Boris Pismenny <borisp@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-13 19:33:39 +08:00
struct tls_offload_context_tx *ctx = tls_offload_ctx_tx(tls_ctx);
net/tls: Add generic NIC offload infrastructure This patch adds a generic infrastructure to offload TLS crypto to a network device. It enables the kernel TLS socket to skip encryption and authentication operations on the transmit side of the data path. Leaving those computationally expensive operations to the NIC. The NIC offload infrastructure builds TLS records and pushes them to the TCP layer just like the SW KTLS implementation and using the same API. TCP segmentation is mostly unaffected. Currently the only exception is that we prevent mixed SKBs where only part of the payload requires offload. In the future we are likely to add a similar restriction following a change cipher spec record. The notable differences between SW KTLS and NIC offloaded TLS implementations are as follows: 1. The offloaded implementation builds "plaintext TLS record", those records contain plaintext instead of ciphertext and place holder bytes instead of authentication tags. 2. The offloaded implementation maintains a mapping from TCP sequence number to TLS records. Thus given a TCP SKB sent from a NIC offloaded TLS socket, we can use the tls NIC offload infrastructure to obtain enough context to encrypt the payload of the SKB. A TLS record is released when the last byte of the record is ack'ed, this is done through the new icsk_clean_acked callback. The infrastructure should be extendable to support various NIC offload implementations. However it is currently written with the implementation below in mind: The NIC assumes that packets from each offloaded stream are sent as plaintext and in-order. It keeps track of the TLS records in the TCP stream. When a packet marked for offload is transmitted, the NIC encrypts the payload in-place and puts authentication tags in the relevant place holders. The responsibility for handling out-of-order packets (i.e. TCP retransmission, qdisc drops) falls on the netdev driver. The netdev driver keeps track of the expected TCP SN from the NIC's perspective. If the next packet to transmit matches the expected TCP SN, the driver advances the expected TCP SN, and transmits the packet with TLS offload indication. If the next packet to transmit does not match the expected TCP SN. The driver calls the TLS layer to obtain the TLS record that includes the TCP of the packet for transmission. Using this TLS record, the driver posts a work entry on the transmit queue to reconstruct the NIC TLS state required for the offload of the out-of-order packet. It updates the expected TCP SN accordingly and transmits the now in-order packet. The same queue is used for packet transmission and TLS context reconstruction to avoid the need for flushing the transmit queue before issuing the context reconstruction request. Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com> Signed-off-by: Boris Pismenny <borisp@mellanox.com> Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-30 15:16:16 +08:00
int payload_len = skb->len - tcp_payload_offset;
void *buf, *iv, *aad, *dummy_buf;
struct aead_request *aead_req;
struct sk_buff *nskb = NULL;
int buf_len;
aead_req = tls_alloc_aead_request(ctx->aead_send, GFP_ATOMIC);
if (!aead_req)
return NULL;
buf_len = TLS_CIPHER_AES_GCM_128_SALT_SIZE +
TLS_CIPHER_AES_GCM_128_IV_SIZE +
TLS_AAD_SPACE_SIZE +
sync_size +
TLS_CIPHER_AES_GCM_128_TAG_SIZE;
buf = kmalloc(buf_len, GFP_ATOMIC);
if (!buf)
goto free_req;
iv = buf;
tls: zero the crypto information from tls_context before freeing This contains key material in crypto_send_aes_gcm_128 and crypto_recv_aes_gcm_128. Introduce union tls_crypto_context, and replace the two identical unions directly embedded in struct tls_context with it. We can then use this union to clean up the memory in the new tls_ctx_free() function. Fixes: 3c4d7559159b ("tls: kernel TLS support") Signed-off-by: Sabrina Dubroca <sd@queasysnail.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-12 23:44:42 +08:00
memcpy(iv, tls_ctx->crypto_send.aes_gcm_128.salt,
net/tls: Add generic NIC offload infrastructure This patch adds a generic infrastructure to offload TLS crypto to a network device. It enables the kernel TLS socket to skip encryption and authentication operations on the transmit side of the data path. Leaving those computationally expensive operations to the NIC. The NIC offload infrastructure builds TLS records and pushes them to the TCP layer just like the SW KTLS implementation and using the same API. TCP segmentation is mostly unaffected. Currently the only exception is that we prevent mixed SKBs where only part of the payload requires offload. In the future we are likely to add a similar restriction following a change cipher spec record. The notable differences between SW KTLS and NIC offloaded TLS implementations are as follows: 1. The offloaded implementation builds "plaintext TLS record", those records contain plaintext instead of ciphertext and place holder bytes instead of authentication tags. 2. The offloaded implementation maintains a mapping from TCP sequence number to TLS records. Thus given a TCP SKB sent from a NIC offloaded TLS socket, we can use the tls NIC offload infrastructure to obtain enough context to encrypt the payload of the SKB. A TLS record is released when the last byte of the record is ack'ed, this is done through the new icsk_clean_acked callback. The infrastructure should be extendable to support various NIC offload implementations. However it is currently written with the implementation below in mind: The NIC assumes that packets from each offloaded stream are sent as plaintext and in-order. It keeps track of the TLS records in the TCP stream. When a packet marked for offload is transmitted, the NIC encrypts the payload in-place and puts authentication tags in the relevant place holders. The responsibility for handling out-of-order packets (i.e. TCP retransmission, qdisc drops) falls on the netdev driver. The netdev driver keeps track of the expected TCP SN from the NIC's perspective. If the next packet to transmit matches the expected TCP SN, the driver advances the expected TCP SN, and transmits the packet with TLS offload indication. If the next packet to transmit does not match the expected TCP SN. The driver calls the TLS layer to obtain the TLS record that includes the TCP of the packet for transmission. Using this TLS record, the driver posts a work entry on the transmit queue to reconstruct the NIC TLS state required for the offload of the out-of-order packet. It updates the expected TCP SN accordingly and transmits the now in-order packet. The same queue is used for packet transmission and TLS context reconstruction to avoid the need for flushing the transmit queue before issuing the context reconstruction request. Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com> Signed-off-by: Boris Pismenny <borisp@mellanox.com> Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-30 15:16:16 +08:00
TLS_CIPHER_AES_GCM_128_SALT_SIZE);
aad = buf + TLS_CIPHER_AES_GCM_128_SALT_SIZE +
TLS_CIPHER_AES_GCM_128_IV_SIZE;
dummy_buf = aad + TLS_AAD_SPACE_SIZE;
nskb = alloc_skb(skb_headroom(skb) + skb->len, GFP_ATOMIC);
if (!nskb)
goto free_buf;
skb_reserve(nskb, skb_headroom(skb));
fill_sg_out(sg_out, buf, tls_ctx, nskb, tcp_payload_offset,
payload_len, sync_size, dummy_buf);
if (tls_enc_records(aead_req, ctx->aead_send, sg_in, sg_out, aad, iv,
rcd_sn, sync_size + payload_len) < 0)
goto free_nskb;
complete_skb(nskb, skb, tcp_payload_offset);
/* validate_xmit_skb_list assumes that if the skb wasn't segmented
* nskb->prev will point to the skb itself
*/
nskb->prev = nskb;
free_buf:
kfree(buf);
free_req:
kfree(aead_req);
return nskb;
free_nskb:
kfree_skb(nskb);
nskb = NULL;
goto free_buf;
}
static struct sk_buff *tls_sw_fallback(struct sock *sk, struct sk_buff *skb)
{
int tcp_payload_offset = skb_transport_offset(skb) + tcp_hdrlen(skb);
struct tls_context *tls_ctx = tls_get_ctx(sk);
tls: Refactor tls_offload variable names For symmetry, we rename tls_offload_context to tls_offload_context_tx before we add tls_offload_context_rx. Signed-off-by: Boris Pismenny <borisp@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-13 19:33:39 +08:00
struct tls_offload_context_tx *ctx = tls_offload_ctx_tx(tls_ctx);
net/tls: Add generic NIC offload infrastructure This patch adds a generic infrastructure to offload TLS crypto to a network device. It enables the kernel TLS socket to skip encryption and authentication operations on the transmit side of the data path. Leaving those computationally expensive operations to the NIC. The NIC offload infrastructure builds TLS records and pushes them to the TCP layer just like the SW KTLS implementation and using the same API. TCP segmentation is mostly unaffected. Currently the only exception is that we prevent mixed SKBs where only part of the payload requires offload. In the future we are likely to add a similar restriction following a change cipher spec record. The notable differences between SW KTLS and NIC offloaded TLS implementations are as follows: 1. The offloaded implementation builds "plaintext TLS record", those records contain plaintext instead of ciphertext and place holder bytes instead of authentication tags. 2. The offloaded implementation maintains a mapping from TCP sequence number to TLS records. Thus given a TCP SKB sent from a NIC offloaded TLS socket, we can use the tls NIC offload infrastructure to obtain enough context to encrypt the payload of the SKB. A TLS record is released when the last byte of the record is ack'ed, this is done through the new icsk_clean_acked callback. The infrastructure should be extendable to support various NIC offload implementations. However it is currently written with the implementation below in mind: The NIC assumes that packets from each offloaded stream are sent as plaintext and in-order. It keeps track of the TLS records in the TCP stream. When a packet marked for offload is transmitted, the NIC encrypts the payload in-place and puts authentication tags in the relevant place holders. The responsibility for handling out-of-order packets (i.e. TCP retransmission, qdisc drops) falls on the netdev driver. The netdev driver keeps track of the expected TCP SN from the NIC's perspective. If the next packet to transmit matches the expected TCP SN, the driver advances the expected TCP SN, and transmits the packet with TLS offload indication. If the next packet to transmit does not match the expected TCP SN. The driver calls the TLS layer to obtain the TLS record that includes the TCP of the packet for transmission. Using this TLS record, the driver posts a work entry on the transmit queue to reconstruct the NIC TLS state required for the offload of the out-of-order packet. It updates the expected TCP SN accordingly and transmits the now in-order packet. The same queue is used for packet transmission and TLS context reconstruction to avoid the need for flushing the transmit queue before issuing the context reconstruction request. Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com> Signed-off-by: Boris Pismenny <borisp@mellanox.com> Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-30 15:16:16 +08:00
int payload_len = skb->len - tcp_payload_offset;
struct scatterlist *sg_in, sg_out[3];
struct sk_buff *nskb = NULL;
int sg_in_max_elements;
int resync_sgs = 0;
s32 sync_size = 0;
u64 rcd_sn;
/* worst case is:
* MAX_SKB_FRAGS in tls_record_info
* MAX_SKB_FRAGS + 1 in SKB head and frags.
*/
sg_in_max_elements = 2 * MAX_SKB_FRAGS + 1;
if (!payload_len)
return skb;
sg_in = kmalloc_array(sg_in_max_elements, sizeof(*sg_in), GFP_ATOMIC);
if (!sg_in)
goto free_orig;
sg_init_table(sg_in, sg_in_max_elements);
sg_init_table(sg_out, ARRAY_SIZE(sg_out));
if (fill_sg_in(sg_in, skb, ctx, &rcd_sn, &sync_size, &resync_sgs)) {
/* bypass packets before kernel TLS socket option was set */
if (sync_size < 0 && payload_len <= -sync_size)
nskb = skb_get(skb);
goto put_sg;
}
nskb = tls_enc_skb(tls_ctx, sg_out, sg_in, skb, sync_size, rcd_sn);
put_sg:
while (resync_sgs)
put_page(sg_page(&sg_in[--resync_sgs]));
kfree(sg_in);
free_orig:
net/tls: remove false positive warning It's possible that TCP stack will decide to retransmit a packet right when that packet's data gets acked, especially in presence of packet reordering. This means that packets may be in flight, even though tls_device code has already freed their record state. Make fill_sg_in() and in turn tls_sw_fallback() not generate a warning in that case, and quietly proceed to drop such frames. Make the exit path from tls_sw_fallback() drop monitor friendly, for users to be able to troubleshoot dropped retransmissions. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-06-04 06:17:00 +08:00
if (nskb)
consume_skb(skb);
else
kfree_skb(skb);
net/tls: Add generic NIC offload infrastructure This patch adds a generic infrastructure to offload TLS crypto to a network device. It enables the kernel TLS socket to skip encryption and authentication operations on the transmit side of the data path. Leaving those computationally expensive operations to the NIC. The NIC offload infrastructure builds TLS records and pushes them to the TCP layer just like the SW KTLS implementation and using the same API. TCP segmentation is mostly unaffected. Currently the only exception is that we prevent mixed SKBs where only part of the payload requires offload. In the future we are likely to add a similar restriction following a change cipher spec record. The notable differences between SW KTLS and NIC offloaded TLS implementations are as follows: 1. The offloaded implementation builds "plaintext TLS record", those records contain plaintext instead of ciphertext and place holder bytes instead of authentication tags. 2. The offloaded implementation maintains a mapping from TCP sequence number to TLS records. Thus given a TCP SKB sent from a NIC offloaded TLS socket, we can use the tls NIC offload infrastructure to obtain enough context to encrypt the payload of the SKB. A TLS record is released when the last byte of the record is ack'ed, this is done through the new icsk_clean_acked callback. The infrastructure should be extendable to support various NIC offload implementations. However it is currently written with the implementation below in mind: The NIC assumes that packets from each offloaded stream are sent as plaintext and in-order. It keeps track of the TLS records in the TCP stream. When a packet marked for offload is transmitted, the NIC encrypts the payload in-place and puts authentication tags in the relevant place holders. The responsibility for handling out-of-order packets (i.e. TCP retransmission, qdisc drops) falls on the netdev driver. The netdev driver keeps track of the expected TCP SN from the NIC's perspective. If the next packet to transmit matches the expected TCP SN, the driver advances the expected TCP SN, and transmits the packet with TLS offload indication. If the next packet to transmit does not match the expected TCP SN. The driver calls the TLS layer to obtain the TLS record that includes the TCP of the packet for transmission. Using this TLS record, the driver posts a work entry on the transmit queue to reconstruct the NIC TLS state required for the offload of the out-of-order packet. It updates the expected TCP SN accordingly and transmits the now in-order packet. The same queue is used for packet transmission and TLS context reconstruction to avoid the need for flushing the transmit queue before issuing the context reconstruction request. Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com> Signed-off-by: Boris Pismenny <borisp@mellanox.com> Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-30 15:16:16 +08:00
return nskb;
}
struct sk_buff *tls_validate_xmit_skb(struct sock *sk,
struct net_device *dev,
struct sk_buff *skb)
{
if (dev == tls_get_ctx(sk)->netdev)
return skb;
return tls_sw_fallback(sk, skb);
}
tls: Add rx inline crypto offload This patch completes the generic infrastructure to offload TLS crypto to a network device. It enables the kernel to skip decryption and authentication of some skbs marked as decrypted by the NIC. In the fast path, all packets received are decrypted by the NIC and the performance is comparable to plain TCP. This infrastructure doesn't require a TCP offload engine. Instead, the NIC only decrypts packets that contain the expected TCP sequence number. Out-Of-Order TCP packets are provided unmodified. As a result, at the worst case a received TLS record consists of both plaintext and ciphertext packets. These partially decrypted records must be reencrypted, only to be decrypted. The notable differences between SW KTLS Rx and this offload are as follows: 1. Partial decryption - Software must handle the case of a TLS record that was only partially decrypted by HW. This can happen due to packet reordering. 2. Resynchronization - tls_read_size calls the device driver to resynchronize HW after HW lost track of TLS record framing in the TCP stream. Signed-off-by: Boris Pismenny <borisp@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-13 19:33:43 +08:00
EXPORT_SYMBOL_GPL(tls_validate_xmit_skb);
net/tls: Add generic NIC offload infrastructure This patch adds a generic infrastructure to offload TLS crypto to a network device. It enables the kernel TLS socket to skip encryption and authentication operations on the transmit side of the data path. Leaving those computationally expensive operations to the NIC. The NIC offload infrastructure builds TLS records and pushes them to the TCP layer just like the SW KTLS implementation and using the same API. TCP segmentation is mostly unaffected. Currently the only exception is that we prevent mixed SKBs where only part of the payload requires offload. In the future we are likely to add a similar restriction following a change cipher spec record. The notable differences between SW KTLS and NIC offloaded TLS implementations are as follows: 1. The offloaded implementation builds "plaintext TLS record", those records contain plaintext instead of ciphertext and place holder bytes instead of authentication tags. 2. The offloaded implementation maintains a mapping from TCP sequence number to TLS records. Thus given a TCP SKB sent from a NIC offloaded TLS socket, we can use the tls NIC offload infrastructure to obtain enough context to encrypt the payload of the SKB. A TLS record is released when the last byte of the record is ack'ed, this is done through the new icsk_clean_acked callback. The infrastructure should be extendable to support various NIC offload implementations. However it is currently written with the implementation below in mind: The NIC assumes that packets from each offloaded stream are sent as plaintext and in-order. It keeps track of the TLS records in the TCP stream. When a packet marked for offload is transmitted, the NIC encrypts the payload in-place and puts authentication tags in the relevant place holders. The responsibility for handling out-of-order packets (i.e. TCP retransmission, qdisc drops) falls on the netdev driver. The netdev driver keeps track of the expected TCP SN from the NIC's perspective. If the next packet to transmit matches the expected TCP SN, the driver advances the expected TCP SN, and transmits the packet with TLS offload indication. If the next packet to transmit does not match the expected TCP SN. The driver calls the TLS layer to obtain the TLS record that includes the TCP of the packet for transmission. Using this TLS record, the driver posts a work entry on the transmit queue to reconstruct the NIC TLS state required for the offload of the out-of-order packet. It updates the expected TCP SN accordingly and transmits the now in-order packet. The same queue is used for packet transmission and TLS context reconstruction to avoid the need for flushing the transmit queue before issuing the context reconstruction request. Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com> Signed-off-by: Boris Pismenny <borisp@mellanox.com> Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-30 15:16:16 +08:00
net/tls: export TLS per skb encryption While offloading TLS connections, drivers need to handle the case where out of order packets need to be transmitted. Other drivers obtain the entire TLS record for the specific skb to provide as context to hardware for encryption. However, other designs may also want to keep the hardware state intact and perform the out of order encryption entirely on the host. To achieve this, export the already existing software encryption fallback path so drivers could access this. Signed-off-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-06-06 05:11:40 +08:00
struct sk_buff *tls_encrypt_skb(struct sk_buff *skb)
{
return tls_sw_fallback(skb->sk, skb);
}
EXPORT_SYMBOL_GPL(tls_encrypt_skb);
net/tls: Add generic NIC offload infrastructure This patch adds a generic infrastructure to offload TLS crypto to a network device. It enables the kernel TLS socket to skip encryption and authentication operations on the transmit side of the data path. Leaving those computationally expensive operations to the NIC. The NIC offload infrastructure builds TLS records and pushes them to the TCP layer just like the SW KTLS implementation and using the same API. TCP segmentation is mostly unaffected. Currently the only exception is that we prevent mixed SKBs where only part of the payload requires offload. In the future we are likely to add a similar restriction following a change cipher spec record. The notable differences between SW KTLS and NIC offloaded TLS implementations are as follows: 1. The offloaded implementation builds "plaintext TLS record", those records contain plaintext instead of ciphertext and place holder bytes instead of authentication tags. 2. The offloaded implementation maintains a mapping from TCP sequence number to TLS records. Thus given a TCP SKB sent from a NIC offloaded TLS socket, we can use the tls NIC offload infrastructure to obtain enough context to encrypt the payload of the SKB. A TLS record is released when the last byte of the record is ack'ed, this is done through the new icsk_clean_acked callback. The infrastructure should be extendable to support various NIC offload implementations. However it is currently written with the implementation below in mind: The NIC assumes that packets from each offloaded stream are sent as plaintext and in-order. It keeps track of the TLS records in the TCP stream. When a packet marked for offload is transmitted, the NIC encrypts the payload in-place and puts authentication tags in the relevant place holders. The responsibility for handling out-of-order packets (i.e. TCP retransmission, qdisc drops) falls on the netdev driver. The netdev driver keeps track of the expected TCP SN from the NIC's perspective. If the next packet to transmit matches the expected TCP SN, the driver advances the expected TCP SN, and transmits the packet with TLS offload indication. If the next packet to transmit does not match the expected TCP SN. The driver calls the TLS layer to obtain the TLS record that includes the TCP of the packet for transmission. Using this TLS record, the driver posts a work entry on the transmit queue to reconstruct the NIC TLS state required for the offload of the out-of-order packet. It updates the expected TCP SN accordingly and transmits the now in-order packet. The same queue is used for packet transmission and TLS context reconstruction to avoid the need for flushing the transmit queue before issuing the context reconstruction request. Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com> Signed-off-by: Boris Pismenny <borisp@mellanox.com> Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-30 15:16:16 +08:00
int tls_sw_fallback_init(struct sock *sk,
tls: Refactor tls_offload variable names For symmetry, we rename tls_offload_context to tls_offload_context_tx before we add tls_offload_context_rx. Signed-off-by: Boris Pismenny <borisp@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-13 19:33:39 +08:00
struct tls_offload_context_tx *offload_ctx,
net/tls: Add generic NIC offload infrastructure This patch adds a generic infrastructure to offload TLS crypto to a network device. It enables the kernel TLS socket to skip encryption and authentication operations on the transmit side of the data path. Leaving those computationally expensive operations to the NIC. The NIC offload infrastructure builds TLS records and pushes them to the TCP layer just like the SW KTLS implementation and using the same API. TCP segmentation is mostly unaffected. Currently the only exception is that we prevent mixed SKBs where only part of the payload requires offload. In the future we are likely to add a similar restriction following a change cipher spec record. The notable differences between SW KTLS and NIC offloaded TLS implementations are as follows: 1. The offloaded implementation builds "plaintext TLS record", those records contain plaintext instead of ciphertext and place holder bytes instead of authentication tags. 2. The offloaded implementation maintains a mapping from TCP sequence number to TLS records. Thus given a TCP SKB sent from a NIC offloaded TLS socket, we can use the tls NIC offload infrastructure to obtain enough context to encrypt the payload of the SKB. A TLS record is released when the last byte of the record is ack'ed, this is done through the new icsk_clean_acked callback. The infrastructure should be extendable to support various NIC offload implementations. However it is currently written with the implementation below in mind: The NIC assumes that packets from each offloaded stream are sent as plaintext and in-order. It keeps track of the TLS records in the TCP stream. When a packet marked for offload is transmitted, the NIC encrypts the payload in-place and puts authentication tags in the relevant place holders. The responsibility for handling out-of-order packets (i.e. TCP retransmission, qdisc drops) falls on the netdev driver. The netdev driver keeps track of the expected TCP SN from the NIC's perspective. If the next packet to transmit matches the expected TCP SN, the driver advances the expected TCP SN, and transmits the packet with TLS offload indication. If the next packet to transmit does not match the expected TCP SN. The driver calls the TLS layer to obtain the TLS record that includes the TCP of the packet for transmission. Using this TLS record, the driver posts a work entry on the transmit queue to reconstruct the NIC TLS state required for the offload of the out-of-order packet. It updates the expected TCP SN accordingly and transmits the now in-order packet. The same queue is used for packet transmission and TLS context reconstruction to avoid the need for flushing the transmit queue before issuing the context reconstruction request. Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com> Signed-off-by: Boris Pismenny <borisp@mellanox.com> Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-30 15:16:16 +08:00
struct tls_crypto_info *crypto_info)
{
const u8 *key;
int rc;
offload_ctx->aead_send =
crypto_alloc_aead("gcm(aes)", 0, CRYPTO_ALG_ASYNC);
if (IS_ERR(offload_ctx->aead_send)) {
rc = PTR_ERR(offload_ctx->aead_send);
pr_err_ratelimited("crypto_alloc_aead failed rc=%d\n", rc);
offload_ctx->aead_send = NULL;
goto err_out;
}
key = ((struct tls12_crypto_info_aes_gcm_128 *)crypto_info)->key;
rc = crypto_aead_setkey(offload_ctx->aead_send, key,
TLS_CIPHER_AES_GCM_128_KEY_SIZE);
if (rc)
goto free_aead;
rc = crypto_aead_setauthsize(offload_ctx->aead_send,
TLS_CIPHER_AES_GCM_128_TAG_SIZE);
if (rc)
goto free_aead;
return 0;
free_aead:
crypto_free_aead(offload_ctx->aead_send);
err_out:
return rc;
}