Building with some experimental patches, I came across a warning
in the tls code:
include/linux/compiler.h:215:30: warning: assignment discards 'const' qualifier from pointer target type [-Wdiscarded-qualifiers]
215 | *(volatile typeof(x) *)&(x) = (val); \
| ^
net/tls/tls_main.c:650:4: note: in expansion of macro 'smp_store_release'
650 | smp_store_release(&saved_tcpv4_prot, prot);
This appears to be a legitimate warning about assigning a const pointer
into the non-const 'saved_tcpv4_prot' global. Annotate both the ipv4 and
ipv6 pointers 'const' to make the code internally consistent.
Fixes: 5bb4c45d46 ("net/tls: Read sk_prot once when building tls proto ops")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
sockmap performs lockless writes to sk->sk_prot on the following paths:
tcp_bpf_{recvmsg|sendmsg} / sock_map_unref
sk_psock_put
sk_psock_drop
sk_psock_restore_proto
WRITE_ONCE(sk->sk_prot, proto)
To prevent load/store tearing [1], and to make tooling aware of intentional
shared access [2], we need to annotate other sites that access sk_prot with
READ_ONCE/WRITE_ONCE macros.
Change done with Coccinelle with following semantic patch:
@@
expression E;
identifier I;
struct sock *sk;
identifier sk_prot =~ "^sk_prot$";
@@
(
E =
-sk->sk_prot
+READ_ONCE(sk->sk_prot)
|
-sk->sk_prot = E
+WRITE_ONCE(sk->sk_prot, E)
|
-sk->sk_prot
+READ_ONCE(sk->sk_prot)
->I
)
Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Apart from being a "tremendous" win when it comes to generated machine
code (see bloat-o-meter output for x86-64 below) this mainly prepares
ground for annotating access to sk_prot with READ_ONCE, so that we don't
pepper the code with access annotations and needlessly repeat loads.
add/remove: 0/0 grow/shrink: 0/1 up/down: 0/-46 (-46)
Function old new delta
tls_init 851 805 -46
Total: Before=21063, After=21017, chg -0.22%
Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The helper that builds kTLS proto ops doesn't need to and should not modify
the base proto ops. Annotate the parameter as read-only.
Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann says:
====================
pull-request: bpf-next 2020-02-21
The following pull-request contains BPF updates for your *net-next* tree.
We've added 25 non-merge commits during the last 4 day(s) which contain
a total of 33 files changed, 2433 insertions(+), 161 deletions(-).
The main changes are:
1) Allow for adding TCP listen sockets into sock_map/hash so they can be used
with reuseport BPF programs, from Jakub Sitnicki.
2) Add a new bpf_program__set_attach_target() helper for adding libbpf support
to specify the tracepoint/function dynamically, from Eelco Chaudron.
3) Add bpf_read_branch_records() BPF helper which helps use cases like profile
guided optimizations, from Daniel Xu.
4) Enable bpf_perf_event_read_value() in all tracing programs, from Song Liu.
5) Relax BTF mandatory check if only used for libbpf itself e.g. to process
BTF defined maps, from Andrii Nakryiko.
6) Move BPF selftests -mcpu compilation attribute from 'probe' to 'v3' as it has
been observed that former fails in envs with low memlock, from Yonghong Song.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
sk_msg and ULP frameworks override protocol callbacks pointer in
sk->sk_prot, while tcp accesses it locklessly when cloning the listening
socket, that is with neither sk_lock nor sk_callback_lock held.
Once we enable use of listening sockets with sockmap (and hence sk_msg),
there will be shared access to sk->sk_prot if socket is getting cloned
while being inserted/deleted to/from the sockmap from another CPU:
Read side:
tcp_v4_rcv
sk = __inet_lookup_skb(...)
tcp_check_req(sk)
inet_csk(sk)->icsk_af_ops->syn_recv_sock
tcp_v4_syn_recv_sock
tcp_create_openreq_child
inet_csk_clone_lock
sk_clone_lock
READ_ONCE(sk->sk_prot)
Write side:
sock_map_ops->map_update_elem
sock_map_update_elem
sock_map_update_common
sock_map_link_no_progs
tcp_bpf_init
tcp_bpf_update_sk_prot
sk_psock_update_proto
WRITE_ONCE(sk->sk_prot, ops)
sock_map_ops->map_delete_elem
sock_map_delete_elem
__sock_map_delete
sock_map_unref
sk_psock_put
sk_psock_drop
sk_psock_restore_proto
tcp_update_ulp
WRITE_ONCE(sk->sk_prot, proto)
Mark the shared access with READ_ONCE/WRITE_ONCE annotations.
Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20200218171023.844439-2-jakub@cloudflare.com
Current code doesn't check if tcp sequence number is starting from (/after)
1st record's start sequnce number. It only checks if seq number is before
1st record's end sequnce number. This problem will always be a possibility
in re-transmit case. If a record which belongs to a requested seq number is
already deleted, tls_get_record will start looking into list and as per the
check it will look if seq number is before the end seq of 1st record, which
will always be true and will return 1st record always, it should in fact
return NULL.
As part of the fix, start looking each record only if the sequence number
lies in the list else return NULL.
There is one more check added, driver look for the start marker record to
handle tcp packets which are before the tls offload start sequence number,
hence return 1st record if the record is tls start marker and seq number is
before the 1st record's starting sequence number.
Fixes: e8f6979981 ("net/tls: Add generic NIC offload infrastructure")
Signed-off-by: Rohit Maheshwari <rohitm@chelsio.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann says:
====================
pull-request: bpf 2020-01-15
The following pull-request contains BPF updates for your *net* tree.
We've added 12 non-merge commits during the last 9 day(s) which contain
a total of 13 files changed, 95 insertions(+), 43 deletions(-).
The main changes are:
1) Fix refcount leak for TCP time wait and request sockets for socket lookup
related BPF helpers, from Lorenz Bauer.
2) Fix wrong verification of ARSH instruction under ALU32, from Daniel Borkmann.
3) Batch of several sockmap and related TLS fixes found while operating
more complex BPF programs with Cilium and OpenSSL, from John Fastabend.
4) Fix sockmap to read psock's ingress_msg queue before regular sk_receive_queue()
to avoid purging data upon teardown, from Lingpeng Chen.
5) Fix printing incorrect pointer in bpftool's btf_dump_ptr() in order to properly
dump a BPF map's value with BTF, from Martin KaFai Lau.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
When user returns SK_DROP we need to reset the number of copied bytes
to indicate to the user the bytes were dropped and not sent. If we
don't reset the copied arg sendmsg will return as if those bytes were
copied giving the user a positive return value.
This works as expected today except in the case where the user also
pops bytes. In the pop case the sg.size is reduced but we don't correctly
account for this when copied bytes is reset. The popped bytes are not
accounted for and we return a small positive value potentially confusing
the user.
The reason this happens is due to a typo where we do the wrong comparison
when accounting for pop bytes. In this fix notice the if/else is not
needed and that we have a similar problem if we push data except its not
visible to the user because if delta is larger the sg.size we return a
negative value so it appears as an error regardless.
Fixes: 7246d8ed4d ("bpf: helper to pop data from messages")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Jonathan Lemon <jonathan.lemon@gmail.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/bpf/20200111061206.8028-9-john.fastabend@gmail.com
Its possible through a set of push, pop, apply helper calls to construct
a skmsg, which is just a ring of scatterlist elements, with the start
value larger than the end value. For example,
end start
|_0_|_1_| ... |_n_|_n+1_|
Where end points at 1 and start points and n so that valid elements is
the set {n, n+1, 0, 1}.
Currently, because we don't build the correct chain only {n, n+1} will
be sent. This adds a check and sg_chain call to correctly submit the
above to the crypto and tls send path.
Fixes: d3b18ad31f ("tls: add bpf support to sk_msg handling")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Jonathan Lemon <jonathan.lemon@gmail.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/bpf/20200111061206.8028-8-john.fastabend@gmail.com
It is possible to build a plaintext buffer using push helper that is larger
than the allocated encrypt buffer. When this record is pushed to crypto
layers this can result in a NULL pointer dereference because the crypto
API expects the encrypt buffer is large enough to fit the plaintext
buffer. Kernel splat below.
To resolve catch the cases this can happen and split the buffer into two
records to send individually. Unfortunately, there is still one case to
handle where the split creates a zero sized buffer. In this case we merge
the buffers and unmark the split. This happens when apply is zero and user
pushed data beyond encrypt buffer. This fixes the original case as well
because the split allocated an encrypt buffer larger than the plaintext
buffer and the merge simply moves the pointers around so we now have
a reference to the new (larger) encrypt buffer.
Perhaps its not ideal but it seems the best solution for a fixes branch
and avoids handling these two cases, (a) apply that needs split and (b)
non apply case. The are edge cases anyways so optimizing them seems not
necessary unless someone wants later in next branches.
[ 306.719107] BUG: kernel NULL pointer dereference, address: 0000000000000008
[...]
[ 306.747260] RIP: 0010:scatterwalk_copychunks+0x12f/0x1b0
[...]
[ 306.770350] Call Trace:
[ 306.770956] scatterwalk_map_and_copy+0x6c/0x80
[ 306.772026] gcm_enc_copy_hash+0x4b/0x50
[ 306.772925] gcm_hash_crypt_remain_continue+0xef/0x110
[ 306.774138] gcm_hash_crypt_continue+0xa1/0xb0
[ 306.775103] ? gcm_hash_crypt_continue+0xa1/0xb0
[ 306.776103] gcm_hash_assoc_remain_continue+0x94/0xa0
[ 306.777170] gcm_hash_assoc_continue+0x9d/0xb0
[ 306.778239] gcm_hash_init_continue+0x8f/0xa0
[ 306.779121] gcm_hash+0x73/0x80
[ 306.779762] gcm_encrypt_continue+0x6d/0x80
[ 306.780582] crypto_gcm_encrypt+0xcb/0xe0
[ 306.781474] crypto_aead_encrypt+0x1f/0x30
[ 306.782353] tls_push_record+0x3b9/0xb20 [tls]
[ 306.783314] ? sk_psock_msg_verdict+0x199/0x300
[ 306.784287] bpf_exec_tx_verdict+0x3f2/0x680 [tls]
[ 306.785357] tls_sw_sendmsg+0x4a3/0x6a0 [tls]
test_sockmap test signature to trigger bug,
[TEST]: (1, 1, 1, sendmsg, pass,redir,start 1,end 2,pop (1,2),ktls,):
Fixes: d3b18ad31f ("tls: add bpf support to sk_msg handling")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Jonathan Lemon <jonathan.lemon@gmail.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/bpf/20200111061206.8028-7-john.fastabend@gmail.com
When sockmap sock with TLS enabled is removed we cleanup bpf/psock state
and call tcp_update_ulp() to push updates to TLS ULP on top. However, we
don't push the write_space callback up and instead simply overwrite the
op with the psock stored previous op. This may or may not be correct so
to ensure we don't overwrite the TLS write space hook pass this field to
the ULP and have it fixup the ctx.
This completes a previous fix that pushed the ops through to the ULP
but at the time missed doing this for write_space, presumably because
write_space TLS hook was added around the same time.
Fixes: 95fa145479 ("bpf: sockmap/tls, close can race with map free")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Jakub Sitnicki <jakub@cloudflare.com>
Acked-by: Jonathan Lemon <jonathan.lemon@gmail.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/bpf/20200111061206.8028-4-john.fastabend@gmail.com
Mallesham reports the TLS with async accelerator was broken by
commit d10523d0b3 ("net/tls: free the record on encryption error")
because encryption can return -EINPROGRESS in such setups, which
should not be treated as an error.
The error is also present in the BPF path (likely copied from there).
Reported-by: Mallesham Jatharakonda <mallesham.jatharakonda@oneconvergence.com>
Fixes: d3b18ad31f ("tls: add bpf support to sk_msg handling")
Fixes: d10523d0b3 ("net/tls: free the record on encryption error")
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>
When device loses sync mid way through a record - kernel
has to re-encrypt the part of the record which the device
already decrypted to be able to decrypt and authenticate
the record in its entirety.
The re-encryption piggy backs on the decryption routine,
but obviously because the partially decrypted record can't
be authenticated crypto API returns an error which is then
ignored by tls_device_reencrypt().
Commit 5c5ec66858 ("net/tls: add TlsDecryptError stat")
added a statistic to count decryption errors, this statistic
can't be incremented when we see the expected re-encryption
error. Move the inc to the caller.
Reported-and-tested-by: David Beckett <david.beckett@netronome.com>
Fixes: 5c5ec66858 ("net/tls: add TlsDecryptError stat")
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>
There is currently no way for driver to reliably check that
the socket it has looked up is in fact RX offloaded. Add
a helper. This allows drivers to catch misbehaving firmware.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
ENOTSUPP is not available in userspace, for example:
setsockopt failed, 524, Unknown error 524
Signed-off-by: Valentin Vidic <vvidic@valentin-vidic.from.hr>
Acked-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Partially sent record cleanup path increments an SG entry
directly instead of using sg_next(). This should not be a
problem today, as encrypted messages should be always
allocated as arrays. But given this is a cleanup path it's
easy to miss was this ever to change. Use sg_next(), and
simplify the code.
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>
Looks like when BPF support was added by commit d3b18ad31f
("tls: add bpf support to sk_msg handling") and
commit d829e9c411 ("tls: convert to generic sk_msg interface")
it broke/removed the support for in-place crypto as added by
commit 4e6d47206c ("tls: Add support for inplace records
encryption").
The inplace_crypto member of struct tls_rec is dead, inited
to zero, and sometimes set to zero again. It used to be
set to 1 when record was allocated, but the skmsg code doesn't
seem to have been written with the idea of in-place crypto
in mind.
Since non trivial effort is required to bring the feature back
and we don't really have the HW to measure the benefit just
remove the left over support for now to avoid confusing readers.
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>
When tls_do_encryption() fails the SG lists are left with the
SG_END and SG_CHAIN marks in place. One could hope that once
encryption fails we will never see the record again, but that
is in fact not true. Commit d3b18ad31f ("tls: add bpf support
to sk_msg handling") added special handling to ENOMEM and ENOSPC
errors which mean we may see the same record re-submitted.
As suggested by John free the record, the BPF code is already
doing just that.
Reported-by: syzbot+df0d4ec12332661dd1f9@syzkaller.appspotmail.com
Fixes: d3b18ad31f ("tls: add bpf support to sk_msg handling")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
bpf_exec_tx_verdict() may free the record if tls_push_record()
fails, or if the entire record got consumed by BPF. Re-check
ctx->open_rec before touching the data.
Fixes: d3b18ad31f ("tls: add bpf support to sk_msg handling")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Minor conflict in drivers/s390/net/qeth_l2_main.c, kept the lock
from commit c8183f5489 ("s390/qeth: fix potential deadlock on
workqueue flush"), removed the code which was removed by commit
9897d583b0 ("s390/qeth: consolidate some duplicated HW cmd code").
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Bring back tls_sw_sendpage_locked. sk_msg redirection into a socket
with TLS_TX takes the following path:
tcp_bpf_sendmsg_redir
tcp_bpf_push_locked
tcp_bpf_push
kernel_sendpage_locked
sock->ops->sendpage_locked
Also update the flags test in tls_sw_sendpage_locked to allow flag
MSG_NO_SHARED_FRAGS. bpf_tcp_sendmsg sets this.
Link: https://lore.kernel.org/netdev/CA+FuTSdaAawmZ2N8nfDDKu3XLpXBbMtcCT0q4FntDD2gn8ASUw@mail.gmail.com/T/#t
Link: https://github.com/wdebruij/kerneltools/commits/icept.2
Fixes: 0608c69c9a ("bpf: sk_msg, sock{map|hash} redirect through ULP")
Fixes: f3de19af0f ("Revert \"net/tls: remove unused function tls_sw_sendpage_locked\"")
Signed-off-by: Willem de Bruijn <willemb@google.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If PROC_FS is not set, gcc warning this:
net/tls/tls_proc.c:23:12: warning:
'tls_statistics_seq_show' defined but not used [-Wunused-function]
Use #ifdef to guard this.
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Acked-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
One conflict in the BPF samples Makefile, some fixes in 'net' whilst
we were converting over to Makefile.target rules in 'net-next'.
Signed-off-by: David S. Miller <davem@davemloft.net>
TLS TX needs to release and re-acquire the socket lock if send buffer
fills up.
TLS SW TX path currently depends on only allowing one thread to enter
the function by the abuse of sk_write_pending. If another writer is
already waiting for memory no new ones are allowed in.
This has two problems:
- writers don't wake other threads up when they leave the kernel;
meaning that this scheme works for single extra thread (second
application thread or delayed work) because memory becoming
available will send a wake up request, but as Mallesham and
Pooja report with larger number of threads it leads to threads
being put to sleep indefinitely;
- the delayed work does not get _scheduled_ but it may _run_ when
other writers are present leading to crashes as writers don't
expect state to change under their feet (same records get pushed
and freed multiple times); it's hard to reliably bail from the
work, however, because the mere presence of a writer does not
guarantee that the writer will push pending records before exiting.
Ensuring wakeups always happen will make the code basically open
code a mutex. Just use a mutex.
The TLS HW TX path does not have any locking (not even the
sk_write_pending hack), yet it uses a per-socket sg_tx_data
array to push records.
Fixes: a42055e8d2 ("net/tls: Add support for async encryption of records for performance")
Reported-by: Mallesham Jatharakonda <mallesh537@gmail.com>
Reported-by: Pooja Trivedi <poojatrivedi@gmail.com>
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>
sk_write_pending being not zero does not guarantee that partial
record will be pushed. If the thread waiting for memory times out
the pending record may get stuck.
In case of tls_device there is no path where parial record is
set and writer present in the first place. Partial record is
set only in tls_push_sg() and tls_push_sg() will return an
error immediately. All tls_device callers of tls_push_sg()
will return (and not wait for memory) if it failed.
Fixes: a42055e8d2 ("net/tls: Add support for async encryption of records for performance")
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>
Use a single bit instead of boolean to remember if packet
was already decrypted.
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>
Store async_capable on a single bit instead of a full integer
to save space.
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>
Avoid unnecessary pointer chasing and calculations, callers already
have most of the state tls_device_decrypted() needs.
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>
Make sure GCC realizes it's unlikely that allocations will fail.
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>
Tell GCC sk->err is not likely to be set.
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>
Add a statistic for number of RX resyncs sent down to the NIC.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a statistic for TLS record decryption errors.
Since devices are supposed to pass records as-is when they
encounter errors this statistic will count bad records in
both pure software and inline crypto configurations.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add SNMP stats for number of sockets with successfully
installed sessions. Break them down to software and
hardware ones. Note that if hardware offload fails
stack uses software implementation, and counts the
session appropriately.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a skeleton structure for adding TLS statistics.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a tracepoint to the TLS offload's fast path. This tracepoint
can be used to track the decrypted and encrypted status of received
records. Records decrypted by the device should have decrypted set
to 1, records which have neither decrypted nor decrypted set are
partially decrypted, require re-encryption and therefore are most
expensive to deal with.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add tracing of device-related interaction to aid performance
analysis, especially around resync:
tls:tls_device_offload_set
tls:tls_device_rx_resync_send
tls:tls_device_rx_resync_nh_schedule
tls:tls_device_rx_resync_nh_delay
tls:tls_device_tx_resync_req
tls:tls_device_tx_resync_send
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
TLS "record layer offload" requires TOE, and bypasses most of
the normal networking stack. It is also significantly less
maintained. Allow users to compile it out to avoid issues.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The tls_hw_* functions are quite confusingly named, since they
are related to the TOE-offload, not TLS_HW offload which doesn't
require TOE. Rename them.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Move tls_hw_* functions to a new, separate source file
to avoid confusion with normal, non-TOE offload.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Move tls_build_proto() so that TOE offload doesn't have to call it
mid way through its bypass enable path.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Rename struct tls_device to struct tls_toe_device to avoid
confusion with normal, non-TOE offload.
No functional changes.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Move tls_device structure and register/unregister functions
to a new header to avoid confusion with normal, non-TOE offload.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Unlike normal TCP code TLS has to touch the cache lines
it copies into to fill header info. On memory-heavy workloads
having non temporal stores and normal accesses targeting
the same cache line leads to significant overhead.
Measured 3% overhead running 3600 round robin connections
with additional memory heavy workload.
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>
For TLS device offload the tag/message authentication code are
filled in by the device. The kernel merely reserves space for
them. Because device overwrites it, the contents of the tag make
do no matter. Current code tries to save space by reusing the
header as the tag. This, however, leads to an additional frag
being created and defeats buffer coalescing (which trickles
all the way down to the drivers).
Remove this optimization, and try to allocate the space for
the tag in the usual way, leave the memory uninitialized.
If memory allocation fails rewind the record pointer so that
we use the already copied user data as tag.
Note that the optimization was actually buggy, as the tag
for TLS 1.2 is 16 bytes, but header is just 13, so the reuse
may had looked past the end of the page..
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>
All modifications to TLS record list happen under the socket
lock. Since records form an ordered queue readers are only
concerned about elements being removed, additions can happen
concurrently.
Use RCU primitives to ensure the correct access types
(READ_ONCE/WRITE_ONCE).
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>
It's generally more cache friendly to walk arrays in order,
especially those which are likely not in cache.
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>
If retransmit record hint fall into the cleanup window we will
free it by just walking the list. No need to duplicate the code.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
TLS code has a number of #ifdefs which make the code a little
harder to follow. Recent fixes removed the ifdef around the
TLS_HW define, so we can switch to the often used pattern
of defining tls_device functions as empty static inlines
in the header when CONFIG_TLS_DEVICE=n.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
On setsockopt path we need to hold device_offload_lock from
the moment we check netdev is up until the context is fully
ready to be added to the tls_device_list.
No need to hold it around the get_netdev_for_sock().
Change the code and remove the confusing comment.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Reusing parts of error path for normal exit will make
next commit harder to read, untangle the two.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since we already have the pointer to the full original sk_proto
stored use that instead of storing all individual callback
pointers as well.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When an application configures kernel TLS on top of a TCP socket, it's
now possible for inet_diag_handler() to collect information regarding the
protocol version, the cipher type and TX / RX configuration, in case
INET_DIAG_INFO is requested.
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We need to make sure context does not get freed while diag
code is interrogating it. Free struct tls_context with
kfree_rcu().
We add the __rcu annotation directly in icsk, and cast it
away in the datapath accessor. Presumably all ULPs will
do a similar thing.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ctx->sk_write_space pointer is only set when TLS tx mode is enabled.
When running without TX mode its a null pointer but we still set the
sk sk_write_space pointer on close().
Fix the close path to only overwrite sk->sk_write_space when the current
pointer is to the tls_write_space function indicating the tls module should
clean it up properly as well.
Reported-by: Hillf Danton <hdanton@sina.com>
Cc: Ying Xue <ying.xue@windriver.com>
Cc: Andrey Konovalov <andreyknvl@google.com>
Fixes: 57c722e932 ("net/tls: swap sk_write_space on close")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Now that we swap the original proto and clear the ULP pointer
on close we have to make sure no callback will try to access
the freed state. sk_write_space is not part of sk_prot, remember
to swap it.
Reported-by: syzbot+dcdc9deefaec44785f32@syzkaller.appspotmail.com
Fixes: 95fa145479 ("bpf: sockmap/tls, close can race with map free")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
sk_validate_xmit_skb() and drivers depend on the sk member of
struct sk_buff to identify segments requiring encryption.
Any operation which removes or does not preserve the original TLS
socket such as skb_orphan() or skb_clone() will cause clear text
leaks.
Make the TCP socket underlying an offloaded TLS connection
mark all skbs as decrypted, if TLS TX is in offload mode.
Then in sk_validate_xmit_skb() catch skbs which have no socket
(or a socket with no validation) and decrypted flag set.
Note that CONFIG_SOCK_VALIDATE_XMIT, CONFIG_TLS_DEVICE and
sk->sk_validate_xmit_skb are slightly interchangeable right now,
they all imply TLS offload. The new checks are guarded by
CONFIG_TLS_DEVICE because that's the option guarding the
sk_buff->decrypted member.
Second, smaller issue with orphaning is that it breaks
the guarantee that packets will be delivered to device
queues in-order. All TLS offload drivers depend on that
scheduling property. This means skb_orphan_partial()'s
trick of preserving partial socket references will cause
issues in the drivers. We need a full orphan, and as a
result netem delay/throttling will cause all TLS offload
skbs to be dropped.
Reusing the sk_buff->decrypted flag also protects from
leaking clear text when incoming, decrypted skb is redirected
(e.g. by TC).
See commit 0608c69c9a ("bpf: sk_msg, sock{map|hash} redirect
through ULP") for justification why the internal flag is safe.
The only location which could leak the flag in is tcp_bpf_sendmsg(),
which is taken care of by clearing the previously unused bit.
v2:
- remove superfluous decrypted mark copy (Willem);
- remove the stale doc entry (Boris);
- rely entirely on EOR marking to prevent coalescing (Boris);
- use an internal sendpages flag instead of marking the socket
(Boris).
v3 (Willem):
- reorganize the can_skb_orphan_partial() condition;
- fix the flag leak-in through tcp_bpf_sendmsg.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Boris Pismenny <borisp@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Looks like we were slightly overzealous with the shutdown()
cleanup. Even though the sock->sk_state can reach CLOSED again,
socket->state will not got back to SS_UNCONNECTED once
connections is ESTABLISHED. Meaning we will see EISCONN if
we try to reconnect, and EINVAL if we try to listen.
Only listen sockets can be shutdown() and reused, but since
ESTABLISHED sockets can never be re-connected() or used for
listen() we don't need to try to clean up the ULP state early.
Fixes: 32857cf57f ("net/tls: fix transition through disconnect with close")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use accessor functions for skb fragment's page_offset instead
of direct references, in preparation for bvec conversion.
Signed-off-by: Jonathan Lemon <jonathan.lemon@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In preparation for unifying the skb_frag and bio_vec, use the fine
accessors which already exist and use skb_frag_t instead of
struct skb_frag_struct.
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
When a map free is called and in parallel a socket is closed we
have two paths that can potentially reset the socket prot ops, the
bpf close() path and the map free path. This creates a problem
with which prot ops should be used from the socket closed side.
If the map_free side completes first then we want to call the
original lowest level ops. However, if the tls path runs first
we want to call the sockmap ops. Additionally there was no locking
around prot updates in TLS code paths so the prot ops could
be changed multiple times once from TLS path and again from sockmap
side potentially leaving ops pointed at either TLS or sockmap
when psock and/or tls context have already been destroyed.
To fix this race first only update ops inside callback lock
so that TLS, sockmap and lowest level all agree on prot state.
Second and a ULP callback update() so that lower layers can
inform the upper layer when they are being removed allowing the
upper layer to reset prot ops.
This gets us close to allowing sockmap and tls to be stacked
in arbitrary order but will save that patch for *next trees.
v4:
- make sure we don't free things for device;
- remove the checks which swap the callbacks back
only if TLS is at the top.
Reported-by: syzbot+06537213db7ba2745c4a@syzkaller.appspotmail.com
Fixes: 02c558b2d5 ("bpf: sockmap, support for msg_peek in sk_msg with redirect ingress")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
It is possible (via shutdown()) for TCP socks to go through TCP_CLOSE
state via tcp_disconnect() without actually calling tcp_close which
would then call the tls close callback. Because of this a user could
disconnect a socket then put it in a LISTEN state which would break
our assumptions about sockets always being ESTABLISHED state.
More directly because close() can call unhash() and unhash is
implemented by sockmap if a sockmap socket has TLS enabled we can
incorrectly destroy the psock from unhash() and then call its close
handler again. But because the psock (sockmap socket representation)
is already destroyed we call close handler in sk->prot. However,
in some cases (TLS BASE/BASE case) this will still point at the
sockmap close handler resulting in a circular call and crash reported
by syzbot.
To fix both above issues implement the unhash() routine for TLS.
v4:
- add note about tls offload still needing the fix;
- move sk_proto to the cold cache line;
- split TX context free into "release" and "free",
otherwise the GC work itself is in already freed
memory;
- more TX before RX for consistency;
- reuse tls_ctx_free();
- schedule the GC work after we're done with context
to avoid UAF;
- don't set the unhash in all modes, all modes "inherit"
TLS_BASE's callbacks anyway;
- disable the unhash hook for TLS_HW.
Fixes: 3c4d755915 ("tls: kernel TLS support")
Reported-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
The tls close() callback currently drops the sock lock to call
strp_done(). Split up the RX cleanup into stopping the strparser
and releasing most resources, syncing strparser and finally
freeing the context.
To avoid the need for a strp_done() call on the cleanup path
of device offload make sure we don't arm the strparser until
we are sure init will be successful.
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
The tls close() callback currently drops the sock lock, makes a
cancel_delayed_work_sync() call, and then relocks the sock.
By restructuring the code we can avoid droping lock and then
reclaiming it. To simplify this we do the following,
tls_sk_proto_close
set_bit(CLOSING)
set_bit(SCHEDULE)
cancel_delay_work_sync() <- cancel workqueue
lock_sock(sk)
...
release_sock(sk)
strp_done()
Setting the CLOSING bit prevents the SCHEDULE bit from being
cleared by any workqueue items e.g. if one happens to be
scheduled and run between when we set SCHEDULE bit and cancel
work. Then because SCHEDULE bit is set now no new work will
be scheduled.
Tested with net selftests and bpf selftests.
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
The deprecated TOE offload doesn't actually do anything in
tls_sk_proto_close() - all TLS code is skipped and context
not freed. Remove the callback to make it easier to refactor
tls_sk_proto_close().
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
In tls_set_device_offload_rx() we prepare the software context
for RX fallback and proceed to add the connection to the device.
Unfortunately, software context prep includes arming strparser
so in case of a later error we have to release the socket lock
to call strp_done().
In preparation for not releasing the socket lock half way through
callbacks move arming strparser into a separate function.
Following patches will make use of that.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
netem runs skb_orphan_partial() which "disconnects" the skb
from normal TCP write memory accounting. We should not adjust
sk->sk_wmem_alloc on the fallback path for such skbs.
Fixes: e8f6979981 ("net/tls: Add generic NIC offload infrastructure")
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>
Turns out TLS_TX in HW offload mode does not initialize tls_prot_info.
Since commit 9cd81988cc ("net/tls: use version from prot") we actually
use this field on the datapath. Luckily we always compare it to TLS 1.3,
and assume 1.2 otherwise. So since zero is not equal to 1.3, everything
worked fine.
Fixes: 9cd81988cc ("net/tls: use version from prot")
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>
Introduce a return code for the tls_dev_resync callback.
When the driver TX resync fails, kernel can retry the resync again
until it succeeds. This prevents drivers from attempting to offload
TLS packets if the connection is known to be out of sync.
We don't worry about the RX resync since they will be retried naturally
as more encrypted records get received.
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>
David reports that RPC applications which use epoll() occasionally
get stuck, and that TLS ULP causes the kernel to not wake applications,
even though read() will return data.
This is indeed true. The ctx->rx_list which holds partially copied
records is not consulted when deciding whether socket is readable.
Note that SO_RCVLOWAT with epoll() is and has always been broken for
kernel TLS. We'd need to parse all records from the TCP layer, instead
of just the first one.
Fixes: 692d7b5d1f ("tls: Fix recvmsg() to be able to peek across multiple records")
Reported-by: David Beckett <david.beckett@netronome.com>
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>
Commit 86029d10af ("tls: zero the crypto information from tls_context
before freeing") added memzero_explicit() calls to clear the key material
before freeing struct tls_context, but it missed tls_device.c has its
own way of freeing this structure. Replace the missing free.
Fixes: 86029d10af ("tls: zero the crypto information from tls_context before freeing")
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>
Neither drivers nor the tls offload code currently supports TLS
version 1.3. Check the TLS version when installing connection
state. TLS 1.3 will just fallback to the kernel crypto for now.
Fixes: 130b392c6c ("net: tls: Add tls 1.3 support")
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>
The new route handling in ip_mc_finish_output() from 'net' overlapped
with the new support for returning congestion notifications from BPF
programs.
In order to handle this I had to take the dev_loopback_xmit() calls
out of the switch statement.
The aquantia driver conflicts were simple overlapping changes.
Signed-off-by: David S. Miller <davem@davemloft.net>
tls_sw_do_sendpage needs to return the total number of bytes sent
regardless of how many sk_msgs are allocated. Unfortunately, copied
(the value we return up the stack) is zero'd before each new sk_msg
is allocated so we only return the copied size of the last sk_msg used.
The caller (splice, etc.) of sendpage will then believe only part
of its data was sent and send the missing chunks again. However,
because the data actually was sent the receiver will get multiple
copies of the same data.
To reproduce this do multiple sendfile calls with a length close to
the max record size. This will in turn call splice/sendpage, sendpage
may use multiple sk_msg in this case and then returns the incorrect
number of bytes. This will cause splice to resend creating duplicate
data on the receiver. Andre created a C program that can easily
generate this case so we will push a similar selftest for this to
bpf-next shortly.
The fix is to _not_ zero the copied field so that the total sent
bytes is returned.
Reported-by: Steinar H. Gunderson <steinar+kernel@gunderson.no>
Reported-by: Andre Tomt <andre@tomt.net>
Tested-by: Andre Tomt <andre@tomt.net>
Fixes: d829e9c411 ("tls: convert to generic sk_msg interface")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
TLS offload drivers keep track of TCP seq numbers to make sure
the packets are fed into the HW in order.
When packets get dropped on the way through the stack, the driver
will get out of sync and have to use fallback encryption, but unless
TCP seq number is resynced it will never match the packets correctly
(or even worse - use incorrect record sequence number after TCP seq
wraps).
Existing drivers (mlx5) feed the entire record on every out-of-order
event, allowing FW/HW to always be in sync.
This patch adds an alternative, more akin to the RX resync. When
driver sees a frame which is past its expected sequence number the
stream must have gotten out of order (if the sequence number is
smaller than expected its likely a retransmission which doesn't
require resync). Driver will ask the stack to perform TX sync
before it submits the next full record, and fall back to software
crypto until stack has performed the sync.
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>
Currently only RX direction is ever resynced, however, TX may
also get out of sequence if packets get dropped on the way to
the driver. Rename the resync callback and add a direction
parameter.
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>
TLS offload device may lose sync with the TCP stream if packets
arrive out of order. Drivers can currently request a resync at
a specific TCP sequence number. When a record is found starting
at that sequence number kernel will inform the device of the
corresponding record number.
This requires the device to constantly scan the stream for a
known pattern (constant bytes of the header) after sync is lost.
This patch adds an alternative approach which is entirely under
the control of the kernel. Kernel tracks records it had to fully
decrypt, even though TLS socket is in TLS_HW mode. If multiple
records did not have any decrypted parts - it's a pretty strong
indication that the device is out of sync.
We choose the min number of fully encrypted records to be 2,
which should hopefully be more than will get retransmitted at
a time.
After kernel decides the device is out of sync it schedules a
resync request. If the TCP socket is empty the resync gets
performed immediately. If socket is not empty we leave the
record parser to resync when next record comes.
Before resync in message parser we peek at the TCP socket and
don't attempt the sync if the socket already has some of the
next record queued.
On resync failure (encrypted data continues to flow in) we
retry with exponential backoff, up to once every 128 records
(with a 16k record thats at most once every 2M of data).
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>
handle_device_resync() doesn't describe the function very well.
The function checks if resync should be issued upon parsing of
a new record.
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>
TLS offload code casts record number to a u64. The buffer
should be aligned to 8 bytes, but its actually a __be64, and
the rest of the TLS code treats it as big int. Make the
offload callbacks take a byte array, drivers can make the
choice to do the ugly cast if they want to.
Prepare for copying the record number onto the stack by
defining a constant for max size of the byte array.
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>
We subtract "TLS_HEADER_SIZE - 1" from req_seq, then if they
match we add the same constant to seq. Just add it to seq,
and we don't have to touch req_seq.
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>
Some ISDN files that got removed in net-next had some changes
done in mainline, take the removals.
Signed-off-by: David S. Miller <davem@davemloft.net>
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>
All callers pass prot->version as the last parameter
of tls_advance_record_sn(), yet tls_advance_record_sn()
itself needs a pointer to prot. Pass prot from callers.
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>
ctx->prot holds the same information as per-direction contexts.
Almost all code gets TLS version from this structure, convert
the last two stragglers, this way we can improve the cache
utilization by moving the per-direction data into cold cache lines.
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>
tls_device_decrypted() is only called from decrypt_skb_update(),
when ctx->decrypted == false, there is no need to re-check the bit.
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>
If the RX config of a TLS socket is SW, there is no point iterating
over the fragments and checking if frame is decrypted. It will
always be fully encrypted. Note that in fully encrypted case
the function doesn't actually touch any offload-related state,
so it's safe to call for TLS_SW, today. Soon we will introduce
code which can only be called for offloaded contexts.
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>
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>
In light of recent bugs, we should make a better effort of
checking return values. In theory none of the functions should
fail today.
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>
Commit 38030d7cb7 ("net/tls: avoid NULL-deref on resync during device removal")
tried to fix a potential NULL-dereference by taking the
context rwsem. Unfortunately the RX resync may get called
from soft IRQ, so we can't use the rwsem to protect from
the device disappearing. Because we are guaranteed there
can be only one resync at a time (it's called from strparser)
use a bit to indicate resync is busy and make device
removal wait for the bit to get cleared.
Note that there is a leftover "flags" field in struct
tls_context already.
Fixes: 4799ac81e5 ("tls: Add rx inline crypto offload")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This reverts commit 38030d7cb7.
Unfortunately the RX resync may get called from soft IRQ,
so we can't take the rwsem to protect from the device
disappearing.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When tls_sw_recvmsg() partially copies a record it pops that
record from ctx->recv_pkt and places it on rx_list.
Next iteration of tls_sw_recvmsg() reads from rx_list via
process_rx_list() before it enters the decryption loop.
If there is no more records to be read tls_wait_data()
will put the process on the wait queue and got to sleep.
This is incorrect, because some data was already copied
in process_rx_list().
In case of RPC connections process may never get woken up,
because peer also simply blocks in read().
I think this may also fix a similar issue when BPF is at
play, because after __tcp_bpf_recvmsg() returns some data
we subtract it from len and use continue to restart the
loop, but len could have just reached 0, so again we'd
sleep unnecessarily. That's added by:
commit d3b18ad31f ("tls: add bpf support to sk_msg handling")
Fixes: 692d7b5d1f ("tls: Fix recvmsg() to be able to peek across multiple records")
Reported-by: David Beckett <david.beckett@netronome.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Tested-by: David Beckett <david.beckett@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If some of the data came from the previous record, i.e. from
the rx_list it had already been decrypted, so it's not counted
towards the "decrypted" variable, but the "copied" variable.
Take that into account when checking lowat.
When calculating lowat target we need to pass the original len.
E.g. if lowat is at 80, len is 100 and we had 30 bytes on rx_list
target would currently be incorrectly calculated as 70, even though
we only need 50 more bytes to make up the 80.
Fixes: 692d7b5d1f ("tls: Fix recvmsg() to be able to peek across multiple records")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Tested-by: David Beckett <david.beckett@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
On device surprise removal path (the notifier) we can't
bail just because the features are disabled. They may
have been enabled during the lifetime of the device.
This bug leads to leaking netdev references and
use-after-frees if there are active connections while
device features are cleared.
Fixes: e8f6979981 ("net/tls: Add generic NIC offload infrastructure")
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>
TLS offload drivers shouldn't (and currently don't) block
the TLS offload feature changes based on whether there are
active offloaded connections or not.
This seems to be a good idea, because we want the admin to
be able to disable the TLS offload at any time, and there
is no clean way of disabling it for active connections
(TX side is quite problematic). So if features are cleared
existing connections will stay offloaded until they close,
and new connections will not attempt offload to a given
device.
However, the offload state removal handling is currently
broken if feature flags get cleared while there are
active TLS offloads.
RX side will completely bail from cleanup, even on normal
remove path, leaving device state dangling, potentially
causing issues when the 5-tuple is reused. It will also
fail to release the netdev reference.
Remove the RX-side warning message, in next release cycle
it should be printed when features are disabled, rather
than when connection dies, but for that we need a more
efficient method of finding connection of a given netdev
(a'la BPF offload code).
Fixes: 4799ac81e5 ("tls: Add rx inline crypto offload")
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>
Arnd Bergmann
Jakub Sitnicki
David S. Miller
Rohit Maheshwari
John Fastabend
Jakub Kicinski
Valentin Vidic
Willem de Bruijn
YueHaibing
Davide Caratti
Jonathan Lemon
Matthew Wilcox (Oracle)
Dirk van der Merwe