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>
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>
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>
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>
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>
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>
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>
When device refuses the offload in tls_set_device_offload_rx()
it calls tls_sw_free_resources_rx() to clean up software context
state.
Unfortunately, tls_sw_free_resources_rx() does not free all
the state tls_set_sw_offload() allocated - it leaks IV and
sequence number buffers. All other code paths which lead to
tls_sw_release_resources_rx() (which tls_sw_free_resources_rx()
calls) free those right before the call.
Avoid the leak by moving freeing of iv and rec_seq into
tls_sw_release_resources_rx().
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>
buildbot noticed that TLS_HW is not defined if CONFIG_TLS_DEVICE=n.
Wrap the cleanup branch into an ifdef, tls_device_free_resources_tx()
wouldn't be compiled either in this case.
Fixes: 35b71a34ad ("net/tls: don't leak partially sent record in device mode")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
David reports that tls triggers warnings related to
sk->sk_forward_alloc not being zero at destruction time:
WARNING: CPU: 5 PID: 6831 at net/core/stream.c:206 sk_stream_kill_queues+0x103/0x110
WARNING: CPU: 5 PID: 6831 at net/ipv4/af_inet.c:160 inet_sock_destruct+0x15b/0x170
When sender fills up the write buffer and dies from
SIGPIPE. This is due to the device implementation
not cleaning up the partially_sent_record.
This is because commit a42055e8d2 ("net/tls: Add support for async encryption of records for performance")
moved the partial record cleanup to the SW-only path.
Fixes: a42055e8d2 ("net/tls: Add support for async encryption of records for performance")
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>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Added support for AES128-CCM based record encryption. AES128-CCM is
similar to AES128-GCM. Both of them have same salt/iv/mac size. The
notable difference between the two is that while invoking AES128-CCM
operation, the salt||nonce (which is passed as IV) has to be prefixed
with a hardcoded value '2'. Further, CCM implementation in kernel
requires IV passed in crypto_aead_request() to be full '16' bytes.
Therefore, the record structure 'struct tls_rec' has been modified to
reserve '16' bytes for IV. This works for both GCM and CCM based cipher.
Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
A previous fix ("tls: Fix write space handling") assumed that
user space application gets informed about the socket send buffer
availability when tls_push_sg() gets called. Inside tls_push_sg(), in
case do_tcp_sendpages() returns 0, the function returns without calling
ctx->sk_write_space. Further, the new function tls_sw_write_space()
did not invoke ctx->sk_write_space. This leads to situation that user
space application encounters a lockup always waiting for socket send
buffer to become available.
Rather than call ctx->sk_write_space from tls_push_sg(), it should be
called from tls_write_space. So whenever tcp stack invokes
sk->sk_write_space after freeing socket send buffer, we always declare
the same to user space by the way of invoking ctx->sk_write_space.
Fixes: 7463d3a2db ("tls: Fix write space handling")
Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Reviewed-by: Boris Pismenny <borisp@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
TLS device cannot use the sw context. This patch returns the original
tls device write space handler and moves the sw/device specific portions
to the relevant files.
Also, we remove the write_space call for the tls_sw flow, because it
handles partial records in its delayed tx work handler.
Fixes: a42055e8d2 ("net/tls: Add support for async encryption of records for performance")
Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Reviewed-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Each tls context maintains two cipher contexts (one each for tx and rx
directions). For each tls session, the constants such as protocol
version, ciphersuite, iv size, associated data size etc are same for
both the directions and need to be stored only once per tls context.
Hence these are moved from 'struct cipher_context' to 'struct
tls_prot_info' and stored only once in 'struct tls_context'.
Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
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>
Wire up support for 256 bit keys from the setsockopt to the crypto
framework
Signed-off-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
build protos is required for tls_hw_prot also hence moved to
'tls_build_proto' and called as required from tls_init
and tls_hw_proto. This is required since build_protos
for v4 is moved from tls_register to tls_init in
commit <28cb6f1eaffdc5a6a9707cac55f4a43aa3fd7895>
Signed-off-by: Atul Gupta <atul.gupta@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann says:
====================
pull-request: bpf-next 2018-12-21
The following pull-request contains BPF updates for your *net-next* tree.
There is a merge conflict in test_verifier.c. Result looks as follows:
[...]
},
{
"calls: cross frame pruning",
.insns = {
[...]
.prog_type = BPF_PROG_TYPE_SOCKET_FILTER,
.errstr_unpriv = "function calls to other bpf functions are allowed for root only",
.result_unpriv = REJECT,
.errstr = "!read_ok",
.result = REJECT,
},
{
"jset: functional",
.insns = {
[...]
{
"jset: unknown const compare not taken",
.insns = {
BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
BPF_FUNC_get_prandom_u32),
BPF_JMP_IMM(BPF_JSET, BPF_REG_0, 1, 1),
BPF_LDX_MEM(BPF_B, BPF_REG_8, BPF_REG_9, 0),
BPF_EXIT_INSN(),
},
.prog_type = BPF_PROG_TYPE_SOCKET_FILTER,
.errstr_unpriv = "!read_ok",
.result_unpriv = REJECT,
.errstr = "!read_ok",
.result = REJECT,
},
[...]
{
"jset: range",
.insns = {
[...]
},
.prog_type = BPF_PROG_TYPE_SOCKET_FILTER,
.result_unpriv = ACCEPT,
.result = ACCEPT,
},
The main changes are:
1) Various BTF related improvements in order to get line info
working. Meaning, verifier will now annotate the corresponding
BPF C code to the error log, from Martin and Yonghong.
2) Implement support for raw BPF tracepoints in modules, from Matt.
3) Add several improvements to verifier state logic, namely speeding
up stacksafe check, optimizations for stack state equivalence
test and safety checks for liveness analysis, from Alexei.
4) Teach verifier to make use of BPF_JSET instruction, add several
test cases to kselftests and remove nfp specific JSET optimization
now that verifier has awareness, from Jakub.
5) Improve BPF verifier's slot_type marking logic in order to
allow more stack slot sharing, from Jiong.
6) Add sk_msg->size member for context access and add set of fixes
and improvements to make sock_map with kTLS usable with openssl
based applications, from John.
7) Several cleanups and documentation updates in bpftool as well as
auto-mount of tracefs for "bpftool prog tracelog" command,
from Quentin.
8) Include sub-program tags from now on in bpf_prog_info in order to
have a reliable way for user space to get all tags of the program
e.g. needed for kallsyms correlation, from Song.
9) Add BTF annotations for cgroup_local_storage BPF maps and
implement bpf fs pretty print support, from Roman.
10) Fix bpftool in order to allow for cross-compilation, from Ivan.
11) Update of bpftool license to GPLv2-only + BSD-2-Clause in order
to be compatible with libbfd and allow for Debian packaging,
from Jakub.
12) Remove an obsolete prog->aux sanitation in dump and get rid of
version check for prog load, from Daniel.
13) Fix a memory leak in libbpf's line info handling, from Prashant.
14) Fix cpumap's frame alignment for build_skb() so that skb_shared_info
does not get unaligned, from Jesper.
15) Fix test_progs kselftest to work with older compilers which are less
smart in optimizing (and thus throwing build error), from Stanislav.
16) Cleanup and simplify AF_XDP socket teardown, from Björn.
17) Fix sk lookup in BPF kselftest's test_sock_addr with regards
to netns_id argument, from Andrey.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
The existing code did not expect users would initialize the TLS ULP
without subsequently calling the TLS TX enabling socket option.
If the application tries to send data after the TLS ULP enable op
but before the TLS TX enable op the BPF sk_msg verdict program is
skipped. This patch resolves this by converting the ipv4 sock ops
to be calculated at init time the same way ipv6 ops are done. This
pulls in any changes to the sock ops structure that have been made
after the socket was created including the changes from adding the
socket to a sock{map|hash}.
This was discovered by running OpenSSL master branch which calls
the TLS ULP setsockopt early in TLS handshake but only enables
the TLS TX path once the handshake has completed. As a result the
datapath missed the initial handshake messages.
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: Daniel Borkmann <daniel@iogearbox.net>
HW unhash within mutex for registered tls devices cause sleep
when called from tcp_set_state for TCP_CLOSE. Release lock and
re-acquire after function call with ref count incr/dec.
defined kref and fp release for tls_device to ensure device
is not released outside lock.
BUG: sleeping function called from invalid context at
kernel/locking/mutex.c:748
in_atomic(): 1, irqs_disabled(): 0, pid: 0, name: swapper/7
INFO: lockdep is turned off.
CPU: 7 PID: 0 Comm: swapper/7 Tainted: G W O
Call Trace:
<IRQ>
dump_stack+0x5e/0x8b
___might_sleep+0x222/0x260
__mutex_lock+0x5c/0xa50
? vprintk_emit+0x1f3/0x440
? kmem_cache_free+0x22d/0x2a0
? tls_hw_unhash+0x2f/0x80
? printk+0x52/0x6e
? tls_hw_unhash+0x2f/0x80
tls_hw_unhash+0x2f/0x80
tcp_set_state+0x5f/0x180
tcp_done+0x2e/0xe0
tcp_rcv_state_process+0x92c/0xdd3
? lock_acquire+0xf5/0x1f0
? tcp_v4_rcv+0xa7c/0xbe0
? tcp_v4_do_rcv+0x70/0x1e0
Signed-off-by: Atul Gupta <atul.gupta@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
create_ctx is called from tls_init and tls_hw_prot
hence initialize function pointers in common routine.
Signed-off-by: Atul Gupta <atul.gupta@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
They are not used anymore and therefore should be removed.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Instead of re-implementing poll routine use the poll callback to
trigger read from kTLS, we reuse the stream_memory_read callback
which is simpler and achieves the same. This helps to align sockmap
and kTLS so we can more easily embed BPF in kTLS.
Joint work with Daniel.
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
On processors with multi-engine crypto accelerators, it is possible that
multiple records get encrypted in parallel and their encryption
completion is notified to different cpus in multicore processor. This
leads to the situation where tls_encrypt_done() starts executing in
parallel on different cores. In current implementation, encrypted
records are queued to tx_ready_list in tls_encrypt_done(). This requires
addition to linked list 'tx_ready_list' to be protected. As
tls_decrypt_done() could be executing in irq content, it is not possible
to protect linked list addition operation using a lock.
To fix the problem, we remove linked list addition operation from the
irq context. We do tx_ready_list addition/removal operation from
application context only and get rid of possible multiple access to
the linked list. Before starting encryption on the record, we add it to
the tail of tx_ready_list. To prevent tls_tx_records() from transmitting
it, we mark the record with a new flag 'tx_ready' in 'struct tls_rec'.
When record encryption gets completed, tls_encrypt_done() has to only
update the 'tx_ready' flag to true & linked list add operation is not
required.
The changed logic brings some other side benefits. Since the records
are always submitted in tls sequence number order for encryption, the
tx_ready_list always remains sorted and addition of new records to it
does not have to traverse the linked list.
Lastly, we renamed tx_ready_list in 'struct tls_sw_context_tx' to
'tx_list'. This is because now, the some of the records at the tail are
not ready to transmit.
Fixes: a42055e8d2 ("net/tls: Add support for async encryption")
Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In current implementation, tls records are encrypted & transmitted
serially. Till the time the previously submitted user data is encrypted,
the implementation waits and on finish starts transmitting the record.
This approach of encrypt-one record at a time is inefficient when
asynchronous crypto accelerators are used. For each record, there are
overheads of interrupts, driver softIRQ scheduling etc. Also the crypto
accelerator sits idle most of time while an encrypted record's pages are
handed over to tcp stack for transmission.
This patch enables encryption of multiple records in parallel when an
async capable crypto accelerator is present in system. This is achieved
by allowing the user space application to send more data using sendmsg()
even while previously issued data is being processed by crypto
accelerator. This requires returning the control back to user space
application after submitting encryption request to accelerator. This
also means that zero-copy mode of encryption cannot be used with async
accelerator as we must be done with user space application buffer before
returning from sendmsg().
There can be multiple records in flight to/from the accelerator. Each of
the record is represented by 'struct tls_rec'. This is used to store the
memory pages for the record.
After the records are encrypted, they are added in a linked list called
tx_ready_list which contains encrypted tls records sorted as per tls
sequence number. The records from tx_ready_list are transmitted using a
newly introduced function called tls_tx_records(). The tx_ready_list is
polled for any record ready to be transmitted in sendmsg(), sendpage()
after initiating encryption of new tls records. This achieves parallel
encryption and transmission of records when async accelerator is
present.
There could be situation when crypto accelerator completes encryption
later than polling of tx_ready_list by sendmsg()/sendpage(). Therefore
we need a deferred work context to be able to transmit records from
tx_ready_list. The deferred work context gets scheduled if applications
are not sending much data through the socket. If the applications issue
sendmsg()/sendpage() in quick succession, then the scheduling of
tx_work_handler gets cancelled as the tx_ready_list would be polled from
application's context itself. This saves scheduling overhead of deferred
work.
The patch also brings some side benefit. We are able to get rid of the
concept of CLOSED record. This is because the records once closed are
either encrypted and then placed into tx_ready_list or if encryption
fails, the socket error is set. This simplifies the kernel tls
sendpath. However since tls_device.c is still using macros, accessory
functions for CLOSED records have been retained.
Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
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: 3c4d755915 ("tls: kernel TLS support")
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, the lower protocols sk_write_space handler is not called if
TLS is sending a scatterlist via tls_push_sg. However, normally
tls_push_sg calls do_tcp_sendpage, which may be under memory pressure,
that in turn may trigger a wait via sk_wait_event. Typically, this
happens when the in-flight bytes exceed the sdnbuf size. In the normal
case when enough ACKs are received sk_write_space() will be called and
the sk_wait_event will be woken up allowing it to send more data
and/or return to the user.
But, in the TLS case because the sk_write_space() handler does not
wake up the events the above send will wait until the sndtimeo is
exceeded. By default this is MAX_SCHEDULE_TIMEOUT so it look like a
hang to the user (especially this impatient user). To fix this pass
the sk_write_space event to the lower layers sk_write_space event
which in the TCP case will wake any pending events.
I observed the above while integrating sockmap and ktls. It
initially appeared as test_sockmap (modified to use ktls) occasionally
hanging. To reliably reproduce this reduce the sndbuf size and stress
the tls layer by sending many 1B sends. This results in every byte
needing a header and each byte individually being sent to the crypto
layer.
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Acked-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Lets not turn the TCP ULP lookup into an arbitrary module loader as
we only intend to load ULP modules through this mechanism, not other
unrelated kernel modules:
[root@bar]# cat foo.c
#include <sys/types.h>
#include <sys/socket.h>
#include <linux/tcp.h>
#include <linux/in.h>
int main(void)
{
int sock = socket(PF_INET, SOCK_STREAM, 0);
setsockopt(sock, IPPROTO_TCP, TCP_ULP, "sctp", sizeof("sctp"));
return 0;
}
[root@bar]# gcc foo.c -O2 -Wall
[root@bar]# lsmod | grep sctp
[root@bar]# ./a.out
[root@bar]# lsmod | grep sctp
sctp 1077248 4
libcrc32c 16384 3 nf_conntrack,nf_nat,sctp
[root@bar]#
Fix it by adding module alias to TCP ULP modules, so probing module
via request_module() will be limited to tcp-ulp-[name]. The existing
modules like kTLS will load fine given tcp-ulp-tls alias, but others
will fail to load:
[root@bar]# lsmod | grep sctp
[root@bar]# ./a.out
[root@bar]# lsmod | grep sctp
[root@bar]#
Sockmap is not affected from this since it's either built-in or not.
Fixes: 734942cc4e ("tcp: ULP infrastructure")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
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>
The poll() changes were not well thought out, and completely
unexplained. They also caused a huge performance regression, because
"->poll()" was no longer a trivial file operation that just called down
to the underlying file operations, but instead did at least two indirect
calls.
Indirect calls are sadly slow now with the Spectre mitigation, but the
performance problem could at least be largely mitigated by changing the
"->get_poll_head()" operation to just have a per-file-descriptor pointer
to the poll head instead. That gets rid of one of the new indirections.
But that doesn't fix the new complexity that is completely unwarranted
for the regular case. The (undocumented) reason for the poll() changes
was some alleged AIO poll race fixing, but we don't make the common case
slower and more complex for some uncommon special case, so this all
really needs way more explanations and most likely a fundamental
redesign.
[ This revert is a revert of about 30 different commits, not reverted
individually because that would just be unnecessarily messy - Linus ]
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
While hacking on kTLS, I ran into the following panic from an
unprivileged netserver / netperf TCP session:
BUG: unable to handle kernel NULL pointer dereference at 0000000000000000
PGD 800000037f378067 P4D 800000037f378067 PUD 3c0e61067 PMD 0
Oops: 0010 [#1] SMP KASAN PTI
CPU: 1 PID: 2289 Comm: netserver Not tainted 4.17.0+ #139
Hardware name: LENOVO 20FBCTO1WW/20FBCTO1WW, BIOS N1FET47W (1.21 ) 11/28/2016
RIP: 0010: (null)
Code: Bad RIP value.
RSP: 0018:ffff88036abcf740 EFLAGS: 00010246
RAX: dffffc0000000000 RBX: ffff88036f5f6800 RCX: 1ffff1006debed26
RDX: ffff88036abcf920 RSI: ffff8803cb1a4f00 RDI: ffff8803c258c280
RBP: ffff8803c258c280 R08: ffff8803c258c280 R09: ffffed006f559d48
R10: ffff88037aacea43 R11: ffffed006f559d49 R12: ffff8803c258c280
R13: ffff8803cb1a4f20 R14: 00000000000000db R15: ffffffffc168a350
FS: 00007f7e631f4700(0000) GS:ffff8803d1c80000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: ffffffffffffffd6 CR3: 00000003ccf64005 CR4: 00000000003606e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
? tls_sw_poll+0xa4/0x160 [tls]
? sock_poll+0x20a/0x680
? do_select+0x77b/0x11a0
? poll_schedule_timeout.constprop.12+0x130/0x130
? pick_link+0xb00/0xb00
? read_word_at_a_time+0x13/0x20
? vfs_poll+0x270/0x270
? deref_stack_reg+0xad/0xe0
? __read_once_size_nocheck.constprop.6+0x10/0x10
[...]
Debugging further, it turns out that calling into ctx->sk_poll() is
invalid since sk_poll itself is NULL which was saved from the original
TCP socket in order for tls_sw_poll() to invoke it.
Looks like the recent conversion from poll to poll_mask callback started
in 1525242310 ("net: add support for ->poll_mask in proto_ops") missed
to eventually convert kTLS, too: TCP's ->poll was converted over to the
->poll_mask in commit 2c7d3daceb ("net/tcp: convert to ->poll_mask")
and therefore kTLS wrongly saved the ->poll old one which is now NULL.
Convert kTLS over to use ->poll_mask instead. Also instead of POLLIN |
POLLRDNORM use the proper EPOLLIN | EPOLLRDNORM bits as the case in
tcp_poll_mask() as well that is mangled here.
Fixes: 2c7d3daceb ("net/tcp: convert to ->poll_mask")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Dave Watson <davejwatson@fb.com>
Tested-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The bpf syscall and selftests conflicts were trivial
overlapping changes.
The r8169 change involved moving the added mdelay from 'net' into a
different function.
A TLS close bug fix overlapped with the splitting of the TLS state
into separate TX and RX parts. I just expanded the tests in the bug
fix from "ctx->conf == X" into "ctx->tx_conf == X && ctx->rx_conf
== X".
Signed-off-by: David S. Miller <davem@davemloft.net>
In the case of writing a partial tls record we forgot to clear the
ctx->in_tcp_sendpages flag, causing some connections to stall.
Fixes: c212d2c7fc ("net/tls: Don't recursively call push_record during tls_write_space callbacks")
Signed-off-by: Andre Tomt <andre@tomt.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Willem de Bruijn