tls: get psock ref after taking rxlock to avoid leak

At the start of tls_sw_recvmsg, we take a reference on the psock, and
then call tls_rx_reader_lock. If that fails, we return directly
without releasing the reference.

Instead of adding a new label, just take the reference after locking
has succeeded, since we don't need it before.

Fixes: 4cbc325ed6b4 ("tls: rx: allow only one reader at a time")
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/fe2ade22d030051ce4c3638704ed58b67d0df643.1711120964.git.sd@queasysnail.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: adjust recv return with async crypto and failed copy to userspace

process_rx_list may not copy as many bytes as we want to the userspace
buffer, for example in case we hit an EFAULT during the copy. If this
happens, we should only count the bytes that were actually copied,
which may be 0.

Subtracting async_copy_bytes is correct in both peek and !peek cases,
because decrypted == async_copy_bytes + peeked for the peek case: peek
is always !ZC, and we can go through either the sync or async path. In
the async case, we add chunk to both decrypted and
async_copy_bytes. In the sync case, we add chunk to both decrypted and
peeked. I missed that in commit 6caaf104423d ("tls: fix peeking with
sync+async decryption").

Fixes: 4d42cd6bc2ac ("tls: rx: fix return value for async crypto")
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/1b5a1eaab3c088a9dd5d9f1059ceecd7afe888d1.1711120964.git.sd@queasysnail.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: recv: process_rx_list shouldn't use an offset with kvec

Only MSG_PEEK needs to copy from an offset during the final
process_rx_list call, because the bytes we copied at the beginning of
tls_sw_recvmsg were left on the rx_list. In the KVEC case, we removed
data from the rx_list as we were copying it, so there's no need to use
an offset, just like in the normal case.

Fixes: 692d7b5d1f91 ("tls: Fix recvmsg() to be able to peek across multiple records")
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/e5487514f828e0347d2b92ca40002c62b58af73d.1711120964.git.sd@queasysnail.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: fix use-after-free on failed backlog decryption

When the decrypt request goes to the backlog and crypto_aead_decrypt
returns -EBUSY, tls_do_decryption will wait until all async
decryptions have completed. If one of them fails, tls_do_decryption
will return -EBADMSG and tls_decrypt_sg jumps to the error path,
releasing all the pages. But the pages have been passed to the async
callback, and have already been released by tls_decrypt_done.

The only true async case is when crypto_aead_decrypt returns
-EINPROGRESS. With -EBUSY, we already waited so we can tell
tls_sw_recvmsg that the data is available for immediate copy, but we
need to notify tls_decrypt_sg (via the new ->async_done flag) that the
memory has already been released.

Fixes: 859054147318 ("net: tls: handle backlogging of crypto requests")
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Link: https://lore.kernel.org/r/4755dd8d9bebdefaa19ce1439b833d6199d4364c.1709132643.git.sd@queasysnail.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: separate no-async decryption request handling from async

If we're not doing async, the handling is much simpler. There's no
reference counting, we just need to wait for the completion to wake us
up and return its result.

We should preferably also use a separate crypto_wait. I'm not seeing a
UAF as I did in the past, I think aec7961916f3 ("tls: fix race between
async notify and socket close") took care of it.

This will make the next fix easier.

Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Link: https://lore.kernel.org/r/47bde5f649707610eaef9f0d679519966fc31061.1709132643.git.sd@queasysnail.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: fix peeking with sync+async decryption

If we peek from 2 records with a currently empty rx_list, and the
first record is decrypted synchronously but the second record is
decrypted async, the following happens:
1. decrypt record 1 (sync)
2. copy from record 1 to the userspace's msg
3. queue the decrypted record to rx_list for future read(!PEEK)
4. decrypt record 2 (async)
5. queue record 2 to rx_list
6. call process_rx_list to copy data from the 2nd record

We currently pass copied=0 as skip offset to process_rx_list, so we
end up copying once again from the first record. We should skip over
the data we've already copied.

Seen with selftest tls.12_aes_gcm.recv_peek_large_buf_mult_recs

Fixes: 692d7b5d1f91 ("tls: Fix recvmsg() to be able to peek across multiple records")
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Link: https://lore.kernel.org/r/1b132d2b2b99296bfde54e8a67672d90d6d16e71.1709132643.git.sd@queasysnail.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: decrement decrypt_pending if no async completion will be called

With mixed sync/async decryption, or failures of crypto_aead_decrypt,
we increment decrypt_pending but we never do the corresponding
decrement since tls_decrypt_done will not be called. In this case, we
should decrement decrypt_pending immediately to avoid getting stuck.

For example, the prequeue prequeue test gets stuck with mixed
modes (one async decrypt + one sync decrypt).

Fixes: 94524d8fc965 ("net/tls: Add support for async decryption of tls records")
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Link: https://lore.kernel.org/r/c56d5fc35543891d5319f834f25622360e1bfbec.1709132643.git.sd@queasysnail.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: don't skip over different type records from the rx_list

If we queue 3 records:
- record 1, type DATA
- record 2, some other type
- record 3, type DATA
and do a recv(PEEK), the rx_list will contain the first two records.

The next large recv will walk through the rx_list and copy data from
record 1, then stop because record 2 is a different type. Since we
haven't filled up our buffer, we will process the next available
record. It's also DATA, so we can merge it with the current read.

We shouldn't do that, since there was a record in between that we
ignored.

Add a flag to let process_rx_list inform tls_sw_recvmsg that it had
more data available.

Fixes: 692d7b5d1f91 ("tls: Fix recvmsg() to be able to peek across multiple records")
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Link: https://lore.kernel.org/r/f00c0c0afa080c60f016df1471158c1caf983c34.1708007371.git.sd@queasysnail.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: stop recv() if initial process_rx_list gave us non-DATA

If we have a non-DATA record on the rx_list and another record of the
same type still on the queue, we will end up merging them:
- process_rx_list copies the non-DATA record
- we start the loop and process the first available record since it's
of the same type
- we break out of the loop since the record was not DATA

Just check the record type and jump to the end in case process_rx_list
did some work.

Fixes: 692d7b5d1f91 ("tls: Fix recvmsg() to be able to peek across multiple records")
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Link: https://lore.kernel.org/r/bd31449e43bd4b6ff546f5c51cf958c31c511deb.1708007371.git.sd@queasysnail.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: break out of main loop when PEEK gets a non-data record

PEEK needs to leave decrypted records on the rx_list so that we can
receive them later on, so it jumps back into the async code that
queues the skb. Unfortunately that makes us skip the
TLS_RECORD_TYPE_DATA check at the bottom of the main loop, so if two
records of the same (non-DATA) type are queued, we end up merging
them.

Add the same record type check, and make it unlikely to not penalize
the async fastpath. Async decrypt only applies to data record, so this
check is only needed for PEEK.

process_rx_list also has similar issues.

Fixes: 692d7b5d1f91 ("tls: Fix recvmsg() to be able to peek across multiple records")
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Link: https://lore.kernel.org/r/3df2eef4fdae720c55e69472b5bea668772b45a2.1708007371.git.sd@queasysnail.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: tls: fix returned read length with async decrypt

We double count async, non-zc rx data. The previous fix was
lucky because if we fully zc async_copy_bytes is 0 so we add 0.
Decrypted already has all the bytes we handled, in all cases.
We don't have to adjust anything, delete the erroneous line.

Fixes: 4d42cd6bc2ac ("tls: rx: fix return value for async crypto")
Co-developed-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: tls: fix use-after-free with partial reads and async decrypt

tls_decrypt_sg doesn't take a reference on the pages from clear_skb,
so the put_page() in tls_decrypt_done releases them, and we trigger
a use-after-free in process_rx_list when we try to read from the
partially-read skb.

Fixes: fd31f3996af2 ("tls: rx: decrypt into a fresh skb")
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: tls: handle backlogging of crypto requests

Since we're setting the CRYPTO_TFM_REQ_MAY_BACKLOG flag on our
requests to the crypto API, crypto_aead_{encrypt,decrypt} can return
-EBUSY instead of -EINPROGRESS in valid situations. For example, when
the cryptd queue for AESNI is full (easy to trigger with an
artificially low cryptd.cryptd_max_cpu_qlen), requests will be enqueued
to the backlog but still processed. In that case, the async callback
will also be called twice: first with err == -EINPROGRESS, which it
seems we can just ignore, then with err == 0.

Compared to Sabrina's original patch this version uses the new
tls_*crypt_async_wait() helpers and converts the EBUSY to
EINPROGRESS to avoid having to modify all the error handling
paths. The handling is identical.

Fixes: a54667f6728c ("tls: Add support for encryption using async offload accelerator")
Fixes: 94524d8fc965 ("net/tls: Add support for async decryption of tls records")
Co-developed-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Link: https://lore.kernel.org/netdev/9681d1febfec295449a62300938ed2ae66983f28.1694018970.git.sd@queasysnail.net/
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: fix race between tx work scheduling and socket close

Similarly to previous commit, the submitting thread (recvmsg/sendmsg)
may exit as soon as the async crypto handler calls complete().
Reorder scheduling the work before calling complete().
This seems more logical in the first place, as it's
the inverse order of what the submitting thread will do.

Reported-by: valis <sec@valis.email>
Fixes: a42055e8d2c3 ("net/tls: Add support for async encryption of records for performance")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: fix race between async notify and socket close

The submitting thread (one which called recvmsg/sendmsg)
may exit as soon as the async crypto handler calls complete()
so any code past that point risks touching already freed data.

Try to avoid the locking and extra flags altogether.
Have the main thread hold an extra reference, this way
we can depend solely on the atomic ref counter for
synchronization.

Don't futz with reiniting the completion, either, we are now
tightly controlling when completion fires.

Reported-by: valis <sec@valis.email>
Fixes: 0cada33241d9 ("net/tls: fix race condition causing kernel panic")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: tls: factor out tls_*crypt_async_wait()

Factor out waiting for async encrypt and decrypt to finish.
There are already multiple copies and a subsequent fix will
need more. No functional changes.

Note that crypto_wait_req() returns wait->err

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: tls, fix WARNIING in __sk_msg_free

A splice with MSG_SPLICE_PAGES will cause tls code to use the
tls_sw_sendmsg_splice path in the TLS sendmsg code to move the user
provided pages from the msg into the msg_pl. This will loop over the
msg until msg_pl is full, checked by sk_msg_full(msg_pl). The user
can also set the MORE flag to hint stack to delay sending until receiving
more pages and ideally a full buffer.

If the user adds more pages to the msg than can fit in the msg_pl
scatterlist (MAX_MSG_FRAGS) we should ignore the MORE flag and send
the buffer anyways.

What actually happens though is we abort the msg to msg_pl scatterlist
setup and then because we forget to set 'full record' indicating we
can no longer consume data without a send we fallthrough to the 'continue'
path which will check if msg_data_left(msg) has more bytes to send and
then attempts to fit them in the already full msg_pl. Then next
iteration of sender doing send will encounter a full msg_pl and throw
the warning in the syzbot report.

To fix simply check if we have a full_record in splice code path and
if not send the msg regardless of MORE flag.

Reported-and-tested-by: syzbot+f2977222e0e95cec15c8@syzkaller.appspotmail.com
Reported-by: Edward Adam Davis <eadavis@qq.com>
Fixes: fe1e81d4f73b ("tls/sw: Support MSG_SPLICE_PAGES")
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: tls, update curr on splice as well

The curr pointer must also be updated on the splice similar to how
we do this for other copy types.

Fixes: d829e9c4112b ("tls: convert to generic sk_msg interface")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Reported-by: Jann Horn <jannh@google.com>
Link: https://lore.kernel.org/r/20231206232706.374377-2-john.fastabend@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: fix NULL deref on tls_sw_splice_eof() with empty record

syzkaller discovered that if tls_sw_splice_eof() is executed as part of
sendfile() when the plaintext/ciphertext sk_msg are empty, the send path
gets confused because the empty ciphertext buffer does not have enough
space for the encryption overhead. This causes tls_push_record() to go on
the `split = true` path (which is only supposed to be used when interacting
with an attached BPF program), and then get further confused and hit the
tls_merge_open_record() path, which then assumes that there must be at
least one populated buffer element, leading to a NULL deref.

It is possible to have empty plaintext/ciphertext buffers if we previously
bailed from tls_sw_sendmsg_locked() via the tls_trim_both_msgs() path.
tls_sw_push_pending_record() already handles this case correctly; let's do
the same check in tls_sw_splice_eof().

Fixes: df720d288dbb ("tls/sw: Use splice_eof() to flush")
Cc: stable@vger.kernel.org
Reported-by: syzbot+40d43509a099ea756317@syzkaller.appspotmail.com
Signed-off-by: Jann Horn <jannh@google.com>
Link: https://lore.kernel.org/r/20231122214447.675768-1-jannh@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: don't reset prot->aad_size and prot->tail_size for TLS_HW

Prior to commit 1a074f7618e8 ("tls: also use init_prot_info in
tls_set_device_offload"), setting TLS_HW on TX didn't touch
prot->aad_size and prot->tail_size. They are set to 0 during context
allocation (tls_prot_info is embedded in tls_context, kzalloc'd by
tls_ctx_create).

When the RX key is configured, tls_set_sw_offload is called (for both
TLS_SW and TLS_HW). If the TX key is configured in TLS_HW mode after
the RX key has been installed, init_prot_info will now overwrite the
correct values of aad_size and tail_size, breaking SW decryption and
causing -EBADMSG errors to be returned to userspace.

Since TLS_HW doesn't use aad_size and tail_size at all (for TLS1.2,
tail_size is always 0, and aad_size is equal to TLS_HEADER_SIZE +
rec_seq_size), we can simply drop this hunk.

Fixes: 1a074f7618e8 ("tls: also use init_prot_info in tls_set_device_offload")
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Tested-by: Ran Rozenstein <ranro@nvidia.com>
Link: https://lore.kernel.org/r/979d2f89a6a994d5bb49cae49a80be54150d094d.1697653889.git.sd@queasysnail.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: remove tls_context argument from tls_set_sw_offload

It's not really needed since we end up refetching it as tls_ctx. We
can also remove the NULL check, since we have already dereferenced ctx
in do_tls_setsockopt_conf.

Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: also use init_prot_info in tls_set_device_offload

Most values are shared. Nonce size turns out to be equal to IV size
for all offloadable ciphers.

Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: move tls_prot_info initialization out of tls_set_sw_offload

Simplify tls_set_sw_offload, and allow reuse for the tls_device code.

Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: extract context alloc/initialization out of tls_set_sw_offload

Simplify tls_set_sw_offload a bit.

Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: store iv directly within cipher_context

TLS_MAX_IV_SIZE + TLS_MAX_SALT_SIZE is 20B, we don't get much benefit
in cipher_context's size and can simplify the init code a bit.

Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rename MAX_IV_SIZE to TLS_MAX_IV_SIZE

It's defined in include/net/tls.h, avoid using an overly generic name.

Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: store rec_seq directly within cipher_context

TLS_MAX_REC_SEQ_SIZE is 8B, we don't get anything by using kmalloc.

Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: Use size_add() in call to struct_size()

If, for any reason, the open-coded arithmetic causes a wraparound,
the protection that `struct_size()` adds against potential integer
overflows is defeated. Fix this by hardening call to `struct_size()`
with `size_add()`.

Fixes: b89fec54fd61 ("tls: rx: wrap decrypt params in a struct")
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tcp: allow again tcp_disconnect() when threads are waiting

As reported by Tom, .NET and applications build on top of it rely
on connect(AF_UNSPEC) to async cancel pending I/O operations on TCP
socket.

The blamed commit below caused a regression, as such cancellation
can now fail.

As suggested by Eric, this change addresses the problem explicitly
causing blocking I/O operation to terminate immediately (with an error)
when a concurrent disconnect() is executed.

Instead of tracking the number of threads blocked on a given socket,
track the number of disconnect() issued on such socket. If such counter
changes after a blocking operation releasing and re-acquiring the socket
lock, error out the current operation.

Fixes: 4faeee0cf8a5 ("tcp: deny tcp_disconnect() when threads are waiting")
Reported-by: Tom Deseyn <tdeseyn@redhat.com>
Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1886305
Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/f3b95e47e3dbed840960548aebaa8d954372db41.1697008693.git.pabeni@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net/tls: do not free tls_rec on async operation in bpf_exec_tx_verdict()

I got the below warning when do fuzzing test:
BUG: KASAN: null-ptr-deref in scatterwalk_copychunks+0x320/0x470
Read of size 4 at addr 0000000000000008 by task kworker/u8:1/9

CPU: 0 PID: 9 Comm: kworker/u8:1 Tainted: G OE
Hardware name: linux,dummy-virt (DT)
Workqueue: pencrypt_parallel padata_parallel_worker
Call trace:
dump_backtrace+0x0/0x420
show_stack+0x34/0x44
dump_stack+0x1d0/0x248
__kasan_report+0x138/0x140
kasan_report+0x44/0x6c
__asan_load4+0x94/0xd0
scatterwalk_copychunks+0x320/0x470
skcipher_next_slow+0x14c/0x290
skcipher_walk_next+0x2fc/0x480
skcipher_walk_first+0x9c/0x110
skcipher_walk_aead_common+0x380/0x440
skcipher_walk_aead_encrypt+0x54/0x70
ccm_encrypt+0x13c/0x4d0
crypto_aead_encrypt+0x7c/0xfc
pcrypt_aead_enc+0x28/0x84
padata_parallel_worker+0xd0/0x2dc
process_one_work+0x49c/0xbdc
worker_thread+0x124/0x880
kthread+0x210/0x260
ret_from_fork+0x10/0x18

This is because the value of rec_seq of tls_crypto_info configured by the
user program is too large, for example, 0xffffffffffffff. In addition, TLS
is asynchronously accelerated. When tls_do_encryption() returns
-EINPROGRESS and sk->sk_err is set to EBADMSG due to rec_seq overflow,
skmsg is released before the asynchronous encryption process ends. As a
result, the UAF problem occurs during the asynchronous processing of the
encryption module.

If the operation is asynchronous and the encryption module returns
EINPROGRESS, do not free the record information.

Fixes: 635d93981786 ("net/tls: free record only on encryption error")
Signed-off-by: Liu Jian <liujian56@huawei.com>
Reviewed-by: Sabrina Dubroca <sd@queasysnail.net>
Link: https://lore.kernel.org/r/20230909081434.2324940-1-liujian56@huawei.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

tls: get cipher_name from cipher_desc in tls_set_sw_offload

tls_cipher_desc also contains the algorithm name needed by
crypto_alloc_aead, use it.

Finally, use get_cipher_desc to check if the cipher_type coming from
userspace is valid, and remove the cipher_type switch.

Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Link: https://lore.kernel.org/r/53d021d80138aa125a9cef4468aa5ce531975a7b.1692977948.git.sd@queasysnail.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: use tls_cipher_desc to access per-cipher crypto_info in tls_set_sw_offload

The crypto_info_* helpers allow us to fetch pointers into the
per-cipher crypto_info's data.

Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Link: https://lore.kernel.org/r/c23af110caf0af6b68de2f86c58064913e2e902a.1692977948.git.sd@queasysnail.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: use tls_cipher_desc to get per-cipher sizes in tls_set_sw_offload

We can get rid of some local variables, but we have to keep nonce_size
because tls1.3 uses nonce_size = 0 for all ciphers.

We can also drop the runtime sanity checks on iv/rec_seq/tag size,
since we have compile time checks on those values.

Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Link: https://lore.kernel.org/r/deed9c4430a62c31751a72b8c03ad66ffe710717.1692977948.git.sd@queasysnail.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net/tls: avoid TCP window full during ->read_sock()

When flushing the backlog after decoding a record we don't really
know how much data the caller want us to evaluate, so use INT_MAX
and 0 as arguments to tls_read_flush_backlog() to ensure we flush
at 128k of data. Otherwise we might be reading too much data and
trigger a TCP window full.

Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Link: https://lore.kernel.org/r/20230807071022.10091-1-hare@suse.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net/tls: implement ->read_sock()

Implement ->read_sock() function for use with nvme-tcp.

Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Cc: Boris Pismenny <boris.pismenny@gmail.com>
Link: https://lore.kernel.org/r/20230726191556.41714-7-hare@suse.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net/tls: split tls_rx_reader_lock

Split tls_rx_reader_{lock,unlock} into an 'acquire/release' and
the actual locking part.
With that we can use the tls_rx_reader_lock in situations where
the socket is already locked.

Suggested-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Link: https://lore.kernel.org/r/20230726191556.41714-6-hare@suse.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net/tls: handle MSG_EOR for tls_sw TX flow

tls_sw_sendmsg() already handles MSG_MORE, but bails
out on MSG_EOR.
Seeing that MSG_EOR is basically the opposite of
MSG_MORE this patch adds handling MSG_EOR by treating
it as the negation of MSG_MORE.
And erroring out if MSG_EOR is specified with MSG_MORE.

Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Link: https://lore.kernel.org/r/20230726191556.41714-2-hare@suse.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: Kill MSG_SENDPAGE_NOTLAST

Now that ->sendpage() has been removed, MSG_SENDPAGE_NOTLAST can be cleaned
up. Things were converted to use MSG_MORE instead, but the protocol
sendpage stubs still convert MSG_SENDPAGE_NOTLAST to MSG_MORE, which is now
unnecessary.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
cc: linux-afs@lists.infradead.org
cc: mptcp@lists.linux.dev
cc: rds-devel@oss.oracle.com
cc: tipc-discussion@lists.sourceforge.net
cc: virtualization@lists.linux-foundation.org
Link: https://lore.kernel.org/r/20230623225513.2732256-17-dhowells@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

sock: Remove ->sendpage*() in favour of sendmsg(MSG_SPLICE_PAGES)

Remove ->sendpage() and ->sendpage_locked(). sendmsg() with
MSG_SPLICE_PAGES should be used instead. This allows multiple pages and
multipage folios to be passed through.

Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Marc Kleine-Budde <mkl@pengutronix.de> # for net/can
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
cc: linux-afs@lists.infradead.org
cc: mptcp@lists.linux.dev
cc: rds-devel@oss.oracle.com
cc: tipc-discussion@lists.sourceforge.net
cc: virtualization@lists.linux-foundation.org
Link: https://lore.kernel.org/r/20230623225513.2732256-16-dhowells@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls/sw: Convert tls_sw_sendpage() to use MSG_SPLICE_PAGES

Convert tls_sw_sendpage() and tls_sw_sendpage_locked() to use sendmsg()
with MSG_SPLICE_PAGES rather than directly splicing in the pages itself.

[!] Note that tls_sw_sendpage_locked() appears to have the wrong locking
upstream. I think the caller will only hold the socket lock, but it
should hold tls_ctx->tx_lock too.

This allows ->sendpage() to be replaced by something that can handle
multiple multipage folios in a single transaction.

Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
cc: Chuck Lever <chuck.lever@oracle.com>
cc: Boris Pismenny <borisp@nvidia.com>
cc: John Fastabend <john.fastabend@gmail.com>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls/sw: Support MSG_SPLICE_PAGES

Make TLS's sendmsg() support MSG_SPLICE_PAGES. This causes pages to be
spliced from the source iterator if possible.

This allows ->sendpage() to be replaced by something that can handle
multiple multipage folios in a single transaction.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Chuck Lever <chuck.lever@oracle.com>
cc: Boris Pismenny <borisp@nvidia.com>
cc: John Fastabend <john.fastabend@gmail.com>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls/sw: Use splice_eof() to flush

Allow splice to end a TLS record after prematurely ending a splice/sendfile
due to getting an EOF condition (->splice_read() returned 0) after splice
had called TLS with a sendmsg() with MSG_MORE set when the user didn't set
MSG_MORE.

Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/r/CAHk-=wh=V579PDYvkpnTobCLGczbgxpMgGmmhqiTyE34Cpi5Gg@mail.gmail.com/
Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
cc: Chuck Lever <chuck.lever@oracle.com>
cc: Boris Pismenny <borisp@nvidia.com>
cc: John Fastabend <john.fastabend@gmail.com>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: Allow MSG_SPLICE_PAGES but treat it as normal sendmsg

Allow MSG_SPLICE_PAGES to be specified to sendmsg() but treat it as normal
sendmsg for now. This means the data will just be copied until
MSG_SPLICE_PAGES is handled.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Chuck Lever <chuck.lever@oracle.com>
cc: Boris Pismenny <borisp@nvidia.com>
cc: John Fastabend <john.fastabend@gmail.com>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: improve lockless access safety of tls_err_abort()

Most protos' poll() methods insert a memory barrier between
writes to sk_err and sk_error_report(). This dates back to
commit a4d258036ed9 ("tcp: Fix race in tcp_poll").

I guess we should do the same thing in TLS, tcp_poll() does
not hold the socket lock.

Fixes: 3c4d7559159b ("tls: kernel TLS support")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: strp: don't use GFP_KERNEL in softirq context

When receive buffer is small, or the TCP rx queue looks too
complicated to bother using it directly - we allocate a new
skb and copy data into it.

We already use sk->sk_allocation... but nothing actually
sets it to GFP_ATOMIC on the ->sk_data_ready() path.

Users of HW offload are far more likely to experience problems
due to scheduling while atomic. "Copy mode" is very rarely
triggered with SW crypto.

Fixes: 84c61fe1a75b ("tls: rx: do not use the standard strparser")
Tested-by: Shai Amiram <samiram@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: tls: avoid hanging tasks on the tx_lock

syzbot sent a hung task report and Eric explains that adversarial
receiver may keep RWIN at 0 for a long time, so we are not guaranteed
to make forward progress. Thread which took tx_lock and went to sleep
may not release tx_lock for hours. Use interruptible sleep where
possible and reschedule the work if it can't take the lock.

Testing: existing selftest passes

Reported-by: syzbot+9c0268252b8ef967c62e@syzkaller.appspotmail.com
Fixes: 79ffe6087e91 ("net/tls: add a TX lock")
Link: https://lore.kernel.org/all/000000000000e412e905f5b46201@google.com/
Cc: stable@vger.kernel.org # wait 4 weeks
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20230301002857.2101894-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: rx: fix return value for async crypto

Gaurav reports that TLS Rx is broken with async crypto
accelerators. The commit under fixes missed updating
the retval byte counting logic when updating how records
are stored. Even tho both before and after the change
'decrypted' was updated inside the main loop, it was
completely overwritten when processing the async
completions. Now that the rx_list only holds
non-zero-copy records we need to add, not overwrite.

Reported-and-bisected-by: Gaurav Jain <gaurav.jain@nxp.com>
Fixes: cbbdee9918a2 ("tls: rx: async: don't put async zc on the list")
Link: https://bugzilla.kernel.org/show_bug.cgi?id=217064
Tested-by: Gaurav Jain <gaurav.jain@nxp.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230227181201.1793772-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: Pass rec instead of aead_req into tls_encrypt_done

The function tls_encrypt_done only uses aead_req to get ahold of
the tls_rec object. So we could pass that in instead of aead_req
to simplify the code.

Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

tls: Remove completion function scaffolding

This patch removes the temporary scaffolding now that the comletion
function signature has been converted.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

tls: Only use data field in crypto completion function

The crypto_async_request passed to the completion is not guaranteed
to be the original request object. Only the data field can be relied
upon.

Fix this by storing the socket pointer with the AEAD request.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

net/sock: Introduce trace_sk_data_ready()

As suggested by Cong, introduce a tracepoint for all ->sk_data_ready()
callback implementations. For example:

<...>
iperf-609 [002] ..... 70.660425: sk_data_ready: family=2 protocol=6 func=sock_def_readable
iperf-609 [002] ..... 70.660436: sk_data_ready: family=2 protocol=6 func=sock_def_readable
<...>

Suggested-by: Cong Wang <cong.wang@bytedance.com>
Signed-off-by: Peilin Ye <peilin.ye@bytedance.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: tls_is_tx_ready() checked list_entry

tls_is_tx_ready() checks that list_first_entry() does not return NULL.
This condition can never happen. For empty lists, list_first_entry()
returns the list_entry() of the head, which is a type confusion.
Use list_first_entry_or_null() which returns NULL in case of empty
lists.

Fixes: a42055e8d2c3 ("net/tls: Add support for async encryption of records for performance")
Signed-off-by: Pietro Borrello <borrello@diag.uniroma1.it>
Link: https://lore.kernel.org/r/20230128-list-entry-null-check-tls-v1-1-525bbfe6f0d0@diag.uniroma1.it
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

bpf, sockmap: Fix missing BPF_F_INGRESS flag when using apply_bytes

When redirecting, we use sk_msg_to_ingress() to get the BPF_F_INGRESS
flag from the msg->flags. If apply_bytes is used and it is larger than
the current data being processed, sk_psock_msg_verdict() will not be
called when sendmsg() is called again. At this time, the msg->flags is 0,
and we lost the BPF_F_INGRESS flag.

So we need to save the BPF_F_INGRESS flag in sk_psock and use it when
redirection.

Fixes: 8934ce2fd081 ("bpf: sockmap redirect ingress support")
Signed-off-by: Pengcheng Yang <yangpc@wangsu.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Jakub Sitnicki <jakub@cloudflare.com>
Link: https://lore.kernel.org/bpf/1669718441-2654-3-git-send-email-yangpc@wangsu.com

net: tls: Add ARIA-GCM algorithm

RFC 6209 describes ARIA for TLS 1.2.
ARIA-128-GCM and ARIA-256-GCM are defined in RFC 6209.

This patch would offer performance increment and an opportunity for
hardware offload.

Benchmark results:
iperf-ssl are used.
CPU: intel i3-12100.

TLS(openssl-3.0-dev)
[ 3] 0.0- 1.0 sec 185 MBytes 1.55 Gbits/sec
[ 3] 1.0- 2.0 sec 186 MBytes 1.56 Gbits/sec
[ 3] 2.0- 3.0 sec 186 MBytes 1.56 Gbits/sec
[ 3] 3.0- 4.0 sec 186 MBytes 1.56 Gbits/sec
[ 3] 4.0- 5.0 sec 186 MBytes 1.56 Gbits/sec
[ 3] 0.0- 5.0 sec 927 MBytes 1.56 Gbits/sec
kTLS(aria-generic)
[ 3] 0.0- 1.0 sec 198 MBytes 1.66 Gbits/sec
[ 3] 1.0- 2.0 sec 194 MBytes 1.62 Gbits/sec
[ 3] 2.0- 3.0 sec 194 MBytes 1.63 Gbits/sec
[ 3] 3.0- 4.0 sec 194 MBytes 1.63 Gbits/sec
[ 3] 4.0- 5.0 sec 194 MBytes 1.62 Gbits/sec
[ 3] 0.0- 5.0 sec 974 MBytes 1.63 Gbits/sec
kTLS(aria-avx wirh GFNI)
[ 3] 0.0- 1.0 sec 632 MBytes 5.30 Gbits/sec
[ 3] 1.0- 2.0 sec 657 MBytes 5.51 Gbits/sec
[ 3] 2.0- 3.0 sec 657 MBytes 5.51 Gbits/sec
[ 3] 3.0- 4.0 sec 656 MBytes 5.50 Gbits/sec
[ 3] 4.0- 5.0 sec 656 MBytes 5.50 Gbits/sec
[ 3] 0.0- 5.0 sec 3.18 GBytes 5.47 Gbits/sec

Signed-off-by: Taehee Yoo <ap420073@gmail.com>
Reviewed-by: Vadim Fedorenko <vfedorenko@novek.ru>
Link: https://lore.kernel.org/r/20220925150033.24615-1-ap420073@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: rx: react to strparser initialization errors

Even though the normal strparser's init function has a return
value we got away with ignoring errors until now, as it only
validates the parameters and we were passing correct parameters.

tls_strp can fail to init on memory allocation errors, which
syzbot duly induced and reported.

Reported-by: syzbot+abd45eb849b05194b1b6@syzkaller.appspotmail.com
Fixes: 84c61fe1a75b ("tls: rx: do not use the standard strparser")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

iov_iter: advancing variants of iov_iter_get_pages{,_alloc}()

Most of the users immediately follow successful iov_iter_get_pages()
with advancing by the amount it had returned.

Provide inline wrappers doing that, convert trivial open-coded
uses of those.

BTW, iov_iter_get_pages() never returns more than it had been asked
to; such checks in cifs ought to be removed someday...

Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

tls: rx: fix the false positive warning

I went too far in the accessor conversion, we can't use tls_strp_msg()
after decryption because the message may not be ready. What we care
about on this path is that the output skb is detached, i.e. we didn't
somehow just turn around and used the input skb with its TCP data
still attached. So look at the anchor directly.

Fixes: 84c61fe1a75b ("tls: rx: do not use the standard strparser")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: rx: don't consider sock_rcvtimeo() cumulative

Eric indicates that restarting rcvtimeo on every wait may be fine.
I thought that we should consider it cumulative, and made
tls_rx_reader_lock() return the remaining timeo after acquiring
the reader lock.

tls_rx_rec_wait() gets its timeout passed in by value so it
does not keep track of time previously spent.

Make the lock waiting consistent with tls_rx_rec_wait() - don't
keep track of time spent.

Read the timeo fresh in tls_rx_rec_wait().
It's unclear to me why callers are supposed to cache the value.

Link: https://lore.kernel.org/all/CANn89iKcmSfWgvZjzNGbsrndmCch2HC_EPZ7qmGboDNaWoviNQ@mail.gmail.com/
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: rx: do not use the standard strparser

TLS is a relatively poor fit for strparser. We pause the input
every time a message is received, wait for a read which will
decrypt the message, start the parser, repeat. strparser is
built to delineate the messages, wrap them in individual skbs
and let them float off into the stack or a different socket.
TLS wants the data pages and nothing else. There's no need
for TLS to keep cloning (and occasionally skb_unclone()'ing)
the TCP rx queue.

This patch uses a pre-allocated skb and attaches the skbs
from the TCP rx queue to it as frags. TLS is careful never
to modify the input skb without CoW'ing / detaching it first.

Since we call TCP rx queue cleanup directly we also get back
the benefit of skb deferred free.

Overall this results in a 6% gain in my benchmarks.

Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: rx: device: keep the zero copy status with offload

The non-zero-copy path assumes a full skb with decrypted contents.
This means the device offload would have to CoW the data. Try
to keep the zero-copy status instead, copy the data to user space.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: rx: don't free the output in case of zero-copy

In the future we'll want to reuse the input skb in case of
zero-copy so we shouldn't always free darg.skb. Move the
freeing of darg.skb into the non-zc cases. All cases will
now free ctx->recv_pkt (inside let tls_rx_rec_done()).

Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: rx: factor SW handling out of tls_rx_one_record()

After recent changes the SW side of tls_rx_one_record() can
be nicely encapsulated in its own function. Move the pad handling
as well. This will be useful for ->zc handling in tls_decrypt_device().

Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: rx: wrap recv_pkt accesses in helpers

To allow for the logic to change later wrap accesses
which interrogate the input skb in helper functions.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: rx: release the sock lock on locking timeout

Eric reports we should release the socket lock if the entire
"grab reader lock" operation has failed. The callers assume
they don't have to release it or otherwise unwind.

Reported-by: Eric Dumazet <edumazet@google.com>
Reported-by: syzbot+16e72110feb2b653ef27@syzkaller.appspotmail.com
Fixes: 4cbc325ed6b4 ("tls: rx: allow only one reader at a time")
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20220720203701.2179034-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: rx: decrypt into a fresh skb

We currently CoW Rx skbs whenever we can't decrypt to a user
space buffer. The skbs can be enormous (64kB) and CoW does
a linear alloc which has a strong chance of failing under
memory pressure. Or even without, skb_cow_data() assumes
GFP_ATOMIC.

Allocate a new frag'd skb and decrypt into it. We finally
take advantage of the decrypted skb getting returned via
darg.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: async: don't put async zc on the list

The "zero-copy" path in SW TLS will engage either for no skbs or
for all but last. If the recvmsg parameters are right and the
socket can do ZC we'll ZC until the iterator can't fit a full
record at which point we'll decrypt one more record and copy
over the necessary bits to fill up the request.

The only reason we hold onto the ZC skbs which went thru the async
path until the end of recvmsg() is to count bytes. We need an accurate
count of zc'ed bytes so that we can calculate how much of the non-zc'd
data to copy. To allow freeing input skbs on the ZC path count only
how much of the list we'll need to consume.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: async: hold onto the input skb

Async crypto currently benefits from the fact that we decrypt
in place. When we allow input and output to be different skbs
we will have to hang onto the input while we move to the next
record. Clone the inputs and keep them on a list.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: async: adjust record geometry immediately

Async crypto TLS Rx currently waits for crypto to be done
in order to strip the TLS header and tailer. Simplify
the code by moving the pointers immediately, since only
TLS 1.2 is supported here there is no message padding.

This simplifies the decryption into a new skb in the next
patch as we don't have to worry about input vs output
skb in the decrypt_done() handler any more.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: return the decrypted skb via darg

Instead of using ctx->recv_pkt after decryption read the skb
from darg.skb. This moves the decision of what the "output skb"
is to the decrypt handlers. For now after decrypt handler returns
successfully ctx->recv_pkt is simply moved to darg.skb, but it
will change soon.

Note that tls_decrypt_sg() cannot clear the ctx->recv_pkt
because it gets called to re-encrypt (i.e. by the device offload).
So we need an awkward temporary if() in tls_rx_one_record().

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: read the input skb from ctx->recv_pkt

Callers always pass ctx->recv_pkt into decrypt_skb_update(),
and it propagates it to its callees. This may give someone
the false impression that those functions can accept any valid
skb containing a TLS record. That's not the case, the record
sequence number is read from the context, and they can only
take the next record coming out of the strp.

Let the functions get the skb from the context instead of
passing it in. This will also make it cleaner to return
a different skb than ctx->recv_pkt as the decrypted one
later on.

Since we're touching the definition of decrypt_skb_update()
use this as an opportunity to rename it.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: factor out device darg update

I already forgot to transform darg from input to output
semantics once on the NIC inline crypto fastpath. To
avoid this happening again create a device equivalent
of decrypt_internal(). A function responsible for decryption
and transforming darg.

While at it rename decrypt_internal() to a hopefully slightly
more meaningful name.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: remove the message decrypted tracking

We no longer allow a decrypted skb to remain linked to ctx->recv_pkt.
Anything on the list is decrypted, anything on ctx->recv_pkt needs
to be decrypted.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: don't keep decrypted skbs on ctx->recv_pkt

Detach the skb from ctx->recv_pkt after decryption is done,
even if we can't consume it.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: don't try to keep the skbs always on the list

I thought that having the skb either always on the ctx->rx_list
or ctx->recv_pkt will simplify the handling, as we would not
have to remember to flip it from one to the other on exit paths.

This became a little harder to justify with the fix for BPF
sockmaps. Subsequent changes will make the situation even worse.
Queue the skbs only when really needed.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: allow only one reader at a time

recvmsg() in TLS gets data from the skb list (rx_list) or fresh
skbs we read from TCP via strparser. The former holds skbs which were
already decrypted for peek or decrypted and partially consumed.

tls_wait_data() only notices appearance of fresh skbs coming out
of TCP (or psock). It is possible, if there is a concurrent call
to peek() and recv() that the peek() will move the data from input
to rx_list without recv() noticing. recv() will then read data out
of order or never wake up.

This is not a practical use case/concern, but it makes the self
tests less reliable. This patch solves the problem by allowing
only one reader in.

Because having multiple processes calling read()/peek() is not
normal avoid adding a lock and try to fast-path the single reader
case.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: add counter for NoPad violations

As discussed with Maxim add a counter for true NoPad violations.
This should help deployments catch unexpected padded records vs
just control records which always need re-encryption.

https: //lore.kernel.org/all/b111828e6ac34baad9f4e783127eba8344ac252d.camel@nvidia.com/
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: fix spelling of MIB

MIN -> MIB

Fixes: 88527790c079 ("tls: rx: add sockopt for enabling optimistic decrypt with TLS 1.3")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: rx: make tls_wait_data() return an recvmsg retcode

tls_wait_data() sets the return code as an output parameter
and always returns ctx->recv_pkt on success.

Return the error code directly and let the caller read the skb
from the context. Use positive return code to indicate ctx->recv_pkt
is ready.

While touching the definition of the function rename it.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: create an internal header

include/net/tls.h is getting a little long, and is probably hard
for driver authors to navigate. Split out the internals into a
header which will live under net/tls/. While at it move some
static inlines with a single user into the source files, add
a few tls_ prefixes and fix spelling of 'proccess'.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: rx: coalesce exit paths in tls_decrypt_sg()

Jump to the free() call, instead of having to remember
to free the memory in multiple places.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: rx: wrap decrypt params in a struct

The max size of iv + aad + tail is 22B. That's smaller
than a single sg entry (32B). Don't bother with the
memory packing, just create a struct which holds the
max size of those members.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: rx: always allocate max possible aad size for decrypt

AAD size is either 5 or 13. Really no point complicating
the code for the 8B of difference. This will also let us
turn the chunked up buffer into a sane struct.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: rx: periodically flush socket backlog

We continuously hold the socket lock during large reads and writes.
This may inflate RTT and negatively impact TCP performance.
Flush the backlog periodically. I tried to pick a flush period (128kB)
which gives significant benefit but the max Bps rate is not yet visibly
impacted.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: add sockopt for enabling optimistic decrypt with TLS 1.3

Since optimisitic decrypt may add extra load in case of retries
require socket owner to explicitly opt-in.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: support optimistic decrypt to user buffer with TLS 1.3

We currently don't support decrypt to user buffer with TLS 1.3
because we don't know the record type and how much padding
record contains before decryption. In practice data records
are by far most common and padding gets used rarely so
we can assume data record, no padding, and if we find out
that wasn't the case - retry the crypto in place (decrypt
to skb).

To safeguard from user overwriting content type and padding
before we can check it attach a 1B sg entry where last byte
of the record will land.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: don't include tail size in data_len

To make future patches easier to review make data_len
contain the length of the data, without the tail.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Revert "tls: rx: move counting TlsDecryptErrors for sync"

This reverts commit 284b4d93daee56dff3e10029ddf2e03227f50dbf.
When using TLS device offload and coming from tls_device_reencrypt()
flow, -EBADMSG error in tls_do_decryption() should not be counted
towards the TLSTlsDecryptError counter.

Move the counter increase back to the decrypt_internal() call site in
decrypt_skb_update().
This also fixes an issue where:
if (n_sgin < 1)
return -EBADMSG;

Errors in decrypt_internal() were not counted after the cited patch.

Fixes: 284b4d93daee ("tls: rx: move counting TlsDecryptErrors for sync")
Cc: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: tls: fix messing up lists when bpf enabled

Artem points out that skb may try to take over the skb and
queue it to its own list. Unlink the skb before calling out.

Fixes: b1a2c1786330 ("tls: rx: clear ctx->recv_pkt earlier")
Reported-by: Artem Savkov <asavkov@redhat.com>
Tested-by: Artem Savkov <asavkov@redhat.com>
Link: https://lore.kernel.org/r/20220518205644.2059468-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: tls: fix async vs NIC crypto offload

When NIC takes care of crypto (or the record has already
been decrypted) we forget to update darg->async. ->async
is supposed to mean whether record is async capable on
input and whether record has been queued for async crypto
on output.

Reported-by: Gal Pressman <gal@nvidia.com>
Fixes: 3547a1f9d988 ("tls: rx: use async as an in-out argument")
Tested-by: Gal Pressman <gal@nvidia.com>
Link: https://lore.kernel.org/r/20220425233309.344858-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: generalize skb freeing deferral to per-cpu lists

Logic added in commit f35f821935d8 ("tcp: defer skb freeing after socket
lock is released") helped bulk TCP flows to move the cost of skbs
frees outside of critical section where socket lock was held.

But for RPC traffic, or hosts with RFS enabled, the solution is far from
being ideal.

For RPC traffic, recvmsg() has to return to user space right after
skb payload has been consumed, meaning that BH handler has no chance
to pick the skb before recvmsg() thread. This issue is more visible
with BIG TCP, as more RPC fit one skb.

For RFS, even if BH handler picks the skbs, they are still picked
from the cpu on which user thread is running.

Ideally, it is better to free the skbs (and associated page frags)
on the cpu that originally allocated them.

This patch removes the per socket anchor (sk->defer_list) and
instead uses a per-cpu list, which will hold more skbs per round.

This new per-cpu list is drained at the end of net_action_rx(),
after incoming packets have been processed, to lower latencies.

In normal conditions, skbs are added to the per-cpu list with
no further action. In the (unlikely) cases where the cpu does not
run net_action_rx() handler fast enough, we use an IPI to raise
NET_RX_SOFTIRQ on the remote cpu.

Also, we do not bother draining the per-cpu list from dev_cpu_dead()
This is because skbs in this list have no requirement on how fast
they should be freed.

Note that we can add in the future a small per-cpu cache
if we see any contention on sd->defer_lock.

Tested on a pair of hosts with 100Gbit NIC, RFS enabled,
and /proc/sys/net/ipv4/tcp_rmem[2] tuned to 16MB to work around
page recycling strategy used by NIC driver (its page pool capacity
being too small compared to number of skbs/pages held in sockets
receive queues)

Note that this tuning was only done to demonstrate worse
conditions for skb freeing for this particular test.
These conditions can happen in more general production workload.

10 runs of one TCP_STREAM flow

Before:
Average throughput: 49685 Mbit.

Kernel profiles on cpu running user thread recvmsg() show high cost for
skb freeing related functions (*)

57.81% [kernel] [k] copy_user_enhanced_fast_string
(*) 12.87% [kernel] [k] skb_release_data
(*) 4.25% [kernel] [k] __free_one_page
(*) 3.57% [kernel] [k] __list_del_entry_valid
1.85% [kernel] [k] __netif_receive_skb_core
1.60% [kernel] [k] __skb_datagram_iter
(*) 1.59% [kernel] [k] free_unref_page_commit
(*) 1.16% [kernel] [k] __slab_free
1.16% [kernel] [k] _copy_to_iter
(*) 1.01% [kernel] [k] kfree
(*) 0.88% [kernel] [k] free_unref_page
0.57% [kernel] [k] ip6_rcv_core
0.55% [kernel] [k] ip6t_do_table
0.54% [kernel] [k] flush_smp_call_function_queue
(*) 0.54% [kernel] [k] free_pcppages_bulk
0.51% [kernel] [k] llist_reverse_order
0.38% [kernel] [k] process_backlog
(*) 0.38% [kernel] [k] free_pcp_prepare
0.37% [kernel] [k] tcp_recvmsg_locked
(*) 0.37% [kernel] [k] __list_add_valid
0.34% [kernel] [k] sock_rfree
0.34% [kernel] [k] _raw_spin_lock_irq
(*) 0.33% [kernel] [k] __page_cache_release
0.33% [kernel] [k] tcp_v6_rcv
(*) 0.33% [kernel] [k] __put_page
(*) 0.29% [kernel] [k] __mod_zone_page_state
0.27% [kernel] [k] _raw_spin_lock

After patch:
Average throughput: 73076 Mbit.

Kernel profiles on cpu running user thread recvmsg() looks better:

81.35% [kernel] [k] copy_user_enhanced_fast_string
1.95% [kernel] [k] _copy_to_iter
1.95% [kernel] [k] __skb_datagram_iter
1.27% [kernel] [k] __netif_receive_skb_core
1.03% [kernel] [k] ip6t_do_table
0.60% [kernel] [k] sock_rfree
0.50% [kernel] [k] tcp_v6_rcv
0.47% [kernel] [k] ip6_rcv_core
0.45% [kernel] [k] read_tsc
0.44% [kernel] [k] _raw_spin_lock_irqsave
0.37% [kernel] [k] _raw_spin_lock
0.37% [kernel] [k] native_irq_return_iret
0.33% [kernel] [k] __inet6_lookup_established
0.31% [kernel] [k] ip6_protocol_deliver_rcu
0.29% [kernel] [k] tcp_rcv_established
0.29% [kernel] [k] llist_reverse_order

v2: kdoc issue (kernel bots)
do not defer if (alloc_cpu == smp_processor_id()) (Paolo)
replace the sk_buff_head with a single-linked list (Jakub)
add a READ_ONCE()/WRITE_ONCE() for the lockless read of sd->defer_list

Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Link: https://lore.kernel.org/r/20220422201237.416238-1-eric.dumazet@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: rx: only copy IV from the packet for TLS 1.2

TLS 1.3 and ChaChaPoly don't carry IV in the packet.
The code before this change would copy out iv_size
worth of whatever followed the TLS header in the packet
and then for TLS 1.3 | ChaCha overwrite that with
the sequence number. Waste of cycles especially
with TLS 1.2 being close to dead and TLS 1.3 being
the common case.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: use MAX_IV_SIZE for allocations

IVs are 8 or 16 bytes, no point reading out the exact value
for quantities this small.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: use async as an in-out argument

Propagating EINPROGRESS thru multiple layers of functions is
error prone. Use darg->async as an in/out argument, like we
use darg->zc today. On input it tells the code if async is
allowed, on output if it took place.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: return the already-copied data on crypto error

async crypto handler will report the socket error no need
to report it again. We can, however, let the data we already
copied be reported to user space but we need to make sure
the error will be reported next time around.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: treat process_rx_list() errors as transient

process_rx_list() only fails if it can't copy data to user
space. There is no point recording the error onto sk->sk_err
or giving up on the data which was read partially. Treat
the return value like a normal socket partial read.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: assume crypto always calls our callback

If crypto didn't always invoke our callback for async
we'd not be clearing skb->sk and would crash in the
skb core when freeing it. This if must be dead code.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: don't handle TLS 1.3 in the async crypto callback

Async crypto never worked with TLS 1.3 and was explicitly disabled in
commit 8497ded2d16c ("net/tls: Disable async decrytion for tls1.3").
There's no need for us to handle TLS 1.3 padding in the async cb.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: move counting TlsDecryptErrors for sync

Move counting TlsDecryptErrors to tls_do_decryption()
where differences between sync and async crypto are
reconciled.

No functional changes, this code just always gave
me a pause.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: reuse leave_on_list label for psock

The code is identical, we can save a few LoC.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: consistently use unlocked accessors for rx_list

rx_list is protected by the socket lock, no need to take
the built-in spin lock on accesses.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: remove noblock parameter from recvmsg() entities

The internal recvmsg() functions have two parameters 'flags' and 'noblock'
that were merged inside skb_recv_datagram(). As a follow up patch to commit
f4b41f062c42 ("net: remove noblock parameter from skb_recv_datagram()")
this patch removes the separate 'noblock' parameter for recvmsg().

Analogue to the referenced patch for skb_recv_datagram() the 'flags' and
'noblock' parameters are unnecessarily split up with e.g.

err = sk->sk_prot->recvmsg(sk, msg, size, flags & MSG_DONTWAIT,
flags & ~MSG_DONTWAIT, &addr_len);

or in

err = INDIRECT_CALL_2(sk->sk_prot->recvmsg, tcp_recvmsg, udp_recvmsg,
sk, msg, size, flags & MSG_DONTWAIT,
flags & ~MSG_DONTWAIT, &addr_len);

instead of simply using only flags all the time and check for MSG_DONTWAIT
where needed (to preserve for the formerly separated no(n)block condition).

Signed-off-by: Oliver Hartkopp <socketcan@hartkopp.net>
Link: https://lore.kernel.org/r/20220411124955.154876-1-socketcan@hartkopp.net
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

tls: rx: jump out for cases which need to leave skb on list

The current invese logic is harder to follow (and adds extra
tests to the fast path). We have to enumerate all cases which
need to keep the skb before consuming it. It's simpler to
jump out of the full record flow as we detect those cases.

This makes it clear that partial consumption and peek can
only reach end of the function thru the !zc case so move
the code up there.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: clear ctx->recv_pkt earlier

Whatever we do in the loop the skb should not remain on as
ctx->recv_pkt afterwards. We can clear that pointer and
restart strparser earlier.

This adds overhead of extra linking and unlinking to rx_list
but that's not large (upcoming change will switch to unlocked
skb list operations).

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: inline consuming the skb at the end of the loop

tls_sw_advance_skb() always consumes the skb at the end of the loop.

To fall here the following must be true:

!async && !is_peek && !retain_skb
retain_skb => !zc && rxm->full_len > len
# but non-full record implies !zc, so above can be simplified as
retain_skb => rxm->full_len > len

!async && !is_peek && !(rxm->full_len > len)
!async && !is_peek && rxm->full_len <= len

tls_sw_advance_skb() returns false if len < rxm->full_len
which can't be true given conditions above.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: pull most of zc check out of the loop

Most of the conditions deciding if zero-copy can be used
do not change throughout the iterations, so pre-calculate
them.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: don't track the async count

We track both if the last record was handled by async crypto
and how many records were async. This is not necessary. We
implicitly assume once crypto goes async it will stay that
way, otherwise we'd reorder records. So just track if we're
in async mode, the exact number of records is not necessary.

This change also forces us into "async" mode more consistently
in case crypto ever decided to interleave async and sync.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: don't handle async in tls_sw_advance_skb()

tls_sw_advance_skb() caters to the async case when skb argument
is NULL. In that case it simply unpauses the strparser.

These are surprising semantics to a person reading the code,
and result in higher LoC, so inline the __strp_unpause and
only call tls_sw_advance_skb() when we actually move past
an skb.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: factor out writing ContentType to cmsg

cmsg can be filled in during rx_list processing or normal
receive. Consolidate the code.

We don't need to keep the boolean to track if the cmsg was
created. 0 is an invalid content type.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: simplify async wait

Since we are protected from async completions by decrypt_compl_lock
we can drop the async_notify and reinit the completion before we
start waiting.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: wrap decryption arguments in a structure

We pass zc as a pointer to bool a few functions down as an in/out
argument. This is error prone since C will happily evalue a pointer
as a boolean (IOW forgetting *zc and writing zc leads to loss of
developer time..). Wrap the arguments into a structure.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: don't report text length from the bowels of decrypt

We plumb pointer to chunk all the way to the decryption method.
It's set to the length of the text when decrypt_skb_update()
returns.

I think the code is written this way because original TLS
implementation passed &chunk to zerocopy_from_iter() and this
was carried forward as the code gotten more complex, without
any refactoring.

The fix for peek() introduced a new variable - to_decrypt
which for all practical purposes is what chunk is going to
get set to. Spare ourselves the pointer passing, use to_decrypt.

Use this opportunity to clean things up a little further.

Note that chunk / to_decrypt was mostly needed for the async
path, since the sync path would access rxm->full_len (decryption
transforms full_len from record size to text size). Use the
right source of truth more explicitly.

We have three cases:
- async - it's TLS 1.2 only, so chunk == to_decrypt, but we
need the min() because to_decrypt is a whole record
and we don't want to underflow len. Note that we can't
handle partial record by falling back to sync as it
would introduce reordering against records in flight.
- zc - again, TLS 1.2 only for now, so chunk == to_decrypt,
we don't do zc if len < to_decrypt, no need to check again.
- normal - it already handles chunk > len, we can factor out the
assignment to rxm->full_len and share it with zc.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: drop unnecessary arguments from tls_setup_from_iter()

sk is unused, remove it to make it clear the function
doesn't poke at the socket.

size_used is always 0 on input and @length on success.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: hw: rx: use return value of tls_device_decrypted() to carry status

Instead of tls_device poking into internals of the message
return 1 from tls_device_decrypted() if the device handled
the decryption.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: refactor decrypt_skb_update()

Use early return and a jump label to remove two indentation levels.
No functional changes.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: don't issue wake ups when data is decrypted

We inform the applications that data is available when
the record is received. Decryption happens inline inside
recvmsg or splice call. Generating another wakeup inside
the decryption handler seems pointless as someone must
be actively reading the socket if we are executing this
code.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: replace 'back' with 'offset'

The padding length TLS 1.3 logic is searching for content_type from
the end of text. IMHO the code is easier to parse if we calculate
offset and decrement it rather than try to maintain positive offset
from the end of the record called "back".

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: use a define for tag length

TLS 1.3 has to strip padding, and it starts out 16 bytes
from the end of the record. Make it clear this is because
of the auth tag.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: init decrypted status in tls_read_size()

We set the record type in tls_read_size(), can as well init
the tlm->decrypted field there.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: don't store the decryption status in socket context

Similar justification to previous change, the information
about decryption status belongs in the skb.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: don't store the record type in socket context

Original TLS implementation was handling one record at a time.
It stashed the type of the record inside tls context (per socket
structure) for convenience. When async crypto support was added
[1] the author had to use skb->cb to store the type per-message.

The use of skb->cb overlaps with strparser, however, so a hybrid
approach was taken where type is stored in context while parsing
(since we parse a message at a time) but once parsed its copied
to skb->cb.

Recently a workaround for sockmaps [2] exposed the previously
private struct _strp_msg and started a trend of adding user
fields directly in strparser's header. This is cleaner than
storing information about an skb in the context.

This change is not strictly necessary, but IMHO the ownership
of the context field is confusing. Information naturally
belongs to the skb.

[1] commit 94524d8fc965 ("net/tls: Add support for async decryption of tls records")
[2] commit b2c4618162ec ("bpf, sockmap: sk_skb data_end access incorrect when src_reg = dst_reg")

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: drop pointless else after goto

Pointless else branch after goto makes the code harder to refactor
down the line.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: rx: jump to a more appropriate label

'recv_end:' checks num_async and decrypted, and is then followed
by the 'end' label. Since we know that decrypted and num_async
are 0 at the start we can jump to 'end'.

Move the init of decrypted and num_async to let the compiler
catch if I'm wrong.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: fix slab-out-of-bounds bug in decrypt_internal

The memory size of tls_ctx->rx.iv for AES128-CCM is 12 setting in
tls_set_sw_offload(). The return value of crypto_aead_ivsize()
for "ccm(aes)" is 16. So memcpy() require 16 bytes from 12 bytes
memory space will trigger slab-out-of-bounds bug as following:

==================================================================
BUG: KASAN: slab-out-of-bounds in decrypt_internal+0x385/0xc40 [tls]
Read of size 16 at addr ffff888114e84e60 by task tls/10911

Call Trace:
<TASK>
dump_stack_lvl+0x34/0x44
print_report.cold+0x5e/0x5db
? decrypt_internal+0x385/0xc40 [tls]
kasan_report+0xab/0x120
? decrypt_internal+0x385/0xc40 [tls]
kasan_check_range+0xf9/0x1e0
memcpy+0x20/0x60
decrypt_internal+0x385/0xc40 [tls]
? tls_get_rec+0x2e0/0x2e0 [tls]
? process_rx_list+0x1a5/0x420 [tls]
? tls_setup_from_iter.constprop.0+0x2e0/0x2e0 [tls]
decrypt_skb_update+0x9d/0x400 [tls]
tls_sw_recvmsg+0x3c8/0xb50 [tls]

Allocated by task 10911:
kasan_save_stack+0x1e/0x40
__kasan_kmalloc+0x81/0xa0
tls_set_sw_offload+0x2eb/0xa20 [tls]
tls_setsockopt+0x68c/0x700 [tls]
__sys_setsockopt+0xfe/0x1b0

Replace the crypto_aead_ivsize() with prot->iv_size + prot->salt_size
when memcpy() iv value in TLS_1_3_VERSION scenario.

Fixes: f295b3ae9f59 ("net/tls: Add support of AES128-CCM based ciphers")
Signed-off-by: Ziyang Xuan <william.xuanziyang@huawei.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: cap the output scatter list to something reasonable

TLS recvmsg() passes user pages as destination for decrypt.
The decrypt operation is repeated record by record, each
record being 16kB, max. TLS allocates an sg_table and uses
iov_iter_get_pages() to populate it with enough pages to
fit the decrypted record.

Even though we decrypt a single message at a time we size
the sg_table based on the entire length of the iovec.
This leads to unnecessarily large allocations, risking
triggering OOM conditions.

Use iov_iter_truncate() / iov_iter_reexpand() to construct
a "capped" version of iov_iter_npages(). Alternatively we
could parametrize iov_iter_npages() to take the size as
arg instead of using i->count, or do something else..

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Fix another skb memory leak when running kTLS traffic

This patch is a followup to
commit ffef737fd037 ("net/tls: Fix skb memory leak when running kTLS traffic")

Which was missing another sk_defer_free_flush() call in
tls_sw_splice_read().

Fixes: f35f821935d8 ("tcp: defer skb freeing after socket lock is released")
Signed-off-by: Gal Pressman <gal@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Fix skb memory leak when running kTLS traffic

The cited Fixes commit introduced a memory leak when running kTLS
traffic (with/without hardware offloads).
I'm running nginx on the server side and wrk on the client side and get
the following:

unreferenced object 0xffff8881935e9b80 (size 224):
comm "softirq", pid 0, jiffies 4294903611 (age 43.204s)
hex dump (first 32 bytes):
80 9b d0 36 81 88 ff ff 00 00 00 00 00 00 00 00 ...6............
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
backtrace:
[<00000000efe2a999>] build_skb+0x1f/0x170
[<00000000ef521785>] mlx5e_skb_from_cqe_mpwrq_linear+0x2bc/0x610 [mlx5_core]
[<00000000945d0ffe>] mlx5e_handle_rx_cqe_mpwrq+0x264/0x9e0 [mlx5_core]
[<00000000cb675b06>] mlx5e_poll_rx_cq+0x3ad/0x17a0 [mlx5_core]
[<0000000018aac6a9>] mlx5e_napi_poll+0x28c/0x1b60 [mlx5_core]
[<000000001f3369d1>] __napi_poll+0x9f/0x560
[<00000000cfa11f72>] net_rx_action+0x357/0xa60
[<000000008653b8d7>] __do_softirq+0x282/0x94e
[<00000000644923c6>] __irq_exit_rcu+0x11f/0x170
[<00000000d4085f8f>] irq_exit_rcu+0xa/0x20
[<00000000d412fef4>] common_interrupt+0x7d/0xa0
[<00000000bfb0cebc>] asm_common_interrupt+0x1e/0x40
[<00000000d80d0890>] default_idle+0x53/0x70
[<00000000f2b9780e>] default_idle_call+0x8c/0xd0
[<00000000c7659e15>] do_idle+0x394/0x450

I'm not familiar with these areas of the code, but I've added this
sk_defer_free_flush() to tls_sw_recvmsg() based on a hunch and it
resolved the issue.

Fixes: f35f821935d8 ("tcp: defer skb freeing after socket lock is released")
Signed-off-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20220102081253.9123-1-gal@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net/tls: simplify the tls_set_sw_offload function

Assigning crypto_info variables in advance can simplify the logic
of accessing value and move related local variables to a smaller
scope.

Signed-off-by: Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Fix authentication failure in CCM mode

When the TLS cipher suite uses CCM mode, including AES CCM and
SM4 CCM, the first byte of the B0 block is flags, and the real
IV starts from the second byte. The XOR operation of the IV and
rec_seq should be skip this byte, that is, add the iv_offset.

Fixes: f295b3ae9f59 ("net/tls: Add support of AES128-CCM based ciphers")
Signed-off-by: Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
Cc: Vakul Garg <vakul.garg@nxp.com>
Cc: stable@vger.kernel.org # v5.2+
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: splice_read: fix accessing pre-processed records

recvmsg() will put peek()ed and partially read records onto the rx_list.
splice_read() needs to consult that list otherwise it may miss data.
Align with recvmsg() and also put partially-read records onto rx_list.
tls_sw_advance_skb() is pretty pointless now and will be removed in
net-next.

Fixes: 692d7b5d1f91 ("tls: Fix recvmsg() to be able to peek across multiple records")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

tls: splice_read: fix record type check

We don't support splicing control records. TLS 1.3 changes moved
the record type check into the decrypt if(). The skb may already
be decrypted and still be an alert.

Note that decrypt_skb_update() is idempotent and updates ctx->decrypted
so the if() is pointless.

Reorder the check for decryption errors with the content type check
while touching them. This part is not really a bug, because if
decryption failed in TLS 1.3 content type will be DATA, and for
TLS 1.2 it will be correct. Nevertheless its strange to touch output
before checking if the function has failed.

Fixes: fedf201e1296 ("net: tls: Refactor control message handling on recv")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net/tls: support SM4 CCM algorithm

The IV of CCM mode has special requirements, this patch supports CCM
mode of SM4 algorithm.

Signed-off-by: Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: support SM4 GCM/CCM algorithm

The RFC8998 specification defines the use of the ShangMi algorithm
cipher suites in TLS 1.3, and also supports the GCM/CCM mode using
the SM4 algorithm.

Signed-off-by: Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Fix flipped sign in async_wait.err assignment

sk->sk_err contains a positive number, yet async_wait.err wants the
opposite. Fix the missed sign flip, which Jakub caught by inspection.

Fixes: a42055e8d2c3 ("net/tls: Add support for async encryption of records for performance")
Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Daniel Jordan <daniel.m.jordan@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Fix flipped sign in tls_err_abort() calls

sk->sk_err appears to expect a positive value, a convention that ktls
doesn't always follow and that leads to memory corruption in other code.
For instance,

[kworker]
tls_encrypt_done(..., err=<negative error from crypto request>)
tls_err_abort(.., err)
sk->sk_err = err;

[task]
splice_from_pipe_feed
...
tls_sw_do_sendpage
if (sk->sk_err) {
ret = -sk->sk_err; // ret is positive

splice_from_pipe_feed (continued)
ret = actor(...) // ret is still positive and interpreted as bytes
// written, resulting in underflow of buf->len and
// sd->len, leading to huge buf->offset and bogus
// addresses computed in later calls to actor()

Fix all tls_err_abort() callers to pass a negative error code
consistently and centralize the error-prone sign flip there, throwing in
a warning to catch future misuse and uninlining the function so it
really does only warn once.

Cc: stable@vger.kernel.org
Fixes: c46234ebb4d1e ("tls: RX path for ktls")
Reported-by: syzbot+b187b77c8474f9648fae@syzkaller.appspotmail.com
Signed-off-by: Daniel Jordan <daniel.m.jordan@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: Rename ->stream_memory_read to ->sock_is_readable

The proto ops ->stream_memory_read() is currently only used
by TCP to check whether psock queue is empty or not. We need
to rename it before reusing it for non-TCP protocols, and
adjust the exsiting users accordingly.

Signed-off-by: Cong Wang <cong.wang@bytedance.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20211008203306.37525-2-xiyou.wangcong@gmail.com

tls: prevent oversized sendfile() hangs by ignoring MSG_MORE

We got multiple reports that multi_chunk_sendfile test
case from tls selftest fails. This was sort of expected,
as the original fix was never applied (see it in the first
Link:). The test in question uses sendfile() with count
larger than the size of the underlying file. This will
make splice set MSG_MORE on all sendpage calls, meaning
TLS will never close and flush the last partial record.

Eric seem to have addressed a similar problem in
commit 35f9c09fe9c7 ("tcp: tcp_sendpages() should call tcp_push() once")
by introducing MSG_SENDPAGE_NOTLAST. Unlike MSG_MORE
MSG_SENDPAGE_NOTLAST is not set on the last call
of a "pipefull" of data (PIPE_DEF_BUFFERS == 16,
so every 16 pages or whenever we run out of data).

Having a break every 16 pages should be fine, TLS
can pack exactly 4 pages into a record, so for
aligned reads there should be no difference,
unaligned may see one extra record per sendpage().

Sticking to TCP semantics seems preferable to modifying
splice, but we can revisit it if real life scenarios
show a regression.

Reported-by: Vadim Fedorenko <vfedorenko@novek.ru>
Reported-by: Seth Forshee <seth.forshee@canonical.com>
Link: https://lore.kernel.org/netdev/1591392508-14592-1-git-send-email-pooja.trivedi@stackpath.com/
Fixes: 3c4d7559159b ("tls: kernel TLS support")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Tested-by: Seth Forshee <seth.forshee@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls splice: remove inappropriate flags checking for MSG_PEEK

In function tls_sw_splice_read, before call tls_sw_advance_skb
it checks likely(!(flags & MSG_PEEK)), while MSG_PEEK is used
for recvmsg, splice supports SPLICE_F_NONBLOCK, SPLICE_F_MOVE,
SPLICE_F_MORE, should remove this checking.

Signed-off-by: Jim Ma <majinjing3@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls splice: check SPLICE_F_NONBLOCK instead of MSG_DONTWAIT

In tls_sw_splice_read, checkout MSG_* is inappropriate, should use
SPLICE_*, update tls_wait_data to accept nonblock arguments instead
of flags for recvmsg and splice.

Fixes: c46234ebb4d1 ("tls: RX path for ktls")
Signed-off-by: Jim Ma <majinjing3@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

skmsg: Extract __tcp_bpf_recvmsg() and tcp_bpf_wait_data()

Although these two functions are only used by TCP, they are not
specific to TCP at all, both operate on skmsg and ingress_msg,
so fit in net/core/skmsg.c very well.

And we will need them for non-TCP, so rename and move them to
skmsg.c and export them to modules.

Signed-off-by: Cong Wang <cong.wang@bytedance.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20210331023237.41094-13-xiyou.wangcong@gmail.com

net/tls: add CHACHA20-POLY1305 configuration

Add ChaCha-Poly specific configuration code.

Signed-off-by: Vadim Fedorenko <vfedorenko@novek.ru>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net/tls: add CHACHA20-POLY1305 specific behavior

RFC 7905 defines special behavior for ChaCha-Poly TLS sessions.
The differences are in the calculation of nonce and the absence
of explicit IV. This behavior is like TLSv1.3 partly.

Signed-off-by: Vadim Fedorenko <vfedorenko@novek.ru>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net/tls: make inline helpers protocol-aware

Inline functions defined in tls.h have a lot of AES-specific
constants. Remove these constants and change argument to struct
tls_prot_info to have an access to cipher type in later patches

Signed-off-by: Vadim Fedorenko <vfedorenko@novek.ru>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net/tls: missing received data after fast remote close

In case when tcp socket received FIN after some data and the
parser haven't started before reading data caller will receive
an empty buffer. This behavior differs from plain TCP socket and
leads to special treating in user-space.
The flow that triggers the race is simple. Server sends small
amount of data right after the connection is configured to use TLS
and closes the connection. In this case receiver sees TLS Handshake
data, configures TLS socket right after Change Cipher Spec record.
While the configuration is in process, TCP socket receives small
Application Data record, Encrypted Alert record and FIN packet. So
the TCP socket changes sk_shutdown to RCV_SHUTDOWN and sk_flag with
SK_DONE bit set. The received data is not parsed upon arrival and is
never sent to user-space.

Patch unpauses parser directly if we have unparsed data in tcp
receive queue.

Fixes: fcf4793e278e ("tls: check RCV_SHUTDOWN in tls_wait_data")
Signed-off-by: Vadim Fedorenko <vfedorenko@novek.ru>
Link: https://lore.kernel.org/r/1605801588-12236-1-git-send-email-vfedorenko@novek.ru
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net/tls: fix corrupted data in recvmsg

If tcp socket has more data than Encrypted Handshake Message then
tls_sw_recvmsg will try to decrypt next record instead of returning
full control message to userspace as mentioned in comment. The next
message - usually Application Data - gets corrupted because it uses
zero copy for decryption that's why the data is not stored in skb
for next iteration. Revert check to not decrypt next record if
current is not Application Data.

Fixes: 692d7b5d1f91 ("tls: Fix recvmsg() to be able to peek across multiple records")
Signed-off-by: Vadim Fedorenko <vfedorenko@novek.ru>
Link: https://lore.kernel.org/r/1605413760-21153-1-git-send-email-vfedorenko@novek.ru
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net/tls: race causes kernel panic

BUG: kernel NULL pointer dereference, address: 00000000000000b8
#PF: supervisor read access in kernel mode
#PF: error_code(0x0000) - not-present page
PGD 80000008b6fef067 P4D 80000008b6fef067 PUD 8b6fe6067 PMD 0
Oops: 0000 [#1] SMP PTI
CPU: 12 PID: 23871 Comm: kworker/12:80 Kdump: loaded Tainted: G S
5.9.0-rc3+ #1
Hardware name: Supermicro X10SRA-F/X10SRA-F, BIOS 2.1 03/29/2018
Workqueue: events tx_work_handler [tls]
RIP: 0010:tx_work_handler+0x1b/0x70 [tls]
Code: dc fe ff ff e8 16 d4 a3 f6 66 0f 1f 44 00 00 0f 1f 44 00 00 55 53 48 8b
6f 58 48 8b bd a0 04 00 00 48 85 ff 74 1c 48 8b 47 28 <48> 8b 90 b8 00 00 00 83
e2 02 75 0c f0 48 0f ba b0 b8 00 00 00 00
RSP: 0018:ffffa44ace61fe88 EFLAGS: 00010286
RAX: 0000000000000000 RBX: ffff91da9e45cc30 RCX: dead000000000122
RDX: 0000000000000001 RSI: ffff91da9e45cc38 RDI: ffff91d95efac200
RBP: ffff91da133fd780 R08: 0000000000000000 R09: 000073746e657665
R10: 8080808080808080 R11: 0000000000000000 R12: ffff91dad7d30700
R13: ffff91dab6561080 R14: 0ffff91dad7d3070 R15: ffff91da9e45cc38
FS: 0000000000000000(0000) GS:ffff91dad7d00000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00000000000000b8 CR3: 0000000906478003 CR4: 00000000003706e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
process_one_work+0x1a7/0x370
worker_thread+0x30/0x370
? process_one_work+0x370/0x370
kthread+0x114/0x130
? kthread_park+0x80/0x80
ret_from_fork+0x22/0x30

tls_sw_release_resources_tx() waits for encrypt_pending, which
can have race, so we need similar changes as in commit
0cada33241d9de205522e3858b18e506ca5cce2c here as well.

Fixes: a42055e8d2c3 ("net/tls: Add support for async encryption of records for performance")
Signed-off-by: Rohit Maheshwari <rohitm@chelsio.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: allow MSG_CMSG_COMPAT in sendmsg

Trying to use ktls on a system with 32-bit userspace and 64-bit kernel
results in a EOPNOTSUPP message during sendmsg:

setsockopt(3, SOL_TLS, TLS_TX, …, 40) = 0
sendmsg(3, …, msg_flags=0}, 0) = -1 EOPNOTSUPP (Operation not supported)

The tls_sw implementation does strict flag checking and does not allow
the MSG_CMSG_COMPAT flag, which is set if the message comes in through
the compat syscall.

This patch adds MSG_CMSG_COMPAT to the flag check to allow the usage of
the TLS SW implementation on systems using the compat syscall path.

Note that the same check is present in the sendmsg path for the TLS
device implementation, however the flag hasn't been added there for lack
of testing hardware.

Signed-off-by: Rouven Czerwinski <r.czerwinski@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

treewide: Remove uninitialized_var() usage

Using uninitialized_var() is dangerous as it papers over real bugs[1]
(or can in the future), and suppresses unrelated compiler warnings
(e.g. "unused variable"). If the compiler thinks it is uninitialized,
either simply initialize the variable or make compiler changes.

In preparation for removing[2] the[3] macro[4], remove all remaining
needless uses with the following script:

git grep '\buninitialized_var\b' | cut -d: -f1 | sort -u | \
xargs perl -pi -e \
's/\buninitialized_var\(([^\)]+)\)/\1/g;
s:\s*/\* (GCC be quiet|to make compiler happy) \*/$::g;'

drivers/video/fbdev/riva/riva_hw.c was manually tweaked to avoid
pathological white-space.

No outstanding warnings were found building allmodconfig with GCC 9.3.0
for x86_64, i386, arm64, arm, powerpc, powerpc64le, s390x, mips, sparc64,
alpha, and m68k.

[1] https://lore.kernel.org/lkml/20200603174714.192027-1-glider@google.com/
[2] https://lore.kernel.org/lkml/CA+55aFw+Vbj0i=1TGqCR5vQkCzWJ0QxK6CernOU6eedsudAixw@mail.gmail.com/
[3] https://lore.kernel.org/lkml/CA+55aFwgbgqhbp1fkxvRKEpzyR5J8n1vKT1VZdz9knmPuXhOeg@mail.gmail.com/
[4] https://lore.kernel.org/lkml/CA+55aFz2500WfbKXAx8s67wrm9=yVJu65TpLgN_ybYNv0VEOKA@mail.gmail.com/

Reviewed-by: Leon Romanovsky <leonro@mellanox.com> # drivers/infiniband and mlx4/mlx5
Acked-by: Jason Gunthorpe <jgg@mellanox.com> # IB
Acked-by: Kalle Valo <kvalo@codeaurora.org> # wireless drivers
Reviewed-by: Chao Yu <yuchao0@huawei.com> # erofs
Signed-off-by: Kees Cook <keescook@chromium.org>

bpf: Fix running sk_skb program types with ktls

KTLS uses a stream parser to collect TLS messages and send them to
the upper layer tls receive handler. This ensures the tls receiver
has a full TLS header to parse when it is run. However, when a
socket has BPF_SK_SKB_STREAM_VERDICT program attached before KTLS
is enabled we end up with two stream parsers running on the same
socket.

The result is both try to run on the same socket. First the KTLS
stream parser runs and calls read_sock() which will tcp_read_sock
which in turn calls tcp_rcv_skb(). This dequeues the skb from the
sk_receive_queue. When this is done KTLS code then data_ready()
callback which because we stacked KTLS on top of the bpf stream
verdict program has been replaced with sk_psock_start_strp(). This
will in turn kick the stream parser again and eventually do the
same thing KTLS did above calling into tcp_rcv_skb() and dequeuing
a skb from the sk_receive_queue.

At this point the data stream is broke. Part of the stream was
handled by the KTLS side some other bytes may have been handled
by the BPF side. Generally this results in either missing data
or more likely a "Bad Message" complaint from the kTLS receive
handler as the BPF program steals some bytes meant to be in a
TLS header and/or the TLS header length is no longer correct.

We've already broke the idealized model where we can stack ULPs
in any order with generic callbacks on the TX side to handle this.
So in this patch we do the same thing but for RX side. We add
a sk_psock_strp_enabled() helper so TLS can learn a BPF verdict
program is running and add a tls_sw_has_ctx_rx() helper so BPF
side can learn there is a TLS ULP on the socket.

Then on BPF side we omit calling our stream parser to avoid
breaking the data stream for the KTLS receiver. Then on the
KTLS side we call BPF_SK_SKB_STREAM_VERDICT once the KTLS
receiver is done with the packet but before it posts the
msg to userspace. This gives us symmetry between the TX and
RX halfs and IMO makes it usable again. On the TX side we
process packets in this order BPF -> TLS -> TCP and on
the receive side in the reverse order TCP -> TLS -> BPF.

Discovered while testing OpenSSL 3.0 Alpha2.0 release.

Fixes: d829e9c4112b5 ("tls: convert to generic sk_msg interface")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/159079361946.5745.605854335665044485.stgit@john-Precision-5820-Tower
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

net/tls: fix race condition causing kernel panic

tls_sw_recvmsg() and tls_decrypt_done() can be run concurrently.
// tls_sw_recvmsg()
if (atomic_read(&ctx->decrypt_pending))
crypto_wait_req(-EINPROGRESS, &ctx->async_wait);
else
reinit_completion(&ctx->async_wait.completion);

//tls_decrypt_done()
pending = atomic_dec_return(&ctx->decrypt_pending);

if (!pending && READ_ONCE(ctx->async_notify))
complete(&ctx->async_wait.completion);

Consider the scenario tls_decrypt_done() is about to run complete()

if (!pending && READ_ONCE(ctx->async_notify))

and tls_sw_recvmsg() reads decrypt_pending == 0, does reinit_completion(),
then tls_decrypt_done() runs complete(). This sequence of execution
results in wrong completion. Consequently, for next decrypt request,
it will not wait for completion, eventually on connection close, crypto
resources freed, there is no way to handle pending decrypt response.

This race condition can be avoided by having atomic_read() mutually
exclusive with atomic_dec_return(),complete().Intoduced spin lock to
ensure the mutual exclution.

Addressed similar problem in tx direction.

v1->v2:
- More readable commit message.
- Corrected the lock to fix new race scenario.
- Removed barrier which is not needed now.

Fixes: a42055e8d2c3 ("net/tls: Add support for async encryption of records for performance")
Signed-off-by: Vinay Kumar Yadav <vinay.yadav@chelsio.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: free record only on encryption error

We cannot free record on any transient error because it leads to
losing previos data. Check socket error to know whether record must
be freed or not.

Fixes: d10523d0b3d7 ("net/tls: free the record on encryption error")
Signed-off-by: Vadim Fedorenko <vfedorenko@novek.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: fix encryption error checking

bpf_exec_tx_verdict() can return negative value for copied
variable. In that case this value will be pushed back to caller
and the real error code will be lost. Fix it using signed type and
checking for positive value.

Fixes: d10523d0b3d7 ("net/tls: free the record on encryption error")
Fixes: d3b18ad31f93 ("tls: add bpf support to sk_msg handling")
Signed-off-by: Vadim Fedorenko <vfedorenko@novek.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Fix sk_psock refcnt leak when in tls_data_ready()

tls_data_ready() invokes sk_psock_get(), which returns a reference of
the specified sk_psock object to "psock" with increased refcnt.

When tls_data_ready() returns, local variable "psock" becomes invalid,
so the refcount should be decreased to keep refcount balanced.

The reference counting issue happens in one exception handling path of
tls_data_ready(). When "psock->ingress_msg" is empty but "psock" is not
NULL, the function forgets to decrease the refcnt increased by
sk_psock_get(), causing a refcnt leak.

Fix this issue by calling sk_psock_put() on all paths when "psock" is
not NULL.

Signed-off-by: Xiyu Yang <xiyuyang19@fudan.edu.cn>
Signed-off-by: Xin Tan <tanxin.ctf@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Fix sk_psock refcnt leak in bpf_exec_tx_verdict()

bpf_exec_tx_verdict() invokes sk_psock_get(), which returns a reference
of the specified sk_psock object to "psock" with increased refcnt.

When bpf_exec_tx_verdict() returns, local variable "psock" becomes
invalid, so the refcount should be decreased to keep refcount balanced.

The reference counting issue happens in one exception handling path of
bpf_exec_tx_verdict(). When "policy" equals to NULL but "psock" is not
NULL, the function forgets to decrease the refcnt increased by
sk_psock_get(), causing a refcnt leak.

Fix this issue by calling sk_psock_put() on this error path before
bpf_exec_tx_verdict() returns.

Signed-off-by: Xiyu Yang <xiyuyang19@fudan.edu.cn>
Signed-off-by: Xin Tan <tanxin.ctf@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

bpf: Sockmap/tls, fix pop data with SK_DROP return code

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: 7246d8ed4dcce ("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

bpf: Sockmap/tls, skmsg can have wrapped skmsg that needs extra chaining

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: d3b18ad31f93d ("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

bpf: Sockmap/tls, tls_sw can create a plaintext buf > encrypt buf

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: d3b18ad31f93d ("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

net/tls: fix async operation

Mallesham reports the TLS with async accelerator was broken by
commit d10523d0b3d7 ("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: d3b18ad31f93 ("tls: add bpf support to sk_msg handling")
Fixes: d10523d0b3d7 ("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>

net/tls: avoid spurious decryption error with HW resync

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 5c5ec6685806 ("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: 5c5ec6685806 ("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>

net/tls: Fix return values to avoid ENOTSUPP

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>

net/tls: use sg_next() to walk sg entries

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>

net/tls: remove the dead inplace_crypto code

Looks like when BPF support was added by commit d3b18ad31f93
("tls: add bpf support to sk_msg handling") and
commit d829e9c4112b ("tls: convert to generic sk_msg interface")
it broke/removed the support for in-place crypto as added by
commit 4e6d47206c32 ("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>

net/tls: free the record on encryption error

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 d3b18ad31f93 ("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: d3b18ad31f93 ("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>

net/tls: take into account that bpf_exec_tx_verdict() may free the record

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: d3b18ad31f93 ("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>

net/tls: enable sk_msg redirect to tls socket egress

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: 0608c69c9a80 ("bpf: sk_msg, sock{map|hash} redirect through ULP")
Fixes: f3de19af0f5b ("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>

net/tls: add a TX lock

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: a42055e8d2c3 ("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>

net/tls: don't pay attention to sk_write_pending when pushing partial records

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: a42055e8d2c3 ("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>

net/tls: store decrypted on a single bit

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>

net/tls: store async_capable on a single bit

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>

net/tls: pass context to tls_device_decrypted()

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>

net/tls: add TlsDecryptError stat

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>

net/tls: clean up the number of #ifdefs for CONFIG_TLS_DEVICE

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>

net/tls: remove sock unlock/lock around strp_done()

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>

net/tls: remove close callback sock unlock/lock around TX work flush

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>

net/tls: don't arm strparser immediately in tls_set_sw_offload()

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>

net/tls: fix poll ignoring partially copied records

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: 692d7b5d1f91 ("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>

net: tls, correctly account for copied bytes with multiple sk_msgs

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: d829e9c4112b ("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>

net/tls: add kernel-driven TLS RX resync

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>

net/tls: rename handle_device_resync()

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>

net/tls: pass record number as a byte array

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>

net/tls: don't pass version to tls_advance_record_sn()

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>

net/tls: don't look for decrypted frames on non-offloaded sockets

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>

net/tls: fix no wakeup on partial reads

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 d3b18ad31f93 ("tls: add bpf support to sk_msg handling")

Fixes: 692d7b5d1f91 ("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>

net/tls: fix lowat calculation if some data came from previous record

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: 692d7b5d1f91 ("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>

net/tls: handle errors from padding_length()

At the time padding_length() is called the record header
is still part of the message. If malicious TLS 1.3 peer
sends an all-zero record padding_length() will stop at
the record header, and return full length of the data
including the tail_size.

Subsequent subtraction of prot->overhead_size from rxm->full_len
will cause rxm->full_len to turn negative. skb accessors,
however, will always catch resulting out-of-bounds operation,
so in practice this fix comes down to returning the correct
error code. It also fixes a set but not used warning.

This code was added by commit 130b392c6cd6 ("net: tls: Add tls 1.3 support").

CC: Dave Watson <davejwatson@fb.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>

net/tls: don't leak IV and record seq when offload fails

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: 4799ac81e52a ("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>

net/tls: don't leak partially sent record in device mode

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 a42055e8d2c3 ("net/tls: Add support for async encryption of records for performance")
moved the partial record cleanup to the SW-only path.

Fixes: a42055e8d2c3 ("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>

net: tls: prevent false connection termination with offload

Only decrypt_internal() performs zero copy on rx, all paths
which don't hit decrypt_internal() must set zc to false,
otherwise tls_sw_recvmsg() may return 0 causing the application
to believe that that connection got closed.

Currently this happens with device offload when new record
is first read from.

Fixes: d069b780e367 ("tls: Fix tls_device receive")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Reported-by: David Beckett <david.beckett@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Replace kfree_skb() with consume_skb()

To free the skb in normal course of processing, consume_skb() should be
used. Only for failure paths, skb_free() is intended to be used.

https://www.kernel.org/doc/htmldocs/networking/API-consume-skb.html

Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Add support of AES128-CCM based ciphers

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>

tls: Fix tls_device receive

Currently, the receive function fails to handle records already
decrypted by the device due to the commit mentioned below.

This commit advances the TLS record sequence number and prepares the context
to handle the next record.

Fixes: fedf201e1296 ("net: tls: Refactor control message handling on recv")
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>

tls: Fix mixing between async capable and async

Today, tls_sw_recvmsg is capable of using asynchronous mode to handle
application data TLS records. Moreover, it assumes that if the cipher
can be handled asynchronously, then all packets will be processed
asynchronously.

However, this assumption is not always true. Specifically, for AES-GCM
in TLS1.2, it causes data corruption, and breaks user applications.

This patch fixes this problem by separating the async capability from
the decryption operation result.

Fixes: c0ab4732d4c6 ("net/tls: Do not use async crypto for non-data records")
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Reviewed-by: Boris Pismenny <borisp@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: Fix write space handling

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: a42055e8d2c3 ("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>

tls: Return type of non-data records retrieved using MSG_PEEK in recvmsg

The patch enables returning 'type' in msghdr for records that are
retrieved with MSG_PEEK in recvmsg. Further it prevents records peeked
from socket from getting clubbed with any other record of different
type when records are subsequently dequeued from strparser.

For each record, we now retain its type in sk_buff's control buffer
cb[]. Inside control buffer, record's full length and offset are already
stored by strparser in 'struct strp_msg'. We store record type after
'struct strp_msg' inside 'struct tls_msg'. For tls1.2, the type is
stored just after record dequeue. For tls1.3, the type is stored after
record has been decrypted.

Inside process_rx_list(), before processing a non-data record, we check
that we must be able to return back the record type to the user
application. If not, the decrypted records in tls context's rx_list is
left there without consuming any data.

Fixes: 692d7b5d1f912 ("tls: Fix recvmsg() to be able to peek across multiple records")
Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Move protocol constants from cipher context to tls context

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>

net/tls: Do not use async crypto for non-data records

Addition of tls1.3 support broke tls1.2 handshake when async crypto
accelerator is used. This is because the record type for non-data
records is not propagated to user application. Also when async
decryption happens, the decryption does not stop when two different
types of records get dequeued and submitted for decryption. To address
it, we decrypt tls1.2 non-data records in synchronous way. We check
whether the record we just processed has same type as the previous one
before checking for async condition and jumping to dequeue next record.

Fixes: 130b392c6cd6b ("net: tls: Add tls 1.3 support")
Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Disable async decrytion for tls1.3

Function tls_sw_recvmsg() dequeues multiple records from stream parser
and decrypts them. In case the decryption is done by async accelerator,
the records may get submitted for decryption while the previous ones may
not have been decryted yet. For tls1.3, the record type is known only
after decryption. Therefore, for tls1.3, tls_sw_recvmsg() may submit
records for decryption even if it gets 'handshake' records after 'data'
records. These intermediate 'handshake' records may do a key updation.
By the time new keys are given to ktls by userspace, it is possible that
ktls has already submitted some records i(which are encrypted with new
keys) for decryption using old keys. This would lead to decrypt failure.
Therefore, async decryption of records should be disabled for tls1.3.

Fixes: 130b392c6cd6b ("net: tls: Add tls 1.3 support")
Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: tls: Set async_capable for tls zerocopy only if we see EINPROGRESS

Currently we don't zerocopy if the crypto framework async bit is set.
However some crypto algorithms (such as x86 AESNI) support async,
but in the context of sendmsg, will never run asynchronously. Instead,
check for actual EINPROGRESS return code before assuming algorithm is
async.

Signed-off-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: tls: Add tls 1.3 support

TLS 1.3 has minor changes from TLS 1.2 at the record layer.

* Header now hardcodes the same version and application content type in
the header.
* The real content type is appended after the data, before encryption (or
after decryption).
* The IV is xored with the sequence number, instead of concatinating four
bytes of IV with the explicit IV.
* Zero-padding: No exlicit length is given, we search backwards from the
end of the decrypted data for the first non-zero byte, which is the
content type. Currently recv supports reading zero-padding, but there
is no way for send to add zero padding.

Signed-off-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: tls: Refactor control message handling on recv

For TLS 1.3, the control message is encrypted. Handle control
message checks after decryption.

Signed-off-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: tls: Refactor tls aad space size calculation

TLS 1.3 has a different AAD size, use a variable in the code to
make TLS 1.3 support easy.

Signed-off-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: tls: Support 256 bit keys

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>

net: tls: Fix deadlock in free_resources tx

If there are outstanding async tx requests (when crypto returns EINPROGRESS),
there is a potential deadlock: the tx work acquires the lock, while we
cancel_delayed_work_sync() while holding the lock. Drop the lock while waiting
for the work to complete.

Fixes: a42055e8d2c30 ("Add support for async encryption of records...")
Signed-off-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: tls: Save iv in tls_rec for async crypto requests

aead_request_set_crypt takes an iv pointer, and we change the iv
soon after setting it. Some async crypto algorithms don't save the iv,
so we need to save it in the tls_rec for async requests.

Found by hardcoding x64 aesni to use async crypto manager (to test the async
codepath), however I don't think this combination can happen in the wild.
Presumably other hardware offloads will need this fix, but there have been
no user reports.

Fixes: a42055e8d2c30 ("Add support for async encryption of records...")
Signed-off-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: Fix recvmsg() to be able to peek across multiple records

This fixes recvmsg() to be able to peek across multiple tls records.
Without this patch, the tls's selftests test case
'recv_peek_large_buf_mult_recs' fails. Each tls receive context now
maintains a 'rx_list' to retain incoming skb carrying tls records. If a
tls record needs to be retained e.g. for peek case or for the case when
the buffer passed to recvmsg() has a length smaller than decrypted
record length, then it is added to 'rx_list'. Additionally, records are
added in 'rx_list' if the crypto operation runs in async mode. The
records are dequeued from 'rx_list' after the decrypted data is consumed
by copying into the buffer passed to recvmsg(). In case, the MSG_PEEK
flag is used in recvmsg(), then records are not consumed or removed
from the 'rx_list'.

Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Make function tls_sw_do_sendpage static

Fixes the following sparse warning:

net/tls/tls_sw.c:1023:5: warning:
symbol 'tls_sw_do_sendpage' was not declared. Should it be static?

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: remove unused function tls_sw_sendpage_locked

There are no in-tree callers.

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: Do not call sk_memcopy_from_iter with zero length

In some conditions e.g. when tls_clone_plaintext_msg() returns -ENOSPC,
the number of bytes to be copied using subsequent function
sk_msg_memcopy_from_iter() becomes zero. This causes function
sk_msg_memcopy_from_iter() to fail which in turn causes tls_sw_sendmsg()
to return failure. To prevent it, do not call sk_msg_memcopy_from_iter()
when number of bytes to copy (indicated by 'try_to_copy') is zero.

Fixes: d829e9c4112b ("tls: convert to generic sk_msg interface")
Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

bpf: sk_msg, sock{map|hash} redirect through ULP

A sockmap program that redirects through a kTLS ULP enabled socket
will not work correctly because the ULP layer is skipped. This
fixes the behavior to call through the ULP layer on redirect to
ensure any operations required on the data stream at the ULP layer
continue to be applied.

To do this we add an internal flag MSG_SENDPAGE_NOPOLICY to avoid
calling the BPF layer on a redirected message. This is
required to avoid calling the BPF layer multiple times (possibly
recursively) which is not the current/expected behavior without
ULPs. In the future we may add a redirect flag if users _do_
want the policy applied again but this would need to work for both
ULP and non-ULP sockets and be opt-in to avoid breaking existing
programs.

Also to avoid polluting the flag space with an internal flag we
reuse the flag space overlapping MSG_SENDPAGE_NOPOLICY with
MSG_WAITFORONE. Here WAITFORONE is specific to recv path and
SENDPAGE_NOPOLICY is only used for sendpage hooks. The last thing
to verify is user space API is masked correctly to ensure the flag
can not be set by user. (Note this needs to be true regardless
because we have internal flags already in-use that user space
should not be able to set). But for completeness we have two UAPI
paths into sendpage, sendfile and splice.

In the sendfile case the function do_sendfile() zero's flags,

./fs/read_write.c:
static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos,
size_t count, loff_t max)
{
...
fl = 0;
#if 0
/*
* We need to debate whether we can enable this or not. The
* man page documents EAGAIN return for the output at least,
* and the application is arguably buggy if it doesn't expect
* EAGAIN on a non-blocking file descriptor.
*/
if (in.file->f_flags & O_NONBLOCK)
fl = SPLICE_F_NONBLOCK;
#endif
file_start_write(out.file);
retval = do_splice_direct(in.file, &pos, out.file, &out_pos, count, fl);
}

In the splice case the pipe_to_sendpage "actor" is used which
masks flags with SPLICE_F_MORE.

./fs/splice.c:
static int pipe_to_sendpage(struct pipe_inode_info *pipe,
struct pipe_buffer *buf, struct splice_desc *sd)
{
...
more = (sd->flags & SPLICE_F_MORE) ? MSG_MORE : 0;
...
}

Confirming what we expect that internal flags are in fact internal
to socket side.

Fixes: d3b18ad31f93 ("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>

bpf: helper to pop data from messages

This adds a BPF SK_MSG program helper so that we can pop data from a
msg. We use this to pop metadata from a previous push data call.

Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>

iov_iter: Use accessor function

Use accessor functions to access an iterator's type and direction. This
allows for the possibility of using some other method of determining the
type of iterator than if-chains with bitwise-AND conditions.

Signed-off-by: David Howells <dhowells@redhat.com>

bpf: sockmap, support for msg_peek in sk_msg with redirect ingress

This adds support for the MSG_PEEK flag when doing redirect to ingress
and receiving on the sk_msg psock queue. Previously the flag was
being ignored which could confuse applications if they expected the
flag to work as normal.

Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>

tls: add bpf support to sk_msg handling

This work adds BPF sk_msg verdict program support to kTLS
allowing BPF and kTLS to be combined together. Previously kTLS
and sk_msg verdict programs were mutually exclusive in the
ULP layer which created challenges for the orchestrator when
trying to apply TCP based policy, for example. To resolve this,
leveraging the work from previous patches that consolidates
the use of sk_msg, we can finally enable BPF sk_msg verdict
programs so they continue to run after the kTLS socket is
created. No change in behavior when kTLS is not used in
combination with BPF, the kselftest suite for kTLS also runs
successfully.

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>

tls: replace poll implementation with read hook

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>

tls: convert to generic sk_msg interface

Convert kTLS over to make use of sk_msg interface for plaintext and
encrypted scattergather data, so it reuses all the sk_msg helpers
and data structure which later on in a second step enables to glue
this to BPF.

This also allows to remove quite a bit of open coded helpers which
are covered by the sk_msg API. Recent changes in kTLs 80ece6a03aaf
("tls: Remove redundant vars from tls record structure") and
4e6d47206c32 ("tls: Add support for inplace records encryption")
changed the data path handling a bit; while we've kept the latter
optimization intact, we had to undo the former change to better
fit the sk_msg model, hence the sg_aead_in and sg_aead_out have
been brought back and are linked into the sk_msg sgs. Now the kTLS
record contains a msg_plaintext and msg_encrypted sk_msg each.

In the original code, the zerocopy_from_iter() has been used out
of TX but also RX path. For the strparser skb-based RX path,
we've left the zerocopy_from_iter() in decrypt_internal() mostly
untouched, meaning it has been moved into tls_setup_from_iter()
with charging logic removed (as not used from RX). Given RX path
is not based on sk_msg objects, we haven't pursued setting up a
dummy sk_msg to call into sk_msg_zerocopy_from_iter(), but it
could be an option to prusue in a later step.

Joint work with John.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

tls: Add support for inplace records encryption

Presently, for non-zero copy case, separate pages are allocated for
storing plaintext and encrypted text of records. These pages are stored
in sg_plaintext_data and sg_encrypted_data scatterlists inside record
structure. Further, sg_plaintext_data & sg_encrypted_data are passed
to cryptoapis for record encryption. Allocating separate pages for
plaintext and encrypted text is inefficient from both required memory
and performance point of view.

This patch adds support of inplace encryption of records. For non-zero
copy case, we reuse the pages from sg_encrypted_data scatterlist to
copy the application's plaintext data. For the movement of pages from
sg_encrypted_data to sg_plaintext_data scatterlists, we introduce a new
function move_to_plaintext_sg(). This function add pages into
sg_plaintext_data from sg_encrypted_data scatterlists.

tls_do_encryption() is modified to pass the same scatterlist as both
source and destination into aead_request_set_crypt() if inplace crypto
has been enabled. A new ariable 'inplace_crypto' has been introduced in
record structure to signify whether the same scatterlist can be used.
By default, the inplace_crypto is enabled in get_rec(). If zero-copy is
used (i.e. plaintext data is not copied), inplace_crypto is set to '0'.

Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Reviewed-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: Remove redundant vars from tls record structure

Structure 'tls_rec' contains sg_aead_in and sg_aead_out which point
to a aad_space and then chain scatterlists sg_plaintext_data,
sg_encrypted_data respectively. Rather than using chained scatterlists
for plaintext and encrypted data in aead_req, it is efficient to store
aad_space in sg_encrypted_data and sg_plaintext_data itself in the
first index and get rid of sg_aead_in, sg_aead_in and further chaining.

This requires increasing size of sg_encrypted_data & sg_plaintext_data
arrarys by 1 to accommodate entry for aad_space. The code which uses
sg_encrypted_data and sg_plaintext_data has been modified to skip first
index as it points to aad_space.

Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Make function get_rec() static

Fixes the following sparse warning:

net/tls/tls_sw.c:655:16: warning:
symbol 'get_rec' was not declared. Should it be static?

Fixes: a42055e8d2c3 ("net/tls: Add support for async encryption of records for performance")
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: Fixed a memory leak during socket close

During socket close, if there is a open record with tx context, it needs
to be be freed apart from freeing up plaintext and encrypted scatter
lists. This patch frees up the open record if present in tx context.

Also tls_free_both_sg() has been renamed to tls_free_open_rec() to
indicate that the free record in tx context is being freed inside the
function.

Fixes: a42055e8d2c3 ("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>

tls: Fix socket mem accounting error under async encryption

Current async encryption implementation sometimes showed up socket
memory accounting error during socket close. This results in kernel
warning calltrace. The root cause of the problem is that socket var
sk_forward_alloc gets corrupted due to access in sk_mem_charge()
and sk_mem_uncharge() being invoked from multiple concurrent contexts
in multicore processor. The apis sk_mem_charge() and sk_mem_uncharge()
are called from functions alloc_plaintext_sg(), free_sg() etc. It is
required that memory accounting apis are called under a socket lock.

The plaintext sg data sent for encryption is freed using free_sg() in
tls_encryption_done(). It is wrong to call free_sg() from this function.
This is because this function may run in irq context. We cannot acquire
socket lock in this function.

We remove calling of function free_sg() for plaintext data from
tls_encryption_done() and defer freeing up of plaintext data to the time
when the record is picked up from tx_list and transmitted/freed. When
tls_tx_records() gets called, socket is already locked and thus there is
no concurrent access problem.

Fixes: a42055e8d2c3 ("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>

tls: Fixed uninitialised vars warning

In tls_sw_sendmsg() and tls_sw_sendpage(), it is possible that the
uninitialised variable 'ret' gets passed to sk_stream_error(). So
initialise local variable 'ret' to '0. The warnings were detected by
'smatch' tool.

Fixes: a42055e8d2c3 ("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>

net/tls: Fixed race condition in async encryption

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: a42055e8d2c3 ("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>

net/tls: Add support for async encryption of records for performance

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>

tls: fix currently broken MSG_PEEK behavior

In kTLS MSG_PEEK behavior is currently failing, strace example:

[pid 2430] socket(AF_INET, SOCK_STREAM, IPPROTO_IP) = 3
[pid 2430] socket(AF_INET, SOCK_STREAM, IPPROTO_IP) = 4
[pid 2430] bind(4, {sa_family=AF_INET, sin_port=htons(0), sin_addr=inet_addr("0.0.0.0")}, 16) = 0
[pid 2430] listen(4, 10) = 0
[pid 2430] getsockname(4, {sa_family=AF_INET, sin_port=htons(38855), sin_addr=inet_addr("0.0.0.0")}, [16]) = 0
[pid 2430] connect(3, {sa_family=AF_INET, sin_port=htons(38855), sin_addr=inet_addr("0.0.0.0")}, 16) = 0
[pid 2430] setsockopt(3, SOL_TCP, 0x1f /* TCP_??? */, [7564404], 4) = 0
[pid 2430] setsockopt(3, 0x11a /* SOL_?? */, 1, "\3\0033\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 40) = 0
[pid 2430] accept(4, {sa_family=AF_INET, sin_port=htons(49636), sin_addr=inet_addr("127.0.0.1")}, [16]) = 5
[pid 2430] setsockopt(5, SOL_TCP, 0x1f /* TCP_??? */, [7564404], 4) = 0
[pid 2430] setsockopt(5, 0x11a /* SOL_?? */, 2, "\3\0033\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 40) = 0
[pid 2430] close(4) = 0
[pid 2430] sendto(3, "test_read_peek", 14, 0, NULL, 0) = 14
[pid 2430] sendto(3, "_mult_recs\0", 11, 0, NULL, 0) = 11
[pid 2430] recvfrom(5, "test_read_peektest_read_peektest"..., 64, MSG_PEEK, NULL, NULL) = 64

As can be seen from strace, there are two TLS records sent,
i) 'test_read_peek' and ii) '_mult_recs\0' where we end up
peeking 'test_read_peektest_read_peektest'. This is clearly
wrong, and what happens is that given peek cannot call into
tls_sw_advance_skb() to unpause strparser and proceed with
the next skb, we end up looping over the current one, copying
the 'test_read_peek' over and over into the user provided
buffer.

Here, we can only peek into the currently held skb (current,
full TLS record) as otherwise we would end up having to hold
all the original skb(s) (depending on the peek depth) in a
separate queue when unpausing strparser to process next
records, minimally intrusive is to return only up to the
current record's size (which likely was what c46234ebb4d1
("tls: RX path for ktls") originally intended as well). Thus,
after patch we properly peek the first record:

[pid 2046] wait4(2075, <unfinished ...>
[pid 2075] socket(AF_INET, SOCK_STREAM, IPPROTO_IP) = 3
[pid 2075] socket(AF_INET, SOCK_STREAM, IPPROTO_IP) = 4
[pid 2075] bind(4, {sa_family=AF_INET, sin_port=htons(0), sin_addr=inet_addr("0.0.0.0")}, 16) = 0
[pid 2075] listen(4, 10) = 0
[pid 2075] getsockname(4, {sa_family=AF_INET, sin_port=htons(55115), sin_addr=inet_addr("0.0.0.0")}, [16]) = 0
[pid 2075] connect(3, {sa_family=AF_INET, sin_port=htons(55115), sin_addr=inet_addr("0.0.0.0")}, 16) = 0
[pid 2075] setsockopt(3, SOL_TCP, 0x1f /* TCP_??? */, [7564404], 4) = 0
[pid 2075] setsockopt(3, 0x11a /* SOL_?? */, 1, "\3\0033\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 40) = 0
[pid 2075] accept(4, {sa_family=AF_INET, sin_port=htons(45732), sin_addr=inet_addr("127.0.0.1")}, [16]) = 5
[pid 2075] setsockopt(5, SOL_TCP, 0x1f /* TCP_??? */, [7564404], 4) = 0
[pid 2075] setsockopt(5, 0x11a /* SOL_?? */, 2, "\3\0033\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 40) = 0
[pid 2075] close(4) = 0
[pid 2075] sendto(3, "test_read_peek", 14, 0, NULL, 0) = 14
[pid 2075] sendto(3, "_mult_recs\0", 11, 0, NULL, 0) = 11
[pid 2075] recvfrom(5, "test_read_peek", 64, MSG_PEEK, NULL, NULL) = 14

Fixes: c46234ebb4d1 ("tls: RX path for ktls")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: async support causes out-of-bounds access in crypto APIs

When async support was added it needed to access the sk from the async
callback to report errors up the stack. The patch tried to use space
after the aead request struct by directly setting the reqsize field in
aead_request. This is an internal field that should not be used
outside the crypto APIs. It is used by the crypto code to define extra
space for private structures used in the crypto context. Users of the
API then use crypto_aead_reqsize() and add the returned amount of
bytes to the end of the request memory allocation before posting the
request to encrypt/decrypt APIs.

So this breaks (with general protection fault and KASAN error, if
enabled) because the request sent to decrypt is shorter than required
causing the crypto API out-of-bounds errors. Also it seems unlikely the
sk is even valid by the time it gets to the callback because of memset
in crypto layer.

Anyways, fix this by holding the sk in the skb->sk field when the
callback is set up and because the skb is already passed through to
the callback handler via void* we can access it in the handler. Then
in the handler we need to be careful to NULL the pointer again before
kfree_skb. I added comments on both the setup (in tls_do_decryption)
and when we clear it from the crypto callback handler
tls_decrypt_done(). After this selftests pass again and fixes KASAN
errors/warnings.

Fixes: 94524d8fc965 ("net/tls: Add support for async decryption of tls records")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Reviewed-by: Vakul Garg <Vakul.garg@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: zero the crypto information from tls_context before freeing

This contains key material in crypto_send_aes_gcm_128 and
crypto_recv_aes_gcm_128.

Introduce union tls_crypto_context, and replace the two identical
unions directly embedded in struct tls_context with it. We can then
use this union to clean up the memory in the new tls_ctx_free()
function.

Fixes: 3c4d7559159b ("tls: kernel TLS support")
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: don't copy the key out of tls12_crypto_info_aes_gcm_128

There's no need to copy the key to an on-stack buffer before calling
crypto_aead_setkey().

Fixes: 3c4d7559159b ("tls: kernel TLS support")
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Fixed return value when tls_complete_pending_work() fails

In tls_sw_sendmsg() and tls_sw_sendpage(), the variable 'ret' has
been set to return value of tls_complete_pending_work(). This allows
return of proper error code if tls_complete_pending_work() fails.

Fixes: 3c4d7559159b ("tls: kernel TLS support")
Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Set count of SG entries if sk_alloc_sg returns -ENOSPC

tls_sw_sendmsg() allocates plaintext and encrypted SG entries using
function sk_alloc_sg(). In case the number of SG entries hit
MAX_SKB_FRAGS, sk_alloc_sg() returns -ENOSPC and sets the variable for
current SG index to '0'. This leads to calling of function
tls_push_record() with 'sg_encrypted_num_elem = 0' and later causes
kernel crash. To fix this, set the number of SG elements to the number
of elements in plaintext/encrypted SG arrays in case sk_alloc_sg()
returns -ENOSPC.

Fixes: 3c4d7559159b ("tls: kernel TLS support")
Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Add support for async decryption of tls records

When tls records are decrypted using asynchronous acclerators such as
NXP CAAM engine, the crypto apis return -EINPROGRESS. Presently, on
getting -EINPROGRESS, the tls record processing stops till the time the
crypto accelerator finishes off and returns the result. This incurs a
context switch and is not an efficient way of accessing the crypto
accelerators. Crypto accelerators work efficient when they are queued
with multiple crypto jobs without having to wait for the previous ones
to complete.

The patch submits multiple crypto requests without having to wait for
for previous ones to complete. This has been implemented for records
which are decrypted in zero-copy mode. At the end of recvmsg(), we wait
for all the asynchronous decryption requests to complete.

The references to records which have been sent for async decryption are
dropped. For cases where record decryption is not possible in zero-copy
mode, asynchronous decryption is not used and we wait for decryption
crypto api to complete.

For crypto requests executing in async fashion, the memory for
aead_request, sglists and skb etc is freed from the decryption
completion handler. The decryption completion handler wakesup the
sleeping user context when recvmsg() flags that it has done sending
all the decryption requests and there are no more decryption requests
pending to be completed.

Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Reviewed-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Calculate nsg for zerocopy path without skb_cow_data.

decrypt_skb fails if the number of sg elements required to map it
is greater than MAX_SKB_FRAGS. nsg must always be calculated, but
skb_cow_data adds unnecessary memcpy's for the zerocopy case.

The new function skb_nsg calculates the number of scatterlist elements
required to map the skb without the extra overhead of skb_cow_data.
This patch reduces memcpy by 50% on my encrypted NBD benchmarks.

Reported-by: Vakul Garg <Vakul.garg@nxp.com>
Reviewed-by: Vakul Garg <Vakul.garg@nxp.com>
Tested-by: Vakul Garg <Vakul.garg@nxp.com>
Signed-off-by: Doron Roberts-Kedes <doronrk@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Combined memory allocation for decryption request

For preparing decryption request, several memory chunks are required
(aead_req, sgin, sgout, iv, aad). For submitting the decrypt request to
an accelerator, it is required that the buffers which are read by the
accelerator must be dma-able and not come from stack. The buffers for
aad and iv can be separately kmalloced each, but it is inefficient.
This patch does a combined allocation for preparing decryption request
and then segments into aead_req || sgin || sgout || iv || aad.

Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Mark the end in scatterlist table

Function zerocopy_from_iter() unmarks the 'end' in input sgtable while
adding new entries in it. The last entry in sgtable remained unmarked.
This results in KASAN error report on using apis like sg_nents(). Before
returning, the function needs to mark the 'end' in the last entry it
adds.

Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Acked-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Use kmemdup to simplify the code

Kmemdup is better than kmalloc+memcpy. So replace them.

Signed-off-by: zhong jiang <zhongjiang@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Use socket data_ready callback on record availability

On receipt of a complete tls record, use socket's saved data_ready
callback instead of state_change callback. In function tls_queue(),
the TLS record is queued in encrypted state. But the decryption
happen inline when tls_sw_recvmsg() or tls_sw_splice_read() get invoked.
So it should be ok to notify the waiting context about the availability
of data as soon as we could collect a full TLS record. For new data
availability notification, sk_data_ready callback is more appropriate.
It points to sock_def_readable() which wakes up specifically for EPOLLIN
event. This is in contrast to the socket callback sk_state_change which
points to sock_def_wakeup() which issues a wakeup unconditionally
(without event mask).

Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: Fix improper revert in zerocopy_from_iter

The current code is problematic because the iov_iter is reverted and
never advanced in the non-error case. This patch skips the revert in the
non-error case. This patch also fixes the amount by which the iov_iter
is reverted. Currently, iov_iter is reverted by size, which can be
greater than the amount by which the iter was actually advanced.
Instead, only revert by the amount that the iter was advanced.

Fixes: 4718799817c5 ("tls: Fix zerocopy_from_iter iov handling")
Signed-off-by: Doron Roberts-Kedes <doronrk@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: Remove dead code in tls_sw_sendmsg

tls_push_record either returns 0 on success or a negative value on failure.
This patch removes code that would only be executed if tls_push_record
were to return a positive value.

Signed-off-by: Doron Roberts-Kedes <doronrk@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: Skip zerocopy path for ITER_KVEC

The zerocopy path ultimately calls iov_iter_get_pages, which defines the
step function for ITER_KVECs as simply, return -EFAULT. Taking the
non-zerocopy path for ITER_KVECs avoids the unnecessary fallback.

See https://lore.kernel.org/lkml/20150401023311.GL29656@ZenIV.linux.org.uk/T/#u
for a discussion of why zerocopy for vmalloc data is not a good idea.

Discovered while testing NBD traffic encrypted with ktls.

Fixes: c46234ebb4d1 ("tls: RX path for ktls")
Signed-off-by: Doron Roberts-Kedes <doronrk@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Removed redundant checks for non-NULL

Removed checks against non-NULL before calling kfree_skb() and
crypto_free_aead(). These functions are safe to be called with NULL
as an argument.

Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Acked-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: check RCV_SHUTDOWN in tls_wait_data

The current code does not check sk->sk_shutdown & RCV_SHUTDOWN.
tls_sw_recvmsg may return a positive value in the case where bytes have
already been copied when the socket is shutdown. sk->sk_err has been
cleared, causing the tls_wait_data to hang forever on a subsequent
invocation. Checking sk->sk_shutdown & RCV_SHUTDOWN, as in tcp_recvmsg,
fixes this problem.

Fixes: c46234ebb4d1 ("tls: RX path for ktls")
Acked-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: Doron Roberts-Kedes <doronrk@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: Stricter error checking in zerocopy sendmsg path

In the zerocopy sendmsg() path, there are error checks to revert
the zerocopy if we get any error code. syzkaller has discovered
that tls_push_record can return -ECONNRESET, which is fatal, and
happens after the point at which it is safe to revert the iter,
as we've already passed the memory to do_tcp_sendpages.

Previously this code could return -ENOMEM and we would want to
revert the iter, but AFAIK this no longer returns ENOMEM after
a447da7d004 ("tls: fix waitall behavior in tls_sw_recvmsg"),
so we fail for all error codes.

Reported-by: syzbot+c226690f7b3126c5ee04@syzkaller.appspotmail.com
Reported-by: syzbot+709f2810a6a05f11d4d3@syzkaller.appspotmail.com
Signed-off-by: Dave Watson <davejwatson@fb.com>
Fixes: 3c4d7559159b ("tls: kernel TLS support")
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: Fix zerocopy_from_iter iov handling

zerocopy_from_iter iterates over the message, but it doesn't revert the
updates made by the iov iteration. This patch fixes it. Now, the iov can
be used after calling zerocopy_from_iter.

Fixes: 3c4d75591 ("tls: kernel TLS support")
Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: Add rx inline crypto offload

This patch completes the generic infrastructure to offload TLS crypto to a
network device. It enables the kernel to skip decryption and
authentication of some skbs marked as decrypted by the NIC. In the fast
path, all packets received are decrypted by the NIC and the performance
is comparable to plain TCP.

This infrastructure doesn't require a TCP offload engine. Instead, the
NIC only decrypts packets that contain the expected TCP sequence number.
Out-Of-Order TCP packets are provided unmodified. As a result, at the
worst case a received TLS record consists of both plaintext and ciphertext
packets. These partially decrypted records must be reencrypted,
only to be decrypted.

The notable differences between SW KTLS Rx and this offload are as
follows:
1. Partial decryption - Software must handle the case of a TLS record
that was only partially decrypted by HW. This can happen due to packet
reordering.
2. Resynchronization - tls_read_size calls the device driver to
resynchronize HW after HW lost track of TLS record framing in
the TCP stream.

Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: Fill software context without allocation

This patch allows tls_set_sw_offload to fill the context in case it was
already allocated previously.

We will use it in TLS_DEVICE to fill the RX software context.

Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: Split tls_sw_release_resources_rx

This patch splits tls_sw_release_resources_rx into two functions one
which releases all inner software tls structures and another that also
frees the containing structure.

In TLS_DEVICE we will need to release the software structures without
freeeing the containing structure, which contains other information.

Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: Split decrypt_skb to two functions

Previously, decrypt_skb also updated the TLS context.
Now, decrypt_skb only decrypts the payload using the current context,
while decrypt_skb_update also updates the state.

Later, in the tls_device Rx flow, we will use decrypt_skb directly.

Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Use aead_request_alloc/free for request alloc/free

Instead of kzalloc/free for aead_request allocation and free, use
functions aead_request_alloc(), aead_request_free(). It ensures that
any sensitive crypto material held in crypto transforms is securely
erased from memory.

Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Acked-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: fix skb_to_sgvec returning unhandled error.

The current code does not inspect the return value of skb_to_sgvec. This
can cause a nullptr kernel panic when the malformed sgvec is passed into
the crypto request.

Checking the return value of skb_to_sgvec and skipping decryption if it
is negative fixes this problem.

Fixes: c46234ebb4d1 ("tls: RX path for ktls")
Acked-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: Doron Roberts-Kedes <doronrk@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Revert changes to convert to ->poll_mask() and aio IOCB_CMD_POLL

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>

net/tls: Remove VLA usage on nonce

It looks like the prior VLA removal, commit b16520f7493d ("net/tls: Remove
VLA usage"), and a new VLA addition, commit c46234ebb4d1e ("tls: RX path
for ktls"), passed in the night. This removes the newly added VLA, which
happens to have its bounds based on the same max value.

Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: Removed unused variable

Removed unused variable 'rxm' from tls_queue().

Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: fix waitall behavior in tls_sw_recvmsg

Current behavior in tls_sw_recvmsg() is to wait for incoming tls
messages and copy up to exactly len bytes of data that the user
provided. This is problematic in the sense that i) if no packet
is currently queued in strparser we keep waiting until one has been
processed and pushed into tls receive layer for tls_wait_data() to
wake up and push the decrypted bits to user space. Given after
tls decryption, we're back at streaming data, use sock_rcvlowat()
hint from tcp socket instead. Retain current behavior with MSG_WAITALL
flag and otherwise use the hint target for breaking the loop and
returning to application. This is done if currently no ctx->recv_pkt
is ready, otherwise continue to process it from our strparser
backlog.

Fixes: c46234ebb4d1 ("tls: RX path for ktls")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: fix use-after-free in tls_push_record

syzkaller managed to trigger a use-after-free in tls like the
following:

BUG: KASAN: use-after-free in tls_push_record.constprop.15+0x6a2/0x810 [tls]
Write of size 1 at addr ffff88037aa08000 by task a.out/2317

CPU: 3 PID: 2317 Comm: a.out Not tainted 4.17.0+ #144
Hardware name: LENOVO 20FBCTO1WW/20FBCTO1WW, BIOS N1FET47W (1.21 ) 11/28/2016
Call Trace:
dump_stack+0x71/0xab
print_address_description+0x6a/0x280
kasan_report+0x258/0x380
? tls_push_record.constprop.15+0x6a2/0x810 [tls]
tls_push_record.constprop.15+0x6a2/0x810 [tls]
tls_sw_push_pending_record+0x2e/0x40 [tls]
tls_sk_proto_close+0x3fe/0x710 [tls]
? tcp_check_oom+0x4c0/0x4c0
? tls_write_space+0x260/0x260 [tls]
? kmem_cache_free+0x88/0x1f0
inet_release+0xd6/0x1b0
__sock_release+0xc0/0x240
sock_close+0x11/0x20
__fput+0x22d/0x660
task_work_run+0x114/0x1a0
do_exit+0x71a/0x2780
? mm_update_next_owner+0x650/0x650
? handle_mm_fault+0x2f5/0x5f0
? __do_page_fault+0x44f/0xa50
? mm_fault_error+0x2d0/0x2d0
do_group_exit+0xde/0x300
__x64_sys_exit_group+0x3a/0x50
do_syscall_64+0x9a/0x300
? page_fault+0x8/0x30
entry_SYSCALL_64_after_hwframe+0x44/0xa9

This happened through fault injection where aead_req allocation in
tls_do_encryption() eventually failed and we returned -ENOMEM from
the function. Turns out that the use-after-free is triggered from
tls_sw_sendmsg() in the second tls_push_record(). The error then
triggers a jump to waiting for memory in sk_stream_wait_memory()
resp. returning immediately in case of MSG_DONTWAIT. What follows is
the trim_both_sgl(sk, orig_size), which drops elements from the sg
list added via tls_sw_sendmsg(). Now the use-after-free gets triggered
when the socket is being closed, where tls_sk_proto_close() callback
is invoked. The tls_complete_pending_work() will figure that there's
a pending closed tls record to be flushed and thus calls into the
tls_push_pending_closed_record() from there. ctx->push_pending_record()
is called from the latter, which is the tls_sw_push_pending_record()
from sw path. This again calls into tls_push_record(). And here the
tls_fill_prepend() will panic since the buffer address has been freed
earlier via trim_both_sgl(). One way to fix it is to move the aead
request allocation out of tls_do_encryption() early into tls_push_record().
This means we don't prep the tls header and advance state to the
TLS_PENDING_CLOSED_RECORD before allocation which could potentially
fail happened. That fixes the issue on my side.

Fixes: 3c4d7559159b ("tls: kernel TLS support")
Reported-by: syzbot+5c74af81c547738e1684@syzkaller.appspotmail.com
Reported-by: syzbot+709f2810a6a05f11d4d3@syzkaller.appspotmail.com
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: fix NULL pointer dereference on poll

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 152524231023 ("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 2c7d3dacebd4 ("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: 2c7d3dacebd4 ("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>

strparser: Add __strp_unpause and use it in ktls.

strp_unpause queues strp_work in order to parse any messages that
arrived while the strparser was paused. However, the process invoking
strp_unpause could eagerly parse a buffered message itself if it held
the sock lock.

__strp_unpause is an alternative to strp_pause that avoids the scheduling
overhead that results when a receiving thread unpauses the strparser
and waits for the next message to be delivered by the workqueue thread.

This patch more than doubled the IOPS achieved in a benchmark of NBD
traffic encrypted using ktls.

Signed-off-by: Doron Roberts-Kedes <doronrk@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: don't use stack memory in a scatterlist

scatterlist code expects virt_to_page() to work, which fails with
CONFIG_VMAP_STACK=y.

Fixes: c46234ebb4d1e ("tls: RX path for ktls")
Signed-off-by: Matt Mullins <mmullins@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Split conf to rx + tx

In TLS inline crypto, we can have one direction in software
and another in hardware. Thus, we split the TLS configuration to separate
structures for receive and transmit.

Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: remove redundant second null check on sgout

A duplicated null check on sgout is redundant as it is known to be
already true because of the identical earlier check. Remove it.
Detected by cppcheck:

net/tls/tls_sw.c:696: (warning) Identical inner 'if' condition is always
true.

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/tls: Remove VLA usage

In the quest to remove VLAs from the kernel[1], this replaces the VLA
size with the only possible size used in the code, and adds a mechanism
to double-check future IV sizes.

[1] https://lkml.kernel.org/r/CA+55aFzCG-zNmZwX4A2FQpadafLfEzK6CC=qPXydAacU1RqZWA@mail.gmail.com

Signed-off-by: Kees Cook <keescook@chromium.org>
Acked-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: RX path for ktls

Add rx path for tls software implementation.

recvmsg, splice_read, and poll implemented.

An additional sockopt TLS_RX is added, with the same interface as
TLS_TX. Either TLX_RX or TLX_TX may be provided separately, or
together (with two different setsockopt calls with appropriate keys).

Control messages are passed via CMSG in a similar way to transmit.
If no cmsg buffer is passed, then only application data records
will be passed to userspace, and EIO is returned for other types of
alerts.

EBADMSG is passed for decryption errors, and EMSGSIZE is passed for
framing too big, and EBADMSG for framing too small (matching openssl
semantics). EINVAL is returned for TLS versions that do not match the
original setsockopt call. All are unrecoverable.

strparser is used to parse TLS framing. Decryption is done directly
in to userspace buffers if they are large enough to support it, otherwise
sk_cow_data is called (similar to ipsec), and buffers are decrypted in
place and copied. splice_read always decrypts in place, since no
buffers are provided to decrypt in to.

sk_poll is overridden, and only returns POLLIN if a full TLS message is
received. Otherwise we wait for strparser to finish reading a full frame.
Actual decryption is only done during recvmsg or splice_read calls.

Signed-off-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: Pass error code explicitly to tls_err_abort

Pass EBADMSG explicitly to tls_err_abort. Receive path will
pass additional codes - EMSGSIZE if framing is larger than max
TLS record size, EINVAL if TLS version mismatch.

Signed-off-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: Move cipher info to a separate struct

Separate tx crypto parameters to a separate cipher_context struct.
The same parameters will be used for rx using the same struct.

tls_advance_record_sn is modified to only take the cipher info.

Signed-off-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tls: Generalize zerocopy_from_iter

Refactor zerocopy_from_iter to take arguments for pages and size,
such that it can be used for both tx and rx. RX will also support
zerocopy direct to output iter, as long as the full message can
be copied at once (a large enough userspace buffer was provided).

Signed-off-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: generalize sk_alloc_sg to work with scatterlist rings

The current implementation of sk_alloc_sg expects scatterlist to always
start at entry 0 and complete at entry MAX_SKB_FRAGS.

Future patches will want to support starting at arbitrary offset into
scatterlist so add an additional sg_start parameters and then default
to the current values in TLS code paths.

Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>

sock: make static tls function alloc_sg generic sock helper

The TLS ULP module builds scatterlists from a sock using
page_frag_refill(). This is going to be useful for other ULPs
so move it into sock file for more general use.

In the process remove useless goto at end of while loop.

Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>

tls: Add support for encryption using async offload accelerator

Async crypto accelerators (e.g. drivers/crypto/caam) support offloading
GCM operation. If they are enabled, crypto_aead_encrypt() return error
code -EINPROGRESS. In this case tls_do_encryption() needs to wait on a
completion till the time the response for crypto offload request is
received.

Signed-off-by: Vakul Garg <vakul.garg@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>