check-format.pl: improve preprocessor directive handling; re-order state variables

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/18812)

PKCS7: add notes to pkcs7.h.in and minor code cleanup in crypto/{pkcs7,cms}/

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/18915)

Fix many inconsistencies in doc of CMS_verify() and PKC7_verify() etc.

Also change B< to I< in {CMS,PKCS7}_verify.pod, PKCS7_sign{,_add_signer}.pod

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/18915)

TX key update support, RX time and PN reporting, general refactoring

- Adds an RX time field to the OSSL_QRX_PKT structure.

- Adds a timekeeping argument to ossl_demux_new which is used to determine
  packet reception time.

- Adds a decoded PN field to the OSSL_QRX_PKT structure.
  This has to be decoded by the QRX anyway, and its omission was an oversight.

- Key update support for the TX side.

- Minor refactoring.

Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18949)

QUIC Record Layer (Refactor and TX Side)

Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18949)

QUIC Demuxer and Record Layer (RX Side)

Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18949)

Document the return value of OSSL_LIB_CTX_load_config()

Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19101)

Fix documented max. length for info buffer in HKDF

This limit was increased with 20c2876f24d0 ("Increase
HKDF_MAXBUF from 1024 to 2048").

CLA: trivial

Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19085)

PKCS7_dataVerify(): fix missing use of CRLs in PKCS#7 message

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/18913)

Remove unused dtls1_bitmap.c

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19088)

BIO_dgram support for BIO_sendmmsg/BIO_recvmmsg

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18270)

Improve chacha20 perfomance on aarch64 by interleaving scalar with SVE/SVE2

The patch will process one extra block by scalar in addition to
blocks by SVE/SVE2 in parallel. This is esp. helpful in the
scenario where we only have 128-bit vector length.

The actual uplift to performance is complicated, depending on the
vector length and input data size. SVE/SVE2 implementation don't
always perform better than  Neon, but it should prevail in most
cases

On a CPU with 256-bit SVE/SVE2, interleaved processing can
handle 9 blocks in parallel (8 blocks by SVE and 1 by Scalar).
on 128-bit SVE/SVE2 it is 5 blocks. Input size that is a multiple
of 9/5 blocks on respective CPU can be typically handled at
maximum speed.

Here are test data for 256-bit and 128-bit SVE/SVE2 by running
"openssl speed -evp chacha20 -bytes 576" (and other size)

----------------------------------+---------------------------------
                256-bit SVE       |        128-bit SVE2
----------------------------------|---------------------------------
Input  576 bytes     512 bytes    |  320 bytes        256 bytes
----------------------------------|---------------------------------
SVE    1716361.91k   1556699.18k  |  1615789.06k      1302864.40k
----------------------------------|---------------------------------
Neon   1262643.44k   1509044.05k  |  680075.67k       1060532.31k
----------------------------------+---------------------------------

If the input size gets very large, the advantage of SVE/SVE2 over
Neon will fade out.

Signed-off-by: Daniel Hu <Daniel.Hu@arm.com>
Change-Id: Ieedfcb767b9c08280d7c8c9a8648919c69728fab

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18901)

Fix memory leak with TLS1.2 compression

Leak sanitizer reports following leak for ssl-test-new subtest
4-tlsv1_2-both-compress:

==335733==ERROR: LeakSanitizer: detected memory leaks
Direct leak of 17728 byte(s) in 1 object(s) allocated from:
    #0 0x3ff9fbba251 in malloc (/usr/lib64/libasan.so.8+0xba251)
    #1 0x3ff9f71744f in tls_do_uncompress ssl/record/methods/tls_common.c:868
    #2 0x3ff9f7175bd in tls_default_post_process_record ssl/record/methods/tls_common.c:896
    #3 0x3ff9f715ee7 in tls_get_more_records ssl/record/methods/tls_common.c:773
    #4 0x3ff9f712209 in tls_read_record ssl/record/methods/tls_common.c:958
    #5 0x3ff9f6ef73f in ssl3_read_bytes ssl/record/rec_layer_s3.c:1235
    #6 0x3ff9f776165 in tls_get_message_header ssl/statem/statem_lib.c:1198
    #7 0x3ff9f74709b in read_state_machine ssl/statem/statem.c:624
    #8 0x3ff9f74709b in state_machine ssl/statem/statem.c:478
    #9 0x3ff9f662e61 in SSL_do_handshake ssl/ssl_lib.c:4430
    #10 0x100c55d in do_handshake_step test/helpers/handshake.c:775
    #11 0x100c55d in do_connect_step test/helpers/handshake.c:1134
    #12 0x100e85b in do_handshake_internal test/helpers/handshake.c:1544
    #13 0x1011715 in do_handshake test/helpers/handshake.c:1738
    #14 0x101d1a7 in test_handshake test/ssl_test.c:543
    #15 0x1027875 in run_tests test/testutil/driver.c:370
    #16 0x1008393 in main test/testutil/main.c:30
    #17 0x3ff9cc2b871 in __libc_start_call_main (/usr/lib64/libc.so.6+0x2b871)
    #18 0x3ff9cc2b94f in __libc_start_main_alias_2 (/usr/lib64/libc.so.6+0x2b94f)
    #19 0x100864f  (/code/openssl/test/ssl_test+0x100864f)
Direct leak of 17728 byte(s) in 1 object(s) allocated from:
    #0 0x3ff9fbba251 in malloc (/usr/lib64/libasan.so.8+0xba251)
    #1 0x3ff9f71744f in tls_do_uncompress ssl/record/methods/tls_common.c:868
    #2 0x3ff9f7175bd in tls_default_post_process_record ssl/record/methods/tls_common.c:896
    #3 0x3ff9f715ee7 in tls_get_more_records ssl/record/methods/tls_common.c:773
    #4 0x3ff9f712209 in tls_read_record ssl/record/methods/tls_common.c:958
    #5 0x3ff9f6ef73f in ssl3_read_bytes ssl/record/rec_layer_s3.c:1235
    #6 0x3ff9f776165 in tls_get_message_header ssl/statem/statem_lib.c:1198
    #7 0x3ff9f74709b in read_state_machine ssl/statem/statem.c:624
    #8 0x3ff9f74709b in state_machine ssl/statem/statem.c:478
    #9 0x3ff9f662e61 in SSL_do_handshake ssl/ssl_lib.c:4430
    #10 0x100c55d in do_handshake_step test/helpers/handshake.c:775
    #11 0x100c55d in do_connect_step test/helpers/handshake.c:1134
    #12 0x1010b09 in do_handshake_internal test/helpers/handshake.c:1550
    #13 0x1011715 in do_handshake test/helpers/handshake.c:1738
    #14 0x101d1a7 in test_handshake test/ssl_test.c:543
    #15 0x1027875 in run_tests test/testutil/driver.c:370
    #16 0x1008393 in main test/testutil/main.c:30
    #17 0x3ff9cc2b871 in __libc_start_call_main (/usr/lib64/libc.so.6+0x2b871)
    #18 0x3ff9cc2b94f in __libc_start_main_alias_2 (/usr/lib64/libc.so.6+0x2b94f)
    #19 0x100864f  (/code/openssl/test/ssl_test+0x100864f)
SUMMARY: AddressSanitizer: 35456 byte(s) leaked in 2 allocation(s).

Fix this by freeing the SSL3_RECORD structure inside the OSSL_RECORD_LAYER.

Signed-off-by: Juergen Christ <jchrist@linux.ibm.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19030)

apps/speed.c: Wait for generated children

In multi-mode, speed fork()s off several children but does not wait for them.
On Linux, this leads to wrong accounting information of getrusage used by
tools to extract running time and page faults.

Wait for every children and check the return code and termination signal.

Signed-off-by: Juergen Christ <jchrist@linux.ibm.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19093)

Fix man page to indicate SHA256 MAC for PKCS12

CLA: trivial

Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19079)

Test that we ignore a bad record version in a plaintext TLSv1.3 record

The RFC requires us to ignore this field in plaintext records - so even
if it is set incorrectly we should tolerate it.

Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19058)

Tolerate a bad record version in TLSv1.3 plaintext records

When a server responds to a second TLSv1.3 ClientHello it is required to
set the legacy_record_version to 0x0303 (TLSv1.2). The client is required
to ignore that field even if it is wrong. The recent changes to the read
record layer in PR #18132 made the record layer stricter and it was
checking that the legacy_record_version was the correct value. This
caused connection failures when talking to buggy servers that set the
wrong legacy_record_version value.

We make us more tolerant again.

Fixes #19051

Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19058)

Add design requirements for QUIC packet demuxer

Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18249)

ERR: Make CRYPTO_malloc() and friends report ERR_R_MALLOC_FAILURE

Fixes #6251

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/14833)

Convert serverinfo in SSL_CTX_use_serverinfo() to v2.

Fixes openssl#18183.

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18614)

Add missing ')' to command help

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19049)

Drop the optimisation level for ppc64le cross-compile

The default cross compiler (gcc 9.4.0) for ppc64le on Ubunut 20.04 seems
buggy and causes a seg fault in sslapitest. This doesn't impact any other
CI cross compile platforms and does not seem to impact the gcc 10.3.0 cross
compiler.

We just drop the optimisation level on that platform.

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19056)

Updates for OSSL_TIME changes

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18676)

QUIC ACK Manager, Statistics Manager and Congestion Control API

This is the initial implementation of the ACK Manager for OpenSSL's QUIC
support, with supporting design documentation and tests.

Because the ACK Manager also depends on the Statistics Manager, it is
also implemented here. The Statistics Manager is quite simple, so this
does not amount to a large amount of extra code.

Because the ACK Manager depends on a congestion controller, it adds a
no-op congestion controller, which uses the previously workshopped
congestion control API.

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18676)

CMP: correct handling of fallback subject in OSSL_CMP_CTX_setup_CRM() and its doc

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/18929)

CMP: fix crash in check_transactionID_or_nonce() on 'actual' being NULL

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/18929)

CRMF: make create_popo_signature() check that pubkey and pkey match

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/18930)

add missing CRMF API function OSSL_CRMF_CERTTEMPLATE_get0_publicKey()

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/18930)

X509: clean up doc and implementation of X509{,_REQ}_check_private_key()

Also constify X509_REQ_get0_pubkey() and X509_REQ_check_private_key().

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/18930)

x509_att.c: improve error checking and reporting and coding style

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/18931)

add missing doc of X509_REQ_get_extensions() and X509_REQ_add_extensions{,_nid}()

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/18931)

X509_REQ_get_extensions: add error queue entry on ill-formed extensions attribute

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/18931)

util/wrap.pl.in: If the subprocess died with a signal, let's re-signal it

A simple 'kill' of the same signal on our own process should do it.
This will allow the shell that this is running under to catch it
properly, and output something if it usually does that.

Fixes #19041

Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19042)

Fix the return type for the rlayer_skip_early_data callback

There was a copy & paste error in the definition of the
rlayer_skip_early_data callback. The return type is supposed to
be "int" but it was defined as a pointer type. This was causing
test failures on some platforms.

Fixes #19037

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/19048)

rl->enc_ctx must be non-NULL and cipher must be set

Otherwise ssl3_cipher() cannot work properly.

Fixes Coverity CID 1509401

Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19027)

Do not use RLAYERfatal on NULL RLAYER

or on record layer that is to be freed anyway.

Fixes Coverity CID 1509402, 1509403

Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19027)

Coverity 1508532: out of bounds access

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19033)

OSSL_HTTP_REQ_CTX_nbio: add support for partial content-type string matching

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/18675)

app_http_tls_cb() and tls_error_hint(): code cleanup

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/18675)

Makefile: Generate crypto objects only as far as needed

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/15224)

Makefile: Call mknum.pl on 'make ordinals' only if needed

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/15224)

Support decode SM2 parameters

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18819)

Moving notify check after the no time check

CLA: trivial

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19007)

openssl speed fails in FIPS mode

...because it uses md5 for HMAC tests. Skip md5 in case of its
unavailability.

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17331)

Fix typo in migration guide

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19032)

Always automatically add -DPEDANTIC with enable-ubsan

To avoid reports like: #19028

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19029)

Fix doc-nits

PR #19031 updated options that that were listed as commands, these
options were already in openssl-list.pod.in, so they are redundant
in openssl.pod.

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19044)

evp: Use functions instead of direct structure field references

AES and chacha20poly1305 also have some codes that directly reference
the fields in the EVP_CIPHER_CTX structure, such as 'ctx->buf' and
'ctx->encrypt', in order to make the code style uniform, use the
corresponding interface API instead of direct field references.

Signed-off-by: Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16465)

evp: Simplify ARIA aead cipher definition

Remove fixed macro variables, only keep the cipher mode name and
key length.

Signed-off-by: Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16465)

Add BSD-armv4 target based on linux-armv4

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Todd Short <todd.short@me.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18910)

Clarify dashes are required for openssl list command

Fixes #19013

CLA: trivial

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19031)

Update gitignore

Add test/timing_load_creds

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19021)

BIO_sendmmsg/BIO_recvmmsg (API only)

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18923)

Limit the size of various MAXCHUNK definitions

The current code has issues when sizeof(long) <> sizeof(size_t).  The two
types are assumed to be interchangeable and them being different will
cause crashes and endless loops.

This fix limits the maximum chunk size for many of the symmetric ciphers
to 2^30 bytes.  This chunk size limits the amount of data that will
be encrypted/decrypted in one lump.  The code internally handles block
of data later than the chunk limit, so this will present no difference
to the caller.  Any loss of efficiency due to limiting the chunking to
1Gbyte rather than more should be insignificant.

Fixes Coverity issues:
    1508498, 1508500 - 1508505, 1508507 - 1508527, 1508529 - 1508533,
    1508535 - 1508537, 1508539, 1508541 - 1508549, 1508551 - 1508569 &
    1508571 - 1508582.

Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18997)

Coverity: misuses of time_t

Coverity 1508506:

    Fixes a bug in the cookie code which would have caused problems for
    ten minutes before and after the lower 32 bits of time_t rolled over.

Coverity 1508534 & 1508540:

    Avoid problems when the lower 32 bits of time_t roll over by delaying
    the cast to integer until after the time delta has been computed.

Reviewed-by: Ben Kaduk <kaduk@mit.edu>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19004)

Check record layer callbacks are non-null

The current libssl code always ensures that the callbacks are non-null.
However, the record layer itself wasn't checkthing this. We ensure it does.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Formatting cleanups

Some minor formatting cleanups and other minor tweaks.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Remove redefinition of macros

Some macros were redefined in ssl3_cbc.c. We remove the redefinitions

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Remove redefinition of SSL_AD_NO_ALERT

The SSL_AD_NO_ALERT value was defined in two places. We centralise its
definition.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Ensure the rrl object is set to NULL after it is freed

Once we free the rrl object we should NULL it to prevent a dangling ref
to it. Otherwise we could get a double free.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Rename some functions to be more consistent

Some functions in the record layer were called rlayer_*, but most were
called tls_*. We standardise on the latter.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Rename DTLS1_BITMAP to DTLS_BITMAP

The 1 in DTLS1 is confusing and is removed. We also tweak the structure
to always be able to track 64 packets regardless of whether we are on a
32 bit or 64 bit system.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Remove ssl/record/README.md

This file contains design details for the old record layer and is no
longer relevant for the new design.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Standardise type for epoch

The value for epoch was being represented internally via various types:
uint16_t, unsigned short, unsigned int, unsigned long

We standardise on uint16_t

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Remove the SSL3_RECORD read field

The read field is no longer used and can be safely removed.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Add a CHANGES.md entry for the record layer changes

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Clean up some SCTP releated issues

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Ensure various record layer options can be updated

We would like the capability for the options/mode/read_ahead settings
to be updateable after the record layer object has been instantiated.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Remove some redundant code

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Update the tls13encryptiontest for new read record layer

The tls13encryption is an internal test that reaches inside libssl
to test encryption/decryption of records. It needs to be amended for the
new code structure so that it is testing the equivalent things as before.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Implement a human readable state function for the record layer

This allows querying of the record layer to get a human readable state
string out. This resolves two outstanding TODO comments and enables us
to remove the rstate variable from s->rlayer.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Correct some formatting errors in tls1_meth.c

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Move SSL_MODE_RELEASE_BUFFERS into the read record layer

This resolves an outstanding "TODO" item.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Remove a redundant TODO

If read_ahead data is left over when a record layer closes down, there is
already code present to push into the "next" BIO. So the TODO removed here
is no longer relevant.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Cleanse the SSLv3 MAC secret when we clean up the read record layer

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Move some fields out of the SSL object and into the record layer object

Fields such as rrlmethod and rrl are entirely related to the record layer,
and so should be in that object.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Remove an unnecessary setup of the read buffer

Now everything has been moved to the record layer the additional check
for setting up the read buffer is not needed and can be removed.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Clear away some unused fields and cruft in the record layer

Now that the read record layer has moved to the new architecture we can
clear some of the old stuff away.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Remove some references to rlayer.rstate

This also fixes ssl3_pending while we are at it

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Ensure we use a dgram mem BIO for the DTLS record layer

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Remove some more redundant TODO(RECLAYER) comments

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Resolve a TODO(RECLAYER) in the SSLv3 code

We remove some code outside of the record layer which is no longer
relevant since its functions are now performed by the new record layer
code. This removes a TODO(RECLAYER) as a result.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Remove some TODO(RECLAYER) comments

Some TODO(RECLAYER) comments are no longer necessary and can be removed.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Remove some unnecessary function pointers from OSSL_RECORD_METHOD

We had some temporary function pointers in OSSL_RECORD_METHOD which were
only necessary during the process of refactoring the read record layer.
These are no longer required so can be removed.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Standardise some DTLS record layer naming

Remove references to dtls1_* and just use dtls_*

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Remove some remaining SSL object references from DTLS record layer

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Push unprocessed DTLS records from one record layer object to next

We add unprocessed DTLS records to the unprocessed record queue. When
the record layer closes down we write the unprocessed records to the
next record layer object.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Remove reliance on the SSL object from the DTLS read record layer code

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Move some DTLS read code into the read record layer

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Remove some final references to the SSL object in the record layer

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Remove the separation betweeen enc_read_ctx and enc_write_ctx

Similarly with read_hash and write_hash. In the new model we have a
separate record layer object for reading and writing. Therefore we don't
need to distinguish between reading and writing inside the record layer
object in the encryption and md ctxs.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Enable the record layer to call the ssl_security callback

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Use a record layer specific message callback

Don't use the message callback from the SSL object. Instead we use a
wrapper callback so that the record layer does not need to be aware of the
SSL object.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Move early data counting out of the SSL object and into the record layer

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Move the sequence number into the OSSL_RECORD_LAYER object

This removes some references to the SSL object from the record layer.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Remove use of SSL object for fragment length checking in record layer

Pass the max fragment length to the record layer when it is applicable
to avoid the need to go through the SSL object.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Remove use of ossl_statem_in_error() from the record layer

We remove the dependency on the SSL object. Instead we check if the
record layer has already set an alert code.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Make the record layer directly aware of EtM

We no longer have to go through the SSL object to discover whether EtM has
been negotiated.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Remove SSL_USE_EXPLICT_IV() from the record layer methods

Instead we introduce RLAYER_USE_EXPLICIT_IV(). This removes a dependency
on the SSL object.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)

Remove some use of SSL object from record layer

Remove SSL_IS_TLS13() and hello_retry_request

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18132)