OpenSSL 3 supports a modular architecture, allowing different providers
to bring specific implementations of cryptographical algorithms. One
such provider, "fips", ships with OpenSSL 3 directly, and groups
algorithms that can be FIPS 140-2 validated.
The import of OpenSSL 3.0.9 was building this provider incorrectly,
missing symbols required for proper operation.
In addition, without the change in OpenSSL's crypto/bn/bn_const.c, the
FIPS module fails loading: `Undefined symbol "ossl_bignum_modp_1536_p"`.
This change is consistent with crypto/bn/bn_dh.c though.
Sponsored by: The FreeBSD Foundation
Pull Request: https://github.com/freebsd/freebsd-src/pull/787
They break the !amd64 builds due to an underspecified include path and
will be re-applied once that's fixed.
Reported by: Ronald Klop <ronald-lists@klop.ws>
OpenSSL 3 supports a modular architecture, allowing different providers
to bring specific implementations of cryptographical algorithms. One
such provider, "fips", ships with OpenSSL 3 directly, and groups
algorithms that can be FIPS 140-2 validated.
The import of OpenSSL 3.0.9 was building this provider incorrectly,
missing symbols required for proper operation.
In addition, without the change in OpenSSL's crypto/bn/bn_const.c, the
FIPS module fails loading: `Undefined symbol "ossl_bignum_modp_1536_p"`.
This change is consistent with crypto/bn/bn_dh.c though.
Sponsored by: The FreeBSD Foundation
Pull Request: https://github.com/freebsd/freebsd-src/pull/787
Migrate to OpenSSL 3.0 in advance of FreeBSD 14.0. OpenSSL 1.1.1 (the
version we were previously using) will be EOL as of 2023-09-11.
Most of the base system has already been updated for a seamless switch
to OpenSSL 3.0. For many components we've added
`-DOPENSSL_API_COMPAT=0x10100000L` to CFLAGS to specify the API version,
which avoids deprecation warnings from OpenSSL 3.0. Changes have also
been made to avoid OpenSSL APIs that were already deprecated in OpenSSL
1.1.1. The process of updating to contemporary APIs can continue after
this merge.
Additional changes are still required for libarchive and Kerberos-
related libraries or tools; workarounds will immediately follow this
commit. Fixes are in progress in the upstream projects and will be
incorporated when those are next updated.
There are some performance regressions in benchmarks (certain tests in
`openssl speed`) and in some OpenSSL consumers in ports (e.g. haproxy).
Investigation will continue for these.
Netflix's testing showed no functional regression and a rather small,
albeit statistically significant, increase in CPU consumption with
OpenSSL 3.0.
Thanks to ngie@ and des@ for updating base system components, to
antoine@ and bofh@ for ports exp-runs and port fixes/workarounds, and to
Netflix and everyone who tested prior to commit or contributed to this
update in other ways.
PR: 271615
PR: 271656 [exp-run]
Relnotes: Yes
Sponsored by: The FreeBSD Foundation
"adrl" is a pseudo-instruction used to calculate an address relative
to PC. It's not recognized by clang resulting in a compilation error.
I've stumbled upon it when trying to integrate the bsaes-armv7 assmebly
logic into FreeBSD kernel, which uses clang as it's default compiler.
Note that this affect the build only if BSAES_ASM_EXTENDED_KEY is
defined, which is not the default option in OpenSSL.
The solution here is to replace it with an add instruction.
This mimics what has already been done in !BSAES_ASM_EXTENDED_KEY logic.
Because of that I've marked this as trivial CLA.
No objections from: jkim
Obtained from: OpenSSL commit 27093ba73372935fe4ef91d0a45ce6ea90a1ac8e
Differential Revision: https://reviews.freebsd.org/D39091
This removes a guard condition that prevents KTLS being enabled for
receiving in TLS 1.3. Use the correct sequence number and BIO for
receive vs transmit offload.
Approved by: jkim
Obtained from: OpenSSL commit 7c78932b9a4330fb7c8db72b3fb37cbff1401f8b
MFC after: 1 week
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D34976
- Don't unpad records, check the outer record type, or extract the
inner record type from TLS 1.3 records handled by the kernel. KTLS
performs all of these steps and returns the inner record type in the
TLS header.
- When checking the length of a received TLS 1.3 record don't allow
for the extra byte for the nested record type when KTLS is used.
- Pass a pointer to the record type in the TLS header to the
SSL3_RT_INNER_CONTENT_TYPE message callback. For KTLS, the old
pointer pointed to the last byte of payload rather than the record
type. For the non-KTLS case, the TLS header has been updated with
the inner type before this callback is invoked.
Approved by: jkim
Obtained from: OpenSSL commit a5fb9605329fb939abb536c1604d44a511741624
MFC after: 1 week
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D34975
When KTLS receive is enabled, pending data may still be present due to
read ahead. This data must still be processed the same as records
received without KTLS. To ease readability (especially in
consideration of additional checks which will be added for TLS 1.3),
add a helper variable 'using_ktls' that is true when the KTLS receive
path is being used to receive a record.
Approved by: jkim
Obtained from: OpenSSL commit 031132c297e54cbc20404a0bf8de6ed863196399
MFC after: 1 week
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D34974
KTLS implementations currently assume that the start of the in-kernel
socket buffer is aligned with the start of a TLS record for the
receive side. The socket option to enable KTLS specifies the TLS
sequence number of this initial record.
When read ahead is enabled, data can be pending in the SSL read buffer
after negotiating session keys. This pending data must be examined to
ensurs that the kernel's socket buffer does not contain a partial TLS
record as well as to determine the correct sequence number of the
first TLS record to be processed by the kernel.
In preparation for enabling receive kernel offload for TLS 1.3, move
the existing logic to handle read ahead from t1_enc.c into ktls.c and
invoke it from ktls_configure_crypto().
Approved by: jkim
Obtained from: OpenSSL commit 85773128d0e80cd8dcc772a6931d385b8cf4acd1
MFC after: 1 week
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D34973
In some corner cases the check for packets
which exceed the allowed record length was missing
when KTLS is initially enabled, when some
unprocessed packets are still pending.
Approved by: jkim
Obtained from: OpenSSL commit 8fff986d52606e1a33f9404504535e2e2aee3e8b
MFC after: 1 week
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D34972
Also add comment to the public header to avoid
making another conflict in future.
Reviewed by: jkim
Obtained from: OpenSSL commit 5d4975ecd88ac17d0749513a8fac9a7c7befd900
MFC after: 1 week
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D34135
OpenSSL assumes the same value for AT_HWCAP=16 (Linux)
So it ends up calling elf_auxv_info() with AT_CANARY which
returns ENOENT, and all acceleration features are disabled.
With this, my ARM64 test machine runs the benchmark
`openssl speed -evp aes-256-gcm` nearly 20x faster
going from 100 MB/sec to 2000 MB/sec
It also improves sha256 from 300 MB/sec to 1800 MB/sec
This fix has been accepted but not yet merged upstream:
https://github.com/openssl/openssl/pull/17082
PR: 259937
Reviewed by: manu, imp
MFC after: immediate
Relnotes: yes
Fixes: 88e852c0b5 ("OpenSSL: Merge OpenSSL 1.1.1j")
Sponsored by: Ampere Computing LLC
Sponsored by: Klara Inc.
Differential Revision: https://reviews.freebsd.org/D33060
FreeBSD's kernel TLS supports Chacha20 for both TLS 1.2 and TLS 1.3.
NB: This commit has not yet been merged upstream as it is deemed a new
feature and did not make the feature freeze cutoff for OpenSSL 3.0.
Reviewed by: jkim
MFC after: 5 days
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D31443
Most of this upstream commit touched tests not included in the
vendor import. The one change merged in is to remove a constant
only present in an internal header to appease the older tests.
Reviewed by: jkim
Obtained from: OpenSSL (e1fdd5262e4a45ce3aaa631768e877ee7b6da21b)
MFC after: 5 days
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D31442
KTLS support has been changed to be off by default, and configuration is
via a single "option" rather two "modes". Documentation is updated
accordingly.
Reviewed by: jkim
Obtained from: OpenSSL (6878f4300213cfd7d4f01e26a8b97f70344da100)
MFC after: 5 days
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D31441
It has always been the case that KTLS is not compiled by default. However
if it is compiled then it was automatically used unless specifically
configured not to. This is problematic because it avoids any crypto
implementations from providers. A user who configures all crypto to use
the FIPS provider may unexpectedly find that TLS related crypto is actually
being performed outside of the FIPS boundary.
Instead we change KTLS so that it is disabled by default.
We also swap to using a single "option" (i.e. SSL_OP_ENABLE_KTLS) rather
than two separate "modes", (i.e. SSL_MODE_NO_KTLS_RX and
SSL_MODE_NO_KTLS_TX).
Reviewed by: jkim
Obtained from: OpenSSL (a3a54179b6754fbed6d88e434baac710a83aaf80)
MFC after: 5 days
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D31440
Linux kernel is going to support ChaCha20-Poly1305 in TLS offload.
Add support for this cipher.
Reviewed by: jkim
Obtained from: OpenSSL (3aa7212e0a4fd1533c8a28b8587dd8b022f3a66f)
MFC after: 5 days
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D31439
BIO_get_ktls_send() and BIO_get_ktls_recv() are documented as
returning either 0 or 1. However, they were actually returning the
internal value of the associated BIO flag for the true case instead of
1.
Also trim redundant ternary operators.
Reviewed by: jkim
Obtained from: OpenSSL (f16e52b67c9261bdc7e1284a50502a802921ac6d)
MFC after: 5 days
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D31438
Add a handler for EBUSY sendfile error in addition to
EAGAIN. With EBUSY returned the data still can be partially
sent and user code has to be notified about it, otherwise it
may try to send data multiple times.
PR: 251969
Reviewed by: jkim
Obtained from: OpenSSL (dfcfd17f2818cf520ce6381aed9ec3d2fc12170d)
MFC after: 1 week
Sponsored by: Netflix (merging to FreeBSD)
Differential Revision: https://reviews.freebsd.org/D28714
This merges upstream patches from OpenSSL's master branch to add
KTLS infrastructure for TLS 1.0-1.3 including both RX and TX
offload and SSL_sendfile support on both Linux and FreeBSD.
Note that TLS 1.3 only supports TX offload.
A new WITH/WITHOUT_OPENSSL_KTLS determines if OpenSSL is built with
KTLS support. It defaults to enabled on amd64 and disabled on all
other architectures.
Reviewed by: jkim (earlier version)
Approved by: secteam
Obtained from: OpenSSL (patches from master)
MFC after: 1 week
Relnotes: yes
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D28273
OpenSSL commit 3db2c9f3:
Complain if we are attempting to encode with an invalid ASN.1 template
OpenSSL commit 43a7033:
Check that multi-strings/CHOICE types don't use implicit tagging
OpenSSL commit f960d812:
Correctly compare EdiPartyName in GENERAL_NAME_cmp()
Obtained from: OpenSSL 3db2c9f3, 43a7033, f960d812
Security: CVE-2020-1971
