We don't typically print anything when a subsystem initializes itself,
and KTLS is currently disabled by default anyway.
Reviewed by: jhb
MFC after: 1 week
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D29097
Maintain a cache of physically contiguous runs of pages for use as
output buffers when software encryption is configured and in-place
encryption is not possible. This makes allocation and free cheaper
since in the common case we avoid touching the vm_page structures for
the buffer, and fewer calls into UMA are needed. gallatin@ reports a
~10% absolute decrease in CPU usage with sendfile/KTLS on a Xeon after
this change.
It is possible that we will not be able to allocate these buffers if
physical memory is fragmented. To avoid frequently calling into the
physical memory allocator in this scenario, rate-limit allocation
attempts after a failure. In the failure case we fall back to the old
behaviour of allocating a page at a time.
N.B.: this scheme could be simplified, either by simply using malloc()
and looking up the PAs of the pages backing the buffer, or by falling
back to page by page allocation and creating a mapping in the cache
zone. This requires some way to save a mapping of an M_EXTPG page array
in the mbuf, though. m_data is not really appropriate. The second
approach may be possible by saving the mapping in the plinks union of
the first vm_page structure of the array, but this would force a vm_page
access when freeing an mbuf.
Reviewed by: gallatin, jhb
Tested by: gallatin
Sponsored by: Ampere Computing
Submitted by: Klara, Inc.
Differential Revision: https://reviews.freebsd.org/D28556
I missed updating this counter when rebasing the changes in
9c64fc4029 after the switch to
COUNTER_U64_DEFINE_EARLY in 1755b2b989.
Fixes: 9c64fc4029 Add Chacha20-Poly1305 as a KTLS cipher suite.
Sponsored by: Netflix
Chacha20-Poly1305 for TLS is an AEAD cipher suite for both TLS 1.2 and
TLS 1.3 (RFCs 7905 and 8446). For both versions, Chacha20 uses the
server and client IVs as implicit nonces xored with the record
sequence number to generate the per-record nonce matching the
construction used with AES-GCM for TLS 1.3.
Reviewed by: gallatin
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D27839
This makes it a bit more straightforward to add new counters when
debugging. No functional change intended.
Reviewed by: jhb
Sponsored by: Ampere Computing
Submitted by: Klara, Inc.
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D28498
Originally IFCAP_NOMAP meant that the mbuf has external storage pointer
that points to unmapped address. Then, this was extended to array of
such pointers. Then, such mbufs were augmented with header/trailer.
Basically, extended mbufs are extended, and set of features is subject
to change. The new name should be generic enough to avoid further
renaming.
- Only check for empty domains if we actually tried to configure domain
affinity in the first place. Otherwise setting bind_threads=1 will
always cause the sysctl value to be reported as zero. This is
harmless since the threads end up being bound, but it's confusing.
- Try to improve the sysctl description a bit.
Reviewed by: gallatin, jhb
Submitted by: Klara, Inc.
Sponsored by: Ampere Computing
Differential Revision: https://reviews.freebsd.org/D28161
When ktls_bind_thread is 2, we pick a ktls worker thread that is
bound to the same domain as the TCP connection associated with
the socket. We use roughly the same code as netinet/tcp_hpts.c to
do this. This allows crypto to run on the same domain as the TCP
connection is associated with. Assuming TCP_REUSPORT_LB_NUMA
(D21636) is in place & in use, this ensures that the crypto source
and destination buffers are local to the same NUMA domain as we're
running crypto on.
This change (when TCP_REUSPORT_LB_NUMA, D21636, is used) reduces
cross-domain traffic from over 37% down to about 13% as measured
by pcm.x on a dual-socket Xeon using nginx and a Netflix workload.
Reviewed by: jhb
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D21648
This gives a more uniform API for send tag life cycle management.
Reviewed by: gallatin, hselasky
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D27000
- Add a new send tag type for a send tag that supports both rate
limiting (packet pacing) and TLS offload (mostly similar to D22669
but adds a separate structure when allocating the new tag type).
- When allocating a send tag for TLS offload, check to see if the
connection already has a pacing rate. If so, allocate a tag that
supports both rate limiting and TLS offload rather than a plain TLS
offload tag.
- When setting an initial rate on an existing ifnet KTLS connection,
set the rate in the TCP control block inp and then reset the TLS
send tag (via ktls_output_eagain) to reallocate a TLS + ratelimit
send tag. This allocates the TLS send tag asynchronously from a
task queue, so the TLS rate limit tag alloc is always sleepable.
- When modifying a rate on a connection using KTLS, look for a TLS
send tag. If the send tag is only a plain TLS send tag, assume we
failed to allocate a TLS ratelimit tag (either during the
TCP_TXTLS_ENABLE socket option, or during the send tag reset
triggered by ktls_output_eagain) and ignore the new rate. If the
send tag is a ratelimit TLS send tag, change the rate on the TLS tag
and leave the inp tag alone.
- Lock the inp lock when setting sb_tls_info for a socket send buffer
so that the routines in tcp_ratelimit can safely dereference the
pointer without needing to grab the socket buffer lock.
- Add an IFCAP_TXTLS_RTLMT capability flag and associated
administrative controls in ifconfig(8). TLS rate limit tags are
only allocated if this capability is enabled. Note that TLS offload
(whether unlimited or rate limited) always requires IFCAP_TXTLS[46].
Reviewed by: gallatin, hselasky
Relnotes: yes
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D26691
The TF_TOE flag is the check used in the rest of the network stack to
determine if TOE is active on a socket. There is at least one path in
the cxgbe(4) TOE driver that can leave the tod pointer non-NULL on a
socket not using TOE.
Reported by: Sony Arpita Das <sonyarpitad@chelsio.com>
Reviewed by: np
Sponsored by: Chelsio Communications
Differential Revision: https://reviews.freebsd.org/D26803
Due to a weakness in the TLS 1.0 protocol, OpenSSL will periodically
send empty TLS records ("empty fragments"). These TLS records have no
payload (and thus a page count of zero). m_uiotombuf_nomap() was
returning NULL instead of an empty mbuf, and a few places needed to be
updated to treat an empty TLS record as having a page count of "1" as
0 means "no work to do" (e.g. nothing to encrypt, or nothing to mark
ready via sbready()).
Reviewed by: gallatin
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D26729
m_segments() was added with r363464 but never used. Remove it to
avoid warnings when compiling kernels.
Reported by: rmacklem (also says jhb)
Reviewed by: gallatin, jhb
Differential Revision: https://reviews.freebsd.org/D26330
When using ifnet ktls, and when ktls_reset_send_tag()
fails to allocate a replacement tag, it leaves
the tls session's snd_tag pointer NULL. ktls_cleanup()
tries to release the send tag, and will trip over
this NULL pointer and panic unless NULL is checked for.
Reviewed by: jhb
Sponsored by: Netflix
Allow TLS records to be decrypted in the kernel after being received
by a NIC. At a high level this is somewhat similar to software KTLS
for the transmit path except in reverse. Protocols enqueue mbufs
containing encrypted TLS records (or portions of records) into the
tail of a socket buffer and the KTLS layer decrypts those records
before returning them to userland applications. However, there is an
important difference:
- In the transmit case, the socket buffer is always a single "record"
holding a chain of mbufs. Not-yet-encrypted mbufs are marked not
ready (M_NOTREADY) and released to protocols for transmit by marking
mbufs ready once their data is encrypted.
- In the receive case, incoming (encrypted) data appended to the
socket buffer is still a single stream of data from the protocol,
but decrypted TLS records are stored as separate records in the
socket buffer and read individually via recvmsg().
Initially I tried to make this work by marking incoming mbufs as
M_NOTREADY, but there didn't seemed to be a non-gross way to deal with
picking a portion of the mbuf chain and turning it into a new record
in the socket buffer after decrypting the TLS record it contained
(along with prepending a control message). Also, such mbufs would
also need to be "pinned" in some way while they are being decrypted
such that a concurrent sbcut() wouldn't free them out from under the
thread performing decryption.
As such, I settled on the following solution:
- Socket buffers now contain an additional chain of mbufs (sb_mtls,
sb_mtlstail, and sb_tlscc) containing encrypted mbufs appended by
the protocol layer. These mbufs are still marked M_NOTREADY, but
soreceive*() generally don't know about them (except that they will
block waiting for data to be decrypted for a blocking read).
- Each time a new mbuf is appended to this TLS mbuf chain, the socket
buffer peeks at the TLS record header at the head of the chain to
determine the encrypted record's length. If enough data is queued
for the TLS record, the socket is placed on a per-CPU TLS workqueue
(reusing the existing KTLS workqueues and worker threads).
- The worker thread loops over the TLS mbuf chain decrypting records
until it runs out of data. Each record is detached from the TLS
mbuf chain while it is being decrypted to keep the mbufs "pinned".
However, a new sb_dtlscc field tracks the character count of the
detached record and sbcut()/sbdrop() is updated to account for the
detached record. After the record is decrypted, the worker thread
first checks to see if sbcut() dropped the record. If so, it is
freed (can happen when a socket is closed with pending data).
Otherwise, the header and trailer are stripped from the original
mbufs, a control message is created holding the decrypted TLS
header, and the decrypted TLS record is appended to the "normal"
socket buffer chain.
(Side note: the SBCHECK() infrastucture was very useful as I was
able to add assertions there about the TLS chain that caught several
bugs during development.)
Tested by: rmacklem (various versions)
Relnotes: yes
Sponsored by: Chelsio Communications
Differential Revision: https://reviews.freebsd.org/D24628
In addition to reducing lines of code, this also ensures that the full
allocation is always zeroed avoiding possible bugs with incorrect
lengths passed to explicit_bzero().
Suggested by: cem
Reviewed by: cem, delphij
Approved by: csprng (cem)
Sponsored by: Chelsio Communications
Differential Revision: https://reviews.freebsd.org/D25435
next commit brings in second flag, so let them already be in the
future namespace.
Reviewed by: gallatin
Differential Revision: https://reviews.freebsd.org/D24598
The following series of patches addresses three things:
Now that array of pages is embedded into mbuf, we no longer need
separate structure to pass around, so struct mbuf_ext_pgs is an
artifact of the first implementation. And struct mbuf_ext_pgs_data
is a crutch to accomodate the main idea r359919 with minimal churn.
Also, M_EXT of type EXT_PGS are just a synonym of M_NOMAP.
The namespace for the newfeature is somewhat inconsistent and
sometimes has a lengthy prefixes. In these patches we will
gradually bring the namespace to "m_epg" prefix for all mbuf
fields and most functions.
Step 1 of 4:
o Anonymize mbuf_ext_pgs_data, embed in m_ext
o Embed mbuf_ext_pgs
o Start documenting all this entanglement
Reviewed by: gallatin
Differential Revision: https://reviews.freebsd.org/D24598
- Add a new TCP_RXTLS_ENABLE socket option to set the encryption and
authentication algorithms and keys as well as the initial sequence
number.
- When reading from a socket using KTLS receive, applications must use
recvmsg(). Each successful call to recvmsg() will return a single
TLS record. A new TCP control message, TLS_GET_RECORD, will contain
the TLS record header of the decrypted record. The regular message
buffer passed to recvmsg() will receive the decrypted payload. This
is similar to the interface used by Linux's KTLS RX except that
Linux does not return the full TLS header in the control message.
- Add plumbing to the TOE KTLS interface to request either transmit
or receive KTLS sessions.
- When a socket is using receive KTLS, redirect reads from
soreceive_stream() into soreceive_generic().
- Note that this interface is currently only defined for TLS 1.1 and
1.2, though I believe we will be able to reuse the same interface
and structures for 1.3.
This change is build on top of nexthop objects introduced in r359823.
Nexthops are separate datastructures, containing all necessary information
to perform packet forwarding such as gateway interface and mtu. Nexthops
are shared among the routes, providing more pre-computed cache-efficient
data while requiring less memory. Splitting the LPM code and the attached
data solves multiple long-standing problems in the routing layer,
drastically reduces the coupling with outher parts of the stack and allows
to transparently introduce faster lookup algorithms.
Route caching was (re)introduced to minimise (slow) routing lookups, allowing
for notably better performance for large TCP senders. Caching works by
acquiring rtentry reference, which is protected by per-rtentry mutex.
If the routing table is changed (checked by comparing the rtable generation id)
or link goes down, cache record gets withdrawn.
Nexthops have the same reference counting interface, backed by refcount(9).
This change merely replaces rtentry with the actual forwarding nextop as a
cached object, which is mostly mechanical. Other moving parts like cache
cleanup on rtable change remains the same.
Differential Revision: https://reviews.freebsd.org/D24340
While the original implementation of unmapped mbufs was a large
step forward in terms of reducing cache misses by enabling mbufs
to carry more than a single page for sendfile, they are rather
cache unfriendly when accessing the ext_pgs metadata and
data. This is because the ext_pgs part of the mbuf is allocated
separately, and almost guaranteed to be cold in cache.
This change takes advantage of the fact that unmapped mbufs
are never used at the same time as pkthdr mbufs. Given this
fact, we can overlap the ext_pgs metadata with the mbuf
pkthdr, and carry the ext_pgs meta directly in the mbuf itself.
Similarly, we can carry the ext_pgs data (TLS hdr/trailer/array
of pages) directly after the existing m_ext.
In order to be able to carry 5 pages (which is the minimum
required for a 16K TLS record which is not perfectly aligned) on
LP64, I've had to steal ext_arg2. The only user of this in the
xmit path is sendfile, and I've adjusted it to use arg1 when
using unmapped mbufs.
This change is almost entirely mechanical, except that we
change mb_alloc_ext_pgs() to no longer allow allocating
pkthdrs, the change to avoid ext_arg2 as mentioned above,
and the removal of the ext_pgs zone,
This change saves roughly 2% "raw" CPU (~59% -> 57%), or over
3% "scaled" CPU on a Netflix 100% software kTLS workload at
90+ Gb/s on Broadwell Xeons.
In a follow-on commit, I plan to remove some hacks to avoid
access ext_pgs fields of mbufs, since they will now be in
cache.
Many thanks to glebius for helping to make this better in
the Netflix tree.
Reviewed by: hselasky, jhb, rrs, glebius (early version)
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D24213
- The linked list of cryptoini structures used in session
initialization is replaced with a new flat structure: struct
crypto_session_params. This session includes a new mode to define
how the other fields should be interpreted. Available modes
include:
- COMPRESS (for compression/decompression)
- CIPHER (for simply encryption/decryption)
- DIGEST (computing and verifying digests)
- AEAD (combined auth and encryption such as AES-GCM and AES-CCM)
- ETA (combined auth and encryption using encrypt-then-authenticate)
Additional modes could be added in the future (e.g. if we wanted to
support TLS MtE for AES-CBC in the kernel we could add a new mode
for that. TLS modes might also affect how AAD is interpreted, etc.)
The flat structure also includes the key lengths and algorithms as
before. However, code doesn't have to walk the linked list and
switch on the algorithm to determine which key is the auth key vs
encryption key. The 'csp_auth_*' fields are always used for auth
keys and settings and 'csp_cipher_*' for cipher. (Compression
algorithms are stored in csp_cipher_alg.)
- Drivers no longer register a list of supported algorithms. This
doesn't quite work when you factor in modes (e.g. a driver might
support both AES-CBC and SHA2-256-HMAC separately but not combined
for ETA). Instead, a new 'crypto_probesession' method has been
added to the kobj interface for symmteric crypto drivers. This
method returns a negative value on success (similar to how
device_probe works) and the crypto framework uses this value to pick
the "best" driver. There are three constants for hardware
(e.g. ccr), accelerated software (e.g. aesni), and plain software
(cryptosoft) that give preference in that order. One effect of this
is that if you request only hardware when creating a new session,
you will no longer get a session using accelerated software.
Another effect is that the default setting to disallow software
crypto via /dev/crypto now disables accelerated software.
Once a driver is chosen, 'crypto_newsession' is invoked as before.
- Crypto operations are now solely described by the flat 'cryptop'
structure. The linked list of descriptors has been removed.
A separate enum has been added to describe the type of data buffer
in use instead of using CRYPTO_F_* flags to make it easier to add
more types in the future if needed (e.g. wired userspace buffers for
zero-copy). It will also make it easier to re-introduce separate
input and output buffers (in-kernel TLS would benefit from this).
Try to make the flags related to IV handling less insane:
- CRYPTO_F_IV_SEPARATE means that the IV is stored in the 'crp_iv'
member of the operation structure. If this flag is not set, the
IV is stored in the data buffer at the 'crp_iv_start' offset.
- CRYPTO_F_IV_GENERATE means that a random IV should be generated
and stored into the data buffer. This cannot be used with
CRYPTO_F_IV_SEPARATE.
If a consumer wants to deal with explicit vs implicit IVs, etc. it
can always generate the IV however it needs and store partial IVs in
the buffer and the full IV/nonce in crp_iv and set
CRYPTO_F_IV_SEPARATE.
The layout of the buffer is now described via fields in cryptop.
crp_aad_start and crp_aad_length define the boundaries of any AAD.
Previously with GCM and CCM you defined an auth crd with this range,
but for ETA your auth crd had to span both the AAD and plaintext
(and they had to be adjacent).
crp_payload_start and crp_payload_length define the boundaries of
the plaintext/ciphertext. Modes that only do a single operation
(COMPRESS, CIPHER, DIGEST) should only use this region and leave the
AAD region empty.
If a digest is present (or should be generated), it's starting
location is marked by crp_digest_start.
Instead of using the CRD_F_ENCRYPT flag to determine the direction
of the operation, cryptop now includes an 'op' field defining the
operation to perform. For digests I've added a new VERIFY digest
mode which assumes a digest is present in the input and fails the
request with EBADMSG if it doesn't match the internally-computed
digest. GCM and CCM already assumed this, and the new AEAD mode
requires this for decryption. The new ETA mode now also requires
this for decryption, so IPsec and GELI no longer do their own
authentication verification. Simple DIGEST operations can also do
this, though there are no in-tree consumers.
To eventually support some refcounting to close races, the session
cookie is now passed to crypto_getop() and clients should no longer
set crp_sesssion directly.
- Assymteric crypto operation structures should be allocated via
crypto_getkreq() and freed via crypto_freekreq(). This permits the
crypto layer to track open asym requests and close races with a
driver trying to unregister while asym requests are in flight.
- crypto_copyback, crypto_copydata, crypto_apply, and
crypto_contiguous_subsegment now accept the 'crp' object as the
first parameter instead of individual members. This makes it easier
to deal with different buffer types in the future as well as
separate input and output buffers. It's also simpler for driver
writers to use.
- bus_dmamap_load_crp() loads a DMA mapping for a crypto buffer.
This understands the various types of buffers so that drivers that
use DMA do not have to be aware of different buffer types.
- Helper routines now exist to build an auth context for HMAC IPAD
and OPAD. This reduces some duplicated work among drivers.
- Key buffers are now treated as const throughout the framework and in
device drivers. However, session key buffers provided when a session
is created are expected to remain alive for the duration of the
session.
- GCM and CCM sessions now only specify a cipher algorithm and a cipher
key. The redundant auth information is not needed or used.
- For cryptosoft, split up the code a bit such that the 'process'
callback now invokes a function pointer in the session. This
function pointer is set based on the mode (in effect) though it
simplifies a few edge cases that would otherwise be in the switch in
'process'.
It does split up GCM vs CCM which I think is more readable even if there
is some duplication.
- I changed /dev/crypto to support GMAC requests using CRYPTO_AES_NIST_GMAC
as an auth algorithm and updated cryptocheck to work with it.
- Combined cipher and auth sessions via /dev/crypto now always use ETA
mode. The COP_F_CIPHER_FIRST flag is now a no-op that is ignored.
This was actually documented as being true in crypto(4) before, but
the code had not implemented this before I added the CIPHER_FIRST
flag.
- I have not yet updated /dev/crypto to be aware of explicit modes for
sessions. I will probably do that at some point in the future as well
as teach it about IV/nonce and tag lengths for AEAD so we can support
all of the NIST KAT tests for GCM and CCM.
- I've split up the exising crypto.9 manpage into several pages
of which many are written from scratch.
- I have converted all drivers and consumers in the tree and verified
that they compile, but I have not tested all of them. I have tested
the following drivers:
- cryptosoft
- aesni (AES only)
- blake2
- ccr
and the following consumers:
- cryptodev
- IPsec
- ktls_ocf
- GELI (lightly)
I have not tested the following:
- ccp
- aesni with sha
- hifn
- kgssapi_krb5
- ubsec
- padlock
- safe
- armv8_crypto (aarch64)
- glxsb (i386)
- sec (ppc)
- cesa (armv7)
- cryptocteon (mips64)
- nlmsec (mips64)
Discussed with: cem
Relnotes: yes
Sponsored by: Chelsio Communications
Differential Revision: https://reviews.freebsd.org/D23677
When I did the use_numa support, I missed the fact that there is
a separate hash function for send tag nic selection. So when
use_numa is enabled, ktls offload does not work properly, as it
does not reliably allocate a send tag on the proper egress nic
since different egress nics are selected for send-tag allocation
and packet transmit. To fix this, this change:
- refectors lacp_select_tx_port_by_hash() and
lacp_select_tx_port() to make lacp_select_tx_port_by_hash()
always called by lacp_select_tx_port()
- pre-shifts flowids to convert them to hashes when calling lacp_select_tx_port_by_hash()
- adds a numa_domain field to if_snd_tag_alloc_params
- plumbs the numa domain into places where we allocate send tags
In testing with NIC TLS setup on a NUMA machine, I see thousands
of output errors before the change when enabling
kern.ipc.tls.ifnet.permitted=1. After the change, I see no
errors, and I see the NIC sysctl counters showing active TLS
offload sessions.
Reviewed by: rrs, hselasky, jhb
Sponsored by: Netflix
The results of ktls_get_cpu() are stored in u_int and NETISR_CPUID_NONE
requires u_int. Adjust uint16_t to uint_t in order to make RSS kernels
compile some more again.
HPTS still has to be fixed, which is a bit more complicated.
Reviewed by: jhb, gallatin, rrs
Differential Revision: https://reviews.freebsd.org/D23726
r357614 added CTLFLAG_NEEDGIANT to make it easier to find nodes that are
still not MPSAFE (or already are but aren’t properly marked).
Use it in preparation for a general review of all nodes.
This is non-functional change that adds annotations to SYSCTL_NODE and
SYSCTL_PROC nodes using one of the soon-to-be-required flags.
Mark all obvious cases as MPSAFE. All entries that haven't been marked
as MPSAFE before are by default marked as NEEDGIANT
Approved by: kib (mentor, blanket)
Commented by: kib, gallatin, melifaro
Differential Revision: https://reviews.freebsd.org/D23718
The use of the uma trash procedures is automatic, there's no need to
pass them explicitly here.
Reviewed by: markj
Sponsored by: Dell EMC Isilon
Differential Revision: https://reviews.freebsd.org/D22582
This permits constructing the entire TLS header in ktls_frame() rather
than ktls_seq(). This also matches the approach used by OpenSSL which
uses an incrementing nonce as the explicit IV rather than the sequence
number.
Reviewed by: gallatin
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D22117
This adds the glue to allocate TLS sessions and invokes it from
the TLS enable socket option handler. This also adds some counters
for active TOE sessions.
The TOE KTLS mode is returned by getsockopt(TLSTX_TLS_MODE) when
TOE KTLS is in use on a socket, but cannot be set via setsockopt().
To simplify various checks, a TLS session now includes an explicit
'mode' member set to the value returned by TLSTX_TLS_MODE. Various
places that used to check 'sw_encrypt' against NULL to determine
software vs ifnet (NIC) TLS now check 'mode' instead.
Reviewed by: np, gallatin
Sponsored by: Chelsio Communications
Differential Revision: https://reviews.freebsd.org/D21891
Software Kernel TLS needs to allocate a new destination crypto
buffer when encrypting data from the page cache, so as to avoid
overwriting shared clear-text file data with encrypted data
specific to a single socket. When the data is anonymous, eg, not
tied to a file, then we can encrypt in place and avoid allocating
a new page. This fixes a bug where the existing code always
assumes the data is private, and never encrypts in place. This
results in unneeded page allocations and potentially more memory
bandwidth consumption when doing socket writes.
When the code was written at Netflix, ktls_encrypt() looked at
private sendfile flags to determine if the pages being encrypted
where part of the page cache (coming from sendfile) or
anonymous (coming from sosend). This was broken internally at
Netflix when the sendfile flags were made private, and the
M_WRITABLE() check was added. Unfortunately, M_WRITABLE() will
always be false for M_NOMAP mbufs, since one cannot just mtod()
them.
This change introduces a new flags field to the mbuf_ext_pgs
struct by stealing a byte from the tls hdr. Note that the current
header is still 2 bytes larger than the largest header we
support: AES-CBC with explicit IV. We set MBUF_PEXT_FLAG_ANON
when creating an unmapped mbuf in m_uiotombuf_nomap() (which is
the path that socket writes take), and we check for that flag in
ktls_encrypt() when looking for anon pages.
Reviewed by: jhb
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D21796
TLS 1.3 requires a few changes because 1.3 pretends to be 1.2
with a record type of application data. The "real" record type is
then included at the end of the user-supplied plaintext
data. This required adding a field to the mbuf_ext_pgs struct to
save the record type, and passing the real record type to the
sw_encrypt() ktls backend functions.
Reviewed by: jhb, hselasky
Sponsored by: Netflix
Differential Revision: D21801
KTLS adds support for in-kernel framing and encryption of Transport
Layer Security (1.0-1.2) data on TCP sockets. KTLS only supports
offload of TLS for transmitted data. Key negotation must still be
performed in userland. Once completed, transmit session keys for a
connection are provided to the kernel via a new TCP_TXTLS_ENABLE
socket option. All subsequent data transmitted on the socket is
placed into TLS frames and encrypted using the supplied keys.
Any data written to a KTLS-enabled socket via write(2), aio_write(2),
or sendfile(2) is assumed to be application data and is encoded in TLS
frames with an application data type. Individual records can be sent
with a custom type (e.g. handshake messages) via sendmsg(2) with a new
control message (TLS_SET_RECORD_TYPE) specifying the record type.
At present, rekeying is not supported though the in-kernel framework
should support rekeying.
KTLS makes use of the recently added unmapped mbufs to store TLS
frames in the socket buffer. Each TLS frame is described by a single
ext_pgs mbuf. The ext_pgs structure contains the header of the TLS
record (and trailer for encrypted records) as well as references to
the associated TLS session.
KTLS supports two primary methods of encrypting TLS frames: software
TLS and ifnet TLS.
Software TLS marks mbufs holding socket data as not ready via
M_NOTREADY similar to sendfile(2) when TLS framing information is
added to an unmapped mbuf in ktls_frame(). ktls_enqueue() is then
called to schedule TLS frames for encryption. In the case of
sendfile_iodone() calls ktls_enqueue() instead of pru_ready() leaving
the mbufs marked M_NOTREADY until encryption is completed. For other
writes (vn_sendfile when pages are available, write(2), etc.), the
PRUS_NOTREADY is set when invoking pru_send() along with invoking
ktls_enqueue().
A pool of worker threads (the "KTLS" kernel process) encrypts TLS
frames queued via ktls_enqueue(). Each TLS frame is temporarily
mapped using the direct map and passed to a software encryption
backend to perform the actual encryption.
(Note: The use of PHYS_TO_DMAP could be replaced with sf_bufs if
someone wished to make this work on architectures without a direct
map.)
KTLS supports pluggable software encryption backends. Internally,
Netflix uses proprietary pure-software backends. This commit includes
a simple backend in a new ktls_ocf.ko module that uses the kernel's
OpenCrypto framework to provide AES-GCM encryption of TLS frames. As
a result, software TLS is now a bit of a misnomer as it can make use
of hardware crypto accelerators.
Once software encryption has finished, the TLS frame mbufs are marked
ready via pru_ready(). At this point, the encrypted data appears as
regular payload to the TCP stack stored in unmapped mbufs.
ifnet TLS permits a NIC to offload the TLS encryption and TCP
segmentation. In this mode, a new send tag type (IF_SND_TAG_TYPE_TLS)
is allocated on the interface a socket is routed over and associated
with a TLS session. TLS records for a TLS session using ifnet TLS are
not marked M_NOTREADY but are passed down the stack unencrypted. The
ip_output_send() and ip6_output_send() helper functions that apply
send tags to outbound IP packets verify that the send tag of the TLS
record matches the outbound interface. If so, the packet is tagged
with the TLS send tag and sent to the interface. The NIC device
driver must recognize packets with the TLS send tag and schedule them
for TLS encryption and TCP segmentation. If the the outbound
interface does not match the interface in the TLS send tag, the packet
is dropped. In addition, a task is scheduled to refresh the TLS send
tag for the TLS session. If a new TLS send tag cannot be allocated,
the connection is dropped. If a new TLS send tag is allocated,
however, subsequent packets will be tagged with the correct TLS send
tag. (This latter case has been tested by configuring both ports of a
Chelsio T6 in a lagg and failing over from one port to another. As
the connections migrated to the new port, new TLS send tags were
allocated for the new port and connections resumed without being
dropped.)
ifnet TLS can be enabled and disabled on supported network interfaces
via new '[-]txtls[46]' options to ifconfig(8). ifnet TLS is supported
across both vlan devices and lagg interfaces using failover, lacp with
flowid enabled, or lacp with flowid enabled.
Applications may request the current KTLS mode of a connection via a
new TCP_TXTLS_MODE socket option. They can also use this socket
option to toggle between software and ifnet TLS modes.
In addition, a testing tool is available in tools/tools/switch_tls.
This is modeled on tcpdrop and uses similar syntax. However, instead
of dropping connections, -s is used to force KTLS connections to
switch to software TLS and -i is used to switch to ifnet TLS.
Various sysctls and counters are available under the kern.ipc.tls
sysctl node. The kern.ipc.tls.enable node must be set to true to
enable KTLS (it is off by default). The use of unmapped mbufs must
also be enabled via kern.ipc.mb_use_ext_pgs to enable KTLS.
KTLS is enabled via the KERN_TLS kernel option.
This patch is the culmination of years of work by several folks
including Scott Long and Randall Stewart for the original design and
implementation; Drew Gallatin for several optimizations including the
use of ext_pgs mbufs, the M_NOTREADY mechanism for TLS records
awaiting software encryption, and pluggable software crypto backends;
and John Baldwin for modifications to support hardware TLS offload.
Reviewed by: gallatin, hselasky, rrs
Obtained from: Netflix
Sponsored by: Netflix, Chelsio Communications
Differential Revision: https://reviews.freebsd.org/D21277
