This function combines crypto_cursor_segbase() and
crypto_cursor_seglen() into a single function. This is mostly
beneficial in the unmapped mbuf case where back to back calls of these
two functions have to iterate over the sub-components of unmapped
mbufs twice.
Bump __FreeBSD_version for crypto drivers in ports.
Suggested by: markj
Reviewed by: markj
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D30445
This function appends the contents of a single mbuf to an sglist
rather than an entire mbuf chain.
Reviewed by: gallatin, markj
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D30135
KASAN enables the use of LLVM's AddressSanitizer in the kernel. This
feature makes use of compiler instrumentation to validate memory
accesses in the kernel and detect several types of bugs, including
use-after-frees and out-of-bounds accesses. It is particularly
effective when combined with test suites or syzkaller. KASAN has high
CPU and memory usage overhead and so is not suited for production
environments.
The runtime and pmap maintain a shadow of the kernel map to store
information about the validity of memory mapped at a given kernel
address.
The runtime implements a number of functions defined by the compiler
ABI. These are prefixed by __asan. The compiler emits calls to
__asan_load*() and __asan_store*() around memory accesses, and the
runtime consults the shadow map to determine whether a given access is
valid.
kasan_mark() is called by various kernel allocators to update state in
the shadow map. Updates to those allocators will come in subsequent
commits.
The runtime also defines various interceptors. Some low-level routines
are implemented in assembly and are thus not amenable to compiler
instrumentation. To handle this, the runtime implements these routines
on behalf of the rest of the kernel. The sanitizer implementation
validates memory accesses manually before handing off to the real
implementation.
The sanitizer in a KASAN-configured kernel can be disabled by setting
the loader tunable debug.kasan.disable=1.
Obtained from: NetBSD
MFC after: 2 weeks
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D29416
There haven't been any non-obscure drivers that supported this
functionality and it has been impossible to test to ensure that it
still works. The only known consumer of this interface was the engine
in OpenSSL < 1.1. Modern OpenSSL versions do not include support for
this interface as it was not well-documented.
Reviewed by: cem
Sponsored by: Chelsio Communications
Differential Revision: https://reviews.freebsd.org/D29736
MFC after: 3 days
Sponsored by: The FreeBSD Foundation
Reviewed by: bcr
Approved by: philip (mentor)
Differential Revision: https://reviews.freebsd.org/D29408
config_intrhook_drain will remove the hook from the list as
config_intrhook_disestablish does if the hook hasn't been called. If it has,
config_intrhook_drain will wait for the hook to be disestablished in the normal
course (or expedited, it's up to the driver to decide how and when
to call config_intrhook_disestablish).
This is intended for removable devices that use config_intrhook and might be
attached early in boot, but that may be removed before the kernel can call the
config_intrhook or before it ends. To prevent all races, the detach routine will
need to call config_intrhook_train.
Sponsored by: Netflix, Inc
Reviewed by: jhb, mav, gde (in D29006 for man page)
Differential Revision: https://reviews.freebsd.org/D29005
The zero_region() kernel interface was previously undocumented.
Add a new zero_region(9) manual page to document it.
Submitted by: Ka Ho Ng <khng@freebsdfoundation.org>
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D28914
This is a simple subsystem that allow drivers to register as a backlight.
Each backlight creates a device node under /dev/backlight/backlightX and
an alias based on the name provided.
Relnotes: yes
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D26250
Document the calls to send messages to userland via devctl.
devctl_notify will create a message for the specified system,
subsystem and type, optionally adding additional information.
Reviewed by: bcr
Differential Revision: https://reviews.freebsd.org/D26520
This routine centralizes the knowledge needed for properly quoting
'value' in all key="value" items that appear in devctl messages.
Reviewed by: bcr
Differential Revision: https://reviews.freebsd.org/D26520
This adds the getenv_bool() function, to parse a boolean value from a
kernel environment variable or tunable. This works for traditional
boolean values like "0" and "1", and also "true" and "false"
(case-insensitive). These semantics do not yet apply to sysctls declared
using SYSCTL_BOOL with CTLFLAG_TUN (they still only parse 1 and 0).
Also added are two wrapper functions, getenv_is_true() and
getenv_is_false(). These are slightly simpler for callers wishing to
perform a single check of a configuration variable.
Reviewed by: jhb (slightly earlier version)
Sponsored by: NetApp, Inc.
Sponsored by: Klara, Inc.
Differential Revision: https://reviews.freebsd.org/D26270
Submitted by: Ka Ho Ng <khng300@gmail.com>
Reviewed by: markj
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D26372
Add prng(9) as a replacement for random(9) in the kernel.
There are two major differences from random(9) and random(3):
- General prng(9) APIs (prng32(9), etc) do not guarantee an
implementation or particular sequence; they should not be used for
repeatable simulations.
- However, specific named API families are also exposed (for now: PCG),
and those are expected to be repeatable (when so-guaranteed by the named
algorithm).
Some minor differences from random(3) and earlier random(9):
- PRNG state for the general prng(9) APIs is per-CPU; this eliminates
contention on PRNG state in SMP workloads. Each PCPU generator in an
SMP system produces a unique sequence.
- Better statistical properties than the Park-Miller ("minstd") PRNG
(longer period, uniform distribution in all bits, passes
BigCrush/PractRand analysis).
- Faster than Park-Miller ("minstd") PRNG -- no division is required to
step PCG-family PRNGs.
For now, random(9) becomes a thin shim around prng32(). Eventually I
would like to mechanically switch consumers over to the explicit API.
Reviewed by: kib, markj (previous version both)
Discussed with: markm
Differential Revision: https://reviews.freebsd.org/D25916
The current scheme of calling VOP_GETATTR adds avoidable overhead.
An example with tmpfs doing fstat (ops/s):
before: 7488958
after: 7913833
Reviewed by: kib (previous version)
Differential Revision: https://reviews.freebsd.org/D25910
These functions were introduced before UMA started ensuring that freed
memory gets placed in domain-local caches. They no longer serve any
purpose since UMA now provides their functionality by default. Remove
them to simplyify the kernel memory allocator interfaces a bit.
Reviewed by: cem, kib
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D25937
These routines are similar to crypto_getreq() and crypto_freereq() but
operate on caller-supplied storage instead of allocating crypto
requests from a UMA zone.
Reviewed by: markj
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D25691
Some crypto consumers such as GELI and KTLS for file-backed sendfile
need to store their output in a separate buffer from the input.
Currently these consumers copy the contents of the input buffer into
the output buffer and queue an in-place crypto operation on the output
buffer. Using a separate output buffer avoids this copy.
- Create a new 'struct crypto_buffer' describing a crypto buffer
containing a type and type-specific fields. crp_ilen is gone,
instead buffers that use a flat kernel buffer have a cb_buf_len
field for their length. The length of other buffer types is
inferred from the backing store (e.g. uio_resid for a uio).
Requests now have two such structures: crp_buf for the input buffer,
and crp_obuf for the output buffer.
- Consumers now use helper functions (crypto_use_*,
e.g. crypto_use_mbuf()) to configure the input buffer. If an output
buffer is not configured, the request still modifies the input
buffer in-place. A consumer uses a second set of helper functions
(crypto_use_output_*) to configure an output buffer.
- Consumers must request support for separate output buffers when
creating a crypto session via the CSP_F_SEPARATE_OUTPUT flag and are
only permitted to queue a request with a separate output buffer on
sessions with this flag set. Existing drivers already reject
sessions with unknown flags, so this permits drivers to be modified
to support this extension without requiring all drivers to change.
- Several data-related functions now have matching versions that
operate on an explicit buffer (e.g. crypto_apply_buf,
crypto_contiguous_subsegment_buf, bus_dma_load_crp_buf).
- Most of the existing data-related functions operate on the input
buffer. However crypto_copyback always writes to the output buffer
if a request uses a separate output buffer.
- For the regions in input/output buffers, the following conventions
are followed:
- AAD and IV are always present in input only and their
fields are offsets into the input buffer.
- payload is always present in both buffers. If a request uses a
separate output buffer, it must set a new crp_payload_start_output
field to the offset of the payload in the output buffer.
- digest is in the input buffer for verify operations, and in the
output buffer for compute operations. crp_digest_start is relative
to the appropriate buffer.
- Add a crypto buffer cursor abstraction. This is a more general form
of some bits in the cryptosoft driver that tried to always use uio's.
However, compared to the original code, this avoids rewalking the uio
iovec array for requests with multiple vectors. It also avoids
allocate an iovec array for mbufs and populating it by instead walking
the mbuf chain directly.
- Update the cryptosoft(4) driver to support separate output buffers
making use of the cursor abstraction.
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D24545
The sole in-tree user of this flag has been retired, so remove this
complexity from all drivers. While here, add a helper routine drivers
can use to read the current request's IV into a local buffer. Use
this routine to replace duplicated code in nearly all drivers.
Reviewed by: cem
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D24450
- 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
Delay the attachment of children, when requested, until after interrutps are
running. This is often needed to allow children to run transactions on i2c or
spi busses. It's a common enough idiom that it will be useful to have its own
wrapper.
Reviewed by: ian
Differential Revision: https://reviews.freebsd.org/D21465
s/BIT_NAND/BIT_ANDNOT/, and for CPU and DOMAINSET too. The actual
implementation is "and not" (or "but not"), i.e. A but not B.
Fortunately this does appear to be what all existing callers want.
Don't supply a NAND (not (A and B)) operation at this time.
Discussed with: jeff
Reviewed by: cem
Sponsored by: Dell EMC Isilon
Differential Revision: https://reviews.freebsd.org/D22791
This adds basic documentation on what the superio driver is and how
other drivers can interact with it. I decided to also document
superio's ivar accessors.
Reviewed by: bcr, brueffer (both manual contents only)
MFC after: 2 weeks
Differential Revision: https://reviews.freebsd.org/D21958
There are several mechanisms by which a vm_page reference is held,
preventing the page from being freed back to the page allocator. In
particular, holding the page's object lock is sufficient to prevent the
page from being freed; holding the busy lock or a wiring is sufficent as
well. These references are protected by the page lock, which must
therefore be acquired for many per-page operations. This results in
false sharing since the page locks are external to the vm_page
structures themselves and each lock protects multiple structures.
Transition to using an atomically updated per-page reference counter.
The object's reference is counted using a flag bit in the counter. A
second flag bit is used to atomically block new references via
pmap_extract_and_hold() while removing managed mappings of a page.
Thus, the reference count of a page is guaranteed not to increase if the
page is unbusied, unmapped, and the object's write lock is held. As
a consequence of this, the page lock no longer protects a page's
identity; operations which move pages between objects are now
synchronized solely by the objects' locks.
The vm_page_wire() and vm_page_unwire() KPIs are changed. The former
requires that either the object lock or the busy lock is held. The
latter no longer has a return value and may free the page if it releases
the last reference to that page. vm_page_unwire_noq() behaves the same
as before; the caller is responsible for checking its return value and
freeing or enqueuing the page as appropriate. vm_page_wire_mapped() is
introduced for use in pmap_extract_and_hold(). It fails if the page is
concurrently being unmapped, typically triggering a fallback to the
fault handler. vm_page_wire() no longer requires the page lock and
vm_page_unwire() now internally acquires the page lock when releasing
the last wiring of a page (since the page lock still protects a page's
queue state). In particular, synchronization details are no longer
leaked into the caller.
The change excises the page lock from several frequently executed code
paths. In particular, vm_object_terminate() no longer bounces between
page locks as it releases an object's pages, and direct I/O and
sendfile(SF_NOCACHE) completions no longer require the page lock. In
these latter cases we now get linear scalability in the common scenario
where different threads are operating on different files.
__FreeBSD_version is bumped. The DRM ports have been updated to
accomodate the KPI changes.
Reviewed by: jeff (earlier version)
Tested by: gallatin (earlier version), pho
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D20486
The page daemon periodically invokes uma_reclaim() to reclaim cached
items from each zone when the system is under memory pressure. This
is important since the size of these caches is unbounded by default.
However it also results in bursts of high latency when allocating from
heavily used zones as threads miss in the per-CPU caches and must
access the keg in order to allocate new items.
With r340405 we maintain an estimate of each zone's usage of its
(per-NUMA domain) cache of full buckets. Start making use of this
estimate to avoid reclaiming the entire cache when under memory
pressure. In particular, introduce TRIM, DRAIN and DRAIN_CPU
verbs for uma_reclaim() and uma_zone_reclaim(). When trimming, only
items in excess of the estimate are reclaimed. Draining a zone
reclaims all of the cached full buckets (the previous behaviour of
uma_reclaim()), and may further drain the per-CPU caches in extreme
cases.
Now, when under memory pressure, the page daemon will trim zones
rather than draining them. As a result, heavily used zones do not incur
bursts of bucket cache misses following reclamation, but large, unused
caches will be reclaimed as before.
Reviewed by: jeff
Tested by: pho (an earlier version)
MFC after: 2 months
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D16667
- Fix warnings from igor and mandoc.
- Provide a brief description of the separation between zones and their
backend slab allocators.
- Document cache zones and secondary zones.
- Document the kernel config options added in r350659.
- Document the uma_zalloc_pcpu() and uma_zfree_pcpu() wrappers.
- Document uma_zone_reserve(), uma_zone_reserve_kva() and
uma_zone_prealloc().
- Document uma_zone_alloc() and uma_zone_freef().
- Add some missing MLINKs and Xrefs.
MFC after: 2 weeks
Similar to what was done for device_printfs in r347229.
Convert g_print_bio() to a thin shim around g_format_bio(), which acts on an
sbuf; documented in g_bio.9.
Reviewed by: markj
Discussed with: rlibby
Sponsored by: Dell EMC Isilon
Differential Revision: https://reviews.freebsd.org/D21165
r350315 created a Linux compatible copy_file_range(2) syscall.
It uses a VOP method called VOP_COPY_FILE_RANGE so that file systems,
such as the NFSv4.2 client can do file system specific copying.
For NFSv4.2, this allows the copying to be done locally on the NFS server,
avoiding transferring the data across the wire twice.
This is a new man page (content changed).
Reviewed by: kib, asomers
Relnotes: yes
Differential Revision: https://reviews.freebsd.org/D20584
We don't split the other man pages in their own package so do the same for runtime.
Reviewed by: bapt, gjb
Differential Revision: https://reviews.freebsd.org/D20962
The hold_count and wire_count fields of struct vm_page are separate
reference counters with similar semantics. The remaining essential
differences are that holds are not counted as a reference with respect
to LRU, and holds have an implicit free-on-last unhold semantic whereas
vm_page_unwire() callers must explicitly determine whether to free the
page once the last reference to the page is released.
This change removes the KPIs which directly manipulate hold_count.
Functions such as vm_fault_quick_hold_pages() now return wired pages
instead. Since r328977 the overhead of maintaining LRU for wired pages
is lower, and in many cases vm_fault_quick_hold_pages() callers would
swap holds for wirings on the returned pages anyway, so with this change
we remove a number of page lock acquisitions.
No functional change is intended. __FreeBSD_version is bumped.
Reviewed by: alc, kib
Discussed with: jeff
Discussed with: jhb, np (cxgbe)
Tested by: pho (previous version)
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D19247
Unmapped mbufs allow sendfile to carry multiple pages of data in a
single mbuf, without mapping those pages. It is a requirement for
Netflix's in-kernel TLS, and provides a 5-10% CPU savings on heavy web
serving workloads when used by sendfile, due to effectively
compressing socket buffers by an order of magnitude, and hence
reducing cache misses.
For this new external mbuf buffer type (EXT_PGS), the ext_buf pointer
now points to a struct mbuf_ext_pgs structure instead of a data
buffer. This structure contains an array of physical addresses (this
reduces cache misses compared to an earlier version that stored an
array of vm_page_t pointers). It also stores additional fields needed
for in-kernel TLS such as the TLS header and trailer data that are
currently unused. To more easily detect these mbufs, the M_NOMAP flag
is set in m_flags in addition to M_EXT.
Various functions like m_copydata() have been updated to safely access
packet contents (using uiomove_fromphys()), to make things like BPF
safe.
NIC drivers advertise support for unmapped mbufs on transmit via a new
IFCAP_NOMAP capability. This capability can be toggled via the new
'nomap' and '-nomap' ifconfig(8) commands. For NIC drivers that only
transmit packet contents via DMA and use bus_dma, adding the
capability to if_capabilities and if_capenable should be all that is
required.
If a NIC does not support unmapped mbufs, they are converted to a
chain of mapped mbufs (using sf_bufs to provide the mapping) in
ip_output or ip6_output. If an unmapped mbuf requires software
checksums, it is also converted to a chain of mapped mbufs before
computing the checksum.
Submitted by: gallatin (earlier version)
Reviewed by: gallatin, hselasky, rrs
Discussed with: ae, kp (firewalls)
Relnotes: yes
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D20616
The epoch_drain_callbacks() function is used to drain all pending
callbacks which have been invoked by prior epoch_call() function calls
on the same epoch. This function is useful when there are shared
memory structure(s) referred to by the epoch callback(s) which are not
refcounted and are rarely freed. The typical place for calling this
function is right before freeing or invalidating the shared
resource(s) used by the epoch callback(s). This function can sleep and
is not optimized for performance.
Differential Revision: https://reviews.freebsd.org/D20109
MFC after: 1 week
Sponsored by: Mellanox Technologies
