This ioctl exposes VOP_BMAP information to userland. It can be used by
programs like fragmentation analyzers and optimized cp implementations. But
I'm using it to test fusefs's VOP_BMAP implementation. The "2" in the name
distinguishes it from the similar but incompatible FIBMAP ioctls in NetBSD
and Linux. FIOBMAP2 differs from FIBMAP in that it uses a 64-bit block
number instead of 32-bit, and it also returns runp and runb.
Reviewed by: mckusick
MFC after: 2 weeks
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D20705
wakeup_one() and underlying sleepq_signal() spend additional time trying
to be fair, waking thread with highest priority, sleeping longest time.
But in case of taskqueue there are many absolutely identical threads, and
any fairness between them is quite pointless. It makes even worse, since
round-robin wakeups not only make previous CPU affinity in scheduler quite
useless, but also hide from user chance to see CPU bottlenecks, when
sequential workload with one request at a time looks evenly distributed
between multiple threads.
This change adds new SLEEPQ_UNFAIR flag to sleepq_signal(), making it wakeup
thread that went to sleep last, but no longer in context switch (to avoid
immediate spinning on the thread lock). On top of that new wakeup_any()
function is added, equivalent to wakeup_one(), but setting the flag.
On top of that taskqueue(9) is switchied to wakeup_any() to wakeup its
threads.
As result, on 72-core Xeon v4 machine sequential ZFS write to 12 ZVOLs
with 16KB block size spend 34% less time in wakeup_any() and descendants
then it was spending in wakeup_one(), and total write throughput increased
by ~10% with the same as before CPU usage.
Reviewed by: markj, mmacy
MFC after: 2 weeks
Sponsored by: iXsystems, Inc.
Differential Revision: https://reviews.freebsd.org/D20669
When it comes to megabytes of text, difference between sbuf_printf() and
sbuf_cat() becomes substantial.
MFC after: 2 weeks
Sponsored by: iXsystems, Inc.
in response to SACKs. The default behavior is unchanged; however, the limit
can be activated by changing the new net.inet.tcp.rack.split_limit sysctl.
Submitted by: Peter Lei <peterlei@netflix.com>
Reported by: jtl
Reviewed by: lstewart (earlier version)
Security: CVE-2019-5599
at runtime. This change removes a dependency on a barrel shifter pass
before branch resolution, while reducing the instruction stream size
by 9 bytes on amd64.
MFC after: 3 days
translation units with differing capabilities
From the author via Bugzilla:
---
When an attempt is made to passthrough a PCI device to a bhyve VM
(causing initialisation of IOMMU) on certain Intel chipsets using
VT-d the PCI bus stops working entirely. This issue occurs on the
E3-1275 v5 processor on C236 chipset and has also been encountered
by others on the forums with different hardware in the Skylake
series.
The chipset has two VT-d translation units. The issue is caused by
an attempt to use the VT-d device-IOTLB capability that is
supported by only the first unit for devices attached to the
second unit which lacks that capability. Only the capabilities of
the first unit are checked and are assumed to be the same for all
units.
Attached is a patch to rectify this issue by determining which
unit is responsible for the device being added to a domain and
then checking that unit's device-IOTLB capability. In addition to
this a few fixes have been made to other instances where the first
unit's capabilities are assumed for all units for domains they
share. In these cases a mutual set of capabilities is determined.
The patch should hopefully fix any bugs for current/future
hardware with multiple translation units supporting different
capabilities.
A description is on the forums at
https://forums.freebsd.org/threads/pci-passthrough-bhyve-usb-xhci.65235
The thread includes observations by other users of the bug
occurring, and description as well as confirmation of the fix.
I'd also like to thank Ordoban for their help.
---
Personally tested on a Skylake laptop, Skylake Xeon server, and
a Xeon-D-1541, passing through XHCI and NVMe functions. Passthru
is hit-or-miss to the point of being unusable without this
patch.
PR: 229852
Submitted by: callum@aitchison.org
MFC after: 1 week
On large systems those sysctls may generate megabytes of output. Before
this change sbuf(9) code was resizing buffer by 4KB each time many times,
generating tons of TLB shootdowns. Unfortunately in this case existing
sbuf_new_for_sysctl() mechanism, supposed to help with this issue, is not
applicable, since all the sbuf writes are done in different kernel thread.
This change improves situation in two ways:
- on first sysctl call, not providing any output buffer, it sets special
sbuf drain function, just counting the data and so not needing big buffer;
- on second sysctl call it uses as initial buffer size value saved on
previous call, so that in most cases there will be no reallocation, unless
GEOM topology changed significantly.
MFC after: 1 week
Sponsored by: iXsystems, Inc.
In r349154, random device reads of size < 16 bytes (AES block size) were
accidentally broken to loop forever. Correct the loop condition for small
reads.
Reported by: pho
Reviewed by: delphij
Approved by: secteam(delphij)
Differential Revision: https://reviews.freebsd.org/D20686
In all supported (and most unsupported) FreeBSD versions the appropriate
header for Capsicum is sys/capsicum.h. Software including sys/capability.h
is most likely looking for Linux capabilities based on the withdrawn
POSIX.1e draft.
This header was previously removed in r334929 and r340156, but reverted
each time due to ports failures. These issues have now (broadly) been
addressed.
PR: 228878 [exp-run]
Submitted by: eadler (r334929)
Relnotes: Yes
Sponsored by: The FreeBSD Foundation
This adds ACPI device path on devinfo(8) output and
show value of _UPC(usb port capabilities), _PLD (physical location of device)
when hw.usb.debug >= 1 .
Reviewed by: hselasky
Differential Revision: https://reviews.freebsd.org/D20630
Add experimental feature to increase concurrency in Fortuna. As this
diverges slightly from canonical Fortuna, and due to the security
sensitivity of random(4), it is off by default. To enable it, set the
tunable kern.random.fortuna.concurrent_read="1". The rest of this commit
message describes the behavior when enabled.
Readers continue to update shared Fortuna state under global mutex, as they
do in the status quo implementation of the algorithm, but shift the actual
PRF generation out from under the global lock. This massively reduces the
CPU time readers spend holding the global lock, allowing for increased
concurrency on SMP systems and less bullying of the harvestq kthread.
It is somewhat of a deviation from FS&K. I think the primary difference is
that the specific sequence of AES keys will differ if READ_RANDOM_UIO is
accessed concurrently (as the 2nd thread to take the mutex will no longer
receive a key derived from rekeying the first thread). However, I believe
the goals of rekeying AES are maintained: trivially, we continue to rekey
every 1MB for the statistical property; and each consumer gets a
forward-secret, independent AES key for their PRF.
Since Chacha doesn't need to rekey for sequences of any length, this change
makes no difference to the sequence of Chacha keys and PRF generated when
Chacha is used in place of AES.
On a GENERIC 4-thread VM (so, INVARIANTS/WITNESS, numbers not necessarily
representative), 3x concurrent AES performance jumped from ~55 MiB/s per
thread to ~197 MB/s per thread. Concurrent Chacha20 at 3 threads went from
roughly ~113 MB/s per thread to ~430 MB/s per thread.
Prior to this change, the system was extremely unresponsive with 3-4
concurrent random readers; each thread had high variance in latency and
throughput, depending on who got lucky and won the lock. "rand_harvestq"
thread CPU use was high (double digits), seemingly due to spinning on the
global lock.
After the change, concurrent random readers and the system in general are
much more responsive, and rand_harvestq CPU use dropped to basically zero.
Tests are added to the devrandom suite to ensure the uint128_add64 primitive
utilized by unlocked read functions to specification.
Reviewed by: markm
Approved by: secteam(delphij)
Relnotes: yes
Differential Revision: https://reviews.freebsd.org/D20313
rename the source to gsb_crc32.c.
This is a prerequisite of unifying kernel zlib instances.
PR: 229763
Submitted by: Yoshihiro Ota <ota at j.email.ne.jp>
Differential Revision: https://reviews.freebsd.org/D20193
names. I.e., everything related to pwm now goes in /dev/pwm. This will
make it easier for userland tools to turn an unqualified name into a fully
qualified pathname, whether it's the base pwmcX.Y name or a label name.
At a basic level, remove assumptions about the underlying algorithm (such as
output block size and reseeding requirements) from the algorithm-independent
logic in randomdev.c. Chacha20 does not have many of the restrictions that
AES-ICM does as a PRF (Pseudo-Random Function), because it has a cipher
block size of 512 bits. The motivation is that by generalizing the API,
Chacha is not penalized by the limitations of AES.
In READ_RANDOM_UIO, first attempt to NOWAIT allocate a large enough buffer
for the entire user request, or the maximal input we'll accept between
signal checking, whichever is smaller. The idea is that the implementation
of any randomdev algorithm is then free to divide up large requests in
whatever fashion it sees fit.
As part of this, two responsibilities from the "algorithm-generic" randomdev
code are pushed down into the Fortuna ra_read implementation (and any other
future or out-of-tree ra_read implementations):
1. If an algorithm needs to rekey every N bytes, it is responsible for
handling that in ra_read(). (I.e., Fortuna's 1MB rekey interval for AES
block generation.)
2. If an algorithm uses a block cipher that doesn't tolerate partial-block
requests (again, e.g., AES), it is also responsible for handling that in
ra_read().
Several APIs are changed from u_int buffer length to the more canonical
size_t. Several APIs are changed from taking a blockcount to a bytecount,
to permit PRFs like Chacha20 to directly generate quantities of output that
are not multiples of RANDOM_BLOCKSIZE (AES block size).
The Fortuna algorithm is changed to NOT rekey every 1MiB when in Chacha20
mode (kern.random.use_chacha20_cipher="1"). This is explicitly supported by
the math in FS&K §9.4 (Ferguson, Schneier, and Kohno; "Cryptography
Engineering"), as well as by their conclusion: "If we had a block cipher
with a 256-bit [or greater] block size, then the collisions would not
have been an issue at all."
For now, continue to break up reads into PAGE_SIZE chunks, as they were
before. So, no functional change, mostly.
Reviewed by: markm
Approved by: secteam(delphij)
Differential Revision: https://reviews.freebsd.org/D20312
Add some basic regression tests to verify behavior of both uint128
implementations at typical boundary conditions, to run on all architectures.
Test uint128 increment behavior of Chacha in keystream mode, as used by
'kern.random.use_chacha20_cipher=1' (r344913) to verify assumptions at edge
cases. These assumptions are critical to the safety of using Chacha as a
PRF in Fortuna (as implemented).
(Chacha's use in arc4random is safe regardless of these tests, as it is
limited to far less than 4 billion blocks of output in that API.)
Reviewed by: markm
Approved by: secteam(gordon)
Differential Revision: https://reviews.freebsd.org/D20392
Implement wiring changes on superpage mappings. Previously, a superpage
mapping was unconditionally demoted by pmap_unwire(), even if the wiring
change applied to the entire superpage mapping.
Rewrite a comment to use the arm64 names for bits in a page table entry.
Previously, the bits were referred to by their x86 names.
Use atomic_"op"_64() instead of atomic_"op"_long() to update a page table
entry in order to match the prevailing style in this file.
MFC after: 10 days
Previously, there was a pwmc instance for each instance of pwm hardware
regardless of how many pwm channels that hardware supported. Now there
will be a pwmc instance for each channel when the hardware supports
multiple channels. With a separate instance for each channel, we can have
"named channels" in userland by making devfs alias entries in /dev/pwm.
These changes add support for ivars to pwmbus, and use an ivar to track the
channel number for each child. It also adds support for hinted children.
In pwmc, the driver checks for a label hint, and if present, it's used to
create an alias for the cdev in /dev/pwm. It's not anticipated that hints
will be heavily used, but it's easy to do and allows quick ad-hoc creation
of named channels from userland by using kenv to create hint.pwmc.N.label=
hints. Upcoming changes will add FDT support, and most labels will
probably be specified that way.
Implement protection changes on superpage mappings. Previously, a superpage
mapping was unconditionally demoted by pmap_protect(), even if the
protection change applied to the entire superpage mapping.
Precompute the bit mask describing the protection changes rather than
recomputing it for every page table entry that is changed.
Skip page table entries that already have the requested protection changes
in place.
Reviewed by: andrew, kib
MFC after: 10 days
Differential Revision: https://reviews.freebsd.org/D20657
is nothing left in the file that related to pwmbus at all. It just contains
prototypes for the functions implemented in dev/pwm.ofw_pwm.c, so name it
accordingly and fix the include protect wrappers to match.
A new pwmbus.h will be coming along in a future commit.
The natural place to look for them based on how other SoCs are organized
would be sys/modules/ti, but that's already taken. Drop a clue into
modules/ti/Makefile directing people to modules/arm_ti if they're looking
for ARM modules.
The pwm and pwmbus interfaces were nearly identical, this merges them into a
single pwmbus interface. The pwmbus driver now implements the pwmbus
interface by simply passing all calls through to its parent (the hardware
driver). The channel_count method moves from pwm to pwmbus, and the
get_bus method is deleted (just no longer needed).
The net effect is that the interface for doing pwm stuff is now the same
regardless of whether you're a child of pwmbus, or some random driver
elsewhere in the hierarchy that is bypassing the pwmbus layer and is talking
directly to the hardware driver via cross-hierarchy connections established
using fdt data.
The pwmc driver is now a child of pwmbus, instead of being its sibling
(that's why the get_bus method is no longer needed; pwmc now gets the
device_t of the bus using device_get_parent()).
pollution that was cleaned up recently, and this file got missed in the
cleanup because it's not attached to the build unless you specifically
request this device in a custom kernel config.