With the mac_priority(4) realtime policy active, users and processes in
the realtime group may promote existing threads and processes to
realtime scheduling priority. Extend the privileges granted to
PRIV_SCHED_SETPOLICY which allows explicit creation of new realtime
threads.
One use case of this is when the pthread scheduling policy is set to
SCHED_RR or SCHED_FIFO via pthread_attr_setschedpolicy(...) before
calling pthread_create(...). I ran into this when testing audio software
with realtime threads, particularly audio/ardour6.
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D33393
Add an idletime user group that allows non-root users to run processes
with idle scheduling priority. Privileges are granted by a MAC policy in
the mac_priority module. For this purpose, the kernel privilege
PRIV_SCHED_IDPRIO was added to sys/priv.h (kernel module ABI change).
Deprecate the system wide sysctl(8) knob
security.bsd.unprivileged_idprio which lets any user run idle priority
processes, regardless of context. While the knob is still working, it is
marked as deprecated in the description and in the man pages.
MFC after: 2 weeks
Differential revision: https://reviews.freebsd.org/D33338
The privilege allows the holder to assign idle priority type to thread
or process.
MFC after: 2 weeks
Differential revision: https://reviews.freebsd.org/D33338
This is a MAC policy module that grants scheduling privileges based on
group membership. Users or processes in the group realtime (gid 47) are
allowed to run threads and processes with realtime scheduling priority.
For timing-sensitive, low-latency software like audio/jack, running with
realtime priority helps to avoid stutter and gaps.
PR: 239125
MFC after: 2 weeks
Differential revision: https://reviews.freebsd.org/D33191
fspacectl(2) is a system call to provide space management support to
userspace applications. VOP_DEALLOCATE(9) is a VOP call to perform the
deallocation. vn_deallocate(9) is a public KPI for kmods' use.
The purpose of proposing a new system call, a KPI and a VOP call is to
allow bhyve or other hypervisor monitors to emulate the behavior of SCSI
UNMAP/NVMe DEALLOCATE on a plain file.
fspacectl(2) comprises of cmd and flags parameters to specify the
space management operation to be performed. Currently cmd has to be
SPACECTL_DEALLOC, and flags has to be 0.
fo_fspacectl is added to fileops.
VOP_DEALLOCATE(9) is added as a new VOP call. A trivial implementation
of VOP_DEALLOCATE(9) is provided.
Sponsored by: The FreeBSD Foundation
Reviewed by: kib
Differential Revision: https://reviews.freebsd.org/D28347
mac_veriexec sets its version to 1, but the mac_veriexec_shaX modules which depend on it expect MAC_VERIEXEC_VERSION = 2.
Be consistent and use MAC_VERIEXEC_VERSION everywhere.
This unbreaks loading of mac_veriexec modules at boot time.
Authored by: Kornel Duleba <mindal@semihalf.com>
Obtained from: Semihalf
Sponsored by: Stormshield
Differential Revision: https://reviews.freebsd.org/D31268
When packet is a SYN packet, we don't need to modify any existing PCB.
Normally SYN arrives on a listening socket, we either create a syncache
entry or generate syncookie, but we don't modify anything with the
listening socket or associated PCB. Thus create a new PCB lookup
mode - rlock if listening. This removes the primary contention point
under SYN flood - the listening socket PCB.
Sidenote: when SYN arrives on a synchronized connection, we still
don't need write access to PCB to send a challenge ACK or just to
drop. There is only one exclusion - tcptw recycling. However,
existing entanglement of tcp_input + stacks doesn't allow to make
this change small. Consider this patch as first approach to the problem.
Reviewed by: rrs
Differential revision: https://reviews.freebsd.org/D29576
place -- in the VOP rather than vn_setexttr() -- and that it is for historic
reasons. We might wish to relocate it in due course, but this way at least
we document the asymmetry.
pipes get stated all thet time and this avoidably contributed to contention.
The pipe lock is only held to accomodate MAC and to check the type.
Since normally there is no probe for pipe stat depessimize this by having the
flag.
The pipe_state field gets modified with locks held all the time and it's not
feasible to convert them to use atomic store. Move the type flag away to a
separate variable as a simple cleanup and to provide stable field to read.
Use short for both fields to avoid growing the struct.
While here short-circuit MAC for pipe_poll as well.
I have such a custom kernel configuration and its build failed with:
linking kernel.full
ld: error: undefined symbol: mac_vnode_assert_locked
>>> referenced by mac_framework.h:556 (/usr/devel/git/apu2c4/sys/security/mac/mac_framework.h:556)
>>> tmpfs_vnops.o:(mac_vnode_check_stat)
>>> referenced by mac_framework.h:556 (/usr/devel/git/apu2c4/sys/security/mac/mac_framework.h:556)
>>> vfs_default.o:(mac_vnode_check_stat)
>>> referenced by mac_framework.h:556 (/usr/devel/git/apu2c4/sys/security/mac/mac_framework.h:556)
>>> ufs_vnops.o:(mac_vnode_check_stat)
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
The code would unconditionally lock the vnode to audit or call the
mac hoook, even if neither want to do anything. Pre-check the state
to avoid locking in the common case of nothing to do.
Note this code should not be normally executed anyway as vnodes are
always return ready. However, poll1/2 from will-it-scale use regular
files for benchmarking, presumably to focus on the interface itself
as the vnode handler is not supposed to do almost anything.
This in particular fixes poll2 which passes 128 fds.
$ ./poll2_processes -s 10
before: 134411
after: 271572
Comparing fsid_t objects requires internal knowledge of the fsid structure
and yet this is duplicated across a number of places in the code.
Simplify by creating a fsidcmp function (macro).
Reviewed by: mjg, rmacklem
Approved by: mav (mentor)
MFC after: 1 week
Sponsored by: iXsystems, Inc.
Differential Revision: https://reviews.freebsd.org/D24749
- thr_kill(2) and thr_exit(2) generally (no argument auditing here.
- A set of syscalls for the process descriptor family, specifically:
pdfork(2), pdgetpid(2) and pdkill(2)
For these syscalls, audit the file descriptor. In the case of pdfork(2)
a pointer to an integer (file descriptor) is passed in as an argument.
We audit the post initialized file descriptor (not the random garbage
that would have been passed in). We will also audit the child process
which was created from the fork operation (similar to what is done for
the fork(2) syscall).
pdkill(2) we audit the signal value and fd, and finally pdgetpid(2)
just the file descriptor:
- Following is a sample of the produced audit trails:
header,111,11,pdfork(2),0,Sat May 16 03:07:50 2020, + 394 msec
argument,0,0x39d,child PID
argument,2,0x2,flags
argument,1,0x8,fd
subject,root,root,0,root,0,924,0,0,0.0.0.0
return,success,925
header,79,11,pdgetpid(2),0,Sat May 16 03:07:50 2020, + 394 msec
argument,1,0x8,fd
subject,root,root,0,root,0,924,0,0,0.0.0.0
return,success,0
trailer,79
header,135,11,pdkill(2),0,Sat May 16 03:07:50 2020, + 395 msec
argument,1,0x8,fd
argument,2,0xf,signal
process_ex,root,root,0,root,0,925,0,0,0.0.0.0
subject,root,root,0,root,0,924,0,0,0.0.0.0
return,success,0
trailer,135
MFC after: 1 week
cdir may have simply failed to resolve (e.g. fget_cap failure in namei
leading to NULL dp passed to AUDIT_ARG_UPATH*_VP); restore the pre-rS358191
behavior of setting cpath[0] = '\0' and bailing out instead of panicking.
This was found by inadvertently running the libc/c063 tests with auditing
enabled, resulting in a panic.
Reviewed by: mjg (committed version actually his)
Differential Revision: https://reviews.freebsd.org/D24445
This lock was made unnecessary by the addition of mac_policy_rms in r356120.
Reviewed by: mjg, kib
Differential Revision: https://reviews.freebsd.org/D24283
to the path related tokens not being processed. Restore this behavior and
and move AUE_JAIL_SET in this block, as it may conditionally contain a
path token.
Discovered by: kevans
PR: 244537
Reviewed by: kevans
Differential Revision: https://reviews.freebsd.org/D23929
The new structure is copy-on-write. With the assumption that path lookups are
significantly more frequent than chdirs and chrooting this is a win.
This provides stable root and jail root vnodes without the need to reference
them on lookup, which in turn means less work on globally shared structures.
Note this also happens to fix a bug where jail vnode was never referenced,
meaning subsequent access on lookup could run into use-after-free.
Reviewed by: kib
Differential Revision: https://reviews.freebsd.org/D23884
jail_remove(2) and finally setloginclass(2) are not being converted and
committed into userspace. Add the cases for these syscalls and make sure
they are being converted properly.
Reviewed by: bz, kevans
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D23882
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
realpath(3) is used a lot e.g., by clang and is a major source of getcwd
and fstatat calls. This can be done more efficiently in the kernel.
This works by performing a regular lookup while saving the name and found
parent directory. If the terminal vnode is a directory we can resolve it using
usual means. Otherwise we can use the name saved by lookup and resolve the
parent.
See the review for sample syscall counts.
Reviewed by: kib
Differential Revision: https://reviews.freebsd.org/D23574
This in particular significantly shortens amd64_syscall, which otherwise
keeps jumping forward over 2KB of code in total.
Note some of these branches should be either eliminated altogether or
coalesced.
All checking routines walk a linked list of all modules in order to determine
if given hook is installed. This became a significant problem after mac_ntpd
started being loaded by default.
Implement a way perform checks for select hooks by testing a boolean.
Use it for priv_check and priv_grant, which are constantly called from priv_check.
The real fix would use hotpatching, but the above provides a way to know when
to do it.
There was only one consumer and it was using it incorrectly.
It is given an equivalent hack.
Reviewed by: jeff
Differential Revision: https://reviews.freebsd.org/D23037
Filesystems which want to use it in limited capacity can employ the
VOP_UNLOCK_FLAGS macro.
Reviewed by: kib (previous version)
Differential Revision: https://reviews.freebsd.org/D21427
If any non-static modules are loaded (and mac_ntpd tends to be), the lock is
taken all the time al over the kernel. On platforms like arm64 this results in
an avoidable significant performance degradation. Since write-locking is almost
never needed, use a primitive optimized towards read-locking.
Sample result of building the kernel on tmpfs 11 times:
stock 11142.80s user 6704.44s system 4924% cpu 6:02.42 total
patched 11118.95s user 2374.94s system 4547% cpu 4:56.71 total
entry, when that entry has been seen already, keep the
already-looked-up value in a variable and use that instead of looking
it up again.
Approved by: alc, markj (earlier version), kib (earlier version)
Differential Revision: https://reviews.freebsd.org/D22348
