Count input/output packets and bytes on the interface as well, not just
in openvpn-specific counters.
PR: 292464
MFC after: 2 weeks
Sponsored by: Rubicon Communications, LLC ("Netgate")
This has moved from a generated file in objdir to the source tree.
Remove the old file and any .depend files that reference it.
Reviewed by: bapt (via IRC)
Fixes: 41ccf82b29 ("libc/aarch64: Use MOPS implementations of memcpy/memmove/memset where availble")
Sponsored by: Arm Ltd
Differential Revision: https://reviews.freebsd.org/D54729
Trim white space padding that gets added libxo fields cc and stack when
using -C and -c.
The padding is only visible if you're using multiple stacks or
congestion algorithms.
PR: 292262
MFC after: 5 days
Reviewed by: asomers, tuexen
Approved by: asomers (mentor)
Differential Revision: https://reviews.freebsd.org/D54709
We should free the allocated ASID if smmu_init_cd() fails.
Move the allocation of "domain" to simplify the first error path.
Reported by: Kevin Day <kevin@your.org>
Reviewed by: br
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D54676
On read, these registers' fields return 1 if forwarding of the
corresponding interrupt is enabled, and 0 otherwise. The test in
read_enabler() was inverted.
Reported by: Kevin Day <kevin@your.org>
Reviewed by: andrew
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D54678
- Free the pointer that was returned by the allocator, instead of the
address of the first member. These will be equal in practice, but
it's sketchy and won't work on CHERI with subobject bounds checking.
- Use an anonymous struct, there's no need to name it.
Reviewed by: bz, brooks, emaste
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D54673
If the allocation fails we should free the task struct.
While here get rid of a couple of unnecessary assertions.
Reported by: Kevin Day <kevin@your.org>
Reviewed by: emaste
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D54671
After commit 3bd8fab241 ("vfs: Move DEBUG_VFS_LOCKS checks to
INVARIANTS"), this option has no effect. Let's finish the removal.
There are a couple of additional uses in zfs, I will submit a separate
patch upstream for them.
Reviewed by: mckusick, kib
Differential Revision: https://reviews.freebsd.org/D54662
Add strtonumx(), a companion to strtonum(3) that preserves its safety
and error-reporting semantics while allowing the caller to specify a
conversion base, similar to the strtol(3) family of functions.
Reviewed by: emaste, kib, ziaee
Obtained from: https://www.illumos.org/issues/15365
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D54270
Sometimes the client device needs to manage the IOQ the request goes
to. Expand the interface we have for the request to allow it to be set
for this special use case.
Sponsored by: Netflix
Reviewed by: jhb
Differential Revision: https://reviews.freebsd.org/D54714
Currently, to define a new environment variable or modify an existing
one, we need to use env(1), which may or may not be available inside
the jail, especially in OCI containers created with the scratch
layer (i.e., those containers that are only a single static binary,
plus configuration files and related stuff). With this option, we
can specify environment variables of arbitrary length for the
specified process running inside the jail.
Reviewed by: jamie@
Approved by: jamie@
Differential Revision: https://reviews.freebsd.org/D54660
This ioctl has been marked as "old" starting with the original KAME export
over 20 years ago and has been hidden under #ifdef _KERNEL since. There
is no software that uses it.
pcie_get_speed_cap() has a hard coded skip of 3 devices at the
beginning. It is either called on a pdev or on a result from
pci_upstream_bridge(). In the latter case skipping another three
devices might get us to acpi0 or nexus, neither of which is a
PCI device still and pci_get_vendor() will panic() on that.
Sponsored by: The FreeBSD Foundation (commit)
GHI: https://github.com/freebsd/drm-kmod/issues/393
MFC after: 2 weeks
Differential Revision: https://reviews.freebsd.org/D53862
The LinuxKPI based wireless drivers are currently limited to amd64 and
arm64 (and until cleaned up i386). Adding RISC-V now as we have a
report that iwlwifi(4) works on RISC-V [1]. Factor the LinuxKPI based
wireless drivers out into their own block.
Given RISC-V has no ACPI support yet (though we fixed #includes in
order to keep compiling the drivers without further modifications
where possible) we need to take care of rtw89 which fails to compile
without ACPI enabled. A quick check at the Linux build files indicates
that the depenency is not correctly recorded there either. Disable
compiling rtw89 without ACPI (on RISC-V) for the moment until this is
fixed.
[1] https://mail-archive.freebsd.org/cgi/getmsg.cgi?fetch=5947+0+archive/2026/freebsd-wireless/20260112.freebsd-wireless
Sponsored by: The FreeBSD Foundation
MFC after: 3 days
Reviewed by: emaste
Differential Revision: https://reviews.freebsd.org/D54693
In order to compile iwlwifi(4), which is reported to work on RISC-V,
include the ACPI headers to avoid adding further FreeBSD-specific #ifdefs
to the driver. With this iwlwifi(4) just compiles on RISC-V (at least
if ACPI support is turned off in the module Makefile).
Sponsored by: The FreeBSD Foundation
MFC after: 3 days
Reviewed by: emaste
Differential Revision: https://reviews.freebsd.org/D54692
We have reports that iwlwifi(4) works on RISC-V. While we can turn off
full ACPI-specific files easily, intermittent code still relies on the
header files to be present. In order to not need to completely #ifdef
everything out we want to include acpi.h from LinuxKPI and as a result
need this file. With this the iwlwifi(4) code compiles just fine and
will do the right thing (given the functional ACPI parts are disabled/
unavailable).
Sponsored by: The FreeBSD Foundation
MFC after: 3 days
Reviewed by: mhorne, emaste
Differential Revision: https://reviews.freebsd.org/D54691
The stock assert() works because rtld-libc includes a custom
implementation of __assert().
Reviewed by: imp, kib
Obtained from: CheriBSD
Sponsored by: AFRL, DARPA
Differential Revision: https://reviews.freebsd.org/D54712
This is a no-op on non-CHERI architectures, but is required for CHERI
where Elf_Addr is only an address and not a complete pointer.
While here, consistently use `uintptr_t *` for arrays of init/fini
function pointers.
Reviewed by: imp, kib
Effort: CHERI upstreaming
Obtained from: CheriBSD
Sponsored by: AFRL, DARPA
Differential Revision: https://reviews.freebsd.org/D54711
Store phnum in Obj_Entry instead of phsize and use that to simplify
the terminate expressions when iterating over program headers.
Reviewed by: kib
Obtained from: CheriBSD
Sponsored by: AFRL, DARPA
Differential Revision: https://reviews.freebsd.org/D54710
The __aligned attribute in the previous version applied to the location
of the pointers, not the data the pointers pointed to. While this
could be fixed by applying the attribute to a local typedef of uint16_t,
just using memcpy() for the unaligned access is simpler and ISO C.
This fixes the build on CHERI architectures which do not support
misaligned pointers and were thus failing with:
lib/libc/string/swab.c:12:18: error: alignment (1) of 'const uint16_t *' (aka 'const unsigned short *') is less than the required capability alignment (16) [-Werror,-Wcheri-capability-misuse]
12 | const uint16_t *f __aligned(1) = from;
|
Co-authored by: Jessica Clarke <jrtc27@FreeBSD.org>
Fixes: 02ebbc781f ("swab: Fix implementation to support overlapping copies")
Sponsored by: AFRL, DARPA
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D54399
Add a guard that expects a failure of the test on aarch64.
Reviewed by: emaste
Fixes: df1ea58873 ("tests: Test libexecinfo backtrace call througth signal trampoline")
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D54675
Reviewed by: markj
Fixes: ea675a43f0 ("libexec/kgdb: Add new modules and install them together with debug info")
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D54680
PF configuration files can contains many things.
Using the new page characters (i.e. ^L, \014) to mark the beginning
of parts is useful because many editors such as emacs and vim has
facilities to jump next/previous ones.
PR: 86635
MFC after: 2 weeks
Submitted by: MOROHOSHI Akihiko <moro@remus.dti.ne.jp>
Submitted by: Simon Wollwage <rootnode+freebsd@wollwage.com>
Sponsored by: Rubicon Communications, LLC ("Netgate")
When printing source limiters use the same keywords as we accept on
input, that is use 'entries' for the entries value (not 'limit') and
'limit' for the limit value (and not 'states').
Update the test case to match.
Sponsored by: Rubicon Communications, LLC ("Netgate")
This change extends pf(4) limiters so administrator
can specify action the rule executes when limit is
reached. By default when limit is reached the limiter
overrides action specified by rule to no-match.
If administrator wants to block packet instead then
rule with limiter should be changed to:
pass in from any to any state limiter test (block)
OK dlg@
Obtained from: OpenBSD, sashan <sashan@openbsd.org>, 04394254d9
Sponsored by: Rubicon Communications, LLC ("Netgate")
the code that unwinds state creation when something fails needed
the pf_source variable at the function scope to be set, but this
was masked by a declaration in the scope that sets up the source
limiting. this results in a NULL deref in the unwind code when it
needs to clean up the accounting for a source limiter.
i found it funny that i left this comment for myself in the unwinding
code:
/* who needs KASSERTS when we have NULL derefs */
ok jmatthew@
Obtained from: OpenBSD, dlg <dlg@openbsd.org>, fc9311361f
Sponsored by: Rubicon Communications, LLC ("Netgate")
linked with the plural from other text, and
matches "Source Limiters. From Atanas Vladimirov.
Obtained from: OpenBSD, sthen <sthen@openbsd.org>, d4acfc5766
Sponsored by: Rubicon Communications, LLC ("Netgate")
pfctl_parser.c, ifa_load() should distinguish between broadcast
and PPP peer address when it populates interface table for rule
parser.
OK @claudio, OK @dlg
Obtained from: OpenBSD, sashan <sashan@openbsd.org>, 2e871bec67
Sponsored by: Rubicon Communications, LLC ("Netgate")
The current ifa_load() is not paranoid enough when it deals with
information which comes from kernel. The function just ignores
sa_len member in socket address returned getifaddrs().
The issue has been reported by anton@. The idea for fix here comes
fromy claudio@.
OK @claudio, @deraadt
Obtained from: OpenBSD, sashan <sashan@openbsd.org>, a48d060175
Sponsored by: Rubicon Communications, LLC ("Netgate")
Those finishing touches were supposed to land
with source/state limiter changes. I failed to
spot them during code review.
OK dlg@
Obtained from: OpenBSD, sashan <sashan@openbsd.org>, 098c19176b
Sponsored by: Rubicon Communications, LLC ("Netgate")
John Baldwin