The includes build is kind of funky, as we support either copying or
symlinking files into /usr/include. For `copies`, we were supplying
the include/ ${TAG_ARGS}, which puts packages into `FreeBSD-runtime`,
without any consideration to the fact that we're installing headers.
Let's copy the approach that the `symlinks` target uses for now, and
add ",dev" to the TAG_ARGS so that headers at least end up in
FreeBSD-runtime-dev, which is more appropriate. Some of these includes
are actually technically supposed to be in *other* packages and their
INCSGROUP's PACKAGE setting is actually correct, but this is less
trivial to solve. This is a bandaid to fix the immediate problem of
some headers ending up in two different packages.
PR: 267526
Reviewed by: dfr, manu
Differential Revision: https://reviews.freebsd.org/D37256
Work is ongoing to add support for pfsync over IPv6. This required some
changes to allow for differentiating between the two families in a more
generic way.
This patch converts the relevant ioctls to using nvlists, making future
extensions (such as supporting IPv6 addresses) easier.
Sponsored by: InnoGames GmbH
Differential Revision: https://reviews.freebsd.org/D36277
markj says:
...the assertion is incorrect and should simply be removed.
It has been racy since we removed the use of the page hash
lock to synchronize wiring of pages.
PR: 267621
Reviewed by: markj, Anton Rang <rang@acm.org>
MFC after: 1 week
Sponsored by: Dell Inc.
Differential Revision: https://reviews.freebsd.org/D37320
The kernel handles the managment of UFS/FFS snapshots. Since UFS/FFS
updates filesystem data (rather than always writing changes to new
locations like ZFS), the kernel must check every filesystem write
to see if the block being written is part of a snapshot. If it is
part of a snapshot, then the kernel must make a copy of the old
block value into a newly allocated block for the snapshot before
allowing the write to be done. Similarly, if a block is being freed,
the kernel must check to see if it is part of a snapshot and let
the snapshot claim the block rather than freeing it for future use.
When a snapshot is freed, its blocks need to be offered to older
snapshots and freed only if no older snapshots wish to claim them.
When snapshots were added to UFS/FFS they were integrated into soft
updates and just a small part of the management of snapshots needed
to be added to fsck_ffs(8) as soft updates minimized the set of
snapshot changes that might need correction. When journaling was
added to soft updates a much more complete knowledge of snapshots
needed to be added to fsck_ffs(8) for it to be able to properly
handle the filesystem changes that a journal rollback needs to do
(specifically the freeing and allocation of blocks). Since this
functionality was unavailable, the use of snapshots was disabled
when running with journaled soft updates.
This set of changes imports the kernel code for the management of
snapshots to fsck_ffs(8). With this code in place it will become
possible to enable snapshots when running with journalled soft
updates. The most immediate benefit will be the ability to use
snapshots to take consistent filesystem dumps on live filesystems.
Future work will be done to update fsck_ffs(8) to be able to use
snapshots to run in background on live filesystems running with
journaled soft updates.
Reviewed by: kib
Tested by: Peter Holm
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D36491
This can sometimes happen with broken HDDs.
MFC after: 2 weeks
Sponsored by: Axcient
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D37313
The NAT module use of the tcphdr.th_x2 field now collides with the
use of this TCP header flag as AccECN (AE) bit. Use the topmost
bit instead to allow negotiation of AccECN across a NAT device.
Event: IETF 115 Hackathon
Reviewed By: #transport, tuexen
MFC after: 3 days
Sponsored by: NetApp, Inc.
Differential Revision: https://reviews.freebsd.org/D37300
mta_start_script is used by /etc/rc.d/othermta which only execute
something if this variable is set to something else than
/etc/rc.sendmail
MFC after: 1 week
We already did the necessary $TMPDIR fallback, if it's going to be used.
Skip the later check so that we don't accidentally override our -p
argument.
Fixes: ac6f924e1c ("mktemp: add -p/--tmpdir argument")
Sponsored by: Klara, Inc.
The system call returns the head of the robust futex list. The list head is stored
in the location pointed to by the head argument. When copying data between address
spaces use proper head storage size as it depends on an emulated ABI.
PR: 267616
MFC after: 3 days
The previous commit lost an implicit struct socket * cast. Use an
inline function instead as the macro is already rather long.
Fixes: e1401f7579 cxgbe: use standard sototcpcb() accessor macro to get socket's tcpcb
Sponsored by: Chelsio Communications
The inp_socket is cleared only in in_pcbdetach(), which for TCP is
always accompanied with inp_pcbfree(). An inpcb that went through
in_pcbfree() shall never be returned by any kind of pcb lookup.
Reviewed by: tuexen
Differential revision: https://reviews.freebsd.org/D37062
These functions tcp_notify(), tcp_drop_syn_sent() and tcp_mtudisc()
are called from tcp*_ctlinput*() right after successfull
in_pcblookup*(). They shall never get a pcb that is dropped.
The in_pcbdrop() KPI, which is used solely by TCP, allows to remove a
pcb from hash list and mark it as dropped. The comment suggests that
such pcb won't be returned by lookups. Indeed, every call to
in_pcblookup*() is accompanied by a check for INP_DROPPED. Do what
comment suggests: never return such pcbs and remove unnecessary checks.
Reviewed by: tuexen
Differential revision: https://reviews.freebsd.org/D37061
Add virt_to_phys() as a define to vtophys().
This is used by a wireless driver for dma related work; sigh.
MFC after: 3 days
Reviewed by: emaste
Differential Revision: https://reviews.freebsd.org/D37301
In addition to (*func) tasklet also seems to have a (*callback) with
a different argument. Add support for this and add tasklet_setup()
as well for support in more drivers.
The from_tasklet() definition is duplicated in the tree; hide it there
under #ifndef to avoid a re-definition. People should generally add
LinuxKPI bits to linuxkpi rather than private files if they also rely
on other LinuxKPI bits.
X-MFC: DO NOT MFC, space allocated by drivers not us.
Reviewed by: emaste
Differential Revision: https://reviews.freebsd.org/D37216
Add a static set of cpumasks for all (possible) cpus with only the one
indexed cpu enabled in each set.
This is needed for cpumask_of(_cpuid) which returns a cpumask (cpuset)
with only cpu _cpuid enabled and is used by one wireless driver at least.
MFC after: 3 days
Reviewed by: emaste
Differential Revision: https://reviews.freebsd.org/D37223
This adds the capability for a modular congestion control
to select which variant of ECN-capable-transport it wants to use
when sending out elegible segments. As an initial CC to utilize
this, DCTCP was selected.
Event: IETF 115 Hackathon
Reviewed By: tuexen, #transport
Sponsored by: NetApp, Inc.
Differential Revision: https://reviews.freebsd.org/D24869
Polarity inversion register was mistekanly filled with the value of
the direction configuration register.
Correct that.
Reviewed by: kd
Obtained from: Semihalf
Differential Revision: https://reviews.freebsd.org/D37262
If a -m argument is given to update, it is passed through to arc diff
when updating each review. Note that if an empty message is specified
via -m, arc diff will update the review without adding a note.
If an -m argument is not given, then the user's editor is invoked by
arc to supply a message for each review matching the previous
behavior.
This can be used to simplify the process for updating a set of
reviews, e.g.:
git checkout foo
git rebase main
git arc update -m "Rebase" main..
This will rebase the 'foo' branch and update the reviews for all
commits on the branch without invoking the user's editor separately
for each review.
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D37260
These values come from section 7.7.11 ("ACS Extended Capability") of the
PCI Express Base Specification Revision 6.0, dated 16 Dec 2021.
MFC after: 1 week
Sponsored by: Chelsio Communications
Reviewed by: kib@
Differential Revision: https://reviews.freebsd.org/D37270
On arm64 with per-thread stack canaries enabled (the PERTHREAD_SSP
option), the compiler may load curthread->td_md.md_canary in function
prologues. This is not safe in data_abort(); see commit 2c10be9e06.
Thus, sanitizer runtimes must also avoid accessing the current thread's
canary.
Since SSP has limited utility in sanitizer runtimes, simply disable it
unconditionally to avoid unpleasant surprises in the future.
MFC after: 2 weeks
Sponsored by: Juniper Networks, Inc.
Sponsored by: Klara, Inc.
With PERTHREAD_SSP configured, the compiler's stack-smashing protection
uses a per-thread canary value instead of a global value. The value is
stored in td->td_md.md_canary; the sp_el0 register always contains a
pointer to that value, and certain functions selected by the compiler
will store the canary value on the stack as a part of the function
prologue (and will verify the copy as part of the epilogue). In
particular, the thread structure may be accessed.
This happens to occur in data_abort(), which leads to the same problem
addressed by commit 2c10be9e06 ("arm64: Handle translation faults for
thread structures"). This commit fixes that directly, by disabling SSP
in data_abort() and a couple of related functions by using a function
attribute. It also moves the update of sp_el0 out of C code in case
the compiler decides to start checking the canary in pmap_switch()
someday.
A different solution might be to move the canary value to the PCB, which
currently lives on the kernel stack and isn't subject to the same
problem as thread structures (if only because guard pages inhibit
superpage promotion). However, there isn't any particular reason the
PCB has to live on the stack today; on amd64 it is embedded in struct
thread, reintroducing the same problem. Keeping the reference canary
value at the top of the stack is also rather dubious since it could be
clobbered by a sufficiently large stack overflow.
A third solution could be to go back to the approach of commit
5aa5420ff2, and modify UMA to use the direct map for thread structures
even if KASAN is enabled. But, transient promotions and demotions in
the direct map are possible too.
Reviewed by: alc, kib, andrew
MFC after: 1 month
Sponsored by: Juniper Networks, Inc.
Sponsored by: Klara, Inc.
Differential Revision: https://reviews.freebsd.org/D37255
Old versions of FreeBSD do not support using id 0 to refer to the
current pid for procctl, so pass getpid() explicitly.
Although this is not required in current FreeBSD branches I am merging
it to reduce differences with upstream.
Obtained from: OpenSSH commit 0f7e1eba5525
This is just a dummy enum and struct in order to make drivers compile
more happily as some parts are simply not hidden behind #ifdefs and
this avoids a longer-term maintenance problem.
MFC after: 3 days
Reviewed by: emaste
Differential Revision: https://reviews.freebsd.org/D37215
Add devm_request_irq() needed by a driver. Turns out all we need
is a wrapper with the right arguments to lkpi_request_irq().
MFC after: 3 days
Reviewed by: emaste
Differential Revision: https://reviews.freebsd.org/D37217