I was somehow convinced that insmntque calls insmntque1 with a NULL
destructor. Unfortunately this worked well enough to not immediately
blow up in simple testing.
Keep not using the destructor in previously patched filesystems though
as it avoids unnecessary casts.
Noted by: kib
Reported by: pho
This adds the PT_GETREGSET and PT_SETREGSET ptrace types. These can be
used to access all the registers from a specified core dump note type.
The NT_PRSTATUS and NT_FPREGSET notes are initially supported. Other
machine-dependant types are expected to be added in the future.
The ptrace addr points to a struct iovec pointing at memory to hold the
registers along with its length. On success the length in the iovec is
updated to tell userspace the actual length the kernel wrote or, if the
base address is NULL, the length the kernel would have written.
Because the data field is an int the arguments are backwards when
compared to the Linux PTRACE_GETREGSET call.
Reviewed by: kib
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D19831
Supplement ifindex table with generation count and use it to
serialize & restore an ifnet pointer.
Reviewed by: kp
Differential revision: https://reviews.freebsd.org/D33266
Fun note: git show e6abef09187a
The manpage has contained the following verbiage on the matter for just
under 31 years:
"At least one argument must be present in the array"
Previous to this version, it had been prefaced with the weakening phrase
"By convention."
Carry through and document it the rest of the way. Allowing argc == 0
has been a source of security issues in the past, and it's hard to
imagine a valid use-case for allowing it. Toss back EINVAL if we ended
up not copying in any args for *execve().
The manpage change can be considered "Obtained from: OpenBSD"
Reviewed by: emaste, kib, markj (all previous version)
Differential Revision: https://reviews.freebsd.org/D34045
This function will be used by coming TLS hardware receive offload support.
Differential Revision: https://reviews.freebsd.org/D32356
Discussed with: jhb@
MFC after: 1 week
Sponsored by: NVIDIA Networking
This matches user_getsockname.
Reviewed by: brooks, kib
Sponsored by: The University of Cambridge, Google Inc.
Differential Revision: https://reviews.freebsd.org/D33987
that disables any access to ptrace(2) for all processes.
Reviewed by: emaste
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D33986
Privilege checks in both functions should allow the current process to
infer information about itself, as well as use the interfaces that are
proclaimed 'debugging', for instance, procctl(2).
Note that in p_cansee() case, explicit comparision of curproc and p
avoids a race where the process might change credentials and cause
thread to compare its cached stale credentials against updated process
creds, effectively disallowing the process to observe itself.
Reviewed by: emaste
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D33986
Otherwise we end up copying one uninitialized byte into the socket
buffer.
Reported by: KMSAN
Reviewed by: jhb
MFC after: 1 week
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D33953
tc_counter_mask is an unsigned int and in the TSC timecounter is equal
to UINT_MAX, so the addition tc->tc_counter_mask + 1 can overflow to 0,
resulting in a hang during boot.
Fixes: c2705ceaeb09 ("x86: Speed up clock calibration")
Reviewed by: cperciva
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D33956
Before this change bufdaemon and bufspacedaemon threads used
kthread_shutdown() to stop activity on system shutdown. The problem is
that kthread_shutdown() has no idea about the wait channel and lock used
by specific thread to wake them up reliably. As result, up to 9 threads
could consume up to 9 seconds to shutdown for no good reason.
This change introduces specific shutdown functions, knowing how to
properly wake up specific threads, reducing wait for those threads on
shutdown/reboot from average 4 seconds to effectively zero.
MFC after: 2 weeks
Reviewed by: kib, markj
Differential Revision: https://reviews.freebsd.org/D33936
This provides information about fixed regions of the target process'
user memory map.
Reviewed by: kib
MFC after: 1 month
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D33708
The approach taken by the stack gap implementation was to insert a
random gap between the top of the fixed stack mapping and the true top
of the main process stack. This approach was chosen so as to avoid
randomizing the previously fixed address of certain process metadata
stored at the top of the stack, but had some shortcomings. In
particular, mlockall(2) calls would wire the gap, bloating the process'
memory usage, and RLIMIT_STACK included the size of the gap so small
(< several MB) limits could not be used.
There is little value in storing each process' ps_strings at a fixed
location, as only very old programs hard-code this address; consumers
were converted decades ago to use a sysctl-based interface for this
purpose. Thus, this change re-implements stack address randomization by
simply breaking the convention of storing ps_strings at a fixed
location, and randomizing the location of the entire stack mapping.
This implementation is simpler and avoids the problems mentioned above,
while being unlikely to break compatibility anywhere the default ASLR
settings are used.
The kern.elfN.aslr.stack_gap sysctl is renamed to kern.elfN.aslr.stack,
and is re-enabled by default.
PR: 260303
Reviewed by: kib
Discussed with: emaste, mw
MFC after: 1 month
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D33704
ASLR stack randomization will reappear in a forthcoming commit. Rather
than inserting a random gap into the stack mapping, the entire stack
mapping itself will be randomized in the same way that other mappings
are when ASLR is enabled.
No functional change intended, as the stack gap implementation is
currently disabled by default.
Reviewed by: kib
MFC after: 2 weeks
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D33704
Rather than fetching the ps_strings address directly from a process'
sysentvec, use this macro. With stack address randomization the
ps_strings address is no longer fixed.
Reviewed by: kib
MFC after: 2 weeks
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D33704
This will not be required with a forthcoming reimplementation of ASLR
stack randomization. Moreover, this change was not sufficient to enable
the use of a stack size limit smaller than the stack gap itself.
PR: 260303
Reviewed by: kib
MFC after: 1 week
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D33704
The size of the ps_strings structure varies between ABIs, so this is
useful for computing the address of the ps_strings structure relative to
the top of the stack when stack address randomization is enabled.
Reviewed by: kib
MFC after: 2 weeks
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D33704
The current ASLR stack gap feature will be removed, and with that the
need for the kern.stacktop sysctl is gone. All consumers have been
removed.
This reverts commit a97d697122da2bfb0baae5f0939d118d119dae33.
Reviewed by: kib
MFC after: 1 week
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D33704
Before commit 3cec5c77d617 buf_daemon() went to longer 1s sleep if
numdirtybuffers <= lodirtybuffers. After that commit new condition
!BIT_EMPTY(BUF_DOMAINS, &bdlodirty) got opposite -- true when one
or more more domains is above lodirtybuffers. As result, on freshly
booted system with no dirty buffers buf_daemon() wakes up 10 times
per second and probably only 1 time per second when there is actual
work to do.
MFC after: 1 week
Reviewed by: kib, markj
Tested by: pho
Differential revision: https://reviews.freebsd.org/D33890
VM's have little control over CPU speed, don't make matters worse
by constantly spaming console.
Reviewed by: jhb
Differential Revision: https://reviews.freebsd.org/D33902
Missed issues in truss on at least armv7 and powerpcspe need to be
resolved before recommit.
This reverts commit 3889fb8af0b611e3126dc250ebffb01805152104.
This reverts commit 1544e0f5d1f1e3b8c10a64cb899a936976ca7ea4.
This more clearly differentiates system call arguments from integer
registers and return values. On current architectures it has no effect,
but on architectures where pointers are not integers (CHERI) and may
not even share registers (CHERI-MIPS) it is necessiary to differentiate
between system call arguments (syscallarg_t) and integer register values
(register_t).
Obtained from: CheriBSD
Reviewed by: imp, kib
Differential Revision: https://reviews.freebsd.org/D33780
Prior to this commit, the TSC and local APIC frequencies were calibrated
at boot time by measuring the clocks before and after a one-second sleep.
This was simple and effective, but had the disadvantage of *requiring a
one-second sleep*.
Rather than making two clock measurements (before and after sleeping) we
now perform many measurements; and rather than simply subtracting the
starting count from the ending count, we calculate a best-fit regression
between the target clock and the reference clock (for which the current
best available timecounter is used). While we do this, we keep track
of an estimate of the uncertainty in the regression slope (aka. the ratio
of clock speeds), and stop measuring when we believe the uncertainty is
less than 1 PPM.
In order to avoid the risk of aliasing resulting from the data-gathering
loop synchronizing with (a multiple of) the frequency of the reference
clock, we add some additional spinning depending upon the iteration number.
For numerical stability and simplicity of implementation, we make use of
floating-point arithmetic for the statistical calculations.
On the author's Dell laptop, this reduces the time spent in calibration
from 2000 ms to 29 ms; on an EC2 c5.xlarge instance, it is reduced from
2000 ms to 2.5 ms.
Reviewed by: bde (previous version), kib
MFC after: 1 month
Sponsored by: https://www.patreon.com/cperciva
Differential Revision: https://reviews.freebsd.org/D33802
a helper to remount filesystem from rw to ro.
Tested by: pho
Reviewed by: markj, mckusick
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D33721
- Move busdma_lock_mutex to subr_bus_dma.c.
- Move _busdma_lock_dflt to subr_bus_dma.c. This function was named a
couple of different things previously. It is not a public API but
an internal helper used in place of a NULL pointer. The prototype
is in <sys/bus_dma.h> as not all backends include
<sys/bus_dma_internal.h>.
Reviewed by: kib
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D33694
Move mostly duplicated code in various MD bus_dma backends to support
bounce pages into sys/kern/subr_busdma_bounce.c. This file is
currently #include'd into the backends rather than compiled standalone
since it requires access to internal members of opaque bus_dma
structures such as bus_dmamap_t and bus_dma_tag_t.
Reviewed by: kib
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D33684
Now that each module handles its global and VNET initialization
itself, there is no VNET related stuff left to do in domain_init().
Differential revision: https://reviews.freebsd.org/D33541
There left only three modules that used dom_init(). And netipsec
was the last one to use dom_destroy().
Differential revision: https://reviews.freebsd.org/D33540
The function now modifies pr_usrreqs only, which are always
global. Rename it to pr_usrreqs_init().
Differential revision: https://reviews.freebsd.org/D33538
The historical BSD network stack loop that rolls over domains and
over protocols has no advantages over more modern SYSINIT(9).
While doing the sweep, split global and per-VNET initializers.
Getting rid of pr_init allows to achieve several things:
o Get rid of ifdef's that protect against double foo_init() when
both INET and INET6 are compiled in.
o Isolate initializers statically to the module they init.
o Makes code easier to understand and maintain.
Reviewed by: melifaro
Differential revision: https://reviews.freebsd.org/D33537
Currently intr_pic_add_handler either returns the PIC you gave it (which
is useless and risks causing confusion about whether it's creating
another PIC) or, on error, NULL. Instead, convert it to return an int
error code as one would expect.
Note that the only consumer of this API, arm64's gicv3_its, does not use
the return value, so no uses need updating to work with the revised API.
Reviewed by: markj, mmel
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D33341
When a softclock thread prepares to go off-CPU, the following happens in
the context of the thread:
1. callout state is locked
2. thread state is set to IWAIT
3. thread lock is switched from the tdq lock to the callout lock
4. tdq lock is released
5. sched_switch() sets td_lock to &blocked_lock
6. sched_switch() releases old td_lock (callout lock)
7. sched_switch() removes td from its runqueue
8. cpu_switch() sets td_lock back to the callout lock
Suppose a timer interrupt fires while the softclock thread is switching
off, and callout_process() schedules the softclock thread. Then there
is a window between steps 5 and 8 where callout_process() can call
sched_add() while td_lock is &blocked_lock, but this is not correct
since the thread is not logically locked.
callout_process() thus needs to spin waiting for the softclock thread to
finish switching off (i.e., after step 8 completes) before rescheduling
it, since callout_process() does not acquire the thread lock directly.
Reported by: syzbot+fb44dbf6734ff492c337@syzkaller.appspotmail.com
Fixes: 74cf7cae4d22 ("softclock: Use dedicated ithreads for running callouts.")
Reviewed by: mav, kib, jhb
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D33709
Simplify control flow around handling of the execpath length and signal
trampoline. Cache the sysentvec pointer in a local variable.
No functional change intended.
Reviewed by: kib
MFC after: 2 weeks
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D33703
Rather than using the swi infrastructure, rewrite softclock() as a
thread loop (softclock_thread()) and use it as the main routine of the
softclock threads. The threads use the CC_LOCK as the thread lock
when idle.
Reviewed by: mav, imp, kib
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D33683
The introduction of <sched.h> improved compatibility with some 3rd
party software, but caused the configure scripts of some ports to
assume that they were run in a GLIBC compatible environment.
Parts of sched.h were made conditional on -D_WITH_CPU_SET_T being
added to ports, but there still were compatibility issues due to
invalid assumptions made in autoconfigure scripts.
The differences between the FreeBSD version of macros like CPU_AND,
CPU_OR, etc. and the GLIBC versions was in the number of arguments:
FreeBSD used a 2-address scheme (one source argument is also used as
the destination of the operation), while GLIBC uses a 3-adderess
scheme (2 source operands and a separately passed destination).
The GLIBC scheme provides a super-set of the functionality of the
FreeBSD macros, since it does not prevent passing the same variable
as source and destination arguments. In code that wanted to preserve
both source arguments, the FreeBSD macros required a temporary copy of
one of the source arguments.
This patch set allows to unconditionally provide functions and macros
expected by 3rd party software written for GLIBC based systems, but
breaks builds of externally maintained sources that use any of the
following macros: CPU_AND, CPU_ANDNOT, CPU_OR, CPU_XOR.
One contributed driver (contrib/ofed/libmlx5) has been patched to
support both the old and the new CPU_OR signatures. If this commit
is merged to -STABLE, the version test will have to be extended to
cover more ranges.
Ports that have added -D_WITH_CPU_SET_T to build on -CURRENT do
no longer require that option.
The FreeBSD version has been bumped to 1400046 to reflect this
incompatible change.
Reviewed by: kib
MFC after: 2 weeks
Relnotes: yes
Differential Revision: https://reviews.freebsd.org/D33451
If GEOM is idle but the root device is not yet present when we enter
vfs_mountroot_wait_if_necessary, we call vfs_mountroot_wait to wait
for root holds (e.g. CAM or USB initialization). Upon returning from
vfs_mountroot_wait, we wait 100 ms at a time until the root device
shows up.
Since the root device most likely appeared during vfs_mountroot_wait
-- waiting for subsystems which may be responsible for the root
device is the whole purpose of that function -- it makes sense to
check if the device is now present rather than printing a warning
and pausing for 100 ms before checking.
Reviewed by: trasz
Fixes: a3ba3d09c248 Make root mount wait mechanism smarter
Sponsored by: https://www.patreon.com/cperciva
Differential Revision: https://reviews.freebsd.org/D33593
In the case of a root hold related to the initialization of a disk
device, a flurry of GEOM tasting is likely to take place as soon as
the device is initialized and the root hold is released. If we
don't wait for GEOM idle it's easy for vfs_mountroot to "win" the
race and proceed before the root filesystem GEOM is ready.
Reviewed by: imp
Sponsored by: https://www.patreon.com/cperciva
Differential Revision: https://reviews.freebsd.org/D33592
While the message is technically correct, it's not particularly
helpful in the case where we're only waiting a few ms; this case
occurs frequently on EC2 arm64 instances with CAM initialization
racing to release its root hold before vfs_mountroot reaches this
point. Only print the message if we end up waiting for more than
one second.
Sponsored by: https://www.patreon.com/cperciva
Differential Revision: https://reviews.freebsd.org/D33591
This will always sleep at least once, so it's a slow path by definition.
Reviewed By: kib
Sponsored By: EPSRC
Differential Revision: https://reviews.freebsd.org/D33387
Functions manipulating mbuf tags are using an int type for passing the
'type' parameter, but the internal tag storage is using a 16bit
integer to store it. This leads to the following code:
t = m_tag_alloc(...,0xffffffff,...,...);
m_tag_prepend(m, t);
r = m_tag_locate(m ,...,0xffffffff, NULL);
Returning r == NULL because m_tag_locate doesn't truncate the type
parameter when doing the match. This is unexpected because the type of
the 'type' parameter is int, and the caller doesn't need to know about
the internal truncations.
Fix this by making the 'type' parameter of type uint16_t in order to
match the size of its internal storage and make it obvious to the
caller the actual size of the parameter.
While there also use uint uniformly replacing the existing u_int
instances.
Reviewed by: kp, donner, glebius
Differential revision: https://reviews.freebsd.org/D33680
When a DMA request using bounce pages completes, a swi is triggered to
schedule pending DMA requests using the just-freed bounce pages. For
a long time this bus_dma swi has been tied to a "virtual memory" swi
(swi_vm). However, all of the swi_vm implementations are the same and
consist of checking a flag (busdma_swi_pending) which is always true
and if set calling busdma_swi. I suspect this dates back to the
pre-SMPng days and that the intention was for swi_vm to serve as a
mux. However, in the current scheme there's no need for the mux.
Instead, remove swi_vm and vm_ih. Each bus_dma implementation that
uses bounce pages is responsible for creating its own swi (busdma_ih)
which it now schedules directly. This swi invokes busdma_swi directly
removing the need for busdma_swi_pending.
One consequence is that the swi now works on RISC-V which had previously
failed to invoke busdma_swi from swi_vm.
Reviewed by: imp, kib
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D33447
