Gleb Smirnoff [Thu, 11 Aug 2022 16:19:36 +0000 (09:19 -0700)]
keysock: do not use raw socket code
This makes key socket implementation self contained and removes one
of the last dependencies on the raw socket code and pr_output method.
There are very subtle API visible changes:
- now key socket would return EOPNOTSUPP instead of EINVAL on
syscalls that are not supposed to be called on a key socket.
- key socket buffer sizes are now controlled by net.key sysctls instead
of net.raw. The latter were not documented anywhere, and even Internet
search doesn't find any references or discussions related to them.
Gleb Smirnoff [Thu, 11 Aug 2022 16:19:36 +0000 (09:19 -0700)]
rtsock: do not use raw socket code
This makes routing socket implementation self contained and removes one
of the last dependencies on the raw socket code and pr_output method.
There are very subtle API visible changes:
- now routing socket would return EOPNOTSUPP instead of EINVAL on
syscalls that are not supposed to be called on a routing socket.
- routing socket buffer sizes are now controlled by net.rtsock
sysctls instead of net.raw. The latter were not documented
anywhere, and even Internet search doesn't find any references
or discussions related to these sysctls.
Gleb Smirnoff [Thu, 11 Aug 2022 16:19:36 +0000 (09:19 -0700)]
netinet: get interface event notifications directly via EVENTHANDLER(9)
The old mechanism of getting them via domains/protocols control input
is a relict from the previous century, when nothing like EVENTHANDLER(9)
existed yet. Retire PRC_IFDOWN/PRC_IFUP as netinet was the only one
to use them.
Andrew Turner [Thu, 11 Aug 2022 12:04:58 +0000 (13:04 +0100)]
Adjust function definition in subr_devmap.c to avoid clang 15 warning
With clang 15, the following -Werror warning is produced:
sys/kern/subr_devmap.c:87:19: error: a function declaration without a prototype is deprecated in all versions of C [-Werror,-Wstrict-prototypes]
devmap_print_table()
^
void
This is because devmap_print_table() and devmap_lastaddr() are declared
with a (void) argument list, but defined with an empty argument list.
Make the definition match the declaration.
Andrew Turner [Thu, 11 Aug 2022 12:02:50 +0000 (13:02 +0100)]
Adjust function definition in ofw to avoid clang 15 warning
With clang 15, the following -Werror warning is produced:
sys/dev/ofw/openfirm.c:826:9: error: a function declaration without a prototype is deprecated in all versions of C [-Werror,-Wstrict-prototypes]
OF_enter()
^
void
This is because OF_enter() and OF_exit are declared with a (void)
argument list, but defined with an empty argument list. Make the
definition match the declaration.
Andrew Turner [Thu, 11 Aug 2022 12:02:33 +0000 (13:02 +0100)]
Adjust function definition in hwpmc to avoid clang 15 warning
With clang 15, the following -Werror warning is produced:
sys/dev/hwpmc/hwpmc_arm64.c:530:21: error: a function declaration without a prototype is deprecated in all versions of C [-Werror,-Wstrict-prototypes]
pmc_arm64_initialize()
^
void
This is because many of the functions are declared with a (void)
argument list, but defined with an empty argument list. Make the
definition match the declaration.
routing: populate fibs with interface routes after growing net.fibs.
Currently it is possible to extend number of fibs in runtime, but this
functionality is of limited use when net.add_addrs_all_fibs is
non-zero, as the routing tables are created empty.
This change automatically populate newly-created fibs with the kernel-originated
interface routes (filtered by RTF_PINNED flag) if net.add_addrs_all_fibs
is set.
Kristof Provost [Thu, 11 Aug 2022 08:34:01 +0000 (10:34 +0200)]
if_ovpn tests: Test using a TCP socket for DCO
This used to trigger panics, so try to reproduce it.
Create an if_ovpn interface, set a new peer on it with a TCP fd (as
opposed to the expected UDP) and ensure that this is rejected.
Warner Losh [Thu, 11 Aug 2022 03:19:01 +0000 (21:19 -0600)]
stand: impose 510,000 byte limit for /boot/loader and /boot/pxeldr
The BIOS method of booting imposes an absolute limit of 640k for the
size of the program being run due to btx. In practice, this means that
programs larger than about 500kiB will fail in odd ways as the stack /
heap will overflow.
Pick 510,000 as the cutoff line semi-arbitrarily. loader_lua is now
almost too big and we want to break the build when it crosses this
threshold. In my experience, below 500,000 always works, above 520,000
always seems to fail with things getting bad somewhere between 512,000
to 515,000. 510,000 is as close to the line as I think we can go, though
experience may dictate we need to lower this in the future.
This is at-best a stop-breakage until we have a better way to subset the
boot loader for BIOS booting to allow better, more fined-tuned
/boot/loaders for the many different environments they have to run
in. This likely means we'll have a graphical loader than understands a
few filesystmes for installation, and a non-graphical loader that
understands the most filesystems possible for everything else in the
future. Our build infrastructure needs some work before we can do that,
however.
At this late date, it likely isn't worth the efforts to move parts of
the loader into high memory. There's a number of assumptions about where
the stack is, where buffers reside, etc that are fulfilled when it lives
in the first 640k that would need bounce buffers and/or other counter
measures if we were to split it up. All BIOS calls are done in 16-bit
mode with SEG:OFF addresses, requiring them to be in the first 640k of
RAM. And nearly all machines in the last decade can boot with UEFI
(though there's some exceptions, so it isn't worth killing outright
yet).
Warner Losh [Thu, 11 Aug 2022 03:18:32 +0000 (21:18 -0600)]
stand: i386/amd64: Always use elf format for /boot/loader and pxeldr
The first level boot blocks have understood how to load ELF code since
1999. Switch /boot/loader and /boot/pxeldr over to being ELF format so
that in-tree tools can examine them more closely. In addition, one
could, in theory, now have a 'lo-mem' and a 'hi-mem' segment (though a
lot of work would need to be done with bounce buffers, btx, code segment
marking, etc for an arrangement like that to work).
As far as I can tell, this is the last a.out binary in the tree. There
are several raw binaries left, but everything else is ELF.
Emmanuel Vadot [Fri, 21 May 2021 11:33:34 +0000 (13:33 +0200)]
sdio: Always use increment address for read/write_4
SDIO CMD53 (RW Extented) can either write to the same address (useful for FIFO)
or auto increment the destination address (to write to multiple registers).
It is more logical to have read/write_4 to use incremental mode and make other
helper function for writing to a FIFO destination especially since most FIFO
write/read will be 8bits based and not 32bits based.
Emmanuel Vadot [Fri, 21 May 2021 09:56:39 +0000 (11:56 +0200)]
sdio: Change the sdio helper name and arg order
Do not use b/l but _1/_4 also address comes first and then data.
This makes them closer to something like bus_space_{read,write}
We have no users in the tree.
devd: move all devd notification logic to a separate file.
Currently, subr_bus.c shares logic for (a) maintaining all HW devices
(e.g. discovery/attach/detach logic) and (b) generic devctl notification
layer for devices/PMU/GEOM/interfaces/etc).
These two subsystems share really tiny interaction interface, composed of 3
notification functions. With that in mind, move devctl layer to a
separate file, establishing a clear notification interface between the
sub.c bus layer and the provider (devctl).
The primary driver of this change is netlink implementation (D36002).
The idea is to propagate device-level events to netlink as well, so all
netlink customers can subscribe to these changes.
The long-term goal is to deprecate devctl and to use netlink as the
kernel<> userland transport provided netlink gets enough traction.
routing: move rtentry and subscription code out of route_ctl.c
route_ctl.c size has grown considerably since initial introduction.
Factor out non-relevant parts:
* all rtentry logic, such as creation/destruction and accessors
goes to net/route/route_rtentry.c
* all rtable subscription logic goes to net/route/route_subscription.c
routing: add rib_<add|del>_route_px() functions operating with nexthops.
This change adds public KPI to work with routes using pre-created
nexthops, instead of using data from addrinfo structures. These
functions will be later used for adding/deleting kernel-originated
routes and upcoming netlink protocol.
As a part of providing this KPI, low-level route addition code has been
reworked to provide more control over route creation or change.
Specifically, a number of operation flags
(RTM_F_<CREATE|EXCL|REPLACE|APPEND>) have been added, defining the
desired behaviour the the route already exists (or not exists). This
change required some changes in the multipath addition code, resulting
in moving this code to route_ctl.c, rendering mpath_ctl.c empty.
routing: split nexthop creation and rtentry creation.
This change is required for the upcoming introduction of the next
nexhop-based operations KPI, as it will create rtentry and nexthops
at different stages of route table modification.
* Use same filter func (rib_filter_f_t) for nexhtop groups to
simplify callbacks.
* simplify conditional route deletion & remove the need to pass
rt_addrinfo to the low-level deletion functions
* speedup rib_walk_del() by removing an additional per-prefix lookup
This and the follow-up routing-related changes target to remove or
reduce `struct rt_addrinfo` usage and use recently-landed nhop(9)
KPI instead.
Traditionally `rt_addrinfo` structure has been used to propagate all necessary
information between the protocol/rtsock and a routing layer. Many
functions inside routing subsystem uses it internally. However, using
this structure became somewhat complicated, as there are too many ways
of specifying a single state and verifying data consistency is hard.
For example, arerouting flgs consistent with mask/gateway sockaddr pointers?
Is mask really a host mask? Are sockaddr "valid" (e.g. properly zeroed, masked,
have proper length)? Are they mutable? Is the suggested interface specified
by the interface index embedded into the sockadd_dl gateway, or passed
as RTAX_IFP parameter, or directly provided by rti_ifp or it needs to
be derived from the ifa?
These (and other similar) questions have to be considered every time when
a function has `rt_addrinfo` pointer as an argument.
The new approach is to bring more control back to the protocols and
construct the desired routing objects themselves - in the end, it's the
protocol/subsystem who knows the desired outcome.
This specific diff changes the following:
* add explicit basic low-level radix operations:
add_route() (renamed from add_route_nhop())
delete_route() (factored from change_route_nhop())
change_route() (renamed from change_route_nhop)
* remove "info" parameter from change_route_conditional() as a part
of reducing rt_addrinfo usage in the internal KPIs
* add lookup_prefix_rt() wrapper for doing re-lookups after
RIB lock/unlock
Gleb Smirnoff [Wed, 10 Aug 2022 18:09:34 +0000 (11:09 -0700)]
tcp: utilize new solisten_clone() and solisten_enqueue()
This streamlines cloning of a socket from a listener. Now we do not
drop the inpcb lock during creation of a new socket, do not do useless
state transitions, and put a fully initialized socket+inpcb+tcpcb into
the listen queue.
Before this change, first we would allocate the socket and inpcb+tcpcb via
tcp_usr_attach() as TCPS_CLOSED, link them into global list of pcbs, unlock
pcb and put this onto incomplete queue (see 6f3caa6d815). Then, after
sonewconn() we would lock it again, transition into TCPS_SYN_RECEIVED,
insert into inpcb hash, finalize initialization of tcpcb. And then, in
call into tcp_do_segment() and upon transition to TCPS_ESTABLISHED call
soisconnected(). This call would lock the listening socket once again
with a LOR protection sequence and then we would relocate the socket onto
the complete queue and only now it is ready for accept(2).
Gleb Smirnoff [Wed, 10 Aug 2022 18:09:34 +0000 (11:09 -0700)]
sockets: provide solisten_clone(), solisten_enqueue()
as alternative KPI to sonewconn(). The latter has three stages:
- check the listening socket queue limits
- allocate a new socket
- call into protocol attach method
- link the new socket into the listen queue of the listening socket
The attach method, originally designed for a creation of socket by the
socket(2) syscall has slightly different semantics than attach of a socket
cloned by listener. Make it possible for protocols to call into the
first stage, then perform a different attach, and then call into the
final stage. The first stage, that checks limits and clones a socket
is called solisten_clone(), and the function that enqueues the socket
is solisten_enqueue().
Changing mode on a pin (input/output/pullup/pulldown) is a bit slow.
Improve this by caching what we can.
We need to check if the pin is in gpio mode, do that the first time
that we have a request for this pin and cache the result. We can't do
that at attach as we are a child of rk_pinctrl and it didn't finished
its attach then.
Cache also the flags specific to the pinctrl (pullup or pulldown) if the
pin is in input mode.
Cache the registers that deals with input/output mode and output value. Also
remove some register reads when we change the direction of a pin or when we
change the output value since the bit changed in the registers only affect output
pins.
Gleb Smirnoff [Wed, 10 Aug 2022 14:32:37 +0000 (07:32 -0700)]
tcp: address a wire level race with 2 ACKs at the end of TCP handshake
Imagine we are in SYN-RCVD state and two ACKs arrive at the same time,
both valid, e.g. coming from the same host and with valid sequence.
First packet would locate the listening socket in the inpcb database,
write-lock it and start expanding the syncache entry into a socket.
Meanwhile second packet would wait on the write lock of the listening
socket. First packet will create a new ESTABLISHED socket, free the
syncache entry and unlock the listening socket. Second packet would
call into syncache_expand(), but this time it will fail as there
is no syncache entry. Second packet would generate RST, effectively
resetting the remote connection.
It seems to me, that it is impossible to solve this problem with
just rearranging locks, as the race happens at a wire level.
To solve the problem, for an ACK packet arrived on a listening socket,
that failed syncache lookup, perform a second non-wildcard lookup right
away. That lookup may find the new born socket. Otherwise, we indeed
send RST.
netinet6: allow ND entries creation for all directly-reachable
destinations.
The current assumption is that kernel-handled rtadv prefixes along with
the interface address prefixes are the only prefixes considered in
the ND neighbor eligibility code.
Change this by allowing any non-gatewaye routes to be eligible. This
will allow DHCPv6-controlled routes to be correctly handled by
the ND code.
Refactor nd6_is_new_addr_neighbor() to enable more deterministic
performance in "found" case and remove non-needed
V_rt_add_addr_allfibs handling logic.
Michael Tuexen [Mon, 8 Aug 2022 11:07:10 +0000 (13:07 +0200)]
tcp: improve BBLog for output events when using the FreeBSD stack
Put the return value of ip_output()/ip6_output in the output event
instead of adding another one in case of an error. This improves
consistency with other similar places.
Reviewed by: rscheff
Sponsored by: Netflix, Inc.
Differential Revision: https://reviews.freebsd.org/D36085
Xin LI [Wed, 10 Aug 2022 00:27:54 +0000 (17:27 -0700)]
arc4random(3): Reduce diff with OpenBSD.
The main change was v1.57 by djm@:
Randomise the rekey interval a little. Previously, the chacha20
instance would be rekeyed every 1.6MB. This makes it happen at a
random point somewhere in the 1-2MB range.
Mark Johnston [Tue, 9 Aug 2022 20:08:13 +0000 (16:08 -0400)]
dtrace/amd64: Implement emulation of call instructions
Here, the provider is responsible for updating the trapframe to redirect
control flow and for computing the return address. Once software-saved
registers are restored, the emulation shifts the remaining context down
on the stack to make space for the return address, then copies the
address provided by the invop handler. dtrace_invop() is modified to
allocate temporary storage space on the stack for use by the provider to
return the return address.
This is to support a new provider for amd64 which can instrument
arbitrary instructions, not just function entry and exit instructions as
FBT does.
In collaboration with: christos
Sponsored by: Google, Inc. (GSoC 2022)
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
Mark Johnston [Tue, 9 Aug 2022 20:08:09 +0000 (16:08 -0400)]
fbt/x86: Extract arg1 for return probes from the trapframe
dtrace invop handlers have access to the whole trapframe, just use that
to extract %rax/%eax for return probes instead of relying on an
additional parameter to the handler. No functional change intended.
MFC after: 2 weeks
Sponsored by: The FreeBSD Foundation
Jessica Clarke [Tue, 9 Aug 2022 21:57:47 +0000 (22:57 +0100)]
etcupdate: Add a -N flag to perform a NO_ROOT build
This is in preparation for including an etcupdate tree when performing a
-DNO_ROOT release image build. Although -DNO_ROOT can be passed via -M,
to be useful we need to mangle the resulting METALOG to mirror the
various cleanups to the tree that are done after the build (removing
generated files, empty files and empty directories), so etcupdate needs
its own flag.
Jessica Clarke [Tue, 9 Aug 2022 21:57:22 +0000 (22:57 +0100)]
etcupdate: Prefer POSIX -depth to BSD -d
This is in preparation for building an etcupdate tree on non-FreeBSD
when building release images. The -d option is documented as a
BSD-specific equivalent to the POSIX -depth primary. Whilst GNU find
sort of accepts it in an attempt to be compatible, it still doesn't
permit it coming before the paths, unlike BSD find, and prints a
deprecation warning either way. Thus, use the equivalent POSIX -depth to
ensure it works correctly and without warning everywhere.
Jessica Clarke [Tue, 9 Aug 2022 21:57:01 +0000 (22:57 +0100)]
release: Forward ${MAKE} to etcupdate via the new -m flag
This is in preparation for non-FreeBSD builds where make is GNU make and
so etcupdate needs to know the name of or path to the bmake binary to
use for its own builds.
Jessica Clarke [Tue, 9 Aug 2022 21:56:19 +0000 (22:56 +0100)]
etcupdate: Add a -m flag to change the make binary that's run
This will allow release/Makefile to forward on ${MAKE} to allow building
on non-FreeBSD systems where ${MAKE} is something other than make, as
make is typically GNU make in such situations.
Jessica Clarke [Tue, 9 Aug 2022 21:52:47 +0000 (22:52 +0100)]
release: Use in-tree etcupdate for build
This is in preparation for non-FreeBSD and -DNO_ROOT builds. On
non-FreeBSD there is no host etcupdate to use, and -DNO_ROOT will
require additional flags that may not be supported by the host's
etcupdate when building on FreeBSD. Moreover, there's no guarantee
anyway that the host's etcupdate is quite right for the current tree;
upgrading from source only requires that the host's is good enough for
-p which just manually copies master.passwd and group, the rest of the
upgrade is done post-installworld. For example, should a new set of
autogenerated files be added that etcupdate is taught about, the host
won't know about them and so the bootstrapped current tree will
incorrectly contain them, leading to spurious diffs on the installed
system.
Mark Johnston [Mon, 25 Jul 2022 20:53:21 +0000 (16:53 -0400)]
vm_fault: Shoot down shared mappings in vm_fault_copy_entry()
As in vm_fault_cow(), it's possible, albeit rare, for multiple vm_maps
to share a shadow object. When copying a page from a backing object
into the shadow, all mappings of the source page must therefore be
removed. Otherwise, future operations on the object tree may detect
that the source page is fully shadowed and thus can be freed.
Approved by: so
Security: FreeBSD-SA-22:11.vm
Reviewed by: alc, kib
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D35635
elf_note_prpsinfo: handle more failures from proc_getargv()
Resulting sbuf_len() from proc_getargv() might return 0 if user mangled
ps_strings enough. Also, sbuf_len() API contract is to return -1 if the
buffer overflowed. The later should not occur because get_ps_strings()
checks for catenated length, but check for this subtle detail explicitly
as well to be more resilent.
The end result is that p_comm is used in this situations.
Approved by: so
Security: FreeBSD-SA-22:09.elf
Reported by: Josef 'Jeff' Sipek <jeffpc@josefsipek.net>
Reviewed by: delphij, markj
admbugs: 988
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D35391
This driver supports the auto negotiation mode between the copper and fiber
ports.
This PHY has two independent PHYs (one for copper and other for fiber) but in
this case the functionality is presented as a single PHY for easy management.
Mike Karels [Tue, 9 Aug 2022 12:08:09 +0000 (07:08 -0500)]
netinet tests: Add test for IPv6 mapped-v4 bind problem
Test fix in 637f317c6d9c, verifying that when ports run out, we get
an EADDRNOTAVAIL error from bind() rather than an EADDRINUSE error
from connect(). Use small port range to exhaust ports and see which
error happens.
Eugene Grosbein [Mon, 8 Aug 2022 22:21:02 +0000 (05:21 +0700)]
syslog(3): unbreak log generation using fabricated PID
Recover application ability to supply fabricated PID
embedded into ident that was lost when libc switched
to generation of RFC 5424 log messages, for example:
logger -t "ident[$$]" -p user.notice "test"
It is essential for long running scripts.
Also, this change unbreaks matching resulted entries
by ident in syslog.conf:
!ident
*.* /var/log/ident.log
Without the fix, the log (and matching) was broken:
Aug 1 07:36:58 hostname ident[123][86483]: test
Now it works as expected and worked before breakage:
Emmanuel Vadot [Mon, 8 Aug 2022 18:21:08 +0000 (20:21 +0200)]
linuxkpi: io.h: Only exclude armv6 and armv7 for asm/set_memory.h
Other arches like powerpc* needs it.
Fixes: d387a1b4b1996 ("linuxkpi: io.h: Do not include asm/set_memory.h for armv6 and armv7") Fixes: 789dbdbb48574 ("linuxkpi: Add arch_io_{reserve,free}_memtype_wc")
Sponsored by: Beckhoff Automation GmbH & Co. KG
John Baldwin [Mon, 8 Aug 2022 18:21:54 +0000 (11:21 -0700)]
cxgbe: Handle requests for TLS key allocations with no TLS key storage.
If an adapter advertises support for TLS keys but an empty TLS key
storage area in on-board memory, fail the request rather than invoking
vmem_alloc on an uninitialized vmem.
Per the EHABI32 specification, __cxa_end_cleanup must take care to
preserve registers before calling _Unwind_Resume(). So, libcxxrt uses
an assembly stub which preserves caller-saved registers around the call
to __cxa_get_cleanup(). But:
- it failed to restore them properly,
- it did not preserve the link register.
Fix both of these problems. This is needed to fix exception unwinding
on FreeBSD with LLVM 14. Note that r4 is callee-saved but is pushed
onto the stack to preserve stack pointer alignment.
Sponsored-by: The FreeBSD Foundation
MFC after: 1 week
thread_create(): call cpu_copy_thread() after td_pflags is zeroed
By calling the function too early we might still have the td_pflags
value cached from the previous struct thread use. cpu_copy_thread()
depends on correct value for TDP_KTHREAD at least on x86.
Reported, bisected, and tested by: pho
Reviewed by: markj
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D36069