in_pcblisten() moves an inpcb from the per-group list into the array, at
which point it becomes visible to inpcb lookups in the datapath. It
assumes that there is space in the array for this, but that's not
guaranteed, since in_pcbinslbgrouphash() doesn't reserve space in the
array if the inpcb isn't associated with a listening socket.
We could resize the array in in_pcblisten(), but that would introduce a
failure case where there currently is none. Instead, keep track of the
number of pending inpcbs as well, and modify in_pcbinslbgrouphash() to
reserve space for each pending (i.e., not-yet-listening) inpcb.
Add a regression test.
Reviewed by: glebius
Reported by: netchild
Fixes: 7cbb6b6e28 ("inpcb: Close some SO_REUSEPORT_LB races, part 2")
Differential Revision: https://reviews.freebsd.org/D49100
Kyua and ATF speak different naming styles. In this case, the
unprivileged user property can be named with underscore on the Kyua
side, and with a hyphen on the ATF side. Sometimes it is not obvious
which style should be used in which situation. For instance, a test case
may require this configuration property being set using require.config.
Also, a test case may want to read the property using something like
atf_tc_get_config_var(). Which names should be used in these cases?
From the perspective of the original code, it is expected to be this:
require.config unprivileged-user
atf_tc_get_config_var(tc, "unprivileged-user")
But, as long as Kyua is the main interface, its users expect to work
with kyua.conf(5), which says that it must be named as unprivileged_user
(with underscore). As a result, test authors tend to do this instead:
require.config unprivileged_user
atf_tc_get_config_var(tc, "unprivileged_user")
Kyua already has hacks to understand both unprivileged_user and
unprivileged-user coming from require.config. And this patch covers the
missing second part -- make Kyua pass both names back to ATF as two
identical configuration properties named different ways.
Reviewed by: ngie, asomers
MFC after: 2 weeks
Differential Revision: https://reviews.freebsd.org/D49039
net80211 has inconsistent locking when calling into (*ic_ampdu_rx_stop)().
Make use of 054c5ddf58 and conditionally check if the caller
locked or not and if locked temporary drop the lock to avoid sleeping
on a non-sleepaable lock during the downcall into the driver.
Sponsored by: The FreeBSD Foundation
MFC after: 3 days
ieee80211_sta_recalc_aggregates() shows up in TODO traces but there is
nothing we have to do there until we have active links (MLO support).
Make the TODO conditional for a time when we will get there.
Sponsored by: The FreeBSD Foundation
MFC after: 3 days
Define cfg80211 IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_MASK to
net80211 IEEE80211_VHTCAP_SUPP_CHAN_WIDTH_MASK.
We should likely at some point make a sweep and replace all the
values with the defines from the comments for the matching net80211
version.
Sponsored by: The FreeBSD Foundation
MFC after: 3 days
When setting the per-band supp_rates bitfield check for mandatory
rates only. We cannot easily say at that point for 2Ghz whether
11g is supported so assume these days it is not pure-b.
Sponsored by: The FreeBSD Foundation
MFC after: 3 days
Implement cfg80211_chandef_create() to work with HT. Update enum
with HT channel types. When calling the function from LinuxKPI 802.11
code, pass in NL80211_CHAN_HT20 if HT is supported rather than NO_HT.
Sponsored by: The FreeBSD Foundation
MFC after: 3 days
In the current test case unicast_v4, host one acts as a router, but it
uses if_bridge(4) to verify the reachability to CARP master.
Unfortunately, if_bridge(4) brings the two CARP hosts into the same
broadcast domain, thus with an unusual (undocumented) setup, one host
is in unicast mode but another one in multicast, or two hosts are both
in multicast mode, the test case can falsely report success.
Fix that by propagating route via routed(8).
PR 284872 shows that CARP(4) happens to send packets with multicast
destination when the unicast peer address is in the range x.x.x.224 ~
x.x.x.239. Update the peer address in test case to catch that bug.
PR: 284872
Reviewed by: kp
MFC after: 3 days
Differential Revision: https://reviews.freebsd.org/D49076
UDP allows to sendto(2) on unconnected socket. The original BSD devise
was that such action would create a temporary (for the duration of the
syscall) connection between our inpcb and remote addr:port specified in
sockaddr 'to' of the syscall. This devise was broken in 2002 in
90162a4e87. For more motivation on the removal of the temporary
connection see email [1].
Since the removal of the true temporary connection the sendto(2) on
unconnected socket has the following side effects:
1) After first sendto(2) the "unconnected" socket will receive datagrams
destined to the selected port.
2) All subsequent sendto(2) calls will use the same source port.
Effectively, such sendto(2) acts like a bind(2) to INADDR_ANY:0. Indeed,
if you do this:
s1 = socket(PF_INET, SOCK_DGRAM, 0);
s2 = socket(PF_INET, SOCK_DGRAM, 0);
sendto(s1, ..., &somedestination, ...);
bind(s2, &{ .sin_addr = INADDR_ANY, sin_port = 0 });
And then look into kgdb at resulting inpcbs, you would find them equal in
all means modulo bound to different anonymous ports.
What is even more interesting is that Linux kernel had picked up same
behavior, including that "unconnected" socket will receive datagrams. So
it seems that such behavior is now an undocumented standard, thus I
covered it in recently added tests/sys/netinet/udp_bindings.
Now, with the above knowledge at hand, why are we using
in_pcbconnect_setup() and in_pcbinshash(), which are supposed to be
private to in_pcb.c, to achieve the binding? Let's use public KPI
in_pcbbind() on the first sendto(2) and use in_pcbladdr() on all
sendto(2)s. Apart from finally hiding these two should be private
functions, we no longer acquire global INP_HASH_WLOCK() for every
sendto(2) on unconnected socket as well as remove a couple workarounds.
[1] https://mail-archive.FreeBSD.org/cgi/mid.cgi?200210141935.aa83883
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D49043
1) Set SO_BROADCAST and send to INADDR_BROADCAST.
2) Set SO_BROADCAST, set IP_ONESBCAST and cycle through broadcast capable
interfaces sending a datagram on each.
3) Set SO_BROADCAST and send to local broadcast address.
For each test we bind a second socket to check whether the broadcast
is actually received.
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D49040
When a socket has SO_BROADCAST set and destination address is INADDR_ANY
or INADDR_BROADCAST, the kernel shall pick up first broadcast capable
interface and broadcast the packet out of it. Since this API is not
reliable on a machine with > 1 broadcast capable interfaces, all practical
software seems to use IP_ONESBCAST or other mechanisms to send broadcasts.
This has been broken at least since FreeBSD 6.0, see bug 99558. Back then
the problem was in the fact that in_broadcast() check was always done
against the gateway address, not the destination address. Later, with
90cc51a1ab, a second problem piled on top - we aren't checking for
INADDR_ANY and INADDR_BROADCAST at all.
Better late than never, fix that by checking destination address.
PR: 99558
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D49042
This aligns with existing in_ifaddr_broadcast() and aligns with other
simple functions or macros with bare "in_" prefix that operator just on
struct in_addr and nothing else, e.g. in_nullhost(). No functional
change.
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D49041
If the lock is unowned (i.e., owner == UMUTEX_CONTESTED), we might get a
spurious failure, and in that case we need to retry the loop.
Otherwise, the calling thread can end up sleeping forever.
The same problem exists in do_set_ceiling(), which open-codes
do_lock_pp(), so fix it there too.
Reviewed by: olce
Reported by: Daniel King <dmking@adacore.com>
MFC after: 2 weeks
Sponsored by: Innovate UK
Differential Revision: https://reviews.freebsd.org/D49031
Two different functions in different files do the same thing - fill a
partial page with zeroes. Add that functionality to vm_page.c and
remove it elsewhere to avoid code duplication.
Reviewed by: markj, kib
Differential Revision: https://reviews.freebsd.org/D49096
Create empty hints files for these platforms. They don't normally need a
hints file, but people use them for device instance wiring. It's less
confusing if they always exist.
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D49052
Micron and Samsung also use this standard, though Micron is known to use
a few different keys. Add the ones that public information say are the
same among those the drives that have them.
There's others, but either they aren't publicly documented in datasheets
or nvme-cli code, or they vary betwen models and our code needs a fair
amount of rework needed.
Also print hex value for unknown keys. This allows scripts to more
easily pick bytes out for those keys who have values that are encoded by
packing multiple values into the 48-bit field.
Sponsored by: Netflix
Two different functions in different files do the same thing - fill a
partial page with zeroes. Add that functionality to vm_page.c and
remove it elsewhere to avoid code duplication.
Reviewed by: kib
Differential Revision: https://reviews.freebsd.org/D49096
We already pass struct pf_pdesc to pf_icmp_state_lookup(). There's no need to
also pass the direction.
Sponsored by: Rubicon Communications, LLC ("Netgate")
In most cases, IP fragments do not have an Ethernet padding. So
add a condition to save a useless call to m_adj() and have a paranoid
length check in the other cases.
OK henning@
Obtained from: OpenBSD, bluhm <bluhm@openbsd.org>, fcf0d61153
Obtained from: OpenBSD, chris <chris@openbsd.org>, ebe64b684c
Sponsored by: Rubicon Communications, LLC ("Netgate")
We now support multiple lables on a rule, so 'rule' is an array of strings, not
just one string. Adjust the check for 'is there a label?'.
CID: 1471808
Sponsored by: Rubicon Communications, LLC ("Netgate")
On most platforms time_t is 64 bits wide, but we assign it to 32 bits
variables. Unfortunately these are SNMP fields that are defined to be 32 bits
wide, so we cannot change this. We are similarly unable to prevent the passage
of time, meaning that time_t must remain 64 bits wide as well.
Explicitly cast to tell Coverity that we're aware of these limitations and do
not wish to be reminded of them.
CID: 1553869
CID: 1557025
CID: 1592232
CID: 1592234
Sponsored by: Rubicon Communications, LLC ("Netgate")
after some discussion with henning, document the various log options as
one section; some text was altered to make it read better;
ok henning
Obtained from: OpenBSD, jmc <jmc@openbsd.org>, cea6f9db1f
Sponsored by: Rubicon Communications, LLC ("Netgate")
Also, clean up some usage of predefined strings (which are discouraged by
mandoc_char(7) for portability reasons) and improve spacing in
hostapd.conf(5).
ok schwarze@
Obtained from: OpenBSD, bentley <bentley@openbsd.org>, e369c2e695
Sponsored by: Rubicon Communications, LLC ("Netgate")
change log(matches) semantics slightly to make it more useful. since it
is a debug tool change of semantics not considered problematic.
up until now, log(matches) forced logging on subsequent matching rules,
the actual logging used the log settings from that matched rule.
now, log(matches) causes subsequent matches to be logged with the log settings
from the log(matches) rule. in particular (this was the driving point),
log(matches, to pflog23) allows you to have the trace log going to a seperate
pflog interface, not clobbering your regular pflogs, actually not affecting
them at all.
long conversation with bluhm about it, which didn't lead to a single bit
changed in the diff but was very very helpful. ok bluhm as well.
Obtained from: OpenBSD, henning <henning@openbsd.org>, f61b1efcce
Sponsored by: Rubicon Communications, LLC ("Netgate")
This adds the ESS EDMA driver introduced by the IPQ4018/IPQ4019.
It provides a number of transmit and receive rings which can be mapped
into virtual ethernet devices, which this driver supports.
It's partially integrated into the ar40xx etherswitch which supplies
the port and some filtering/VPN offload functionality. This driver
only currently supports the per-port options which allow for the
virtual ethernet driver mapping.
This was written by reverse engineering the functionality of the
ethernet switch and ethernet driver support provided by Qualcomm
Atheros via their OpenWRT contributions. The code is all originally
authored by myself.
Differential Revision: https://reviews.freebsd.org/D49027
Our class-based login restrictions patch was merged upstream in commit
c276672fc0e9 ("Class-imposed login restrictions").
Sponsored by: The FreeBSD Foundation
Commit 9d63429fa1 ("ssh: move common Makefile boilerplate to a new
ssh.mk") introduced ssh.mk for common OpenSSH paths and flags, as part
of enabling FIDO/U2F. Move duplicated MK_LDNS and MK_TCP_WRAPPERS
handling there.
Reviewed by: kevans
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D31896
Centralize optional krb5_config.h handling in ssh.mk. Do not add
headers (that are committed to the src tree) to SRCS as there is no
need.
Reviewed by: imp, jlduran, kevans (all earlier)
MFC after: 1 month
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D34409
After fixing the obvious build and cast error when moving this from tools
to usr.sbin and removing -Wno-cast-align, re-add it back as at least
powerpc* does not compile without it (amd64, arm64, and i386 were ok).
In general wlanstats may now get a bit more love after 22 years of
being in the tree and being moved at least twice.
But first unbreak things again.
Reported by: mmel, cperciva (weekly snapshots)
Sponsored by: The FreeBSD Foundation
Fixes: 61ce422531
MFC after: 3 days
These libefivar files correspond to the following EDK2 Tianocore
routines:
efivar-dp-format.c -> MdePkg/Library/UefiDevicePathLib/DevicePathToText.c
efivar-dp-parse.c -> MdePkg/Library/UefiDevicePathLib/DevicePathFromText.c
The following changes were apparently applied upstream to appease their
CodeQL workflows on GitHub. Note that some of these changes are
logically contradictory, however avoiding differences between the two
code bases is a priority, as noted in the review.
Commit message from upstream:
Includes changes across the repo for the following CodeQL rules:
- cpp/comparison-with-wider-type
- cpp/overflow-buffer
- cpp/redundant-null-check-param
- cpp/uselesstest
Reviewed by: imp
Approved by: emaste (mentor)
Obtained from: b7735a087a
Differential Revision: https://reviews.freebsd.org/D49059
Add a 'g' flag to allow the replacement to occur more than once on the
same line.
Reviewed by: imp, emaste
Approved by: emaste (mentor)
Differential Revision: https://reviews.freebsd.org/D49058
Some systems have very slow console output and it may be about either
wrong memory attributes are set or gop->Blt() implementation is bad.
We do not have good way to set memory attributes, but we can
choose which Blt() to use (or we can set "gop off" to fall back on
use of SimpleTextOutput protocol).
This update adds argument for "gop" command to switch gop->Blt() use.
Note, this update does not fix the problem, but allows us to try to
understand the possible cause.
PR: 254381
Reported by: Michael Galassi
Reviewed by: manu, imp
Differential Revision: https://reviews.freebsd.org/D49073
Comply with style(9) and andd checks for booleaness when doing
bit tests.
If there is no need for double negated checks simplify them.
This all makes the conditions a lot easier to read.
Slip in a comment about MIC vs. MMIC.
No functional changes.
Sponsored by: The FreeBSD Foundation
MFC after: 3 days
Reviewed by: emaste, adrian
Differential Revision: https://reviews.freebsd.org/D49055
Notable upstream pull request merges:
#16857387ed5ca4 Add recursive dataset mounting and unmounting support
to pam_zfs_key
#16929c2458ba92 optimize recv_fix_encryption_hierarchy()
#1698012f0baf34 Make the vfs.zfs.vdev.raidz_impl sysctl cross-platform
#1698640496514b Expand fragmentation table to reflect larger possibile
allocation sizes
#1700388020b993 Add kstats tracking gang allocations
#1701321205f648 Avoid ARC buffer transfrom operations in prefetch
#17016390f6c119 zio: lock parent zios when updating wait counts on
reexecute
#17029b8c73ab78 zio: do no-op injections just before handing off to vdevs
#170376a2f7b384 Fix metaslab group fragmentation math
#17040b901d4a0b Update the dataset name in handle after zfs_rename
Obtained from: OpenZFS
OpenZFS commit: 6a2f7b3844
During my progress on updating cc_cubic to RFC9438, found such redundancy as:
- W_est: we use the alternative stack local variable `W_est` in
`cubic_ack_received()`.
- cwnd_prior: it is used for Reno-Friendly Region in RFC9438 Section 4.3,
but we use the alternative cwnd from NewReno for Reno-Friendly as
in commit ee45061051.
No functional change intended.
Reviewed by: rscheff, tuexen
Differential Revision: https://reviews.freebsd.org/D49008
In particular, export a "port" entry as well as an array of "host"
entries for each active connection.
Reviewed by: asomers
Sponsored by: Chelsio Communications
Differential Revision: https://reviews.freebsd.org/D48775
