1 Release notes for FreeBSD 13.0.
3 This file describes new user-visible features, changes and updates relevant to
4 users of binary FreeBSD releases. Each entry should describe the change in no
5 more than several sentences and should reference manual pages where an
6 interested user can find more information. Entries should wrap after 80
7 columns. Each entry should begin with one or more commit IDs on one line,
8 specified as a comma separated list and/or range, followed by a colon and a
9 newline. Entries should be separated by a newline.
11 Changes to this file should not be MFCed.
14 Fixes the case where gssd will not startup because /usr is a separate
15 local file system that is not yet mounted. It does not fix the case
16 where /usr is a separately mounted remote file system (such as NFS).
17 This latter case can be fixed by adding mountcritremote to the
18 REQUIRED line. Unfortunately doing so implies that all Kerberized
19 NFS mounts in /etc/fstab will need the "late" mount option.
20 This was not done, since the requirement for "late" would introduce
24 This commit added a new startup scripts variable called
25 nfsv4_server_only which uses the -R option on mountd added by r367026.
26 When nfsv4_server_only is set to "YES" in /etc/rc.conf, the NFS server
27 only handles NFSv4 and does not register with rpcbind. As such, rpcbind
28 does not need to be running. Useful for sites which consider rpcbind a
32 Kernel option ACPI_DMAR was renamed to IOMMU. amd64's IOMMU subsystem
33 was split out from amd64 DMAR support and is now generic, i.e., it can
34 be used by all architectures.
37 A series of commits ending with r364896 added NFS over TLS
38 to the kernel. This is believed to be compatible with
39 the Internet Draft titled "Towards Remote Procedure Call Encryption
40 By Default" (expected to soon become an RFC).
41 The mount_nfs(8) and exports(5) man pages describe the mount and
42 export option(s) related to NFS over TLS.
43 For NFS over TLS to work, the rpctlscd(8) { client } or rpctlssd(8)
44 { server } must be running on a kernel built with "options KERN_TLS"
45 on an architecture where PMAP_HAS_DMAP != 0.
48 Changes to one obscure devd event generated on resume need to
49 be documented. The old form will still be generated in 13, but not
53 Applications using regex(3), e.g. sed/grep, will no longer accept
54 redundant escapes for most ordinary characters.
57 SCTP support has been removed from GENERIC kernel configurations.
58 The SCTP stack is now built as sctp.ko and can be dynamically loaded.
61 Merge sendmail 8.16.1: See contrib/sendmail/RELEASE_NOTES for details.
64 The safexcel(4) crypto offload driver has been added.
67 nc(1) now implements SCTP mode, enabled by specifying the --sctp option.
70 A new implementation of bc and dc has been imported. It offers
71 better standards compliance, performance, localization and comes
72 with extensive test cases that are optionally installed.
73 Use WITHOUT_GH_BC=yes to build and install the world with the
74 previous version instead of the new one, if required.
77 struct export_args has changed so that the "user" specified for
78 the -maproot and -mapall exports(5) options may be in more than
82 sed(1) has learned about hex escapes (e.g. \x27) and will now do the
83 right thing with them, removing the need for printf magic or obnoxious
84 escaping in many scenarios.
86 r361238, r361798, r361799:
87 ZFS will now unconditionally reject read(2) of a directory with EISDIR.
88 Additionally, read(2) of a directory is now rejected with EISDIR by
89 default and may be re-enabled for non-ZFS filesystems that allow it with
90 the sysctl(8) MIB 'security.bsd.allow_read_dir'.
92 Aliases for grep to default to '-d skip' may be desired if commonly
93 non-recursively grepping a list that includes directories and the
94 possibility of EISDIR errors in stderr is not tolerable. Example
95 aliases, commented out, have been installed in /root/.cshrc and
99 Add exec.prepare and exec.release hooks for jail(8) and jail.conf(5).
100 exec.prepare runs before mounts, so can be used to populate new jails.
101 exec.release runs after unmounts, so can be used to remove ephemeral
104 r360920,r360923,r360924,r360927,r360928,r360931,r360933,r360936:
105 Remove support for ARC4, Blowfish, Cast, DES, Triple DES, MD5,
106 MD5-KPDK, MD5-HMAC, SHA1-KPDK, and Skipjack algorithms from
107 the kernel open cryptographic framework (OCF).
110 Remove support for ARC4, Blowfish, Cast, DES, Triple DES,
111 MD5-HMAC, and Skipjack algorithms from /dev/crypto.
114 Remove support for DES, Triple DES, Blowfish, Cast, and
115 Camellia ciphers from IPsec(4). Remove support for MD5-HMAC,
116 Keyed MD5, Keyed SHA1, and RIPEMD160-HMAC from IPsec(4).
119 Remove support for Triple DES, Blowfish, and MD5 HMAC from
123 Remove support for DES, Triple DES, and RC4 from in-kernel GSS
130 init(8), service(8), and cron(8) will now adopt user/class environment
131 variables (excluding PATH, by default, which will be overwritten) by
132 default. Notably, environment variables for all cron jobs and rc
133 services can now be set via login.conf(5).
136 sparc64 has been removed from FreeBSD.
139 Adds support for NFSv4.2 (RFC-7862) and Extended Attributes
140 (RFC-8276) to the NFS client and server.
141 NFSv4.2 is comprised of several optional features that can be supported
142 in addition to NFSv4.1. This patch adds the following optional features:
143 - posix_fadvise(POSIX_FADV_WILLNEED/POSIX_FADV_DONTNEED)
145 - intra server file range copying via the copy_file_range(2) syscall
146 --> Avoiding data tranfer over the wire to/from the NFS client.
147 - lseek(SEEK_DATA/SEEK_HOLE)
148 - Extended attribute syscalls for "user" namespace attributes as defined
151 For the client, NFSv4.2 is only used if the mount command line option
152 minorversion=2 is specified.
153 For the server, two new sysctls called vfs.nfsd.server_min_minorversion4
154 and vfs.nfsd.server_max_minorversion4 have been added that allow
155 sysadmins to limit the minor versions of NFSv4 supported by the nfsd
157 Setting vfs.nfsd.server_max_minorversion4 to 0 or 1 will disable NFSv4.2
161 armv5 support has been removed from FreeBSD.
164 iwm(4) now supports most Intel 9260, 9460 and 9560 Wi-Fi devices.
167 sqlite3 is updated to sqlite3-3.30.1.
170 cron(8) now supports the -n (suppress mail on succesful run) and -q
171 (suppress logging of command execution) options in the crontab format.
172 See the crontab(5) manpage for details.
175 ntpd is no longer by default locked in memory. rlimit memlock 32
176 or rlimit memlock 0 can be used to restore this behaviour.
179 rc.subr(8) now honors ${name}_env in all rc(8) scripts. Previously,
180 environment variables set by a user via ${name}_env were ignored
181 if the service defined a custom *_cmd variable to control the behavior
182 of the run_rc_command function, e.g., start_cmd, instead of relying on
183 the variables like command and command_args,
185 r351770,r352920,r352922,r352923:
186 dd(1) now supports conv=fsync, conv=fdatasync, oflag=fsync, oflag=sync,
187 and iflag=fullblock flags, compatible with illumos and GNU.
190 Add kernel-side support for in-kernel Transport Layer Security
191 (KTLS). KTLS permits using sendfile(2) over sockets using
195 WPA is updated from 2.8 to 2.9.
198 Add probes for lockmgr(9) to the lockstat DTrace provider, add
199 corresponding lockstat(1) events, and document the new probes in
203 Intel RST is a new 'feature' that remaps NVMe devices from
204 their normal location to part of the AHCI bar space. This
205 will eliminate the need to set the BIOS SATA setting from RST
206 to AHCI causing the nvme drive to be erased before FreeBSD
207 will see the nvme drive. FreeBSD will now be able to see the
208 nvme drive now in the default config.
211 Add a vop_stdioctl() call, so that file systems that do not support
212 holes will have a trivial implementation of lseek(SEEK_DATA/SEEK_HOLE).
213 The algorithm appears to be compatible with the POSIX draft and
214 the implementation in Linux for the case of a file system that
215 does not support holes. Prior to this patch, lseek(2) would reply
216 -1 with errno set to ENOTTY for SEEK_DATA/SEEK_HOLE on files in
217 file systems that do not support holes.
218 r351372 maps ENOTTY to EINVAL for lseek(SEEK_DATA/SEEK_HOLE) for
219 any other cases, such as a ENOTTY return from vn_bmap_seekhole().
222 The fuse driver has been renamed to fusefs(5) and been substantially
223 rewritten. The new driver includes many bug fixes and performance
224 enhancements, as well as the following user-visible features:
225 * Optional kernel-side permissions checks (-o default_permissions)
226 * mknod(2), socket(2), and pipe(2) support
227 * server side locking with fcntl(2)
228 * FUSE operations are now interruptible when mounted with -o intr
229 * server side handling of UTIME_NOW during utimensat(2)
230 * mount options may be updated with "mount -u"
231 * fusefs file system may now be exported over NFS
232 * RLIMIT_FSIZE support
233 * support for fuse file systems using protocols as old as 7.4
235 FUSE file system developers should also take note of the following new
237 * The protocol level has been raised from 7.8 to 7.23
238 * kqueue support on /dev/fuse
239 * server-initiated cache invalidation via FUSE_NOTIFY_REPLY
242 gnop(8) can now configure a delay to be applied to read and write
243 request delays. See the -d, -q and -x parameters.
246 Adds a Linux compatible copy_file_range(2) syscall.
249 libcap_random(3) has been removed. Applications can use native
250 APIs to get random data in capability mode.
253 Add support for using unmapped mbufs with sendfile(2).
256 nand(4) and related components have been removed.
259 The UEFI loader now supports HTTP boot.
262 bhyve(8) now implements a High Definition Audio (HDA) driver, allowing
263 guests to play to and record audio data from the host.
266 swapon(8) can now erase a swap device immediately before enabling it,
267 similar to newfs(8)'s -E option. This behaviour can be specified by
268 adding -E to swapon(8)'s command-line parameters, or by adding the
269 "trimonce" option to a swap device's /etc/fstab entry.
272 The following network drivers have been removed: bm(4), cs(4), de(4),
273 ed(4), ep(4), ex(4), fe(4), pcn(4), sf(4), sn(4), tl(4), tx(4), txp(4),
277 Wired page accounting has been split into kernel wirings and user
278 wirings (e.g., by mlock(2)). Kernel wirings no long count towards
279 the global limit, which is renamed to vm.max_user_wired. bhyve -S
280 allocates user-wired memory and is now subject to that limit.