PT_LWPINFO from FreeBSD is almost never intended to be expressed with
PT_LWPINFO in NetBSD. PT_GET_SIGINFO reads siginfo_t with the signal
information about the event, on FreeBSD siginfo_t is merged into
ptrace_lwpinfo and returns the thread that received the event (not the
first one in a list like on NetBSD).
Since binutils 2.15, nm(1) cannot be used for character devices.
We worked around this by a local patch:
http://cvsweb.netbsd.org/bsdweb.cgi/src/gnu/dist/binutils/binutils/Attic/bucomm.c?r1=1.1.1.2&hideattic=0#rev1.2
With recent update of binutils, 'nm /dev/ksyms' got broken again.
This is due to a consistency check involving file size reported by
stat(2), which is always zero for character devices. So, skip this
check if file size is zero.
--- 9.14.8 released ---
5315. [bug] Apply the inital RRSIG expiration spread fixed
to all dynamically created records in the zone
including NSEC3. Also fix the signature clusters
when the server has been offline for prolonged
period of times. [GL #1256]
5314. [func] Added a new statistics variable "tcp-highwater"
that reports the maximum number of simultaneous TCP
clients BIND has handled while running. [GL #1206]
5313. [bug] The default GeoIP2 database location did not match
the ARM. 'named -V' now reports the default
location. [GL #1301]
5310. [bug] TCP failures were affecting EDNS statistics. [GL #1059]
5308. [bug] Don't log DNS_R_UNCHANGED from sync_secure_journal()
at ERROR level in receive_secure_serial(). [GL #1288]
5307. [bug] Fix hang when named-compilezone output is sent to pipe.
Thanks to Tony Finch. [GL !2481]
5306. [security] Set a limit on the number of concurrently served
pipelined TCP queries. (CVE-2019-6477) [GL #1264]
5305. [bug] NSEC Aggressive Cache ("synth-from-dnssec") has been
disabled by default because it was found to have
a significant performance impact on the recursive
service. [GL #1265]
5304. [bug] "dnskey-sig-validity 0;" was not being accepted.
[GL #876]
5302. [bug] Fix checking that "dnstap-output" is defined when
"dnstap" is specified in a view. [GL #1281]
5301. [bug] Detect partial prefixes / incomplete IPv4 address in
acls. [GL #1143]
- port lsan_allocator.h to riscv and ia64.
- remove configure output garbage from ia64/defs.mk
- update README.gcc8 to reality:
- arms mostly work, but not quite
- most ports now build
- some ports have switched
* hooks: STOPPED is now run on timeout and exit
* BSD: Use IP_REVCIF rather than IN_PKTINFO
* DHCP: When rebinding, ensure we have a DHCP ARP state
* RA: Sort routers when reachability changes
* RA: Apply hoplimit, reachable and retrans timer values to kernel
* RA: Warn if advertised MTU > interface MTU
* dhcpcd: Report SSID connection to when we gain carrier
* DHCP: Fix corruption of address flags when renewing
It is plainly obvious that the init value cannot be used (the
var was never used uninit'd - could not be) but gcc apparently cannot
work that out. Revert this if we ever get a compiler with a brain.
* Fix bugs in select-pane and the main-horizontal and main-vertical layouts.
CHANGES FROM 2.8 to 2.9
* Attempt to preserve horizontal cursor position as well as vertical with
reflow.
* Rewrite main-vertical and horizontal and change layouts to better handle the
case where all panes won't fit into the window size, reduce problems with
pane border status lines and fix other bugs mostly found by Thomas Sattler.
* Add format variables for the default formats in the various modes
(tree_mode_format and so on) and add a -a flag to display-message to list
variables with values.
* Add a -v flag to display-message to show verbose messages as the format is
parsed, this allows formats to be debugged
* Add support for HPA (\033[`).
* Add support for origin mode (\033[?6h).
* No longer clear history on RIS.
* Extend the #[] style syntax and use that together with previous format
changes to allow the status line to be entirely configured with a single
option.
Now that it is possible to configure their content, enable the existing code
that lets the status line be multiple lines in height. The status option can
now take a value of 2, 3, 4 or 5 (as well as the previous on or off) to
configure more than one line. The new status-format array option configures
the format of each line, the default just references the existing status-*
options, although some of the more obscure status options may be eliminated
in time.
Additions to the #[] syntax are: "align" to specify alignment (left, centre,
right), "list" for the window list and "range" to configure ranges of text
for the mouse bindings.
The "align" keyword can also be used to specify alignment of entries in tree
mode and the pane status lines.
* Add E: and T: format modifiers to expand a format twice (useful to expand the
value of an option).
* The individual -fg, -bg and -attr options have been removed; they
were superseded by -style options in tmux 1.9.
* Allow more than one mode to be opened in a pane. Modes are kept on a stack
and retrieved if the same mode is entered again. Exiting the active mode goes
back to the previous one.
* When showing command output in copy mode, call it view mode instead (affects
pane_mode format).
* Add -b to display-panes like run-shell.
* Handle UTF-8 in word-separators option.
* New "terminal" colour allowing options to use the terminal default colour
rather than inheriting the default from a parent option.
* Do not move the cursor in copy mode when the mouse wheel is used.
* Use the same working directory rules for jobs as new windows rather than
always starting in the user's home.
* Allow panes to be one line or column in size.
* Go to last line when goto-line number is out of range in copy mode.
* Yank previously cut text if any with C-y in the command prompt, only use the
buffer if no text has been cut.
* Add q: format modifier to quote shell special characters.
* Add StatusLeft and StatusRight mouse locations (keys such as
MouseDown1StatusLeft) for the status-left and status-right areas of the
status line.
* Add -Z to find-window.
* Support for windows larger than the client. This adds two new options,
window-size and default-size, and a new command, resize-window. The
force-width and force-height options and the session_width and session_height
formats have been removed.
The new window-size option tells tmux how to work out the size of windows:
largest means it picks the size of the largest session, smallest the smallest
session (similar to the old behaviour) and manual means that it does not
automatically resize windows. aggressive-resize modifies the choice of
session for largest and smallest as it did before.
If a window is in a session attached to a client that is too small, only part
of the window is shown. tmux attempts to keep the cursor visible, so the part
of the window displayed is changed as the cursor moves (with a small delay,
to try and avoid excess redrawing when applications redraw status lines or
similar that are not currently visible).
Drawing windows which are larger than the client is not as efficient as those
which fit, particularly when the cursor moves, so it is recommended to avoid
using this on slow machines or networks (set window-size to smallest or
manual).
The resize-window command can be used to resize a window manually. If it is
used, the window-size option is automatically set to manual for the window
(undo this with "setw -u window-size"). resize-window works in a similar way
to resize-pane (-U -D -L -R -x -y flags) but also has -a and -A flags. -a
sets the window to the size of the smallest client (what it would be if
window-size was smallest) and -A the largest.
For the same behaviour as force-width or force-height, use resize-window -x
or -y.
If the global window-size option is set to manual, the default-size option is
used for new windows. If -x or -y is used with new-session, that sets the
default-size option for the new session.
The maximum size of a window is 10000x10000. But expect applications to
complain and higher memory use if making a window that big. The minimum size
is the size required for the current layout including borders.
The refresh-client command can be used to pan around a window, -U -D -L -R
moves up, down, left or right and -c returns to automatic cursor
tracking. The position is reset when the current window is changed.
notification of a failed auth attempt, and a notification of a
successful auth attempt.
Implements a third option - notification of abusive behavior, and
accepts, but does not act on a forth type - "bad username". It is
envisioned that a system administrator will configure a small list
of "known bad usernames" that should be blocked immediately.
From Kurt Lidl @ FreeBSD
they end up mis-ordering tm files.
revert the rs6000/netbsd64.h change and put it back as
SUBSUBTARGET_EXTRA_SPECS.
this seems to make the build work again, though mknative will
need a re-run for ppc*.
so that assembly specs get propagated (they were lost before)
- adjust the block register padding (like the other OS's) so that code compiles
- XXX: linker still broken for shared libraries because the os specific spec
is not chosen
PR port-alpha/54307.
If rtree.c and tcache.c are compiled with -O0, userland just works
without problems as far as I can see. Alternately, you can specify
-DJEMALLOC_DEBUG to avoid random crash. Smells like compiler bug,
or wrong coding which relies on some undefined behavior.
Anyway, we need to pull this up into netbsd-9 asap.
it -O0. The tm argument gets passed in as non-null and it is non-null after
the command execution, but then becomes null and causes a null dereference.
libsanitizer needs to match with the compiler concept of
TARGET_ASAN_SHADOW_OFFSET, since the compiler generates instructions
to access memory at that offset in the code, and expects the
sanitizer library to have mapped something there. If there is
disagreement, Mr. Segmentation Fault comes and resolves it for you.
Remove const from the 2nd argument.
const char ** and char ** are incompatible types and it was a cost to keep
the technically incompatible form for a more purist variation. NetBSD was
almost the last alive OS to still keep the const argument (known leftovers:
Minix and Illumos).
Keep the const form for the internal purposes inside citrus and rump.
Address the build breakage fallout in the same change.
There are no ABI changes.
Change accepted by core@.
5299. [security] A flaw in DNSSEC verification when transferring
mirror zones could allow data to be incorrectly
marked valid. (CVE-2019-6475) [GL #16P]
5298. [security] Named could assert if a forwarder returned a
referral, rather than resolving the query, when QNAME
minimization was enabled. (CVE-2019-6476) [GL #1051]
5297. [bug] Check whether a previous QNAME minimization fetch
is still running before starting a new one; return
SERVFAIL and log an error if so. [GL #1191]
5294. [func] Fallback to ACE name on output in locale, which does not
support converting it to unicode. [GL #846]
5293. [bug] On Windows, named crashed upon any attempt to fetch XML
statistics from it. [GL #1245]
5292. [bug] Queue 'rndc nsec3param' requests while signing inline
zone changes. [GL #1205]
--- 9.14.6 released ---
5289. [bug] Address NULL pointer dereference in rpz.c:rpz_detach.
[GL #1210]
5286. [contrib] Address potential NULL pointer dereferences in
dlz_mysqldyn_mod.c. [GL #1207]
5285. [port] win32: implement "-T maxudpXXX". [GL #837]
5283. [bug] When a response-policy zone expires, ensure that
its policies are removed from the RPZ summary
database. [GL #1146]
5282. [bug] Fixed a bug in searching for possible wildcard matches
for query names in the RPZ summary database. [GL #1146]
5281. [cleanup] Don't escape commas when reporting named's command
line. [GL #1189]
5280. [protocol] Add support for displaying EDNS option LLQ. [GL #1201]
5279. [bug] When loading, reject zones containing CDS or CDNSKEY
RRsets at the zone apex if they would cause DNSSEC
validation failures if published in the parent zone
as the DS RRset. [GL #1187]
have been converted to GCC_NO_FORMAT_TRUNCATION rather than
GCC_NO_STRINGOP_TRUNCATION which is what happened. This might unbreak
the build (olr at least get it further).
GCC_NO_FORMAT_TRUNCATION -Wno-format-truncation (GCC 7/8)
GCC_NO_STRINGOP_TRUNCATION -Wno-stringop-truncation (GCC 8)
GCC_NO_STRINGOP_OVERFLOW -Wno-stringop-overflow (GCC 8)
GCC_NO_CAST_FUNCTION_TYPE -Wno-cast-function-type (GCC 8)
use these to turn off warnings for most GCC-8 complaints. many
of these are false positives, most of the real bugs are already
commited, or are yet to come.
we plan to introduce versions of (some?) of these that use the
"-Wno-error=" form, which still displays the warnings but does
not make it an error, and all of the above will be re-considered
as either being "fix me" (warning still displayed) or "warning
is wrong."
to copy the interface name. The kernel requires the interface to be NUL
terminated anyway and will fail with ENAMETOOLONG otherwise. Pointed out
by Robert Swindells.
* Fix carrier status after a route socket overflow
* Allow domain spaced options
* DHCP: Allow not sending Force Renew Nonce or Reconf Accept
* IPv4LL: Now passes Apple Bonjour test versions 1.4 and 1.5
* ARP: Fix a typo and remove pragma (thus working with old gcc)
* DHCP6: Fix a cosmetic issue with infinite leases
* DHCP6: SLA 0 and Prefix Len 0 will now add a delegatd /64 address
* Ignore some virtual interfaces such as Tap and Bridge by default
* BPF: Move validation logic out of BPF and back into dhcpcd
Summary of changes in tzdata2019c (2019-09-11 08:59:48 -0700):
Fiji observes DST from 2019-11-10 to 2020-01-12
Norfolk Island starts observing Australian-style DST
Plus historic corrections to time in Turkey (1940-85)
South Korea (1948-51) Detroit (US) (1967-8), Perry County
(Indiana, US) (pre 1970) Edmonton (CA) (1967, 1969)
Vancouver (CA) (1946), Vienna (AT) (1946), Kaliningrad (1945-6).
Louisville (US) (1946-50). Brussles (BE) (1892).
Hong Kong Winter Time (1941) now listed as being "DST".
Summary of changes in tzdata2019b (2019-07-01 00:09:53 -0700):
Brazil no longer observes DST
Predictions for Morocco extended to 2087.
Panestine (March 2019) time zone change date corrected
(and guesses for future transitions revised).
Historic updates: Honk Kong (1941 - 1947), Italy (1866).
* test/btyacc/big_b.error, test/btyacc/big_l.error, test/btyacc/help.error, test/btyacc/no_b_opt.error, test/btyacc/no_output2.error, test/btyacc/no_p_opt.error, test/btyacc/nostdin.error, test/yacc/big_b.error, test/yacc/big_l.error, test/yacc/help.error, test/yacc/no_b_opt.error, test/yacc/no_output2.error, test/yacc/no_p_opt.error, test/yacc/nostdin.error:
regen
* test/run_test.sh: test "-H" rather than "-D"
2019-06-16 Thomas E. Dickey <dickey@invisible-island.net>
* main.c, yacc.1:
change "-D" option to "-H" (discussion with Ethan Sommer)
* VERSION, package/byacc.spec, package/debian/changelog, package/mingw-byacc.spec, package/pkgsrc/Makefile:
bump
* test/btyacc/defines1.calc.c, test/btyacc/defines2.calc.c, test/btyacc/defines3.calc.c:
RCS_BASE
* test/run_test.sh: in test_defines, save the ".c" file too
* test/yacc/defines3.calc.c, test/yacc/defines2.calc.c, test/yacc/defines1.calc.c:
RCS_BASE
* test/run_test.sh:
output of test_defines should be a header ".h", not ".c"
* test/btyacc/defines1.calc.h, test/btyacc/defines1.error, test/btyacc/defines1.output, test/btyacc/defines2.calc.h, test/btyacc/defines2.error, test/btyacc/defines2.output, test/btyacc/defines3.calc.h, test/btyacc/defines3.error, test/btyacc/defines3.output:
RCS_BASE
* main.c: however, a subsequent -d cancels -D
* test/yacc/defines1.calc.h, test/yacc/defines3.calc.h: RCS_BASE
* main.c: -D option implies -d
* test/yacc/defines1.error, test/yacc/defines1.output, test/yacc/defines2.calc.h, test/yacc/defines2.error, test/yacc/defines2.output, test/yacc/defines3.error, test/yacc/defines3.output:
RCS_BASE
* yacc.1: align macro definitions with my other manpages
* test/run_test.sh: add test for -D after -d or -b options
* test/btyacc/stdin1.calc.c, test/btyacc/stdin1.error, test/btyacc/stdin1.output, test/btyacc/stdin2.calc.c, test/btyacc/stdin2.error, test/btyacc/stdin2.output:
RCS_BASE
* test/btyacc/big_b.error, test/btyacc/big_b.output, test/btyacc/big_l.error, test/btyacc/big_l.output, test/btyacc/help.error, test/btyacc/help.output, test/btyacc/no_b_opt.error, test/btyacc/no_b_opt.output, test/btyacc/no_b_opt1.error, test/btyacc/no_b_opt1.output, test/btyacc/no_code_c.error, test/btyacc/no_code_c.output, test/btyacc/no_defines.error, test/btyacc/no_defines.output, test/btyacc/no_graph.error, test/btyacc/no_graph.output, test/btyacc/no_include.error, test/btyacc/no_include.output, test/btyacc/no_opts.error, test/btyacc/no_opts.output, test/btyacc/no_output.error, test/btyacc/no_output.output, test/btyacc/no_output1.error, test/btyacc/no_output1.output, test/btyacc/no_output2.error, test/btyacc/no_output2.output, test/btyacc/no_p_opt.error, test/btyacc/no_p_opt.output, test/btyacc/no_p_opt1.error, test/btyacc/no_p_opt1.output, test/btyacc/no_verbose.error, test/btyacc/no_verbose.output, test/btyacc/nostdin.error, test/btyacc/nostdin.output, test/yacc/big_b.error, test/yacc/big_b.output, test/yacc/big_l.error, test/yacc/big_l.output, test/yacc/help.error, test/yacc/help.output, test/yacc/no_b_opt.error, test/yacc/no_b_opt.output, test/yacc/no_b_opt1.error, test/yacc/no_b_opt1.output, test/yacc/no_code_c.error, test/yacc/no_code_c.output, test/yacc/no_defines.error, test/yacc/no_defines.output, test/yacc/no_graph.error, test/yacc/no_graph.output, test/yacc/no_include.error, test/yacc/no_include.output, test/yacc/no_opts.error, test/yacc/no_opts.output, test/yacc/no_output.error, test/yacc/no_output.output, test/yacc/no_output1.error, test/yacc/no_output1.output, test/yacc/no_output2.error, test/yacc/no_output2.output, test/yacc/no_p_opt.error, test/yacc/no_p_opt.output, test/yacc/no_p_opt1.error, test/yacc/no_p_opt1.output, test/yacc/no_verbose.error, test/yacc/no_verbose.output, test/yacc/nostdin.error, test/yacc/nostdin.output:
regen
* test/run_test.sh:
add a test for stdin "-" vs end-options "--", and correct a redirection
of stderr in test_flags
* test/yacc/stdin2.output, test/yacc/stdin2.calc.c, test/yacc/stdin1.calc.c, test/yacc/stdin1.error, test/yacc/stdin1.output, test/yacc/stdin2.error:
RCS_BASE
* test/btyacc/big_b.output, test/btyacc/big_l.output, test/btyacc/help.output, test/btyacc/no_b_opt.output, test/btyacc/no_output2.output, test/btyacc/no_p_opt.output, test/btyacc/nostdin.output, test/yacc/big_b.output, test/yacc/big_l.output, test/yacc/help.output, test/yacc/no_b_opt.output, test/yacc/no_output2.output, test/yacc/no_p_opt.output, test/yacc/nostdin.output:
regen
* main.c: add -D option, to specify filename vs y.tab.h for -d
* defs.h: add dflag2, for -D option
* yacc.1: document -D option
* config_h.in: updated with autoheader-252
* configure: regen
* package/debian/copyright: bump
* aclocal.m4: add CF_GETOPT_HEADER
* aclocal.m4: Improved autoconf macros:
+ CF_CC_ENV_FLAGS
putting preprocessor flags in CFLAGS also is a nuisance, which can be
addressed in the same way.
+ CF_GCC_WARNINGS
factor out workaround for XTSTRINGDEFINES as CF_CONST_X_STRING
+ CF_GNU_SOURCE
The check for _DEFAULT_SOURCE should apply to "recent" Cygwin (since early 2016),
and except for "NEWLIB" vs "GLIBC" in the test, acts the same if I pretend
that "newlib" is the GNU C library. Without this, the check falls through
to the _XOPEN_SOURCE test, which breaks the pseudoterminal checks for xterm.
+ CF_POSIX_C_SOURCE
add/use CF_POSIX_VISIBLE
+ CF_TRY_XOPEN_SOURCE
use CF_APPEND_TEXT
+ CF_WITH_MAN2HTML
use sed to work around non-POSIX tail utility
+ CF_XOPEN_SOURCE
use CF_APPEND_TEXT
add/use CF_POSIX_VISIBLE
* VERSION, package/byacc.spec, package/debian/changelog, package/mingw-byacc.spec, package/pkgsrc/Makefile:
bump
2019-06-10 Thomas E. Dickey <dickey@invisible-island.net>
* config.guess: 2019-06-10
2019-05-22 Thomas E. Dickey <dickey@invisible-island.net>
* config.sub: 2019-05-22
zfs_netbsd_{create,mknod,link,etc..} that call functions called
zfs_{create,mknod,link,etc..}. These later functions may return a
error code along with a *vpp that is NULL. This situation was not
handled by the zfs_netbsd_* functions and would result in a panic in a
number of cases. The simplest to trigger it was filling up a dataset
or pool resulting in a over quota condition. An attempt to create
another file, or directory at that point would panic.
- file has looks bogus maybe-uninitialized
- llvm triggers an attribute violation:
ScheduleDAGInstrs.cpp:1430:14: error: declaration of
'llvm::raw_ostream& llvm::operator<<(llvm::raw_ostream&, const llvm::ILPValue&)'
with attribute 'noinline' follows inline declaration [-Werror=attributes]
- ntp and pkg_install have obvious restrict violations, should be
fixed but i'm avoiding patching upstream code in this pass
- tftp has an array bounds that doesn't seem real issue
- sysinst's partman.c has major problem with passing the same
string as source and dest in snprintf, as a way to strcat
with formatting which trip restrict violations. non trivial
to fix so for now the warning is elided.
- Xext's XEVI.c has similar issue as partman.c
everyone and GCC 8 gets these warnings turned off for now:
-Wno-format-truncation
-Wno-stringop-overflow
-Wno-stringop-truncation
-Wno-cast-function-type
as they trip a large amount of code. most of them should be
investigated, but the few i looked at were not finding actually
real bugs, vs instances of poor coding, so skipping for now.
avoid passing the same pointer in multiple arguments for restrict
marked arguments:
- sigaction() wants separate in/out
- use memmove() not memcpy() for overlapping regions (this may fix
a real bug in nvi -- but it seems unlikely)
- select() wants separate read/write/except
- sigprocmask() wants separate set/oset
- netbsd/arm64 uses 64 byte malloc alignment
- make lsan compile on sparc*, mips*, ppc and arm64 again
- add missing sparc, alpha and i386 abi compat for struct __sanitizer_addrinfo
- avoid linux includes on arm64
- avoid multiply defined __ubsan_handle_cfi_bad_type when UBSAN_CAN_USE_CXXABI
isn't defined, and, undefine it
- bad_array_length.cc and bad_array_new.cc lose special build rules
- regen arm64 mknative
Synchromize the struct field format with other BSDs and Darwin.
No ABI change between older and newer struct form on the ports.
The change will require no changes to most C users during the transition
period as the header keeps a caller cast.
Discussed with core@ and there were no objections for this move.
- sanitizer_procmaps_netbsd.cc is obsolete.
- fix merge botches where upstrem has as slightly different version
is upstream than prior local
- libstdc++ default is now gnu++17
- ubsan needs UBSAN_CAN_USE_CXXABI set
- properly use $G_RTL_BASE_H not (empty) $RTL_BASE_H
- libbackend HH gains new generated insn-modes-inline.h, and read-md.c
gains the HH dep.
- update sanitizer makefiles
also, reduce diffs to upstream:
- remove or1k support
- re-order various lines to match upstream
- move regex map code into file-prefix-map.[ch]
XXX: our change for e500 has moved into 'powerpcspe' port, which has
XXX: been marked deprecated in GCC 8. this may affect what ppc ports
XXX: can update to GCC 8 easily, and we may need to add support for
XXX: 'powerpcspe' while we can.
Summary for 1.9.1 libpcap release
Mention pcap_get_required_select_timeout() in the main pcap man page
Fix pcap-usb-linux.c build on systems with musl
Fix assorted man page and other documentation issues
Plug assorted memory leaks
Documentation changes to use https:
Changes to how time stamp calculations are done
Lots of tweaks to make newer compilers happier and warning-free and
to fix instances of C undefined behavior
Warn if AC_PROG_CC_C99 can't enable C99 support
Rename pcap_set_protocol() to pcap_set_protocol_linux().
Align pcap_t private data on an 8-byte boundary.
Fix various error messages
Use 64-bit clean API in dag_findalldevs()
Fix cleaning up after some errors
Work around some ethtool ioctl bugs in newer Linux kernels (GitHub
issue #689)
Add backwards compatibility sections to some man pages (GitHub issue
#745)
Fix autotool configuration on AIX and macOS
Don't export bpf_filter_with_aux_data() or struct bpf_aux_data;
they're internal-only and subject to change
Fix pcapng block size checking
On macOS, don't build rpcapd or test programs any fatter than they
need to be
Fix reading of capture statistics for Linux USB
Fix packet size values for Linux USB packets (GitHub issue #808)
Check only VID in VLAN test in filterss (GitHub issue #461)
Fix pcap_list_datalinks on 802.11 devices on macOS
Fix overflows with very large snapshot length in pcap file
Improve parsing of rpcapd configuration file (GitHub issue #767)
Handle systems without strlcpy() or strlcat() better
Fix crashes and other errors with invalid filter expressions
Fix use of uninitialized file descriptor in remote capture
Fix some CMake issues
Fix some divide-by-zero issues with the filter compiler
Work around a GNU libc bug in pcap_nametonetaddr()
Add support for DLT_LINUX_SLL2
Fix handling of the packet-count argument for Myricom SNF devices
Fix --disable-rdma in configure script (GitHub issue #782)
Fix compilation of TurboCap support (GitHub issue #764)
Constify first argument to pcap_findalldevs_ex()
Fix a number of issues when running rpcapd as an inetd-style daemon
Fix CMake issues with D-Bus libraries
In rpcapd, clean up termination of a capture session
Redo remote capture protocol negotiation
In rpcapd, report the same error for "invalid user name" and
"invalid password", to make brute-forcing harder
For remote captures, add an error code for "the server requires TLS"
Fix pcap_dump_fopen() on Windows to avoid clashes between
{Win,N}Pcap and application C runtimes
Fix exporting of functions from Windows DLLs (GitHub issue #810)
Fix building as part of Npcap
Allow rpcapd to rebind more rapidly
Fix building shared libpcap library on midipix (midipix.org)
Fix hack to detect UTF-16LE adapter names on Windows not to go past
the end of the string
Fix handling of "wireless WAN" (mobile phone network modems) on
Windows with WinPcap/Npcap (GitHub issue #824)
Have pcap_dump_open_append() create the dump file if it doesn't
exists (GitHub issue #247)
Fix the maxmum snapshot length for DLT_USBPCAP
Use -fPIC when building for 64-bit SPARC on Linux (GitHub issue #837)
Fix CMake 64-bit library installation directory on some Linux
distributions
Boost the TPACKET_V3 timeout to the maximum if a timeout of 0 was
specified
Five CVE-2019-15161, CVE-2019-15162, CVE-2019-15163, CVE-2019-15164, CVE-2019-15165
Fixes for CVE-2018-16301, errors in pcapng reading.
PCAPNG reader applies some sanity checks before doing malloc().
- many optimisations improved: inter-procedural, profile-directed,
LTO, loops including user-controllable unroll support, and more.
- columns numbers added to line numbers in dwarf
- gcov extended significantly
- many sanitizer updates
- many new warning messages
- many better hints and more useful error messages
- minor ABI changes on x86-64 libstdc++, and some c++17 modes
- draft c++2a features
- better c++17 experimental support
- Armv8.4-A supported, better 8.2-A and 8.3-A support, including
32 bit arm port. cortex a-55, a-75 and a-55.a-75 combo support.
- in the GCC bugzilla, 8.1 shows 1149 bugs fixed, 8.2 shows 100, and
8.3 shows 158.
