atf/atf-c++/macros_test/detect_unused_tests as expected failures
when using versions of GCC where they are known to fail, with a
reference to PR toolchain/49187.
--
When decompressing with -N or -NT, strip any path from header name.
This uses the path of the compressed file combined with the name
from the header as the name of the decompressed output file. Any
path information in the header name is stripped. This avoids a
possible vulnerability where absolute or descending paths are put
in the gzip header.
--
as it is needed for the "includes" phase. Make it symlink all necessary
headers in obj to decouple it from "includes" itself, breaking the
dependency cycle. Move the do-x11 target between do-lib and do-build, so
that libraries can get the benefit of build_install (correct .WAIT
behavior) and everything else like "includes" uses the plain SUBDIR
entry in external/mit.
From Alan Modra: Wed, 3 Jul 2013 11:51:56 +0930: powerpc64 call stubs:
This brings powerpc64 ld in line with gold regarding calls allowed to
go via a plt call stub or toc-adjusting stub without a following nop.
A long time ago ld allowed tail calls, but this is wrong because we'll
return to the caller's caller with the wrong toc pointer. I fixed
that for shared libraries but allowed tail calls in an executable for
some reason. Probably just muddy thinking on my part, because there's
no difference between an executable and a shared library regarding the
need to restore the toc pointer. Perhaps it was because some testcase
failed, most likely due to the g++ issue I comment on in the patch.
From Alan Modra: 29 Aug 2014 01:22:29 +0000: PowerPC64 call lacks nop error:
Enough people accidentally or otherwise use -mcmodel=small then wonder
why a call that is local is behaving as if it is going via the PLT.
* elf64-ppc.c (ppc64_elf_relocate_section): Report a different
error for calls via a toc adjusting stub without a nop.
problems with compiling atomics for the riscv64 target with gcc. Remove
the now-unneeded workarounds for the 2 files in libasan that were formerly
failing to compile with the optimizer turned on.
- Fix a conditional
- Don't include bsd.lib.mk if we are not building libraries
(otherwise we end up with an empty library name that installed
bogus library files lib.a etc.)
- Allow signals while reading the user crontab file; doing "crontab -"
does not let you abort otherwise, and doing ^Z, kill %1 leaves turds
in /var/cron/tabs
Summary of changes in tzdata2014j (2014-11-10 17:37:11 -0800):
* Turks & Caicos' switch from US eastern time to UTC-4 year-round
did not occur on 2014-11-02 at 02:00.
* Many pre-1989 time stamps have been corrected for Asia/Seoul and
Asia/Pyongyang, based on sources for the Korean-language Wikipedia
entry for time in Korea.
* Some more zones have been turned into links, when they differed
from existing zones only for older (pre-1970) time stamps.
* Changes affecting commentary.
Summary of changes in tzdata2014i (2014-10-21 22:04:57 -0700):
* Pacific/Fiji will observe DST from 2014-11-02 02:00 to 2015-01-18 03:00.
* A new Zone Pacific/Bougainville, for the part of Papua New Guinea
that plans to switch from UTC+10 to UTC+11 on 2014-12-28 at 02:00.
* Since Belarus is not changing its clocks even though Moscow is,
the time zone abbreviation in Europe/Minsk is changing from FET
to its more-traditional value MSK on 2014-10-26 at 01:00.
* The new abbreviation IDT stands for the pre-1976 use of UT+8 in
Indochina, to distinguish it better from ICT (UT+7).
* Many time stamps have been corrected for Asia/Ho_Chi_Minh before 1976.
Asia/Ho_Chi_Minh has been added to zone1970.tab, to give tzselect
users in Vietnam two choices, since north and south Vietnam
disagreed after our 1970 cutoff.
* Asia/Phnom_Penh and Asia/Vientiane have been turned into links, as
they differed from existing zones only for older (pre-1970) time stamps.
* Changes affecting commentary.
** no user visible changes in this release
* version 2.5.38 released 2014-02-14
** internationalization
*** add sr translation from the translation project
*** update da, es, ko, nl, pt_BR, ro, ru, sv, tr, vi, zh_CN translations from th
e translation project
*** rename zh_tw to its proper zh_TW name
* Don't update when nothing has been deleted
* Backup resolv.conf to resolv.conf.bak when it doesn't have an openresolv
signature
Restore it when the new resolv.conf only has the openresolv signature
* Document prepend_search and prepend_nameservers
* Implement append_search and append_nameservers
* Implement replace and replace_sub to allow for keyword/value/replacement
* There is a new "stop-bits" option, which takes an argument of 1 or 2,
indicating the number of stop bits to use for async serial ports.
* Various bug fixes.
* Fixed a potential security issue in parsing option files (CVE-2014-3158).
* There is a new "stop-bits" option, which takes an argument of 1 or 2,
indicating the number of stop bits to use for async serial ports.
* Various bug fixes.
- ENHANCE: When executing a chain, require at least one service
function to succeed. This mitigates fail-open scenarios caused by
misconfigurations or missing modules.
- ENHANCE: Make sure to overwrite buffers which may have contained an
authentication token when they're no longer needed.
- BUGFIX: Under certain circumstances, specifying a non-existent
module (or misspelling the name of a module) in a policy could
result in a fail-open scenario. (CVE-2014-3879)
- FEATURE: Add a search path for modules. This was implemented in
Nummularia but inadvertently left out of the release notes.
- BUGFIX: The is_upper() predicate only accepted the letter A as an
upper-case character instead of the entire A-Z range. As a result,
service and module names containing upper-case letters other than A
would be rejected.
Summary of changes in tzdata2014i (2014-10-21 22:04:57 -0700):
* Pacific/Fiji will observe DST from 2014-11-02 02:00 to 2015-01-18 03:00.
Guess that future years will use a similar pattern.
* A new Zone Pacific/Bougainville, for the part of Papua New Guinea
that plans to switch from UTC+10 to UTC+11 on 2014-12-28 at 02:00.
* Since Belarus is not changing its clocks even though Moscow is,
the time zone abbreviation in Europe/Minsk is changing from FET
to its more-traditional value MSK on 2014-10-26 at 01:00.
* The new abbreviation IDT stands for the pre-1976 use of UT+8 in
Indochina, to distinguish it better from ICT (UT+7).
* Many time stamps have been corrected for Asia/Ho_Chi_Minh before 1976.
Asia/Ho_Chi_Minh has been added to zone1970.tab, since
north and south Vietnam disagreed after our 1970 cutoff.
* Asia/Phnom_Penh and Asia/Vientiane have been turned into links, as
they differed from existing zones only for pre-1970 time stamps.
* Changes affecting commentary.
* dnsmasq subscriber no longer moans if it hasn't written a pidfile
* Ensure that name_server_blacklist works for more than one option.
Thanks to Frederic Barthelery.
* unbound_insecure can disable DNSSEC for all domains processed.
* local_nameservers now defaults to
127.* 0.0.0.0 255.255.255.255 ::1
and is used instead of a hard coded list.
* Allow the disabling of resolvconf or optionally an individual
subscriber.
* Don't wait around trying to create a lock if we don't have
permission.
* resolv_conf_passthrough=NULL will update resolv.conf to match
only what is configured in resolvconf.conf and ignore any
interface configuration.
* Fix an unaligned access error on BeagleBone Black with FreeBSD.
Thanks to Guy Yur for the patch.
* Remove the fast loop trying to up an interface which does not
report carrier.
* Remove vis based encoding - instead validate against option type and
stop at invalid [1]
This removes all shell escaped encoding - dhcpcd will assume that IF
the --script option is a shell, it will quote variables correctly.
The stock dhcpcd-run-hooks does.
* dhcpcd -V now prints how the variables will be decoded.
* Changed some options in dhcpcd-definitions.conf to more sensible defaults.
* Don't daemonise on delegated address dad.
* Don't drop delegated reject route when forking.
* Fix IPv6 handling of link-local addresses on KAME stacks.
* Work on OpenBSD-5.6 without any special interface setup needed.
* Callout to handlecarrier when we don't have real carrier support and
rely on looking at IFF_UP and IFF_RUNNING.
This allows our hooks to know that dhcpcd thinks we have a carrier or not.
[1] DHCP option encodings defined in dhcpcd-definitions.conf
* domain (RFC3397)/dname (string) is strict domain name allowance
(ie, [alnum] with _- (but not at the start or end))
* string is now printable ascii (1-127) until invalid
* ascii is all ascii (1-127) until invalid
* raw is all chars (1-255) until NUL
* binhex is a hex representation of the option including embedded NULs
* ssid is still escpaed octal because it's expected to be human readable
AND can technically be all NUL
* everything else has strict option -> value encoding
Summary of changes in tzdata2014h (2014-09-25 18:59:03 -0700):
* America/Jamaica's 1974 spring-forward transition was Jan. 6, not Apr. 28.
* Shanks says Asia/Novokuznetsk switched from LMT (not "NMT") on 1924-05-01,
not 1920-01-06.
* Some more zones have been turned into links, when they differed
from existing zones only for older time stamps. As usual,
these changes affect UTC offsets in pre-1970 time stamps only.
Their old contents have been moved to the 'backzone' file.
The affected zones are: Africa/Blantyre, Africa/Bujumbura,
Africa/Gaborone, Africa/Harare, Africa/Kigali, Africa/Lubumbashi,
Africa/Lusaka, Africa/Maseru, and Africa/Mbabane.
* Changes affecting documentation and commentary.
* Detect removal of IPv6 routes
* Don't add link-local addresses to POINTOPOINT interfaces
* Don't discard expired DHCPv6 leases when dumping them
* If a DHCPv6 lease has no timers, expire it right away
* Report delegated addresses
* Call dhcpcd-run-hooks correctly when delegated prefixes already exist
* Fix a memory error when ia_* config exists but IPv6 is disabled
* Ensure servername and bootfile are safely exported
* Sanitise the following characters using svis(3) with VIS_CTYLE and
VIS_OCTAL:
| ^ & ; < > ( ) $ ` \ " ' <tab> <newline>
This allows a non buggy unvis(1) to decode it 100% and stays compatible
with how dhcpcd used to handle encoding on most platforms.
For systems that supply svis(3) there is a code reduction, for systems
that do not, a slight code increase. This change mitigates systems
affected by bash CVE-2014-6271 and CVE-2014-7169.
* Many bounds checking fixes from Tobias Stoeckmann
* Improve error when the authentication token cannot be found
* close the IPv4 specific UDP socket when done sending
* Implemented a write queue to the control sockets
* Only send interfaces to control sockets when in a BOUND state
* Add a sample controlgroup directive to dhcpcd.conf to make setup easier
* Add variables if_oneup and if_ipwaited so hook scripts know the overall
state of dhcpcd better
* Pass RC_SVCNAME from enviromment to hooks so that a service hook can
know it's name (may not be dhcpcd)
* Document every variable set for dhcpcd-run-hooks(8)
* Allow SSIDs with non printable characters to be used in ssid selection
in dhcpcd.conf
* Add an unprivileged control socket so that normal users can obtain
dhcpcd running state
Summary of changes in tzdata2014g (2014-08-28 12:31:23 -0700):
* Turks & Caicos is switching from US eastern time to UTC-4 year-round,
modeled as a switch from EST/EDT to AST on 2014-11-02 at 02:00.
* Time in Russia or the USSR before 1926 or so has been corrected by
a few seconds for several zones, and a few minutes for
Asia/Yekaterinburg.
* The Portuguese decree of 1911-05-26 took effect on 1912-01-01, not
on the date of the decree. This affects 1911 time stamps in
Africa/Bissau, Africa/Luanda, Atlantic/Azores, and Atlantic/Madeira.
* Lisbon's pre-1912 GMT offset was -0:36:45 (rounded from -0:36:44.68),
not -0:36:32.
* Asia/Dhaka ended DST on 2009-12-31 at 24:00, not 23:59.
* A new file 'backzone' contains data which may appeal to
connoisseurs of old time stamps, although it is out of scope for
the tz database, is often poorly sourced, and contains some data
that is known to be incorrect.
* Some more zones have been turned into links, when they differed
from existing zones only for older time stamps. As usual,
these changes affect UTC offsets in pre-1970 time stamps only.
Their old contents have been moved to the 'backzone' file.
* Changes affecting documentation and commentary.
its tailq. This required introducing a new flag in the TAGQ structure to
avoid reverting to poking under the hood of the queue.h API.
The concrete bug it solves is that using tags would make vi crash reliably
on exit.
Apply fix that got missed as this file got moved from dist/gcc to dist/libgcc:
Somehow emacs tries to unregister frame info for a symbol it did not register.
Don't abort for now. This worked before because the previous version of gcc
did not bother unregistering. Adding debugging printfs makes emacs core
dump, where other programs work fine.
XXX: pullup 7