(-O1 is fine). This is a new regression introduced in GCC9.
For everyone's safety, extend -O0 hack to dwarf2{expr,loc}.c for
all arm variants with GCC >= 8.
Summary of changes in tzdata2020b (2020-10-06 18:35:04 -0700):
Revised predictions for Morocco's changes starting in 2023.
Canada's Yukon changes to -07 on 2020-11-01, not 2020-03-08.
Macquarie Island has stayed in sync with Tasmania since 2011.
Casey, Antarctica is at +08 in winter and +11 in summer.
GCC 9.3 seems to be able to compile rtree.c with -O2:
- No new regressions in ATF.
- System survives over a night, at least, under heavy loads.
On the other hand, unfortunately, GCC 9.3 still miscompiles tcache.c
with -O2 or -O1. For example, even ``gcc -g hello.c'' fails with ICE
if tcache.c is compiled with -O[12] in libc.
Cherry-pick:
From e2a2a24a8e78427ff8667d625f5befbe88c328bb Mon Sep 17 00:00:00 2001
From: Kamil Rytarowski <n54@gmx.com>
Date: Wed, 23 Sep 2020 04:20:05 +0200
Subject: [PATCH] Preinitialize the sockaddr_un variable to zero
Don't pass random sun_len for the BSD's,
zero the whole structure as recommended for portability.
Reported by Coverity.
gdbsupport/ChangeLog:
* agent.cc (gdb_connect_sync_socket): Preinitialize addr with zeros.
* dhcpcd: Backticks have been removed from quoting filenames
* dhcpcd: Only manipulate stdin, stdout and stderr if they are valid
* duid: Adjust option so the type can be specified
* logerr: Don't leak logfile fd to scripts
* privsep: Run the launcher process in the sandbox
* BSD: Use `ifi_link_state` as the single source of truth about carrier
* BSD: Ignore vether(4) devices by default
fonts included as part of X11 that are also included as scalable fonts
as part of macOS.
Many websites will attempt to select these fonts, which results in
very strange, poor quality rendering.
This should keep legacy X11 applications that want the old "Helvetica"
and "Times" fonts working, but prevent them from being selected by
applications that want scalable fonts.
It should not prevent "Helvetica" or "Times" from being selected
if the user installs a scalable version.
aka "the memcmp bug", from a patch for gcc-10 in
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95189
Tested on x86_64 by running a full build, install and running it.
Testing that the problem has been fixed with one of the unit tests provided.
Cherry-pick and adapt:
From 2a9ce60de98e53198047daaeeec3cf09ece4e693 Mon Sep 17 00:00:00 2001
From: Kamil Rytarowski <n54@gmx.com>
Date: Fri, 2 Oct 2020 16:13:09 +0200
Subject: [PATCH] [compiler-rt] [netbsd] Improve the portability of
ThreadSelfTlsTcb
Use __lwp_gettcb_fast() and __lwp_getprivate_fast(), as _lwp_getprivate()
can be a biased pointer and invalid for use in this function on all CPUs.
- Bug fixes to gmp_snprintf, conversion to double, mpz_powm,
and mpf_set_str.
- New functions for factorial, primorial, fibonacci, mpz_2fac_ui,
and mpz_mfac_uiui.
- MIPS r6 cores are now supported.
- Various speeds ups.
Changed __float128 to the type _Float128 specified in ISO/IEC TS 18661.
__float128 is used as a fallback if _Float128 is not supported.
New function mpfr_get_str_ndigits about conversion to a string of digits.
New function mpfr_dot for the dot product (incomplete, experimental).
New functions mpfr_get_decimal128 and mpfr_set_decimal128 (available
only when MPFR has been built with decimal float support).
New function mpfr_cmpabs_ui.
New function mpfr_total_order_p for the IEEE 754 totalOrder predicate.
The mpfr_out_str function now accepts bases from -2 to -36, in order to
follow mpfr_get_str and GMP's mpf_out_str functions (these cases gave an
assertion failure, as with other invalid bases).
Shared caches: cleanup; really detect lock failures (abort in this case).
Improved mpfr_add and mpfr_sub when all operands have a precision
equal to twice the number of bits per word, e.g., 128 bits on a 64-bit
platform.
Optimized the tuning parameters for various architectures.
0's and svr4_exec_displacement read phdrs_target successfully, and we
end up in the rabbit-hole of 0 displacement for pie binaries and then we
can't load our shared libaries from the core file.
this is an optimization that primarily benefits linux/glibc -
most other systems have this disabled. in netbsd we've tried to
patch around it to make things work, but there still appears to be
some edge cases where libGL mysteriously crashes.
discussed on tech-x11 some time ago. already in place in pkgsrc.
2020-07-24 Aaron Merey <amerey@redhat.com>
* configure: Rebuild.
* configure.ac: Remove AC_DEBUGINFOD.
2020-07-04 Nick Clifton <nickc@redhat.com>
Binutils 2.35 branch created.
2020-04-21 Stephen Casner <casner@acm.org>
PR 25830
* configure.ac (noconfigdirs): Exclude gdb & gprof for pdp11.
* configure: Rebuild.
2020-03-12 Tom Tromey <tom@tromey.com>
* Makefile.in: Rebuild.
* Makefile.def (gdbserver): Depend on gdbsupport.
2020-03-12 Tom Tromey <tom@tromey.com>
* Makefile.in: Rebuild.
* Makefile.def (gdbsupport): Don't depend on bfd.
2020-03-12 Tom Tromey <tom@tromey.com>
* Makefile.in: Rebuild.
* Makefile.def (gdbsupport): Depend on intl.
2020-02-17 Tom Tromey <tom@tromey.com>
* configure: Rebuild.
* configure.ac (configdirs): Add gnulib and gdbsupport when building
gdbserver.
2020-02-14 Tom Tromey <tom@tromey.com>
* Makefile.in: Rebuild.
* Makefile.def: Make gdbserver require gnulib and libiberty.
2020-02-07 Tom Tromey <tom@tromey.com>
Pedro Alves <palves@redhat.com>
* src-release.sh (GDB_SUPPORT_DIRS): Add gdbserver.
* gdbserver: New directory, moved from gdb/gdbserver.
* configure.ac (host_tools): Add gdbserver.
Only build gdbserver on certain systems.
* Makefile.in, configure: Rebuild.
* Makefile.def (host_modules, dependencies): Add gdbserver.
* MAINTAINERS: Add gdbserver.
2020-01-28 Sergio Durigan Junior <sergiodj@redhat.com>
* src-release.sh (getver): Look for gdbsupport's
create-version.sh script at the current directory if tool is
"gdb".
2020-01-19 Simon Marchi <simon.marchi@polymtl.ca>
* remote-sim.c (gdbsim_target::wait): Return
sim_data->remote_sim_ptid instead of inferior_ptid.
2020-01-18 Nick Clifton <nickc@redhat.com>
Binutils 2.34 branch created.
2020-01-18 Nick Clifton <nickc@redhat.com>
Synchronize top level configure files with master version:
2020-01-01 Ben Elliston <bje@gnu.org>
* config.guess: Update copyright years.
* config.sub: Likewise.
2019-12-21 Ben Elliston <bje@gnu.org>
* config.guess (set_cc_for_build): Prevent multiple calls by
checking if $tmp is already set. We can't check CC_FOR_BUILD as
the user may set it externally. Thanks to Torbjörn Granlund for
the bug report.
2019-12-21 Torbjörn Granlund <tg@gmplib.org>
* config.guess (alpha:Linux:*:*): Guard against missing
/proc/cpuinfo by redirecting standard error to /dev/null.
2019-09-12 Daniel Bittman <danielbittman1@gmail.com>
* config.guess (*:Twizzler:*:*): New.
* config.sub (-twizzler*): New.
2019-07-24 Ben Elliston <bje@gnu.org>
* config.guess (mips:OSF1:*.*): Whitespace cleanup.
2019-06-30 Ben Elliston <bje@gnu.org>
* config.sub (case $os): Match nsk* and powerunix. Don't later
match nsk* and set os=nsk which removes the OS version number.
2019-06-30 Ben Elliston <bje@gnu.org>
* config.sub: Recognise os108*.
2019-06-26 Ben Elliston <bje@gnu.org>
* config.sub (hp300): Set $os to hpux.
2019-06-26 Ben Elliston <bje@gnu.org>
* config.sub (vsta): Move into alphabetical order.
2019-06-10 Ben Elliston <bje@gnu.org>
* config.guess (*:OS108:*:*): Recognise new OS.
2019-05-28 Ben Elliston <bje@gnu.org>
* config.guess (*:Darwin:*:*): Run xcode-select to determine if a
system compiler is installed. If not, do not run set_cc_for_build,
as the default cc will open a dialog box asking to install
Xcode. If no C compiler is available, guess based on uname -p and
uname -m.
2019-05-28 Ben Elliston <bje@gnu.org>
* config.guess (*:Darwin:*:*): Simplify UNAME_PROCESSOR.
2020-01-17 Simon Marchi <simon.marchi@efficios.com>
* Makefile.def: Add dependencies of all-gdbsupport on all-bfd.
* Makefile.in: Re-generate.
2020-01-14 Tom Tromey <tom@tromey.com>
* src-release.sh (GDB_SUPPORT_DIRS): Add gdbsupport.
* MAINTAINERS: Add gdbsupport.
* configure: Rebuild.
* configure.ac (configdirs): Add gdbsupport.
* gdbsupport: New directory, move from gdb/gdbsupport.
* Makefile.def (host_modules, dependencies): Add gnulib.
* Makefile.in: Rebuild.
2020-01-09 Aaron Merey <amerey@redhat.com>
* config/debuginfod.m4: New file. Add macro AC_DEBUGINFOD. Adds
new configure option --with-debuginfod.
* configure: Regenerate.
* configure.ac: Call AC_DEBUGINFOD.
2019-12-26 Christian Biesinger <cbiesinger@google.com>
* .gitignore: Add perf.data and perf.data.old.
2019-10-17 Sergio Durigan Junior <sergiodj@redhat.com>
* src-release.sh (GDB_SUPPORT_DIRS): Add libctf.
2019-10-17 Alan Modra <amodra@gmail.com>
PR 29
* src-release.sh (getver): Replace "head -1" with "head -n 1".
2019-07-30 Nick Alcock <nick.alcock@oracle.com>
* Makefile.def (host_modules): libctf is no longer no_install.
* Makefile.in: Regenerated.
2019-07-13 Nick Alcock <nick.alcock@oracle.com>
* Makefile.def (dependencies): all-ld depends on all-libctf.
* Makefile.in: Regenerated.
2019-09-09 Phil Blundell <pb@pbcl.net>
binutils 2.33 branch created
2019-08-19 Tom Tromey <tom@tromey.com>
* configure: Rebuild.
* configure.ac: Add --with-static-standard-libraries.
2019-08-09 Nick Clifton <nickc@redhat.com>
* libiberty: Sync with gcc. Bring in:
2019-08-08 Martin Liska <mliska@suse.cz>
PR bootstrap/91352
* lrealpath.c (is_valid_fd): New function.
2019-07-24 Martin Liska <mliska@suse.cz>
PR lto/91228
* simple-object-elf.c (simple_object_elf_copy_lto_debug_sections):
Find first '\0' starting from gnu_lto + 1.
2019-07-12 Ren Kimura <rkx1209dev@gmail.com>
* simple-object-elf.c (simple_object_elf_match): Check zero value shstrndx.
This fixes a Bug 90924.
2019-07-22 Martin Liska <mliska@suse.cz>
* simple-object-elf.c (simple_object_elf_copy_lto_debug_sections):
Do not search for gnu_lto_v1, but search for first '\0'.
2019-07-18 Eduard-Mihai Burtescu <eddyb@lyken.rs>
* cplus-dem.c: Include rust-demangle.h.
* rust-demangle.c: Include rust-demangle.h.
* rust-demangle.h: New file.
2019-05-31 Michael Forney <mforney@mforney.org>
* cp-demangle.c: Don't define CP_DYNAMIC_ARRAYS if __STDC_NO_VLA__
is non-zero.
2019-04-30 Ben L <bobsayshilol@live.co.uk>
* d-demangle.c (dlang_parse_assocarray): Correctly handle error result.
* testsuite/d-demangle-expected: Add testcase.
* d-demangle.c (dlang_parse_tuple): Correctly handle error result.
* testsuite/d-demangle-expected: Add testcase.
* d-demangle.c (dlang_parse_structlit): Correctly handle error result.
* testsuite/d-demangle-expected: Add testcase.
* d-demangle.c (dlang_parse_arrayliteral): Correctly handle error result.
* testsuite/d-demangle-expected: Add testcase.
* d-demangle.c (dlang_parse_integer): Fix stack underflow.
* testsuite/d-demangle-expected: Add testcase.
* cp-demangle (d_print_comp_inner): Guard against a NULL 'typed_name'.
* testsuite/demangle-expected: Add testcase.
* cp-demangle.c (d_encoding): Guard against NULL return values from
d_right (dc).
* testsuite/demangle-expected: Add testcase.
2019-04-29 Ben L <bobsayshilol@live.co.uk>
* cp-demangle.c (d_expression_1): Don't peek ahead unless the current
char is valid.
* testsuite/demangle-expected: Add testcase.
2019-04-10 Nick Clifton <nickc@redhat.com>
PR 89394
* cp-demangle.c (cplus_demangle_fill_name): Reject negative
lengths.
(d_count_templates_scopes): Replace num_templates and num_scopes
parameters with a struct d_print_info pointer parameter. Adjust
body of the function accordingly. Add recursion counter and check
that the recursion limit is not reached.
(d_print_init): Pass dpi parameter to d_count_templates_scopes.
Reset recursion counter afterwards, unless the recursion limit was
reached.
2019-07-13 Joel Brobecker <brobecker@adacore.com>
* src-release (getver): If $tool/gdbsupport/create-version.sh
exists, use that to determine the version number.
2019-06-21 Andreas Schwab <schwab@linux-m68k.org>
* src-release.sh (GDB_SUPPORT_DIRS): Add gnulib.
2019-06-14 Tom Tromey <tom@tromey.com>
* MAINTAINERS: Add gnulib.
* gnulib: New directory, move from gdb/gnulib.
* configure.ac (host_libs): Add gnulib.
* configure: Rebuild.
* Makefile.def (host_modules, dependencies): Add gnulib.
* Makefile.in: Rebuild.
2019-06-03 Nick Clifton <nickc@redhat.com>
Revert:
2019-05-29 Nick Clifton <nickc@redhat.com>
* configure.ac (noconfigdirs): Add libctf if the target does not use
the ELF file format.
* configure: Regenerate.
2019-05-29 Nick Clifton <nickc@redhat.com>
* src-release.sh (do_proto_toplev): Add libctf to list of
directories that can be disabled.
2019-05-29 Nick Clifton <nickc@redhat.com>
* configure.ac (noconfigdirs): Add libctf if the target does not use
the ELF file format.
* configure: Regenerate.
2019-05-28 Nick Alcock <nick.alcock@oracle.com>
* Makefile.def (dependencies): configure-libctf depends on all-bfd
and all its deps.
* Makefile.in: Regenerated.
2019-05-28 Nick Alcock <nick.alcock@oracle.com>
* MAINTAINERS: Add libctf.
2019-05-28 Nick Alcock <nick.alcock@oracle.com>
* Makefile.def (host_modules): Add libctf.
* Makefile.def (dependencies): Likewise.
libctf depends on zlib, libiberty, and bfd.
* Makefile.in: Regenerated.
* configure.ac (host_libs): Add libctf.
* configure: Regenerated.
2019-05-23 Jose E. Marchesi <jose.marchesi@oracle.com>
* config.guess: Synchronize with config project master sources.
* config.sub: Likewise.
* readline/support/config.guess: Likewise.
* readline/support/config.sub: Likewise.
2019-04-10 Nick Clifton <nickc@redhat.com>
* libiberty: Sync with gcc. Bring in:
2019-04-10 Nick Clifton <nickc@redhat.com>
PR 89394
* cp-demangle.c (cplus_demangle_fill_name): Reject negative
lengths.
(d_count_templates_scopes): Replace num_templates and num_scopes
parameters with a struct d_print_info pointer parameter. Adjust
body of the function accordingly. Add recursion counter and check
that the recursion limit is not reached.
(d_print_init): Pass dpi parameter to d_count_templates_scopes.
Reset recursion counter afterwards, unless the recursion limit was
reached.
2018-06-24 Nick Clifton <nickc@redhat.com>
2.32 branch created.
2019-01-14 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
Merge from GCC:
PR target/88535
* config.guess: Import upstream version 2019-01-03.
* config.sub: Import upstream version 2019-01-01.
2019-01-10 Nick Clifton <nickc@redhat.com>
* libiberty: Sync with gcc. Bring in:
2019-01-09 Sandra Loosemore <sandra@codesourcery.com>
PR other/16615
* cp-demangle.c: Mechanically replace "can not" with "cannot".
* floatformat.c: Likewise.
* strerror.c: Likewise.
2018-12-22 Jason Merrill <jason@redhat.com>
Remove support for demangling GCC 2.x era mangling schemes.
* cplus-dem.c: Remove cplus_mangle_opname, cplus_demangle_opname,
internal_cplus_demangle, and all subroutines.
(libiberty_demanglers): Remove entries for ancient GNU (pre-3.0),
Lucid, ARM, HP, and EDG demangling styles.
(cplus_demangle): Remove 'work' variable. Don't call
internal_cplus_demangle.
2019-01-03 Дилян Палаузов <dilyan.palauzov@aegee.org>
* configure.ac: Don't configure readline if --with-system-readline is
used.
* configure: Re-generate.
Sync tsan_interceptors.cc with upstream file tsan_interceptors_posix.cpp.
Changes:
- define fileno_unlocked() for NetBSD
- use defined __sanitizer_FILE for NetBSD
- handle symbol indirection for vfork and nanosleep
- delete sem_* API, currently moved to common code with sanitizers
- undef SANITIZER_INTERCEPT_PTHREAD_SIGMASK to fix build
- remove locally introduced SANITIZER_NETBSD
With these changes, TSan for GCC builds.
Use the 2020-09-10 snapshot to replace two-years-old GCC9 copy of
interceptors.
With this change, ASan works with the locally patched GCC tree (as there
were IOCTL operations not properly defined).
- alpha is running and no regression in ATF (tests in lib/libc/sys are
skipped due to port-alpha/55652).
- m68000 is successfully built and running at a same level as before.
- m68k (amiga, mac68k, sun3) is running and no regression in a tiny
subset of ATF (kernel, lib/libc/{gen,sys}).
- macppc (oea) and evbppc (booke, ibm4xx) are running and no regression
in ATF.
Cherry-pick the missing change from:
commit 983d7ddd0b278b45d815cbac9197205b39c4860a
Author: Kamil Rytarowski <n54@gmx.com>
Date: Thu Jul 11 06:22:35 2019 +0000
Add NetBSD LSan support
Summary:
Combine few relatively small changes into one:
- implement internal_ptrace() and internal_clone() for NetBSD
- add support for stoptheworld based on the ptrace(2) API
- define COMPILER_RT_HAS_LSAN for NetBSD
- enable tests for NetBSD/amd64
Inspired by the original implementation by Christos Zoulas in netbsd/src for GCC.
The implementation is in theory CPU independent through well defined macros
across all NetBSD ports, however only the x86_64 version was tested.
Reviewers: mgorny, dvyukov, vitalybuka, joerg, jfb
Reviewed By: vitalybuka
Subscribers: dexonsmith, jfb, srhines, kubamracek, llvm-commits, christos
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64057
llvm-svn: 365735
Cherry-pick and adapt:
commit 02519fc7a6f8c528f67975a9f78ce64dabf402b4
Author: Kamil Rytarowski <n54@gmx.com>
Date: Thu Sep 12 18:57:58 2019 +0000
Add getauxval() compat for NetBSD
Summary:
getauxval() is not available on NetBSD and there is no a direct equivalent.
Add a function that implements the same semantics with NetBSD internals.
Reorder the GetPageSize() functions to prefer the sysctl approach for NetBSD.
It no longer makes a difference which approach is better. Avoid changing
conditional code path.
Reviewers: vitalybuka, dvyukov, mgorny, joerg
Reviewed By: vitalybuka
Subscribers: llvm-commits, #sanitizers
Tags: #sanitizers, #llvm
Differential Revision: https://reviews.llvm.org/D67329
llvm-svn: 371758
Cherry-pick and adapt:
commit fc356dcc11c10003ff22acff667b0a9f5e6c1e0f
Author: Kamil Rytarowski <n54@gmx.com>
Date: Tue Dec 24 20:33:54 2019 +0100
[compiler-rt] Adapt stop-the-world for ptrace changes in NetBSD-9.99.30
Handle PT_LWPNEXT for newer kernels and keep PT_LWPINFO for older ones.
Cherry-pick and adapt:
commit 983d7ddd0b278b45d815cbac9197205b39c4860a
Author: Kamil Rytarowski <n54@gmx.com>
Date: Thu Jul 11 06:22:35 2019 +0000
Add NetBSD LSan support
Summary:
Combine few relatively small changes into one:
- implement internal_ptrace() and internal_clone() for NetBSD
- add support for stoptheworld based on the ptrace(2) API
- define COMPILER_RT_HAS_LSAN for NetBSD
- enable tests for NetBSD/amd64
Inspired by the original implementation by Christos Zoulas in netbsd/src for GCC.
The implementation is in theory CPU independent through well defined macros
across all NetBSD ports, however only the x86_64 version was tested.
Reviewers: mgorny, dvyukov, vitalybuka, joerg, jfb
Reviewed By: vitalybuka
Subscribers: dexonsmith, jfb, srhines, kubamracek, llvm-commits, christos
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64057
llvm-svn: 365735
Cherry-pick and adapt:
commit 8827047551570b7ed7088765c3de2a8cce6823b8
Author: Kamil Rytarowski <n54@gmx.com>
Date: Sat Sep 21 07:30:42 2019 +0000
Stop tracking atexit/__cxa_atexit/pthread_atfork allocations in LSan/NetBSD
Summary:
The atexit(3) and __cxa_atexit() calls allocate internally memory and free on exit,
after executing all callback. This causes false positives as DoLeakCheck() is called
from the atexit handler. In the LSan/ASan tests there are strict checks triggering
false positives here.
Intercept all atexit(3) and __cxa_atexit() calls and disable LSan when calling the
real functions.
Stop tracing allocations in pthread_atfork(3) funtions, as there are performed
internal allocations that are not freed for the time of running StopTheWorld()
code. This avoids false-positives.
The same changes have to be replicated in the ASan and LSan runtime.
Non-NetBSD OSs are not tested and this code is restricted to NetBSD only.
Reviewers: dvyukov, joerg, mgorny, vitalybuka, eugenis
Reviewed By: vitalybuka
Subscribers: jfb, llvm-commits, #sanitizers
Tags: #sanitizers, #llvm
Differential Revision: https://reviews.llvm.org/D67331
llvm-svn: 372459
Cherry-pick and adapt:
commit 1b58389428ed07a7322ba9c2bcaeec99807f9457
Author: Kamil Rytarowski <n54@gmx.com>
Date: Sat Sep 21 07:45:02 2019 +0000
Add __lsan::ScopedInterceptorDisabler for strerror(3)
Summary:
strerror(3) on NetBSD uses internally TSD with a destructor that is never
fired for exit(3). It's correctly called for pthread_exit(3) scenarios.
This is a case when a leak on exit(3) is expected, unavoidable and harmless.
Reviewers: joerg, vitalybuka, dvyukov, mgorny
Reviewed By: vitalybuka
Subscribers: dmgreen, kristof.beyls, jfb, llvm-commits, #sanitizers
Tags: #sanitizers, #llvm
Differential Revision: https://reviews.llvm.org/D67337
llvm-svn: 372461
into '_' to meet sh variable name rules) into a shell string processing
loop.
On my test system, this reduces the total elapsed time for the bin/sh ATF
tests from about 109 secs to about 102 (user cpu from 24.5 to 21, sys cpu
from 34 to 30) and the usr.bin/make tests elapsed time from 42.5 to 40
secs (user from a bit over 15 to a bit over 13, and sys from 16+ to 13+).
(Recorded on an AMD64 domU).
These probably exaggerate the effect, as there are a bunch of quite small
tests, which means the ATF overhead (which this change affects) is a greater
proportion of the total test time than for some other tests where most of
the time is spent actually testing.
But I am fairly confident that there will be at least some improvement.
This could be further improved by removing the cmdsub invocation method,
and instead passing the name of a variable containing the string to
normalise (with the result returned in that same var) - but that would
mean altering all the callers as well. Some other time maybe.
Cherry-pick and adapt:
commit 0da15ff318480f30bbc90b9e8ecb9bb9431c7fd3
Author: Kamil Rytarowski <n54@gmx.com>
Date: Fri Jan 3 02:35:33 2020 +0100
[compiler-rt] [netbsd] Switch to syscall for ThreadSelfTlsTcb()
This change avoids using internal, namespaced per-CPU calls that
are not a stable interface to read the TSL base value.
- most builds are working
- arm64, amd64, sparc, evbmips64-eb, and armv7hf-el are known to be
stable and work
- mips64-el needs new binutils, in testing
- vax, sh3*, riscv64 and mipse[lb] builds are broken in various ways
* route: ensure IPv4LL routes come last in priority
* DHCP: fix many issues with extending the last lease
* privsep: don't read control group from config in privsep
* privsep: only the master process responds to signals
* privsep: use a socketpair for stderr/stdin rather than dupping /dev/null
* privsep: right limit stdin/stderr/stdout
* privsep: dumping a lease is now run in a sandbox
* options: check if kernel supports INET or INET6 before enabling default
* options: let clientid override a prior duid
* options: allow -1 to represent infinity for requested lease time
* dhcpcd: fix a crash initing a new interface after route overflow
slightly updated in status. x86-64 and sparc and mips64-eb
all appear to work well enough to complete or get well into
an atf run (still going.)
note first ICE segv on sh3
right now. new address-of-packed-member and format-overflow
warnings have new GCC_NO_ADDR_OF_PACKED_MEMBER amd
GCC_NO_FORMAT_OVERFLOW variables to remove these warnings.
apply to a bunch of the tree. mostly, these are real bugs that
should be fixed, but in many cases, only by removing the 'packed'
attribute from some structure that doesn't really need it. (i
looked at many different ones, and while perhaps 60-80% were
already properly aligned, it wasn't clear to me that the uses
were always coming from sane data vs network alignment, so it
doesn't seem safe to remove packed without careful research for
each affect struct.) clang already warned (and was not erroring)
for many of these cases, but gcc picked up dozens more.
- live patching support
- shell completion help
- generally better diagnostic output (less verbose/more useful)
- diagnostics and optimisation choices can be emitted in json
- asan memory usage reduction
- many general, and specific to switch, inter-procedure,
profile and link-time optimisations. from the release notes:
"Overall compile time of Firefox 66 and LibreOffice 6.2.3 on
an 8-core machine was reduced by about 5% compared to GCC 8.3"
- OpenMP 5.0 support
- better spell-guesser
- partial experimental support for c2x and c++2a
- c++17 is no longer experimental
- arm AAPCS GCC 6-8 structure passing bug fixed, may cause
incompatibility (restored compat with GCC 5 and earlier.)
- openrisc support
This helps distinguising NetBSD-current pages from NetBSD-9.0 pages better
on man.netbsd.org, which is still using nroff. (Using mandoc needs more
tweaking to avoid all page footers reflecting the uname release of the
web server.)
operation kauth_authorize_system().
KAUTH_SYSTEM_MOUNT / KAUTH_REQ_SYSTEM_MOUNT_NEW wants the to be
covered vnode and the mount flags, not the mount structure.
Fix for PR kern/55602: zpool panic on mounting zfs filesystem
else sys.mk will attempt to write it to the source tree,
which may be r/o and fail.
XXX may cause updates builds to fail because a file listed
in .depend won't exist anymore.
add method to transform both the "__foo__" and "@foo@"
version of various transforms, while upstream are converting
from the former to the latter it seems. convert all the
common transforms to this method.
expand the grep for .pc files to look for missing @foo@
fixes to the man page genration, and ignore a few things
(thanks uwe@ - .IN lines, and stuff between tab(@) and .TE.)
xorg-server __default_font_path__ -> @default_font_path@,
and xfree86 transform xconfigdir, xkbdir, and modulepath
using new both method.
Reset STACK_BOUNDARY to default, 16, to fix strange freeze for amiga,
when kernel is compiled by GCC8.
For m68k, the stack pointer is required to be aligned to 16-bit boundary
by architecture. Whereas System V ABI demands it to be aligned to 32-bit
boundary.
According to the document, STACK_BOUNDARY is ``the minimum alignment
enforced by hardware for the stack pointer on this machine.'' Whereas,
PREFERRED_STACK_BOUNDARY should be used ``if you wish to preserve a
certain alignment for the stack pointer, greater than what the hardware
enforces.''
Therefore, STACK_BOUNDARY and PREFERRED_STACK_BOUNDARY should be 16 and
32, respectively, for m68k. This is how Linux/m68k does.
GCC 8 generates codes that wisely allocate 64-bit objects on stack by
using STACK_BOUNDARY. This corrupts the stack frame if it is not properly
aligned.
Since the architecture only guarantees the stack pointer to be aligned to
16-bit boundary, it is not aligned to 32-bit boundary in an instance in
general. If the interrupt occurs at this moment, the interrupt handler
spoils the stack frame as explained above, which results in the mysterious
kernel freezes.
I guess that similar failures can occur even for userland with signal.
With this setting, amiga kernel works just fine as far as I can see.
Furthermore, I've confirmed for amiga, mac68k, and sun3,
(1) Kernel compiled by patched GCC8 works with
(1-a) userland built by GCC7 and non-modified GCC8, and
(1-b) userland built by patched GCC8.
(2) Userland binaries compiled by GCC7 and non-modified GCC8 work fine
with kernel and base libraries built by patched GCC8.
(3) There's no regression observed for tests/kernel, tests/lib/libc/sys,
and tests/lib/libc/gen.
This also fixes sun2 kernel to a considerable extent. With non-modified
GCC8, sun2 kernel crashes in strange ways during the early boot stages.
With this change, it boots singleuser.
OK mrg
Version 4.4.2
22 January 2020
Release Notes
NEW FEATURES
Please note that that ISC DHCP is now licensed under the Mozilla Public License,
MPL 2.0. Please see https://www.mozilla.org/en-US/MPL/2.0/ to read the MPL 2.0
license terms.
While release 4.4.2 is primarily a maintenance release that addresses a number
of defects, it does introduce a few new features:
- Keama - Keama is a migration utility that assists in converting ISC DHCP
server configuration files to Kea configuration files. It is found in the
keama subdirectory and includes a README.md file with instructions on how
to build it as well as a manpage on its usage.
- Two new server parameters related to ping checking were added:
1. ping-cltt-secs which allows the user to specify the number of seconds
that must elapse since CLTT before a ping check is conducted.
2. ping-timeout-ms which allows the user to specify the amount of time the
server waits for a ping-check response in milliseconds rather than in
seconds.
In general, the areas of focus for ISC DHCP 4.4 were:
1. Dynamic DNS additions
2. dhclient improvements
3. Support for dynamic shared libraries
Dynamic DNS Improvements:
- We added three new server configuration parameters which influence DDNS
conflict resolution:
1. ddns-dual-stack-mixed-mode - alters DNS conflict resolution behavior
to mitigate issues with non-compliant clients in dual stack environments.
2. ddns-guard-id-must-match - relaxes the DHCID RR client id matching
requirement of DNS conflict resolution.
3. ddns-other-guard-is-dynamic - alters dual-stack-mixed-mode behavior to
allow unguarded DNS entries to be overwritten in certain cases
- The server now honors update-static-leases parameter for static DHCPv6
hosts.
dhclient Improvements:
- We've added three command line parameters to dhclient:
1. --prefix-len-hint - directs dhclient to use the given length as
the prefix length hint when requesting prefixes
2. --decline-wait-time - instructs the client to wait the given number
of seconds after declining an IPv4 address before issuing a discover
3. --address-prefix-len - specifies the prefix length passed by dhclient
into the client script (via the environment variable ip6_prefixlen) with
each IPv6 address. We added this parameter because we have changed the
default value from 64 to 128 in order to be compliant with RFC3315bis
draft (-09, page 64) and RFC5942, Section 4, point 1.
**WARNING**: The new default value of 128 may not be backwardly compatible
with your environment. If you are operating without a router, such as
between VMs on a host, you may find they cannot see each other with prefix
length of 128. In such cases, you'll need to either provide routing or use
the command line parameter to set the value to 64. Alternatively you may
change the default at compile time by setting DHCLIENT_DEFAULT_PREFIX_LEN
in includes/site.h.
- dhclient will now generate a DHCPv6 DECLINE message when the client script
indicates a DAD failure
Dynamic shared library support:
Configure script, configure.ac+lt, which supports libtool is now provided
with the source tar ball. This script can be used to configure ISC DHCP
to build with libtool and thus use dynamic shared libraries.
Other Highlights:
- The server now supports dhcp-cache-threshold for DHCPv6 operations
- The server now supports DHPv6 address allocation based on EUI-64 DUIDs
- Experimental support for alternate relay port in the both the server
and relay for IPv4, IPv6 and 4o6 (see: draft-ietf-dhc-relay-port-10.txt)
For information on how to install, configure and run this software, as
well as how to find documentation and report bugs, please consult the
README file.
ISC DHCP uses standard GNU configure for installation. Please review the
output of "./configure --help" to see what options are available.
The system has only been tested on Linux, FreeBSD, and Solaris, and may not
work on other platforms. Please report any problems and suggested fixes to
<dhcp-users@isc.org>.
ISC DHCP is open source software maintained by Internet Systems
Consortium. This product includes cryptographic software written
by Eric Young (eay@cryptsoft.com).
Changes since 4.4.2b1 (Bug Fixes)
- Added a clarification on DHCPINFORMs and server authority to
dhcpd.conf.5
[Gitlab #37]
- Only emit lease scrubbing log messages when DEBUG_FAILOVER_MESSAGES
is defined.
[Gitlab #72]
- Added the interface name to socket initialization failure log messages.
Prior to this the log messages stated only the error reason without
stating the target interface.
[Gitlab #75]
- Corrected buffer pointer logic in dhcrelay functions that manipulate
agent relay options. Thanks to Thomas Imbert of MSRC Vulnerabilities
& Mitigations for reporting the issue.
[#71]
- Corrected unresolved symbol errors building relay_unittests when
configured to build using libtool.
[#80]
Changes since 4.4.1 (New Features)
- A new configuration parameter, ping-cltt-secs (v4 operation only), has
been added to allow the user to specify the number of seconds that must
elapse since CLTT before a ping check is conducted. Prior to this, the
value was hard coded at 60 seconds. Please see the server man pages for
a more detailed discussion.
[ISC-Bugs #36283]
- A new configuration parameter, ping-timeout-ms (v4 operation only),
has been added that allows the user to specify the amount of time
the server waits for a ping-check response in milliseconds rather
than in seconds (via ping-timeout). When greater than zero, the value
of ping-timeout-ms will override the value of ping-timeout. Thanks
to Jay Doran from Bluecat Networks for suggesting this feature.
[Gitlab #10]
- An experimental tool called, Keama (KEA Migration Assistant), which helps
translate ISC DHCP configurations to Kea configurations, is now included
in the distribution.
[Gitlab #34]
Changes since 4.4.1 (Bug Fixes)
- Corrected a misuse of the BIND9 DDNS API which caused DDNS updates to be
carried out over TCP rather than UDP. The coding error was exposed by
migration to BIND9 9.11. Thanks to Jinmei Tatuya at Infoblox for
reporting the issue.
[ISC-Bugs #47757]
- Bind9 now defaults to requiring python to build. The Makefile for
building Bind9 when bundled with ISC DHCP was modified to turn off
this dependency.
[Gitlab #3]
- Corrected a dual-stack mixed-mode issue that occurs when both
ddns-guard-id-must-match and ddns-other-guard-is-dynamic
are enabled and that caused the server to incorrectly interpret
the presence of a guard record belonging to another client as
a case of no guard record at all. Thanks to Fernando Soto
from BlueCat Networks for reporting this issue.
[Gitlab #1]
- Corrected a compilation issue that occurred when building without DNS
update ability (e.g. by undefining NSUPDATE).
[Gitlab #16]
- Corrected an issue that was causing the server, when running in
DHPCv4 mode, to segfault when class lease limits are reached.
Thanks to Peter Nagy at Porion-Digital for reporting the matter
and submitting a patch.
[Gitlab #13]
- Made minor changes to eliminate warnings when compiled with GCC 9.
Thanks to Brett Neumeier for bringing the matter to our attention.
[Gitlab #15]
- Fixed potential memory leaks in parser error message generation
spotted by Coverity, CIDs: 1448191, 1448193, 1448194, 1448195
[Gitlab #30]
- Updated URL of IEEE oui.txt in contrib/dhcp-lease-list.pl. Thanks
to Tommy Smith for contributing the patch.
[Gitlab #26]
- Fixed define flags when using SO_BINDTODEVICE. Thanks to Joe LeVeque for
reporting the issue.
[GitLab #19]
- Applied a patch from OpenBSD to always set the scope id of outbound
DHPCv6 packets. Note this change only applies when compiling under
OpenBSD. Thanks to Brad Smith at OpenBSD from bringing it to our
attention.
[Gitlab #33]
- Modified dhclient to not discard config file leases that are
duplicates of server-provided leases and to retain such leases
after they have been used as the fallback active lease and
DHCP service has been restored. This allows them to be used
more than once during the lifetime of a dhclient instance.
This applies to DHCPv4 operation only.
[Gitlab #9]
- Corrected a number of reference counter and zero-length buffer leaks.
Thanks to Christopher Ertl of MSRC Vulnerabilities & Mitigations for
pointing them out.
[Gitlab #57]
- Closed a small window of time between the installation of graceful
shutdown signal handlers and application context startup, during which
the receipt of shutdown signal would cause a REQUIRE() assertion to
occur. Note this issue is only visible when compiling with
ENABLE_GENTLE_SHUTDOWN defined.
[Gitlab #53]
- Corrected a buffer overflow that can occur when retrieving zone
names that are more than 255 characters in length.
[Gitlab #20]
- The "d" domain name option format was incorrectly handled as text
instead of RFC 1035 wire format. Thanks to Jay Doran at BlueCat Networks
for reporting this issue.
[Gitlab #2]
- Improved the error message issued when a host declaration has both
a uid and a dhcp-client-identifier. Server configuration parsing will
now fail if a host declaration specifies more than one uid.
[Gitlab #7]
- Updated developer's documentation on building and running unit tests.
Removed support for --with-atf=bind as BIND9 no longer bundles in ATF
source.
[Gitlab #35]
- Fixed a syntax error in ldap.c which cropped up under Ubuntu
18.04.1/gcc 7.4.0. Thanks to Charles Hedrick for pointing it out.
[Gitlab #51]
- Added clarification to dhcp-options.5 section on ip-address values
describing the first-use DNS resolution of options with hostnames as
values (e.g. next-server).
[Gitlab #28]
- The option format for the server option omapi-key was changed to a
format type 'k' (key name); while server options ldap-port and
ldap-init-retry were changed to 'L' (unsigned 32-bit integer). These
three options were inadvertantly broken when the 'd' format content
was changed to comply with RFC 1035 wire format (see Gitlab #2).
[Gitlab #68]
5458. [bug] Prevent a theoretically possible NULL dereference caused
by a data race between zone_maintenance() and
dns_zone_setview_helper(). [GL #1627]
5455. [bug] named could crash when cleaning dead nodes in
lib/dns/rbtdb.c that were being reused. [GL #1968]
5454. [bug] Address a startup crash that occurred when the server
was under load and the root zone had not yet been
loaded. [GL #1862]
5453. [bug] named crashed on shutdown when a new rndc connection was
received during shutdown. [GL #1747]
5452. [bug] The "blackhole" ACL was accidentally disabled for client
queries. [GL #1936]
5451. [func] Add 'rndc dnssec -status' command. [GL #1612]
5449. [bug] Fix a socket shutdown race in netmgr udp. [GL #1938]
5448. [bug] Fix a race condition in isc__nm_tcpdns_send().
[GL #1937]
5447. [bug] IPv6 addresses ending in "::" could break YAML
parsing. A "0" is now appended to such addresses
in YAML output from dig, mdig, delv, and dnstap-read.
[GL #1952]
5446. [bug] The validator could fail to accept a properly signed
RRset if an unsupported algorithm appeared earlier in
the DNSKEY RRset than a supported algorithm. It could
also stop if it detected a malformed public key.
[GL #1689]
5444. [bug] 'rndc dnstap -roll <value>' did not limit the number of
saved files to <value>. [GL !3728]
5443. [bug] The "primary" and "secondary" keywords, when used
as parameters for "check-names", were not
processed correctly and were being ignored. [GL #1949]
5441. [bug] ${LMDB_CFLAGS} was missing from make/includes.in.
[GL #1955]
5440. [test] Properly handle missing kyua. [GL #1950]
5439. [bug] The DS RRset returned by dns_keynode_dsset() was used in
a non-thread-safe manner. [GL #1926]
--- 9.16.4 released ---
5438. [bug] Fix a race in TCP accepting code. [GL #1930]
5437. [bug] Fix a data race in lib/dns/resolver.c:log_formerr().
[GL #1808]
5436. [security] It was possible to trigger an INSIST when determining
whether a record would fit into a TCP message buffer.
(CVE-2020-8618) [GL #1850]
5435. [tests] Add RFC 4592 responses examples to the wildcard system
test. [GL #1718]
5434. [security] It was possible to trigger an INSIST in
lib/dns/rbtdb.c:new_reference() with a particular zone
content and query patterns. (CVE-2020-8619) [GL #1111]
[GL #1718]
5431. [func] Reject DS records at the zone apex when loading
master files. Log but otherwise ignore attempts to
add DS records at the zone apex via UPDATE. [GL #1798]
5430. [doc] Update docs - with netmgr, a separate listening socket
is created for each IPv6 interface (just as with IPv4).
[GL #1782]
5428. [bug] Clean up GSSAPI resources in nsupdate only after taskmgr
has been destroyed. Thanks to Petr Menšík. [GL !3316]
5426. [bug] Don't abort() when setting SO_INCOMING_CPU on the socket
fails. [GL #1911]
5425. [func] The default value of "max-stale-ttl" has been changed
from 1 week to 12 hours. [GL #1877]
5424. [bug] With KASP, when creating a successor key, the "goal"
state of the current active key (predecessor) was not
changed and thus never removed from the zone. [GL #1846]
5423. [bug] Fix a bug in keymgr_key_has_successor(): it incorrectly
returned true if any other key in the keyring had a
successor. [GL #1845]
5422. [bug] When using dnssec-policy, print correct key timing
metadata. [GL #1843]
5421. [bug] Fix a race that could cause named to crash when looking
up the nodename of an RBT node if the tree was modified.
[GL #1857]
5420. [bug] Add missing isc_{mutex,conditional}_destroy() calls
that caused a memory leak on FreeBSD. [GL #1893]
5418. [bug] delv failed to parse deprecated trusted-keys-style
trust anchors. [GL #1860]
5416. [bug] Fix a lock order inversion in lib/isc/unix/socket.c.
[GL #1859]
5415. [test] Address race in dnssec system test that led to
test failures. [GL #1852]
5414. [test] Adjust time allowed for journal truncation to occur
in nsupdate system test to avoid test failure.
[GL #1855]
5413. [test] Address race in autosign system test that led to
test failures. [GL #1852]
5412. [bug] 'provide-ixfr no;' failed to return up-to-date responses
when the serial was greater than or equal to the
current serial. [GL #1714]
5411. [cleanup] TCP accept code has been refactored to use a single
accept() and pass the accepted socket to child threads
for processing. [GL !3320]
5409. [performance] When looking up NSEC3 data in a zone database, skip the
check for empty non-terminal nodes; the NSEC3 tree does
not have any. [GL #1834]
5408. [protocol] Print Extended DNS Errors if present in OPT record.
[GL #1835]
5407. [func] Zone timers are now exported via statistics channel.
Thanks to Paul Frieden, Verizon Media. [GL #1232]
5405. [bug] 'named-checkconf -p' could include spurious text in
server-addresses statements due to an uninitialized DSCP
value. [GL #1812]
Some highlights in no particular order:
%destructor was somehow lost from the list that follows, it should be
part of it.
Use .Ic for yacc directives when they are defined, .Ql otherwise.
Use explicit .Sq Li (instead of .Ql) in description of %destructor to
make sure the result is consistently quoted. It more readable that
way.
Use .Va and .Vt where appropriate.
C preprocessor directives are marked up with .No (a nop), so that it's
easy to switch them to something else if need be. For now just use
them as plain words.
- add .../xorg subdir to the path
- add dbe and present extensions, both wanted via linkage
.. but maybe these shouldn't be built? they're not in sets.
All these port specific X11R6 servers just used MI wscons, and
all of them have been switched to Xorg servers with wsfb driver etc.
Even if someone wants a smaller monolithic server, it should be
written as a MI "Xwscons" server.
- use ${X11INCS.DIX} in <bsd.x11.mk>
- explicitly undef HAVE_XORG_CONFIG_H to avoid reference to
hw/xfree86/common in mi/miinitext.c
- remove -DHAVE_DIX_CONFIG_H which is already defined in
${X11FLAGS.DIX} and pulled via Makefile.servermod
XXX: -DHAVE_STRCASECMP should also be moved to ${X11FLAGS.DIX},
or should be pulled via include/dix-config.h?
