dash/configure.ac

1874 lines
72 KiB
Plaintext
Raw Normal View History

AC_PREREQ([2.69])
2023-04-15 00:01:04 +02:00
define(_CLIENT_VERSION_MAJOR, 20)
define(_CLIENT_VERSION_MINOR, 1)
define(_CLIENT_VERSION_BUILD, 0)
define(_CLIENT_VERSION_RC, 0)
2018-07-12 11:01:10 +02:00
define(_CLIENT_VERSION_IS_RELEASE, false)
2023-01-12 22:13:43 +01:00
define(_COPYRIGHT_YEAR, 2023)
Merge #7192: Unify product name to as few places as possible 027fdb8 When/if the copyright line does not mention Bitcoin Core developers, add a second line to copyrights in -version, About dialog, and splash screen (Luke Dashjr) cc2095e Rewrite FormatParagraph to handle newlines within input strings correctly (Luke Dashjr) cddffaf Bugfix: Include COPYRIGHT_HOLDERS_SUBSTITUTION in Makefile substitutions so it gets passed to extract-strings correctly (Luke Dashjr) 29598e4 Move PACKAGE_URL to configure.ac (Luke Dashjr) 78ec83d splashscreen: Resize text to fit exactly (Luke Dashjr) 3cae140 Bugfix: Actually use _COPYRIGHT_HOLDERS_SUBSTITUTION everywhere (Luke Dashjr) 4d5a3df Bugfix: gitian-descriptors: Add missing python-setuptools requirement for OS X (biplist module) (Luke Dashjr) e4ab5e5 Bugfix: Correct copyright year in Mac DMG background image (Luke Dashjr) 917b1d0 Set copyright holders displayed in notices separately from the package name (Luke Dashjr) c39a6ff Travis & gitian-osx: Use depends for ds_store and mac_alias modules (Luke Dashjr) 902ccde depends: Add mac_alias to depends (Luke Dashjr) 82a2d98 depends: Add ds_store to depends (Cory Fields) de619a3 depends: Pass PYTHONPATH along to configure (Cory Fields) e611b6e macdeploy: Use rsvg-convert rather than cairosvg (Luke Dashjr) 63bcdc5 More complicated package name substitution for Mac deployment (Luke Dashjr) 1a6c67c Parameterise 2009 in translatable copyright strings (Luke Dashjr) d5f4683 Unify package name to as few places as possible without major changes (Luke Dashjr)
2016-02-04 13:41:58 +01:00
define(_COPYRIGHT_HOLDERS,[The %s developers])
define(_COPYRIGHT_HOLDERS_SUBSTITUTION,[[Dash Core]])
AC_INIT([Dash Core],m4_join([.], _CLIENT_VERSION_MAJOR, _CLIENT_VERSION_MINOR, _CLIENT_VERSION_BUILD)m4_if(_CLIENT_VERSION_RC, [0], [], [rc]_CLIENT_VERSION_RC),[https://github.com/dashpay/dash/issues],[dashcore],[https://dash.org/])
AC_CONFIG_SRCDIR([src/validation.cpp])
AC_CONFIG_HEADERS([src/config/bitcoin-config.h])
AC_CONFIG_AUX_DIR([build-aux])
AC_CONFIG_MACRO_DIR([build-aux/m4])
m4_ifndef([PKG_PROG_PKG_CONFIG], [AC_MSG_ERROR([PKG_PROG_PKG_CONFIG macro not found. Please install pkg-config and re-run autogen.sh])])
PKG_PROG_PKG_CONFIG
if test "x$PKG_CONFIG" = x; then
AC_MSG_ERROR([pkg-config not found])
fi
2017-09-15 13:54:36 +02:00
BITCOIN_DAEMON_NAME=dashd
BITCOIN_GUI_NAME=dash-qt
BITCOIN_CLI_NAME=dash-cli
BITCOIN_TX_NAME=dash-tx
BITCOIN_WALLET_TOOL_NAME=dash-wallet
dnl Unless the user specified ARFLAGS, force it to be cr
2022-06-16 12:14:04 +02:00
dnl This is also the default as-of libtool 2.4.7
AC_ARG_VAR(ARFLAGS, [Flags for the archiver, defaults to <cr> if not set])
if test "x${ARFLAGS+set}" != "xset"; then
ARFLAGS="cr"
fi
AC_CANONICAL_HOST
AH_TOP([#ifndef DASH_CONFIG_H])
AH_TOP([#define DASH_CONFIG_H])
AH_BOTTOM([#endif //DASH_CONFIG_H])
dnl Automake init set-up and checks
Merge #18290: build: Set minimum Automake version to 1.13 ddc7e42d600a0cb3e763cda0dc04a1f2f34e9440 build: Set minimum Automake version to 1.13 (Hennadii Stepanov) Pull request description: This PR suggests to set the required minimum Automake version to `1.13` explicitly for the following reasons: - it guarantees that [CVE-2012-3386](https://lists.gnu.org/archive/html/automake/2012-07/msg00023.html) has been fixed - `AC_CONFIG_MACRO_DIR` macro support, which we already use; from the [release notes](https://lists.gnu.org/archive/html/automake/2012-12/msg00038.html): > Improvements to aclocal and related rebuilds rules: > - Autoconf-provided macros AC_CONFIG_MACRO_DIR and AC_CONFIG_MACRO_DIRS are now traced by aclocal, and can be used to declare the local m4 include directories. Formerly, one had to specify it with an explicit '-I' option to the 'aclocal' invocation. - `AM_SILENT_RULES` macro support (since version `1.11`) Automake `1.13` requires Autoconf `2.65` or greater. We already have `2.69` since #17769. --- For reference, Automake `1.13` was released in [December of 2012](https://lists.gnu.org/archive/html/automake/2012-12/msg00038.html). CentOS 7 uses Automake [`1.13.4`](https://centos.pkgs.org/7/centos-x86_64/automake-1.13.4-3.el7.noarch.rpm.html) See the Automake docs for more info: - [`AM_INIT_AUTOMAKE`](https://www.gnu.org/software/automake/manual/automake.html#Public-Macros) - [List of Automake options](https://www.gnu.org/software/automake/manual/automake.html#List-of-Automake-options) ACKs for top commit: laanwj: so also ACK ddc7e42d600a0cb3e763cda0dc04a1f2f34e9440 fanquake: ACK ddc7e42d600a0cb3e763cda0dc04a1f2f34e9440 - I think adding a minimum required version here is fine. I'd be surprised if someone who is currently building Bitcoin Core was unable to after this change. Tree-SHA512: a1f97864bc3a513450c03d041498f28e823e6f8cd9710d81df081435d72bd4b6cd2f3deb997dbf902f950215a859e48a2ee7ca1f8ebf4271778dd951ab78abf4
2020-03-11 16:31:13 +01:00
AM_INIT_AUTOMAKE([1.13 no-define subdir-objects foreign])
AM_MAINTAINER_MODE([enable])
dnl make the compilation flags quiet unless V=1 is used
Merge #18290: build: Set minimum Automake version to 1.13 ddc7e42d600a0cb3e763cda0dc04a1f2f34e9440 build: Set minimum Automake version to 1.13 (Hennadii Stepanov) Pull request description: This PR suggests to set the required minimum Automake version to `1.13` explicitly for the following reasons: - it guarantees that [CVE-2012-3386](https://lists.gnu.org/archive/html/automake/2012-07/msg00023.html) has been fixed - `AC_CONFIG_MACRO_DIR` macro support, which we already use; from the [release notes](https://lists.gnu.org/archive/html/automake/2012-12/msg00038.html): > Improvements to aclocal and related rebuilds rules: > - Autoconf-provided macros AC_CONFIG_MACRO_DIR and AC_CONFIG_MACRO_DIRS are now traced by aclocal, and can be used to declare the local m4 include directories. Formerly, one had to specify it with an explicit '-I' option to the 'aclocal' invocation. - `AM_SILENT_RULES` macro support (since version `1.11`) Automake `1.13` requires Autoconf `2.65` or greater. We already have `2.69` since #17769. --- For reference, Automake `1.13` was released in [December of 2012](https://lists.gnu.org/archive/html/automake/2012-12/msg00038.html). CentOS 7 uses Automake [`1.13.4`](https://centos.pkgs.org/7/centos-x86_64/automake-1.13.4-3.el7.noarch.rpm.html) See the Automake docs for more info: - [`AM_INIT_AUTOMAKE`](https://www.gnu.org/software/automake/manual/automake.html#Public-Macros) - [List of Automake options](https://www.gnu.org/software/automake/manual/automake.html#List-of-Automake-options) ACKs for top commit: laanwj: so also ACK ddc7e42d600a0cb3e763cda0dc04a1f2f34e9440 fanquake: ACK ddc7e42d600a0cb3e763cda0dc04a1f2f34e9440 - I think adding a minimum required version here is fine. I'd be surprised if someone who is currently building Bitcoin Core was unable to after this change. Tree-SHA512: a1f97864bc3a513450c03d041498f28e823e6f8cd9710d81df081435d72bd4b6cd2f3deb997dbf902f950215a859e48a2ee7ca1f8ebf4271778dd951ab78abf4
2020-03-11 16:31:13 +01:00
AM_SILENT_RULES([yes])
dnl Compiler checks (here before libtool).
if test "x${CXXFLAGS+set}" = "xset"; then
CXXFLAGS_overridden=yes
else
CXXFLAGS_overridden=no
fi
AC_PROG_CXX
dnl By default, libtool for mingw refuses to link static libs into a dll for
dnl fear of mixing pic/non-pic objects, and import/export complications. Since
dnl we have those under control, re-enable that functionality.
case $host in
*mingw*)
lt_cv_deplibs_check_method="pass_all"
;;
esac
AC_ARG_ENABLE([c++20],
[AS_HELP_STRING([--enable-c++20],
[enable compilation in c++20 mode (disabled by default)])],
[use_cxx20=$enableval],
[use_cxx20=no])
dnl Require C++17 or C++20 compiler (no GNU extensions)
if test "x$use_cxx20" = xyes; then
AX_CXX_COMPILE_STDCXX([20], [noext], [mandatory])
else
AX_CXX_COMPILE_STDCXX([17], [noext], [mandatory])
fi
dnl Check if -latomic is required for <std::atomic>
CHECK_ATOMIC
dnl Unless the user specified OBJCXX, force it to be the same as CXX. This ensures
dnl that we get the same -std flags for both.
m4_ifdef([AC_PROG_OBJCXX],[
if test "x${OBJCXX+set}" = "x"; then
OBJCXX="${CXX}"
fi
AC_PROG_OBJCXX
])
dnl OpenBSD ships with 2.4.2
LT_PREREQ([2.4.2])
dnl Libtool init checks.
Merge bitcoin/bitcoin#24112: build: pass win32-dll to LT_INIT() 80e78b6a0479094cae642726f74a17d09b708ddc build: pass win32-dll to LT_INIT() (fanquake) Pull request description: This is the recommended way to support building PE DLLs with modern mingw toolchains and libtool. I made a similar change upstream in the secp256k1 repo: https://github.com/bitcoin-core/secp256k1/pull/1022. Note that we already pass `-no-undefined` to our libtool LDFLAGS. > This option should be used if the package has been ported to build clean > dlls on win32 platforms. > If this macro is not used, libtool will assume that the package libraries > are not dll clean and will build only static libraries on win32 hosts. See: https://www.gnu.org/software/libtool/manual/libtool.html#LT_005fINIT https://www.gnu.org/software/gnulib/manual/html_node/Libtool-and-Windows.html https://autotools.io/libtool/windows.html https://github.com/bitcoin-core/secp256k1/issues/923 Guix Build: ```bash bash-5.1# find guix-build-$(git rev-parse --short=12 HEAD)/output/ -type f -print0 | env LC_ALL=C sort -z | xargs -r0 sha256sum a504bac1c81818e2fa7e14d4b1e2ddf665b9b79683fff4390ec1d76335157012 guix-build-80e78b6a0479/output/aarch64-linux-gnu/SHA256SUMS.part ef9193402c261adb993f6644de3f49858acb6d120002505e0def4827b8772294 guix-build-80e78b6a0479/output/aarch64-linux-gnu/bitcoin-80e78b6a0479-aarch64-linux-gnu-debug.tar.gz 030961da6966a14d3dd7322dc7559dfdc0bdeb9f39d042f379c41bab98303d28 guix-build-80e78b6a0479/output/aarch64-linux-gnu/bitcoin-80e78b6a0479-aarch64-linux-gnu.tar.gz c485e456d325cdd64111eefe36b007e7bcb9e5eb61b1ab752e601023f5853e55 guix-build-80e78b6a0479/output/arm-linux-gnueabihf/SHA256SUMS.part ff0481c57a3ab15c9651ead7b77990c37af9360bfbc1f5285ad5f7d0c66f5acc guix-build-80e78b6a0479/output/arm-linux-gnueabihf/bitcoin-80e78b6a0479-arm-linux-gnueabihf-debug.tar.gz 5f582e30bbdba9df175bf4c5d4aae64e2a1cf572086390ae6962d3ee9f0325a9 guix-build-80e78b6a0479/output/arm-linux-gnueabihf/bitcoin-80e78b6a0479-arm-linux-gnueabihf.tar.gz e96c601af96e851a0351c6f8975feb47623a2dd5e3dd2c15bcdfe8435f845538 guix-build-80e78b6a0479/output/arm64-apple-darwin/SHA256SUMS.part a50db7a8a9b6415842807644760110f2e01665b922b2762634d94e2b497cbd4a guix-build-80e78b6a0479/output/arm64-apple-darwin/bitcoin-80e78b6a0479-arm64-apple-darwin.tar.gz 0f3707a2423483f84be5edff91f8e657cf71ab097d2550f4369760ac8c6a1644 guix-build-80e78b6a0479/output/arm64-apple-darwin/bitcoin-80e78b6a0479-osx-unsigned.dmg 33252a9895c013cfbea06444d6372a23cc555831e4675705b4d7d6b065f06cff guix-build-80e78b6a0479/output/arm64-apple-darwin/bitcoin-80e78b6a0479-osx-unsigned.tar.gz 2ab70177c80c36e98018d07e2aece084c7d3d604e7dc12d2df2e1a077e06b983 guix-build-80e78b6a0479/output/dist-archive/bitcoin-80e78b6a0479.tar.gz cc0237b05948472efa61f7d5a666d8e97b5abeb7f498f3f72d46ff69be38bcf4 guix-build-80e78b6a0479/output/powerpc64-linux-gnu/SHA256SUMS.part b3778fd81bf4e432ad1590792673c91d09c8f8f43daef4cbe0852bca49e1ed57 guix-build-80e78b6a0479/output/powerpc64-linux-gnu/bitcoin-80e78b6a0479-powerpc64-linux-gnu-debug.tar.gz bebe78f0e6a062d943c99470f12bfc520381acec40e0409915cc8d5dccbe5999 guix-build-80e78b6a0479/output/powerpc64-linux-gnu/bitcoin-80e78b6a0479-powerpc64-linux-gnu.tar.gz 350f7b22562d8b6642f37afb3e192d36dbcb360a361c8b834d0f7d50401667b8 guix-build-80e78b6a0479/output/powerpc64le-linux-gnu/SHA256SUMS.part 9a488fbd71c53092feda8dfccecc4ae7d10aa2efe48f99f150cb2322bb28c5e6 guix-build-80e78b6a0479/output/powerpc64le-linux-gnu/bitcoin-80e78b6a0479-powerpc64le-linux-gnu-debug.tar.gz 06d3c472171124d6ca92f95f7d5cb7fc4a523c25396dbbb9522cab920867d3db guix-build-80e78b6a0479/output/powerpc64le-linux-gnu/bitcoin-80e78b6a0479-powerpc64le-linux-gnu.tar.gz 67d591d5f15933d56046d0b8208970dc812ddd240c14a4c3b635cdc256ae5205 guix-build-80e78b6a0479/output/riscv64-linux-gnu/SHA256SUMS.part f9a853d703ac153748f3d9f60d4a74a72c75966dc1d3711b688ebd003ff9389c guix-build-80e78b6a0479/output/riscv64-linux-gnu/bitcoin-80e78b6a0479-riscv64-linux-gnu-debug.tar.gz 07554223c5ab3b940f53f9483054023e639d4e9902810b3d5c1875fd390064ea guix-build-80e78b6a0479/output/riscv64-linux-gnu/bitcoin-80e78b6a0479-riscv64-linux-gnu.tar.gz 294dc1274391b17fb750eac7c76e59c18e972ed3fbf8bccd53ba514843fbc59f guix-build-80e78b6a0479/output/x86_64-apple-darwin/SHA256SUMS.part cc5c1256ca57f80d5ecb93fe2ac477f90945206430545b0463813f7099804f47 guix-build-80e78b6a0479/output/x86_64-apple-darwin/bitcoin-80e78b6a0479-osx-unsigned.dmg c5ff5cf7a8119981f8a1aa2306ff9e84c60e5c9845836eb2562941801495c7de guix-build-80e78b6a0479/output/x86_64-apple-darwin/bitcoin-80e78b6a0479-osx-unsigned.tar.gz 5951712d82391ba0471f253a7f87701b464dad1a90bb1d91866b0c7c51164a8f guix-build-80e78b6a0479/output/x86_64-apple-darwin/bitcoin-80e78b6a0479-osx64.tar.gz 28f84ed57769a642b1679acf90fdaad7ac489d59598d4a8a859021d39a32d878 guix-build-80e78b6a0479/output/x86_64-linux-gnu/SHA256SUMS.part f7c37c47ffaec6fbca36c3c4897b2df2fa7dd5327cb860501eec9d9e987ea5c7 guix-build-80e78b6a0479/output/x86_64-linux-gnu/bitcoin-80e78b6a0479-x86_64-linux-gnu-debug.tar.gz c7c6db897a604e5a85f37938b763538751a133bbab90e80904d4b7198b227d95 guix-build-80e78b6a0479/output/x86_64-linux-gnu/bitcoin-80e78b6a0479-x86_64-linux-gnu.tar.gz 80b1e0e249cefe8941ca0e1a563f479c5e4408da6f0aa02c127182a09310dd8c guix-build-80e78b6a0479/output/x86_64-w64-mingw32/SHA256SUMS.part e7f8b2cd0f0f465e80d96338dcc398306b321a9c99556ca1d39f094752702a21 guix-build-80e78b6a0479/output/x86_64-w64-mingw32/bitcoin-80e78b6a0479-win-unsigned.tar.gz 5191f309c758135fd597df6bc9ef9e2c2746947abb74b38c32e5b6e073fa0995 guix-build-80e78b6a0479/output/x86_64-w64-mingw32/bitcoin-80e78b6a0479-win64-debug.zip 1569de943ca054841141c700f1d4fca2658228b85eee1f44d201b0c881218ef0 guix-build-80e78b6a0479/output/x86_64-w64-mingw32/bitcoin-80e78b6a0479-win64-setup-unsigned.exe c8d78aeedeeaf7af4d325d38fb5e2307965b0080ce08a8cde802afaaa73f157d guix-build-80e78b6a0479/output/x86_64-w64-mingw32/bitcoin-80e78b6a0479-win64.zip ``` ACKs for top commit: hebasto: ACK 80e78b6a0479094cae642726f74a17d09b708ddc Tree-SHA512: fb4a6a443288723776491a9795429273b4a454cfd8230e75570d44fcd71037dc784a2061f6a979322ebc8f9b4131dfbb0494146ab3863f94829e72922be4ec07
2022-02-23 15:17:09 +01:00
LT_INIT([pic-only win32-dll])
dnl Check/return PATH for base programs.
AC_PATH_TOOL(AR, ar)
AC_PATH_TOOL(RANLIB, ranlib)
AC_PATH_TOOL(STRIP, strip)
AC_PATH_TOOL(GCOV, gcov)
AC_PATH_TOOL(LLVM_COV, llvm-cov)
AC_PATH_PROG(LCOV, lcov)
dnl Python 3.8 is specified in .python-version and should be used if available, see doc/dependencies.md
AC_PATH_PROGS([PYTHON], [python3.8 python3.9 python3.10 python3.11 python3.12 python3 python])
AC_PATH_PROG(GENHTML, genhtml)
AC_PATH_PROG([GIT], [git])
AC_PATH_PROG(CCACHE,ccache)
AC_PATH_PROG(XGETTEXT,xgettext)
AC_PATH_PROG(HEXDUMP,hexdump)
AC_PATH_TOOL(OBJCOPY, objcopy)
Enable stacktrace support in gitian builds (#3006) * Remove use of -rdynamic This causes check-symbols to fail horribly and also turned out to be not required when using libbacktrace. It was only required when using "backtrace()" from "<execinfo.h>" * Remove spurious ], from configure.ac * Add -DENABLE_STACKTRACES=1 to CMakeLists.txt * Remove unused method my_backtrace_simple_callback * Use fs::path().filename() instead of basename() * Add static g_exeFileName and g_exeFileBaseName * Use .exe.dbg file when available * Use uint64_t instead of uintptr_t * Implement GetBaseAddress() for unix and win32 * Implement unified crash_info and use it everywhere before printing crash info * Print a serialized version of crash_info when there is no debug info * Implement "-printcrashinfo" command line option * Compile stacktrace support unconditionally and only make crash hooks conditional This also renames the --enable-stacktraces option to --enable-crash-hooks * Enable crash hooks in win/linux Gitian builds * Try to load .debug file on MacOS and enable crash hooks for osx Gitian builds * Check for dsymutil and if it needs --flat * Create .debug files in osx Gitian build * Handle review comments * Also print crash description when no stacktrace is available * Unconditionally add -g1 debug information Instead of making it dependent on "--enable-crash-hooks". We will need the debug info every time now, even in release builds. * Put MacOS debug info into dSYM symbols instead of plain .debug files * Implement MacOS specific GetBaseAddress
2019-07-02 06:16:11 +02:00
AC_PATH_TOOL(DSYMUTIL, dsymutil)
AC_PATH_PROG(DOXYGEN, doxygen)
AM_CONDITIONAL([HAVE_DOXYGEN], [test -n "$DOXYGEN"])
Merge #7192: Unify product name to as few places as possible 027fdb8 When/if the copyright line does not mention Bitcoin Core developers, add a second line to copyrights in -version, About dialog, and splash screen (Luke Dashjr) cc2095e Rewrite FormatParagraph to handle newlines within input strings correctly (Luke Dashjr) cddffaf Bugfix: Include COPYRIGHT_HOLDERS_SUBSTITUTION in Makefile substitutions so it gets passed to extract-strings correctly (Luke Dashjr) 29598e4 Move PACKAGE_URL to configure.ac (Luke Dashjr) 78ec83d splashscreen: Resize text to fit exactly (Luke Dashjr) 3cae140 Bugfix: Actually use _COPYRIGHT_HOLDERS_SUBSTITUTION everywhere (Luke Dashjr) 4d5a3df Bugfix: gitian-descriptors: Add missing python-setuptools requirement for OS X (biplist module) (Luke Dashjr) e4ab5e5 Bugfix: Correct copyright year in Mac DMG background image (Luke Dashjr) 917b1d0 Set copyright holders displayed in notices separately from the package name (Luke Dashjr) c39a6ff Travis & gitian-osx: Use depends for ds_store and mac_alias modules (Luke Dashjr) 902ccde depends: Add mac_alias to depends (Luke Dashjr) 82a2d98 depends: Add ds_store to depends (Cory Fields) de619a3 depends: Pass PYTHONPATH along to configure (Cory Fields) e611b6e macdeploy: Use rsvg-convert rather than cairosvg (Luke Dashjr) 63bcdc5 More complicated package name substitution for Mac deployment (Luke Dashjr) 1a6c67c Parameterise 2009 in translatable copyright strings (Luke Dashjr) d5f4683 Unify package name to as few places as possible without major changes (Luke Dashjr)
2016-02-04 13:41:58 +01:00
AC_ARG_VAR(PYTHONPATH, Augments the default search path for python module files)
AC_ARG_ENABLE([wallet],
[AS_HELP_STRING([--disable-wallet],
[disable wallet (enabled by default)])],
[enable_wallet=$enableval],
[enable_wallet=auto])
AC_ARG_WITH([sqlite],
[AS_HELP_STRING([--with-sqlite=yes|no|auto],
[enable sqlite wallet support (default: auto, i.e., enabled if wallet is enabled and sqlite is found)])],
[use_sqlite=$withval],
[use_sqlite=auto])
AC_ARG_WITH([bdb],
[AS_HELP_STRING([--without-bdb],
[disable bdb wallet support (default is enabled if wallet is enabled)])],
[use_bdb=$withval],
[use_bdb=auto])
AC_ARG_ENABLE([ebpf],
[AS_HELP_STRING([--enable-ebpf],
[enable eBPF tracing (default is yes if sys/sdt.h is found)])],
[use_ebpf=$enableval],
[use_ebpf=yes])
AC_ARG_WITH([miniupnpc],
[AS_HELP_STRING([--with-miniupnpc],
[enable UPNP (default is yes if libminiupnpc is found)])],
[use_upnp=$withval],
[use_upnp=auto])
AC_ARG_WITH([natpmp],
[AS_HELP_STRING([--with-natpmp],
[enable NAT-PMP (default is yes if libnatpmp is found)])],
[use_natpmp=$withval],
[use_natpmp=auto])
AC_ARG_ENABLE(tests,
AS_HELP_STRING([--disable-tests],[do not compile tests (default is to compile)]),
[use_tests=$enableval],
[use_tests=yes])
AC_ARG_ENABLE(gui-tests,
AS_HELP_STRING([--disable-gui-tests],[do not compile GUI tests (default is to compile if GUI and tests enabled)]),
[use_gui_tests=$enableval],
[use_gui_tests=$use_tests])
AC_ARG_ENABLE(bench,
AS_HELP_STRING([--disable-bench],[do not compile benchmarks (default is to compile)]),
[use_bench=$enableval],
[use_bench=yes])
AC_ARG_ENABLE([extended-functional-tests],
AS_HELP_STRING([--enable-extended-functional-tests],[enable expensive functional tests when using lcov (default no)]),
[use_extended_functional_tests=$enableval],
[use_extended_functional_tests=no])
AC_ARG_ENABLE([fuzz],
AS_HELP_STRING([--enable-fuzz],
[build for fuzzing (default no). enabling this will disable all other targets and override --{enable,disable}-fuzz-binary]),
[enable_fuzz=$enableval],
[enable_fuzz=no])
AC_ARG_ENABLE([fuzz-binary],
AS_HELP_STRING([--enable-fuzz-binary],
[enable building of fuzz binary (default yes).]),
[enable_fuzz_binary=$enableval],
[enable_fuzz_binary=yes])
AC_ARG_ENABLE([danger_fuzz_link_all],
AS_HELP_STRING([--enable-danger-fuzz-link-all],
[Danger! Modifies source code. Needs git and gnu sed installed. Link each fuzz target (default no).]),
[enable_danger_fuzz_link_all=$enableval],
[enable_danger_fuzz_link_all=no])
AC_ARG_WITH([qrencode],
[AS_HELP_STRING([--with-qrencode],
[enable QR code support (default is yes if qt is enabled and libqrencode is found)])],
[use_qr=$withval],
[use_qr=auto])
AC_ARG_ENABLE([hardening],
[AS_HELP_STRING([--disable-hardening],
[do not attempt to harden the resulting executables (default is to harden when possible)])],
[use_hardening=$enableval],
[use_hardening=auto])
AC_ARG_ENABLE([reduce-exports],
[AS_HELP_STRING([--enable-reduce-exports],
[attempt to reduce exported symbols in the resulting executables (default is no)])],
[use_reduce_exports=$enableval],
[use_reduce_exports=no])
AC_ARG_ENABLE([ccache],
[AS_HELP_STRING([--disable-ccache],
[do not use ccache for building (default is to use if found)])],
[use_ccache=$enableval],
[use_ccache=auto])
dnl Suppress warnings from external headers (e.g. Boost, Qt).
dnl May be useful if warnings from external headers clutter the build output
dnl too much, so that it becomes difficult to spot Bitcoin Core warnings
dnl or if they cause a build failure with --enable-werror.
AC_ARG_ENABLE([suppress-external-warnings],
[AS_HELP_STRING([--enable-suppress-external-warnings],
[Suppress warnings from external headers (default is no)])],
[suppress_external_warnings=$enableval],
[suppress_external_warnings=no])
AC_ARG_ENABLE([lcov],
[AS_HELP_STRING([--enable-lcov],
[enable lcov testing (default is no)])],
[use_lcov=$enableval],
[use_lcov=no])
Merge #13458: gui: Drop qt4 support af6ac3b677454644364fd24d0df0c02ac9b8c8db doc: Remove mention of Qt4 from build docs (Wladimir J. van der Laan) 462c71f71b5d753bc8327ab833dea23679450ca3 test: Update travis to not test Qt4 anymore (Wladimir J. van der Laan) 907f73bbc5b6c98b01d7c6088a294dea66634a3f gui: Remove QT_VERSION fallbacks for Qt < 5 (Wladimir J. van der Laan) bad068ad9f4bc60bfc10e27d4ffaec92d7df8491 build: Build system changes to support only Qt5 (Wladimir J. van der Laan) Pull request description: Implements #8263. Qt4.x has been EOL since 2015, and at least Gentoo has, or is going to drop support for it. I wouldn't be surprised if other Linux distributions follow. This removes Qt4 detection from the build system, as well as removes all Qt4 fallbacks from the code. Turns out there's more than I expected: this is going to make maintenance of the GUI code, as well as adding new features significantly easier. (I know there's still some references left to qt4 in RPM and Debian build script, but I don't have the knowledge how to fix them) Tree-SHA512: d495924fd4dda6f6566ba44ee96be7cbe62e69ba1ca993b80a8449f78da852b7f1bd3e8200d57cfa1d72233c340eeff4596fb0032ecbddc715d99aea63817d3f (cherry picked from commit dc53f7f2514026db8a28632371e73e6dbf858083) # Conflicts: # .travis.yml # build-aux/m4/bitcoin_qt.m4 # configure.ac # doc/build-osx.md # doc/build-unix.md # doc/dependencies.md # src/qt/README.md # src/qt/bitcoingui.cpp # src/qt/coincontroldialog.cpp # src/qt/dash.cpp # src/qt/guiutil.cpp # src/qt/macdockiconhandler.mm # src/qt/networkstyle.cpp # src/qt/openuridialog.cpp # src/qt/peertablemodel.cpp # src/qt/signverifymessagedialog.cpp # src/qt/test/wallettests.cpp # src/qt/transactionview.cpp
2018-06-24 15:56:00 +02:00
AC_ARG_ENABLE([lcov-branch-coverage],
[AS_HELP_STRING([--enable-lcov-branch-coverage],
[enable lcov testing branch coverage (default is no)])],
[use_lcov_branch=yes],
[use_lcov_branch=no])
AC_ARG_ENABLE([threadlocal],
[AS_HELP_STRING([--enable-threadlocal],
[enable features that depend on the c++ thread_local keyword (currently just thread names in debug logs). (default is to enable if there is platform support)])],
[use_thread_local=$enableval],
[use_thread_local=auto])
AC_ARG_ENABLE([asm],
[AS_HELP_STRING([--disable-asm],
[disable assembly routines (enabled by default)])],
[use_asm=$enableval],
[use_asm=yes])
if test "x$use_asm" = xyes; then
AC_DEFINE(USE_ASM, 1, [Define this symbol to build in assembly routines])
fi
AC_ARG_ENABLE([zmq],
[AS_HELP_STRING([--disable-zmq],
[disable ZMQ notifications])],
[use_zmq=$enableval],
[use_zmq=yes])
AC_ARG_ENABLE(man,
[AS_HELP_STRING([--disable-man],
[do not install man pages (default is to install)])],,
enable_man=yes)
AM_CONDITIONAL(ENABLE_MAN, test "$enable_man" != no)
dnl Enable debug
2014-03-04 06:52:42 +01:00
AC_ARG_ENABLE([debug],
[AS_HELP_STRING([--enable-debug],
[use compiler flags and macros suited for debugging (default is no)])],
2014-03-04 06:52:42 +01:00
[enable_debug=$enableval],
[enable_debug=no])
dnl Enable exception stacktraces
AC_ARG_ENABLE([stacktraces],
[AS_HELP_STRING([--enable-stacktraces],
[gather and print exception stack traces (default is yes)])],
[enable_stacktraces=$enableval],
[enable_stacktraces=yes])
dnl Enable crash hooks
Enable stacktrace support in gitian builds (#3006) * Remove use of -rdynamic This causes check-symbols to fail horribly and also turned out to be not required when using libbacktrace. It was only required when using "backtrace()" from "<execinfo.h>" * Remove spurious ], from configure.ac * Add -DENABLE_STACKTRACES=1 to CMakeLists.txt * Remove unused method my_backtrace_simple_callback * Use fs::path().filename() instead of basename() * Add static g_exeFileName and g_exeFileBaseName * Use .exe.dbg file when available * Use uint64_t instead of uintptr_t * Implement GetBaseAddress() for unix and win32 * Implement unified crash_info and use it everywhere before printing crash info * Print a serialized version of crash_info when there is no debug info * Implement "-printcrashinfo" command line option * Compile stacktrace support unconditionally and only make crash hooks conditional This also renames the --enable-stacktraces option to --enable-crash-hooks * Enable crash hooks in win/linux Gitian builds * Try to load .debug file on MacOS and enable crash hooks for osx Gitian builds * Check for dsymutil and if it needs --flat * Create .debug files in osx Gitian build * Handle review comments * Also print crash description when no stacktrace is available * Unconditionally add -g1 debug information Instead of making it dependent on "--enable-crash-hooks". We will need the debug info every time now, even in release builds. * Put MacOS debug info into dSYM symbols instead of plain .debug files * Implement MacOS specific GetBaseAddress
2019-07-02 06:16:11 +02:00
AC_ARG_ENABLE([crash-hooks],
[AS_HELP_STRING([--enable-crash-hooks],
[hook into exception/signal/assert handling to gather stack traces (default is no)])],
[enable_crashhooks=$enableval],
[enable_crashhooks=no])
dnl Enable in-wallet miner
AC_ARG_ENABLE([miner],
[AS_HELP_STRING([--enable-miner],
[enable in-wallet miner (default is yes)])],
[enable_miner=$enableval],
[enable_miner=yes])
AM_CONDITIONAL([ENABLE_MINER], [test x$enable_miner = xyes])
if test "x$enable_miner" = xyes; then
AC_DEFINE(ENABLE_MINER, 1, [Define this symbol if in-wallet miner should be enabled])
fi
dnl Enable different -fsanitize options
AC_ARG_WITH([sanitizers],
[AS_HELP_STRING([--with-sanitizers],
[comma separated list of extra sanitizers to build with (default is none enabled)])],
[use_sanitizers=$withval])
dnl Enable gprof profiling
AC_ARG_ENABLE([gprof],
[AS_HELP_STRING([--enable-gprof],
[use gprof profiling compiler flags (default is no)])],
[enable_gprof=$enableval],
[enable_gprof=no])
dnl Turn warnings into errors
AC_ARG_ENABLE([werror],
[AS_HELP_STRING([--enable-werror],
[Treat certain compiler warnings as errors (default is no)])],
[enable_werror=$enableval],
[enable_werror=no])
AC_LANG_PUSH([C++])
dnl Check for a flag to turn compiler warnings into errors. This is helpful for checks which may
dnl appear to succeed because by default they merely emit warnings when they fail.
dnl
dnl Note that this is not necessarily a check to see if -Werror is supported, but rather to see if
dnl a compile with -Werror can succeed. This is important because the compiler may already be
dnl warning about something unrelated, for example about some path issue. If that is the case,
dnl -Werror cannot be used because all of those warnings would be turned into errors.
AX_CHECK_COMPILE_FLAG([-Werror],[CXXFLAG_WERROR="-Werror"],[CXXFLAG_WERROR=""])
dnl Check for a flag to turn linker warnings into errors. When flags are passed to linkers via the
dnl compiler driver using a -Wl,-foo flag, linker warnings may be swallowed rather than bubbling up.
dnl See note above, the same applies here as well.
dnl
dnl LDFLAG_WERROR Should only be used when testing -Wl,*
case $host in
*darwin*)
AX_CHECK_LINK_FLAG([-Wl,-fatal_warnings],[LDFLAG_WERROR="-Wl,-fatal_warnings"],[LDFLAG_WERROR=""])
;;
*)
AX_CHECK_LINK_FLAG([-Wl,--fatal-warnings],[LDFLAG_WERROR="-Wl,--fatal-warnings"],[LDFLAG_WERROR=""])
;;
esac
2014-03-04 06:52:42 +01:00
if test "x$enable_debug" = xyes; then
dnl Clear default -g -O2 flags
if test "x$CXXFLAGS_overridden" = xno; then
CXXFLAGS=""
fi
dnl Disable all optimizations
AX_CHECK_COMPILE_FLAG([-O0], [[DEBUG_CXXFLAGS="$DEBUG_CXXFLAGS -O0"]],,[[$CXXFLAG_WERROR]])
dnl Prefer -g3, fall back to -g if that is unavailable.
AX_CHECK_COMPILE_FLAG(
[-g3],
[[DEBUG_CXXFLAGS="$DEBUG_CXXFLAGS -g3"]],
[AX_CHECK_COMPILE_FLAG([-g],[[DEBUG_CXXFLAGS="$DEBUG_CXXFLAGS -g"]],,[[$CXXFLAG_WERROR]])],
[[$CXXFLAG_WERROR]])
AX_CHECK_PREPROC_FLAG([-DDEBUG_CORE],[[DEBUG_CPPFLAGS="$DEBUG_CPPFLAGS -DDEBUG_CORE"]],,[[$CXXFLAG_WERROR]])
AX_CHECK_PREPROC_FLAG([-DDEBUG_LOCKORDER],[[DEBUG_CPPFLAGS="$DEBUG_CPPFLAGS -DDEBUG_LOCKORDER"]],,[[$CXXFLAG_WERROR]])
AX_CHECK_PREPROC_FLAG([-DABORT_ON_FAILED_ASSUME],[[DEBUG_CPPFLAGS="$DEBUG_CPPFLAGS -DABORT_ON_FAILED_ASSUME"]],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-ftrapv],[DEBUG_CXXFLAGS="$DEBUG_CXXFLAGS -ftrapv"],,[[$CXXFLAG_WERROR]])
Enable stacktrace support in gitian builds (#3006) * Remove use of -rdynamic This causes check-symbols to fail horribly and also turned out to be not required when using libbacktrace. It was only required when using "backtrace()" from "<execinfo.h>" * Remove spurious ], from configure.ac * Add -DENABLE_STACKTRACES=1 to CMakeLists.txt * Remove unused method my_backtrace_simple_callback * Use fs::path().filename() instead of basename() * Add static g_exeFileName and g_exeFileBaseName * Use .exe.dbg file when available * Use uint64_t instead of uintptr_t * Implement GetBaseAddress() for unix and win32 * Implement unified crash_info and use it everywhere before printing crash info * Print a serialized version of crash_info when there is no debug info * Implement "-printcrashinfo" command line option * Compile stacktrace support unconditionally and only make crash hooks conditional This also renames the --enable-stacktraces option to --enable-crash-hooks * Enable crash hooks in win/linux Gitian builds * Try to load .debug file on MacOS and enable crash hooks for osx Gitian builds * Check for dsymutil and if it needs --flat * Create .debug files in osx Gitian build * Handle review comments * Also print crash description when no stacktrace is available * Unconditionally add -g1 debug information Instead of making it dependent on "--enable-crash-hooks". We will need the debug info every time now, even in release builds. * Put MacOS debug info into dSYM symbols instead of plain .debug files * Implement MacOS specific GetBaseAddress
2019-07-02 06:16:11 +02:00
else
# We always enable at at least -g1 debug info to support proper stacktraces in crash infos
# Stacktraces will be suboptimal due to optimization, but better than nothing. Also, -fno-omit-frame-pointer
# mitigates this a little bit
if test "x$GCC" = xyes; then
CFLAGS="$CFLAGS -g1 -fno-omit-frame-pointer"
fi
if test "x$GXX" = xyes; then
CXXFLAGS="$CXXFLAGS -g1 -fno-omit-frame-pointer"
fi
fi
2014-03-04 06:52:42 +01:00
if test "x$enable_stacktraces" != xno; then
AC_CHECK_HEADERS([execinfo.h], [], [enable_stacktraces=no])
fi
AM_CONDITIONAL([ENABLE_STACKTRACES], [test x$enable_stacktraces = xyes])
if test "x$enable_stacktraces" = xyes; then
AC_DEFINE(ENABLE_STACKTRACES, 1, [Define this symbol if stacktraces should be enabled])
else
enable_crashhooks=no
fi
Enable stacktrace support in gitian builds (#3006) * Remove use of -rdynamic This causes check-symbols to fail horribly and also turned out to be not required when using libbacktrace. It was only required when using "backtrace()" from "<execinfo.h>" * Remove spurious ], from configure.ac * Add -DENABLE_STACKTRACES=1 to CMakeLists.txt * Remove unused method my_backtrace_simple_callback * Use fs::path().filename() instead of basename() * Add static g_exeFileName and g_exeFileBaseName * Use .exe.dbg file when available * Use uint64_t instead of uintptr_t * Implement GetBaseAddress() for unix and win32 * Implement unified crash_info and use it everywhere before printing crash info * Print a serialized version of crash_info when there is no debug info * Implement "-printcrashinfo" command line option * Compile stacktrace support unconditionally and only make crash hooks conditional This also renames the --enable-stacktraces option to --enable-crash-hooks * Enable crash hooks in win/linux Gitian builds * Try to load .debug file on MacOS and enable crash hooks for osx Gitian builds * Check for dsymutil and if it needs --flat * Create .debug files in osx Gitian build * Handle review comments * Also print crash description when no stacktrace is available * Unconditionally add -g1 debug information Instead of making it dependent on "--enable-crash-hooks". We will need the debug info every time now, even in release builds. * Put MacOS debug info into dSYM symbols instead of plain .debug files * Implement MacOS specific GetBaseAddress
2019-07-02 06:16:11 +02:00
AM_CONDITIONAL([ENABLE_CRASH_HOOKS], [test x$enable_crashhooks = xyes])
if test "x$enable_crashhooks" = xyes; then
AC_DEFINE(ENABLE_CRASH_HOOKS, 1, [Define this symbol if crash hooks should be enabled])
fi
AX_CHECK_LINK_FLAG([-Wl,-wrap=__cxa_allocate_exception], [LINK_WRAP_SUPPORTED=yes],,,)
Enable stacktrace support in gitian builds (#3006) * Remove use of -rdynamic This causes check-symbols to fail horribly and also turned out to be not required when using libbacktrace. It was only required when using "backtrace()" from "<execinfo.h>" * Remove spurious ], from configure.ac * Add -DENABLE_STACKTRACES=1 to CMakeLists.txt * Remove unused method my_backtrace_simple_callback * Use fs::path().filename() instead of basename() * Add static g_exeFileName and g_exeFileBaseName * Use .exe.dbg file when available * Use uint64_t instead of uintptr_t * Implement GetBaseAddress() for unix and win32 * Implement unified crash_info and use it everywhere before printing crash info * Print a serialized version of crash_info when there is no debug info * Implement "-printcrashinfo" command line option * Compile stacktrace support unconditionally and only make crash hooks conditional This also renames the --enable-stacktraces option to --enable-crash-hooks * Enable crash hooks in win/linux Gitian builds * Try to load .debug file on MacOS and enable crash hooks for osx Gitian builds * Check for dsymutil and if it needs --flat * Create .debug files in osx Gitian build * Handle review comments * Also print crash description when no stacktrace is available * Unconditionally add -g1 debug information Instead of making it dependent on "--enable-crash-hooks". We will need the debug info every time now, even in release builds. * Put MacOS debug info into dSYM symbols instead of plain .debug files * Implement MacOS specific GetBaseAddress
2019-07-02 06:16:11 +02:00
AM_CONDITIONAL([CRASH_HOOKS_WRAPPED_CXX_ABI],[test x$LINK_WRAP_SUPPORTED = xyes])
if test x$LINK_WRAP_SUPPORTED = "xyes"; then
Enable stacktrace support in gitian builds (#3006) * Remove use of -rdynamic This causes check-symbols to fail horribly and also turned out to be not required when using libbacktrace. It was only required when using "backtrace()" from "<execinfo.h>" * Remove spurious ], from configure.ac * Add -DENABLE_STACKTRACES=1 to CMakeLists.txt * Remove unused method my_backtrace_simple_callback * Use fs::path().filename() instead of basename() * Add static g_exeFileName and g_exeFileBaseName * Use .exe.dbg file when available * Use uint64_t instead of uintptr_t * Implement GetBaseAddress() for unix and win32 * Implement unified crash_info and use it everywhere before printing crash info * Print a serialized version of crash_info when there is no debug info * Implement "-printcrashinfo" command line option * Compile stacktrace support unconditionally and only make crash hooks conditional This also renames the --enable-stacktraces option to --enable-crash-hooks * Enable crash hooks in win/linux Gitian builds * Try to load .debug file on MacOS and enable crash hooks for osx Gitian builds * Check for dsymutil and if it needs --flat * Create .debug files in osx Gitian build * Handle review comments * Also print crash description when no stacktrace is available * Unconditionally add -g1 debug information Instead of making it dependent on "--enable-crash-hooks". We will need the debug info every time now, even in release builds. * Put MacOS debug info into dSYM symbols instead of plain .debug files * Implement MacOS specific GetBaseAddress
2019-07-02 06:16:11 +02:00
AC_DEFINE(CRASH_HOOKS_WRAPPED_CXX_ABI, 1, [Define this symbol to use wrapped CXX ABIs for exception stacktraces])
fi
# Needed for MinGW targets when debug symbols are enabled as compiled objects get very large
AX_CHECK_COMPILE_FLAG([-Wa,-mbig-obj], [CXXFLAGS="$CXXFLAGS -Wa,-mbig-obj"],,,)
if test x$use_sanitizers != x; then
dnl First check if the compiler accepts flags. If an incompatible pair like
dnl -fsanitize=address,thread is used here, this check will fail. This will also
dnl fail if a bad argument is passed, e.g. -fsanitize=undfeined
AX_CHECK_COMPILE_FLAG(
[[-fsanitize=$use_sanitizers]],
[[SANITIZER_CXXFLAGS=-fsanitize=$use_sanitizers]],
[AC_MSG_ERROR([compiler did not accept requested flags])])
dnl Some compilers (e.g. GCC) require additional libraries like libasan,
dnl libtsan, libubsan, etc. Make sure linking still works with the sanitize
dnl flag. This is a separate check so we can give a better error message when
dnl the sanitize flags are supported by the compiler but the actual sanitizer
dnl libs are missing.
AX_CHECK_LINK_FLAG(
[[-fsanitize=$use_sanitizers]],
[[SANITIZER_LDFLAGS=-fsanitize=$use_sanitizers]],
[AC_MSG_ERROR([linker did not accept requested flags, you are missing required libraries])],
[],
[AC_LANG_PROGRAM([[
#include <cstdint>
#include <cstddef>
extern "C" int LLVMFuzzerTestOneInput(const uint8_t *data, size_t size) { return 0; }
__attribute__((weak)) // allow for libFuzzer linking
]],[[]])])
fi
ERROR_CXXFLAGS=
if test "x$enable_werror" = "xyes"; then
if test "x$CXXFLAG_WERROR" = "x"; then
AC_MSG_ERROR("enable-werror set but -Werror is not usable")
fi
AX_CHECK_COMPILE_FLAG([-Werror=reorder],[ERROR_CXXFLAGS="$ERROR_CXXFLAGS -Werror=reorder"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Werror=gnu],[ERROR_CXXFLAGS="$ERROR_CXXFLAGS -Werror=gnu"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Werror=vla],[ERROR_CXXFLAGS="$ERROR_CXXFLAGS -Werror=vla"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Werror=shadow-field],[ERROR_CXXFLAGS="$ERROR_CXXFLAGS -Werror=shadow-field"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Werror=switch],[ERROR_CXXFLAGS="$ERROR_CXXFLAGS -Werror=switch"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Werror=thread-safety],[ERROR_CXXFLAGS="$ERROR_CXXFLAGS -Werror=thread-safety"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Werror=range-loop-analysis],[ERROR_CXXFLAGS="$ERROR_CXXFLAGS -Werror=range-loop-analysis"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Werror=unused-variable],[ERROR_CXXFLAGS="$ERROR_CXXFLAGS -Werror=unused-variable"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Werror=date-time],[ERROR_CXXFLAGS="$ERROR_CXXFLAGS -Werror=date-time"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Werror=return-type],[ERROR_CXXFLAGS="$ERROR_CXXFLAGS -Werror=return-type"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Werror=conditional-uninitialized],[ERROR_CXXFLAGS="$ERROR_CXXFLAGS -Werror=conditional-uninitialized"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Werror=sign-compare],[ERROR_CXXFLAGS="$ERROR_CXXFLAGS -Werror=sign-compare"],,[[$CXXFLAG_WERROR]])
dnl -Wsuggest-override is broken with GCC before 9.2
dnl https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78010
AX_CHECK_COMPILE_FLAG([-Werror=suggest-override],[ERROR_CXXFLAGS="$ERROR_CXXFLAGS -Werror=suggest-override"],,[[$CXXFLAG_WERROR]],
[AC_LANG_SOURCE([[struct A { virtual void f(); }; struct B : A { void f() final; };]])])
AX_CHECK_COMPILE_FLAG([-Werror=unreachable-code-loop-increment],[ERROR_CXXFLAGS="$ERROR_CXXFLAGS -Werror=unreachable-code-loop-increment"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Werror=mismatched-tags], [ERROR_CXXFLAGS="$ERROR_CXXFLAGS -Werror=mismatched-tags"], [], [$CXXFLAG_WERROR])
fi
if test "x$CXXFLAGS_overridden" = "xno"; then
AX_CHECK_COMPILE_FLAG([-Wall],[WARN_CXXFLAGS="$WARN_CXXFLAGS -Wall"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Wextra],[WARN_CXXFLAGS="$WARN_CXXFLAGS -Wextra"],,[[$CXXFLAG_WERROR]])
Merge #18088: build: ensure we aren't using GNU extensions 0ae8f18dfe143051fec6ae10ea7df10142e3ff2f build: add -Wgnu to compile flags (fanquake) 3a0fd7726b8b916de6cce33bb67f48990575f923 Remove use of non-standard zero variadic macros (Ben Woosley) 49f6178c3e5e3ad54a419da9d8523207da17fc64 Drop unused LOG_TIME_MICROS helper (Ben Woosley) 5d4999951ee32e333b511245862628e80f83b703 prevector: Avoid unnamed struct, which is a GNU extension (DesWurstes) Pull request description: Since we [started using](https://github.com/bitcoin/bitcoin/pull/7165) the `ax_cxx_compile_stdcxx.m4` macro we've been passing `[noext]` to indicate that we don't want to use an extended mode, i.e GNU extensions. Speaking to Cory he clarified that the intention was to "require only vanilla c++11 and turn _off_ extension support so they would fail to compile". However in the codebase we are currently making use of some GNU extensions. We should either remove there usage, or at least amend our CXX compiler checks. I'd prefer the former. #### anonymous structs ```bash ./prevector.h:153:9: warning: anonymous structs are a GNU extension [-Wgnu-anonymous-struct] struct { ``` This is fixed in https://github.com/bitcoin/bitcoin/commit/b849212c1ec01cc8633b8cdcd390da9b1051be0d. #### variadic macros ```bash ./undo.h:57:50: warning: must specify at least one argument for '...' parameter of variadic macro [-Wgnu-zero-variadic-macro-arguments] ::Unserialize(s, VARINT(nVersionDummy)); ``` This is taken care of in #18087. The `LOG_TIME_*` macros introduced in #16805 make use of a [GNU extension](https://gcc.gnu.org/onlinedocs/cpp/Variadic-Macros.html). ```bash In file included from validation.cpp:22: ./logging/timer.h:99:99: warning: token pasting of ',' and __VA_ARGS__ is a GNU extension [-Wgnu-zero-variadic-macro-arguments] BCLog::Timer<std::chrono::milliseconds> PASTE2(logging_timer, __COUNTER__)(__func__, end_msg, ## __VA_ARGS__) ^ ./logging/timer.h:99:99: warning: token pasting of ',' and __VA_ARGS__ is a GNU extension [-Wgnu-zero-variadic-macro-arguments] ./logging/timer.h:99:99: warning: token pasting of ',' and __VA_ARGS__ is a GNU extension [-Wgnu-zero-variadic-macro-arguments] ./logging/timer.h:99:99: warning: token pasting of ',' and __VA_ARGS__ is a GNU extension [-Wgnu-zero-variadic-macro-arguments] ./logging/timer.h:99:99: warning: token pasting of ',' and __VA_ARGS__ is a GNU extension [-Wgnu-zero-variadic-macro-arguments] ./logging/timer.h:101:92: warning: token pasting of ',' and __VA_ARGS__ is a GNU extension [-Wgnu-zero-variadic-macro-arguments] BCLog::Timer<std::chrono::seconds> PASTE2(logging_timer, __COUNTER__)(__func__, end_msg, ## __VA_ARGS__) ^ 6 warnings generated. ``` This is fixed in 081a0ab64eb442bc85c4d4a4d3bc2c8e97ac2a6d and 612e8e138b97fc5ad2f38847300132a8fc423c3f. #### prevention To ensure that usage doesn't creep back in we can add [`-Wgnu`](https://clang.llvm.org/docs/DiagnosticsReference.html#wgnu) to our compile time flags, which will make Clang warn whenever it encounters GNU extensions. This would close #14130. Also related to #17230, where it's suggested we use a GNU extension, the `gnu::pure` attribute. ACKs for top commit: practicalswift: ACK 0ae8f18dfe143051fec6ae10ea7df10142e3ff2f -- diff looks correct MarcoFalke: ACK 0ae8f18dfe143051fec6ae10ea7df10142e3ff2f vasild: utACK 0ae8f18df dongcarl: ACK 0ae8f18dfe143051fec6ae10ea7df10142e3ff2f Tree-SHA512: c517404681ef8edf04c785731d26105bac9f3c9c958605aa24cbe399c649e7c5ee0c4aa8e714fd2b2d335e2fbea4d571e09b0dec36678ef871f0a6683ba6bb7f
2020-05-05 01:19:44 +02:00
AX_CHECK_COMPILE_FLAG([-Wgnu],[WARN_CXXFLAGS="$WARN_CXXFLAGS -Wgnu"],,[[$CXXFLAG_WERROR]])
2020-07-15 13:51:02 +02:00
dnl some compilers will ignore -Wformat-security without -Wformat, so just combine the two here.
AX_CHECK_COMPILE_FLAG([-Wformat -Wformat-security],[WARN_CXXFLAGS="$WARN_CXXFLAGS -Wformat -Wformat-security"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Wvla],[WARN_CXXFLAGS="$WARN_CXXFLAGS -Wvla"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Wshadow-field],[WARN_CXXFLAGS="$WARN_CXXFLAGS -Wshadow-field"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Wswitch],[WARN_CXXFLAGS="$WARN_CXXFLAGS -Wswitch"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Wthread-safety],[WARN_CXXFLAGS="$WARN_CXXFLAGS -Wthread-safety"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Wrange-loop-analysis],[WARN_CXXFLAGS="$WARN_CXXFLAGS -Wrange-loop-analysis"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Wredundant-decls],[WARN_CXXFLAGS="$WARN_CXXFLAGS -Wredundant-decls"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Wunused-variable],[WARN_CXXFLAGS="$WARN_CXXFLAGS -Wunused-variable"],,[[$CXXFLAG_WERROR]])
Merge #19846: build: enable unused member function diagnostic 819d03b932134ee91df3b0fe98a481a331ce57bf refactor: took out unused member functions (Zero) ed69213c2b2a99023bdee5168614cb8b71990f5f build: enable unused member function diagnostic (Zero) Pull request description: This PR enables the `-Wunused-member-function` compiler diagnostic, as discussed in #19702. > **Notice**: The `unused-member-function` diagnostic is only available on clang. Therefore, clang should be used to test this PR. - [x] Include the `-Wunused-member-function`diagnostic in `./configure.ac`. (ed69213c2b2a99023bdee5168614cb8b71990f5f) - [x] Resolve the reported warnings. (819d03b932134ee91df3b0fe98a481a331ce57bf) Currently, enabling this flag no longer reports the following warnings: > **Note**: output from `make 2>&1 | grep "warning: unused member function" | sort | uniq -c` ``` 1 index/blockfilterindex.cpp:54:5: warning: unused member function 'DBHeightKey' [-Wunused-member-function] 2 script/bitcoinconsensus.cpp:50:9: warning: unused member function 'GetType' [-Wunused-member-function] 1 test/util_tests.cpp:1975:14: warning: unused member function 'operator=' [-Wunused-member-function] ``` All tests have passed locally (from `make check` & `src/test/test_bitcoin`). This PR closes #19702. ACKs for top commit: practicalswift: ACK 819d03b932134ee91df3b0fe98a481a331ce57bf - patch still looks correct :) MarcoFalke: ACK 819d03b932134ee91df3b0fe98a481a331ce57bf pox: Tested ACK 819d03b932134ee91df3b0fe98a481a331ce57bf with clang after `make clean`. No unused member function warnings. theStack: tested ACK 819d03b932134ee91df3b0fe98a481a331ce57bf Tree-SHA512: 5fdfbbb02b3dc618a90a874a5caa5e01e596fc1d14a209e75a6981f01b253f9bca0cfac8fdd758dd7151986609fb76571c3745124a29cfd4f8cbb8d82a07272e
2021-01-05 12:05:39 +01:00
AX_CHECK_COMPILE_FLAG([-Wunused-member-function],[WARN_CXXFLAGS="$WARN_CXXFLAGS -Wunused-member-function"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Wdate-time],[WARN_CXXFLAGS="$WARN_CXXFLAGS -Wdate-time"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Wconditional-uninitialized],[WARN_CXXFLAGS="$WARN_CXXFLAGS -Wconditional-uninitialized"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Wsign-compare],[WARN_CXXFLAGS="$WARN_CXXFLAGS -Wsign-compare"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Wduplicated-branches],[WARN_CXXFLAGS="$WARN_CXXFLAGS -Wduplicated-branches"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Wduplicated-cond],[WARN_CXXFLAGS="$WARN_CXXFLAGS -Wduplicated-cond"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Wlogical-op],[WARN_CXXFLAGS="$WARN_CXXFLAGS -Wlogical-op"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Woverloaded-virtual],[WARN_CXXFLAGS="$WARN_CXXFLAGS -Woverloaded-virtual"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Wsuggest-override],[WARN_CXXFLAGS="$WARN_CXXFLAGS -Wsuggest-override"],,[[$CXXFLAG_WERROR]],
[AC_LANG_SOURCE([[struct A { virtual void f(); }; struct B : A { void f() final; };]])])
AX_CHECK_COMPILE_FLAG([-Wunreachable-code-loop-increment],[WARN_CXXFLAGS="$WARN_CXXFLAGS -Wunreachable-code-loop-increment"],,[[$CXXFLAG_WERROR]])
dnl Some compilers (gcc) ignore unknown -Wno-* options, but warn about all
dnl unknown options if any other warning is produced. Test the -Wfoo case, and
dnl set the -Wno-foo case if it works.
AX_CHECK_COMPILE_FLAG([-Wunused-parameter],[NOWARN_CXXFLAGS="$NOWARN_CXXFLAGS -Wno-unused-parameter"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Wself-assign],[NOWARN_CXXFLAGS="$NOWARN_CXXFLAGS -Wno-self-assign"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Wunused-local-typedef],[NOWARN_CXXFLAGS="$NOWARN_CXXFLAGS -Wno-unused-local-typedef"],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-Wimplicit-fallthrough],[NOWARN_CXXFLAGS="$NOWARN_CXXFLAGS -Wno-implicit-fallthrough"],,[[$CXXFLAG_WERROR]])
if test x$suppress_external_warnings != xyes ; then
AX_CHECK_COMPILE_FLAG([-Wdeprecated-copy],[NOWARN_CXXFLAGS="$NOWARN_CXXFLAGS -Wno-deprecated-copy"],,[[$CXXFLAG_WERROR]])
fi
fi
dnl Don't allow extended (non-ASCII) symbols in identifiers. This is easier for code review.
AX_CHECK_COMPILE_FLAG([-fno-extended-identifiers],[[CXXFLAGS="$CXXFLAGS -fno-extended-identifiers"]],,[[$CXXFLAG_WERROR]])
enable_sse42=no
enable_sse41=no
enable_avx2=no
enable_x86_shani=no
if test "x$use_asm" = "xyes"; then
dnl Check for optional instruction set support. Enabling these does _not_ imply that all code will
dnl be compiled with them, rather that specific objects/libs may use them after checking for runtime
dnl compatibility.
dnl x86
AX_CHECK_COMPILE_FLAG([-msse4.2],[[SSE42_CXXFLAGS="-msse4.2"]],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-msse4.1],[[SSE41_CXXFLAGS="-msse4.1"]],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-mavx -mavx2],[[AVX2_CXXFLAGS="-mavx -mavx2"]],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-msse4 -msha],[[X86_SHANI_CXXFLAGS="-msse4 -msha"]],,[[$CXXFLAG_WERROR]])
TEMP_CXXFLAGS="$CXXFLAGS"
CXXFLAGS="$CXXFLAGS $SSE42_CXXFLAGS"
AC_MSG_CHECKING(for SSE4.2 intrinsics)
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[
#include <stdint.h>
#if defined(_MSC_VER)
#include <intrin.h>
#elif defined(__GNUC__) && defined(__SSE4_2__)
#include <nmmintrin.h>
#endif
]],[[
uint64_t l = 0;
l = _mm_crc32_u8(l, 0);
l = _mm_crc32_u32(l, 0);
l = _mm_crc32_u64(l, 0);
return l;
]])],
[ AC_MSG_RESULT(yes); enable_sse42=yes],
[ AC_MSG_RESULT(no)]
)
CXXFLAGS="$TEMP_CXXFLAGS"
Merge #13191: Specialized double-SHA256 with 64 byte inputs with SSE4.1 and AVX2 4defdfab94504018f822dc34a313ad26cedc8255 [MOVEONLY] Move unused Merkle branch code to tests (Pieter Wuille) 4437d6e1f3107a20a8c7b66be8b4b972a82e3b28 8-way AVX2 implementation for double SHA256 on 64-byte inputs (Pieter Wuille) 230294bf5fdeba7213471cd0b795fb7aa36e5717 4-way SSE4.1 implementation for double SHA256 on 64-byte inputs (Pieter Wuille) 1f0e7ca09c9d7c5787c218156fa5096a1bdf2ea8 Use SHA256D64 in Merkle root computation (Pieter Wuille) d0c96328833127284574bfef26f96aa2e4afc91a Specialized double sha256 for 64 byte inputs (Pieter Wuille) 57f34630fb6c3e218bd19535ac607008cb894173 Refactor SHA256 code (Pieter Wuille) 0df017889b4f61860092e1d54e271092cce55f62 Benchmark Merkle root computation (Pieter Wuille) Pull request description: This introduces a framework for specialized double-SHA256 with 64 byte inputs. 4 different implementations are provided: * Generic C++ (reusing the normal SHA256 code) * Specialized C++ for 64-byte inputs, but no special instructions * 4-way using SSE4.1 intrinsics * 8-way using AVX2 intrinsics On my own system (AVX2 capable), I get these benchmarks for computing the Merkle root of 9001 leaves (supported lengths / special instructions / parallellism): * 7.2 ms with varsize/naive/1way (master, non-SSE4 hardware) * 5.8 ms with size64/naive/1way (this PR, non-SSE4 capable systems) * 4.8 ms with varsize/SSE4/1way (master, SSE4 hardware) * 2.9 ms with size64/SSE4/4way (this PR, SSE4 hardware) * 1.1 ms with size64/AVX2/8way (this PR, AVX2 hardware) Tree-SHA512: efa32d48b32820d9ce788ead4eb583949265be8c2e5f538c94bc914e92d131a57f8c1ee26c6f998e81fb0e30675d4e2eddc3360bcf632676249036018cff343e
2018-06-04 09:11:18 +02:00
TEMP_CXXFLAGS="$CXXFLAGS"
CXXFLAGS="$CXXFLAGS $SSE41_CXXFLAGS"
AC_MSG_CHECKING(for SSE4.1 intrinsics)
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[
#include <stdint.h>
#include <immintrin.h>
]],[[
__m128i l = _mm_set1_epi32(0);
return _mm_extract_epi32(l, 3);
]])],
[ AC_MSG_RESULT(yes); enable_sse41=yes; AC_DEFINE(ENABLE_SSE41, 1, [Define this symbol to build code that uses SSE4.1 intrinsics]) ],
[ AC_MSG_RESULT(no)]
)
CXXFLAGS="$TEMP_CXXFLAGS"
TEMP_CXXFLAGS="$CXXFLAGS"
CXXFLAGS="$CXXFLAGS $AVX2_CXXFLAGS"
AC_MSG_CHECKING(for AVX2 intrinsics)
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[
#include <stdint.h>
#include <immintrin.h>
]],[[
__m256i l = _mm256_set1_epi32(0);
return _mm256_extract_epi32(l, 7);
]])],
[ AC_MSG_RESULT(yes); enable_avx2=yes; AC_DEFINE(ENABLE_AVX2, 1, [Define this symbol to build code that uses AVX2 intrinsics]) ],
[ AC_MSG_RESULT(no)]
)
CXXFLAGS="$TEMP_CXXFLAGS"
Merge #13386: SHA256 implementations based on Intel SHA Extensions 66b2cf1ccfad545a8ec3f2a854e23f647322bf30 Use immintrin.h everywhere for intrinsics (Pieter Wuille) 4c935e2eee456ff66cdfb908b0edffdd1e8a6c04 Add SHA256 implementation using using Intel SHA intrinsics (Pieter Wuille) 268400d3188200c9e3dcd3482c4853354388a721 [Refactor] CPU feature detection logic for SHA256 (Pieter Wuille) Pull request description: Based on #13191. This adds SHA256 implementations that use Intel's SHA Extension instructions (using intrinsics). This needs GCC 4.9 or Clang 3.4. In addition to #13191, two extra implementations are provided: * (a) A variable-length SHA256 implementation using SHA extensions. * (b) A 2-way 64-byte input double-SHA256 implementation using SHA extensions. Benchmarks for 9001-element Merkle tree root computation on an AMD Ryzen 1800X system: * Using generic C++ code (pre-#10821): 6.1ms * Using SSE4 (master, #10821): 4.6ms * Using 4-way SSE4 specialized for 64-byte inputs (#13191): 2.8ms * Using 8-way AVX2 specialized for 64-byte inputs (#13191): 2.1ms * Using 2-way SHA-NI specialized for 64-byte inputs (this PR): 0.56ms Benchmarks for 32-byte SHA256 on the same system: * Using SSE4 (master, #10821): 190ns * Using SHA-NI (this PR): 53ns Benchmarks for 1000000-byte SHA256 on the same system: * Using SSE4 (master, #10821): 2.5ms * Using SHA-NI (this PR): 0.51ms Tree-SHA512: 2b319e33b22579f815d91f9daf7994a5e1e799c4f73c13e15070dd54ba71f3f6438ccf77ae9cbd1ce76f972d9cbeb5f0edfea3d86f101bbc1055db70e42743b7
2018-07-09 20:33:41 +02:00
TEMP_CXXFLAGS="$CXXFLAGS"
CXXFLAGS="$CXXFLAGS $X86_SHANI_CXXFLAGS"
AC_MSG_CHECKING(for x86 SHA-NI intrinsics)
Merge #13386: SHA256 implementations based on Intel SHA Extensions 66b2cf1ccfad545a8ec3f2a854e23f647322bf30 Use immintrin.h everywhere for intrinsics (Pieter Wuille) 4c935e2eee456ff66cdfb908b0edffdd1e8a6c04 Add SHA256 implementation using using Intel SHA intrinsics (Pieter Wuille) 268400d3188200c9e3dcd3482c4853354388a721 [Refactor] CPU feature detection logic for SHA256 (Pieter Wuille) Pull request description: Based on #13191. This adds SHA256 implementations that use Intel's SHA Extension instructions (using intrinsics). This needs GCC 4.9 or Clang 3.4. In addition to #13191, two extra implementations are provided: * (a) A variable-length SHA256 implementation using SHA extensions. * (b) A 2-way 64-byte input double-SHA256 implementation using SHA extensions. Benchmarks for 9001-element Merkle tree root computation on an AMD Ryzen 1800X system: * Using generic C++ code (pre-#10821): 6.1ms * Using SSE4 (master, #10821): 4.6ms * Using 4-way SSE4 specialized for 64-byte inputs (#13191): 2.8ms * Using 8-way AVX2 specialized for 64-byte inputs (#13191): 2.1ms * Using 2-way SHA-NI specialized for 64-byte inputs (this PR): 0.56ms Benchmarks for 32-byte SHA256 on the same system: * Using SSE4 (master, #10821): 190ns * Using SHA-NI (this PR): 53ns Benchmarks for 1000000-byte SHA256 on the same system: * Using SSE4 (master, #10821): 2.5ms * Using SHA-NI (this PR): 0.51ms Tree-SHA512: 2b319e33b22579f815d91f9daf7994a5e1e799c4f73c13e15070dd54ba71f3f6438ccf77ae9cbd1ce76f972d9cbeb5f0edfea3d86f101bbc1055db70e42743b7
2018-07-09 20:33:41 +02:00
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[
#include <stdint.h>
#include <immintrin.h>
]],[[
__m128i i = _mm_set1_epi32(0);
__m128i j = _mm_set1_epi32(1);
__m128i k = _mm_set1_epi32(2);
return _mm_extract_epi32(_mm_sha256rnds2_epu32(i, i, k), 0);
]])],
[ AC_MSG_RESULT(yes); enable_x86_shani=yes; AC_DEFINE(ENABLE_X86_SHANI, 1, [Define this symbol to build code that uses x86 SHA-NI intrinsics]) ],
Merge #13386: SHA256 implementations based on Intel SHA Extensions 66b2cf1ccfad545a8ec3f2a854e23f647322bf30 Use immintrin.h everywhere for intrinsics (Pieter Wuille) 4c935e2eee456ff66cdfb908b0edffdd1e8a6c04 Add SHA256 implementation using using Intel SHA intrinsics (Pieter Wuille) 268400d3188200c9e3dcd3482c4853354388a721 [Refactor] CPU feature detection logic for SHA256 (Pieter Wuille) Pull request description: Based on #13191. This adds SHA256 implementations that use Intel's SHA Extension instructions (using intrinsics). This needs GCC 4.9 or Clang 3.4. In addition to #13191, two extra implementations are provided: * (a) A variable-length SHA256 implementation using SHA extensions. * (b) A 2-way 64-byte input double-SHA256 implementation using SHA extensions. Benchmarks for 9001-element Merkle tree root computation on an AMD Ryzen 1800X system: * Using generic C++ code (pre-#10821): 6.1ms * Using SSE4 (master, #10821): 4.6ms * Using 4-way SSE4 specialized for 64-byte inputs (#13191): 2.8ms * Using 8-way AVX2 specialized for 64-byte inputs (#13191): 2.1ms * Using 2-way SHA-NI specialized for 64-byte inputs (this PR): 0.56ms Benchmarks for 32-byte SHA256 on the same system: * Using SSE4 (master, #10821): 190ns * Using SHA-NI (this PR): 53ns Benchmarks for 1000000-byte SHA256 on the same system: * Using SSE4 (master, #10821): 2.5ms * Using SHA-NI (this PR): 0.51ms Tree-SHA512: 2b319e33b22579f815d91f9daf7994a5e1e799c4f73c13e15070dd54ba71f3f6438ccf77ae9cbd1ce76f972d9cbeb5f0edfea3d86f101bbc1055db70e42743b7
2018-07-09 20:33:41 +02:00
[ AC_MSG_RESULT(no)]
)
CXXFLAGS="$TEMP_CXXFLAGS"
# ARM
AX_CHECK_COMPILE_FLAG([-march=armv8-a+crc+crypto],[[ARM_CRC_CXXFLAGS="-march=armv8-a+crc+crypto"]],,[[$CXXFLAG_WERROR]])
AX_CHECK_COMPILE_FLAG([-march=armv8-a+crc+crypto], [ARM_SHANI_CXXFLAGS="-march=armv8-a+crc+crypto"], [], [$CXXFLAG_WERROR])
TEMP_CXXFLAGS="$CXXFLAGS"
CXXFLAGS="$CXXFLAGS $ARM_CRC_CXXFLAGS"
AC_MSG_CHECKING(for ARMv8 CRC32 intrinsics)
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[
#include <arm_acle.h>
#include <arm_neon.h>
]],[[
#ifdef __aarch64__
__crc32cb(0, 0); __crc32ch(0, 0); __crc32cw(0, 0); __crc32cd(0, 0);
vmull_p64(0, 0);
#else
#error "crc32c library does not support hardware acceleration on 32-bit ARM"
#endif
]])],
[ AC_MSG_RESULT(yes); enable_arm_crc=yes; ],
[ AC_MSG_RESULT(no)]
)
CXXFLAGS="$TEMP_CXXFLAGS"
TEMP_CXXFLAGS="$CXXFLAGS"
CXXFLAGS="$CXXFLAGS $ARM_SHANI_CXXFLAGS"
AC_MSG_CHECKING([for ARMv8 SHA-NI intrinsics])
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[
#include <arm_acle.h>
#include <arm_neon.h>
]],[[
uint32x4_t a, b, c;
vsha256h2q_u32(a, b, c);
vsha256hq_u32(a, b, c);
vsha256su0q_u32(a, b);
vsha256su1q_u32(a, b, c);
]])],
[ AC_MSG_RESULT([yes]); enable_arm_shani=yes; AC_DEFINE([ENABLE_ARM_SHANI], [1], [Define this symbol to build code that uses ARMv8 SHA-NI intrinsics]) ],
[ AC_MSG_RESULT([no])]
)
CXXFLAGS="$TEMP_CXXFLAGS"
fi
CPPFLAGS="$CPPFLAGS -DHAVE_BUILD_INFO"
AC_ARG_WITH([utils],
[AS_HELP_STRING([--with-utils],
[build dash-cli dash-tx dash-wallet (default=yes)])],
[build_bitcoin_utils=$withval],
[build_bitcoin_utils=yes])
AC_ARG_ENABLE([util-cli],
[AS_HELP_STRING([--enable-util-cli],
[build dash-cli])],
[build_bitcoin_cli=$enableval],
[build_bitcoin_cli=$build_bitcoin_utils])
AC_ARG_ENABLE([util-tx],
[AS_HELP_STRING([--enable-util-tx],
[build dash-tx])],
[build_bitcoin_tx=$enableval],
[build_bitcoin_tx=$build_bitcoin_utils])
AC_ARG_ENABLE([util-wallet],
[AS_HELP_STRING([--enable-util-wallet],
[build dash-wallet])],
[build_bitcoin_wallet=$enableval],
[build_bitcoin_wallet=$build_bitcoin_utils])
AC_ARG_WITH([libs],
[AS_HELP_STRING([--with-libs],
[build libraries (default=yes)])],
[build_bitcoin_libs=$withval],
[build_bitcoin_libs=yes])
AC_ARG_WITH([daemon],
[AS_HELP_STRING([--with-daemon],
2015-04-03 00:51:08 +02:00
[build dashd daemon (default=yes)])],
[build_bitcoind=$withval],
[build_bitcoind=yes])
case $host in
*mingw*)
TARGET_OS=windows
AC_CHECK_LIB([kernel32], [GetModuleFileNameA],, AC_MSG_ERROR(libkernel32 missing))
AC_CHECK_LIB([user32], [main],, AC_MSG_ERROR(libuser32 missing))
AC_CHECK_LIB([gdi32], [main],, AC_MSG_ERROR(libgdi32 missing))
AC_CHECK_LIB([comdlg32], [main],, AC_MSG_ERROR(libcomdlg32 missing))
AC_CHECK_LIB([winmm], [main],, AC_MSG_ERROR(libwinmm missing))
AC_CHECK_LIB([shell32], [SHGetSpecialFolderPathW],, AC_MSG_ERROR(libshell32 missing))
AC_CHECK_LIB([comctl32], [main],, AC_MSG_ERROR(libcomctl32 missing))
AC_CHECK_LIB([ole32], [CoCreateInstance],, AC_MSG_ERROR(libole32 missing))
AC_CHECK_LIB([oleaut32], [main],, AC_MSG_ERROR(liboleaut32 missing))
AC_CHECK_LIB([uuid], [main],, AC_MSG_ERROR(libuuid missing))
AC_CHECK_LIB([advapi32], [CryptAcquireContextW],, AC_MSG_ERROR(libadvapi32 missing))
AC_CHECK_LIB([ws2_32], [WSAStartup],, AC_MSG_ERROR(libws2_32 missing))
AC_CHECK_LIB([shlwapi], [PathRemoveFileSpecW],, AC_MSG_ERROR(libshlwapi missing))
AC_CHECK_LIB([iphlpapi], [GetAdaptersAddresses],, AC_MSG_ERROR(libiphlpapi missing))
AC_CHECK_LIB([psapi], [GetProcessMemoryInfo],, [AC_MSG_ERROR([libpsapi missing])])
AC_CHECK_LIB([bcrypt], [main],, [AC_MSG_ERROR([libbcrypt missing])])
dnl -static is interpreted by libtool, where it has a different meaning.
dnl In libtool-speak, it's -all-static.
AX_CHECK_LINK_FLAG([[-static]],[LIBTOOL_APP_LDFLAGS="$LIBTOOL_APP_LDFLAGS -all-static"])
AC_PATH_PROG([MAKENSIS], [makensis], none)
if test x$MAKENSIS = xnone; then
AC_MSG_WARN("makensis not found. Cannot create installer.")
fi
AC_PATH_TOOL(WINDRES, windres, none)
if test x$WINDRES = xnone; then
AC_MSG_ERROR("windres not found")
fi
CPPFLAGS="$CPPFLAGS -D_MT -DWIN32 -D_WINDOWS -DBOOST_THREAD_USE_LIB -D_WIN32_WINNT=0x0601"
dnl libtool insists upon adding -nostdlib and a list of objects/libs to link against.
dnl That breaks our ability to build dll's with static libgcc/libstdc++/libssp. Override
dnl its command here, with the predeps/postdeps removed, and -static inserted. Postdeps are
dnl also overridden to prevent their insertion later.
dnl This should only affect dll's.
archive_cmds_CXX="\$CC -shared \$libobjs \$deplibs \$compiler_flags -static -o \$output_objdir/\$soname \${wl}--enable-auto-image-base -Xlinker --out-implib -Xlinker \$lib"
postdeps_CXX=
dnl We require Windows 7 (NT 6.1) or later
AX_CHECK_LINK_FLAG([[-Wl,--major-subsystem-version -Wl,6 -Wl,--minor-subsystem-version -Wl,1]],[LDFLAGS="$LDFLAGS -Wl,--major-subsystem-version -Wl,6 -Wl,--minor-subsystem-version -Wl,1"],,[[$LDFLAG_WERROR]])
;;
*darwin*)
TARGET_OS=darwin
if test x$cross_compiling != xyes; then
BUILD_OS=darwin
AC_CHECK_PROG([BREW],brew, brew)
if test x$BREW = xbrew; then
dnl These Homebrew packages may be keg-only, meaning that they won't be found
dnl in expected paths because they may conflict with system files. Ask
dnl Homebrew where each one is located, then adjust paths accordingly.
dnl It's safe to add these paths even if the functionality is disabled by
dnl the user (--without-wallet or --without-gui for example).
if test "x$use_bdb" != xno && $BREW list --versions berkeley-db4 >/dev/null && test "x$BDB_CFLAGS" = "x" && test "x$BDB_LIBS" = "x"; then
bdb_prefix=$($BREW --prefix berkeley-db4 2>/dev/null)
dnl This must precede the call to BITCOIN_FIND_BDB48 below.
BDB_CFLAGS="-I$bdb_prefix/include"
BDB_LIBS="-L$bdb_prefix/lib -ldb_cxx-4.8"
fi
if test "x$use_sqlite" != xno && $BREW list --versions sqlite3 >/dev/null; then
export PKG_CONFIG_PATH="$($BREW --prefix sqlite3 2>/dev/null)/lib/pkgconfig:$PKG_CONFIG_PATH"
fi
if $BREW list --versions qt5 >/dev/null; then
export PKG_CONFIG_PATH="$($BREW --prefix qt5 2>/dev/null)/lib/pkgconfig:$PKG_CONFIG_PATH"
fi
gmp_prefix=$($BREW --prefix gmp 2>/dev/null)
if test x$gmp_prefix != x; then
CPPFLAGS="$CPPFLAGS -I$gmp_prefix/include"
LDFLAGS="$LDFLAGS -L$gmp_prefix/lib"
fi
case $host in
*aarch64*)
dnl The preferred Homebrew prefix for Apple Silicon is /opt/homebrew.
dnl Therefore, as we do not use pkg-config to detect miniupnpc and libnatpmp
dnl packages, we should set the CPPFLAGS and LDFLAGS variables for them
dnl explicitly.
if test "x$use_upnp" != xno && $BREW list --versions miniupnpc >/dev/null; then
miniupnpc_prefix=$($BREW --prefix miniupnpc 2>/dev/null)
if test "x$suppress_external_warnings" != xno; then
CPPFLAGS="$CPPFLAGS -isystem $miniupnpc_prefix/include"
else
CPPFLAGS="$CPPFLAGS -I$miniupnpc_prefix/include"
fi
LDFLAGS="$LDFLAGS -L$miniupnpc_prefix/lib"
fi
if test "x$use_natpmp" != xno && $BREW list --versions libnatpmp >/dev/null; then
libnatpmp_prefix=$($BREW --prefix libnatpmp 2>/dev/null)
if test "x$suppress_external_warnings" != xno; then
CPPFLAGS="$CPPFLAGS -isystem $libnatpmp_prefix/include"
else
CPPFLAGS="$CPPFLAGS -I$libnatpmp_prefix/include"
fi
LDFLAGS="$LDFLAGS -L$libnatpmp_prefix/lib"
fi
;;
esac
fi
else
case $build_os in
*darwin*)
BUILD_OS=darwin
;;
*)
AC_PATH_TOOL([DSYMUTIL], [dsymutil], dsymutil)
AC_PATH_TOOL([INSTALLNAMETOOL], [install_name_tool], install_name_tool)
AC_PATH_TOOL([OTOOL], [otool], otool)
AC_PATH_PROGS([XORRISOFS], [xorrisofs], xorrisofs)
AC_PATH_PROGS([DMG], [dmg], dmg)
dnl libtool will try to strip the static lib, which is a problem for
dnl cross-builds because strip attempts to call a hard-coded ld,
dnl which may not exist in the path. Stripping the .a is not
dnl necessary, so just disable it.
old_striplib=
;;
esac
fi
AX_CHECK_LINK_FLAG([[-Wl,-headerpad_max_install_names]], [LDFLAGS="$LDFLAGS -Wl,-headerpad_max_install_names"],, [[$LDFLAG_WERROR]])
CPPFLAGS="$CPPFLAGS -DMAC_OSX -DOBJC_OLD_DISPATCH_PROTOTYPES=0"
OBJCXXFLAGS="$CXXFLAGS"
;;
*android*)
dnl make sure android stays above linux for hosts like *linux-android*
TARGET_OS=android
case $host in
*x86_64*)
ANDROID_ARCH=x86_64
;;
*aarch64*)
ANDROID_ARCH=arm64-v8a
;;
*armv7a*)
ANDROID_ARCH=armeabi-v7a
;;
*) AC_MSG_ERROR("Could not determine Android arch") ;;
esac
;;
*linux*)
TARGET_OS=linux
;;
esac
if test x$use_extended_functional_tests != xno; then
AC_SUBST(EXTENDED_FUNCTIONAL_TESTS, --extended)
fi
if test x$use_lcov = xyes; then
if test x$LCOV = x; then
AC_MSG_ERROR("lcov testing requested but lcov not found")
fi
if test x$PYTHON = x; then
AC_MSG_ERROR("lcov testing requested but python not found")
fi
if test x$GENHTML = x; then
AC_MSG_ERROR("lcov testing requested but genhtml not found")
fi
AC_MSG_CHECKING([whether compiler is Clang])
AC_PREPROC_IFELSE([AC_LANG_SOURCE([[
#if defined(__clang__) && defined(__llvm__)
// Compiler is Clang
#else
# error Compiler is not Clang
#endif
]])],[
AC_MSG_RESULT([yes])
if test x$LLVM_COV = x; then
AC_MSG_ERROR([lcov testing requested but llvm-cov not found])
fi
COV_TOOL="$LLVM_COV gcov"
],[
AC_MSG_RESULT([no])
if test x$GCOV = x; then
AC_MSG_ERROR([lcov testing requested but gcov not found])
fi
COV_TOOL="$GCOV"
])
AC_SUBST(COV_TOOL)
AC_SUBST(COV_TOOL_WRAPPER, "cov_tool_wrapper.sh")
LCOV="$LCOV --gcov-tool $(pwd)/$COV_TOOL_WRAPPER"
AX_CHECK_LINK_FLAG([[--coverage]], [LDFLAGS="$LDFLAGS --coverage"],
[AC_MSG_ERROR("lcov testing requested but --coverage linker flag does not work")])
AX_CHECK_COMPILE_FLAG([--coverage],[CXXFLAGS="$CXXFLAGS --coverage"],
[AC_MSG_ERROR("lcov testing requested but --coverage flag does not work")])
CXXFLAGS="$CXXFLAGS -Og"
fi
if test x$use_lcov_branch != xno; then
AC_SUBST(LCOV_OPTS, "$LCOV_OPTS --rc lcov_branch_coverage=1")
fi
dnl Check for endianness
AC_C_BIGENDIAN
dnl Check for pthread compile/link requirements
AX_PTHREAD
dnl The following macro will add the necessary defines to bitcoin-config.h, but
dnl they also need to be passed down to any subprojects. Pull the results out of
dnl the cache and add them to CPPFLAGS.
AC_SYS_LARGEFILE
dnl detect POSIX or GNU variant of strerror_r
AC_FUNC_STRERROR_R
if test x$ac_cv_sys_file_offset_bits != x &&
test x$ac_cv_sys_file_offset_bits != xno &&
test x$ac_cv_sys_file_offset_bits != xunknown; then
CPPFLAGS="$CPPFLAGS -D_FILE_OFFSET_BITS=$ac_cv_sys_file_offset_bits"
fi
if test x$ac_cv_sys_large_files != x &&
test x$ac_cv_sys_large_files != xno &&
test x$ac_cv_sys_large_files != xunknown; then
CPPFLAGS="$CPPFLAGS -D_LARGE_FILES=$ac_cv_sys_large_files"
fi
2014-10-14 23:31:16 +02:00
AX_GCC_FUNC_ATTRIBUTE([visibility])
AX_GCC_FUNC_ATTRIBUTE([dllexport])
AX_GCC_FUNC_ATTRIBUTE([dllimport])
AC_SEARCH_LIBS([clock_gettime],[rt])
if test "x$enable_gprof" = xyes; then
dnl -pg is incompatible with -pie. Since hardening and profiling together doesn't make sense,
dnl we simply make them mutually exclusive here. Additionally, hardened toolchains may force
dnl -pie by default, in which case it needs to be turned off with -no-pie.
if test x$use_hardening = xyes; then
AC_MSG_ERROR(gprof profiling is not compatible with hardening. Reconfigure with --disable-hardening or --disable-gprof)
fi
use_hardening=no
AX_CHECK_COMPILE_FLAG([-pg],[GPROF_CXXFLAGS="-pg"],
[AC_MSG_ERROR(gprof profiling requested but not available)], [[$CXXFLAG_WERROR]])
AX_CHECK_LINK_FLAG([[-no-pie]], [GPROF_LDFLAGS="-no-pie"])
AX_CHECK_LINK_FLAG([[-pg]],[GPROF_LDFLAGS="$GPROF_LDFLAGS -pg"],
[AC_MSG_ERROR(gprof profiling requested but not available)], [[$GPROF_LDFLAGS]])
fi
if test x$TARGET_OS != xwindows; then
dnl All windows code is PIC, forcing it on just adds useless compile warnings
AX_CHECK_COMPILE_FLAG([-fPIC],[PIC_FLAGS="-fPIC"])
fi
dnl All versions of gcc that we commonly use for building are subject to bug
dnl https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90348. To work around that, set
dnl -fstack-reuse=none for all gcc builds. (Only gcc understands this flag)
AX_CHECK_COMPILE_FLAG([-fstack-reuse=none],[HARDENED_CXXFLAGS="$HARDENED_CXXFLAGS -fstack-reuse=none"])
if test x$use_hardening != xno; then
use_hardening=yes
AX_CHECK_COMPILE_FLAG([-Wstack-protector],[HARDENED_CXXFLAGS="$HARDENED_CXXFLAGS -Wstack-protector"])
AX_CHECK_COMPILE_FLAG([-fstack-protector-all],[HARDENED_CXXFLAGS="$HARDENED_CXXFLAGS -fstack-protector-all"])
Merge #20720: build: more robustly check for fcf-protection support e9189a750b237eba1befc6b16c12c2cee3e0176c build: more robustly check for fcf-protection support (fanquake) Pull request description: When using Clang 7, we may end up trying to use the flag when it won't work properly, which can lead to confusing errors. i.e: ```bash /usr/bin/ld: error: ... <corrupt x86 feature size: 0x8> ``` Use `AX_CHECK_LINK_FLAG` & `--fatal-warnings` to ensure we wont use the flag in this case. We do this as even when the error is emitted, compilation succeeds, and the binaries produced will run. This means we can't just check if the compiler accepts the flag, or if compilation succeeds (without or without `-Werror`, and/or passing `-Wl,--fatal-warnings`, which may not be passed through to the linker). This was reported by someone configuring for fuzzing, on Debian 10, where Clang 7 is the default. See here for a minimal example of the problematic behaviour: https://gist.github.com/fanquake/9b33555fcfebef8eb8c0795a71732bc6 ACKs for top commit: pstratem: tested ACK e9189a750b237eba1befc6b16c12c2cee3e0176c MarcoFalke: not an ACK e9189a750b237eba1befc6b16c12c2cee3e0176c , I only tested configure on my system (gcc-10, clang-11): hebasto: ACK e9189a750b237eba1befc6b16c12c2cee3e0176c, tested with clang-7, clang-10 and gcc: the `-fcf-protection=full` is not applied for clang-7, but applied for others compilers. Tree-SHA512: ec24b0cc5523b90139c96cbb33bb98d1e6a24d858c466aa7dfb3c474caf8c50aca53e570fdbc0ff88378406b0ac5d687542452637b1b5fa062e829291b886fc1
2021-02-08 06:33:59 +01:00
dnl -fcf-protection used with Clang 7 causes ld to emit warnings:
dnl ld: error: ... <corrupt x86 feature size: 0x8>
dnl Use CHECK_LINK_FLAG & --fatal-warnings to ensure we wont use the flag in this case.
AX_CHECK_LINK_FLAG([-fcf-protection=full],[HARDENED_CXXFLAGS="$HARDENED_CXXFLAGS -fcf-protection=full"],, [[$LDFLAG_WERROR]])
Merge #18921: build: add stack-clash and control-flow protection options to hardening flags b536813cefc13f5c54a28a7c2fce8c69e89d6624 build: add -fstack-clash-protection to hardening flags (fanquake) 076183b36b76a11438463883ff916f17aef9e001 build: add -fcf-protection=full to hardening options (fanquake) Pull request description: Beginning with Ubuntu `19.10`, it's packaged GCC now has some additional hardening options enabled by default (in addition to existing defaults like `-fstack-protector-strong` and reducing the minimum ssp buffer size). The new additions are`-fcf-protection=full` and `-fstack-clash-protection`. > -fcf-protection=[full|branch|return|none] > Enable code instrumentation of control-flow transfers to increase program security by checking that target addresses of control-flow transfer instructions (such as indirect function call, function return, indirect jump) are valid. This prevents diverting the flow of control to an unexpected target. This is intended to protect against such threats as Return-oriented Programming (ROP), and similarly call/jmp-oriented programming (COP/JOP). > -fstack-clash-protection > Generate code to prevent stack clash style attacks. When this option is enabled, the compiler will only allocate one page of stack space at a time and each page is accessed immediately after allocation. Thus, it prevents allocations from jumping over any stack guard page provided by the operating system. If your interested you can grab `gcc-9_9.3.0-10ubuntu2.debian.tar.xz` from https://packages.ubuntu.com/focal/g++-9. The relevant changes are part of the `gcc-distro-specs` patches, along with the relevant additions to the gcc manages: > NOTE: In Ubuntu 19.10 and later versions, -fcf-protection is enabled by default for C, C++, ObjC, ObjC++, if none of -fno-cf-protection nor -fcf-protection=* are found. > NOTE: In Ubuntu 19.10 and later versions, -fstack-clash-protection is enabled by default for C, C++, ObjC, ObjC++, unless -fno-stack-clash-protection is found. So, if you're C++ using GCC on Ubuntu 19.10 or later, these options will be active unless you explicitly opt out. This can be observed with a small test: ```c++ int main() { return 0; } ``` ```bash g++ --version g++ (Ubuntu 9.3.0-10ubuntu2) 9.3.0 g++ test.cpp objdump -dC a.out .. 0000000000001129 <main>: 1129: f3 0f 1e fa endbr64 112d: 55 push %rbp 112e: 48 89 e5 mov %rsp,%rbp 1131: b8 00 00 00 00 mov $0x0,%eax 1136: 5d pop %rbp 1137: c3 retq 1138: 0f 1f 84 00 00 00 00 nopl 0x0(%rax,%rax,1) 113f: 00 # recompile opting out of control flow protection g++ test.cpp -fcf-protection=none objdump -dC a.out ... 0000000000001129 <main>: 1129: 55 push %rbp 112a: 48 89 e5 mov %rsp,%rbp 112d: b8 00 00 00 00 mov $0x0,%eax 1132: 5d pop %rbp 1133: c3 retq 1134: 66 2e 0f 1f 84 00 00 nopw %cs:0x0(%rax,%rax,1) 113b: 00 00 00 113e: 66 90 xchg %ax,%ax ``` Note the insertion of an `endbr64` instruction when compiling and _not_ opting out. This instruction is part of the Intel Control-flow Enforcement Technology [spec](https://software.intel.com/sites/default/files/managed/4d/2a/control-flow-enforcement-technology-preview.pdf), which the GCC control flow implementation is based on. If we're still doing gitian builds for the `0.21.0` and `0.22.0` releases, we'd likely update the gitian image to Ubuntu Focal, which would mean that the GCC used for gitian builds would also be using these options by default. So we should decide whether we want to explicitly turn these options on as part of our hardening options (although not just for this reason), or, we should be opting-out. GCC has supported both options since 8.0.0. Clang has supported `-fcf-protection` from 7.0.0 and will support `-fstack-clash-protection` in it's upcoming [11.0.0 release](https://clang.llvm.org/docs/ReleaseNotes.html#id6). ACKs for top commit: jamesob: ACK b536813cefc13f5c54a28a7c2fce8c69e89d6624 ([`jamesob/ackr/18921.1.fanquake.build_add_stack_clash_an`](https://github.com/jamesob/bitcoin/tree/ackr/18921.1.fanquake.build_add_stack_clash_an)) laanwj: Code review ACK b536813cefc13f5c54a28a7c2fce8c69e89d6624 Tree-SHA512: abc9adf23cdf1be384f5fb9aa5bfffdda86b9ecd671064298d4cda0440828b509f070f9b19c88c7ce50ead9ff32afff9f14c5e78d75f01241568fbfa077be0b7
2020-08-29 13:40:38 +02:00
case $host in
*mingw*)
dnl stack-clash-protection doesn't currently work, and likely should just be skipped for Windows.
dnl See https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90458 for more details.
;;
*)
Merge bitcoin/bitcoin#21788: build: Silence [-Wunused-command-line-argument] warnings e9f948c72790136656df6056fd9e3698f360e077 build: Convert warnings into errors when testing for -fstack-clash-protection (Hennadii Stepanov) Pull request description: Apple clang version 12.0.5 (clang-1205.0.22.9) that is a part of Xcode 12.5, and is based on LLVM clang 11.1.0, fires spammy warnings: ``` clang: warning: argument unused during compilation: '-fstack-clash-protection' [-Wunused-command-line-argument] ``` From the https://github.com/apple/llvm-project: ``` $ git log --oneline | grep 'stack-clash-protection' 00065d5cbd02 Revert "-fstack-clash-protection: Return an actual error when used on unsupported OS" 4d59c8fdb955 -fstack-clash-protection: Return an actual error when used on unsupported OS df3bfaa39071 [Driver] Change -fnostack-clash-protection to -fno-stack-clash-protection 68e07da3e5d5 [clang][PowerPC] Enable -fstack-clash-protection option for ppc64 515bfc66eace [SystemZ] Implement -fstack-clash-protection e67cbac81211 Support -fstack-clash-protection for x86 454621160066 Revert "Support -fstack-clash-protection for x86" 0fd51a4554f5 Support -fstack-clash-protection for x86 658495e6ecd4 Revert "Support -fstack-clash-protection for x86" e229017732bc Support -fstack-clash-protection for x86 b03c3d8c6209 Revert "Support -fstack-clash-protection for x86" 4a1a0690ad68 Support -fstack-clash-protection for x86 f6d98429fcdb Revert "Support -fstack-clash-protection for x86" 39f50da2a357 Support -fstack-clash-protection for x86 ``` I suppose, that Apple clang-1205.0.22.9 ends with on of the "Revert..." commits. This PR prevents using of the `-fstack-clash-protection` flag if it causes warnings. --- System: macOS Big Sur 11.3 (20E232). ACKs for top commit: jarolrod: re-ACK e9f948c72790136656df6056fd9e3698f360e077 Sjors: tACK e9f948c72790136656df6056fd9e3698f360e077 on macOS 11.3.1 Tree-SHA512: 30186da67f9b0f34418014860c766c2e7f622405520f1cbbc1095d4aa4038b0a86014d76076f318a4b1b09170a96d8167c21d7f53a760e26017f486e1a7d39d4
2021-05-25 10:02:57 +02:00
AX_CHECK_COMPILE_FLAG([-fstack-clash-protection], [HARDENED_CXXFLAGS="$HARDENED_CXXFLAGS -fstack-clash-protection"], [], [$CXXFLAG_WERROR])
;;
esac
Merge #18921: build: add stack-clash and control-flow protection options to hardening flags b536813cefc13f5c54a28a7c2fce8c69e89d6624 build: add -fstack-clash-protection to hardening flags (fanquake) 076183b36b76a11438463883ff916f17aef9e001 build: add -fcf-protection=full to hardening options (fanquake) Pull request description: Beginning with Ubuntu `19.10`, it's packaged GCC now has some additional hardening options enabled by default (in addition to existing defaults like `-fstack-protector-strong` and reducing the minimum ssp buffer size). The new additions are`-fcf-protection=full` and `-fstack-clash-protection`. > -fcf-protection=[full|branch|return|none] > Enable code instrumentation of control-flow transfers to increase program security by checking that target addresses of control-flow transfer instructions (such as indirect function call, function return, indirect jump) are valid. This prevents diverting the flow of control to an unexpected target. This is intended to protect against such threats as Return-oriented Programming (ROP), and similarly call/jmp-oriented programming (COP/JOP). > -fstack-clash-protection > Generate code to prevent stack clash style attacks. When this option is enabled, the compiler will only allocate one page of stack space at a time and each page is accessed immediately after allocation. Thus, it prevents allocations from jumping over any stack guard page provided by the operating system. If your interested you can grab `gcc-9_9.3.0-10ubuntu2.debian.tar.xz` from https://packages.ubuntu.com/focal/g++-9. The relevant changes are part of the `gcc-distro-specs` patches, along with the relevant additions to the gcc manages: > NOTE: In Ubuntu 19.10 and later versions, -fcf-protection is enabled by default for C, C++, ObjC, ObjC++, if none of -fno-cf-protection nor -fcf-protection=* are found. > NOTE: In Ubuntu 19.10 and later versions, -fstack-clash-protection is enabled by default for C, C++, ObjC, ObjC++, unless -fno-stack-clash-protection is found. So, if you're C++ using GCC on Ubuntu 19.10 or later, these options will be active unless you explicitly opt out. This can be observed with a small test: ```c++ int main() { return 0; } ``` ```bash g++ --version g++ (Ubuntu 9.3.0-10ubuntu2) 9.3.0 g++ test.cpp objdump -dC a.out .. 0000000000001129 <main>: 1129: f3 0f 1e fa endbr64 112d: 55 push %rbp 112e: 48 89 e5 mov %rsp,%rbp 1131: b8 00 00 00 00 mov $0x0,%eax 1136: 5d pop %rbp 1137: c3 retq 1138: 0f 1f 84 00 00 00 00 nopl 0x0(%rax,%rax,1) 113f: 00 # recompile opting out of control flow protection g++ test.cpp -fcf-protection=none objdump -dC a.out ... 0000000000001129 <main>: 1129: 55 push %rbp 112a: 48 89 e5 mov %rsp,%rbp 112d: b8 00 00 00 00 mov $0x0,%eax 1132: 5d pop %rbp 1133: c3 retq 1134: 66 2e 0f 1f 84 00 00 nopw %cs:0x0(%rax,%rax,1) 113b: 00 00 00 113e: 66 90 xchg %ax,%ax ``` Note the insertion of an `endbr64` instruction when compiling and _not_ opting out. This instruction is part of the Intel Control-flow Enforcement Technology [spec](https://software.intel.com/sites/default/files/managed/4d/2a/control-flow-enforcement-technology-preview.pdf), which the GCC control flow implementation is based on. If we're still doing gitian builds for the `0.21.0` and `0.22.0` releases, we'd likely update the gitian image to Ubuntu Focal, which would mean that the GCC used for gitian builds would also be using these options by default. So we should decide whether we want to explicitly turn these options on as part of our hardening options (although not just for this reason), or, we should be opting-out. GCC has supported both options since 8.0.0. Clang has supported `-fcf-protection` from 7.0.0 and will support `-fstack-clash-protection` in it's upcoming [11.0.0 release](https://clang.llvm.org/docs/ReleaseNotes.html#id6). ACKs for top commit: jamesob: ACK b536813cefc13f5c54a28a7c2fce8c69e89d6624 ([`jamesob/ackr/18921.1.fanquake.build_add_stack_clash_an`](https://github.com/jamesob/bitcoin/tree/ackr/18921.1.fanquake.build_add_stack_clash_an)) laanwj: Code review ACK b536813cefc13f5c54a28a7c2fce8c69e89d6624 Tree-SHA512: abc9adf23cdf1be384f5fb9aa5bfffdda86b9ecd671064298d4cda0440828b509f070f9b19c88c7ce50ead9ff32afff9f14c5e78d75f01241568fbfa077be0b7
2020-08-29 13:40:38 +02:00
dnl When enable_debug is yes, all optimizations are disabled.
dnl However, FORTIFY_SOURCE requires that there is some level of optimization, otherwise it does nothing and just creates a compiler warning.
dnl Since FORTIFY_SOURCE is a no-op without optimizations, do not enable it when enable_debug is yes.
if test x$enable_debug != xyes; then
AX_CHECK_PREPROC_FLAG([-D_FORTIFY_SOURCE=2],[
AX_CHECK_PREPROC_FLAG([-U_FORTIFY_SOURCE],[
HARDENED_CPPFLAGS="$HARDENED_CPPFLAGS -U_FORTIFY_SOURCE"
])
HARDENED_CPPFLAGS="$HARDENED_CPPFLAGS -D_FORTIFY_SOURCE=2"
])
fi
AX_CHECK_LINK_FLAG([[-Wl,--enable-reloc-section]], [HARDENED_LDFLAGS="$HARDENED_LDFLAGS -Wl,--enable-reloc-section"],, [[$LDFLAG_WERROR]])
AX_CHECK_LINK_FLAG([[-Wl,--dynamicbase]], [HARDENED_LDFLAGS="$HARDENED_LDFLAGS -Wl,--dynamicbase"],, [[$LDFLAG_WERROR]])
AX_CHECK_LINK_FLAG([[-Wl,--nxcompat]], [HARDENED_LDFLAGS="$HARDENED_LDFLAGS -Wl,--nxcompat"],, [[$LDFLAG_WERROR]])
AX_CHECK_LINK_FLAG([[-Wl,--high-entropy-va]], [HARDENED_LDFLAGS="$HARDENED_LDFLAGS -Wl,--high-entropy-va"],, [[$LDFLAG_WERROR]])
AX_CHECK_LINK_FLAG([[-Wl,-z,relro]], [HARDENED_LDFLAGS="$HARDENED_LDFLAGS -Wl,-z,relro"],, [[$LDFLAG_WERROR]])
AX_CHECK_LINK_FLAG([[-Wl,-z,now]], [HARDENED_LDFLAGS="$HARDENED_LDFLAGS -Wl,-z,now"],, [[$LDFLAG_WERROR]])
AX_CHECK_LINK_FLAG([[-Wl,-z,separate-code]], [HARDENED_LDFLAGS="$HARDENED_LDFLAGS -Wl,-z,separate-code"],, [[$LDFLAG_WERROR]])
AX_CHECK_LINK_FLAG([[-fPIE -pie]], [PIE_FLAGS="-fPIE"; HARDENED_LDFLAGS="$HARDENED_LDFLAGS -pie"],, [[$CXXFLAG_WERROR]])
case $host in
*mingw*)
AC_CHECK_LIB([ssp], [main],, AC_MSG_ERROR(libssp missing))
;;
esac
fi
dnl These flags are specific to ld64, and may cause issues with other linkers.
dnl For example: GNU ld will interpret -dead_strip as -de and then try and use
dnl "ad_strip" as the symbol for the entry point.
if test x$TARGET_OS = xdarwin; then
AX_CHECK_LINK_FLAG([[-Wl,-dead_strip]], [LDFLAGS="$LDFLAGS -Wl,-dead_strip"],, [[$LDFLAG_WERROR]])
AX_CHECK_LINK_FLAG([[-Wl,-dead_strip_dylibs]], [LDFLAGS="$LDFLAGS -Wl,-dead_strip_dylibs"],, [[$LDFLAG_WERROR]])
AX_CHECK_LINK_FLAG([[-Wl,-bind_at_load]], [HARDENED_LDFLAGS="$HARDENED_LDFLAGS -Wl,-bind_at_load"],, [[$LDFLAG_WERROR]])
fi
AC_CHECK_HEADERS([endian.h sys/endian.h byteswap.h stdio.h stdlib.h unistd.h strings.h sys/types.h sys/stat.h sys/select.h sys/prctl.h sys/sysctl.h vm/vm_param.h sys/vmmeter.h sys/resources.h])
AC_CHECK_DECLS([getifaddrs, freeifaddrs],[CHECK_SOCKET],,
[#include <sys/types.h>
#include <ifaddrs.h>]
)
AC_CHECK_DECLS([strnlen])
dnl Check for daemon(3), unrelated to --with-daemon (although used by it)
AC_CHECK_DECLS([daemon])
Merge #21250: build: make HAVE_O_CLOEXEC available outside LevelDB (bugfix) 9bac71350d98580cc7441957fc7c3fa2f4158553 build: make HAVE_O_CLOEXEC available outside LevelDB (bugfix) (Sebastian Falbesoner) 584fd91d2d294883e6896dbd64a2176528e94581 init: only use pipe2 if availabile, check in configure (Sebastian Falbesoner) Pull request description: The result of the O_CLOEXEC availability check is currently only set in the Makefile and passed to LevelDB (see `LEVELDB_CPPFLAGS_INT` in `src/Makefile.leveldb.include`), but not defined to be used in our codebase. This means that code within the preprocessor conditional `#if HAVE_O_CLOEXEC` was actually never compiled. On the master branch this is currently used for pipe creation in `src/shutdown.cpp`, PR #21007 moves this part to a new module (I found the issue while testing that PR). The fix is similar to the one in #19803, which solved the same problem for HAVE_FDATASYNC. In the course of working on the PR it turned out that pipe2 is not available an all platforms, hence a configure check and a corresponding define HAVE_PIPE2 is introduced and used. The PR can be tested by anyone with a system that has pipe2 and O_CLOEXEC available by putting gibberish into the HAVE_O_CLOEXEC block: on master, everything should compile fine, on PR, the compiler should abort with an error. At least that's my naive way of testing preprocessor logic, happy to hear more sophisticated ways :-) ACKs for top commit: laanwj: Code review ACK 9bac71350d98580cc7441957fc7c3fa2f4158553 Tree-SHA512: aec89faf6ba52b6f014c610ebef7b725d9e967207d58b42a4a71afc9f1268fcb673ecc85b33a2a3debba8105a304dd7edaba4208c5373fcef2ab83e48a170051
2021-02-23 18:51:13 +01:00
AC_CHECK_DECLS([pipe2])
AC_CHECK_DECLS([le16toh, le32toh, le64toh, htole16, htole32, htole64, be16toh, be32toh, be64toh, htobe16, htobe32, htobe64],,,
[#if HAVE_ENDIAN_H
#include <endian.h>
#elif HAVE_SYS_ENDIAN_H
#include <sys/endian.h>
#endif])
AC_CHECK_DECLS([bswap_16, bswap_32, bswap_64],,,
[#if HAVE_BYTESWAP_H
#include <byteswap.h>
#endif])
AC_CHECK_DECLS([__builtin_clz, __builtin_clzl, __builtin_clzll])
dnl Check for mallopt(M_ARENA_MAX) (to set glibc arenas)
AC_MSG_CHECKING(for mallopt M_ARENA_MAX)
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <malloc.h>]],
[[ mallopt(M_ARENA_MAX, 1); ]])],
[ AC_MSG_RESULT(yes); AC_DEFINE(HAVE_MALLOPT_ARENA_MAX, 1,[Define this symbol if you have mallopt with M_ARENA_MAX]) ],
[ AC_MSG_RESULT(no)]
)
dnl Check for malloc_info (for memory statistics information in getmemoryinfo)
AC_MSG_CHECKING(for getmemoryinfo)
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <malloc.h>]],
[[ int f = malloc_info(0, NULL); ]])],
[ AC_MSG_RESULT(yes); AC_DEFINE(HAVE_MALLOC_INFO, 1,[Define this symbol if you have malloc_info]) ],
[ AC_MSG_RESULT(no)]
)
dnl Check for posix_fallocate
AC_MSG_CHECKING(for posix_fallocate)
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[
// same as in src/util/system.cpp
#ifdef __linux__
#ifdef _POSIX_C_SOURCE
#undef _POSIX_C_SOURCE
#endif
#define _POSIX_C_SOURCE 200112L
#endif // __linux__
#include <fcntl.h>]],
[[ int f = posix_fallocate(0, 0, 0); ]])],
[ AC_MSG_RESULT(yes); AC_DEFINE(HAVE_POSIX_FALLOCATE, 1,[Define this symbol if you have posix_fallocate]) ],
[ AC_MSG_RESULT(no)]
)
AC_MSG_CHECKING([for visibility attribute])
AC_LINK_IFELSE([AC_LANG_SOURCE([
int foo_def( void ) __attribute__((visibility("default")));
int main(){}
])],
[
AC_DEFINE(HAVE_VISIBILITY_ATTRIBUTE,1,[Define if the visibility attribute is supported.])
AC_MSG_RESULT(yes)
],
[
AC_MSG_RESULT(no)
if test x$use_reduce_exports = xyes; then
2015-02-24 02:18:54 +01:00
AC_MSG_ERROR([Cannot find a working visibility attribute. Use --disable-reduce-exports.])
fi
]
)
if test "x$use_thread_local" = xyes || test "x$use_thread_local" = xauto; then
TEMP_LDFLAGS="$LDFLAGS"
LDFLAGS="$TEMP_LDFLAGS $PTHREAD_CFLAGS"
AC_MSG_CHECKING([for thread_local support])
AC_LINK_IFELSE([AC_LANG_SOURCE([
#include <thread>
static thread_local int foo = 0;
static void run_thread() { foo++;}
int main(){
for(int i = 0; i < 10; i++) { std::thread(run_thread).detach();}
return foo;
}
])],
[
case $host in
*mingw*)
dnl mingw32's implementation of thread_local has also been shown to behave
dnl erroneously under concurrent usage; see:
dnl https://gist.github.com/jamesob/fe9a872051a88b2025b1aa37bfa98605
AC_MSG_RESULT(no)
;;
*freebsd*)
dnl FreeBSD's implementation of thread_local is also buggy (per
dnl https://groups.google.com/d/msg/bsdmailinglist/22ncTZAbDp4/Dii_pII5AwAJ)
AC_MSG_RESULT(no)
;;
*)
AC_DEFINE(HAVE_THREAD_LOCAL,1,[Define if thread_local is supported.])
AC_MSG_RESULT(yes)
;;
esac
],
[
AC_MSG_RESULT(no)
]
)
LDFLAGS="$TEMP_LDFLAGS"
fi
dnl check for gmtime_r(), fallback to gmtime_s() if that is unavailable
dnl fail if neither are available.
AC_MSG_CHECKING(for gmtime_r)
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <ctime>]],
[[ gmtime_r((const time_t *) nullptr, (struct tm *) nullptr); ]])],
[ AC_MSG_RESULT(yes); AC_DEFINE(HAVE_GMTIME_R, 1, [Define this symbol if gmtime_r is available]) ],
[ AC_MSG_RESULT(no);
AC_MSG_CHECKING(for gmtime_s);
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <ctime>]],
[[ gmtime_s((struct tm *) nullptr, (const time_t *) nullptr); ]])],
[ AC_MSG_RESULT(yes)],
[ AC_MSG_RESULT(no); AC_MSG_ERROR(Both gmtime_r and gmtime_s are unavailable) ]
)
]
)
dnl Check for different ways of gathering OS randomness
AC_MSG_CHECKING(for Linux getrandom syscall)
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <unistd.h>
#include <sys/syscall.h>
#include <linux/random.h>]],
[[ syscall(SYS_getrandom, nullptr, 32, 0); ]])],
[ AC_MSG_RESULT(yes); AC_DEFINE(HAVE_SYS_GETRANDOM, 1,[Define this symbol if the Linux getrandom system call is available]) ],
[ AC_MSG_RESULT(no)]
)
AC_MSG_CHECKING(for getentropy)
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <unistd.h>]],
[[ getentropy(nullptr, 32) ]])],
[ AC_MSG_RESULT(yes); AC_DEFINE(HAVE_GETENTROPY, 1,[Define this symbol if the BSD getentropy system call is available]) ],
[ AC_MSG_RESULT(no)]
)
AC_MSG_CHECKING(for getentropy via random.h)
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <unistd.h>
#include <sys/random.h>]],
[[ getentropy(nullptr, 32) ]])],
[ AC_MSG_RESULT(yes); AC_DEFINE(HAVE_GETENTROPY_RAND, 1,[Define this symbol if the BSD getentropy system call is available with sys/random.h]) ],
[ AC_MSG_RESULT(no)]
)
AC_MSG_CHECKING(for sysctl)
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <sys/types.h>
#include <sys/sysctl.h>]],
[[ #ifdef __linux__
#error "Don't use sysctl on Linux, it's deprecated even when it works"
#endif
sysctl(nullptr, 2, nullptr, nullptr, nullptr, 0); ]])],
[ AC_MSG_RESULT(yes); AC_DEFINE(HAVE_SYSCTL, 1,[Define this symbol if the BSD sysctl() is available]) ],
[ AC_MSG_RESULT(no)]
)
AC_MSG_CHECKING(for sysctl KERN_ARND)
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <sys/types.h>
#include <sys/sysctl.h>]],
[[ #ifdef __linux__
#error "Don't use sysctl on Linux, it's deprecated even when it works"
#endif
static int name[2] = {CTL_KERN, KERN_ARND};
sysctl(name, 2, nullptr, nullptr, nullptr, 0); ]])],
[ AC_MSG_RESULT(yes); AC_DEFINE(HAVE_SYSCTL_ARND, 1,[Define this symbol if the BSD sysctl(KERN_ARND) is available]) ],
[ AC_MSG_RESULT(no)]
)
AC_MSG_CHECKING(for if type char equals int8_t)
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <stdint.h>
#include <type_traits>]],
[[ static_assert(std::is_same<int8_t, char>::value, ""); ]])],
[ AC_MSG_RESULT(yes); AC_DEFINE(CHAR_EQUALS_INT8, 1,[Define this symbol if type char equals int8_t]) ],
[ AC_MSG_RESULT(no)]
)
dnl ensure backtrace() is found, check -lexecinfo if necessary
if test x$TARGET_OS != xwindows; then
if test "x$enable_stacktraces" != xno; then
AC_SEARCH_LIBS([backtrace], [execinfo], [], [
AC_MSG_ERROR([Unable to find backtrace()])
])
fi
fi
AC_MSG_CHECKING(for fdatasync)
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <unistd.h>]],
[[ fdatasync(0); ]])],
[ AC_MSG_RESULT(yes); HAVE_FDATASYNC=1 ],
[ AC_MSG_RESULT(no); HAVE_FDATASYNC=0 ]
)
AC_DEFINE_UNQUOTED([HAVE_FDATASYNC], [$HAVE_FDATASYNC], [Define to 1 if fdatasync is available.])
AC_MSG_CHECKING(for F_FULLFSYNC)
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <fcntl.h>]],
[[ fcntl(0, F_FULLFSYNC, 0); ]])],
[ AC_MSG_RESULT(yes); HAVE_FULLFSYNC=1 ],
[ AC_MSG_RESULT(no); HAVE_FULLFSYNC=0 ]
)
AC_MSG_CHECKING(for O_CLOEXEC)
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <fcntl.h>]],
[[ open("", O_CLOEXEC); ]])],
[ AC_MSG_RESULT(yes); HAVE_O_CLOEXEC=1 ],
[ AC_MSG_RESULT(no); HAVE_O_CLOEXEC=0 ]
)
Merge #21250: build: make HAVE_O_CLOEXEC available outside LevelDB (bugfix) 9bac71350d98580cc7441957fc7c3fa2f4158553 build: make HAVE_O_CLOEXEC available outside LevelDB (bugfix) (Sebastian Falbesoner) 584fd91d2d294883e6896dbd64a2176528e94581 init: only use pipe2 if availabile, check in configure (Sebastian Falbesoner) Pull request description: The result of the O_CLOEXEC availability check is currently only set in the Makefile and passed to LevelDB (see `LEVELDB_CPPFLAGS_INT` in `src/Makefile.leveldb.include`), but not defined to be used in our codebase. This means that code within the preprocessor conditional `#if HAVE_O_CLOEXEC` was actually never compiled. On the master branch this is currently used for pipe creation in `src/shutdown.cpp`, PR #21007 moves this part to a new module (I found the issue while testing that PR). The fix is similar to the one in #19803, which solved the same problem for HAVE_FDATASYNC. In the course of working on the PR it turned out that pipe2 is not available an all platforms, hence a configure check and a corresponding define HAVE_PIPE2 is introduced and used. The PR can be tested by anyone with a system that has pipe2 and O_CLOEXEC available by putting gibberish into the HAVE_O_CLOEXEC block: on master, everything should compile fine, on PR, the compiler should abort with an error. At least that's my naive way of testing preprocessor logic, happy to hear more sophisticated ways :-) ACKs for top commit: laanwj: Code review ACK 9bac71350d98580cc7441957fc7c3fa2f4158553 Tree-SHA512: aec89faf6ba52b6f014c610ebef7b725d9e967207d58b42a4a71afc9f1268fcb673ecc85b33a2a3debba8105a304dd7edaba4208c5373fcef2ab83e48a170051
2021-02-23 18:51:13 +01:00
AC_DEFINE_UNQUOTED([HAVE_O_CLOEXEC], [$HAVE_O_CLOEXEC], [Define to 1 if O_CLOEXEC flag is available.])
dnl crc32c platform checks
AC_MSG_CHECKING(for __builtin_prefetch)
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[ ]], [[
char data = 0;
const char* address = &data;
__builtin_prefetch(address, 0, 0);
]])],
[ AC_MSG_RESULT(yes); HAVE_BUILTIN_PREFETCH=1 ],
[ AC_MSG_RESULT(no); HAVE_BUILTIN_PREFETCH=0 ]
)
AC_MSG_CHECKING(for _mm_prefetch)
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <xmmintrin.h>]], [[
char data = 0;
const char* address = &data;
_mm_prefetch(address, _MM_HINT_NTA);
]])],
[ AC_MSG_RESULT(yes); HAVE_MM_PREFETCH=1 ],
[ AC_MSG_RESULT(no); HAVE_MM_PREFETCH=0 ]
)
AC_MSG_CHECKING(for strong getauxval support in the system headers)
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[
#include <sys/auxv.h>
]], [[
getauxval(AT_HWCAP);
]])],
[ AC_MSG_RESULT(yes); HAVE_STRONG_GETAUXVAL=1; AC_DEFINE(HAVE_STRONG_GETAUXVAL, 1, [Define this symbol to build code that uses getauxval)]) ],
[ AC_MSG_RESULT(no); HAVE_STRONG_GETAUXVAL=0 ]
)
dnl Check for reduced exports
if test x$use_reduce_exports = xyes; then
AX_CHECK_COMPILE_FLAG([-fvisibility=hidden],[RE_CXXFLAGS="-fvisibility=hidden"],
2015-02-24 02:18:54 +01:00
[AC_MSG_ERROR([Cannot set default symbol visibility. Use --disable-reduce-exports.])])
fi
AC_MSG_CHECKING([for std::system])
AC_LINK_IFELSE(
[ AC_LANG_PROGRAM(
[[ #include <cstdlib> ]],
[[ int nErr = std::system(""); ]]
)],
[ AC_MSG_RESULT(yes); AC_DEFINE(HAVE_STD__SYSTEM, 1, Define to 1 if you have the `std::system' function.)],
[ AC_MSG_RESULT(no) ]
)
AC_MSG_CHECKING([for ::_wsystem])
AC_LINK_IFELSE(
[ AC_LANG_PROGRAM(
[[ ]],
[[ int nErr = ::_wsystem(""); ]]
)],
[ AC_MSG_RESULT(yes); AC_DEFINE(HAVE_WSYSTEM, 1, Define to 1 if you have the `::wsystem' function.)],
[ AC_MSG_RESULT(no) ]
)
# Define to 1 if std::system or ::wsystem (Windows) is available
AC_DEFINE([HAVE_SYSTEM], [HAVE_STD__SYSTEM || HAVE_WSYSTEM], [std::system or ::wsystem])
dnl SUPPRESSED_CPPFLAGS=SUPPRESS_WARNINGS([$SOME_CPPFLAGS])
dnl Replace -I with -isystem in $SOME_CPPFLAGS to suppress warnings from
dnl headers from its include directories and return the result.
dnl See -isystem documentation:
dnl https://gcc.gnu.org/onlinedocs/gcc/Directory-Options.html
dnl https://clang.llvm.org/docs/ClangCommandLineReference.html#cmdoption-clang-isystem-directory
dnl Do not change "-I/usr/include" to "-isystem /usr/include" because that
dnl is not necessary (/usr/include is already a system directory) and because
dnl it would break GCC's #include_next.
AC_DEFUN([SUPPRESS_WARNINGS],
[[$(echo $1 |${SED} -E -e 's/(^| )-I/\1-isystem /g' -e 's;-isystem /usr/include/*( |$);-I/usr/include\1;g')]])
dnl enable-fuzz should disable all other targets
if test "x$enable_fuzz" = "xyes"; then
AC_MSG_WARN(enable-fuzz will disable all other targets and force --enable-fuzz-binary=yes)
build_bitcoin_utils=no
build_bitcoin_cli=no
build_bitcoin_tx=no
build_bitcoin_wallet=no
build_bitcoind=no
build_bitcoin_libs=no
bitcoin_enable_qt=no
bitcoin_enable_qt_test=no
bitcoin_enable_qt_dbus=no
enable_wallet=no
use_bench=no
use_upnp=no
use_natpmp=no
use_zmq=no
enable_fuzz_binary=yes
AX_CHECK_PREPROC_FLAG([-DABORT_ON_FAILED_ASSUME],[[DEBUG_CPPFLAGS="$DEBUG_CPPFLAGS -DABORT_ON_FAILED_ASSUME"]],,[[$CXXFLAG_WERROR]])
AC_MSG_CHECKING([whether main function is needed for fuzz binary])
AX_CHECK_LINK_FLAG(
[[-fsanitize=$use_sanitizers]],
[AC_MSG_RESULT([no])],
[AC_MSG_RESULT([yes])
CPPFLAGS="$CPPFLAGS -DPROVIDE_FUZZ_MAIN_FUNCTION"],
[],
[AC_LANG_PROGRAM([[
#include <cstdint>
#include <cstddef>
extern "C" int LLVMFuzzerTestOneInput(const uint8_t *data, size_t size) { return 0; }
/* comment to remove the main function ...
]],[[
*/ int not_main() {
]])])
else
BITCOIN_QT_INIT
dnl sets $bitcoin_enable_qt, $bitcoin_enable_qt_test, $bitcoin_enable_qt_dbus
BITCOIN_QT_CONFIGURE([5.11.3])
dnl Keep a copy of the original $QT_INCLUDES and use it when invoking qt's moc
QT_INCLUDES_UNSUPPRESSED=$QT_INCLUDES
if test x$suppress_external_warnings != xno ; then
QT_INCLUDES=SUPPRESS_WARNINGS($QT_INCLUDES)
QT_DBUS_INCLUDES=SUPPRESS_WARNINGS($QT_DBUS_INCLUDES)
QT_TEST_INCLUDES=SUPPRESS_WARNINGS($QT_TEST_INCLUDES)
fi
CPPFLAGS="$CPPFLAGS -DPROVIDE_FUZZ_MAIN_FUNCTION"
fi
if test x$enable_wallet != xno; then
dnl Check for libdb_cxx only if wallet enabled
if test "x$use_bdb" != "xno"; then
BITCOIN_FIND_BDB48
if test x$suppress_external_warnings != xno ; then
BDB_CPPFLAGS=SUPPRESS_WARNINGS($BDB_CPPFLAGS)
fi
fi
dnl Check for sqlite3
if test "x$use_sqlite" != "xno"; then
PKG_CHECK_MODULES([SQLITE], [sqlite3 >= 3.7.17], [have_sqlite=yes], [have_sqlite=no])
fi
AC_MSG_CHECKING([whether to build wallet with support for sqlite])
if test "x$use_sqlite" = "xno"; then
use_sqlite=no
elif test "x$have_sqlite" = "xno"; then
if test "x$use_sqlite" = "xyes"; then
AC_MSG_ERROR([sqlite support requested but cannot be built. Use --without-sqlite])
fi
use_sqlite=no
else
if test x$use_sqlite != xno; then
AC_DEFINE([USE_SQLITE],[1],[Define if sqlite support should be compiled in])
use_sqlite=yes
fi
fi
AC_MSG_RESULT([$use_sqlite])
dnl Disable wallet if both --without-bdb and --without-sqlite
if test "x$use_bdb$use_sqlite" = "xnono"; then
if test "x$enable_wallet" = "xyes"; then
AC_MSG_ERROR([wallet functionality requested but no BDB or SQLite support available.])
fi
enable_wallet=no
fi
fi
if test x$use_ebpf != xno; then
Merge bitcoin/bitcoin#22238: build: improve detection of eBPF support 8f7704d0321a71c1691837a6bd3b4e05f84d3031 build: improve detection of eBPF support (fanquake) Pull request description: Just checking for the `sys/sdt.h` header isn't enough, as systems like macOS have the header, but it doesn't actually have the `DTRACE_PROBE*` probes, which leads to [compile failures](https://github.com/bitcoin/bitcoin/pull/22006#issuecomment-859559004). The contents of `sys/sdt.h` in the macOS SDK is: ```bash #ifndef _SYS_SDT_H #define _SYS_SDT_H /* * This is a wrapper header that wraps the mach visible sdt.h header so that * the header file ends up visible where software expects it to be. We also * do the C/C++ symbol wrapping here, since Mach headers are technically C * interfaces. * * Note: The process of adding USDT probes to code is slightly different * than documented in the "Solaris Dynamic Tracing Guide". * The DTRACE_PROBE*() macros are not supported on Mac OS X -- instead see * "BUILDING CODE CONTAINING USDT PROBES" in the dtrace(1) manpage * */ #include <sys/cdefs.h> __BEGIN_DECLS #include <mach/sdt.h> __END_DECLS #endif /* _SYS_SDT_H */ ``` The `BUILDING CODE CONTAINING USDT PROBES` section from the dtrace manpage is available [here](https://gist.github.com/fanquake/e56c9866d53b326646d04ab43a8df9e2), and outlines the more involved process of using USDT probes on macOS. ACKs for top commit: jb55: utACK 8f7704d0321a71c1691837a6bd3b4e05f84d3031 practicalswift: cr ACK 8f7704d0321a71c1691837a6bd3b4e05f84d3031 hebasto: ACK 8f7704d0321a71c1691837a6bd3b4e05f84d3031, tested on macOS Big Sur 11.4 (20F71) and on Linux Mint 20.1 (x86_64) with depends. Tree-SHA512: 5f1351d0ac2e655fccb22a5454f415906404fdaa336fd89b54ef49ca50a442c44ab92d063cba3f161cb8ea0679c92ae3cd6cfbbcb19728cac21116247a017df5
2021-06-18 09:15:33 +02:00
AC_MSG_CHECKING([whether eBPF tracepoints are supported])
AC_COMPILE_IFELSE([
AC_LANG_PROGRAM(
[#include <sys/sdt.h>],
[DTRACE_PROBE("context", "event");]
)],
[AC_MSG_RESULT(yes); have_sdt=yes; AC_DEFINE([ENABLE_TRACING], [1], [Define to 1 to enable eBPF user static defined tracepoints])],
[AC_MSG_RESULT(no); have_sdt=no;]
)
fi
dnl Check for libminiupnpc (optional)
if test x$use_upnp != xno; then
AC_CHECK_HEADERS(
[miniupnpc/miniupnpc.h miniupnpc/upnpcommands.h miniupnpc/upnperrors.h],
[AC_CHECK_LIB([miniupnpc], [upnpDiscover], [MINIUPNPC_LIBS=-lminiupnpc], [have_miniupnpc=no])],
[have_miniupnpc=no]
)
Merge #15993: net: Drop support of the insecure miniUPnPc versions 59cb722fd050393a69f1e0df97d857c893d19d80 Update configure to reject unsafe miniUPnPc API ver (Hennadii Stepanov) ab2190557ec2757fa48b52855b05561854af49af doc: Add release notes for 15993 (Hennadii Stepanov) 02709e95601c6020a87a6a05ee1d00c13fc38f9b Align formatting with clang-format (Hennadii Stepanov) 91a1b8508358d04685391651aea303ebce1c3d05 Use PACKAGE_NAME in UPnP description (Hennadii Stepanov) 9f76e45b9d6671e2074fb7a3885db703045a791f Drop support of insecure miniUPnPc versions (Hennadii Stepanov) Pull request description: 1. Minimum supported miniUPnPc API version is set to 10: - https://packages.ubuntu.com/xenial/libminiupnpc-dev - https://packages.debian.org/jessie/libminiupnpc-dev Refs: - #6583 - #6789 - #10414 2. The hardcoded "Bitcoin" replaced with `PACKAGE_NAME`: ![Screenshot from 2019-05-06 23-10-29](https://user-images.githubusercontent.com/32963518/57253178-afc60780-7056-11e9-83c9-e85670c58c1e.png) 3. Also style-only commit applied. Pardon: could not reopen my previous PR #15966. ACKs for top commit: ryanofsky: utACK 59cb722fd050393a69f1e0df97d857c893d19d80. Changes since last review: adding a new commit which updates configure script to fall back to disabling upnp if version is too old, adding a requested comment explaining static_assert condition, and fixing a spelling (jessy/jessie) Tree-SHA512: 42ed11bc2fb2ec83d5dd58e2383da5444a24fd572707f6cf10b622cb8943e28adfcca4750d06801024c4472625b5ea9279516fbd9d2ccebc9bbaafe1d148e80d
2019-07-29 16:41:54 +02:00
dnl The minimum supported miniUPnPc API version is set to 10. This keeps compatibility
dnl with Ubuntu 16.04 LTS and Debian 8 libminiupnpc-dev packages.
if test x$have_miniupnpc != xno; then
AC_MSG_CHECKING([whether miniUPnPc API version is supported])
AC_PREPROC_IFELSE([AC_LANG_PROGRAM([[
@%:@include <miniupnpc/miniupnpc.h>
]], [[
#if MINIUPNPC_API_VERSION >= 10
// Everything is okay
#else
# error miniUPnPc API version is too old
#endif
]])],[
AC_MSG_RESULT(yes)
],[
AC_MSG_RESULT(no)
AC_MSG_WARN([miniUPnPc API version < 10 is unsupported, disabling UPnP support.])
have_miniupnpc=no
])
fi
fi
dnl Check for libnatpmp (optional).
if test "x$use_natpmp" != xno; then
AC_CHECK_HEADERS([natpmp.h],
[AC_CHECK_LIB([natpmp], [initnatpmp], [NATPMP_LIBS=-lnatpmp], [have_natpmp=no])],
[have_natpmp=no])
fi
if test x$build_bitcoin_wallet$build_bitcoin_cli$build_bitcoin_tx$build_bitcoind$bitcoin_enable_qt$use_tests$use_bench = xnonononononono; then
use_boost=no
else
use_boost=yes
fi
if test x$use_boost = xyes; then
dnl Check for Boost headers
AX_BOOST_BASE([1.64.0],[],[AC_MSG_ERROR([Boost is not available!])])
if test x$want_boost = xno; then
AC_MSG_ERROR([[only libdashconsensus can be built without boost]])
fi
AX_BOOST_FILESYSTEM
AX_BOOST_THREAD
if test x$suppress_external_warnings != xno; then
Merge bitcoin/bitcoin#25436: build: GCC-12 build improvements 880d4aaf81f3d5d7fbb915905c2e61b816a6a747 build: use BOOST_NO_CXX98_FUNCTION_BASE to suppress warnings (fanquake) 1bdbbbdc46c4e50bf07bc362e7e391ea1a53ea2f build: suppress array-bounds errors in libxkbcommon (fanquake) Pull request description: 2 changes to better support building with GCC 12, which out of the box, is currently broken if you want to build using depends. Prevent `-Warray-bounds` errors when building libxkbcommon. i.e: ```bash src/xkbcomp/ast-build.c:82:27: error: array subscript 'ExprDef[0]' is partly outside array bounds of 'unsigned char[32]' [-Werror=array-bounds] 82 | expr->expr.value_type = type; | ~~~~~~~~~~~~~~~~~~~~~~^~~~~~ src/xkbcomp/ast-build.c:75:21: note: object of size 32 allocated by 'malloc' 75 | ExprDef *expr = malloc(size); | ^~~~~~~~~~~~ ``` It might be the case that these would be fixed by updating the package, but that would also require installing new build tools (meson), as well as potentially more dependencies (wayland), and it'd need testing with Qt. For now, just turn the errors into wanrings. Define `BOOST_NO_CXX98_FUNCTION_BASE` to prevent GCC warning about the use of `std::unary_function`. i.e: ```bash /bitcoin/depends/aarch64-unknown-linux-gnu/include/boost/container_hash/hash.hpp:131:33: warning: 'template<class _Arg, class _Result> struct std::unary_function' is deprecated [-Wdeprecated-declarations] 131 | struct hash_base : std::unary_function<T, std::size_t> {}; | ^~~~~~~~~~~~~~ In file included from /usr/include/c++/12/bits/unique_ptr.h:37, from /usr/include/c++/12/memory:76, from ./init.h:10, from init.cpp:10: /usr/include/c++/12/bits/stl_function.h:117:12: note: declared here 117 | struct unary_function ``` Boost `container_hash` (included via functional -> multi_index) uses [`std::unary_function`, which was deprecated in C++11](https://en.cppreference.com/w/cpp/utility/functional/unary_function), and "removed" in C++17. It's use causes warnings with newer compilers, i.e GCC 12.1. Use the MACRO outlined in https://github.com/boostorg/container_hash/issues/22, and added to Boost Config for GCC 12 in https://github.com/boostorg/config/pull/430, to prevent it's use. [BOOST_NO_CXX98_FUNCTION_BASE](https://www.boost.org/doc/libs/master/libs/config/doc/html/boost_config/boost_macro_reference.html): > The standard library no longer supports std::unary_function and std::binary_function. > They were deprecated in C++11 and is removed from C++14. Guix Build (x86_64): ```bash ``` Guix Build (arm64): ```bash ``` ACKs for top commit: laanwj: Code review ACK 880d4aaf81f3d5d7fbb915905c2e61b816a6a747 Tree-SHA512: 10c4679c3eb788e9279acc4960731c55ae1568bd3df525d3c46f97d8b0319e7d8450b1638b6777d98111b5991dba5c787e95d80b1ac932e0b4779d4b8e74875e
2022-06-27 11:12:57 +02:00
dnl Prevent use of std::unary_function, which was removed in C++17,
dnl and will generate warnings with newer compilers.
dnl See: https://github.com/boostorg/container_hash/issues/22.
BOOST_CPPFLAGS="$BOOST_CPPFLAGS -DBOOST_NO_CXX98_FUNCTION_BASE"
BOOST_CPPFLAGS=SUPPRESS_WARNINGS($BOOST_CPPFLAGS)
fi
BOOST_LIBS="$BOOST_LDFLAGS $BOOST_SYSTEM_LIB $BOOST_FILESYSTEM_LIB $BOOST_THREAD_LIB"
fi
if test x$use_reduce_exports = xyes; then
CXXFLAGS="$CXXFLAGS $RE_CXXFLAGS"
AX_CHECK_LINK_FLAG([[-Wl,--exclude-libs,ALL]], [RELDFLAGS="-Wl,--exclude-libs,ALL"],, [[$LDFLAG_WERROR]])
fi
if test x$use_tests = xyes; then
if test x$HEXDUMP = x; then
AC_MSG_ERROR(hexdump is required for tests)
fi
if test x$use_boost = xyes; then
AX_BOOST_UNIT_TEST_FRAMEWORK
dnl Determine if -DBOOST_TEST_DYN_LINK is needed
AC_MSG_CHECKING([for dynamic linked boost test])
TEMP_LIBS="$LIBS"
LIBS="$LIBS $BOOST_LDFLAGS $BOOST_UNIT_TEST_FRAMEWORK_LIB"
TEMP_CPPFLAGS="$CPPFLAGS"
CPPFLAGS="$CPPFLAGS $BOOST_CPPFLAGS"
AC_LINK_IFELSE([AC_LANG_SOURCE([
#define BOOST_TEST_DYN_LINK
#define BOOST_TEST_MAIN
#include <boost/test/unit_test.hpp>
])],
[AC_MSG_RESULT(yes)]
[TESTDEFS="$TESTDEFS -DBOOST_TEST_DYN_LINK"],
[AC_MSG_RESULT(no)])
LIBS="$TEMP_LIBS"
CPPFLAGS="$TEMP_CPPFLAGS"
fi
fi
dnl libevent check
if test x$build_bitcoin_cli$build_bitcoind$bitcoin_enable_qt$use_tests$use_bench != xnonononono; then
PKG_CHECK_MODULES([EVENT], [libevent >= 2.0.21], [use_libevent=yes], [AC_MSG_ERROR([libevent version 2.0.21 or greater not found.])])
if test x$TARGET_OS != xwindows; then
PKG_CHECK_MODULES([EVENT_PTHREADS], [libevent_pthreads >= 2.0.21],, [AC_MSG_ERROR([libevent_pthreads version 2.0.21 or greater not found.])])
fi
fi
dnl QR Code encoding library check
if test "x$use_qr" != xno; then
BITCOIN_QT_CHECK([PKG_CHECK_MODULES([QR], [libqrencode], [have_qrencode=yes], [have_qrencode=no])])
fi
dnl ZMQ check
if test "x$use_zmq" = xyes; then
PKG_CHECK_MODULES([ZMQ], [libzmq >= 4],
AC_DEFINE([ENABLE_ZMQ], [1], [Define to 1 to enable ZMQ functions]),
[AC_DEFINE([ENABLE_ZMQ], [0], [Define to 1 to enable ZMQ functions])
AC_MSG_WARN([libzmq version 4.x or greater not found, disabling])
use_zmq=no])
else
AC_DEFINE_UNQUOTED([ENABLE_ZMQ], [0], [Define to 1 to enable ZMQ functions])
fi
if test "x$use_zmq" = xyes; then
dnl Assume libzmq was built for static linking
case $host in
*mingw*)
ZMQ_CFLAGS="$ZMQ_CFLAGS -DZMQ_STATIC"
;;
esac
fi
dnl check if libgmp is present
AC_CHECK_HEADER([gmp.h],, AC_MSG_ERROR(libgmp headers missing))
AC_CHECK_LIB([gmp], [__gmpz_init],GMP_LIBS=-lgmp, AC_MSG_ERROR(libgmp missing))
AC_MSG_CHECKING([whether to build dashd])
AM_CONDITIONAL([BUILD_BITCOIND], [test x$build_bitcoind = xyes])
AC_MSG_RESULT($build_bitcoind)
AC_MSG_CHECKING([whether to build dash-cli])
AM_CONDITIONAL([BUILD_BITCOIN_CLI], [test x$build_bitcoin_cli = xyes])
AC_MSG_RESULT($build_bitcoin_cli)
AC_MSG_CHECKING([whether to build dash-tx])
AM_CONDITIONAL([BUILD_BITCOIN_TX], [test x$build_bitcoin_tx = xyes])
AC_MSG_RESULT($build_bitcoin_tx)
AC_MSG_CHECKING([whether to build dash-wallet])
AM_CONDITIONAL([BUILD_BITCOIN_WALLET], [test x$build_bitcoin_wallet = xyes])
AC_MSG_RESULT($build_bitcoin_wallet)
AC_MSG_CHECKING([whether to build libraries])
AM_CONDITIONAL([BUILD_BITCOIN_LIBS], [test x$build_bitcoin_libs = xyes])
if test x$build_bitcoin_libs = xyes; then
AC_DEFINE(HAVE_CONSENSUS_LIB, 1, [Define this symbol if the consensus lib has been built])
AC_CONFIG_FILES([libdashconsensus.pc:libdashconsensus.pc.in])
fi
AC_MSG_RESULT($build_bitcoin_libs)
AC_LANG_POP
if test "x$use_ccache" != "xno"; then
AC_MSG_CHECKING(if ccache should be used)
if test x$CCACHE = x; then
if test "x$use_ccache" = "xyes"; then
AC_MSG_ERROR([ccache not found.]);
else
use_ccache=no
fi
else
use_ccache=yes
CC="$ac_cv_path_CCACHE $CC"
CXX="$ac_cv_path_CCACHE $CXX"
fi
AC_MSG_RESULT($use_ccache)
if test "x$use_ccache" = "xyes"; then
AX_CHECK_COMPILE_FLAG([-fdebug-prefix-map=A=B],[DEBUG_CXXFLAGS="$DEBUG_CXXFLAGS -fdebug-prefix-map=\$(abs_top_srcdir)=."],,[[$CXXFLAG_WERROR]])
AX_CHECK_PREPROC_FLAG([-fmacro-prefix-map=A=B],[DEBUG_CPPFLAGS="$DEBUG_CPPFLAGS -fmacro-prefix-map=\$(abs_top_srcdir)=."],,[[$CXXFLAG_WERROR]])
fi
fi
dnl enable wallet
AC_MSG_CHECKING([if wallet should be enabled])
if test x$enable_wallet != xno; then
AC_MSG_RESULT(yes)
AC_DEFINE_UNQUOTED([ENABLE_WALLET],[1],[Define to 1 to enable wallet functions])
enable_wallet=yes
else
AC_MSG_RESULT(no)
fi
dnl enable upnp support
AC_MSG_CHECKING([whether to build with support for UPnP])
if test x$have_miniupnpc = xno; then
if test x$use_upnp = xyes; then
Merge #15993: net: Drop support of the insecure miniUPnPc versions 59cb722fd050393a69f1e0df97d857c893d19d80 Update configure to reject unsafe miniUPnPc API ver (Hennadii Stepanov) ab2190557ec2757fa48b52855b05561854af49af doc: Add release notes for 15993 (Hennadii Stepanov) 02709e95601c6020a87a6a05ee1d00c13fc38f9b Align formatting with clang-format (Hennadii Stepanov) 91a1b8508358d04685391651aea303ebce1c3d05 Use PACKAGE_NAME in UPnP description (Hennadii Stepanov) 9f76e45b9d6671e2074fb7a3885db703045a791f Drop support of insecure miniUPnPc versions (Hennadii Stepanov) Pull request description: 1. Minimum supported miniUPnPc API version is set to 10: - https://packages.ubuntu.com/xenial/libminiupnpc-dev - https://packages.debian.org/jessie/libminiupnpc-dev Refs: - #6583 - #6789 - #10414 2. The hardcoded "Bitcoin" replaced with `PACKAGE_NAME`: ![Screenshot from 2019-05-06 23-10-29](https://user-images.githubusercontent.com/32963518/57253178-afc60780-7056-11e9-83c9-e85670c58c1e.png) 3. Also style-only commit applied. Pardon: could not reopen my previous PR #15966. ACKs for top commit: ryanofsky: utACK 59cb722fd050393a69f1e0df97d857c893d19d80. Changes since last review: adding a new commit which updates configure script to fall back to disabling upnp if version is too old, adding a requested comment explaining static_assert condition, and fixing a spelling (jessy/jessie) Tree-SHA512: 42ed11bc2fb2ec83d5dd58e2383da5444a24fd572707f6cf10b622cb8943e28adfcca4750d06801024c4472625b5ea9279516fbd9d2ccebc9bbaafe1d148e80d
2019-07-29 16:41:54 +02:00
AC_MSG_ERROR("UPnP requested but cannot be built. Use --without-miniupnpc.")
fi
AC_MSG_RESULT(no)
Merge #15993: net: Drop support of the insecure miniUPnPc versions 59cb722fd050393a69f1e0df97d857c893d19d80 Update configure to reject unsafe miniUPnPc API ver (Hennadii Stepanov) ab2190557ec2757fa48b52855b05561854af49af doc: Add release notes for 15993 (Hennadii Stepanov) 02709e95601c6020a87a6a05ee1d00c13fc38f9b Align formatting with clang-format (Hennadii Stepanov) 91a1b8508358d04685391651aea303ebce1c3d05 Use PACKAGE_NAME in UPnP description (Hennadii Stepanov) 9f76e45b9d6671e2074fb7a3885db703045a791f Drop support of insecure miniUPnPc versions (Hennadii Stepanov) Pull request description: 1. Minimum supported miniUPnPc API version is set to 10: - https://packages.ubuntu.com/xenial/libminiupnpc-dev - https://packages.debian.org/jessie/libminiupnpc-dev Refs: - #6583 - #6789 - #10414 2. The hardcoded "Bitcoin" replaced with `PACKAGE_NAME`: ![Screenshot from 2019-05-06 23-10-29](https://user-images.githubusercontent.com/32963518/57253178-afc60780-7056-11e9-83c9-e85670c58c1e.png) 3. Also style-only commit applied. Pardon: could not reopen my previous PR #15966. ACKs for top commit: ryanofsky: utACK 59cb722fd050393a69f1e0df97d857c893d19d80. Changes since last review: adding a new commit which updates configure script to fall back to disabling upnp if version is too old, adding a requested comment explaining static_assert condition, and fixing a spelling (jessy/jessie) Tree-SHA512: 42ed11bc2fb2ec83d5dd58e2383da5444a24fd572707f6cf10b622cb8943e28adfcca4750d06801024c4472625b5ea9279516fbd9d2ccebc9bbaafe1d148e80d
2019-07-29 16:41:54 +02:00
use_upnp=no
else
if test x$use_upnp != xno; then
AC_MSG_RESULT(yes)
use_upnp=yes
Merge bitcoin/bitcoin#26896: build: Remove port-forwarding runtime setting options from configure d51f0fa4b7b19281efe65aacf414845c661d0a13 doc: add release notes for 26896 (fanquake) 2b248798d96f794db08b7725730b5fb4e00b9b10 build: remove --enable-upnp-default from configure (fanquake) 02f5a5e7b5fd7ba35e407d4409202a0e0fed003c build: remove --enable-natpmp-default from configure (fanquake) 25a0e8ba0b31d8bd265df0589fe49241a60d0fc2 Remove configure-time setting of DEFAULT_UPNP (fanquake) 06562e5fa771dab275a9cab4914cd64d961a52bc Remove configure-time setting of DEFAULT_NATPMP (fanquake) Pull request description: This PR removes the `--enable-upnp-default` and `--enable-natpmp-default` options from configure. It's odd to me that we maintain configure-time options for setting the default port-forwarding runtime state (but no other similar options), and I'm not sure what use-case it satisfies, that can't be achieved by multiple other means. I also doubt that we'll ever restart using these in release builds, or turning on any of this by default. I think the only scenario these options would be used is when you want to compile your own binaries (we don't use them in Guix), with port-forwarding on by default, but otherwise can't or don't want to use a `.conf` file, can't or don't want to pass command line options at runtime, and also don't want to modify the source code? ACKs for top commit: hebasto: ACK d51f0fa4b7b19281efe65aacf414845c661d0a13, rebased and comments have been addressed since my recent [review](https://github.com/bitcoin/bitcoin/pull/26896#pullrequestreview-1273910740). TheCharlatan: ACK d51f0fa4b7b19281efe65aacf414845c661d0a13 Tree-SHA512: 481decd8bddd8b03b7319591e3acf189f7b6b96c9a9a8c5bc1a3f8ec00d0b8f9b52d2f5c28a298a2ec947cfe9611cfd184e393ccb2e4e21bfce86ca7d4de60d3
2023-01-30 11:45:00 +01:00
AC_DEFINE([USE_UPNP], [1], [Define to 1 if UPnP support should be compiled in.]) if test x$TARGET_OS = xwindows; then
MINIUPNPC_CPPFLAGS="-DSTATICLIB -DMINIUPNP_STATICLIB"
fi
else
AC_MSG_RESULT(no)
fi
fi
dnl Enable NAT-PMP support.
AC_MSG_CHECKING([whether to build with support for NAT-PMP])
if test "x$have_natpmp" = xno; then
if test "x$use_natpmp" = xyes; then
AC_MSG_ERROR([NAT-PMP requested but cannot be built. Use --without-natpmp])
fi
AC_MSG_RESULT([no])
use_natpmp=no
else
if test "x$use_natpmp" != xno; then
AC_MSG_RESULT([yes])
use_natpmp=yes
Merge bitcoin/bitcoin#26896: build: Remove port-forwarding runtime setting options from configure d51f0fa4b7b19281efe65aacf414845c661d0a13 doc: add release notes for 26896 (fanquake) 2b248798d96f794db08b7725730b5fb4e00b9b10 build: remove --enable-upnp-default from configure (fanquake) 02f5a5e7b5fd7ba35e407d4409202a0e0fed003c build: remove --enable-natpmp-default from configure (fanquake) 25a0e8ba0b31d8bd265df0589fe49241a60d0fc2 Remove configure-time setting of DEFAULT_UPNP (fanquake) 06562e5fa771dab275a9cab4914cd64d961a52bc Remove configure-time setting of DEFAULT_NATPMP (fanquake) Pull request description: This PR removes the `--enable-upnp-default` and `--enable-natpmp-default` options from configure. It's odd to me that we maintain configure-time options for setting the default port-forwarding runtime state (but no other similar options), and I'm not sure what use-case it satisfies, that can't be achieved by multiple other means. I also doubt that we'll ever restart using these in release builds, or turning on any of this by default. I think the only scenario these options would be used is when you want to compile your own binaries (we don't use them in Guix), with port-forwarding on by default, but otherwise can't or don't want to use a `.conf` file, can't or don't want to pass command line options at runtime, and also don't want to modify the source code? ACKs for top commit: hebasto: ACK d51f0fa4b7b19281efe65aacf414845c661d0a13, rebased and comments have been addressed since my recent [review](https://github.com/bitcoin/bitcoin/pull/26896#pullrequestreview-1273910740). TheCharlatan: ACK d51f0fa4b7b19281efe65aacf414845c661d0a13 Tree-SHA512: 481decd8bddd8b03b7319591e3acf189f7b6b96c9a9a8c5bc1a3f8ec00d0b8f9b52d2f5c28a298a2ec947cfe9611cfd184e393ccb2e4e21bfce86ca7d4de60d3
2023-01-30 11:45:00 +01:00
AC_DEFINE([USE_NATPMP], [1], [Define to 1 if UPnP support should be compiled in.])
if test x$TARGET_OS = xwindows; then
NATPMP_CPPFLAGS="-DSTATICLIB -DNATPMP_STATICLIB"
fi
else
AC_MSG_RESULT([no])
fi
fi
dnl these are only used when qt is enabled
BUILD_TEST_QT=""
if test x$bitcoin_enable_qt != xno; then
dnl enable dbus support
AC_MSG_CHECKING([whether to build GUI with support for D-Bus])
if test x$bitcoin_enable_qt_dbus != xno; then
AC_DEFINE([USE_DBUS],[1],[Define if dbus support should be compiled in])
fi
AC_MSG_RESULT($bitcoin_enable_qt_dbus)
dnl enable qr support
AC_MSG_CHECKING([whether to build GUI with support for QR codes])
if test x$have_qrencode = xno; then
if test x$use_qr = xyes; then
AC_MSG_ERROR([QR support requested but cannot be built. Use --without-qrencode])
fi
use_qr=no
else
if test x$use_qr != xno; then
AC_DEFINE([USE_QRCODE],[1],[Define if QR support should be compiled in])
use_qr=yes
fi
fi
AC_MSG_RESULT([$use_qr])
if test x$XGETTEXT = x; then
AC_MSG_WARN("xgettext is required to update qt translations")
fi
AC_MSG_CHECKING([whether to build test_dash-qt])
if test x$use_gui_tests$bitcoin_enable_qt_test = xyesyes; then
AC_MSG_RESULT([yes])
BUILD_TEST_QT="yes"
else
AC_MSG_RESULT([no])
fi
fi
AM_CONDITIONAL([ENABLE_ZMQ], [test "x$use_zmq" = "xyes"])
AC_MSG_CHECKING([whether to build test_dash])
if test x$use_tests = xyes; then
if test "x$enable_fuzz" = "xyes"; then
AC_MSG_RESULT([no, because fuzzing is enabled])
else
AC_MSG_RESULT([yes])
fi
BUILD_TEST="yes"
else
AC_MSG_RESULT([no])
BUILD_TEST=""
fi
AC_MSG_CHECKING([whether to reduce exports])
if test x$use_reduce_exports = xyes; then
AC_MSG_RESULT([yes])
else
AC_MSG_RESULT([no])
fi
Enable stacktrace support in gitian builds (#3006) * Remove use of -rdynamic This causes check-symbols to fail horribly and also turned out to be not required when using libbacktrace. It was only required when using "backtrace()" from "<execinfo.h>" * Remove spurious ], from configure.ac * Add -DENABLE_STACKTRACES=1 to CMakeLists.txt * Remove unused method my_backtrace_simple_callback * Use fs::path().filename() instead of basename() * Add static g_exeFileName and g_exeFileBaseName * Use .exe.dbg file when available * Use uint64_t instead of uintptr_t * Implement GetBaseAddress() for unix and win32 * Implement unified crash_info and use it everywhere before printing crash info * Print a serialized version of crash_info when there is no debug info * Implement "-printcrashinfo" command line option * Compile stacktrace support unconditionally and only make crash hooks conditional This also renames the --enable-stacktraces option to --enable-crash-hooks * Enable crash hooks in win/linux Gitian builds * Try to load .debug file on MacOS and enable crash hooks for osx Gitian builds * Check for dsymutil and if it needs --flat * Create .debug files in osx Gitian build * Handle review comments * Also print crash description when no stacktrace is available * Unconditionally add -g1 debug information Instead of making it dependent on "--enable-crash-hooks". We will need the debug info every time now, even in release builds. * Put MacOS debug info into dSYM symbols instead of plain .debug files * Implement MacOS specific GetBaseAddress
2019-07-02 06:16:11 +02:00
# When compiled natively on MacOS, we need to specify -flat to avoid producing a dSYM bundle
# When cross-compiled on linux, we're using a different version of the tool that only supports flat symbol files
AC_MSG_CHECKING([whether dsymutil needs -flat])
if test x$DSYMUTIL != x && ($DSYMUTIL --help | grep -q \\-flat); then
AC_MSG_RESULT([yes])
DSYMUTIL_FLAT="$DSYMUTIL -flat"
else
AC_MSG_RESULT([no])
DSYMUTIL_FLAT="$DSYMUTIL"
fi
AC_MSG_RESULT($dsymutil_needs_flat)
if test x$build_bitcoin_wallet$build_bitcoin_cli$build_bitcoin_tx$build_bitcoind$bitcoin_enable_qt$use_tests$use_bench = xnonononononono; then
AC_MSG_ERROR([No targets! Please specify at least one of: --with-utils --with-libs --with-daemon --with-gui --enable-bench or --enable-tests])
fi
if test x$enable_fuzz_binary = xyes; then
CHECK_RUNTIME_LIB
fi
AM_CONDITIONAL([TARGET_DARWIN], [test x$TARGET_OS = xdarwin])
AM_CONDITIONAL([BUILD_DARWIN], [test x$BUILD_OS = xdarwin])
AM_CONDITIONAL([TARGET_LINUX], [test x$TARGET_OS = xlinux])
AM_CONDITIONAL([TARGET_WINDOWS], [test x$TARGET_OS = xwindows])
AM_CONDITIONAL([ENABLE_WALLET],[test x$enable_wallet = xyes])
AM_CONDITIONAL([USE_SQLITE], [test "x$use_sqlite" = "xyes"])
AM_CONDITIONAL([USE_BDB], [test "x$use_bdb" = "xyes"])
AM_CONDITIONAL([ENABLE_TRACING],[test x$have_sdt = xyes])
AM_CONDITIONAL([ENABLE_TESTS],[test x$BUILD_TEST = xyes])
AM_CONDITIONAL([ENABLE_FUZZ],[test x$enable_fuzz = xyes])
AM_CONDITIONAL([ENABLE_FUZZ_BINARY],[test x$enable_fuzz_binary = xyes])
AM_CONDITIONAL([ENABLE_FUZZ_LINK_ALL],[test x$enable_danger_fuzz_link_all = xyes])
AM_CONDITIONAL([ENABLE_QT],[test x$bitcoin_enable_qt = xyes])
AM_CONDITIONAL([ENABLE_QT_TESTS],[test x$BUILD_TEST_QT = xyes])
AM_CONDITIONAL([ENABLE_BENCH],[test x$use_bench = xyes])
AM_CONDITIONAL([USE_QRCODE], [test x$use_qr = xyes])
AM_CONDITIONAL([USE_LCOV],[test x$use_lcov = xyes])
AM_CONDITIONAL([USE_LIBEVENT],[test x$use_libevent = xyes])
AM_CONDITIONAL([HARDEN],[test x$use_hardening = xyes])
AM_CONDITIONAL([ENABLE_SSE42],[test x$enable_sse42 = xyes])
Merge #13191: Specialized double-SHA256 with 64 byte inputs with SSE4.1 and AVX2 4defdfab94504018f822dc34a313ad26cedc8255 [MOVEONLY] Move unused Merkle branch code to tests (Pieter Wuille) 4437d6e1f3107a20a8c7b66be8b4b972a82e3b28 8-way AVX2 implementation for double SHA256 on 64-byte inputs (Pieter Wuille) 230294bf5fdeba7213471cd0b795fb7aa36e5717 4-way SSE4.1 implementation for double SHA256 on 64-byte inputs (Pieter Wuille) 1f0e7ca09c9d7c5787c218156fa5096a1bdf2ea8 Use SHA256D64 in Merkle root computation (Pieter Wuille) d0c96328833127284574bfef26f96aa2e4afc91a Specialized double sha256 for 64 byte inputs (Pieter Wuille) 57f34630fb6c3e218bd19535ac607008cb894173 Refactor SHA256 code (Pieter Wuille) 0df017889b4f61860092e1d54e271092cce55f62 Benchmark Merkle root computation (Pieter Wuille) Pull request description: This introduces a framework for specialized double-SHA256 with 64 byte inputs. 4 different implementations are provided: * Generic C++ (reusing the normal SHA256 code) * Specialized C++ for 64-byte inputs, but no special instructions * 4-way using SSE4.1 intrinsics * 8-way using AVX2 intrinsics On my own system (AVX2 capable), I get these benchmarks for computing the Merkle root of 9001 leaves (supported lengths / special instructions / parallellism): * 7.2 ms with varsize/naive/1way (master, non-SSE4 hardware) * 5.8 ms with size64/naive/1way (this PR, non-SSE4 capable systems) * 4.8 ms with varsize/SSE4/1way (master, SSE4 hardware) * 2.9 ms with size64/SSE4/4way (this PR, SSE4 hardware) * 1.1 ms with size64/AVX2/8way (this PR, AVX2 hardware) Tree-SHA512: efa32d48b32820d9ce788ead4eb583949265be8c2e5f538c94bc914e92d131a57f8c1ee26c6f998e81fb0e30675d4e2eddc3360bcf632676249036018cff343e
2018-06-04 09:11:18 +02:00
AM_CONDITIONAL([ENABLE_SSE41],[test x$enable_sse41 = xyes])
AM_CONDITIONAL([ENABLE_AVX2],[test x$enable_avx2 = xyes])
AM_CONDITIONAL([ENABLE_X86_SHANI],[test x$enable_x86_shani = xyes])
AM_CONDITIONAL([ENABLE_ARM_CRC],[test x$enable_arm_crc = xyes])
AM_CONDITIONAL([ENABLE_ARM_SHANI], [test "$enable_arm_shani" = "yes"])
AM_CONDITIONAL([USE_ASM],[test x$use_asm = xyes])
AM_CONDITIONAL([WORDS_BIGENDIAN],[test x$ac_cv_c_bigendian = xyes])
AM_CONDITIONAL([USE_NATPMP],[test x$use_natpmp = xyes])
AM_CONDITIONAL([USE_UPNP],[test x$use_upnp = xyes])
AC_DEFINE(CLIENT_VERSION_MAJOR, _CLIENT_VERSION_MAJOR, [Major version])
AC_DEFINE(CLIENT_VERSION_MINOR, _CLIENT_VERSION_MINOR, [Minor version])
AC_DEFINE(CLIENT_VERSION_BUILD, _CLIENT_VERSION_BUILD, [Version Build])
AC_DEFINE(CLIENT_VERSION_IS_RELEASE, _CLIENT_VERSION_IS_RELEASE, [Version is release])
AC_DEFINE(COPYRIGHT_YEAR, _COPYRIGHT_YEAR, [Copyright year])
Merge #7192: Unify product name to as few places as possible 027fdb8 When/if the copyright line does not mention Bitcoin Core developers, add a second line to copyrights in -version, About dialog, and splash screen (Luke Dashjr) cc2095e Rewrite FormatParagraph to handle newlines within input strings correctly (Luke Dashjr) cddffaf Bugfix: Include COPYRIGHT_HOLDERS_SUBSTITUTION in Makefile substitutions so it gets passed to extract-strings correctly (Luke Dashjr) 29598e4 Move PACKAGE_URL to configure.ac (Luke Dashjr) 78ec83d splashscreen: Resize text to fit exactly (Luke Dashjr) 3cae140 Bugfix: Actually use _COPYRIGHT_HOLDERS_SUBSTITUTION everywhere (Luke Dashjr) 4d5a3df Bugfix: gitian-descriptors: Add missing python-setuptools requirement for OS X (biplist module) (Luke Dashjr) e4ab5e5 Bugfix: Correct copyright year in Mac DMG background image (Luke Dashjr) 917b1d0 Set copyright holders displayed in notices separately from the package name (Luke Dashjr) c39a6ff Travis & gitian-osx: Use depends for ds_store and mac_alias modules (Luke Dashjr) 902ccde depends: Add mac_alias to depends (Luke Dashjr) 82a2d98 depends: Add ds_store to depends (Cory Fields) de619a3 depends: Pass PYTHONPATH along to configure (Cory Fields) e611b6e macdeploy: Use rsvg-convert rather than cairosvg (Luke Dashjr) 63bcdc5 More complicated package name substitution for Mac deployment (Luke Dashjr) 1a6c67c Parameterise 2009 in translatable copyright strings (Luke Dashjr) d5f4683 Unify package name to as few places as possible without major changes (Luke Dashjr)
2016-02-04 13:41:58 +01:00
AC_DEFINE(COPYRIGHT_HOLDERS, "_COPYRIGHT_HOLDERS", [Copyright holder(s) before %s replacement])
AC_DEFINE(COPYRIGHT_HOLDERS_SUBSTITUTION, "_COPYRIGHT_HOLDERS_SUBSTITUTION", [Replacement for %s in copyright holders string])
define(_COPYRIGHT_HOLDERS_FINAL, [patsubst(_COPYRIGHT_HOLDERS, [%s], [_COPYRIGHT_HOLDERS_SUBSTITUTION])])
Merge #7192: Unify product name to as few places as possible 027fdb8 When/if the copyright line does not mention Bitcoin Core developers, add a second line to copyrights in -version, About dialog, and splash screen (Luke Dashjr) cc2095e Rewrite FormatParagraph to handle newlines within input strings correctly (Luke Dashjr) cddffaf Bugfix: Include COPYRIGHT_HOLDERS_SUBSTITUTION in Makefile substitutions so it gets passed to extract-strings correctly (Luke Dashjr) 29598e4 Move PACKAGE_URL to configure.ac (Luke Dashjr) 78ec83d splashscreen: Resize text to fit exactly (Luke Dashjr) 3cae140 Bugfix: Actually use _COPYRIGHT_HOLDERS_SUBSTITUTION everywhere (Luke Dashjr) 4d5a3df Bugfix: gitian-descriptors: Add missing python-setuptools requirement for OS X (biplist module) (Luke Dashjr) e4ab5e5 Bugfix: Correct copyright year in Mac DMG background image (Luke Dashjr) 917b1d0 Set copyright holders displayed in notices separately from the package name (Luke Dashjr) c39a6ff Travis & gitian-osx: Use depends for ds_store and mac_alias modules (Luke Dashjr) 902ccde depends: Add mac_alias to depends (Luke Dashjr) 82a2d98 depends: Add ds_store to depends (Cory Fields) de619a3 depends: Pass PYTHONPATH along to configure (Cory Fields) e611b6e macdeploy: Use rsvg-convert rather than cairosvg (Luke Dashjr) 63bcdc5 More complicated package name substitution for Mac deployment (Luke Dashjr) 1a6c67c Parameterise 2009 in translatable copyright strings (Luke Dashjr) d5f4683 Unify package name to as few places as possible without major changes (Luke Dashjr)
2016-02-04 13:41:58 +01:00
AC_DEFINE(COPYRIGHT_HOLDERS_FINAL, "_COPYRIGHT_HOLDERS_FINAL", [Copyright holder(s)])
AC_SUBST(CLIENT_VERSION_MAJOR, _CLIENT_VERSION_MAJOR)
AC_SUBST(CLIENT_VERSION_MINOR, _CLIENT_VERSION_MINOR)
AC_SUBST(CLIENT_VERSION_BUILD, _CLIENT_VERSION_BUILD)
AC_SUBST(CLIENT_VERSION_IS_RELEASE, _CLIENT_VERSION_IS_RELEASE)
AC_SUBST(COPYRIGHT_YEAR, _COPYRIGHT_YEAR)
Merge #7192: Unify product name to as few places as possible 027fdb8 When/if the copyright line does not mention Bitcoin Core developers, add a second line to copyrights in -version, About dialog, and splash screen (Luke Dashjr) cc2095e Rewrite FormatParagraph to handle newlines within input strings correctly (Luke Dashjr) cddffaf Bugfix: Include COPYRIGHT_HOLDERS_SUBSTITUTION in Makefile substitutions so it gets passed to extract-strings correctly (Luke Dashjr) 29598e4 Move PACKAGE_URL to configure.ac (Luke Dashjr) 78ec83d splashscreen: Resize text to fit exactly (Luke Dashjr) 3cae140 Bugfix: Actually use _COPYRIGHT_HOLDERS_SUBSTITUTION everywhere (Luke Dashjr) 4d5a3df Bugfix: gitian-descriptors: Add missing python-setuptools requirement for OS X (biplist module) (Luke Dashjr) e4ab5e5 Bugfix: Correct copyright year in Mac DMG background image (Luke Dashjr) 917b1d0 Set copyright holders displayed in notices separately from the package name (Luke Dashjr) c39a6ff Travis & gitian-osx: Use depends for ds_store and mac_alias modules (Luke Dashjr) 902ccde depends: Add mac_alias to depends (Luke Dashjr) 82a2d98 depends: Add ds_store to depends (Cory Fields) de619a3 depends: Pass PYTHONPATH along to configure (Cory Fields) e611b6e macdeploy: Use rsvg-convert rather than cairosvg (Luke Dashjr) 63bcdc5 More complicated package name substitution for Mac deployment (Luke Dashjr) 1a6c67c Parameterise 2009 in translatable copyright strings (Luke Dashjr) d5f4683 Unify package name to as few places as possible without major changes (Luke Dashjr)
2016-02-04 13:41:58 +01:00
AC_SUBST(COPYRIGHT_HOLDERS, "_COPYRIGHT_HOLDERS")
AC_SUBST(COPYRIGHT_HOLDERS_SUBSTITUTION, "_COPYRIGHT_HOLDERS_SUBSTITUTION")
AC_SUBST(COPYRIGHT_HOLDERS_FINAL, "_COPYRIGHT_HOLDERS_FINAL")
AC_SUBST(BITCOIN_DAEMON_NAME)
AC_SUBST(BITCOIN_GUI_NAME)
AC_SUBST(BITCOIN_CLI_NAME)
AC_SUBST(BITCOIN_TX_NAME)
AC_SUBST(BITCOIN_WALLET_TOOL_NAME)
AC_SUBST(RELDFLAGS)
AC_SUBST(DEBUG_CPPFLAGS)
AC_SUBST(WARN_CXXFLAGS)
AC_SUBST(NOWARN_CXXFLAGS)
AC_SUBST(DEBUG_CXXFLAGS)
AC_SUBST(ERROR_CXXFLAGS)
AC_SUBST(GPROF_CXXFLAGS)
AC_SUBST(GPROF_LDFLAGS)
AC_SUBST(HARDENED_CXXFLAGS)
AC_SUBST(HARDENED_CPPFLAGS)
AC_SUBST(HARDENED_LDFLAGS)
AC_SUBST(PIC_FLAGS)
AC_SUBST(PIE_FLAGS)
AC_SUBST(SANITIZER_CXXFLAGS)
AC_SUBST(SANITIZER_LDFLAGS)
AC_SUBST(SSE42_CXXFLAGS)
Merge #13191: Specialized double-SHA256 with 64 byte inputs with SSE4.1 and AVX2 4defdfab94504018f822dc34a313ad26cedc8255 [MOVEONLY] Move unused Merkle branch code to tests (Pieter Wuille) 4437d6e1f3107a20a8c7b66be8b4b972a82e3b28 8-way AVX2 implementation for double SHA256 on 64-byte inputs (Pieter Wuille) 230294bf5fdeba7213471cd0b795fb7aa36e5717 4-way SSE4.1 implementation for double SHA256 on 64-byte inputs (Pieter Wuille) 1f0e7ca09c9d7c5787c218156fa5096a1bdf2ea8 Use SHA256D64 in Merkle root computation (Pieter Wuille) d0c96328833127284574bfef26f96aa2e4afc91a Specialized double sha256 for 64 byte inputs (Pieter Wuille) 57f34630fb6c3e218bd19535ac607008cb894173 Refactor SHA256 code (Pieter Wuille) 0df017889b4f61860092e1d54e271092cce55f62 Benchmark Merkle root computation (Pieter Wuille) Pull request description: This introduces a framework for specialized double-SHA256 with 64 byte inputs. 4 different implementations are provided: * Generic C++ (reusing the normal SHA256 code) * Specialized C++ for 64-byte inputs, but no special instructions * 4-way using SSE4.1 intrinsics * 8-way using AVX2 intrinsics On my own system (AVX2 capable), I get these benchmarks for computing the Merkle root of 9001 leaves (supported lengths / special instructions / parallellism): * 7.2 ms with varsize/naive/1way (master, non-SSE4 hardware) * 5.8 ms with size64/naive/1way (this PR, non-SSE4 capable systems) * 4.8 ms with varsize/SSE4/1way (master, SSE4 hardware) * 2.9 ms with size64/SSE4/4way (this PR, SSE4 hardware) * 1.1 ms with size64/AVX2/8way (this PR, AVX2 hardware) Tree-SHA512: efa32d48b32820d9ce788ead4eb583949265be8c2e5f538c94bc914e92d131a57f8c1ee26c6f998e81fb0e30675d4e2eddc3360bcf632676249036018cff343e
2018-06-04 09:11:18 +02:00
AC_SUBST(SSE41_CXXFLAGS)
AC_SUBST(AVX2_CXXFLAGS)
AC_SUBST(X86_SHANI_CXXFLAGS)
AC_SUBST(ARM_CRC_CXXFLAGS)
AC_SUBST(ARM_SHANI_CXXFLAGS)
AC_SUBST(LIBTOOL_APP_LDFLAGS)
AC_SUBST(USE_SQLITE)
AC_SUBST(USE_BDB)
AC_SUBST(USE_UPNP)
AC_SUBST(USE_QRCODE)
AC_SUBST(BOOST_LIBS)
AC_SUBST(SQLITE_LIBS)
AC_SUBST(TESTDEFS)
AC_SUBST(MINIUPNPC_CPPFLAGS)
AC_SUBST(MINIUPNPC_LIBS)
AC_SUBST(GMP_LIBS)
AC_SUBST(NATPMP_CPPFLAGS)
AC_SUBST(NATPMP_LIBS)
AC_SUBST(EVENT_LIBS)
AC_SUBST(EVENT_PTHREADS_LIBS)
AC_SUBST(ZMQ_LIBS)
AC_SUBST(QR_LIBS)
Enable stacktrace support in gitian builds (#3006) * Remove use of -rdynamic This causes check-symbols to fail horribly and also turned out to be not required when using libbacktrace. It was only required when using "backtrace()" from "<execinfo.h>" * Remove spurious ], from configure.ac * Add -DENABLE_STACKTRACES=1 to CMakeLists.txt * Remove unused method my_backtrace_simple_callback * Use fs::path().filename() instead of basename() * Add static g_exeFileName and g_exeFileBaseName * Use .exe.dbg file when available * Use uint64_t instead of uintptr_t * Implement GetBaseAddress() for unix and win32 * Implement unified crash_info and use it everywhere before printing crash info * Print a serialized version of crash_info when there is no debug info * Implement "-printcrashinfo" command line option * Compile stacktrace support unconditionally and only make crash hooks conditional This also renames the --enable-stacktraces option to --enable-crash-hooks * Enable crash hooks in win/linux Gitian builds * Try to load .debug file on MacOS and enable crash hooks for osx Gitian builds * Check for dsymutil and if it needs --flat * Create .debug files in osx Gitian build * Handle review comments * Also print crash description when no stacktrace is available * Unconditionally add -g1 debug information Instead of making it dependent on "--enable-crash-hooks". We will need the debug info every time now, even in release builds. * Put MacOS debug info into dSYM symbols instead of plain .debug files * Implement MacOS specific GetBaseAddress
2019-07-02 06:16:11 +02:00
AC_SUBST(DSYMUTIL_FLAT)
AC_SUBST(HAVE_FDATASYNC)
AC_SUBST(HAVE_FULLFSYNC)
AC_SUBST(HAVE_O_CLOEXEC)
AC_SUBST(HAVE_BUILTIN_PREFETCH)
AC_SUBST(HAVE_MM_PREFETCH)
AC_SUBST(HAVE_STRONG_GETAUXVAL)
AC_SUBST(HAVE_GMTIME_R)
AC_SUBST(ANDROID_ARCH)
AC_CONFIG_FILES([Makefile src/Makefile doc/man/Makefile share/setup.nsi share/qt/Info.plist test/config.ini])
AC_CONFIG_FILES([contrib/devtools/split-debug.sh],[chmod +x contrib/devtools/split-debug.sh])
AM_COND_IF([HAVE_DOXYGEN], [AC_CONFIG_FILES([doc/Doxyfile])])
AC_CONFIG_LINKS([contrib/devtools/security-check.py:contrib/devtools/security-check.py])
AC_CONFIG_LINKS([contrib/devtools/test-security-check.py:contrib/devtools/test-security-check.py])
AC_CONFIG_LINKS([contrib/devtools/symbol-check.py:contrib/devtools/symbol-check.py])
AC_CONFIG_LINKS([contrib/devtools/test-symbol-check.py:contrib/devtools/test-symbol-check.py])
2017-10-02 13:35:08 +02:00
AC_CONFIG_LINKS([contrib/filter-lcov.py:contrib/filter-lcov.py])
AC_CONFIG_LINKS([contrib/macdeploy/background.tiff:contrib/macdeploy/background.tiff])
AC_CONFIG_LINKS([test/functional/test_runner.py:test/functional/test_runner.py])
AC_CONFIG_LINKS([test/fuzz/test_runner.py:test/fuzz/test_runner.py])
AC_CONFIG_LINKS([test/util/bitcoin-util-test.py:test/util/bitcoin-util-test.py])
AC_CONFIG_LINKS([test/util/rpcauth-test.py:test/util/rpcauth-test.py])
2014-11-05 16:58:37 +01:00
dnl boost's m4 checks do something really nasty: they export these vars. As a
dnl result, they leak into secp256k1's configure and crazy things happen.
dnl Until this is fixed upstream and we've synced, we'll just un-export them.
CPPFLAGS_TEMP="$CPPFLAGS"
2014-11-05 20:48:11 +01:00
unset CPPFLAGS
CPPFLAGS="$CPPFLAGS_TEMP"
LDFLAGS_TEMP="$LDFLAGS"
2014-11-05 20:48:11 +01:00
unset LDFLAGS
LDFLAGS="$LDFLAGS_TEMP"
LIBS_TEMP="$LIBS"
unset LIBS
LIBS="$LIBS_TEMP"
ac_configure_args="${ac_configure_args} --disable-shared --with-pic --enable-benchmark=no --enable-module-recovery --disable-module-ecdh"
AC_CONFIG_SUBDIRS([src/dashbls src/secp256k1])
2014-11-05 16:58:37 +01:00
AC_OUTPUT
dnl Replace the BUILDDIR path with the correct Windows path if compiling on Native Windows
case ${OS} in
*Windows*)
sed 's/BUILDDIR="\/\([[a-z]]\)/BUILDDIR="\1:/' test/config.ini > test/config-2.ini
mv test/config-2.ini test/config.ini
;;
esac
Merge #13458: gui: Drop qt4 support af6ac3b677454644364fd24d0df0c02ac9b8c8db doc: Remove mention of Qt4 from build docs (Wladimir J. van der Laan) 462c71f71b5d753bc8327ab833dea23679450ca3 test: Update travis to not test Qt4 anymore (Wladimir J. van der Laan) 907f73bbc5b6c98b01d7c6088a294dea66634a3f gui: Remove QT_VERSION fallbacks for Qt < 5 (Wladimir J. van der Laan) bad068ad9f4bc60bfc10e27d4ffaec92d7df8491 build: Build system changes to support only Qt5 (Wladimir J. van der Laan) Pull request description: Implements #8263. Qt4.x has been EOL since 2015, and at least Gentoo has, or is going to drop support for it. I wouldn't be surprised if other Linux distributions follow. This removes Qt4 detection from the build system, as well as removes all Qt4 fallbacks from the code. Turns out there's more than I expected: this is going to make maintenance of the GUI code, as well as adding new features significantly easier. (I know there's still some references left to qt4 in RPM and Debian build script, but I don't have the knowledge how to fix them) Tree-SHA512: d495924fd4dda6f6566ba44ee96be7cbe62e69ba1ca993b80a8449f78da852b7f1bd3e8200d57cfa1d72233c340eeff4596fb0032ecbddc715d99aea63817d3f (cherry picked from commit dc53f7f2514026db8a28632371e73e6dbf858083) # Conflicts: # .travis.yml # build-aux/m4/bitcoin_qt.m4 # configure.ac # doc/build-osx.md # doc/build-unix.md # doc/dependencies.md # src/qt/README.md # src/qt/bitcoingui.cpp # src/qt/coincontroldialog.cpp # src/qt/dash.cpp # src/qt/guiutil.cpp # src/qt/macdockiconhandler.mm # src/qt/networkstyle.cpp # src/qt/openuridialog.cpp # src/qt/peertablemodel.cpp # src/qt/signverifymessagedialog.cpp # src/qt/test/wallettests.cpp # src/qt/transactionview.cpp
2018-06-24 15:56:00 +02:00
echo
echo "Options used to compile and link:"
echo " with wallet = $enable_wallet"
echo " with gui / qt = $bitcoin_enable_qt"
if test "x$enable_wallet" != "xno"; then
echo " with sqlite = $use_sqlite"
echo " with bdb = $use_bdb"
fi
if test x$bitcoin_enable_qt != xno; then
echo " with qr = $use_qr"
fi
echo " with zmq = $use_zmq"
if test x$enable_fuzz == xno; then
echo " with test = $use_tests"
else
echo " with test = not building test_dash because fuzzing is enabled"
echo " with fuzz = $enable_fuzz"
fi
echo " with bench = $use_bench"
echo " with upnp = $use_upnp"
echo " with natpmp = $use_natpmp"
echo " use asm = $use_asm"
echo " ebpf tracing = $have_sdt"
echo " sanitizers = $use_sanitizers"
echo " debug enabled = $enable_debug"
echo " stacktraces enabled = $enable_stacktraces"
Enable stacktrace support in gitian builds (#3006) * Remove use of -rdynamic This causes check-symbols to fail horribly and also turned out to be not required when using libbacktrace. It was only required when using "backtrace()" from "<execinfo.h>" * Remove spurious ], from configure.ac * Add -DENABLE_STACKTRACES=1 to CMakeLists.txt * Remove unused method my_backtrace_simple_callback * Use fs::path().filename() instead of basename() * Add static g_exeFileName and g_exeFileBaseName * Use .exe.dbg file when available * Use uint64_t instead of uintptr_t * Implement GetBaseAddress() for unix and win32 * Implement unified crash_info and use it everywhere before printing crash info * Print a serialized version of crash_info when there is no debug info * Implement "-printcrashinfo" command line option * Compile stacktrace support unconditionally and only make crash hooks conditional This also renames the --enable-stacktraces option to --enable-crash-hooks * Enable crash hooks in win/linux Gitian builds * Try to load .debug file on MacOS and enable crash hooks for osx Gitian builds * Check for dsymutil and if it needs --flat * Create .debug files in osx Gitian build * Handle review comments * Also print crash description when no stacktrace is available * Unconditionally add -g1 debug information Instead of making it dependent on "--enable-crash-hooks". We will need the debug info every time now, even in release builds. * Put MacOS debug info into dSYM symbols instead of plain .debug files * Implement MacOS specific GetBaseAddress
2019-07-02 06:16:11 +02:00
echo " crash hooks enabled = $enable_crashhooks"
echo " miner enabled = $enable_miner"
echo " gprof enabled = $enable_gprof"
echo " werror = $enable_werror"
Merge #13458: gui: Drop qt4 support af6ac3b677454644364fd24d0df0c02ac9b8c8db doc: Remove mention of Qt4 from build docs (Wladimir J. van der Laan) 462c71f71b5d753bc8327ab833dea23679450ca3 test: Update travis to not test Qt4 anymore (Wladimir J. van der Laan) 907f73bbc5b6c98b01d7c6088a294dea66634a3f gui: Remove QT_VERSION fallbacks for Qt < 5 (Wladimir J. van der Laan) bad068ad9f4bc60bfc10e27d4ffaec92d7df8491 build: Build system changes to support only Qt5 (Wladimir J. van der Laan) Pull request description: Implements #8263. Qt4.x has been EOL since 2015, and at least Gentoo has, or is going to drop support for it. I wouldn't be surprised if other Linux distributions follow. This removes Qt4 detection from the build system, as well as removes all Qt4 fallbacks from the code. Turns out there's more than I expected: this is going to make maintenance of the GUI code, as well as adding new features significantly easier. (I know there's still some references left to qt4 in RPM and Debian build script, but I don't have the knowledge how to fix them) Tree-SHA512: d495924fd4dda6f6566ba44ee96be7cbe62e69ba1ca993b80a8449f78da852b7f1bd3e8200d57cfa1d72233c340eeff4596fb0032ecbddc715d99aea63817d3f (cherry picked from commit dc53f7f2514026db8a28632371e73e6dbf858083) # Conflicts: # .travis.yml # build-aux/m4/bitcoin_qt.m4 # configure.ac # doc/build-osx.md # doc/build-unix.md # doc/dependencies.md # src/qt/README.md # src/qt/bitcoingui.cpp # src/qt/coincontroldialog.cpp # src/qt/dash.cpp # src/qt/guiutil.cpp # src/qt/macdockiconhandler.mm # src/qt/networkstyle.cpp # src/qt/openuridialog.cpp # src/qt/peertablemodel.cpp # src/qt/signverifymessagedialog.cpp # src/qt/test/wallettests.cpp # src/qt/transactionview.cpp
2018-06-24 15:56:00 +02:00
echo
echo " target os = $TARGET_OS"
echo " build os = $build_os"
echo
echo " CC = $CC"
echo " CFLAGS = $PTHREAD_CFLAGS $CFLAGS"
echo " CPPFLAGS = $DEBUG_CPPFLAGS $HARDENED_CPPFLAGS $CPPFLAGS"
echo " CXX = $CXX"
echo " CXXFLAGS = $DEBUG_CXXFLAGS $HARDENED_CXXFLAGS $WARN_CXXFLAGS $NOWARN_CXXFLAGS $ERROR_CXXFLAGS $GPROF_CXXFLAGS $CXXFLAGS"
echo " LDFLAGS = $PTHREAD_LIBS $HARDENED_LDFLAGS $GPROF_LDFLAGS $LDFLAGS"
echo " ARFLAGS = $ARFLAGS"
Merge #13458: gui: Drop qt4 support af6ac3b677454644364fd24d0df0c02ac9b8c8db doc: Remove mention of Qt4 from build docs (Wladimir J. van der Laan) 462c71f71b5d753bc8327ab833dea23679450ca3 test: Update travis to not test Qt4 anymore (Wladimir J. van der Laan) 907f73bbc5b6c98b01d7c6088a294dea66634a3f gui: Remove QT_VERSION fallbacks for Qt < 5 (Wladimir J. van der Laan) bad068ad9f4bc60bfc10e27d4ffaec92d7df8491 build: Build system changes to support only Qt5 (Wladimir J. van der Laan) Pull request description: Implements #8263. Qt4.x has been EOL since 2015, and at least Gentoo has, or is going to drop support for it. I wouldn't be surprised if other Linux distributions follow. This removes Qt4 detection from the build system, as well as removes all Qt4 fallbacks from the code. Turns out there's more than I expected: this is going to make maintenance of the GUI code, as well as adding new features significantly easier. (I know there's still some references left to qt4 in RPM and Debian build script, but I don't have the knowledge how to fix them) Tree-SHA512: d495924fd4dda6f6566ba44ee96be7cbe62e69ba1ca993b80a8449f78da852b7f1bd3e8200d57cfa1d72233c340eeff4596fb0032ecbddc715d99aea63817d3f (cherry picked from commit dc53f7f2514026db8a28632371e73e6dbf858083) # Conflicts: # .travis.yml # build-aux/m4/bitcoin_qt.m4 # configure.ac # doc/build-osx.md # doc/build-unix.md # doc/dependencies.md # src/qt/README.md # src/qt/bitcoingui.cpp # src/qt/coincontroldialog.cpp # src/qt/dash.cpp # src/qt/guiutil.cpp # src/qt/macdockiconhandler.mm # src/qt/networkstyle.cpp # src/qt/openuridialog.cpp # src/qt/peertablemodel.cpp # src/qt/signverifymessagedialog.cpp # src/qt/test/wallettests.cpp # src/qt/transactionview.cpp
2018-06-24 15:56:00 +02:00
echo