oxen-io
/
oxen-core
mirror of https://github.com/oxen-io/oxen-core.git
2
1
Fork 0
Code Issues Projects Releases Wiki Activity
oxen-core/CMakeLists.txt

989 lines
36 KiB
CMake
Raw Normal View History

Update cmakelists and misc refs to Loki
2018-04-25 08:13:38 +02:00
# Copyright (c) 2018, The Loki Project
Add more 2019 license updates for Monero
2019-04-12 06:38:30 +02:00
# Copyright (c) 2014-2019, The Monero Project
Replace tabs with two spaces for consistency with rest of codebase Remove trailing whitespace in same files.
2015-12-14 05:54:39 +01:00
#
added license to CMake
2014-09-11 08:25:07 +02:00
# All rights reserved.
Replace tabs with two spaces for consistency with rest of codebase Remove trailing whitespace in same files.
2015-12-14 05:54:39 +01:00
#
added license to CMake
2014-09-11 08:25:07 +02:00
# Redistribution and use in source and binary forms, with or without modification, are
# permitted provided that the following conditions are met:
Replace tabs with two spaces for consistency with rest of codebase Remove trailing whitespace in same files.
2015-12-14 05:54:39 +01:00
#
added license to CMake
2014-09-11 08:25:07 +02:00
# 1. Redistributions of source code must retain the above copyright notice, this list of
# conditions and the following disclaimer.
Replace tabs with two spaces for consistency with rest of codebase Remove trailing whitespace in same files.
2015-12-14 05:54:39 +01:00
#
added license to CMake
2014-09-11 08:25:07 +02:00
# 2. Redistributions in binary form must reproduce the above copyright notice, this list
# of conditions and the following disclaimer in the documentation and/or other
# materials provided with the distribution.
Replace tabs with two spaces for consistency with rest of codebase Remove trailing whitespace in same files.
2015-12-14 05:54:39 +01:00
#
added license to CMake
2014-09-11 08:25:07 +02:00
# 3. Neither the name of the copyright holder nor the names of its contributors may be
# used to endorse or promote products derived from this software without specific
# prior written permission.
Replace tabs with two spaces for consistency with rest of codebase Remove trailing whitespace in same files.
2015-12-14 05:54:39 +01:00
#
added license to CMake
2014-09-11 08:25:07 +02:00
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY
# EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
# MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL
# THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
# PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
# INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
# STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF
# THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
Replace tabs with two spaces for consistency with rest of codebase Remove trailing whitespace in same files.
2015-12-14 05:54:39 +01:00
#
added license to CMake
2014-09-11 08:25:07 +02:00
# Parts of this file are originally copyright (c) 2012-2013 The Cryptonote developers
Add the iwyu target to get information on unused headers With this change, building the iwyu target will provide information on unused headers, which can be removed to reduce compilation times.
2019-09-05 17:27:12 +02:00
set(CMAKE_EXPORT_COMPILE_COMMANDS ON)
Add misc hardening flags to the cmake machinery See https://wiki.debian.org/Hardening#User_Space
2017-12-22 22:48:31 +01:00
list(INSERT CMAKE_MODULE_PATH 0
"${CMAKE_SOURCE_DIR}/cmake")
Enable ccache by default It can still be explicitly disabled with -DCMAKE_CXX_COMPILER_LAUNCHER= -DCMAKE_C_COMPILER_LAUNCHER= if someone doesn't want it for some reason.
2020-06-11 20:12:30 +02:00
find_program(CCACHE_PROGRAM ccache)
if(CCACHE_PROGRAM)
foreach(lang C CXX)
if(NOT DEFINED CMAKE_${lang}_COMPILER_LAUNCHER AND NOT CMAKE_${lang}_COMPILER MATCHES ".*/ccache")
message(STATUS "Enabling ccache for ${lang}")
set(CMAKE_${lang}_COMPILER_LAUNCHER ${CCACHE_PROGRAM} CACHE STRING "")
endif()
endforeach()
endif()
Bump oxen-logging to stable release; fix imported fmt/spdlog targets The tagged 1.0.0 release includes a fix for the fmt::fmt/spdlog::spdlog targets not being transitively available. (It does require a cmake version bump to 3.13, though, to make it work).
2022-09-22 19:07:09 +02:00
cmake_minimum_required(VERSION 3.13...3.24)
build: use ARCH 'native' by default, allow to configure and override it
2018-10-13 11:46:19 +02:00
message(STATUS "CMake version ${CMAKE_VERSION}")
moved all stuff to github
2014-03-03 23:07:58 +01:00
MacOS 10.14 compatibility
2020-10-13 23:13:08 +02:00
# Has to be set before `project()`, and ignored on non-macos:
Target macos 10.12 When targetting macos <10.14 macos won't allow use of anything from C++17 that throws, such as: - std::get on a variant - std::visit - std::optional::value() - std::any_cast This avoids all of these. For std::get, we either replace with std::get_if (where appropriate), or else use a `var::get` implementation of std::get added to lokimq (also updated here). (This `var` namespace is just an `std` alias everywhere *except* old target macos). For std::visit, likewise lokimq adds an var::visit implementation for old macos that we use. std::optional::value() uses weren't useful anyway as everywhere it calls them we've already checked that the option has a value, in which case we can use `*opt` (which doesn't check for contents and throw). std::any just has to be avoided as far as I can tell, but the one place we used it is only ever a block, so I just replaced it with a `const block*`.
2020-10-16 00:44:36 +02:00
set(CMAKE_OSX_DEPLOYMENT_TARGET 10.12 CACHE STRING "macOS deployment target (Apple clang only)")
MacOS 10.14 compatibility
2020-10-13 23:13:08 +02:00
executable names changed
2021-01-04 04:19:42 +01:00
project(oxen
Change dev version to 11.0.0
2022-09-01 23:57:23 +02:00
VERSION 11.0.0
Move version into top-level CMakeLists.txt CMake already provides variables to handle the version major/minor/patch if we give it the dotted version in the `project()` command. Using it significantly reduces the amount of macro stuff we have to do in version.cpp.in, and it seems a little nicer to have it defined in the project top level rather than buried in a needs-to-beprocessed .cpp file. This moves the release codename there, too, so that it stays being defined in essentially the same place as the version. This change here requires some minor tweaking of the version generation code to do it in two steps (when we have git): the first (`src/version.cpp.in` -> `build/version.cpp.in`) replaces all the main version variables during cmake configuration, the second (`build/version.cpp.in` -> `build/version.cpp`) then replaces the VERSIONTAG at build time. (Before this commit, there was only version tag replacement that only happened at build time). Also bumps up the version here (since I'm moving it anyway) to match master's 7.1.8.
2020-05-03 20:45:26 +02:00
LANGUAGES CXX C)
Update release codename
2022-05-30 22:13:27 +02:00
set(OXEN_RELEASE_CODENAME "Wistful Wagyu")
Oxen 10.2 version bumps & mandatory upgrade date - 10.2.0 oxend version - keep lokinet at 0.9.9 - require storage server 2.4.0 - schedule 10.2.0 mandatory date as 14 September 00:00 UTC (10am in Australia Eastern Time; 8pm (on the 13th) in US Eastern time.
2022-08-22 23:09:31 +02:00
# Version update notes:
# - bump the version above.
# - for a new major release, make a new codename
# - (if update needed) required SS/lokinet versions are in cryptonote_core/service_node_rules.h
cmake: set the project name
2014-10-23 20:03:54 +02:00
Add a LOKI_RELEASE_SUFFIX for adding "-dev" to dev branch builds This will require a one-time change when we merge to master (to delete the `-dev`) but after that it should just sit there without needing any modifications on `master` or `dev`. (And letting it be specified via cmake arguments will let me slightly simplify the debs which currently have to add a vaguely similar patch to get the debian version into the version string).
2020-05-04 03:10:09 +02:00
# String value to append to the full version string; this is intended to easily identify whether a
# binary was build from the release or development branches. This should be permanently set to an
Remove -dev suffix for stable branch
2020-10-02 02:21:37 +02:00
# empty string on `stable`, "-dev" on the `dev` branch, and can be set externally (via cmake
Add a LOKI_RELEASE_SUFFIX for adding "-dev" to dev branch builds This will require a one-time change when we merge to master (to delete the `-dev`) but after that it should just sit there without needing any modifications on `master` or `dev`. (And letting it be specified via cmake arguments will let me slightly simplify the debs which currently have to add a vaguely similar patch to get the debian version into the version string).
2020-05-04 03:10:09 +02:00
# arguments) where it makes sense to take some other branch release with an extra value.
executable names changed
2021-01-04 04:19:42 +01:00
if(NOT DEFINED OXEN_RELEASE_SUFFIX)
set(OXEN_RELEASE_SUFFIX "-dev")
Add a LOKI_RELEASE_SUFFIX for adding "-dev" to dev branch builds This will require a one-time change when we merge to master (to delete the `-dev`) but after that it should just sit there without needing any modifications on `master` or `dev`. (And letting it be specified via cmake arguments will let me slightly simplify the debs which currently have to add a vaguely similar patch to get the debian version into the version string).
2020-05-04 03:10:09 +02:00
endif()
cmake: set the project name
2014-10-23 20:03:54 +02:00
Fix cmake policy warnings The policy needs to be set before the `link_dep_libs` function (it was before when it was in `external`, but this started generating warnings again when I moved it from there to the top level CMakeLists.txt).
2020-03-20 18:49:25 +01:00
if(POLICY CMP0079)
cmake_policy(SET CMP0079 NEW)
endif()
C++17 Switch loki dev branch to C++17 compilation, and update the code with various C++17 niceties. - stop including the (deprecated) lokimq/string_view.h header and instead switch everything to use std::string_view and `""sv` instead of `""_sv`. - std::string_view is much nicer than epee::span, so updated various loki-specific code to use it instead. - made epee "portable storage" serialization accept a std::string_view instead of const lvalue std::string so that we can avoid copying. - switched from mapbox::variant to std::variant - use `auto [a, b] = whatever()` instead of `T1 a; T2 b; std::tie(a, b) = whatever()` in a couple places (in the wallet code). - switch to std::lock(...) instead of boost::lock(...) for simultaneous lock acquisition. boost::lock() won't compile in C++17 mode when given locks of different types. - removed various pre-C++17 workarounds, e.g. for fold expressions, unused argument attributes, and byte-spannable object detection. - class template deduction means lock types no longer have to specify the mutex, so `std::unique_lock<std::mutex> lock{mutex}` can become `std::unique_lock lock{mutex}`. This will make switching any mutex types (e.g. from boost to std mutexes) far easier as you just have to update the type in the header and everything should work. This also makes the tools::unique_lock and tools::shared_lock methods redundant (which were a sort of poor-mans-pre-C++17 way to eliminate the redundancy) so they are now gone and replaced with direct unique_lock or shared_lock constructions. - Redid the LNS validation using a string_view; instead of using raw char pointers the code now uses a string view and chops off parts of the view as it validates. So, for instance, it starts with "abcd.loki", validates the ".loki" and chops the view to "abcd", then validates the first character and chops to "bcd", validates the last and chops to "bc", then can just check everything remaining for is-valid-middle-char. - LNS validation gained a couple minor validation checks in the process: - slightly tightened the requirement on lokinet addresses to require that the last character of the mapped address is 'y' or 'o' (the last base32z char holds only one significant bit). - In parse_owner_to_generic_owner made sure that the owner value has the correct size (otherwise we could up end not filling or overfilling the pubkey buffer). - Replaced base32z/base64/hex conversions with lokimq's versions which have a nicer interface, are better optimized, and don't depend on epee.
2020-05-13 20:12:49 +02:00
set(CMAKE_CXX_STANDARD 17)
Set C++14 compilation mode properly CMake has standard interfaces for this; hacking it into CXX_FLAGS is just gross.
2020-06-12 18:16:33 +02:00
set(CMAKE_CXX_STANDARD_REQUIRED ON)
set(CMAKE_CXX_EXTENSIONS OFF)
set(CMAKE_C_STANDARD 11)
set(CMAKE_C_STANDARD_REQUIRED ON)
set(CMAKE_C_EXTENSIONS OFF)
Use cmake variable for enabling PIC With the current approach -fPIC wasn't getting used when building external libraries. I tried moving add_subdirectory(external) down below where we set flags, but that results in a slew or warning because it *also* turn on a bunch of warnings that aren't safe in the various external code.
2020-07-16 19:14:19 +02:00
set(CMAKE_POSITION_INDEPENDENT_CODE TRUE)
Modernize how -pie gets set Let cmake deal with setting up -pie; this ought to avoid a bunch of unused argument errors on apple clang where -pie is being forced on incorrectly.
2021-05-12 16:33:53 +02:00
if(CMAKE_VERSION VERSION_GREATER_EQUAL 3.14)
cmake_policy(SET CMP0083 NEW)
include(CheckPIESupported)
check_pie_supported(OUTPUT_VARIABLE pie_error LANGUAGES CXX C)
if(NOT CMAKE_CXX_LINK_PIE_SUPPORTED)
message(WARNING "PIE linking is not supported: ${pie_error}")
endif()
else()
message(WARNING "PIE disabled: cmake 3.14+ is required for proper PIE linking support")
endif()
Use cmake variable for enabling PIC With the current approach -fPIC wasn't getting used when building external libraries. I tried moving add_subdirectory(external) down below where we set flags, but that results in a slew or warning because it *also* turn on a bunch of warnings that aren't safe in the various external code.
2020-07-16 19:14:19 +02:00
std::filesystem Converts all use of boost::filesystem to std::filesystem. For macos and potentially other exotic systems where std::filesystem isn't available, we use ghc::filesystem instead (which is a drop-in replacement for std::filesystem, unlike boost::filesystem). This also greatly changes how we handle filenames internally by holding them in filesystem::path objects as soon as possible (using fs::u8path()), rather than strings, which avoids a ton of issues around unicode filenames. As a result this lets us drop the boost::locale dependency on Windows along with a bunch of messy Windows ifdef code, and avoids the need for doing gross boost locale codecvt calls.
2020-10-22 19:55:33 +02:00
include(CheckCCompilerFlag)
include(CheckCXXCompilerFlag)
include(CheckLinkerFlag)
include(CheckLibraryExists)
include(CheckFunctionExists)
Add misc hardening flags to the cmake machinery See https://wiki.debian.org/Hardening#User_Space
2017-12-22 22:48:31 +01:00
function (add_c_flag_if_supported flag var)
string(REPLACE "-" "_" supported ${flag}_c)
check_c_compiler_flag(${flag} ${supported})
if(${${supported}})
set(${var} "${${var}} ${flag}" PARENT_SCOPE)
endif()
endfunction()
function (add_cxx_flag_if_supported flag var)
string(REPLACE "-" "_" supported ${flag}_cxx)
check_cxx_compiler_flag(${flag} ${supported})
if(${${supported}})
set(${var} "${${var}} ${flag}" PARENT_SCOPE)
endif()
endfunction()
function (add_linker_flag_if_supported flag var)
string(REPLACE "-" "_" supported ${flag}_ld)
string(REPLACE "," "_" supported ${flag}_ld)
check_linker_flag(${flag} ${supported})
if(${${supported}})
set(${var} "${${var}} ${flag}" PARENT_SCOPE)
endif()
endfunction()
cmake: add -DHAVE_ defines to the command line since there's no config.h here
2018-01-28 09:57:51 +01:00
function (add_definition_if_function_found function var)
string(REPLACE "-" "_" supported ${function}_function)
check_function_exists(${function} ${supported})
if(${${supported}})
add_definitions("-D${var}")
endif()
endfunction()
function (add_definition_if_library_exists library function header var)
string(REPLACE "-" "_" supported ${function}_library)
check_library_exists(${library} ${function} ${header} ${supported})
if(${${supported}})
add_definitions("-D${var}")
endif()
endfunction()
Various contrib/depends fixes - set WITH_SYSTEMD=OFF in contrib/depends so that we don't try to go look for it on the host system. - link icu libs properly - don't build embedded zmq but rather set up a target so that loki-mq uses the externally built one - fix OpenSSL::Crypto not properly depending on ws2_32 - build boost atomic in the built boost because boost::thread depends on it (and we were just getting lucky before by not happening to touch anything that needed it) - make bundled unbound link against the `extra` interface for various required windows crap (ws2_32 and other stuff).
2020-03-17 01:56:48 +01:00
# Properly links a target to a list of library names by finding the given libraries. Takes:
# - a target
# - a linktype (e.g. INTERFACE, PUBLIC, PRIVATE)
# - a library search path (or "" for defaults)
# - any number of library names
function(link_dep_libs target linktype libdirs)
foreach(lib ${ARGN})
find_library(link_lib-${lib} NAMES ${lib} PATHS ${libdirs})
if(link_lib-${lib})
target_link_libraries(${target} ${linktype} ${link_lib-${lib}})
endif()
endforeach()
endfunction()
cmake: define a default build type Lets 'cmake .. && make' be sufficient to build.
2016-08-28 12:53:46 +02:00
if(NOT CMAKE_BUILD_TYPE)
set(CMAKE_BUILD_TYPE Release CACHE STRING "Build type" FORCE)
message(STATUS "Setting default build type: ${CMAKE_BUILD_TYPE}")
endif()
Set up LTO properly CMake 3.9+ has generic LTO enabling code, switch to it. Update required cmake version to 3.10 (3.9 is probably sufficient, but 3.10 is bionic's version that we're actually testing).
2020-06-12 22:24:03 +02:00
cmake_policy(SET CMP0069 NEW)
SET(CMAKE_POLICY_DEFAULT_CMP0069 NEW)
Android build infrastructure updates - Various static deps updated needed for properly cross-compiling libraries for android. - disable LTO because it doesn't work reliably with the android NDK. - allow building without miniupnpc, and default to not building it under android. - make sure we build translation tools on the native arch when cross-compiling for android. - don't build ncurses, libusb, hidapi when doing an android build
2020-10-24 20:29:47 +02:00
if(CMAKE_BUILD_TYPE STREQUAL Release AND NOT MINGW AND NOT ANDROID)
Static builds: make usable binaries from cmake This adds a static dependency script for libraries like boost, unbound, etc. to cmake, invokable with: cmake .. -DBUILD_STATIC_DEPS=ON which downloads and builds static versions of all our required dependencies (boost, unbound, openssl, ncurses, etc.). It also implies -DSTATIC=ON to build other vendored deps (like miniupnpc, lokimq) as static as well. Unlike the contrib/depends system, this is easier to maintain (one script using nicer cmake with functions instead of raw Makefile spaghetti code), and isn't concerned with reproducible builds -- this doesn't rebuild the compiler, for instance. It also works with the existing build system so that it is simply another way to invoke the cmake build scripts but doesn't require any external tooling. This works on Linux, Mac, and Windows. Some random comments on this commit (for preserving history): - Don't use target_link_libraries on imported targets. Newer cmake is fine with it, but Bionic's cmake doesn't like it but seems okay with setting the properties directly. - This rebuilds libzmq and libsodium, even though there is some provision already within loki-core to do so: however, the existing embedded libzmq fails with the static deps because it uses libzmq's cmake build script, which relies on pkg-config to find libsodium which ends up finding the system one (or not finding any), rather than the one we build with DownloadLibSodium. Since both libsodium and libzmq are faily simple builds it seemed easiest to just add them to the cmake static build rather than trying to shoehorn the current code into the static build script. - Half of the protobuf build system ignores CC/CXX just because Google, and there's no documentation anywhere except for a random closed bug report about needing to set these other variables (CC_FOR_BUILD, CXX_FOR_BUILD) instead, but you need to. Thanks Google. - The boost build is set to output very little because even the minimum -d1 output level spams ~15k lines of output just for the headers it installs.
2020-06-11 20:19:44 +02:00
set(USE_LTO_DEFAULT ON)
else()
set(USE_LTO_DEFAULT OFF)
Set up LTO properly CMake 3.9+ has generic LTO enabling code, switch to it. Update required cmake version to 3.10 (3.9 is probably sufficient, but 3.10 is bionic's version that we're actually testing).
2020-06-12 22:24:03 +02:00
endif()
Static builds: make usable binaries from cmake This adds a static dependency script for libraries like boost, unbound, etc. to cmake, invokable with: cmake .. -DBUILD_STATIC_DEPS=ON which downloads and builds static versions of all our required dependencies (boost, unbound, openssl, ncurses, etc.). It also implies -DSTATIC=ON to build other vendored deps (like miniupnpc, lokimq) as static as well. Unlike the contrib/depends system, this is easier to maintain (one script using nicer cmake with functions instead of raw Makefile spaghetti code), and isn't concerned with reproducible builds -- this doesn't rebuild the compiler, for instance. It also works with the existing build system so that it is simply another way to invoke the cmake build scripts but doesn't require any external tooling. This works on Linux, Mac, and Windows. Some random comments on this commit (for preserving history): - Don't use target_link_libraries on imported targets. Newer cmake is fine with it, but Bionic's cmake doesn't like it but seems okay with setting the properties directly. - This rebuilds libzmq and libsodium, even though there is some provision already within loki-core to do so: however, the existing embedded libzmq fails with the static deps because it uses libzmq's cmake build script, which relies on pkg-config to find libsodium which ends up finding the system one (or not finding any), rather than the one we build with DownloadLibSodium. Since both libsodium and libzmq are faily simple builds it seemed easiest to just add them to the cmake static build rather than trying to shoehorn the current code into the static build script. - Half of the protobuf build system ignores CC/CXX just because Google, and there's no documentation anywhere except for a random closed bug report about needing to set these other variables (CC_FOR_BUILD, CXX_FOR_BUILD) instead, but you need to. Thanks Google. - The boost build is set to output very little because even the minimum -d1 output level spams ~15k lines of output just for the headers it installs.
2020-06-11 20:19:44 +02:00
option(USE_LTO "Use Link-Time Optimization" ${USE_LTO_DEFAULT})
Set up LTO properly CMake 3.9+ has generic LTO enabling code, switch to it. Update required cmake version to 3.10 (3.9 is probably sufficient, but 3.10 is bionic's version that we're actually testing).
2020-06-12 22:24:03 +02:00
if(USE_LTO)
include(CheckIPOSupported)
check_ipo_supported(RESULT IPO_ENABLED OUTPUT ipo_error)
if(IPO_ENABLED)
message(STATUS "LTO enabled")
else()
message(WARNING "LTO not supported by compiler: ${ipo_error}")
endif()
else()
message(STATUS "LTO disabled")
set(IPO_ENABLED OFF)
endif()
if(IPO_ENABLED AND NOT DEFINED CMAKE_INTERPROCEDURAL_OPTIMIZATION)
set(CMAKE_INTERPROCEDURAL_OPTIMIZATION ON)
endif()
Fix designated initializers + add CD/CI badge (#100) * add gitlab config file for CD/CI * formatting * Update readme to display CI badge * Replace designated initializers with aggregate initializers * CD/CI for mac * use git fetch for cli instead of clone * CD/CI for windows
2018-07-20 07:16:12 +02:00
# On Darwin, ensure the user-defined paths are used to find PCSC
# before falling back to the system frameworks.
set(CMAKE_FIND_FRAMEWORK "LAST")
cmake: don't set ARCH from CMAKE_SYSTEM_PROCESSOR It is not correct to do so, because ARCH should only take values supported by the -march argument, with the exception of 'default' which denotes not passing -march at all. ARCH defines the target architecture for builds that are intended to be portable to other machines.
2016-08-28 07:37:34 +02:00
# ARCH defines the target architecture, either by an explicit identifier or
# one of the following two keywords. By default, ARCH a value of 'native':
# target arch = host arch, binary is not portable. When ARCH is set to the
# string 'default', no -march arg is passed, which creates a binary that is
# portable across processors in the same family as host processor. In cases
# when ARCH is not set to an explicit identifier, cmake's builtin is used
# to identify the target architecture, to direct logic in this cmake script.
# Since ARCH is a cached variable, it will not be set on first cmake invocation.
Don't set ARCH_ID if it's already set Usually it's unset, but depends toolchain files set it explicitly for cross-compiling. Don't override preset values. But make sure it's set for all x86 / x86_64 builds, not just Linux. Also make sure -std=c++11 is set for Darwin builds, not all submodules were getting it consistently before.
2019-09-25 05:28:20 +02:00
if (NOT ARCH_ID)
cmake: don't set ARCH from CMAKE_SYSTEM_PROCESSOR It is not correct to do so, because ARCH should only take values supported by the -march argument, with the exception of 'default' which denotes not passing -march at all. ARCH defines the target architecture for builds that are intended to be portable to other machines.
2016-08-28 07:37:34 +02:00
if (NOT ARCH OR ARCH STREQUAL "" OR ARCH STREQUAL "native" OR ARCH STREQUAL "default")
cmake: ARCH_ID fixes for cross compilation
2019-02-22 16:18:32 +01:00
if(CMAKE_SYSTEM_PROCESSOR STREQUAL "")
set(CMAKE_SYSTEM_PROCESSOR ${CMAKE_HOST_SYSTEM_PROCESSOR})
endif()
cmake: don't set ARCH from CMAKE_SYSTEM_PROCESSOR It is not correct to do so, because ARCH should only take values supported by the -march argument, with the exception of 'default' which denotes not passing -march at all. ARCH defines the target architecture for builds that are intended to be portable to other machines.
2016-08-28 07:37:34 +02:00
set(ARCH_ID "${CMAKE_SYSTEM_PROCESSOR}")
else()
set(ARCH_ID "${ARCH}")
cmake: initialize ARCH for native builds We need ARCH, because it needs to be set for ARM7, ARM6 to be initialized. Strangely, on different machines (both ARMv7, Arch), ${ARCH} var is either empty or 'native'. Handle both cases.
2016-06-22 02:07:19 +02:00
endif()
Don't set ARCH_ID if it's already set Usually it's unset, but depends toolchain files set it explicitly for cross-compiling. Don't override preset values. But make sure it's set for all x86 / x86_64 builds, not just Linux. Also make sure -std=c++11 is set for Darwin builds, not all submodules were getting it consistently before.
2019-09-25 05:28:20 +02:00
endif()
cmake: quotes around ARCH_ID string It's only blank only if somebody running cmake in MSYS/MinGW (Windows) manually forgets to add -D ARCH, but when it is blank, without quotes those lines are invalid cmake syntax.
2016-09-17 20:31:25 +02:00
string(TOLOWER "${ARCH_ID}" ARM_ID)
Fix test for 'ARM_ID' as caught by @radfish in #1088
2016-09-18 23:56:38 +02:00
string(SUBSTRING "${ARM_ID}" 0 3 ARM_TEST)
cmake: define ARM var for all ARM arch variants This is refactoring only. No behavior change.
2016-08-28 13:28:05 +02:00
if (ARM_TEST STREQUAL "arm")
set(ARM 1)
Fix test for 'ARM_ID' as caught by @radfish in #1088
2016-09-18 23:56:38 +02:00
string(SUBSTRING "${ARM_ID}" 0 5 ARM_TEST)
cmake: define ARM var for all ARM arch variants This is refactoring only. No behavior change.
2016-08-28 13:28:05 +02:00
if (ARM_TEST STREQUAL "armv6")
set(ARM6 1)
endif()
if (ARM_TEST STREQUAL "armv7")
set(ARM7 1)
Replace tabs with two spaces for consistency with rest of codebase Remove trailing whitespace in same files.
2015-12-14 05:54:39 +01:00
endif()
** CHANGES ARE EXPERIMENTAL (FOR TESTING ONLY) Bockchain: 1. Optim: Multi-thread long-hash computation when encountering groups of blocks. 2. Optim: Cache verified txs and return result from cache instead of re-checking whenever possible. 3. Optim: Preload output-keys when encoutering groups of blocks. Sort by amount and global-index before bulk querying database and multi-thread when possible. 4. Optim: Disable double spend check on block verification, double spend is already detected when trying to add blocks. 5. Optim: Multi-thread signature computation whenever possible. 6. Patch: Disable locking (recursive mutex) on called functions from check_tx_inputs which causes slowdowns (only seems to happen on ubuntu/VMs??? Reason: TBD) 7. Optim: Removed looped full-tx hash computation when retrieving transactions from pool (???). 8. Optim: Cache difficulty/timestamps (735 blocks) for next-difficulty calculations so that only 2 db reads per new block is needed when a new block arrives (instead of 1470 reads). Berkeley-DB: 1. Fix: 32-bit data errors causing wrong output global indices and failure to send blocks to peers (etc). 2. Fix: Unable to pop blocks on reorganize due to transaction errors. 3. Patch: Large number of transaction aborts when running multi-threaded bulk queries. 4. Patch: Insufficient locks error when running full sync. 5. Patch: Incorrect db stats when returning from an immediate exit from "pop block" operation. 6. Optim: Add bulk queries to get output global indices. 7. Optim: Modified output_keys table to store public_key+unlock_time+height for single transaction lookup (vs 3) 8. Optim: Used output_keys table retrieve public_keys instead of going through output_amounts->output_txs+output_indices->txs->output:public_key 9. Optim: Added thread-safe buffers used when multi-threading bulk queries. 10. Optim: Added support for nosync/write_nosync options for improved performance (*see --db-sync-mode option for details) 11. Mod: Added checkpoint thread and auto-remove-logs option. 12. *Now usable on 32-bit systems like RPI2. LMDB: 1. Optim: Added custom comparison for 256-bit key tables (minor speed-up, TBD: get actual effect) 2. Optim: Modified output_keys table to store public_key+unlock_time+height for single transaction lookup (vs 3) 3. Optim: Used output_keys table retrieve public_keys instead of going through output_amounts->output_txs+output_indices->txs->output:public_key 4. Optim: Added support for sync/writemap options for improved performance (*see --db-sync-mode option for details) 5. Mod: Auto resize to +1GB instead of multiplier x1.5 ETC: 1. Minor optimizations for slow-hash for ARM (RPI2). Incomplete. 2. Fix: 32-bit saturation bug when computing next difficulty on large blocks. [PENDING ISSUES] 1. Berkely db has a very slow "pop-block" operation. This is very noticeable on the RPI2 as it sometimes takes > 10 MINUTES to pop a block during reorganization. This does not happen very often however, most reorgs seem to take a few seconds but it possibly depends on the number of outputs present. TBD. 2. Berkeley db, possible bug "unable to allocate memory". TBD. [NEW OPTIONS] (*Currently all enabled for testing purposes) 1. --fast-block-sync arg=[0:1] (default: 1) a. 0 = Compute long hash per block (may take a while depending on CPU) b. 1 = Skip long-hash and verify blocks based on embedded known good block hashes (faster, minimal CPU dependence) 2. --db-sync-mode arg=[[safe|fast|fastest]:[sync|async]:[nblocks_per_sync]] (default: fastest:async:1000) a. safe = fdatasync/fsync (or equivalent) per stored block. Very slow, but safest option to protect against power-out/crash conditions. b. fast/fastest = Enables asynchronous fdatasync/fsync (or equivalent). Useful for battery operated devices or STABLE systems with UPS and/or systems with battery backed write cache/solid state cache. Fast - Write meta-data but defer data flush. Fastest - Defer meta-data and data flush. Sync - Flush data after nblocks_per_sync and wait. Async - Flush data after nblocks_per_sync but do not wait for the operation to finish. 3. --prep-blocks-threads arg=[n] (default: 4 or system max threads, whichever is lower) Max number of threads to use when computing long-hash in groups. 4. --show-time-stats arg=[0:1] (default: 1) Show benchmark related time stats. 5. --db-auto-remove-logs arg=[0:1] (default: 1) For berkeley-db only. Auto remove logs if enabled. **Note: lmdb and berkeley-db have changes to the tables and are not compatible with official git head version. At the moment, you need a full resync to use this optimized version. [PERFORMANCE COMPARISON] **Some figures are approximations only. Using a baseline machine of an i7-2600K+SSD+(with full pow computation): 1. The optimized lmdb/blockhain core can process blocks up to 585K for ~1.25 hours + download time, so it usually takes 2.5 hours to sync the full chain. 2. The current head with memory can process blocks up to 585K for ~4.2 hours + download time, so it usually takes 5.5 hours to sync the full chain. 3. The current head with lmdb can process blocks up to 585K for ~32 hours + download time and usually takes 36 hours to sync the full chain. Averate procesing times (with full pow computation): lmdb-optimized: 1. tx_ave = 2.5 ms / tx 2. block_ave = 5.87 ms / block memory-official-repo: 1. tx_ave = 8.85 ms / tx 2. block_ave = 19.68 ms / block lmdb-official-repo (0f4a036437fd41a5498ee5e74e2422ea6177aa3e) 1. tx_ave = 47.8 ms / tx 2. block_ave = 64.2 ms / block **Note: The following data denotes processing times only (does not include p2p download time) lmdb-optimized processing times (with full pow computation): 1. Desktop, Quad-core / 8-threads 2600k (8Mb) - 1.25 hours processing time (--db-sync-mode=fastest:async:1000). 2. Laptop, Dual-core / 4-threads U4200 (3Mb) - 4.90 hours processing time (--db-sync-mode=fastest:async:1000). 3. Embedded, Quad-core / 4-threads Z3735F (2x1Mb) - 12.0 hours processing time (--db-sync-mode=fastest:async:1000). lmdb-optimized processing times (with per-block-checkpoint) 1. Desktop, Quad-core / 8-threads 2600k (8Mb) - 10 minutes processing time (--db-sync-mode=fastest:async:1000). berkeley-db optimized processing times (with full pow computation) 1. Desktop, Quad-core / 8-threads 2600k (8Mb) - 1.8 hours processing time (--db-sync-mode=fastest:async:1000). 2. RPI2. Improved from estimated 3 months(???) into 2.5 days (*Need 2AMP supply + Clock:1Ghz + [usb+ssd] to achieve this speed) (--db-sync-mode=fastest:async:1000). berkeley-db optimized processing times (with per-block-checkpoint) 1. RPI2. 12-15 hours (*Need 2AMP supply + Clock:1Ghz + [usb+ssd] to achieve this speed) (--db-sync-mode=fastest:async:1000).
2015-07-10 22:09:32 +02:00
endif()
Fix ARM64 identification The actual arch flag the compiler recognizes is "armv8-a". This is true for both gcc and clang.
2017-04-13 00:54:33 +02:00
if (ARM_ID STREQUAL "aarch64" OR ARM_ID STREQUAL "arm64" OR ARM_ID STREQUAL "armv8-a")
Add ARMv8 Handling to CMakeLists.txt - version 2 Adds 64-bit ARMv8 handling to CMakeLists.txt and implements GCC flags for two errata on the Cortex-A53 present on every chip in revision 0: https://developer.arm.com/docs/epm048406/latest/arm-processor-cortex-a53-mpcore-product-revision-r0-software-developers-errata-notice 835769 affects 64-bit multiply accumulate 843419 affects internal page addressing Rewritten to build on @radfish's changes Updated to address @radfish's points
2016-08-29 12:13:55 +02:00
set(ARM 1)
set(ARM8 1)
Fix ARM64 identification The actual arch flag the compiler recognizes is "armv8-a". This is true for both gcc and clang.
2017-04-13 00:54:33 +02:00
set(ARCH "armv8-a")
Add ARMv8 Handling to CMakeLists.txt - version 2 Adds 64-bit ARMv8 handling to CMakeLists.txt and implements GCC flags for two errata on the Cortex-A53 present on every chip in revision 0: https://developer.arm.com/docs/epm048406/latest/arm-processor-cortex-a53-mpcore-product-revision-r0-software-developers-errata-notice 835769 affects 64-bit multiply accumulate 843419 affects internal page addressing Rewritten to build on @radfish's changes Updated to address @radfish's points
2016-08-29 12:13:55 +02:00
endif()
enable monero build on ppc64le architecture Fix wrong flags for ppc64le: -maes - not avaiable -march - changed by -mcpu that does the same for powerpc
2017-06-27 19:06:01 +02:00
if(ARCH_ID STREQUAL "ppc64le")
set(PPC64LE 1)
Fixed for compilation in big endian PPC. "Everything is broken thought due to little endian logic."
2018-05-20 00:03:59 +02:00
set(PPC64 0)
set(PPC 0)
endif()
if(ARCH_ID STREQUAL "powerpc64" OR ARCH_ID STREQUAL "ppc64")
set(PPC64LE 0)
set(PPC64 1)
set(PPC 0)
enable monero build on ppc64le architecture Fix wrong flags for ppc64le: -maes - not avaiable -march - changed by -mcpu that does the same for powerpc
2017-06-27 19:06:01 +02:00
endif()
build: ubuntu 16 ppc build fixes
2018-11-01 04:11:09 +01:00
if(ARCH_ID STREQUAL "powerpc" OR ARCH_ID STREQUAL "ppc")
Fixed for compilation in big endian PPC. "Everything is broken thought due to little endian logic."
2018-05-20 00:03:59 +02:00
set(PPC64LE 0)
set(PPC64 0)
set(PPC 1)
endif()
disable AES on s390x architecture rebased by moneromooo
2018-05-01 07:13:10 +02:00
if(ARCH_ID STREQUAL "s390x")
set(S390X 1)
endif()
Optionally query moneropulse DNS records to check for updates It just checks and prints a message if there is a new version for now.
2017-02-18 17:30:29 +01:00
# BUILD_TAG is used to select the build type to check for a new version
if(BUILD_TAG)
message(STATUS "Building build tag ${BUILD_TAG}")
add_definitions("-DBUILD_TAG=${BUILD_TAG}")
else()
message(STATUS "Building without build tag")
endif()
Update and submodule gtest The gtest version bundled inside tests/ is ancient (7 years old) and doesn't build properly for some compilers. Replace it with a current gtest submodule in external/.
2020-06-10 00:54:13 +02:00
enable_testing()
option(BUILD_DOCUMENTATION "Build the Doxygen documentation." ON)
option(BUILD_TESTS "Build tests." OFF)
if (BUILD_TESTS)
add_definitions(-DUNIT_TEST)
endif()
Disable Trezor by default Trezor support currently doesn't work (Trezor's hardware wallet is hard-coded for a Monero net-id), and this saves needing to muck around with protobuf.
2021-01-08 20:42:33 +01:00
OPTION(USE_DEVICE_TREZOR "Build Trezor support (currently non-functional)" OFF)
option to exclude making pybind on android and windows by default
2022-12-05 21:33:56 +01:00
if(NOT MINGW AND NOT ANDROID)
set(USE_PYBIND_DEFAULT ON)
else()
set(USE_PYBIND_DEFAULT OFF)
endif()
option(BUILD_PYBIND "build python bindings" ${USE_PYBIND_DEFAULT})
Static builds: make usable binaries from cmake This adds a static dependency script for libraries like boost, unbound, etc. to cmake, invokable with: cmake .. -DBUILD_STATIC_DEPS=ON which downloads and builds static versions of all our required dependencies (boost, unbound, openssl, ncurses, etc.). It also implies -DSTATIC=ON to build other vendored deps (like miniupnpc, lokimq) as static as well. Unlike the contrib/depends system, this is easier to maintain (one script using nicer cmake with functions instead of raw Makefile spaghetti code), and isn't concerned with reproducible builds -- this doesn't rebuild the compiler, for instance. It also works with the existing build system so that it is simply another way to invoke the cmake build scripts but doesn't require any external tooling. This works on Linux, Mac, and Windows. Some random comments on this commit (for preserving history): - Don't use target_link_libraries on imported targets. Newer cmake is fine with it, but Bionic's cmake doesn't like it but seems okay with setting the properties directly. - This rebuilds libzmq and libsodium, even though there is some provision already within loki-core to do so: however, the existing embedded libzmq fails with the static deps because it uses libzmq's cmake build script, which relies on pkg-config to find libsodium which ends up finding the system one (or not finding any), rather than the one we build with DownloadLibSodium. Since both libsodium and libzmq are faily simple builds it seemed easiest to just add them to the cmake static build rather than trying to shoehorn the current code into the static build script. - Half of the protobuf build system ignores CC/CXX just because Google, and there's no documentation anywhere except for a random closed bug report about needing to set these other variables (CC_FOR_BUILD, CXX_FOR_BUILD) instead, but you need to. Thanks Google. - The boost build is set to output very little because even the minimum -d1 output level spams ~15k lines of output just for the headers it installs.
2020-06-11 20:19:44 +02:00
find_package(Git)
Add check if submodules need to be updated Adds CMake check that pulls from the different git remotes and checks if there is any output.
2018-10-04 15:07:00 +02:00
if(NOT MANUAL_SUBMODULES)
if(GIT_FOUND)
build: fixed submodule check
2018-10-16 16:43:56 +02:00
function (check_submodule relative_path)
build: older git versions don't have '-C' flag
2018-10-30 06:04:57 +01:00
execute_process(COMMAND git rev-parse "HEAD" WORKING_DIRECTORY ${CMAKE_CURRENT_SOURCE_DIR}/${relative_path} OUTPUT_VARIABLE localHead)
execute_process(COMMAND git rev-parse "HEAD:${relative_path}" WORKING_DIRECTORY ${CMAKE_CURRENT_SOURCE_DIR} OUTPUT_VARIABLE checkedHead)
build: fixed submodule check
2018-10-16 16:43:56 +02:00
string(COMPARE EQUAL "${localHead}" "${checkedHead}" upToDate)
if (upToDate)
message(STATUS "Submodule '${relative_path}' is up-to-date")
else()
Convert quorumnet to loki-mq This adds the loki-mq dependency and replaces SNNetwork with it (along with some syntax updates for how loki-mq changed a bit from SNNetwork). This also replaces common/hex.h and common/string_view.h with loki-mq's faster (hex) and more complete and tested (string_view) implementations.
2020-03-03 05:38:18 +01:00
message(FATAL_ERROR "Submodule '${relative_path}' is not up-to-date. Please update with\ngit submodule update --init --recursive\nor run cmake with -DMANUAL_SUBMODULES=1")
build: fixed submodule check
2018-10-16 16:43:56 +02:00
endif()
Convert quorumnet to loki-mq This adds the loki-mq dependency and replaces SNNetwork with it (along with some syntax updates for how loki-mq changed a bit from SNNetwork). This also replaces common/hex.h and common/string_view.h with loki-mq's faster (hex) and more complete and tested (string_view) implementations.
2020-03-03 05:38:18 +01:00
# Extra arguments check nested submodules
foreach(submod ${ARGN})
execute_process(COMMAND git rev-parse "HEAD" WORKING_DIRECTORY ${CMAKE_CURRENT_SOURCE_DIR}/${relative_path}/${submod} OUTPUT_VARIABLE localHead)
execute_process(COMMAND git rev-parse "HEAD:${submod}" WORKING_DIRECTORY ${CMAKE_CURRENT_SOURCE_DIR}/${relative_path} OUTPUT_VARIABLE checkedHead)
string(COMPARE EQUAL "${localHead}" "${checkedHead}" upToDate)
if (NOT upToDate)
message(FATAL_ERROR "Nested submodule '${relative_path}/${submod}' is not up-to-date. Please update with\ngit submodule update --init --recursive\nor run cmake with -DMANUAL_SUBMODULES=1")
endif()
endforeach()
build: fixed submodule check
2018-10-16 16:43:56 +02:00
endfunction ()
Add check if submodules need to be updated Adds CMake check that pulls from the different git remotes and checks if there is any output.
2018-10-04 15:07:00 +02:00
message(STATUS "Checking submodules")
build: fixed submodule check
2018-10-16 16:43:56 +02:00
check_submodule(external/rapidjson)
device/trezor: trezor support added
2018-08-23 23:50:53 +02:00
check_submodule(external/trezor-common)
RandomX integration Support RandomX PoW algorithm
2019-04-23 21:32:27 +02:00
check_submodule(external/randomx)
Clean up sys-or-submodule code, support system fmt, rename loki-mq Synced up with oxen-storage-server. Also rename loki-mq submodule to oxen-mq.
2022-05-30 20:23:06 +02:00
check_submodule(external/oxen-mq cppzmq)
Batching of service node rewards This updates the coinbase transactions to reward service nodes periodically rather than every block. If you recieve a service node reward this reward will be delayed x blocks, if you receive another reward to the same wallet before those blocks have been completed it will be added to your total and all will be paid out after those x blocks has passed. For example if our batching interval is 2 blocks: Block 1 - Address A receives reward of 10 oxen - added to batch Block 2 - Address A receives reward of 10 oxen - added to batch Block 3 - Address A is paid out 20 oxen. Batching accumulates a small reward for all nodes every block The batching of service node rewards allows us to drip feed rewards to service nodes. Rather than accruing each service node 16.5 oxen every time they are pulse block leader we now reward every node the 16.5 / num_service_nodes every block and pay each wallet the full amount that has been accrued after a period of time (Likely 3.5 days). To spread each payment evenly we now pay the rewards based on the address of the recipient. This modulus of their address determines which block the address should be paid and by setting the interval to our service_node_batching interval we can guarantee they will be paid out regularly and evenly distribute the payments for all wallets over this
2022-04-29 01:51:14 +02:00
check_submodule(external/SQLiteCpp)
Update and submodule gtest The gtest version bundled inside tests/ is ancient (7 years old) and doesn't build properly for some compilers. Replace it with a current gtest submodule in external/.
2020-06-10 00:54:13 +02:00
if(BUILD_TESTS)
check_submodule(external/googletest)
Wallet3 squashed initial commit tx scanning for basic transactions working - TODO: subadresses. The scanning code is there, but it does not currently know/care about any subaddresses. daemon comms for basic syncing working (multi-)wallet sync more or less working properly - seem to have a dangling shared_ptr somewhere when removing a wallet from daemon comms, so not working perfectly yet. Lots of TODOs and cleanup needed, as well as further features of course.
2021-07-19 20:42:38 +02:00
check_submodule(external/Catch2)
Update and submodule gtest The gtest version bundled inside tests/ is ancient (7 years old) and doesn't build properly for some compilers. Replace it with a current gtest submodule in external/.
2020-06-10 00:54:13 +02:00
endif()
Replace epee http rpc server with uWebSockets This replaces the NIH epee http server which does not work all that well with an external C++ library called uWebSockets. Fundamentally this gives the following advantages: - Much less code to maintain - Just one thread for handling HTTP connections versus epee's pool of threads - Uses existing LokiMQ job server and existing thread pool for handling the actual tasks; they are processed/scheduled in the same "rpc" or "admin" queues as lokimq rpc calls. One notable benefit is that "admin" rpc commands get their own queue (and thus cannot be delayed by long rpc commands). Currently the lokimq threads and the http rpc thread pool and the p2p thread pool and the job queue thread pool and the dns lookup thread pool and... are *all* different thread pools; this is a step towards consolidating them. - Very little mutex contention (which has been a major problem with epee RPC in the past): there is one mutex (inside uWebSockets) for putting responses back into the thread managing the connection; everything internally gets handled through (lock-free) lokimq inproc sockets. - Faster RPC performance on average, and much better worst case performance. Epee's http interface seems to have some race condition that ocassionally stalls a request (even a very simple one) for a dozen or more seconds for no good reason. - Long polling gets redone here to no longer need threads; instead we just store the request and respond when the thread pool, or else in a timer (that runs once/second) for timing out long polls. --- The basic idea of how this works from a high level: We launch a single thread to handle HTTP RPC requests and response data. This uWebSockets thread is essentially running an event loop: it never actually handles any logic; it only serves to shuttle data that arrives in a request to some other thread, and then, at some later point, to send some reply back to that waiting connection. Everything is asynchronous and non-blocking here: the basic uWebSockets event loop just operates as things arrive, passes it off immediately, and goes back to waiting for the next thing to arrive. The basic flow is like this: 0. uWS thread -- listens on localhost:22023 1. uWS thread -- incoming request on localhost:22023 2. uWS thread -- fires callback, which injects the task into the LokiMQ job queue 3. LMQ main loop -- schedules it as an RPC job 4. LMQ rpc thread -- Some LokiMQ thread runs it, gets the result 5. LMQ rpc thread -- Result gets queued up for the uWS thread 6. uWS thread -- takes the request and starts sending it (asynchronously) back to the requestor. In more detail: uWebSockets has registered has registered handlers for non-jsonrpc requests (legacy JSON or binary). If the port is restricted then admin commands get mapped to a "Access denied" response handler, otherwise public commands (and admin commands on an unrestricted port) go to the rpc command handler. POST requests to /json_rpc have their own handler; this is a little different than the above because it has to parse the request before it can determine whether it is allowed or not, but once this is done it continues roughly the same as legacy/binary requests. uWebSockets then listens on the given IP/port for new incoming requests, and starts listening for requests in a thread (we own this thread). When a request arrives, it fires the event handler for that request. (This may happen multiple times, if the client is sending a bunch of data in a POST request). Once we have the full request, we then queue the job in LokiMQ, putting it in the "rpc" or "admin" command categories. (The one practical different here is that "admin" is configured to be allowed to start up its own thread if all other threads are busy, while "rpc" commands are prioritized along with everything else.) LokiMQ then schedules this, along with native LokiMQ "rpc." or "admin." requests. When a LMQ worker thread becomes available, the RPC command gets called in it and runs. Whatever output it produces (or error message, if it throws) then gets wrapped up in jsonrpc boilerplate (if necessary), and delivered to the uWebSockets thread to be sent in reply to that request. uWebSockets picks up the data and sends whatever it can without blocking, then buffers whatever it couldn't send to be sent again in a later event loop iteration once the requestor can accept more data. (This part is outside lokid; we only have to give uWS the data and let it worry about delivery). --- PR specifics: Things removed from this PR: 1. ssl settings; with this PR the HTTP RPC interface is plain-text. The previous default generated a self-signed certificate for the server on startup and then the client accepted any certificate. This is actually *worse* than unencrypted because it is entirely MITM-readable and yet might make people think that their RPC communication is encrypted, and setting up actual certificates is difficult enough that I think most people don't bother. uWebSockets *does* support HTTPS, and we could glue the existing options into it, but I'm not convinced it's worthwhile: it works much better to put HTTPS in a front-end proxy holding the certificate that proxies requests to the backend (which can then listen in restricted mode on some localhost port). One reason this is better is that it is much easier to reload and/or restart such a front-end server, while certificate updates with lokid require a full restart. Another reason is that you get an error page instead of a timeout if something is wrong with the backend. Finally we also save having to generate a temporary certificate on *every* lokid invocation. 2. HTTP Digest authentication. Digest authentication is obsolete (and was already obsolete when it got added to Monero). HTTP-Digest was originally an attempt to provide a password authentication mechanism that does not leak the password in transit, but still required that the server know the password. It only has marginal value against replay attacks, and is made entirely obsolete by sending traffic over HTTPS instead. No client out there supports Digest but *not* Basic auth, and so given the limited usefulness it seems pointless to support more than Basic auth for HTTP RPC login. What's worse is that epee's HTTP Digest authentication is a terrible implementation: it uses boost::spirit -- a recursive descent parser meant for building complex language grammars -- just to parse a single HTTP header for Digest auth. This is a big load of crap that should never have been accepted upstream, and that we should get rid of (even if we wanted to support Digest auth it takes less than 100 lines of code to do it when *not* using a recursive descent parser).
2020-06-29 01:23:06 +02:00
check_submodule(external/uWebSockets uSockets)
std::filesystem Converts all use of boost::filesystem to std::filesystem. For macos and potentially other exotic systems where std::filesystem isn't available, we use ghc::filesystem instead (which is a drop-in replacement for std::filesystem, unlike boost::filesystem). This also greatly changes how we handle filenames internally by holding them in filesystem::path objects as soon as possible (using fs::u8path()), rather than strings, which avoids a ton of issues around unicode filenames. As a result this lets us drop the boost::locale dependency on Windows along with a bunch of messy Windows ifdef code, and avoids the need for doing gross boost locale codecvt calls.
2020-10-22 19:55:33 +02:00
check_submodule(external/ghc-filesystem)
wallet3 cli This introduces wallet3 cli to the codebase, build using python and wraps the c++ wallet3 core using pybind.
2022-10-23 22:08:52 +02:00
check_submodule(external/SQLiteCpp)
check_submodule(external/pybind11)
Add check if submodules need to be updated Adds CMake check that pulls from the different git remotes and checks if there is any output.
2018-10-04 15:07:00 +02:00
endif()
endif()
** CHANGES ARE EXPERIMENTAL (FOR TESTING ONLY) Bockchain: 1. Optim: Multi-thread long-hash computation when encountering groups of blocks. 2. Optim: Cache verified txs and return result from cache instead of re-checking whenever possible. 3. Optim: Preload output-keys when encoutering groups of blocks. Sort by amount and global-index before bulk querying database and multi-thread when possible. 4. Optim: Disable double spend check on block verification, double spend is already detected when trying to add blocks. 5. Optim: Multi-thread signature computation whenever possible. 6. Patch: Disable locking (recursive mutex) on called functions from check_tx_inputs which causes slowdowns (only seems to happen on ubuntu/VMs??? Reason: TBD) 7. Optim: Removed looped full-tx hash computation when retrieving transactions from pool (???). 8. Optim: Cache difficulty/timestamps (735 blocks) for next-difficulty calculations so that only 2 db reads per new block is needed when a new block arrives (instead of 1470 reads). Berkeley-DB: 1. Fix: 32-bit data errors causing wrong output global indices and failure to send blocks to peers (etc). 2. Fix: Unable to pop blocks on reorganize due to transaction errors. 3. Patch: Large number of transaction aborts when running multi-threaded bulk queries. 4. Patch: Insufficient locks error when running full sync. 5. Patch: Incorrect db stats when returning from an immediate exit from "pop block" operation. 6. Optim: Add bulk queries to get output global indices. 7. Optim: Modified output_keys table to store public_key+unlock_time+height for single transaction lookup (vs 3) 8. Optim: Used output_keys table retrieve public_keys instead of going through output_amounts->output_txs+output_indices->txs->output:public_key 9. Optim: Added thread-safe buffers used when multi-threading bulk queries. 10. Optim: Added support for nosync/write_nosync options for improved performance (*see --db-sync-mode option for details) 11. Mod: Added checkpoint thread and auto-remove-logs option. 12. *Now usable on 32-bit systems like RPI2. LMDB: 1. Optim: Added custom comparison for 256-bit key tables (minor speed-up, TBD: get actual effect) 2. Optim: Modified output_keys table to store public_key+unlock_time+height for single transaction lookup (vs 3) 3. Optim: Used output_keys table retrieve public_keys instead of going through output_amounts->output_txs+output_indices->txs->output:public_key 4. Optim: Added support for sync/writemap options for improved performance (*see --db-sync-mode option for details) 5. Mod: Auto resize to +1GB instead of multiplier x1.5 ETC: 1. Minor optimizations for slow-hash for ARM (RPI2). Incomplete. 2. Fix: 32-bit saturation bug when computing next difficulty on large blocks. [PENDING ISSUES] 1. Berkely db has a very slow "pop-block" operation. This is very noticeable on the RPI2 as it sometimes takes > 10 MINUTES to pop a block during reorganization. This does not happen very often however, most reorgs seem to take a few seconds but it possibly depends on the number of outputs present. TBD. 2. Berkeley db, possible bug "unable to allocate memory". TBD. [NEW OPTIONS] (*Currently all enabled for testing purposes) 1. --fast-block-sync arg=[0:1] (default: 1) a. 0 = Compute long hash per block (may take a while depending on CPU) b. 1 = Skip long-hash and verify blocks based on embedded known good block hashes (faster, minimal CPU dependence) 2. --db-sync-mode arg=[[safe|fast|fastest]:[sync|async]:[nblocks_per_sync]] (default: fastest:async:1000) a. safe = fdatasync/fsync (or equivalent) per stored block. Very slow, but safest option to protect against power-out/crash conditions. b. fast/fastest = Enables asynchronous fdatasync/fsync (or equivalent). Useful for battery operated devices or STABLE systems with UPS and/or systems with battery backed write cache/solid state cache. Fast - Write meta-data but defer data flush. Fastest - Defer meta-data and data flush. Sync - Flush data after nblocks_per_sync and wait. Async - Flush data after nblocks_per_sync but do not wait for the operation to finish. 3. --prep-blocks-threads arg=[n] (default: 4 or system max threads, whichever is lower) Max number of threads to use when computing long-hash in groups. 4. --show-time-stats arg=[0:1] (default: 1) Show benchmark related time stats. 5. --db-auto-remove-logs arg=[0:1] (default: 1) For berkeley-db only. Auto remove logs if enabled. **Note: lmdb and berkeley-db have changes to the tables and are not compatible with official git head version. At the moment, you need a full resync to use this optimized version. [PERFORMANCE COMPARISON] **Some figures are approximations only. Using a baseline machine of an i7-2600K+SSD+(with full pow computation): 1. The optimized lmdb/blockhain core can process blocks up to 585K for ~1.25 hours + download time, so it usually takes 2.5 hours to sync the full chain. 2. The current head with memory can process blocks up to 585K for ~4.2 hours + download time, so it usually takes 5.5 hours to sync the full chain. 3. The current head with lmdb can process blocks up to 585K for ~32 hours + download time and usually takes 36 hours to sync the full chain. Averate procesing times (with full pow computation): lmdb-optimized: 1. tx_ave = 2.5 ms / tx 2. block_ave = 5.87 ms / block memory-official-repo: 1. tx_ave = 8.85 ms / tx 2. block_ave = 19.68 ms / block lmdb-official-repo (0f4a036437fd41a5498ee5e74e2422ea6177aa3e) 1. tx_ave = 47.8 ms / tx 2. block_ave = 64.2 ms / block **Note: The following data denotes processing times only (does not include p2p download time) lmdb-optimized processing times (with full pow computation): 1. Desktop, Quad-core / 8-threads 2600k (8Mb) - 1.25 hours processing time (--db-sync-mode=fastest:async:1000). 2. Laptop, Dual-core / 4-threads U4200 (3Mb) - 4.90 hours processing time (--db-sync-mode=fastest:async:1000). 3. Embedded, Quad-core / 4-threads Z3735F (2x1Mb) - 12.0 hours processing time (--db-sync-mode=fastest:async:1000). lmdb-optimized processing times (with per-block-checkpoint) 1. Desktop, Quad-core / 8-threads 2600k (8Mb) - 10 minutes processing time (--db-sync-mode=fastest:async:1000). berkeley-db optimized processing times (with full pow computation) 1. Desktop, Quad-core / 8-threads 2600k (8Mb) - 1.8 hours processing time (--db-sync-mode=fastest:async:1000). 2. RPI2. Improved from estimated 3 months(???) into 2.5 days (*Need 2AMP supply + Clock:1Ghz + [usb+ssd] to achieve this speed) (--db-sync-mode=fastest:async:1000). berkeley-db optimized processing times (with per-block-checkpoint) 1. RPI2. 12-15 hours (*Need 2AMP supply + Clock:1Ghz + [usb+ssd] to achieve this speed) (--db-sync-mode=fastest:async:1000).
2015-07-10 22:09:32 +02:00
# set this to 0 if per-block checkpoint needs to be disabled
cmake modernization The archaic (i.e. decade old) cmake usage here really got in the way of trying to properly use newer libraries (like lokimq), so this undertakes overhauling it considerably to make it much more sane (and significantly reduce the size). I left more of the architecture-specific bits in the top-level CMakeLists.txt intact; most of the efforts here are about properly loading dependencies, specifying dependencies and avoiding a whole pile of cmake antipatterns. This bumps the required cmake version to 3.5, which is what xenial comes with. - extensive use of interface libraries to include libraries, definitions, and include paths - use Boost::whatever instead of ${Boost_WHATEVER_LIBRARY}. The interface targets are (again) much better as they also give you any needed include or linking flags without needing to worry about them. - don't list header files when building things. This has *never* been correct cmake usage (cmake has always known how to wallet_rpc_headers the headers that .cpp files include to know about build changes). - remove the loki_add_library monstrosity; it breaks target names and makes compiling less efficient because the author couldn't figure out how to link things together. - make loki_add_executable take the output filename, and set the output path to bin/ and install to bin because *every single usage* of loki_add_executable was immediately followed by setting the output filename and setting the output path to bin/ and installing to bin. - move a bunch of crap that is only used in one particular src/whatever/CMakeLists.txt into that particular CMakeLists.txt instead of the top level CMakeLists.txt (or src/CMakeLists.txt). - Remove a bunch of redundant dependencies; most of them look like they were just copy-and-pasted in, and many more aren't needed (since they are implied by the PUBLIC linking of other dependencies). - Removed `die` since it just does a FATAL_ERROR, but adds color (which is useless since CMake already makes FATAL_ERRORs perfectly visible). - Change the way LOKI_DAEMON_AND_WALLET_ONLY works to just change the make targets to daemon and simplewallet rather than changing the build process (this should make it faster, too, since there are various other things that will be excluded).
2020-03-03 04:57:08 +01:00
option(PER_BLOCK_CHECKPOINT "Enables per-block checkpointing" ON)
** CHANGES ARE EXPERIMENTAL (FOR TESTING ONLY) Bockchain: 1. Optim: Multi-thread long-hash computation when encountering groups of blocks. 2. Optim: Cache verified txs and return result from cache instead of re-checking whenever possible. 3. Optim: Preload output-keys when encoutering groups of blocks. Sort by amount and global-index before bulk querying database and multi-thread when possible. 4. Optim: Disable double spend check on block verification, double spend is already detected when trying to add blocks. 5. Optim: Multi-thread signature computation whenever possible. 6. Patch: Disable locking (recursive mutex) on called functions from check_tx_inputs which causes slowdowns (only seems to happen on ubuntu/VMs??? Reason: TBD) 7. Optim: Removed looped full-tx hash computation when retrieving transactions from pool (???). 8. Optim: Cache difficulty/timestamps (735 blocks) for next-difficulty calculations so that only 2 db reads per new block is needed when a new block arrives (instead of 1470 reads). Berkeley-DB: 1. Fix: 32-bit data errors causing wrong output global indices and failure to send blocks to peers (etc). 2. Fix: Unable to pop blocks on reorganize due to transaction errors. 3. Patch: Large number of transaction aborts when running multi-threaded bulk queries. 4. Patch: Insufficient locks error when running full sync. 5. Patch: Incorrect db stats when returning from an immediate exit from "pop block" operation. 6. Optim: Add bulk queries to get output global indices. 7. Optim: Modified output_keys table to store public_key+unlock_time+height for single transaction lookup (vs 3) 8. Optim: Used output_keys table retrieve public_keys instead of going through output_amounts->output_txs+output_indices->txs->output:public_key 9. Optim: Added thread-safe buffers used when multi-threading bulk queries. 10. Optim: Added support for nosync/write_nosync options for improved performance (*see --db-sync-mode option for details) 11. Mod: Added checkpoint thread and auto-remove-logs option. 12. *Now usable on 32-bit systems like RPI2. LMDB: 1. Optim: Added custom comparison for 256-bit key tables (minor speed-up, TBD: get actual effect) 2. Optim: Modified output_keys table to store public_key+unlock_time+height for single transaction lookup (vs 3) 3. Optim: Used output_keys table retrieve public_keys instead of going through output_amounts->output_txs+output_indices->txs->output:public_key 4. Optim: Added support for sync/writemap options for improved performance (*see --db-sync-mode option for details) 5. Mod: Auto resize to +1GB instead of multiplier x1.5 ETC: 1. Minor optimizations for slow-hash for ARM (RPI2). Incomplete. 2. Fix: 32-bit saturation bug when computing next difficulty on large blocks. [PENDING ISSUES] 1. Berkely db has a very slow "pop-block" operation. This is very noticeable on the RPI2 as it sometimes takes > 10 MINUTES to pop a block during reorganization. This does not happen very often however, most reorgs seem to take a few seconds but it possibly depends on the number of outputs present. TBD. 2. Berkeley db, possible bug "unable to allocate memory". TBD. [NEW OPTIONS] (*Currently all enabled for testing purposes) 1. --fast-block-sync arg=[0:1] (default: 1) a. 0 = Compute long hash per block (may take a while depending on CPU) b. 1 = Skip long-hash and verify blocks based on embedded known good block hashes (faster, minimal CPU dependence) 2. --db-sync-mode arg=[[safe|fast|fastest]:[sync|async]:[nblocks_per_sync]] (default: fastest:async:1000) a. safe = fdatasync/fsync (or equivalent) per stored block. Very slow, but safest option to protect against power-out/crash conditions. b. fast/fastest = Enables asynchronous fdatasync/fsync (or equivalent). Useful for battery operated devices or STABLE systems with UPS and/or systems with battery backed write cache/solid state cache. Fast - Write meta-data but defer data flush. Fastest - Defer meta-data and data flush. Sync - Flush data after nblocks_per_sync and wait. Async - Flush data after nblocks_per_sync but do not wait for the operation to finish. 3. --prep-blocks-threads arg=[n] (default: 4 or system max threads, whichever is lower) Max number of threads to use when computing long-hash in groups. 4. --show-time-stats arg=[0:1] (default: 1) Show benchmark related time stats. 5. --db-auto-remove-logs arg=[0:1] (default: 1) For berkeley-db only. Auto remove logs if enabled. **Note: lmdb and berkeley-db have changes to the tables and are not compatible with official git head version. At the moment, you need a full resync to use this optimized version. [PERFORMANCE COMPARISON] **Some figures are approximations only. Using a baseline machine of an i7-2600K+SSD+(with full pow computation): 1. The optimized lmdb/blockhain core can process blocks up to 585K for ~1.25 hours + download time, so it usually takes 2.5 hours to sync the full chain. 2. The current head with memory can process blocks up to 585K for ~4.2 hours + download time, so it usually takes 5.5 hours to sync the full chain. 3. The current head with lmdb can process blocks up to 585K for ~32 hours + download time and usually takes 36 hours to sync the full chain. Averate procesing times (with full pow computation): lmdb-optimized: 1. tx_ave = 2.5 ms / tx 2. block_ave = 5.87 ms / block memory-official-repo: 1. tx_ave = 8.85 ms / tx 2. block_ave = 19.68 ms / block lmdb-official-repo (0f4a036437fd41a5498ee5e74e2422ea6177aa3e) 1. tx_ave = 47.8 ms / tx 2. block_ave = 64.2 ms / block **Note: The following data denotes processing times only (does not include p2p download time) lmdb-optimized processing times (with full pow computation): 1. Desktop, Quad-core / 8-threads 2600k (8Mb) - 1.25 hours processing time (--db-sync-mode=fastest:async:1000). 2. Laptop, Dual-core / 4-threads U4200 (3Mb) - 4.90 hours processing time (--db-sync-mode=fastest:async:1000). 3. Embedded, Quad-core / 4-threads Z3735F (2x1Mb) - 12.0 hours processing time (--db-sync-mode=fastest:async:1000). lmdb-optimized processing times (with per-block-checkpoint) 1. Desktop, Quad-core / 8-threads 2600k (8Mb) - 10 minutes processing time (--db-sync-mode=fastest:async:1000). berkeley-db optimized processing times (with full pow computation) 1. Desktop, Quad-core / 8-threads 2600k (8Mb) - 1.8 hours processing time (--db-sync-mode=fastest:async:1000). 2. RPI2. Improved from estimated 3 months(???) into 2.5 days (*Need 2AMP supply + Clock:1Ghz + [usb+ssd] to achieve this speed) (--db-sync-mode=fastest:async:1000). berkeley-db optimized processing times (with per-block-checkpoint) 1. RPI2. 12-15 hours (*Need 2AMP supply + Clock:1Ghz + [usb+ssd] to achieve this speed) (--db-sync-mode=fastest:async:1000).
2015-07-10 22:09:32 +02:00
cmake: prepend to CMAKE_MODULE_PATH
2014-10-22 21:17:52 +02:00
list(INSERT CMAKE_MODULE_PATH 0
"${CMAKE_SOURCE_DIR}/cmake")
Fixed artifacts from cherry-picking devel->master
2014-09-18 00:37:32 +02:00
Remove cruft
2020-06-12 00:59:20 +02:00
option(BOOST_IGNORE_SYSTEM_PATHS "Ignore boost system paths for local boost installation" OFF)
remerged; commands JSON. logging upgrade. doxygen
2015-04-01 19:00:45 +02:00
moved all stuff to github
2014-03-03 23:07:58 +01:00
set_property(GLOBAL PROPERTY USE_FOLDERS ON)
remerged; commands JSON. logging upgrade. doxygen
2015-04-01 19:00:45 +02:00
moved 32-bit/64-bit detection into main CMakeLists
2015-03-02 16:10:16 +01:00
# Check whether we're on a 32-bit or 64-bit system
if(CMAKE_SIZEOF_VOID_P EQUAL "8")
fixed arch_width option
2015-03-02 17:04:58 +01:00
set(DEFAULT_BUILD_64 ON)
moved 32-bit/64-bit detection into main CMakeLists
2015-03-02 16:10:16 +01:00
else()
fixed arch_width option
2015-03-02 17:04:58 +01:00
set(DEFAULT_BUILD_64 OFF)
endif()
fixed BUILD_64 option
2015-03-03 20:57:40 +01:00
option(BUILD_64 "Build for 64-bit? 'OFF' builds for 32-bit." ${DEFAULT_BUILD_64})
fixed arch_width option
2015-03-02 17:04:58 +01:00
if(BUILD_64)
set(ARCH_WIDTH "64")
else()
set(ARCH_WIDTH "32")
moved 32-bit/64-bit detection into main CMakeLists
2015-03-02 16:10:16 +01:00
endif()
fixed msys2 / mingw folders based on architecture, added license to unbound CMakeList as that is not part of standard Unbound
2015-03-02 16:20:42 +01:00
message(STATUS "Building for a ${ARCH_WIDTH}-bit system")
remerged; commands JSON. logging upgrade. doxygen
2015-04-01 19:00:45 +02:00
exclude local miniupnpc for FreeBSD, install from ports instead
2014-09-09 14:03:42 +02:00
# CMAKE_SYSTEM_NAME checks are commonly known, but specifically taken from libsdl's CMakeLists
Preliminary support for DragonFly BSD - It builds but no further testing has been done.
2016-12-15 11:27:53 +01:00
if(CMAKE_SYSTEM_NAME MATCHES "kFreeBSD.*|FreeBSD")
exclude local miniupnpc for FreeBSD, install from ports instead
2014-09-09 14:03:42 +02:00
set(FREEBSD TRUE)
endif()
Preliminary support for DragonFly BSD - It builds but no further testing has been done.
2016-12-15 11:27:53 +01:00
# Check if we're on DragonFly BSD. See the README.md for build instructions.
if(CMAKE_SYSTEM_NAME MATCHES "DragonFly.*")
set(DRAGONFLY TRUE)
endif()
Added/corrected OpenBSD build instructions.
2016-01-21 19:34:02 +01:00
# Check if we're on OpenBSD. See the README.md for build instructions.
OpenBSD support for Monero.
2016-01-21 19:18:26 +01:00
if(CMAKE_SYSTEM_NAME MATCHES "kOpenBSD.*|OpenBSD.*")
set(OPENBSD TRUE)
endif()
# TODO: check bsdi, NetBSD, to see if they need the same FreeBSD changes
Replace tabs with two spaces for consistency with rest of codebase Remove trailing whitespace in same files.
2015-12-14 05:54:39 +01:00
#
exclude local miniupnpc for FreeBSD, install from ports instead
2014-09-09 14:03:42 +02:00
# elseif(CMAKE_SYSTEM_NAME MATCHES "kNetBSD.*|NetBSD.*")
# set(NETBSD TRUE)
# elseif(CMAKE_SYSTEM_NAME MATCHES ".*BSDI.*")
# set(BSDI TRUE)
std::filesystem Converts all use of boost::filesystem to std::filesystem. For macos and potentially other exotic systems where std::filesystem isn't available, we use ghc::filesystem instead (which is a drop-in replacement for std::filesystem, unlike boost::filesystem). This also greatly changes how we handle filenames internally by holding them in filesystem::path objects as soon as possible (using fs::u8path()), rather than strings, which avoids a ton of issues around unicode filenames. As a result this lets us drop the boost::locale dependency on Windows along with a bunch of messy Windows ifdef code, and avoids the need for doing gross boost locale codecvt calls.
2020-10-22 19:55:33 +02:00
include(cmake/check_for_std_filesystem.cmake)
Set epee include dir via epee target instead of globally
2020-05-30 04:24:19 +02:00
include_directories(external/rapidjson/include src external)
moved all stuff to github
2014-03-03 23:07:58 +01:00
various fixes to allow mac osx compilation
2014-04-30 19:52:21 +02:00
if(APPLE)
mac osx building fixes
2014-04-30 22:50:06 +02:00
include_directories(SYSTEM /usr/include/malloc)
Silence CMake policy warning on macos. See: https://cmake.org/cmake/help/v3.0/policy/CMP0042.html
2016-12-23 08:54:34 +01:00
if(POLICY CMP0042)
cmake_policy(SET CMP0042 NEW)
endif()
various fixes to allow mac osx compilation
2014-04-30 19:52:21 +02:00
endif()
Static builds: make usable binaries from cmake This adds a static dependency script for libraries like boost, unbound, etc. to cmake, invokable with: cmake .. -DBUILD_STATIC_DEPS=ON which downloads and builds static versions of all our required dependencies (boost, unbound, openssl, ncurses, etc.). It also implies -DSTATIC=ON to build other vendored deps (like miniupnpc, lokimq) as static as well. Unlike the contrib/depends system, this is easier to maintain (one script using nicer cmake with functions instead of raw Makefile spaghetti code), and isn't concerned with reproducible builds -- this doesn't rebuild the compiler, for instance. It also works with the existing build system so that it is simply another way to invoke the cmake build scripts but doesn't require any external tooling. This works on Linux, Mac, and Windows. Some random comments on this commit (for preserving history): - Don't use target_link_libraries on imported targets. Newer cmake is fine with it, but Bionic's cmake doesn't like it but seems okay with setting the properties directly. - This rebuilds libzmq and libsodium, even though there is some provision already within loki-core to do so: however, the existing embedded libzmq fails with the static deps because it uses libzmq's cmake build script, which relies on pkg-config to find libsodium which ends up finding the system one (or not finding any), rather than the one we build with DownloadLibSodium. Since both libsodium and libzmq are faily simple builds it seemed easiest to just add them to the cmake static build rather than trying to shoehorn the current code into the static build script. - Half of the protobuf build system ignores CC/CXX just because Google, and there's no documentation anywhere except for a random closed bug report about needing to set these other variables (CC_FOR_BUILD, CXX_FOR_BUILD) instead, but you need to. Thanks Google. - The boost build is set to output very little because even the minimum -d1 output level spams ~15k lines of output just for the headers it installs.
2020-06-11 20:19:44 +02:00
option(BUILD_STATIC_DEPS "Download, build and statically link against core dependencies" OFF)
if(BUILD_STATIC_DEPS)
include(StaticBuild)
endif()
if(MSVC OR MINGW OR BUILD_STATIC_DEPS)
Default to static linking on both MSVC and MINGW
2014-08-06 18:27:16 +02:00
set(DEFAULT_STATIC true)
else()
set(DEFAULT_STATIC false)
endif()
Static builds: make usable binaries from cmake This adds a static dependency script for libraries like boost, unbound, etc. to cmake, invokable with: cmake .. -DBUILD_STATIC_DEPS=ON which downloads and builds static versions of all our required dependencies (boost, unbound, openssl, ncurses, etc.). It also implies -DSTATIC=ON to build other vendored deps (like miniupnpc, lokimq) as static as well. Unlike the contrib/depends system, this is easier to maintain (one script using nicer cmake with functions instead of raw Makefile spaghetti code), and isn't concerned with reproducible builds -- this doesn't rebuild the compiler, for instance. It also works with the existing build system so that it is simply another way to invoke the cmake build scripts but doesn't require any external tooling. This works on Linux, Mac, and Windows. Some random comments on this commit (for preserving history): - Don't use target_link_libraries on imported targets. Newer cmake is fine with it, but Bionic's cmake doesn't like it but seems okay with setting the properties directly. - This rebuilds libzmq and libsodium, even though there is some provision already within loki-core to do so: however, the existing embedded libzmq fails with the static deps because it uses libzmq's cmake build script, which relies on pkg-config to find libsodium which ends up finding the system one (or not finding any), rather than the one we build with DownloadLibSodium. Since both libsodium and libzmq are faily simple builds it seemed easiest to just add them to the cmake static build rather than trying to shoehorn the current code into the static build script. - Half of the protobuf build system ignores CC/CXX just because Google, and there's no documentation anywhere except for a random closed bug report about needing to set these other variables (CC_FOR_BUILD, CXX_FOR_BUILD) instead, but you need to. Thanks Google. - The boost build is set to output very little because even the minimum -d1 output level spams ~15k lines of output just for the headers it installs.
2020-06-11 20:19:44 +02:00
option(STATIC "Try to link external dependencies statically, where possible" ${DEFAULT_STATIC})
if(BUILD_STATIC_DEPS AND NOT STATIC)
message(FATAL_ERROR "Option BUILD_STATIC_DEPS requires STATIC be enabled as well")
endif()
Various linking and build fixes - updating to latest loki-mq (1.0.0 + various linking fixes) - BUILD_SHARED_LIBS was being handled very strangely; make it a full option instead (defaulting to off) that a cmake invoker can specify, as per cmake recommendations. - travis ci tweaks/changes: - Add a static bionic build - Simplify cmake argument code - Add `--version` invocation for lokid and loki-wallet-cli to test that the binaries were linked properly. - always build an embedded sodium statically; if we do it dynamically and an older system one exists we are going to have trouble. - don't force epee and blocks to be static; rather they get controlled by the above BUILD_SHARED_LIBS, just like all the other internal libraries. - use some PkgConfig:: imported targets rather than bunch-of-variables.
2020-03-14 06:21:38 +01:00
option(BUILD_SHARED_LIBS "Build shared internal libraries" OFF)
cmake: support BUILD_SHARED_LIBS built-in option Support building internal libraries as shared. This reduces development time by eliminating the need to re-link all binaries every time non-interface code in the library changes. Instead, can hack on libxyz, then `make libxyz`, and re-run monerod. By default BUILD_SHARED_LIBS is OFF in release build type, and ON in debug build type, but can be overriden with -D.
2016-09-17 20:03:51 +02:00
msys: look in msys' directory for files This tells find_path and find_library to look under this directory first which is what we want on msys2.
2014-11-18 00:09:21 +01:00
if(MINGW)
Fixed MINGW duplicate library errors when MSYS folder cannot be found. *Thanks to luigi1111 for reporting.
2015-07-18 04:49:22 +02:00
string(REGEX MATCH "^[^/]:/[^/]*" msys2_install_path "${CMAKE_C_COMPILER}")
message(STATUS "MSYS location: ${msys2_install_path}")
fixed msys2 / mingw folders based on architecture, added license to unbound CMakeList as that is not part of standard Unbound
2015-03-02 16:20:42 +01:00
set(CMAKE_INCLUDE_PATH "${msys2_install_path}/mingw${ARCH_WIDTH}/include")
cmake: work around a bug with implicit link directories Unfortunately, this is necessary because CMake doesn't detect whether mingw libraries are static or shared and doesn't put a -static flag around the -lfoo argument which then makes the shared library be linked to.
2014-11-18 23:04:47 +01:00
# This is necessary because otherwise CMake will make Boost libraries -lfoo
# rather than a full path. Unfortunately, this makes the shared libraries get
# linked due to a bug in CMake which misses putting -static flags around the
# -lfoo arguments.
Fixed MINGW duplicate library errors when MSYS folder cannot be found. *Thanks to luigi1111 for reporting.
2015-07-18 04:49:22 +02:00
set(DEFLIB ${msys2_install_path}/mingw${ARCH_WIDTH}/lib)
list(REMOVE_ITEM CMAKE_C_IMPLICIT_LINK_DIRECTORIES ${DEFLIB})
list(REMOVE_ITEM CMAKE_CXX_IMPLICIT_LINK_DIRECTORIES ${DEFLIB})
msys: look in msys' directory for files This tells find_path and find_library to look under this directory first which is what we want on msys2.
2014-11-18 00:09:21 +01:00
endif()
cmake: minor cleanups (indentation and typos)
2014-10-21 20:24:49 +02:00
if(STATIC)
if(MSVC)
set(CMAKE_FIND_LIBRARY_SUFFIXES .lib .dll.a .a ${CMAKE_FIND_LIBRARY_SUFFIXES})
else()
set(CMAKE_FIND_LIBRARY_SUFFIXES .a ${CMAKE_FIND_LIBRARY_SUFFIXES})
endif()
endif()
enforce static libraries when building statically
2014-10-06 18:46:18 +02:00
fuzz_testing: build with ASAN (assumed to be available)
2017-12-10 12:25:12 +01:00
if(SANITIZE)
if (MSVC)
message(FATAL_ERROR "Cannot sanitize with MSVC")
else()
message(STATUS "Using ASAN")
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -fsanitize=address")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fsanitize=address")
endif()
endif()
Problem: misleading information about LMDB storage Solution: updated the comments to reflect the current situation in terms of LMDB implementation and no longer recommend 'memory' for blockchain storage in production use.
2017-03-08 14:29:08 +01:00
# Set default blockchain storage location:
Add minor whitespace/delete dead code after diffing Monero and Loki Mostly small whitespace changes that make comparing a diff of the entirety of Loki and Monero less noisy.
2018-10-24 02:33:48 +02:00
# memory was the default in Cryptonote before Monero implemented LMDB, it still works but is unnecessary.
Add compile-time support for both db implementations: in-memory and LMDB Usage: default is lmdb for blockchain branch: $ make release same as: $ DATABASE=lmdb make release for original in-memory implementation: $ DATABASE=memory make release
2015-01-26 06:36:09 +01:00
# set(DATABASE memory)
set(DATABASE lmdb)
if (DEFINED ENV{DATABASE})
set(DATABASE $ENV{DATABASE})
message(STATUS "DATABASE set: ${DATABASE}")
else()
message(STATUS "Could not find DATABASE in env (not required unless you want to change database type from default: ${DATABASE})")
endif()
** CHANGES ARE EXPERIMENTAL (FOR TESTING ONLY) Bockchain: 1. Optim: Multi-thread long-hash computation when encountering groups of blocks. 2. Optim: Cache verified txs and return result from cache instead of re-checking whenever possible. 3. Optim: Preload output-keys when encoutering groups of blocks. Sort by amount and global-index before bulk querying database and multi-thread when possible. 4. Optim: Disable double spend check on block verification, double spend is already detected when trying to add blocks. 5. Optim: Multi-thread signature computation whenever possible. 6. Patch: Disable locking (recursive mutex) on called functions from check_tx_inputs which causes slowdowns (only seems to happen on ubuntu/VMs??? Reason: TBD) 7. Optim: Removed looped full-tx hash computation when retrieving transactions from pool (???). 8. Optim: Cache difficulty/timestamps (735 blocks) for next-difficulty calculations so that only 2 db reads per new block is needed when a new block arrives (instead of 1470 reads). Berkeley-DB: 1. Fix: 32-bit data errors causing wrong output global indices and failure to send blocks to peers (etc). 2. Fix: Unable to pop blocks on reorganize due to transaction errors. 3. Patch: Large number of transaction aborts when running multi-threaded bulk queries. 4. Patch: Insufficient locks error when running full sync. 5. Patch: Incorrect db stats when returning from an immediate exit from "pop block" operation. 6. Optim: Add bulk queries to get output global indices. 7. Optim: Modified output_keys table to store public_key+unlock_time+height for single transaction lookup (vs 3) 8. Optim: Used output_keys table retrieve public_keys instead of going through output_amounts->output_txs+output_indices->txs->output:public_key 9. Optim: Added thread-safe buffers used when multi-threading bulk queries. 10. Optim: Added support for nosync/write_nosync options for improved performance (*see --db-sync-mode option for details) 11. Mod: Added checkpoint thread and auto-remove-logs option. 12. *Now usable on 32-bit systems like RPI2. LMDB: 1. Optim: Added custom comparison for 256-bit key tables (minor speed-up, TBD: get actual effect) 2. Optim: Modified output_keys table to store public_key+unlock_time+height for single transaction lookup (vs 3) 3. Optim: Used output_keys table retrieve public_keys instead of going through output_amounts->output_txs+output_indices->txs->output:public_key 4. Optim: Added support for sync/writemap options for improved performance (*see --db-sync-mode option for details) 5. Mod: Auto resize to +1GB instead of multiplier x1.5 ETC: 1. Minor optimizations for slow-hash for ARM (RPI2). Incomplete. 2. Fix: 32-bit saturation bug when computing next difficulty on large blocks. [PENDING ISSUES] 1. Berkely db has a very slow "pop-block" operation. This is very noticeable on the RPI2 as it sometimes takes > 10 MINUTES to pop a block during reorganization. This does not happen very often however, most reorgs seem to take a few seconds but it possibly depends on the number of outputs present. TBD. 2. Berkeley db, possible bug "unable to allocate memory". TBD. [NEW OPTIONS] (*Currently all enabled for testing purposes) 1. --fast-block-sync arg=[0:1] (default: 1) a. 0 = Compute long hash per block (may take a while depending on CPU) b. 1 = Skip long-hash and verify blocks based on embedded known good block hashes (faster, minimal CPU dependence) 2. --db-sync-mode arg=[[safe|fast|fastest]:[sync|async]:[nblocks_per_sync]] (default: fastest:async:1000) a. safe = fdatasync/fsync (or equivalent) per stored block. Very slow, but safest option to protect against power-out/crash conditions. b. fast/fastest = Enables asynchronous fdatasync/fsync (or equivalent). Useful for battery operated devices or STABLE systems with UPS and/or systems with battery backed write cache/solid state cache. Fast - Write meta-data but defer data flush. Fastest - Defer meta-data and data flush. Sync - Flush data after nblocks_per_sync and wait. Async - Flush data after nblocks_per_sync but do not wait for the operation to finish. 3. --prep-blocks-threads arg=[n] (default: 4 or system max threads, whichever is lower) Max number of threads to use when computing long-hash in groups. 4. --show-time-stats arg=[0:1] (default: 1) Show benchmark related time stats. 5. --db-auto-remove-logs arg=[0:1] (default: 1) For berkeley-db only. Auto remove logs if enabled. **Note: lmdb and berkeley-db have changes to the tables and are not compatible with official git head version. At the moment, you need a full resync to use this optimized version. [PERFORMANCE COMPARISON] **Some figures are approximations only. Using a baseline machine of an i7-2600K+SSD+(with full pow computation): 1. The optimized lmdb/blockhain core can process blocks up to 585K for ~1.25 hours + download time, so it usually takes 2.5 hours to sync the full chain. 2. The current head with memory can process blocks up to 585K for ~4.2 hours + download time, so it usually takes 5.5 hours to sync the full chain. 3. The current head with lmdb can process blocks up to 585K for ~32 hours + download time and usually takes 36 hours to sync the full chain. Averate procesing times (with full pow computation): lmdb-optimized: 1. tx_ave = 2.5 ms / tx 2. block_ave = 5.87 ms / block memory-official-repo: 1. tx_ave = 8.85 ms / tx 2. block_ave = 19.68 ms / block lmdb-official-repo (0f4a036437fd41a5498ee5e74e2422ea6177aa3e) 1. tx_ave = 47.8 ms / tx 2. block_ave = 64.2 ms / block **Note: The following data denotes processing times only (does not include p2p download time) lmdb-optimized processing times (with full pow computation): 1. Desktop, Quad-core / 8-threads 2600k (8Mb) - 1.25 hours processing time (--db-sync-mode=fastest:async:1000). 2. Laptop, Dual-core / 4-threads U4200 (3Mb) - 4.90 hours processing time (--db-sync-mode=fastest:async:1000). 3. Embedded, Quad-core / 4-threads Z3735F (2x1Mb) - 12.0 hours processing time (--db-sync-mode=fastest:async:1000). lmdb-optimized processing times (with per-block-checkpoint) 1. Desktop, Quad-core / 8-threads 2600k (8Mb) - 10 minutes processing time (--db-sync-mode=fastest:async:1000). berkeley-db optimized processing times (with full pow computation) 1. Desktop, Quad-core / 8-threads 2600k (8Mb) - 1.8 hours processing time (--db-sync-mode=fastest:async:1000). 2. RPI2. Improved from estimated 3 months(???) into 2.5 days (*Need 2AMP supply + Clock:1Ghz + [usb+ssd] to achieve this speed) (--db-sync-mode=fastest:async:1000). berkeley-db optimized processing times (with per-block-checkpoint) 1. RPI2. 12-15 hours (*Need 2AMP supply + Clock:1Ghz + [usb+ssd] to achieve this speed) (--db-sync-mode=fastest:async:1000).
2015-07-10 22:09:32 +02:00
Add compile-time support for both db implementations: in-memory and LMDB Usage: default is lmdb for blockchain branch: $ make release same as: $ DATABASE=lmdb make release for original in-memory implementation: $ DATABASE=memory make release
2015-01-26 06:36:09 +01:00
if (DATABASE STREQUAL "lmdb")
change to LMDB as default, even on 32-bit and ARM
2016-03-12 20:20:00 +01:00
message(STATUS "Using LMDB as default DB type")
cmake modernization The archaic (i.e. decade old) cmake usage here really got in the way of trying to properly use newer libraries (like lokimq), so this undertakes overhauling it considerably to make it much more sane (and significantly reduce the size). I left more of the architecture-specific bits in the top-level CMakeLists.txt intact; most of the efforts here are about properly loading dependencies, specifying dependencies and avoiding a whole pile of cmake antipatterns. This bumps the required cmake version to 3.5, which is what xenial comes with. - extensive use of interface libraries to include libraries, definitions, and include paths - use Boost::whatever instead of ${Boost_WHATEVER_LIBRARY}. The interface targets are (again) much better as they also give you any needed include or linking flags without needing to worry about them. - don't list header files when building things. This has *never* been correct cmake usage (cmake has always known how to wallet_rpc_headers the headers that .cpp files include to know about build changes). - remove the loki_add_library monstrosity; it breaks target names and makes compiling less efficient because the author couldn't figure out how to link things together. - make loki_add_executable take the output filename, and set the output path to bin/ and install to bin because *every single usage* of loki_add_executable was immediately followed by setting the output filename and setting the output path to bin/ and installing to bin. - move a bunch of crap that is only used in one particular src/whatever/CMakeLists.txt into that particular CMakeLists.txt instead of the top level CMakeLists.txt (or src/CMakeLists.txt). - Remove a bunch of redundant dependencies; most of them look like they were just copy-and-pasted in, and many more aren't needed (since they are implied by the PUBLIC linking of other dependencies). - Removed `die` since it just does a FATAL_ERROR, but adds color (which is useless since CMake already makes FATAL_ERRORs perfectly visible). - Change the way LOKI_DAEMON_AND_WALLET_ONLY works to just change the make targets to daemon and simplewallet rather than changing the build process (this should make it faster, too, since there are various other things that will be excluded).
2020-03-03 04:57:08 +01:00
set(BLOCKCHAIN_DB "lmdb")
Add compile-time support for both db implementations: in-memory and LMDB Usage: default is lmdb for blockchain branch: $ make release same as: $ DATABASE=lmdb make release for original in-memory implementation: $ DATABASE=memory make release
2015-01-26 06:36:09 +01:00
else()
cmake modernization The archaic (i.e. decade old) cmake usage here really got in the way of trying to properly use newer libraries (like lokimq), so this undertakes overhauling it considerably to make it much more sane (and significantly reduce the size). I left more of the architecture-specific bits in the top-level CMakeLists.txt intact; most of the efforts here are about properly loading dependencies, specifying dependencies and avoiding a whole pile of cmake antipatterns. This bumps the required cmake version to 3.5, which is what xenial comes with. - extensive use of interface libraries to include libraries, definitions, and include paths - use Boost::whatever instead of ${Boost_WHATEVER_LIBRARY}. The interface targets are (again) much better as they also give you any needed include or linking flags without needing to worry about them. - don't list header files when building things. This has *never* been correct cmake usage (cmake has always known how to wallet_rpc_headers the headers that .cpp files include to know about build changes). - remove the loki_add_library monstrosity; it breaks target names and makes compiling less efficient because the author couldn't figure out how to link things together. - make loki_add_executable take the output filename, and set the output path to bin/ and install to bin because *every single usage* of loki_add_executable was immediately followed by setting the output filename and setting the output path to bin/ and installing to bin. - move a bunch of crap that is only used in one particular src/whatever/CMakeLists.txt into that particular CMakeLists.txt instead of the top level CMakeLists.txt (or src/CMakeLists.txt). - Remove a bunch of redundant dependencies; most of them look like they were just copy-and-pasted in, and many more aren't needed (since they are implied by the PUBLIC linking of other dependencies). - Removed `die` since it just does a FATAL_ERROR, but adds color (which is useless since CMake already makes FATAL_ERRORs perfectly visible). - Change the way LOKI_DAEMON_AND_WALLET_ONLY works to just change the make targets to daemon and simplewallet rather than changing the build process (this should make it faster, too, since there are various other things that will be excluded).
2020-03-03 04:57:08 +01:00
message(FATAL_ERROR "Invalid database type: ${DATABASE}")
** CHANGES ARE EXPERIMENTAL (FOR TESTING ONLY) Bockchain: 1. Optim: Multi-thread long-hash computation when encountering groups of blocks. 2. Optim: Cache verified txs and return result from cache instead of re-checking whenever possible. 3. Optim: Preload output-keys when encoutering groups of blocks. Sort by amount and global-index before bulk querying database and multi-thread when possible. 4. Optim: Disable double spend check on block verification, double spend is already detected when trying to add blocks. 5. Optim: Multi-thread signature computation whenever possible. 6. Patch: Disable locking (recursive mutex) on called functions from check_tx_inputs which causes slowdowns (only seems to happen on ubuntu/VMs??? Reason: TBD) 7. Optim: Removed looped full-tx hash computation when retrieving transactions from pool (???). 8. Optim: Cache difficulty/timestamps (735 blocks) for next-difficulty calculations so that only 2 db reads per new block is needed when a new block arrives (instead of 1470 reads). Berkeley-DB: 1. Fix: 32-bit data errors causing wrong output global indices and failure to send blocks to peers (etc). 2. Fix: Unable to pop blocks on reorganize due to transaction errors. 3. Patch: Large number of transaction aborts when running multi-threaded bulk queries. 4. Patch: Insufficient locks error when running full sync. 5. Patch: Incorrect db stats when returning from an immediate exit from "pop block" operation. 6. Optim: Add bulk queries to get output global indices. 7. Optim: Modified output_keys table to store public_key+unlock_time+height for single transaction lookup (vs 3) 8. Optim: Used output_keys table retrieve public_keys instead of going through output_amounts->output_txs+output_indices->txs->output:public_key 9. Optim: Added thread-safe buffers used when multi-threading bulk queries. 10. Optim: Added support for nosync/write_nosync options for improved performance (*see --db-sync-mode option for details) 11. Mod: Added checkpoint thread and auto-remove-logs option. 12. *Now usable on 32-bit systems like RPI2. LMDB: 1. Optim: Added custom comparison for 256-bit key tables (minor speed-up, TBD: get actual effect) 2. Optim: Modified output_keys table to store public_key+unlock_time+height for single transaction lookup (vs 3) 3. Optim: Used output_keys table retrieve public_keys instead of going through output_amounts->output_txs+output_indices->txs->output:public_key 4. Optim: Added support for sync/writemap options for improved performance (*see --db-sync-mode option for details) 5. Mod: Auto resize to +1GB instead of multiplier x1.5 ETC: 1. Minor optimizations for slow-hash for ARM (RPI2). Incomplete. 2. Fix: 32-bit saturation bug when computing next difficulty on large blocks. [PENDING ISSUES] 1. Berkely db has a very slow "pop-block" operation. This is very noticeable on the RPI2 as it sometimes takes > 10 MINUTES to pop a block during reorganization. This does not happen very often however, most reorgs seem to take a few seconds but it possibly depends on the number of outputs present. TBD. 2. Berkeley db, possible bug "unable to allocate memory". TBD. [NEW OPTIONS] (*Currently all enabled for testing purposes) 1. --fast-block-sync arg=[0:1] (default: 1) a. 0 = Compute long hash per block (may take a while depending on CPU) b. 1 = Skip long-hash and verify blocks based on embedded known good block hashes (faster, minimal CPU dependence) 2. --db-sync-mode arg=[[safe|fast|fastest]:[sync|async]:[nblocks_per_sync]] (default: fastest:async:1000) a. safe = fdatasync/fsync (or equivalent) per stored block. Very slow, but safest option to protect against power-out/crash conditions. b. fast/fastest = Enables asynchronous fdatasync/fsync (or equivalent). Useful for battery operated devices or STABLE systems with UPS and/or systems with battery backed write cache/solid state cache. Fast - Write meta-data but defer data flush. Fastest - Defer meta-data and data flush. Sync - Flush data after nblocks_per_sync and wait. Async - Flush data after nblocks_per_sync but do not wait for the operation to finish. 3. --prep-blocks-threads arg=[n] (default: 4 or system max threads, whichever is lower) Max number of threads to use when computing long-hash in groups. 4. --show-time-stats arg=[0:1] (default: 1) Show benchmark related time stats. 5. --db-auto-remove-logs arg=[0:1] (default: 1) For berkeley-db only. Auto remove logs if enabled. **Note: lmdb and berkeley-db have changes to the tables and are not compatible with official git head version. At the moment, you need a full resync to use this optimized version. [PERFORMANCE COMPARISON] **Some figures are approximations only. Using a baseline machine of an i7-2600K+SSD+(with full pow computation): 1. The optimized lmdb/blockhain core can process blocks up to 585K for ~1.25 hours + download time, so it usually takes 2.5 hours to sync the full chain. 2. The current head with memory can process blocks up to 585K for ~4.2 hours + download time, so it usually takes 5.5 hours to sync the full chain. 3. The current head with lmdb can process blocks up to 585K for ~32 hours + download time and usually takes 36 hours to sync the full chain. Averate procesing times (with full pow computation): lmdb-optimized: 1. tx_ave = 2.5 ms / tx 2. block_ave = 5.87 ms / block memory-official-repo: 1. tx_ave = 8.85 ms / tx 2. block_ave = 19.68 ms / block lmdb-official-repo (0f4a036437fd41a5498ee5e74e2422ea6177aa3e) 1. tx_ave = 47.8 ms / tx 2. block_ave = 64.2 ms / block **Note: The following data denotes processing times only (does not include p2p download time) lmdb-optimized processing times (with full pow computation): 1. Desktop, Quad-core / 8-threads 2600k (8Mb) - 1.25 hours processing time (--db-sync-mode=fastest:async:1000). 2. Laptop, Dual-core / 4-threads U4200 (3Mb) - 4.90 hours processing time (--db-sync-mode=fastest:async:1000). 3. Embedded, Quad-core / 4-threads Z3735F (2x1Mb) - 12.0 hours processing time (--db-sync-mode=fastest:async:1000). lmdb-optimized processing times (with per-block-checkpoint) 1. Desktop, Quad-core / 8-threads 2600k (8Mb) - 10 minutes processing time (--db-sync-mode=fastest:async:1000). berkeley-db optimized processing times (with full pow computation) 1. Desktop, Quad-core / 8-threads 2600k (8Mb) - 1.8 hours processing time (--db-sync-mode=fastest:async:1000). 2. RPI2. Improved from estimated 3 months(???) into 2.5 days (*Need 2AMP supply + Clock:1Ghz + [usb+ssd] to achieve this speed) (--db-sync-mode=fastest:async:1000). berkeley-db optimized processing times (with per-block-checkpoint) 1. RPI2. 12-15 hours (*Need 2AMP supply + Clock:1Ghz + [usb+ssd] to achieve this speed) (--db-sync-mode=fastest:async:1000).
2015-07-10 22:09:32 +02:00
endif()
Replace tabs with two spaces for consistency with rest of codebase Remove trailing whitespace in same files.
2015-12-14 05:54:39 +01:00
cmake modernization The archaic (i.e. decade old) cmake usage here really got in the way of trying to properly use newer libraries (like lokimq), so this undertakes overhauling it considerably to make it much more sane (and significantly reduce the size). I left more of the architecture-specific bits in the top-level CMakeLists.txt intact; most of the efforts here are about properly loading dependencies, specifying dependencies and avoiding a whole pile of cmake antipatterns. This bumps the required cmake version to 3.5, which is what xenial comes with. - extensive use of interface libraries to include libraries, definitions, and include paths - use Boost::whatever instead of ${Boost_WHATEVER_LIBRARY}. The interface targets are (again) much better as they also give you any needed include or linking flags without needing to worry about them. - don't list header files when building things. This has *never* been correct cmake usage (cmake has always known how to wallet_rpc_headers the headers that .cpp files include to know about build changes). - remove the loki_add_library monstrosity; it breaks target names and makes compiling less efficient because the author couldn't figure out how to link things together. - make loki_add_executable take the output filename, and set the output path to bin/ and install to bin because *every single usage* of loki_add_executable was immediately followed by setting the output filename and setting the output path to bin/ and installing to bin. - move a bunch of crap that is only used in one particular src/whatever/CMakeLists.txt into that particular CMakeLists.txt instead of the top level CMakeLists.txt (or src/CMakeLists.txt). - Remove a bunch of redundant dependencies; most of them look like they were just copy-and-pasted in, and many more aren't needed (since they are implied by the PUBLIC linking of other dependencies). - Removed `die` since it just does a FATAL_ERROR, but adds color (which is useless since CMake already makes FATAL_ERRORs perfectly visible). - Change the way LOKI_DAEMON_AND_WALLET_ONLY works to just change the make targets to daemon and simplewallet rather than changing the build process (this should make it faster, too, since there are various other things that will be excluded).
2020-03-03 04:57:08 +01:00
set(THREADS_PREFER_PTHREAD_FLAG ON)
find_package(Threads)
make this build on SunOS/Solaris
2017-10-10 15:05:28 +02:00
IOS CMAKE build settings
2017-04-02 12:19:25 +02:00
if (APPLE AND NOT IOS)
Set C++14 compilation mode properly CMake has standard interfaces for this; hacking it into CXX_FLAGS is just gross.
2020-06-12 18:16:33 +02:00
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -march=x86-64 -fvisibility=default")
Various contrib/depends fixes - set WITH_SYSTEMD=OFF in contrib/depends so that we don't try to go look for it on the host system. - link icu libs properly - don't build embedded zmq but rather set up a target so that loki-mq uses the externally built one - fix OpenSSL::Crypto not properly depending on ws2_32 - build boost atomic in the built boost because boost::thread depends on it (and we were just getting lucky before by not happening to touch anything that needed it) - make bundled unbound link against the `extra` interface for various required windows crap (ws2_32 and other stuff).
2020-03-17 01:56:48 +01:00
endif()
cmake: add -DHAVE_ defines to the command line since there's no config.h here
2018-01-28 09:57:51 +01:00
add_definition_if_library_exists(c memset_s "string.h" HAVE_MEMSET_S)
add_definition_if_library_exists(c explicit_bzero "strings.h" HAVE_EXPLICIT_BZERO)
add_definition_if_function_found(strptime HAVE_STRPTIME)
Embed the translation files in the binary If a translation file exists in a "translations" directory located in the same directory as the binary, it is used in priority (this can be useful when working on translations as you don't have to recompile the whole program all the time), and if no such file is found the embedded translation file is used (if it exists).
2017-12-15 18:44:01 +01:00
# Generate header for embedded translations
Don't use external project for no reason Using an external project to build a subdirectory is gross, and moreover it breaks if you are trying to use a custom compiler and uses the wrong one (or just fails if a `c++` binary doesn't exist). Since the builds appear to run just fine without this, just include it via add_subdirectory instead.
2020-06-09 20:47:05 +02:00
add_subdirectory(translations)
Fixed CMake search for libunbound As it turns out, some of CMake's built-in modules for detecting headers and libraries don't work if you have certain compiler flags set, such as -Werror=old-style-definition, as they do "int main()" rather than "int main(void)". Having CMake search for libs before compiler flags are set alleviates this issue, and I believe the underlying issue will be fixed in future releases of CMake.
2014-09-17 23:44:35 +02:00
Back out systemd linking to top-level CMakeLists systemd notifies on startup really belong in daemon, not core, so this paves the way for easily moving the calls there.
2020-04-28 01:17:24 +02:00
add_library(systemd INTERFACE) # Will do nothing unless we find and enable systemd support
Move Sodium detection above external submodules (#1060) Some submodules can request a libsodium version and set SODIUM_* variables before us. Combining with force downloading we can end up building dependencies with differing versions of Sodium.
2020-03-10 06:02:57 +01:00
Static builds: make usable binaries from cmake This adds a static dependency script for libraries like boost, unbound, etc. to cmake, invokable with: cmake .. -DBUILD_STATIC_DEPS=ON which downloads and builds static versions of all our required dependencies (boost, unbound, openssl, ncurses, etc.). It also implies -DSTATIC=ON to build other vendored deps (like miniupnpc, lokimq) as static as well. Unlike the contrib/depends system, this is easier to maintain (one script using nicer cmake with functions instead of raw Makefile spaghetti code), and isn't concerned with reproducible builds -- this doesn't rebuild the compiler, for instance. It also works with the existing build system so that it is simply another way to invoke the cmake build scripts but doesn't require any external tooling. This works on Linux, Mac, and Windows. Some random comments on this commit (for preserving history): - Don't use target_link_libraries on imported targets. Newer cmake is fine with it, but Bionic's cmake doesn't like it but seems okay with setting the properties directly. - This rebuilds libzmq and libsodium, even though there is some provision already within loki-core to do so: however, the existing embedded libzmq fails with the static deps because it uses libzmq's cmake build script, which relies on pkg-config to find libsodium which ends up finding the system one (or not finding any), rather than the one we build with DownloadLibSodium. Since both libsodium and libzmq are faily simple builds it seemed easiest to just add them to the cmake static build rather than trying to shoehorn the current code into the static build script. - Half of the protobuf build system ignores CC/CXX just because Google, and there's no documentation anywhere except for a random closed bug report about needing to set these other variables (CC_FOR_BUILD, CXX_FOR_BUILD) instead, but you need to. Thanks Google. - The boost build is set to output very little because even the minimum -d1 output level spams ~15k lines of output just for the headers it installs.
2020-06-11 20:19:44 +02:00
if(NOT TARGET sodium)
Move Sodium detection above external submodules (#1060) Some submodules can request a libsodium version and set SODIUM_* variables before us. Combining with force downloading we can end up building dependencies with differing versions of Sodium.
2020-03-10 06:02:57 +01:00
# Allow -D DOWNLOAD_SODIUM=FORCE to download without even checking for a local libsodium
Static builds: make usable binaries from cmake This adds a static dependency script for libraries like boost, unbound, etc. to cmake, invokable with: cmake .. -DBUILD_STATIC_DEPS=ON which downloads and builds static versions of all our required dependencies (boost, unbound, openssl, ncurses, etc.). It also implies -DSTATIC=ON to build other vendored deps (like miniupnpc, lokimq) as static as well. Unlike the contrib/depends system, this is easier to maintain (one script using nicer cmake with functions instead of raw Makefile spaghetti code), and isn't concerned with reproducible builds -- this doesn't rebuild the compiler, for instance. It also works with the existing build system so that it is simply another way to invoke the cmake build scripts but doesn't require any external tooling. This works on Linux, Mac, and Windows. Some random comments on this commit (for preserving history): - Don't use target_link_libraries on imported targets. Newer cmake is fine with it, but Bionic's cmake doesn't like it but seems okay with setting the properties directly. - This rebuilds libzmq and libsodium, even though there is some provision already within loki-core to do so: however, the existing embedded libzmq fails with the static deps because it uses libzmq's cmake build script, which relies on pkg-config to find libsodium which ends up finding the system one (or not finding any), rather than the one we build with DownloadLibSodium. Since both libsodium and libzmq are faily simple builds it seemed easiest to just add them to the cmake static build rather than trying to shoehorn the current code into the static build script. - Half of the protobuf build system ignores CC/CXX just because Google, and there's no documentation anywhere except for a random closed bug report about needing to set these other variables (CC_FOR_BUILD, CXX_FOR_BUILD) instead, but you need to. Thanks Google. - The boost build is set to output very little because even the minimum -d1 output level spams ~15k lines of output just for the headers it installs.
2020-06-11 20:19:44 +02:00
option(DOWNLOAD_SODIUM "Allow libsodium to be downloaded and built locally if not found on the system" OFF)
if(NOT DOWNLOAD_SODIUM STREQUAL "FORCE" AND NOT BUILD_STATIC_DEPS)
find_package(PkgConfig REQUIRED)
pkg_check_modules(SODIUM libsodium>=1.0.9 IMPORTED_TARGET)
endif()
Move Sodium detection above external submodules (#1060) Some submodules can request a libsodium version and set SODIUM_* variables before us. Combining with force downloading we can end up building dependencies with differing versions of Sodium.
2020-03-10 06:02:57 +01:00
Static builds: make usable binaries from cmake This adds a static dependency script for libraries like boost, unbound, etc. to cmake, invokable with: cmake .. -DBUILD_STATIC_DEPS=ON which downloads and builds static versions of all our required dependencies (boost, unbound, openssl, ncurses, etc.). It also implies -DSTATIC=ON to build other vendored deps (like miniupnpc, lokimq) as static as well. Unlike the contrib/depends system, this is easier to maintain (one script using nicer cmake with functions instead of raw Makefile spaghetti code), and isn't concerned with reproducible builds -- this doesn't rebuild the compiler, for instance. It also works with the existing build system so that it is simply another way to invoke the cmake build scripts but doesn't require any external tooling. This works on Linux, Mac, and Windows. Some random comments on this commit (for preserving history): - Don't use target_link_libraries on imported targets. Newer cmake is fine with it, but Bionic's cmake doesn't like it but seems okay with setting the properties directly. - This rebuilds libzmq and libsodium, even though there is some provision already within loki-core to do so: however, the existing embedded libzmq fails with the static deps because it uses libzmq's cmake build script, which relies on pkg-config to find libsodium which ends up finding the system one (or not finding any), rather than the one we build with DownloadLibSodium. Since both libsodium and libzmq are faily simple builds it seemed easiest to just add them to the cmake static build rather than trying to shoehorn the current code into the static build script. - Half of the protobuf build system ignores CC/CXX just because Google, and there's no documentation anywhere except for a random closed bug report about needing to set these other variables (CC_FOR_BUILD, CXX_FOR_BUILD) instead, but you need to. Thanks Google. - The boost build is set to output very little because even the minimum -d1 output level spams ~15k lines of output just for the headers it installs.
2020-06-11 20:19:44 +02:00
add_library(sodium INTERFACE)
if(SODIUM_FOUND AND NOT DOWNLOAD_SODIUM STREQUAL "FORCE" AND NOT BUILD_STATIC_DEPS)
target_link_libraries(sodium INTERFACE PkgConfig::SODIUM)
else()
if(NOT DOWNLOAD_SODIUM AND NOT BUILD_STATIC_DEPS)
message(FATAL_ERROR "Could not find libsodium >= 1.0.9; either install it on your system or use -DDOWNLOAD_SODIUM=ON to download and build an internal copy")
endif()
message(STATUS "Sodium >= 1.0.9 not found, but DOWNLOAD_SODIUM specified, so downloading it")
include(DownloadLibSodium)
target_link_libraries(sodium INTERFACE sodium_vendor)
Various linking and build fixes - updating to latest loki-mq (1.0.0 + various linking fixes) - BUILD_SHARED_LIBS was being handled very strangely; make it a full option instead (defaulting to off) that a cmake invoker can specify, as per cmake recommendations. - travis ci tweaks/changes: - Add a static bionic build - Simplify cmake argument code - Add `--version` invocation for lokid and loki-wallet-cli to test that the binaries were linked properly. - always build an embedded sodium statically; if we do it dynamically and an older system one exists we are going to have trouble. - don't force epee and blocks to be static; rather they get controlled by the above BUILD_SHARED_LIBS, just like all the other internal libraries. - use some PkgConfig:: imported targets rather than bunch-of-variables.
2020-03-14 06:21:38 +01:00
endif()
Move Sodium detection above external submodules (#1060) Some submodules can request a libsodium version and set SODIUM_* variables before us. Combining with force downloading we can end up building dependencies with differing versions of Sodium.
2020-03-10 06:02:57 +01:00
Clean up sys-or-submodule code, support system fmt, rename loki-mq Synced up with oxen-storage-server. Also rename loki-mq submodule to oxen-mq.
2022-05-30 20:23:06 +02:00
# Need this target export so that oxen-mq properly picks up sodium
Static builds: make usable binaries from cmake This adds a static dependency script for libraries like boost, unbound, etc. to cmake, invokable with: cmake .. -DBUILD_STATIC_DEPS=ON which downloads and builds static versions of all our required dependencies (boost, unbound, openssl, ncurses, etc.). It also implies -DSTATIC=ON to build other vendored deps (like miniupnpc, lokimq) as static as well. Unlike the contrib/depends system, this is easier to maintain (one script using nicer cmake with functions instead of raw Makefile spaghetti code), and isn't concerned with reproducible builds -- this doesn't rebuild the compiler, for instance. It also works with the existing build system so that it is simply another way to invoke the cmake build scripts but doesn't require any external tooling. This works on Linux, Mac, and Windows. Some random comments on this commit (for preserving history): - Don't use target_link_libraries on imported targets. Newer cmake is fine with it, but Bionic's cmake doesn't like it but seems okay with setting the properties directly. - This rebuilds libzmq and libsodium, even though there is some provision already within loki-core to do so: however, the existing embedded libzmq fails with the static deps because it uses libzmq's cmake build script, which relies on pkg-config to find libsodium which ends up finding the system one (or not finding any), rather than the one we build with DownloadLibSodium. Since both libsodium and libzmq are faily simple builds it seemed easiest to just add them to the cmake static build rather than trying to shoehorn the current code into the static build script. - Half of the protobuf build system ignores CC/CXX just because Google, and there's no documentation anywhere except for a random closed bug report about needing to set these other variables (CC_FOR_BUILD, CXX_FOR_BUILD) instead, but you need to. Thanks Google. - The boost build is set to output very little because even the minimum -d1 output level spams ~15k lines of output just for the headers it installs.
2020-06-11 20:19:44 +02:00
export(TARGETS sodium NAMESPACE sodium:: FILE sodium-exports.cmake)
endif()
Remove unbound submodule There's really no reason to submodule it - we work with pretty much any libunbound version, and it's a very commonly available library (comparable to sqlite3 or boost, which we don't submodule). This removes the submodule and switches it to a hard dependency.
2020-06-11 20:02:21 +02:00
Back out systemd linking to top-level CMakeLists systemd notifies on startup really belong in daemon, not core, so this paves the way for easily moving the calls there.
2020-04-28 01:17:24 +02:00
option(WITH_SYSTEMD "Attempts to link against and enable systemd daemon notification support" ON)
Static builds: make usable binaries from cmake This adds a static dependency script for libraries like boost, unbound, etc. to cmake, invokable with: cmake .. -DBUILD_STATIC_DEPS=ON which downloads and builds static versions of all our required dependencies (boost, unbound, openssl, ncurses, etc.). It also implies -DSTATIC=ON to build other vendored deps (like miniupnpc, lokimq) as static as well. Unlike the contrib/depends system, this is easier to maintain (one script using nicer cmake with functions instead of raw Makefile spaghetti code), and isn't concerned with reproducible builds -- this doesn't rebuild the compiler, for instance. It also works with the existing build system so that it is simply another way to invoke the cmake build scripts but doesn't require any external tooling. This works on Linux, Mac, and Windows. Some random comments on this commit (for preserving history): - Don't use target_link_libraries on imported targets. Newer cmake is fine with it, but Bionic's cmake doesn't like it but seems okay with setting the properties directly. - This rebuilds libzmq and libsodium, even though there is some provision already within loki-core to do so: however, the existing embedded libzmq fails with the static deps because it uses libzmq's cmake build script, which relies on pkg-config to find libsodium which ends up finding the system one (or not finding any), rather than the one we build with DownloadLibSodium. Since both libsodium and libzmq are faily simple builds it seemed easiest to just add them to the cmake static build rather than trying to shoehorn the current code into the static build script. - Half of the protobuf build system ignores CC/CXX just because Google, and there's no documentation anywhere except for a random closed bug report about needing to set these other variables (CC_FOR_BUILD, CXX_FOR_BUILD) instead, but you need to. Thanks Google. - The boost build is set to output very little because even the minimum -d1 output level spams ~15k lines of output just for the headers it installs.
2020-06-11 20:19:44 +02:00
if (WITH_SYSTEMD AND NOT BUILD_STATIC_DEPS)
find_package(PkgConfig REQUIRED)
Back out systemd linking to top-level CMakeLists systemd notifies on startup really belong in daemon, not core, so this paves the way for easily moving the calls there.
2020-04-28 01:17:24 +02:00
pkg_check_modules(SYSTEMD libsystemd IMPORTED_TARGET)
if(SYSTEMD_FOUND)
target_compile_definitions(systemd INTERFACE ENABLE_SYSTEMD)
target_link_libraries(systemd INTERFACE PkgConfig::SYSTEMD)
Only warn about not finding systemd on linux
2020-06-01 19:27:15 +02:00
elseif(CMAKE_SYSTEM_NAME STREQUAL Linux)
Back out systemd linking to top-level CMakeLists systemd notifies on startup really belong in daemon, not core, so this paves the way for easily moving the calls there.
2020-04-28 01:17:24 +02:00
message(WARNING "systemd not found; building without systemd support (use -DWITH_SYSTEMD=OFF to suppress this warning)")
endif()
endif()
Update CMakeLists.txt consolidate the cmake logic
2021-10-05 03:18:08 +02:00
if(NOT BUILD_STATIC_DEPS)
Remove check for unbound We don't use it anymore, but the top-level cmake was still looking for it.
2022-05-31 00:24:24 +02:00
find_package(PkgConfig REQUIRED)
Make sure we have sqlite 3.24.0+ Batching code relies on UPSERTs which were added in 3.24. (This broke initial bionic debs, which needed a sqlite3 backport from newer Ubuntu to address).
2022-06-17 02:40:19 +02:00
pkg_check_modules(SQLITE3 REQUIRED sqlite3>=3.24.0 IMPORTED_TARGET GLOBAL)
sqlitecpp
2021-10-05 02:33:24 +02:00
message(STATUS "Found sqlite3 ${SQLITE3_VERSION}")
sqlitecpp cmake nonsense
2021-10-07 01:04:23 +02:00
add_library(SQLite::SQLite3 ALIAS PkgConfig::SQLITE3)
sqlitecpp
2021-10-05 02:33:24 +02:00
endif()
cmake modernization The archaic (i.e. decade old) cmake usage here really got in the way of trying to properly use newer libraries (like lokimq), so this undertakes overhauling it considerably to make it much more sane (and significantly reduce the size). I left more of the architecture-specific bits in the top-level CMakeLists.txt intact; most of the efforts here are about properly loading dependencies, specifying dependencies and avoiding a whole pile of cmake antipatterns. This bumps the required cmake version to 3.5, which is what xenial comes with. - extensive use of interface libraries to include libraries, definitions, and include paths - use Boost::whatever instead of ${Boost_WHATEVER_LIBRARY}. The interface targets are (again) much better as they also give you any needed include or linking flags without needing to worry about them. - don't list header files when building things. This has *never* been correct cmake usage (cmake has always known how to wallet_rpc_headers the headers that .cpp files include to know about build changes). - remove the loki_add_library monstrosity; it breaks target names and makes compiling less efficient because the author couldn't figure out how to link things together. - make loki_add_executable take the output filename, and set the output path to bin/ and install to bin because *every single usage* of loki_add_executable was immediately followed by setting the output filename and setting the output path to bin/ and installing to bin. - move a bunch of crap that is only used in one particular src/whatever/CMakeLists.txt into that particular CMakeLists.txt instead of the top level CMakeLists.txt (or src/CMakeLists.txt). - Remove a bunch of redundant dependencies; most of them look like they were just copy-and-pasted in, and many more aren't needed (since they are implied by the PUBLIC linking of other dependencies). - Removed `die` since it just does a FATAL_ERROR, but adds color (which is useless since CMake already makes FATAL_ERRORs perfectly visible). - Change the way LOKI_DAEMON_AND_WALLET_ONLY works to just change the make targets to daemon and simplewallet rather than changing the build process (this should make it faster, too, since there are various other things that will be excluded).
2020-03-03 04:57:08 +01:00
add_subdirectory(external)
Code modifications to integrate Ledger HW device into monero-wallet-cli. The basic approach it to delegate all sensitive data (master key, secret ephemeral key, key derivation, ....) and related operations to the device. As device has low memory, it does not keep itself the values (except for view/spend keys) but once computed there are encrypted (with AES are equivalent) and return back to monero-wallet-cli. When they need to be manipulated by the device, they are decrypted on receive. Moreover, using the client for storing the value in encrypted form limits the modification in the client code. Those values are transfered from one C-structure to another one as previously. The code modification has been done with the wishes to be open to any other hardware wallet. To achieve that a C++ class hw::Device has been introduced. Two initial implementations are provided: the "default", which remaps all calls to initial Monero code, and the "Ledger", which delegates all calls to Ledger device.
2018-02-20 17:01:27 +01:00
Disable Trezor by default Trezor support currently doesn't work (Trezor's hardware wallet is hard-coded for a Monero net-id), and this saves needing to muck around with protobuf.
2021-01-08 20:42:33 +01:00
if(USE_DEVICE_TREZOR)
include(CheckTrezor)
endif()
device/trezor: trezor support added
2018-08-23 23:50:53 +02:00
moved all stuff to github
2014-03-03 23:07:58 +01:00
if(MSVC)
add_definitions("/bigobj /MP /W3 /GS- /D_CRT_SECURE_NO_WARNINGS /wd4996 /wd4345 /D_WIN32_WINNT=0x0600 /DWIN32_LEAN_AND_MEAN /DGTEST_HAS_TR1_TUPLE=0 /FIinline_c.h /D__SSE4_1__")
cmake: cleanup logic that sets flags per target/subdir The previous logic that used a COMMON_*_FLAGS intermediate variable and then re-assigned CMAKE_*_FLAGS before including each subdirectory was confusing and ugly. This PR is the right way to do it. This commit is purely refactoring: built binaries unchanged.
2016-07-29 03:19:01 +02:00
# set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} /Dinline=__inline")
set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} /STACK:10485760")
moved all stuff to github
2014-03-03 23:07:58 +01:00
if(STATIC)
foreach(VAR CMAKE_C_FLAGS_DEBUG CMAKE_CXX_FLAGS_DEBUG CMAKE_C_FLAGS_RELEASE CMAKE_CXX_FLAGS_RELEASE)
string(REPLACE "/MD" "/MT" ${VAR} "${${VAR}}")
endforeach()
endif()
include_directories(SYSTEM src/platform/msc)
else()
Updated +crypto option for ARMv8 to modify ${ARCH} instead
2016-09-18 04:23:15 +02:00
include(TestCXXAcceptsFlag)
build: use ARCH 'native' by default, allow to configure and override it
2018-10-13 11:46:19 +02:00
if (NOT ARCH)
set(ARCH native CACHE STRING "CPU to build for: -march value or 'default' to not pass -march at all")
Add support for apple and arm building Add pcsc-lite to linux builds Fixup windows icu4c linking with depends, the static libraries have an 's' appended to them Compiling depends arm-linux-gnueabihf will allow you to compile armv6zk monero binaries
2018-03-20 00:33:49 +01:00
endif()
cmake: don't set ARCH from CMAKE_SYSTEM_PROCESSOR It is not correct to do so, because ARCH should only take values supported by the -march argument, with the exception of 'default' which denotes not passing -march at all. ARCH defines the target architecture for builds that are intended to be portable to other machines.
2016-08-28 07:37:34 +02:00
message(STATUS "Building on ${CMAKE_SYSTEM_PROCESSOR} for ${ARCH}")
if(ARCH STREQUAL "default")
moved all stuff to github
2014-03-03 23:07:58 +01:00
set(ARCH_FLAG "")
enable monero build on ppc64le architecture Fix wrong flags for ppc64le: -maes - not avaiable -march - changed by -mcpu that does the same for powerpc
2017-06-27 19:06:01 +02:00
elseif(PPC64LE)
Fixed for compilation in big endian PPC. "Everything is broken thought due to little endian logic."
2018-05-20 00:03:59 +02:00
set(ARCH_FLAG "-mcpu=power8")
elseif(PPC64)
set(ARCH_FLAG "-mcpu=970")
elseif(PPC)
set(ARCH_FLAG "-mcpu=7400")
CMakeLists.txt - ios/xcode fix
2017-06-23 10:02:07 +02:00
elseif(IOS AND ARCH STREQUAL "arm64")
message(STATUS "IOS: Changing arch from arm64 to armv8")
set(ARCH_FLAG "-march=armv8")
Fix iOS Build (#908)
2019-11-01 02:41:31 +01:00
elseif(IOS AND ARCH STREQUAL "x86_64")
message(STATUS "IOS: Changing arch from x86_64 to x86-64")
set(ARCH_FLAG "-march=x86-64")
moved all stuff to github
2014-03-03 23:07:58 +01:00
else()
Revert "Interpret x86_64 as x86-64 for architecture" This reverts commit 86234921506944dddab4e2f5edf96bf747be6c73. Let's restrict ARCH to values accepted by -march to keep things clear and consistent. ARCH is -march, with only one exception: a value of "default" indicates to not pass -march at all.
2016-08-28 07:42:59 +02:00
set(ARCH_FLAG "-march=${ARCH}")
NetBSD support
2018-09-02 16:51:23 +02:00
if(ARCH STREQUAL "native")
check_c_compiler_flag(-march=native CC_SUPPORTS_MARCH_NATIVE)
if (NOT CC_SUPPORTS_MARCH_NATIVE)
check_c_compiler_flag(-mtune=native CC_SUPPORTS_MTUNE_NATIVE)
if (CC_SUPPORTS_MTUNE_NATIVE)
set(ARCH_FLAG "-mtune=${ARCH}")
else()
set(ARCH_FLAG "")
endif()
endif()
endif()
moved all stuff to github
2014-03-03 23:07:58 +01:00
endif()
Merge pull request #4457 85318e78 build: set ARCH_FLAG before compiler/linker flag checks (xiphon)
2018-09-29 22:16:19 +02:00
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} ${ARCH_FLAG}")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} ${ARCH_FLAG}")
Remove -Wundef from compilation flags This triggers a pile of false positives from gtest and mapbox variant. In the case of gtest, these were being hidden by including gtest as a system include, which was simply disgusting.
2020-06-10 01:09:23 +02:00
set(WARNINGS "-Wall -Wextra -Wpointer-arith -Wvla -Wwrite-strings -Wno-error=extra -Wno-error=deprecated-declarations -Wno-unused-parameter -Wno-unused-variable -Wno-error=unused-variable -Wno-error=uninitialized")
Add WARNINGS_AS_ERRORS flag for release builds
2019-09-16 05:04:03 +02:00
option(WARNINGS_AS_ERRORS "Enable warning as errors" OFF)
Fix WARNING_AS_ERROR -> WARNINGS_AS_ERRORS typo (#929)
2019-11-21 00:04:21 +01:00
if(NOT MINGW AND WARNINGS_AS_ERRORS)
Revert "Remove -Werror" This reverts commit 707c2f836b3136bc88b5aed374a4b3128f9596a7.
2019-03-09 03:04:57 +01:00
set(WARNINGS_AS_ERRORS_FLAG "-Werror")
endif()
Add WARNINGS_AS_ERRORS flag for release builds
2019-09-16 05:04:03 +02:00
Eliminate warnings from unknown warning flags on apple clang
2020-03-14 22:00:36 +01:00
if(CMAKE_C_COMPILER_ID STREQUAL "Clang" OR CMAKE_C_COMPILER_ID STREQUAL "AppleClang")
cmake: define ARM var for all ARM arch variants This is refactoring only. No behavior change.
2016-08-28 13:28:05 +02:00
if(ARM)
cmake: ARM: clang: make warning non-fatal: inline asm Clang issues a warning for some inline asm in stack_trace.cpp. This patch ieaves the warning to be displayed as a reminder to fix the code.
2016-07-11 03:11:18 +02:00
set(WARNINGS "${WARNINGS} -Wno-error=inline-asm")
endif()
moved all stuff to github
2014-03-03 23:07:58 +01:00
else()
Make false positive logical-op warning non-fatal A lokimq header has a range comparison that can trigger a tautological warning under xenial gcc; make it just a warning even with -Werror is turned on.
2020-05-27 20:01:07 +02:00
set(WARNINGS "${WARNINGS} -Wlogical-op -Wno-error=maybe-uninitialized -Wno-error=cpp -Wno-error=logical-op")
moved all stuff to github
2014-03-03 23:07:58 +01:00
endif()
if(MINGW)
Ignore another warning to make mingw happy
2014-08-06 18:32:59 +02:00
set(WARNINGS "${WARNINGS} -Wno-error=unused-value -Wno-error=unused-but-set-variable")
remove POSIX_C_SOURCE and remove dlfcn.h for static builds The former was a faulty "fix" for gmtime_r not existing on Windows. The latter is needed only for dynamic builds, and is not included with msys2, which ends up fine because Windows is only built static at this time.
2016-06-21 18:16:25 +02:00
set(MINGW_FLAG "${MINGW_FLAG} -DWIN32_LEAN_AND_MEAN")
Set the boost thread api for mingw
2014-08-06 18:43:01 +02:00
set(Boost_THREADAPI win32)
moved all stuff to github
2014-03-03 23:07:58 +01:00
include_directories(SYSTEM src/platform/mingw)
cmake: cleanup logic that sets flags per target/subdir The previous logic that used a COMMON_*_FLAGS intermediate variable and then re-assigned CMAKE_*_FLAGS before including each subdirectory was confusing and ugly. This PR is the right way to do it. This commit is purely refactoring: built binaries unchanged.
2016-07-29 03:19:01 +02:00
set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} -Wl,--stack,10485760")
Fix release flags flags were being accumulated instead of being set. Win32 was getting -O2 -DNDEBUG -Ofast -DNDEBUG etc... Also for Win32, stay compatible to WinXP
2016-01-02 20:39:03 +01:00
if(NOT BUILD_64)
add_definitions(-DWINVER=0x0501 -D_WIN32_WINNT=0x0501)
endif()
moved all stuff to github
2014-03-03 23:07:58 +01:00
endif()
set(C_WARNINGS "-Waggregate-return -Wnested-externs -Wold-style-definition -Wstrict-prototypes")
set(CXX_WARNINGS "-Wno-reorder -Wno-missing-field-initializers")
Build wallet with Android NDK
2017-01-05 02:11:05 +01:00
cmake: option+target for profiling for coverage
2016-08-29 20:23:57 +02:00
option(COVERAGE "Enable profiling for test coverage report" 0)
if(COVERAGE)
message(STATUS "Building with profiling for test coverage report")
set(COVERAGE_FLAGS "-fprofile-arcs -ftest-coverage --coverage")
endif()
Add misc hardening flags to the cmake machinery See https://wiki.debian.org/Hardening#User_Space
2017-12-22 22:48:31 +01:00
# if those don't work for your compiler, single it out where appropriate
fix building on OpenBSD, simplify Readme section Based on a patch from @mrme0w (#5569). Fixes #5770
2019-07-23 08:36:59 +02:00
if(CMAKE_BUILD_TYPE STREQUAL "Release" AND NOT OPENBSD)
Add WARNINGS_AS_ERRORS flag for release builds
2019-09-16 05:04:03 +02:00
set(C_SECURITY_FLAGS "${C_SECURITY_FLAGS} -U_FORTIFY_SOURCE -D_FORTIFY_SOURCE=1")
Add misc hardening flags to the cmake machinery See https://wiki.debian.org/Hardening#User_Space
2017-12-22 22:48:31 +01:00
set(C_SECURITY_FLAGS "${C_SECURITY_FLAGS} -U_FORTIFY_SOURCE -D_FORTIFY_SOURCE=1")
set(CXX_SECURITY_FLAGS "${CXX_SECURITY_FLAGS} -U_FORTIFY_SOURCE -D_FORTIFY_SOURCE=1")
endif()
# warnings
add_c_flag_if_supported(-Wformat C_SECURITY_FLAGS)
add_cxx_flag_if_supported(-Wformat CXX_SECURITY_FLAGS)
add_c_flag_if_supported(-Wformat-security C_SECURITY_FLAGS)
add_cxx_flag_if_supported(-Wformat-security CXX_SECURITY_FLAGS)
# -fstack-protector
build: fix building on Windows due to _FORTIFY_SOURCE changes in MSYS2 Also, enable other hardening options that work on Windows with GCC 9.x
2019-10-24 21:35:20 +02:00
if (NOT OPENBSD AND NOT (WIN32 AND (CMAKE_C_COMPILER_ID STREQUAL "GNU" AND CMAKE_C_COMPILER_VERSION VERSION_LESS 9.1)))
CMakeLists.txt: disable -fstack-protector* on windows Current GCC produces broken binaries with these options
2018-08-22 00:40:03 +02:00
add_c_flag_if_supported(-fstack-protector C_SECURITY_FLAGS)
add_cxx_flag_if_supported(-fstack-protector CXX_SECURITY_FLAGS)
add_c_flag_if_supported(-fstack-protector-strong C_SECURITY_FLAGS)
add_cxx_flag_if_supported(-fstack-protector-strong CXX_SECURITY_FLAGS)
endif()
Add misc hardening flags to the cmake machinery See https://wiki.debian.org/Hardening#User_Space
2017-12-22 22:48:31 +01:00
Fix iOS Build (#908)
2019-11-01 02:41:31 +01:00
# -fno-stack-check
if (IOS)
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -fno-stack-check")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fno-stack-check")
endif()
CMakeLists.txt: detect -fcf-protection=full and -fstack-clash-protection Introduced with GCC 8.2
2018-07-26 15:07:04 +02:00
# New in GCC 8.2
build: fix building on Windows due to _FORTIFY_SOURCE changes in MSYS2 Also, enable other hardening options that work on Windows with GCC 9.x
2019-10-24 21:35:20 +02:00
if (NOT OPENBSD AND NOT (WIN32 AND (CMAKE_C_COMPILER_ID STREQUAL "GNU" AND CMAKE_C_COMPILER_VERSION VERSION_LESS 9.1)))
CMakeLists.txt: detect -fcf-protection=full and -fstack-clash-protection Introduced with GCC 8.2
2018-07-26 15:07:04 +02:00
add_c_flag_if_supported(-fcf-protection=full C_SECURITY_FLAGS)
add_cxx_flag_if_supported(-fcf-protection=full CXX_SECURITY_FLAGS)
build: fix building on Windows due to _FORTIFY_SOURCE changes in MSYS2 Also, enable other hardening options that work on Windows with GCC 9.x
2019-10-24 21:35:20 +02:00
endif()
Disable -fstack-clash-protection on macOS It breaks the build on the latest Big Sur Xcode version because it accepts the option but then later produces warnings about it being unused when trying to use it.
2021-05-12 16:34:44 +02:00
if (NOT WIN32 AND NOT OPENBSD AND NOT APPLE)
CMakeLists.txt: detect -fcf-protection=full and -fstack-clash-protection Introduced with GCC 8.2
2018-07-26 15:07:04 +02:00
add_c_flag_if_supported(-fstack-clash-protection C_SECURITY_FLAGS)
add_cxx_flag_if_supported(-fstack-clash-protection CXX_SECURITY_FLAGS)
endif()
Disable stringop-overflow warnings under GCC12 They are really buggy and produce tons of false positives and, as far as I've seen, no true positives so just disable them. I kept this only to 12.x in the hope that gcc 13 will improve the warning's usefulness.
2022-12-20 23:23:39 +01:00
# GCC 12's stringop-overflow warnings are really broken, with tons and tons of false positives all
# over the place (not just in our code, but also in its own stdlibc++ code).
if (CMAKE_CXX_COMPILER_ID STREQUAL "GNU" AND CMAKE_CXX_COMPILER_VERSION VERSION_GREATER 12 AND CMAKE_CXX_COMPILER_VERSION VERSION_LESS 13)
set(CXX_WARNINGS "${CXX_WARNINGS} -Wno-stringop-overflow")
endif()
cmake: do not use -mmitigate-rop on GCC >= 9.1 It was removed, but it still accepted by the compiler, which warns for every file
2019-06-08 20:33:04 +02:00
# Removed in GCC 9.1 (or before ?), but still accepted, so spams the output
cmake simplifications Remove some (already fixed) workarounds and logic that only applies to compiler versions we left behind.
2020-05-13 20:34:25 +02:00
if (NOT CMAKE_C_COMPILER_ID STREQUAL "GNU" OR CMAKE_C_COMPILER_VERSION VERSION_LESS 9.1)
cmake: do not use -mmitigate-rop on GCC >= 9.1 It was removed, but it still accepted by the compiler, which warns for every file
2019-06-08 20:33:04 +02:00
add_c_flag_if_supported(-mmitigate-rop C_SECURITY_FLAGS)
add_cxx_flag_if_supported(-mmitigate-rop CXX_SECURITY_FLAGS)
endif()
CMakeLists.txt: add -mmitigate-rop to security flags if found This is GCC/x86/x86_64 only
2018-08-07 15:05:52 +02:00
Add misc hardening flags to the cmake machinery See https://wiki.debian.org/Hardening#User_Space
2017-12-22 22:48:31 +01:00
# linker
add_linker_flag_if_supported(-Wl,-z,relro LD_SECURITY_FLAGS)
add_linker_flag_if_supported(-Wl,-z,now LD_SECURITY_FLAGS)
add_linker_flag_if_supported(-Wl,-z,noexecstack noexecstack_SUPPORTED)
if (noexecstack_SUPPORTED)
set(LD_SECURITY_FLAGS "${LD_SECURITY_FLAGS} -Wl,-z,noexecstack")
endif()
add_linker_flag_if_supported(-Wl,-z,noexecheap noexecheap_SUPPORTED)
if (noexecheap_SUPPORTED)
set(LD_SECURITY_FLAGS "${LD_SECURITY_FLAGS} -Wl,-z,noexecheap")
endif()
Add glibc back compat code To ensure that the binaries compiled by gitian run across many linux distributions, enforce 2.17 as the minimum libc version supported.
2018-10-13 17:07:56 +02:00
if(BACKCOMPAT)
add_linker_flag_if_supported(-Wl,--wrap=__divmoddi4 LD_BACKCOMPAT_FLAGS)
add_linker_flag_if_supported(-Wl,--wrap=glob LD_BACKCOMPAT_FLAGS)
message(STATUS "Using Lib C back compat flags: ${LD_BACKCOMPAT_FLAGS}")
endif()
Add misc hardening flags to the cmake machinery See https://wiki.debian.org/Hardening#User_Space
2017-12-22 22:48:31 +01:00
# some windows linker bits
if (WIN32)
add_linker_flag_if_supported(-Wl,--dynamicbase LD_SECURITY_FLAGS)
add_linker_flag_if_supported(-Wl,--nxcompat LD_SECURITY_FLAGS)
Windows: enable high-entropy ASLR where available
2019-10-24 21:40:11 +02:00
add_linker_flag_if_supported(-Wl,--high-entropy-va LD_SECURITY_FLAGS)
Add misc hardening flags to the cmake machinery See https://wiki.debian.org/Hardening#User_Space
2017-12-22 22:48:31 +01:00
endif()
message(STATUS "Using C security hardening flags: ${C_SECURITY_FLAGS}")
message(STATUS "Using C++ security hardening flags: ${CXX_SECURITY_FLAGS}")
message(STATUS "Using linker security hardening flags: ${LD_SECURITY_FLAGS}")
Use cmake variable for enabling PIC With the current approach -fPIC wasn't getting used when building external libraries. I tried moving add_subdirectory(external) down below where we set flags, but that results in a slew or warning because it *also* turn on a bunch of warnings that aren't safe in the various external code.
2020-07-16 19:14:19 +02:00
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -D_GNU_SOURCE ${MINGW_FLAG} ${WARNINGS} ${C_WARNINGS} ${COVERAGE_FLAGS} ${C_SECURITY_FLAGS}")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -D_GNU_SOURCE ${MINGW_FLAG} ${WARNINGS} ${CXX_WARNINGS} ${COVERAGE_FLAGS} ${CXX_SECURITY_FLAGS}")
Add glibc back compat code To ensure that the binaries compiled by gitian run across many linux distributions, enforce 2.17 as the minimum libc version supported.
2018-10-13 17:07:56 +02:00
set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} ${LD_SECURITY_FLAGS} ${LD_BACKCOMPAT_FLAGS}")
Updated +crypto option for ARMv8 to modify ${ARCH} instead
2016-09-18 04:23:15 +02:00
Add ARMv8 Handling to CMakeLists.txt - version 2 Adds 64-bit ARMv8 handling to CMakeLists.txt and implements GCC flags for two errata on the Cortex-A53 present on every chip in revision 0: https://developer.arm.com/docs/epm048406/latest/arm-processor-cortex-a53-mpcore-product-revision-r0-software-developers-errata-notice 835769 affects 64-bit multiply accumulate 843419 affects internal page addressing Rewritten to build on @radfish's changes Updated to address @radfish's points
2016-08-29 12:13:55 +02:00
if(ARM)
message(STATUS "Setting FPU Flags for ARM Processors")
#NB NEON hardware does not fully implement the IEEE 754 standard for floating-point arithmetic
#Need custom assembly code to take full advantage of NEON SIMD
#Cortex-A5/9 -mfpu=neon-fp16
#Cortex-A7/15 -mfpu=neon-vfpv4
#Cortex-A8 -mfpu=neon
Updated +crypto option for ARMv8 to modify ${ARCH} instead
2016-09-18 04:23:15 +02:00
#ARMv8 -FP and SIMD on by default for all ARM8v-A series, NO -mfpu setting needed
Add ARMv8 Handling to CMakeLists.txt - version 2 Adds 64-bit ARMv8 handling to CMakeLists.txt and implements GCC flags for two errata on the Cortex-A53 present on every chip in revision 0: https://developer.arm.com/docs/epm048406/latest/arm-processor-cortex-a53-mpcore-product-revision-r0-software-developers-errata-notice 835769 affects 64-bit multiply accumulate 843419 affects internal page addressing Rewritten to build on @radfish's changes Updated to address @radfish's points
2016-08-29 12:13:55 +02:00
#For custom -mtune, processor IDs for ARMv8-A series:
#0xd04 - Cortex-A35
#0xd07 - Cortex-A57
#0xd08 - Cortex-A72
#0xd03 - Cortex-A73
if(NOT ARM8)
CHECK_CXX_ACCEPTS_FLAG(-mfpu=vfp3-d16 CXX_ACCEPTS_VFP3_D16)
CHECK_CXX_ACCEPTS_FLAG(-mfpu=vfp4 CXX_ACCEPTS_VFP4)
CHECK_CXX_ACCEPTS_FLAG(-mfloat-abi=hard CXX_ACCEPTS_MFLOAT_HARD)
CHECK_CXX_ACCEPTS_FLAG(-mfloat-abi=softfp CXX_ACCEPTS_MFLOAT_SOFTFP)
endif()
add arm6 target and flags
2015-04-06 14:00:09 +02:00
Add ARMv8 Handling to CMakeLists.txt - version 2 Adds 64-bit ARMv8 handling to CMakeLists.txt and implements GCC flags for two errata on the Cortex-A53 present on every chip in revision 0: https://developer.arm.com/docs/epm048406/latest/arm-processor-cortex-a53-mpcore-product-revision-r0-software-developers-errata-notice 835769 affects 64-bit multiply accumulate 843419 affects internal page addressing Rewritten to build on @radfish's changes Updated to address @radfish's points
2016-08-29 12:13:55 +02:00
if(ARM8)
CHECK_CXX_ACCEPTS_FLAG(-mfix-cortex-a53-835769 CXX_ACCEPTS_MFIX_CORTEX_A53_835769)
CHECK_CXX_ACCEPTS_FLAG(-mfix-cortex-a53-843419 CXX_ACCEPTS_MFIX_CORTEX_A53_843419)
endif()
if(ARM6)
message(STATUS "Selecting VFP for ARMv6")
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -mfpu=vfp")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mfpu=vfp")
endif(ARM6)
if(ARM7)
if(CXX_ACCEPTS_VFP3_D16 AND NOT CXX_ACCEPTS_VFP4)
message(STATUS "Selecting VFP3 for ARMv7")
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -mfpu=vfp3-d16")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mfpu=vfp3-d16")
endif()
if(CXX_ACCEPTS_VFP4)
message(STATUS "Selecting VFP4 for ARMv7")
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -mfpu=vfp4")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mfpu=vfp4")
endif()
if(CXX_ACCEPTS_MFLOAT_HARD)
message(STATUS "Setting Hardware ABI for Floating Point")
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -mfloat-abi=hard")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mfloat-abi=hard")
endif()
if(CXX_ACCEPTS_MFLOAT_SOFTFP AND NOT CXX_ACCEPTS_MFLOAT_HARD)
message(STATUS "Setting Software ABI for Floating Point")
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -mfloat-abi=softfp")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mfloat-abi=softfp")
endif()
endif(ARM7)
if(ARM8)
if(CXX_ACCEPTS_MFIX_CORTEX_A53_835769)
message(STATUS "Enabling Cortex-A53 workaround 835769")
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -mfix-cortex-a53-835769")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mfix-cortex-a53-835769")
endif()
if(CXX_ACCEPTS_MFIX_CORTEX_A53_843419)
message(STATUS "Enabling Cortex-A53 workaround 843419")
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -mfix-cortex-a53-843419")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mfix-cortex-a53-843419")
endif()
endif(ARM8)
endif(ARM)
** CHANGES ARE EXPERIMENTAL (FOR TESTING ONLY) Bockchain: 1. Optim: Multi-thread long-hash computation when encountering groups of blocks. 2. Optim: Cache verified txs and return result from cache instead of re-checking whenever possible. 3. Optim: Preload output-keys when encoutering groups of blocks. Sort by amount and global-index before bulk querying database and multi-thread when possible. 4. Optim: Disable double spend check on block verification, double spend is already detected when trying to add blocks. 5. Optim: Multi-thread signature computation whenever possible. 6. Patch: Disable locking (recursive mutex) on called functions from check_tx_inputs which causes slowdowns (only seems to happen on ubuntu/VMs??? Reason: TBD) 7. Optim: Removed looped full-tx hash computation when retrieving transactions from pool (???). 8. Optim: Cache difficulty/timestamps (735 blocks) for next-difficulty calculations so that only 2 db reads per new block is needed when a new block arrives (instead of 1470 reads). Berkeley-DB: 1. Fix: 32-bit data errors causing wrong output global indices and failure to send blocks to peers (etc). 2. Fix: Unable to pop blocks on reorganize due to transaction errors. 3. Patch: Large number of transaction aborts when running multi-threaded bulk queries. 4. Patch: Insufficient locks error when running full sync. 5. Patch: Incorrect db stats when returning from an immediate exit from "pop block" operation. 6. Optim: Add bulk queries to get output global indices. 7. Optim: Modified output_keys table to store public_key+unlock_time+height for single transaction lookup (vs 3) 8. Optim: Used output_keys table retrieve public_keys instead of going through output_amounts->output_txs+output_indices->txs->output:public_key 9. Optim: Added thread-safe buffers used when multi-threading bulk queries. 10. Optim: Added support for nosync/write_nosync options for improved performance (*see --db-sync-mode option for details) 11. Mod: Added checkpoint thread and auto-remove-logs option. 12. *Now usable on 32-bit systems like RPI2. LMDB: 1. Optim: Added custom comparison for 256-bit key tables (minor speed-up, TBD: get actual effect) 2. Optim: Modified output_keys table to store public_key+unlock_time+height for single transaction lookup (vs 3) 3. Optim: Used output_keys table retrieve public_keys instead of going through output_amounts->output_txs+output_indices->txs->output:public_key 4. Optim: Added support for sync/writemap options for improved performance (*see --db-sync-mode option for details) 5. Mod: Auto resize to +1GB instead of multiplier x1.5 ETC: 1. Minor optimizations for slow-hash for ARM (RPI2). Incomplete. 2. Fix: 32-bit saturation bug when computing next difficulty on large blocks. [PENDING ISSUES] 1. Berkely db has a very slow "pop-block" operation. This is very noticeable on the RPI2 as it sometimes takes > 10 MINUTES to pop a block during reorganization. This does not happen very often however, most reorgs seem to take a few seconds but it possibly depends on the number of outputs present. TBD. 2. Berkeley db, possible bug "unable to allocate memory". TBD. [NEW OPTIONS] (*Currently all enabled for testing purposes) 1. --fast-block-sync arg=[0:1] (default: 1) a. 0 = Compute long hash per block (may take a while depending on CPU) b. 1 = Skip long-hash and verify blocks based on embedded known good block hashes (faster, minimal CPU dependence) 2. --db-sync-mode arg=[[safe|fast|fastest]:[sync|async]:[nblocks_per_sync]] (default: fastest:async:1000) a. safe = fdatasync/fsync (or equivalent) per stored block. Very slow, but safest option to protect against power-out/crash conditions. b. fast/fastest = Enables asynchronous fdatasync/fsync (or equivalent). Useful for battery operated devices or STABLE systems with UPS and/or systems with battery backed write cache/solid state cache. Fast - Write meta-data but defer data flush. Fastest - Defer meta-data and data flush. Sync - Flush data after nblocks_per_sync and wait. Async - Flush data after nblocks_per_sync but do not wait for the operation to finish. 3. --prep-blocks-threads arg=[n] (default: 4 or system max threads, whichever is lower) Max number of threads to use when computing long-hash in groups. 4. --show-time-stats arg=[0:1] (default: 1) Show benchmark related time stats. 5. --db-auto-remove-logs arg=[0:1] (default: 1) For berkeley-db only. Auto remove logs if enabled. **Note: lmdb and berkeley-db have changes to the tables and are not compatible with official git head version. At the moment, you need a full resync to use this optimized version. [PERFORMANCE COMPARISON] **Some figures are approximations only. Using a baseline machine of an i7-2600K+SSD+(with full pow computation): 1. The optimized lmdb/blockhain core can process blocks up to 585K for ~1.25 hours + download time, so it usually takes 2.5 hours to sync the full chain. 2. The current head with memory can process blocks up to 585K for ~4.2 hours + download time, so it usually takes 5.5 hours to sync the full chain. 3. The current head with lmdb can process blocks up to 585K for ~32 hours + download time and usually takes 36 hours to sync the full chain. Averate procesing times (with full pow computation): lmdb-optimized: 1. tx_ave = 2.5 ms / tx 2. block_ave = 5.87 ms / block memory-official-repo: 1. tx_ave = 8.85 ms / tx 2. block_ave = 19.68 ms / block lmdb-official-repo (0f4a036437fd41a5498ee5e74e2422ea6177aa3e) 1. tx_ave = 47.8 ms / tx 2. block_ave = 64.2 ms / block **Note: The following data denotes processing times only (does not include p2p download time) lmdb-optimized processing times (with full pow computation): 1. Desktop, Quad-core / 8-threads 2600k (8Mb) - 1.25 hours processing time (--db-sync-mode=fastest:async:1000). 2. Laptop, Dual-core / 4-threads U4200 (3Mb) - 4.90 hours processing time (--db-sync-mode=fastest:async:1000). 3. Embedded, Quad-core / 4-threads Z3735F (2x1Mb) - 12.0 hours processing time (--db-sync-mode=fastest:async:1000). lmdb-optimized processing times (with per-block-checkpoint) 1. Desktop, Quad-core / 8-threads 2600k (8Mb) - 10 minutes processing time (--db-sync-mode=fastest:async:1000). berkeley-db optimized processing times (with full pow computation) 1. Desktop, Quad-core / 8-threads 2600k (8Mb) - 1.8 hours processing time (--db-sync-mode=fastest:async:1000). 2. RPI2. Improved from estimated 3 months(???) into 2.5 days (*Need 2AMP supply + Clock:1Ghz + [usb+ssd] to achieve this speed) (--db-sync-mode=fastest:async:1000). berkeley-db optimized processing times (with per-block-checkpoint) 1. RPI2. 12-15 hours (*Need 2AMP supply + Clock:1Ghz + [usb+ssd] to achieve this speed) (--db-sync-mode=fastest:async:1000).
2015-07-10 22:09:32 +02:00
mac osx building fixes
2014-04-30 22:50:06 +02:00
if(APPLE)
Merge pull request #4446 69da14e1 fixes make debug compilation on OSX (Dusan Klinec)
2018-09-29 22:14:28 +02:00
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fvisibility=default -DGTEST_HAS_TR1_TUPLE=0")
mac osx building fixes
2014-04-30 22:50:06 +02:00
endif()
cmake: option+target for profiling for coverage
2016-08-29 20:23:57 +02:00
CMakeLists.txt: add -ftemplate-depth=900
2018-11-12 23:35:21 +01:00
# At least some CLANGs default to not enough for monero
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -ftemplate-depth=900")
cmake: pass -static when STATIC=ON on Windows This gets rid of bitmonerod.exe's dependecy on libwindpthreads-1.dll in build on Windows on x86_64 (via MSYS2 default toolchain). With this patch all DLL dependencies are on DLLs in c:\windows\system32.
2016-08-28 00:47:07 +02:00
if(STATIC)
# STATIC already configures most deps to be linked in statically,
# here we make more deps static if the platform permits it
if (MINGW)
# On Windows, this is as close to fully-static as we get:
# this leaves only deps on /c/Windows/system32/*.dll
set(STATIC_FLAGS "-static")
Preliminary support for DragonFly BSD - It builds but no further testing has been done.
2016-12-15 11:27:53 +01:00
elseif (NOT (APPLE OR FREEBSD OR OPENBSD OR DRAGONFLY))
cmake: pass -static when STATIC=ON on Windows This gets rid of bitmonerod.exe's dependecy on libwindpthreads-1.dll in build on Windows on x86_64 (via MSYS2 default toolchain). With this patch all DLL dependencies are on DLLs in c:\windows\system32.
2016-08-28 00:47:07 +02:00
# On Linux, we don't support fully static build, but these can be static
set(STATIC_FLAGS "-static-libgcc -static-libstdc++")
endif()
set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} ${STATIC_FLAGS} ")
cmake: install throw hook in OSX build too This is an attempt to fix build with STATIC=ON on OSX (#932): [ 95%] Linking CXX executable ../../bin/bitmonerod Undefined symbols for architecture x86_64: "___real___cxa_throw", referenced from: ___wrap___cxa_throw in libcommon.a(stack_trace.cpp.o) ld: symbol(s) not found for architecture x86_64
2016-07-26 04:30:59 +02:00
endif()
moved all stuff to github
2014-03-03 23:07:58 +01:00
endif()
Set up LTO properly CMake 3.9+ has generic LTO enabling code, switch to it. Update required cmake version to 3.10 (3.9 is probably sufficient, but 3.10 is bionic's version that we're actually testing).
2020-06-12 22:24:03 +02:00
2014 network limit 1.0a +utils +toc -doc -drmonero commands and options for network limiting works very well e.g. for 50 KiB/sec up and down ToS (QoS) flag peer number limit TODO some spikes in ingress/download TODO problems when other up and down limit added "otshell utils" - simple logging (with colors, text files channels)
2015-01-05 20:30:17 +01:00
if (${BOOST_IGNORE_SYSTEM_PATHS} STREQUAL "ON")
cmake: minor cleanups (indentation and typos)
2014-10-21 20:24:49 +02:00
set(Boost_NO_SYSTEM_PATHS TRUE)
moved msys2 toolchain, boost libs fix
2014-09-22 12:30:53 +02:00
endif()
Fixed bug in static linking boost on MINGW There was a workaround for linking to boost at all on MINGW, but unfortunately this workaround would not correctly link to boost statically. This workaround for that workaround works around the issue that that workaround had.
2015-03-22 10:26:30 +01:00
set(OLD_LIB_SUFFIXES ${CMAKE_FIND_LIBRARY_SUFFIXES})
moved all stuff to github
2014-03-03 23:07:58 +01:00
if(STATIC)
Fixed bug in static linking boost on MINGW There was a workaround for linking to boost at all on MINGW, but unfortunately this workaround would not correctly link to boost statically. This workaround for that workaround works around the issue that that workaround had.
2015-03-22 10:26:30 +01:00
if(MINGW)
set(CMAKE_FIND_LIBRARY_SUFFIXES .a)
Revert Boost_NO_BOOST_CMAKE, set rpc version 3.4, specify debug utils as option
2020-04-13 03:24:13 +02:00
set(Boost_NO_BOOST_CMAKE ON)
Fixed bug in static linking boost on MINGW There was a workaround for linking to boost at all on MINGW, but unfortunately this workaround would not correctly link to boost statically. This workaround for that workaround works around the issue that that workaround had.
2015-03-22 10:26:30 +01:00
endif()
moved all stuff to github
2014-03-03 23:07:58 +01:00
set(Boost_USE_STATIC_LIBS ON)
set(Boost_USE_STATIC_RUNTIME ON)
endif()
cmake modernization The archaic (i.e. decade old) cmake usage here really got in the way of trying to properly use newer libraries (like lokimq), so this undertakes overhauling it considerably to make it much more sane (and significantly reduce the size). I left more of the architecture-specific bits in the top-level CMakeLists.txt intact; most of the efforts here are about properly loading dependencies, specifying dependencies and avoiding a whole pile of cmake antipatterns. This bumps the required cmake version to 3.5, which is what xenial comes with. - extensive use of interface libraries to include libraries, definitions, and include paths - use Boost::whatever instead of ${Boost_WHATEVER_LIBRARY}. The interface targets are (again) much better as they also give you any needed include or linking flags without needing to worry about them. - don't list header files when building things. This has *never* been correct cmake usage (cmake has always known how to wallet_rpc_headers the headers that .cpp files include to know about build changes). - remove the loki_add_library monstrosity; it breaks target names and makes compiling less efficient because the author couldn't figure out how to link things together. - make loki_add_executable take the output filename, and set the output path to bin/ and install to bin because *every single usage* of loki_add_executable was immediately followed by setting the output filename and setting the output path to bin/ and installing to bin. - move a bunch of crap that is only used in one particular src/whatever/CMakeLists.txt into that particular CMakeLists.txt instead of the top level CMakeLists.txt (or src/CMakeLists.txt). - Remove a bunch of redundant dependencies; most of them look like they were just copy-and-pasted in, and many more aren't needed (since they are implied by the PUBLIC linking of other dependencies). - Removed `die` since it just does a FATAL_ERROR, but adds color (which is useless since CMake already makes FATAL_ERRORs perfectly visible). - Change the way LOKI_DAEMON_AND_WALLET_ONLY works to just change the make targets to daemon and simplewallet rather than changing the build process (this should make it faster, too, since there are various other things that will be excluded).
2020-03-03 04:57:08 +01:00
set(Boost_USE_MULTITHREADED TRUE) # Needed for macOS, at least, and won't hurt elsewhere
Enable Boost::locale for Windows only Not needed anywhere else, this change is mainly for Android, compiling Boost::locale means also compiling ICU which can get complicated and we're able to completely side step.
2020-03-23 01:51:16 +01:00
Static builds: make usable binaries from cmake This adds a static dependency script for libraries like boost, unbound, etc. to cmake, invokable with: cmake .. -DBUILD_STATIC_DEPS=ON which downloads and builds static versions of all our required dependencies (boost, unbound, openssl, ncurses, etc.). It also implies -DSTATIC=ON to build other vendored deps (like miniupnpc, lokimq) as static as well. Unlike the contrib/depends system, this is easier to maintain (one script using nicer cmake with functions instead of raw Makefile spaghetti code), and isn't concerned with reproducible builds -- this doesn't rebuild the compiler, for instance. It also works with the existing build system so that it is simply another way to invoke the cmake build scripts but doesn't require any external tooling. This works on Linux, Mac, and Windows. Some random comments on this commit (for preserving history): - Don't use target_link_libraries on imported targets. Newer cmake is fine with it, but Bionic's cmake doesn't like it but seems okay with setting the properties directly. - This rebuilds libzmq and libsodium, even though there is some provision already within loki-core to do so: however, the existing embedded libzmq fails with the static deps because it uses libzmq's cmake build script, which relies on pkg-config to find libsodium which ends up finding the system one (or not finding any), rather than the one we build with DownloadLibSodium. Since both libsodium and libzmq are faily simple builds it seemed easiest to just add them to the cmake static build rather than trying to shoehorn the current code into the static build script. - Half of the protobuf build system ignores CC/CXX just because Google, and there's no documentation anywhere except for a random closed bug report about needing to set these other variables (CC_FOR_BUILD, CXX_FOR_BUILD) instead, but you need to. Thanks Google. - The boost build is set to output very little because even the minimum -d1 output level spams ~15k lines of output just for the headers it installs.
2020-06-11 20:19:44 +02:00
if(BUILD_STATIC_DEPS)
# StaticBuild.cmake sets Boost targets up for us
Enable Boost::locale for Windows only Not needed anywhere else, this change is mainly for Android, compiling Boost::locale means also compiling ICU which can get complicated and we're able to completely side step.
2020-03-23 01:51:16 +01:00
else()
Remove openssl, unbound, expat, OpenAlias openssl is a miserable dependency to fight with, especially for iOS/Android, and we use it for very little: - it gets (mis)used to write base64 data to a file in wallet2 for things like multisig data. However this mode is *not* enabled by default, basically completely unknown, completely unused, only exists in the cli wallet, and is just dumb. (Honestly the justification given in the PR is that "Windows users might want it", presupposing that there exists Windows users who are capable of generating a multisig wallet in a CLI-only application and yet are incapable of dealing with binary files). - it's a dependency of unbound (in order to do dnssec, I believe). Unbound itself is fairly useless for Oxen, so I've removed it too: - it does OpenAlias lookups, which are a Monero thing that has never been used outside Monero, doesn't work reliably (because it fails if the result isn't DNSSEC validated) and is pointless when we have ONS. - it does DNS resolution on seed nodes, but we have never set seed nodes by name and just use seed node IPs instead (which seems a bit better anyway since the DNS lookup leaks some metadata). - it *was* being used for sha256, but an earlier commit in this PR already replaced that with libsodium (entirely coincidentally). - for static deps, it enables HTTPS support for the wallet. However only the CLI wallet actually supports this (the GUI and mobile wallets don't), and since oxend hasn't support this for a while I have strong doubts it is being used anywhere. (Plus wallet3 will do everything encrypted using zmq/libsodium, so doesn't need this to be secure). Note here that it is *only* removed by this commit for static builds: if doing a system build that links to libcurl supporting HTTPS then HTTPS support will still work. Libexpat is also gone because it was only there for libunbound.
2022-04-15 01:11:36 +02:00
find_package(Boost 1.62 QUIET REQUIRED COMPONENTS system thread serialization program_options)
Enable Boost::locale for Windows only Not needed anywhere else, this change is mainly for Android, compiling Boost::locale means also compiling ICU which can get complicated and we're able to completely side step.
2020-03-23 01:51:16 +01:00
endif()
dependencies enforced, send que error message log level moved
2014-09-29 20:13:15 +02:00
Fixed bug in static linking boost on MINGW There was a workaround for linking to boost at all on MINGW, but unfortunately this workaround would not correctly link to boost statically. This workaround for that workaround works around the issue that that workaround had.
2015-03-22 10:26:30 +01:00
set(CMAKE_FIND_LIBRARY_SUFFIXES ${OLD_LIB_SUFFIXES})
dependencies enforced, send que error message log level moved
2014-09-29 20:13:15 +02:00
if(NOT Boost_FOUND)
cmake modernization The archaic (i.e. decade old) cmake usage here really got in the way of trying to properly use newer libraries (like lokimq), so this undertakes overhauling it considerably to make it much more sane (and significantly reduce the size). I left more of the architecture-specific bits in the top-level CMakeLists.txt intact; most of the efforts here are about properly loading dependencies, specifying dependencies and avoiding a whole pile of cmake antipatterns. This bumps the required cmake version to 3.5, which is what xenial comes with. - extensive use of interface libraries to include libraries, definitions, and include paths - use Boost::whatever instead of ${Boost_WHATEVER_LIBRARY}. The interface targets are (again) much better as they also give you any needed include or linking flags without needing to worry about them. - don't list header files when building things. This has *never* been correct cmake usage (cmake has always known how to wallet_rpc_headers the headers that .cpp files include to know about build changes). - remove the loki_add_library monstrosity; it breaks target names and makes compiling less efficient because the author couldn't figure out how to link things together. - make loki_add_executable take the output filename, and set the output path to bin/ and install to bin because *every single usage* of loki_add_executable was immediately followed by setting the output filename and setting the output path to bin/ and installing to bin. - move a bunch of crap that is only used in one particular src/whatever/CMakeLists.txt into that particular CMakeLists.txt instead of the top level CMakeLists.txt (or src/CMakeLists.txt). - Remove a bunch of redundant dependencies; most of them look like they were just copy-and-pasted in, and many more aren't needed (since they are implied by the PUBLIC linking of other dependencies). - Removed `die` since it just does a FATAL_ERROR, but adds color (which is useless since CMake already makes FATAL_ERRORs perfectly visible). - Change the way LOKI_DAEMON_AND_WALLET_ONLY works to just change the make targets to daemon and simplewallet rather than changing the build process (this should make it faster, too, since there are various other things that will be excluded).
2020-03-03 04:57:08 +01:00
message(FATAL_ERROR "Could not find Boost libraries, please make sure you have installed Boost or libboost-all-dev (>=1.58) or the equivalent")
Report Boost library version during compilation
2016-10-27 22:38:29 +02:00
elseif(Boost_FOUND)
message(STATUS "Found Boost Version: ${Boost_VERSION}")
moved all stuff to github
2014-03-03 23:07:58 +01:00
endif()
dependencies enforced, send que error message log level moved
2014-09-29 20:13:15 +02:00
cmake modernization The archaic (i.e. decade old) cmake usage here really got in the way of trying to properly use newer libraries (like lokimq), so this undertakes overhauling it considerably to make it much more sane (and significantly reduce the size). I left more of the architecture-specific bits in the top-level CMakeLists.txt intact; most of the efforts here are about properly loading dependencies, specifying dependencies and avoiding a whole pile of cmake antipatterns. This bumps the required cmake version to 3.5, which is what xenial comes with. - extensive use of interface libraries to include libraries, definitions, and include paths - use Boost::whatever instead of ${Boost_WHATEVER_LIBRARY}. The interface targets are (again) much better as they also give you any needed include or linking flags without needing to worry about them. - don't list header files when building things. This has *never* been correct cmake usage (cmake has always known how to wallet_rpc_headers the headers that .cpp files include to know about build changes). - remove the loki_add_library monstrosity; it breaks target names and makes compiling less efficient because the author couldn't figure out how to link things together. - make loki_add_executable take the output filename, and set the output path to bin/ and install to bin because *every single usage* of loki_add_executable was immediately followed by setting the output filename and setting the output path to bin/ and installing to bin. - move a bunch of crap that is only used in one particular src/whatever/CMakeLists.txt into that particular CMakeLists.txt instead of the top level CMakeLists.txt (or src/CMakeLists.txt). - Remove a bunch of redundant dependencies; most of them look like they were just copy-and-pasted in, and many more aren't needed (since they are implied by the PUBLIC linking of other dependencies). - Removed `die` since it just does a FATAL_ERROR, but adds color (which is useless since CMake already makes FATAL_ERRORs perfectly visible). - Change the way LOKI_DAEMON_AND_WALLET_ONLY works to just change the make targets to daemon and simplewallet rather than changing the build process (this should make it faster, too, since there are various other things that will be excluded).
2020-03-03 04:57:08 +01:00
# Interface target for random extra system libraries that we need to link everything against
add_library(extra INTERFACE)
target_link_libraries(extra INTERFACE ${CMAKE_DL_LIBS})
target_link_libraries(extra INTERFACE Boost::boost)
target_link_libraries(extra INTERFACE Threads::Threads)
Various linking and build fixes - updating to latest loki-mq (1.0.0 + various linking fixes) - BUILD_SHARED_LIBS was being handled very strangely; make it a full option instead (defaulting to off) that a cmake invoker can specify, as per cmake recommendations. - travis ci tweaks/changes: - Add a static bionic build - Simplify cmake argument code - Add `--version` invocation for lokid and loki-wallet-cli to test that the binaries were linked properly. - always build an embedded sodium statically; if we do it dynamically and an older system one exists we are going to have trouble. - don't force epee and blocks to be static; rather they get controlled by the above BUILD_SHARED_LIBS, just like all the other internal libraries. - use some PkgConfig:: imported targets rather than bunch-of-variables.
2020-03-14 06:21:38 +01:00
if(APPLE AND BUILD_SHARED_LIBS)
# Don't crap out on circular dependencies between internal libs on macOS when doing a shared build
target_link_libraries(extra INTERFACE "-undefined dynamic_lookup")
endif()
cmake modernization The archaic (i.e. decade old) cmake usage here really got in the way of trying to properly use newer libraries (like lokimq), so this undertakes overhauling it considerably to make it much more sane (and significantly reduce the size). I left more of the architecture-specific bits in the top-level CMakeLists.txt intact; most of the efforts here are about properly loading dependencies, specifying dependencies and avoiding a whole pile of cmake antipatterns. This bumps the required cmake version to 3.5, which is what xenial comes with. - extensive use of interface libraries to include libraries, definitions, and include paths - use Boost::whatever instead of ${Boost_WHATEVER_LIBRARY}. The interface targets are (again) much better as they also give you any needed include or linking flags without needing to worry about them. - don't list header files when building things. This has *never* been correct cmake usage (cmake has always known how to wallet_rpc_headers the headers that .cpp files include to know about build changes). - remove the loki_add_library monstrosity; it breaks target names and makes compiling less efficient because the author couldn't figure out how to link things together. - make loki_add_executable take the output filename, and set the output path to bin/ and install to bin because *every single usage* of loki_add_executable was immediately followed by setting the output filename and setting the output path to bin/ and installing to bin. - move a bunch of crap that is only used in one particular src/whatever/CMakeLists.txt into that particular CMakeLists.txt instead of the top level CMakeLists.txt (or src/CMakeLists.txt). - Remove a bunch of redundant dependencies; most of them look like they were just copy-and-pasted in, and many more aren't needed (since they are implied by the PUBLIC linking of other dependencies). - Removed `die` since it just does a FATAL_ERROR, but adds color (which is useless since CMake already makes FATAL_ERRORs perfectly visible). - Change the way LOKI_DAEMON_AND_WALLET_ONLY works to just change the make targets to daemon and simplewallet rather than changing the build process (this should make it faster, too, since there are various other things that will be excluded).
2020-03-03 04:57:08 +01:00
# Interface target for ICU libs (if needed)
add_library(icu INTERFACE)
moved all stuff to github
2014-03-03 23:07:58 +01:00
if(MINGW)
Incorporate Window debug build targets
2018-05-31 11:40:00 +02:00
set(CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} -Wa,-mbig-obj")
cmake modernization The archaic (i.e. decade old) cmake usage here really got in the way of trying to properly use newer libraries (like lokimq), so this undertakes overhauling it considerably to make it much more sane (and significantly reduce the size). I left more of the architecture-specific bits in the top-level CMakeLists.txt intact; most of the efforts here are about properly loading dependencies, specifying dependencies and avoiding a whole pile of cmake antipatterns. This bumps the required cmake version to 3.5, which is what xenial comes with. - extensive use of interface libraries to include libraries, definitions, and include paths - use Boost::whatever instead of ${Boost_WHATEVER_LIBRARY}. The interface targets are (again) much better as they also give you any needed include or linking flags without needing to worry about them. - don't list header files when building things. This has *never* been correct cmake usage (cmake has always known how to wallet_rpc_headers the headers that .cpp files include to know about build changes). - remove the loki_add_library monstrosity; it breaks target names and makes compiling less efficient because the author couldn't figure out how to link things together. - make loki_add_executable take the output filename, and set the output path to bin/ and install to bin because *every single usage* of loki_add_executable was immediately followed by setting the output filename and setting the output path to bin/ and installing to bin. - move a bunch of crap that is only used in one particular src/whatever/CMakeLists.txt into that particular CMakeLists.txt instead of the top level CMakeLists.txt (or src/CMakeLists.txt). - Remove a bunch of redundant dependencies; most of them look like they were just copy-and-pasted in, and many more aren't needed (since they are implied by the PUBLIC linking of other dependencies). - Removed `die` since it just does a FATAL_ERROR, but adds color (which is useless since CMake already makes FATAL_ERRORs perfectly visible). - Change the way LOKI_DAEMON_AND_WALLET_ONLY works to just change the make targets to daemon and simplewallet rather than changing the build process (this should make it faster, too, since there are various other things that will be excluded).
2020-03-03 04:57:08 +01:00
target_link_libraries(extra INTERFACE mswsock ws2_32 iphlpapi crypt32 bcrypt)
Remove dead DEPENDS code We dropped the contrib/depends build system quite a while ago (because it was nasty), but there are still various DEPENDS checks scattered through cmake that are just dead code now. This removes them.
2021-05-12 15:46:57 +02:00
link_dep_libs(icu INTERFACE "" icuio icuin icuuc icudt icutu iconv)
backtrace_symbols() requires -lexecinfo on dragonflybsd freebsd also needs -lexecinfo for backtrace()
2017-02-08 23:50:14 +01:00
elseif(FREEBSD)
cmake modernization The archaic (i.e. decade old) cmake usage here really got in the way of trying to properly use newer libraries (like lokimq), so this undertakes overhauling it considerably to make it much more sane (and significantly reduce the size). I left more of the architecture-specific bits in the top-level CMakeLists.txt intact; most of the efforts here are about properly loading dependencies, specifying dependencies and avoiding a whole pile of cmake antipatterns. This bumps the required cmake version to 3.5, which is what xenial comes with. - extensive use of interface libraries to include libraries, definitions, and include paths - use Boost::whatever instead of ${Boost_WHATEVER_LIBRARY}. The interface targets are (again) much better as they also give you any needed include or linking flags without needing to worry about them. - don't list header files when building things. This has *never* been correct cmake usage (cmake has always known how to wallet_rpc_headers the headers that .cpp files include to know about build changes). - remove the loki_add_library monstrosity; it breaks target names and makes compiling less efficient because the author couldn't figure out how to link things together. - make loki_add_executable take the output filename, and set the output path to bin/ and install to bin because *every single usage* of loki_add_executable was immediately followed by setting the output filename and setting the output path to bin/ and installing to bin. - move a bunch of crap that is only used in one particular src/whatever/CMakeLists.txt into that particular CMakeLists.txt instead of the top level CMakeLists.txt (or src/CMakeLists.txt). - Remove a bunch of redundant dependencies; most of them look like they were just copy-and-pasted in, and many more aren't needed (since they are implied by the PUBLIC linking of other dependencies). - Removed `die` since it just does a FATAL_ERROR, but adds color (which is useless since CMake already makes FATAL_ERRORs perfectly visible). - Change the way LOKI_DAEMON_AND_WALLET_ONLY works to just change the make targets to daemon and simplewallet rather than changing the build process (this should make it faster, too, since there are various other things that will be excluded).
2020-03-03 04:57:08 +01:00
target_link_libraries(extra INTERFACE execinfo)
Preliminary support for DragonFly BSD - It builds but no further testing has been done.
2016-12-15 11:27:53 +01:00
elseif(DRAGONFLY)
find_library(COMPAT compat)
cmake modernization The archaic (i.e. decade old) cmake usage here really got in the way of trying to properly use newer libraries (like lokimq), so this undertakes overhauling it considerably to make it much more sane (and significantly reduce the size). I left more of the architecture-specific bits in the top-level CMakeLists.txt intact; most of the efforts here are about properly loading dependencies, specifying dependencies and avoiding a whole pile of cmake antipatterns. This bumps the required cmake version to 3.5, which is what xenial comes with. - extensive use of interface libraries to include libraries, definitions, and include paths - use Boost::whatever instead of ${Boost_WHATEVER_LIBRARY}. The interface targets are (again) much better as they also give you any needed include or linking flags without needing to worry about them. - don't list header files when building things. This has *never* been correct cmake usage (cmake has always known how to wallet_rpc_headers the headers that .cpp files include to know about build changes). - remove the loki_add_library monstrosity; it breaks target names and makes compiling less efficient because the author couldn't figure out how to link things together. - make loki_add_executable take the output filename, and set the output path to bin/ and install to bin because *every single usage* of loki_add_executable was immediately followed by setting the output filename and setting the output path to bin/ and installing to bin. - move a bunch of crap that is only used in one particular src/whatever/CMakeLists.txt into that particular CMakeLists.txt instead of the top level CMakeLists.txt (or src/CMakeLists.txt). - Remove a bunch of redundant dependencies; most of them look like they were just copy-and-pasted in, and many more aren't needed (since they are implied by the PUBLIC linking of other dependencies). - Removed `die` since it just does a FATAL_ERROR, but adds color (which is useless since CMake already makes FATAL_ERRORs perfectly visible). - Change the way LOKI_DAEMON_AND_WALLET_ONLY works to just change the make targets to daemon and simplewallet rather than changing the build process (this should make it faster, too, since there are various other things that will be excluded).
2020-03-03 04:57:08 +01:00
target_link_libraries(extra INTERFACE ${COMPAT})
make this build on SunOS/Solaris
2017-10-10 15:05:28 +02:00
elseif(CMAKE_SYSTEM_NAME MATCHES "(SunOS|Solaris)")
cmake modernization The archaic (i.e. decade old) cmake usage here really got in the way of trying to properly use newer libraries (like lokimq), so this undertakes overhauling it considerably to make it much more sane (and significantly reduce the size). I left more of the architecture-specific bits in the top-level CMakeLists.txt intact; most of the efforts here are about properly loading dependencies, specifying dependencies and avoiding a whole pile of cmake antipatterns. This bumps the required cmake version to 3.5, which is what xenial comes with. - extensive use of interface libraries to include libraries, definitions, and include paths - use Boost::whatever instead of ${Boost_WHATEVER_LIBRARY}. The interface targets are (again) much better as they also give you any needed include or linking flags without needing to worry about them. - don't list header files when building things. This has *never* been correct cmake usage (cmake has always known how to wallet_rpc_headers the headers that .cpp files include to know about build changes). - remove the loki_add_library monstrosity; it breaks target names and makes compiling less efficient because the author couldn't figure out how to link things together. - make loki_add_executable take the output filename, and set the output path to bin/ and install to bin because *every single usage* of loki_add_executable was immediately followed by setting the output filename and setting the output path to bin/ and installing to bin. - move a bunch of crap that is only used in one particular src/whatever/CMakeLists.txt into that particular CMakeLists.txt instead of the top level CMakeLists.txt (or src/CMakeLists.txt). - Remove a bunch of redundant dependencies; most of them look like they were just copy-and-pasted in, and many more aren't needed (since they are implied by the PUBLIC linking of other dependencies). - Removed `die` since it just does a FATAL_ERROR, but adds color (which is useless since CMake already makes FATAL_ERRORs perfectly visible). - Change the way LOKI_DAEMON_AND_WALLET_ONLY works to just change the make targets to daemon and simplewallet rather than changing the build process (this should make it faster, too, since there are various other things that will be excluded).
2020-03-03 04:57:08 +01:00
target_link_libraries(extra INTERFACE socket nsl resolv)
elseif(APPLE)
Remove dead DEPENDS code We dropped the contrib/depends build system quite a while ago (because it was nasty), but there are still various DEPENDS checks scattered through cmake that are just dead code now. This removes them.
2021-05-12 15:46:57 +02:00
find_library(COREFOUNDATION CoreFoundation)
find_library(IOKIT IOKit)
target_link_libraries(extra INTERFACE ${IOKIT} ${COREFOUNDATION})
elseif(NOT (MSVC OR APPLE OR OPENBSD OR ANDROID))
fix for mingw not playing nicely with libunbound configure, fix for correctly finding static libs on various operating systems
2014-10-06 22:29:07 +02:00
find_library(RT rt)
cmake modernization The archaic (i.e. decade old) cmake usage here really got in the way of trying to properly use newer libraries (like lokimq), so this undertakes overhauling it considerably to make it much more sane (and significantly reduce the size). I left more of the architecture-specific bits in the top-level CMakeLists.txt intact; most of the efforts here are about properly loading dependencies, specifying dependencies and avoiding a whole pile of cmake antipatterns. This bumps the required cmake version to 3.5, which is what xenial comes with. - extensive use of interface libraries to include libraries, definitions, and include paths - use Boost::whatever instead of ${Boost_WHATEVER_LIBRARY}. The interface targets are (again) much better as they also give you any needed include or linking flags without needing to worry about them. - don't list header files when building things. This has *never* been correct cmake usage (cmake has always known how to wallet_rpc_headers the headers that .cpp files include to know about build changes). - remove the loki_add_library monstrosity; it breaks target names and makes compiling less efficient because the author couldn't figure out how to link things together. - make loki_add_executable take the output filename, and set the output path to bin/ and install to bin because *every single usage* of loki_add_executable was immediately followed by setting the output filename and setting the output path to bin/ and installing to bin. - move a bunch of crap that is only used in one particular src/whatever/CMakeLists.txt into that particular CMakeLists.txt instead of the top level CMakeLists.txt (or src/CMakeLists.txt). - Remove a bunch of redundant dependencies; most of them look like they were just copy-and-pasted in, and many more aren't needed (since they are implied by the PUBLIC linking of other dependencies). - Removed `die` since it just does a FATAL_ERROR, but adds color (which is useless since CMake already makes FATAL_ERRORs perfectly visible). - Change the way LOKI_DAEMON_AND_WALLET_ONLY works to just change the make targets to daemon and simplewallet rather than changing the build process (this should make it faster, too, since there are various other things that will be excluded).
2020-03-03 04:57:08 +01:00
target_link_libraries(extra INTERFACE rt)
moved all stuff to github
2014-03-03 23:07:58 +01:00
endif()
cmake modernization The archaic (i.e. decade old) cmake usage here really got in the way of trying to properly use newer libraries (like lokimq), so this undertakes overhauling it considerably to make it much more sane (and significantly reduce the size). I left more of the architecture-specific bits in the top-level CMakeLists.txt intact; most of the efforts here are about properly loading dependencies, specifying dependencies and avoiding a whole pile of cmake antipatterns. This bumps the required cmake version to 3.5, which is what xenial comes with. - extensive use of interface libraries to include libraries, definitions, and include paths - use Boost::whatever instead of ${Boost_WHATEVER_LIBRARY}. The interface targets are (again) much better as they also give you any needed include or linking flags without needing to worry about them. - don't list header files when building things. This has *never* been correct cmake usage (cmake has always known how to wallet_rpc_headers the headers that .cpp files include to know about build changes). - remove the loki_add_library monstrosity; it breaks target names and makes compiling less efficient because the author couldn't figure out how to link things together. - make loki_add_executable take the output filename, and set the output path to bin/ and install to bin because *every single usage* of loki_add_executable was immediately followed by setting the output filename and setting the output path to bin/ and installing to bin. - move a bunch of crap that is only used in one particular src/whatever/CMakeLists.txt into that particular CMakeLists.txt instead of the top level CMakeLists.txt (or src/CMakeLists.txt). - Remove a bunch of redundant dependencies; most of them look like they were just copy-and-pasted in, and many more aren't needed (since they are implied by the PUBLIC linking of other dependencies). - Removed `die` since it just does a FATAL_ERROR, but adds color (which is useless since CMake already makes FATAL_ERRORs perfectly visible). - Change the way LOKI_DAEMON_AND_WALLET_ONLY works to just change the make targets to daemon and simplewallet rather than changing the build process (this should make it faster, too, since there are various other things that will be excluded).
2020-03-03 04:57:08 +01:00
if (WIN32)
target_link_libraries(extra INTERFACE setupapi)
Replace USB-CCID (smartcard) by USB-HID Remove PCSC dependencies which is a bit hard (not user friendly) to install on linux and Mac Split Ledger logic and device IO
2018-08-01 09:24:53 +02:00
endif()
Drop integration test code. This code is bitrotting, doesn't compile, and isn't being maintained anymore. The integration test suite was an interesting idea, in early Loki days, but is no longer being maintained and is quite cumbersome to run (for instance, it is not possible to run it via CI because it depends on xterm to actually run). The code to actually run it (in doy-lee's loki-integration-testing repository) is also a large burden of "janky" code that isn't worth maintaining. Remove this from the code; if someone wants to pick it back up in the future reverting this commit shouldn't be too difficult (though I'd suggest that a much better approach to integration testing would be to run different daemons/wallets via rpc commands, as the network-tests do, rather than trying to feed stdin and parse stdout from running individual oxends/wallets).
2021-08-19 21:38:06 +02:00
option(USE_READLINE "Build with GNU readline support." ON)
if(USE_READLINE AND BUILD_STATIC_DEPS)
# readline target already set up
elseif(USE_READLINE)
find_package(Readline)
if(READLINE_FOUND AND GNU_READLINE_FOUND)
add_library(readline INTERFACE)
target_link_libraries(readline INTERFACE ${GNU_READLINE_LIBRARY})
target_include_directories(readline INTERFACE ${Readline_INCLUDE_DIR})
target_compile_definitions(readline INTERFACE HAVE_READLINE)
message(STATUS "Found readline library at: ${GNU_READLINE_LIBRARY}")
else()
message(STATUS "Could not find GNU readline library so building without readline support")
Add readline support to cli This PR adds readline support to the daemon and monero-wallet-cli. Only GNU readline is supported (e.g. not libedit) and there are cmake checks to ensure this. There is a cmake variable, Readline_ROOT_DIR that can specify a directory to find readline, otherwise some default paths are searched. There is also a cmake option, USE_READLINE, that defaults to ON. If set to ON, if readline is not found, the build continues but without readline support. One negative side effect of using readline is that the color prompt in the wallet-cli now has no color and just uses terminal default. I know how to fix this but it's quite a big change so will tackle another time.
2017-05-30 00:39:49 +02:00
endif()
endif()
cmake set atomic for android
2017-01-21 05:15:00 +01:00
if(ANDROID)
cmake modernization The archaic (i.e. decade old) cmake usage here really got in the way of trying to properly use newer libraries (like lokimq), so this undertakes overhauling it considerably to make it much more sane (and significantly reduce the size). I left more of the architecture-specific bits in the top-level CMakeLists.txt intact; most of the efforts here are about properly loading dependencies, specifying dependencies and avoiding a whole pile of cmake antipatterns. This bumps the required cmake version to 3.5, which is what xenial comes with. - extensive use of interface libraries to include libraries, definitions, and include paths - use Boost::whatever instead of ${Boost_WHATEVER_LIBRARY}. The interface targets are (again) much better as they also give you any needed include or linking flags without needing to worry about them. - don't list header files when building things. This has *never* been correct cmake usage (cmake has always known how to wallet_rpc_headers the headers that .cpp files include to know about build changes). - remove the loki_add_library monstrosity; it breaks target names and makes compiling less efficient because the author couldn't figure out how to link things together. - make loki_add_executable take the output filename, and set the output path to bin/ and install to bin because *every single usage* of loki_add_executable was immediately followed by setting the output filename and setting the output path to bin/ and installing to bin. - move a bunch of crap that is only used in one particular src/whatever/CMakeLists.txt into that particular CMakeLists.txt instead of the top level CMakeLists.txt (or src/CMakeLists.txt). - Remove a bunch of redundant dependencies; most of them look like they were just copy-and-pasted in, and many more aren't needed (since they are implied by the PUBLIC linking of other dependencies). - Removed `die` since it just does a FATAL_ERROR, but adds color (which is useless since CMake already makes FATAL_ERRORs perfectly visible). - Change the way LOKI_DAEMON_AND_WALLET_ONLY works to just change the make targets to daemon and simplewallet rather than changing the build process (this should make it faster, too, since there are various other things that will be excluded).
2020-03-03 04:57:08 +01:00
target_compile_options(extra INTERFACE "-Wno-error=user-defined-warnings")
cmake set atomic for android
2017-01-21 05:15:00 +01:00
endif()
Do not require libatomic on FreeBSD f3e09f36 hooked a dependency on libatomic on 32 bit machines if Clang is used because compilation failed with: `std::__atomic_base<unsigned long long>::load(std::memory_order) const': /usr/bin/../lib/gcc/i686-pc-linux-gnu/6.1.1/../../../../include/c++/6.1.1/bits/atomic_base.h:396: undefined reference to `__atomic_load_8' But that does not happen on FreeBSD. The problem is likely that on Linux Clang tries to use GCC-provided C++11 library. Further, __atomic_load_8() (for 8-byte integers) is not readily available on 32 bit machines. From https://gcc.gnu.org/wiki/Atomic/GCCMM: "When lock free instructions are not available (either through hardware or OS support) atomic operations are left as function calls to be resolved by a library."
2017-11-11 14:29:42 +01:00
if(CMAKE_C_COMPILER_ID STREQUAL "Clang" AND ARCH_WIDTH EQUAL "32" AND NOT IOS AND NOT FREEBSD)
cmake: libatomic only needed for 32-bit Clang builds
2016-07-29 05:26:51 +02:00
find_library(ATOMIC atomic)
CMakeLists.txt: only use libatomic when found
2018-10-20 23:36:51 +02:00
if (ATOMIC_FOUND)
cmake modernization The archaic (i.e. decade old) cmake usage here really got in the way of trying to properly use newer libraries (like lokimq), so this undertakes overhauling it considerably to make it much more sane (and significantly reduce the size). I left more of the architecture-specific bits in the top-level CMakeLists.txt intact; most of the efforts here are about properly loading dependencies, specifying dependencies and avoiding a whole pile of cmake antipatterns. This bumps the required cmake version to 3.5, which is what xenial comes with. - extensive use of interface libraries to include libraries, definitions, and include paths - use Boost::whatever instead of ${Boost_WHATEVER_LIBRARY}. The interface targets are (again) much better as they also give you any needed include or linking flags without needing to worry about them. - don't list header files when building things. This has *never* been correct cmake usage (cmake has always known how to wallet_rpc_headers the headers that .cpp files include to know about build changes). - remove the loki_add_library monstrosity; it breaks target names and makes compiling less efficient because the author couldn't figure out how to link things together. - make loki_add_executable take the output filename, and set the output path to bin/ and install to bin because *every single usage* of loki_add_executable was immediately followed by setting the output filename and setting the output path to bin/ and installing to bin. - move a bunch of crap that is only used in one particular src/whatever/CMakeLists.txt into that particular CMakeLists.txt instead of the top level CMakeLists.txt (or src/CMakeLists.txt). - Remove a bunch of redundant dependencies; most of them look like they were just copy-and-pasted in, and many more aren't needed (since they are implied by the PUBLIC linking of other dependencies). - Removed `die` since it just does a FATAL_ERROR, but adds color (which is useless since CMake already makes FATAL_ERRORs perfectly visible). - Change the way LOKI_DAEMON_AND_WALLET_ONLY works to just change the make targets to daemon and simplewallet rather than changing the build process (this should make it faster, too, since there are various other things that will be excluded).
2020-03-03 04:57:08 +01:00
target_link_libraries(extra INTERFACE ${ATOMIC})
CMakeLists.txt: only use libatomic when found
2018-10-20 23:36:51 +02:00
endif()
Print stack trace on exceptions if libunwind is found. Useful for debugging logs.
2016-03-28 20:00:18 +02:00
endif()
cmake modernization The archaic (i.e. decade old) cmake usage here really got in the way of trying to properly use newer libraries (like lokimq), so this undertakes overhauling it considerably to make it much more sane (and significantly reduce the size). I left more of the architecture-specific bits in the top-level CMakeLists.txt intact; most of the efforts here are about properly loading dependencies, specifying dependencies and avoiding a whole pile of cmake antipatterns. This bumps the required cmake version to 3.5, which is what xenial comes with. - extensive use of interface libraries to include libraries, definitions, and include paths - use Boost::whatever instead of ${Boost_WHATEVER_LIBRARY}. The interface targets are (again) much better as they also give you any needed include or linking flags without needing to worry about them. - don't list header files when building things. This has *never* been correct cmake usage (cmake has always known how to wallet_rpc_headers the headers that .cpp files include to know about build changes). - remove the loki_add_library monstrosity; it breaks target names and makes compiling less efficient because the author couldn't figure out how to link things together. - make loki_add_executable take the output filename, and set the output path to bin/ and install to bin because *every single usage* of loki_add_executable was immediately followed by setting the output filename and setting the output path to bin/ and installing to bin. - move a bunch of crap that is only used in one particular src/whatever/CMakeLists.txt into that particular CMakeLists.txt instead of the top level CMakeLists.txt (or src/CMakeLists.txt). - Remove a bunch of redundant dependencies; most of them look like they were just copy-and-pasted in, and many more aren't needed (since they are implied by the PUBLIC linking of other dependencies). - Removed `die` since it just does a FATAL_ERROR, but adds color (which is useless since CMake already makes FATAL_ERRORs perfectly visible). - Change the way LOKI_DAEMON_AND_WALLET_ONLY works to just change the make targets to daemon and simplewallet rather than changing the build process (this should make it faster, too, since there are various other things that will be excluded).
2020-03-03 04:57:08 +01:00
cmake: build tests last Keep -Werror for src, contrib and do not pass it for tests/
2017-10-09 02:45:45 +02:00
add_subdirectory(contrib)
add_subdirectory(src)
cmake: BUILD_TESTS as option explicitly; added missed dependency
2016-05-27 09:42:31 +02:00
don't build tests unless we're building all-* or *-test builds
2014-12-01 19:00:22 +01:00
if(BUILD_TESTS)
dns_checks: new helper program to check on DNSSEC lookups
2019-04-16 14:46:01 +02:00
message(STATUS "Building tests")
don't build tests unless we're building all-* or *-test builds
2014-12-01 19:00:22 +01:00
add_subdirectory(tests)
dns_checks: new helper program to check on DNSSEC lookups
2019-04-16 14:46:01 +02:00
else()
message(STATUS "Not building tests")
endif()
remerged; commands JSON. logging upgrade. doxygen
2015-04-01 19:00:45 +02:00
if(BUILD_DOCUMENTATION)
set(DOC_GRAPHS "YES" CACHE STRING "Create dependency graphs (needs graphviz)")
set(DOC_FULLGRAPHS "NO" CACHE STRING "Create call/callee graphs (large)")
find_program(DOT_PATH dot)
if (DOT_PATH STREQUAL "DOT_PATH-NOTFOUND")
message("Doxygen: graphviz not found - graphs disabled")
set(DOC_GRAPHS "NO")
endif()
find_package(Doxygen)
if(DOXYGEN_FOUND)
configure_file("cmake/Doxyfile.in" "Doxyfile" @ONLY)
configure_file("cmake/Doxygen.extra.css.in" "Doxygen.extra.css" @ONLY)
add_custom_target(doc
${DOXYGEN_EXECUTABLE} ${CMAKE_CURRENT_BINARY_DIR}/Doxyfile
WORKING_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR}
COMMENT "Generating API documentation with Doxygen.." VERBATIM)
endif()
endif()
Add the iwyu target to get information on unused headers With this change, building the iwyu target will provide information on unused headers, which can be removed to reduce compilation times.
2019-09-05 17:27:12 +02:00
find_package(PythonInterp)
find_program(iwyu_tool_path NAMES iwyu_tool.py iwyu_tool)
if (iwyu_tool_path AND PYTHONINTERP_FOUND)
add_custom_target(iwyu
COMMAND "${PYTHON_EXECUTABLE}" "${iwyu_tool_path}" -p "${CMAKE_BINARY_DIR}" -- --no_fwd_decls
COMMENT "Running include-what-you-use tool"
VERBATIM
)
endif()
create_archive target: add targets to archive the bin/ build dir From a cmake build dir (`make` used for simple example; can also be something else, such as ninja): make create_tarxz make create_zip creater a loki-<OS>[-arch]-x.y.z[-dev]-<GITHASH>.tar.xz or .zip. make create_archive decides what to do based on the build type: creates a .zip for a windows build, a tar.xz for anything else. (We have been distributing the macOS binaries as a .zip but that seems unnecessary: I tested on our dev mac and a .tar.xz offers exactly the same UX as a .zip, but is noticeably smaller). From the top-level makefile there is also a new `make release-full-static-archive` that does a full static build (include all deps) and builds the archive.
2020-06-12 04:46:55 +02:00
Drone: don't build debug tools in static builds (#1281) The strip_binaries target was depending on all binaries, including debug utilities, and so the static builds would end up building and including them. This makes strip_binaries not strip or depend on EXCLUDE_FROM_ALL targets (or targets inside EXCLUDE_FROM_ALL directories) to prevent this. (A separate strip_binaries_all target is set up in case someone builds them and wants to strip them as well).
2020-09-26 04:59:34 +02:00
# Set up a `make strip_binaries` target that strips built binaries. This depends on all
# default-built binaries and strips them after build. (To also build and strip debug utilities
# there is also a `make strip_binaries_all` target.)
executable names changed
2021-01-04 04:19:42 +01:00
get_property(oxen_exec_tgts_all GLOBAL PROPERTY oxen_executable_targets)
set(oxen_exec_tgts "")
create_archive target: add targets to archive the bin/ build dir From a cmake build dir (`make` used for simple example; can also be something else, such as ninja): make create_tarxz make create_zip creater a loki-<OS>[-arch]-x.y.z[-dev]-<GITHASH>.tar.xz or .zip. make create_archive decides what to do based on the build type: creates a .zip for a windows build, a tar.xz for anything else. (We have been distributing the macOS binaries as a .zip but that seems unnecessary: I tested on our dev mac and a .tar.xz offers exactly the same UX as a .zip, but is noticeably smaller). From the top-level makefile there is also a new `make release-full-static-archive` that does a full static build (include all deps) and builds the archive.
2020-06-12 04:46:55 +02:00
set(strip_binaries "")
Drone: don't build debug tools in static builds (#1281) The strip_binaries target was depending on all binaries, including debug utilities, and so the static builds would end up building and including them. This makes strip_binaries not strip or depend on EXCLUDE_FROM_ALL targets (or targets inside EXCLUDE_FROM_ALL directories) to prevent this. (A separate strip_binaries_all target is set up in case someone builds them and wants to strip them as well).
2020-09-26 04:59:34 +02:00
set(strip_binaries_all "")
executable names changed
2021-01-04 04:19:42 +01:00
foreach(tgt ${oxen_exec_tgts_all})
Drone: don't build debug tools in static builds (#1281) The strip_binaries target was depending on all binaries, including debug utilities, and so the static builds would end up building and including them. This makes strip_binaries not strip or depend on EXCLUDE_FROM_ALL targets (or targets inside EXCLUDE_FROM_ALL directories) to prevent this. (A separate strip_binaries_all target is set up in case someone builds them and wants to strip them as well).
2020-09-26 04:59:34 +02:00
list(APPEND strip_binaries_all COMMAND ${CMAKE_STRIP} $<TARGET_FILE:${tgt}>)
# Look for a EXCLUDE_FROM_ALL property:
get_property(tgt_excl_all TARGET ${tgt} PROPERTY EXCLUDE_FROM_ALL)
# Also look for EXCLUDE_FROM_ALL on the target's source directory (this, unfortunately, is not
# inherited into the target itself, hence we check both).
get_property(tgt_dir TARGET ${tgt} PROPERTY SOURCE_DIR)
get_property(tgt_dir_excl_all DIRECTORY ${tgt_dir} PROPERTY EXCLUDE_FROM_ALL)
if (NOT tgt_excl_all AND NOT tgt_dir_excl_all)
executable names changed
2021-01-04 04:19:42 +01:00
list(APPEND oxen_exec_tgts ${tgt})
Drone: don't build debug tools in static builds (#1281) The strip_binaries target was depending on all binaries, including debug utilities, and so the static builds would end up building and including them. This makes strip_binaries not strip or depend on EXCLUDE_FROM_ALL targets (or targets inside EXCLUDE_FROM_ALL directories) to prevent this. (A separate strip_binaries_all target is set up in case someone builds them and wants to strip them as well).
2020-09-26 04:59:34 +02:00
list(APPEND strip_binaries COMMAND ${CMAKE_STRIP} $<TARGET_FILE:${tgt}>)
endif()
create_archive target: add targets to archive the bin/ build dir From a cmake build dir (`make` used for simple example; can also be something else, such as ninja): make create_tarxz make create_zip creater a loki-<OS>[-arch]-x.y.z[-dev]-<GITHASH>.tar.xz or .zip. make create_archive decides what to do based on the build type: creates a .zip for a windows build, a tar.xz for anything else. (We have been distributing the macOS binaries as a .zip but that seems unnecessary: I tested on our dev mac and a .tar.xz offers exactly the same UX as a .zip, but is noticeably smaller). From the top-level makefile there is also a new `make release-full-static-archive` that does a full static build (include all deps) and builds the archive.
2020-06-12 04:46:55 +02:00
endforeach()
executable names changed
2021-01-04 04:19:42 +01:00
add_custom_target(strip_binaries ${strip_binaries} DEPENDS ${oxen_exec_tgts})
add_custom_target(strip_binaries_all ${strip_binaries_all} DEPENDS ${oxen_exec_tgts_all})
create_archive target: add targets to archive the bin/ build dir From a cmake build dir (`make` used for simple example; can also be something else, such as ninja): make create_tarxz make create_zip creater a loki-<OS>[-arch]-x.y.z[-dev]-<GITHASH>.tar.xz or .zip. make create_archive decides what to do based on the build type: creates a .zip for a windows build, a tar.xz for anything else. (We have been distributing the macOS binaries as a .zip but that seems unnecessary: I tested on our dev mac and a .tar.xz offers exactly the same UX as a .zip, but is noticeably smaller). From the top-level makefile there is also a new `make release-full-static-archive` that does a full static build (include all deps) and builds the archive.
2020-06-12 04:46:55 +02:00
execute_process(COMMAND tar --version RESULT_VARIABLE tar_exit_code OUTPUT_VARIABLE tar_vers)
Small tweaks to build system - make tagged and stable branch builds not add git tag to the tar.xz filename - copy apt-get tweaks (to make it less noisy) from deb branches - do a faster shallow submodule clone - remove exclusion for removed is_hdd test - remove deb_builder (deb building drone code is in its own branches) - fix upload dir for a tagged build
2020-10-04 03:55:02 +02:00
set(git_tag "-unknown")
create_archive target: add targets to archive the bin/ build dir From a cmake build dir (`make` used for simple example; can also be something else, such as ninja): make create_tarxz make create_zip creater a loki-<OS>[-arch]-x.y.z[-dev]-<GITHASH>.tar.xz or .zip. make create_archive decides what to do based on the build type: creates a .zip for a windows build, a tar.xz for anything else. (We have been distributing the macOS binaries as a .zip but that seems unnecessary: I tested on our dev mac and a .tar.xz offers exactly the same UX as a .zip, but is noticeably smaller). From the top-level makefile there is also a new `make release-full-static-archive` that does a full static build (include all deps) and builds the archive.
2020-06-12 04:46:55 +02:00
if(GIT_FOUND)
Small tweaks to build system - make tagged and stable branch builds not add git tag to the tar.xz filename - copy apt-get tweaks (to make it less noisy) from deb branches - do a faster shallow submodule clone - remove exclusion for removed is_hdd test - remove deb_builder (deb building drone code is in its own branches) - fix upload dir for a tagged build
2020-10-04 03:55:02 +02:00
execute_process(COMMAND "${GIT_EXECUTABLE}" rev-parse --abbrev-ref HEAD RESULT_VARIABLE ret OUTPUT_VARIABLE branch OUTPUT_STRIP_TRAILING_WHITESPACE)
if(NOT ret AND branch STREQUAL "stable")
Make untagged stable builds add -<n>-<tag> suffix
2020-10-23 05:46:56 +02:00
# Get the tag description; for a tagged release this will be just the tag (v1.2.3); for
# something following a tag this will be something like "v1.2.3-2-abcdef" for something 2
# commits beyond the tag, currently at commit "abcdef".
execute_process(COMMAND "${GIT_EXECUTABLE}" describe --tags --abbrev=6 HEAD RESULT_VARIABLE ret OUTPUT_VARIABLE tag OUTPUT_STRIP_TRAILING_WHITESPACE)
if(NOT ret AND tag MATCHES "v[0-9]+\\.[0-9]+\\.[0-9]+(-.*)")
# We're building something following a tagged release, so append the post-version git tag info
set(git_tag "${CMAKE_MATCH_1}")
else()
set(git_tag "") # No tag appended if we're building a tagged stable branch release
endif()
Small tweaks to build system - make tagged and stable branch builds not add git tag to the tar.xz filename - copy apt-get tweaks (to make it less noisy) from deb branches - do a faster shallow submodule clone - remove exclusion for removed is_hdd test - remove deb_builder (deb building drone code is in its own branches) - fix upload dir for a tagged build
2020-10-04 03:55:02 +02:00
else()
execute_process(COMMAND "${GIT_EXECUTABLE}" rev-parse --short=9 HEAD RESULT_VARIABLE ret OUTPUT_VARIABLE commithash OUTPUT_STRIP_TRAILING_WHITESPACE)
if(NOT ret)
set(git_tag "-${commithash}")
endif()
create_archive target: add targets to archive the bin/ build dir From a cmake build dir (`make` used for simple example; can also be something else, such as ninja): make create_tarxz make create_zip creater a loki-<OS>[-arch]-x.y.z[-dev]-<GITHASH>.tar.xz or .zip. make create_archive decides what to do based on the build type: creates a .zip for a windows build, a tar.xz for anything else. (We have been distributing the macOS binaries as a .zip but that seems unnecessary: I tested on our dev mac and a .tar.xz offers exactly the same UX as a .zip, but is noticeably smaller). From the top-level makefile there is also a new `make release-full-static-archive` that does a full static build (include all deps) and builds the archive.
2020-06-12 04:46:55 +02:00
endif()
endif()
set(tar_os ${CMAKE_SYSTEM_NAME})
set(default_archive create_tarxz)
if(tar_os STREQUAL "Linux")
set(tar_os "linux-${CMAKE_SYSTEM_PROCESSOR}")
elseif(tar_os STREQUAL "Darwin")
set(tar_os "macos")
elseif(tar_os STREQUAL "Windows")
if(CMAKE_CROSSCOMPILING AND ARCH_TRIPLET MATCHES i686-.*mingw)
set(tar_os "win-x86")
elseif(CMAKE_CROSSCOMPILING AND ARCH_TRIPLET MATCHES x86_64-.*mingw)
set(tar_os "win-x64")
else()
set(tar_os "windows") # Don't know what arch
endif()
set(default_archive create_zip) # .tar.xz files are too scary for Windows users
endif()
executable names changed
2021-01-04 04:19:42 +01:00
set(tar_dir "oxen-${tar_os}-${PROJECT_VERSION}${OXEN_RELEASE_SUFFIX}${git_tag}")
create_archive target: add targets to archive the bin/ build dir From a cmake build dir (`make` used for simple example; can also be something else, such as ninja): make create_tarxz make create_zip creater a loki-<OS>[-arch]-x.y.z[-dev]-<GITHASH>.tar.xz or .zip. make create_archive decides what to do based on the build type: creates a .zip for a windows build, a tar.xz for anything else. (We have been distributing the macOS binaries as a .zip but that seems unnecessary: I tested on our dev mac and a .tar.xz offers exactly the same UX as a .zip, but is noticeably smaller). From the top-level makefile there is also a new `make release-full-static-archive` that does a full static build (include all deps) and builds the archive.
2020-06-12 04:46:55 +02:00
add_custom_target(create_tarxz
COMMAND ${CMAKE_COMMAND} -E rename bin "${tar_dir}"
COMMAND ${CMAKE_COMMAND} -E tar cvJ "${tar_dir}.tar.xz" -- "${tar_dir}"
COMMAND ${CMAKE_COMMAND} -E rename "${tar_dir}" bin
executable names changed
2021-01-04 04:19:42 +01:00
DEPENDS ${oxen_exec_tgts})
create_archive target: add targets to archive the bin/ build dir From a cmake build dir (`make` used for simple example; can also be something else, such as ninja): make create_tarxz make create_zip creater a loki-<OS>[-arch]-x.y.z[-dev]-<GITHASH>.tar.xz or .zip. make create_archive decides what to do based on the build type: creates a .zip for a windows build, a tar.xz for anything else. (We have been distributing the macOS binaries as a .zip but that seems unnecessary: I tested on our dev mac and a .tar.xz offers exactly the same UX as a .zip, but is noticeably smaller). From the top-level makefile there is also a new `make release-full-static-archive` that does a full static build (include all deps) and builds the archive.
2020-06-12 04:46:55 +02:00
add_custom_target(create_zip
COMMAND ${CMAKE_COMMAND} -E rename bin "${tar_dir}"
COMMAND ${CMAKE_COMMAND} -E tar cv "${tar_dir}.zip" --format=zip -- "${tar_dir}"
COMMAND ${CMAKE_COMMAND} -E rename "${tar_dir}" bin
executable names changed
2021-01-04 04:19:42 +01:00
DEPENDS ${oxen_exec_tgts})
create_archive target: add targets to archive the bin/ build dir From a cmake build dir (`make` used for simple example; can also be something else, such as ninja): make create_tarxz make create_zip creater a loki-<OS>[-arch]-x.y.z[-dev]-<GITHASH>.tar.xz or .zip. make create_archive decides what to do based on the build type: creates a .zip for a windows build, a tar.xz for anything else. (We have been distributing the macOS binaries as a .zip but that seems unnecessary: I tested on our dev mac and a .tar.xz offers exactly the same UX as a .zip, but is noticeably smaller). From the top-level makefile there is also a new `make release-full-static-archive` that does a full static build (include all deps) and builds the archive.
2020-06-12 04:46:55 +02:00
add_custom_target(create_archive DEPENDS ${default_archive})
wallet3 cli This introduces wallet3 cli to the codebase, build using python and wraps the c++ wallet3 core using pybind.
2022-10-23 22:08:52 +02:00
option to exclude making pybind on android and windows by default
2022-12-05 21:33:56 +01:00
if(BUILD_PYBIND)
add_subdirectory(pybind)
endif()