Openssl for windows x64 установка

(See the Donations section on this page to get your business here)

Источник

Information Security Squad

stay tune stay secure

Как установить OpenSSL на Windows Server 2019

OpenSSL – это библиотека с открытым исходным кодом, которая предоставляет криптографические протоколы, используемые для защиты приложений и передачи информации между системами.

Менеджмент OpenSSL находится под управлением добровольцев по всему миру.

В двух словах, инструментарий OpenSSL реализует протоколы Secure Sockets Layer (SSL v2 / v3) и Transport Layer Security (TLS v1) с полной криптографией.

Это руководство покажет вам, как установить OpenSSL на Windows Server 2019.

Установите OpenSSL на Windows Server 2019

Перейдите на страницу загрузок OpenSSL и скачайте последнюю версию OpenSSL, соответствующую архитектуре вашего процессора.

Для себя я скачаю 64-битную версию.

У вас есть возможность выбрать версию Light или полную версию.

Вы также можете использовать команду curl.exe для загрузки из командной строки.

Смотрите скриншот ниже.

Примите лицензионное соглашение и нажмите «Next».

Выберите папку назначения, в которую будет установлен OpenSSL.

Выберите каталог для ярлыка приложения.

Выберите дополнительные задачи для выполнения.

Нажмите «Install», чтобы начать установку OpenSSL на Windows Server 2019.

Подождите установку несколько минут для завершения.

Нажмите «Finish», чтобы завершить успешную установку.

Наконец добавьте C:\OpenSSL-Win64 в PATH среды Windows.

Для 32-разрядной системы замените OpenSSL-Win64 на OpenSSL-Win32.

Создание сертификатов с OpenSSL на Windows Server 2019

Теперь вы готовы использовать OpenSSL на Windows Server 2019 для создания сертификатов.

Начните с экспорта OPENSSL_CONF.

Для 32-разрядной системы замените OpenSSL-Win64 на OpenSSL-Win32. Давайте создадим тестовый SSL-сертификат для проверки нашей установки.

Источник

Защищенное TLS-соединение с использованием Boost.Asio и OpenSSL под Windows

Введение

Однажды мне потребовалось создать защищенный канал связи между своим сервером и своим приложением. Я помнил, что в документации к Boost Asio упоминалось, что он может работать с защищенными соединениями, используя OpenSSL. Я начал искать информацию по этой теме, но, увы, мало что нашел, тем более под Windows. Так что теперь, разобравшись с этим вопросом, я решил написать эту инструкцию, чтобы другим людям было проще разобраться.

Задача — нужно собрать под Windows сервер и клиент, используя Boost Asio и OpenSSL, чтобы клиент и сервер обменивались информацией по защищенному TLS-каналу. Для пример, я решил взять вот этот клиент и сервер с официального сайта Boost.

Для того чтобы решить эту задачу, нам нужно собрать OpenSSL, подготовить ключи и сертификаты, и собрать оба примера с использованием Boost Asio, OpenSSL.

Установка OpenSSL под Windows

Я взял OpenSSL из официального репозитория: github.com/openssl/openssl

Для сборки OpenSSL я использовал MS Visual Studio 2013, и я собирал статическую библиотеку.

Последовательность сборки следующая:
Сначала нужно сконфигурировать OpenSSL с помощью скрипта на Perl, под Win32. Ниже по тексту я буду считать, что OpenSSL у вас находится в C:\Work\OpenSSL. Вам следует зайти в этот каталог и вызвать скрипт конфигурации:

На этом нужно закрыть обычную командную строку, и запустить командную строку MS Visual Studio, в которой определены дополнительные пути к файлам и дополнительные переменные окружения. Вы можете найти командную строку MS Visual Studio в каталоге C:\Program Files (x86)\Microsoft Visual Studio 12.0\Common7\Tools\Shortcuts.

Из командной строки MS Visual Studio нужно перейти в каталог C:\Work\OpenSSL и запустить сборку с помощью nmake:

Это команда для сборки статической библиотеки, если вы хотите собрать динамическую библиотеку, то нужно запускать ntdll.mak.

После того, как сборка завершилась, нужно скопировать библиотеки и исходные файлы в новый каталог:

На этом сборка OpenSSL под Windows завершена.

Сборка клиента и сервера

Как я и сказал раньше, для примера я решил взять вот эти клиент и сервер из документации по Boost Asio. Однако при попытке сборки я столкнулся с некоторыми проблемами, и в результате мне пришлось модифицировать исходники.

Хотя это место и объявлено как extern «C», и с точки зрения С тут нет никаких ошибок, но Visual Studio не дает нам никакой возможности отключить эту ошибку. Мне пришлось вносить изменения и преобразовывать тип явно:

Исходный код сервера:

Исходный код клиента:

Создание ключей и сертификатов

На этом этапе клиент и сервер запускаются, теперь необходимо проверить их работу. Для этого нужно создать корневой сертификат и подписать им сертификат для сервера.

После сборки в каталоге C:\Work\OpenSSL\output\bin будет лежать openssl.exe, нужно воспользоваться им, чтобы сгенерировать ключи и сертификаты.

Для начала создаем приватный ключ для корневого сертификата:

Потом на основе этого ключа создаем корневой сертификат, действующий 20000 дней:

В интерактивном меню вас попросят ввести двухбуквенный код страны, провинцию, город, организацию, подразделение, Common Name и e-mail адрес. Нужно заполнить все поля на свое усмотрение.

Теперь нужно создать другой сертификат, подписанный корневым сертификатом.

Создаем еще один ключ:

Создаем запрос на подпись:

В интерактивном меню вам потребуется ответить на те же вопросы, что и при создании корневого сертификата. Нужно, чтобы введенный вами Common Name отличался от Common Name у корневого сертификата, это важно!

Теперь подписываем этот запрос корневым сертификатом:

На всякий случай можно проверить, что подписано все правильно:

Первая команда должна вернуть OK, потому что корневой сертификат — самоподписанный.

Вторая команда должна вернуть ОК, потому что user.crt подписан корневым сертификатом.

Последняя команда должна вернуть ошибку, потому что user.crt не является самоподписанным. Если последняя команда возвращает OK, значит что-то пошло не так. В моем случае для исправления нужно было всего лишь сделать Common Name у обоих сертификатов различающимся.

И напоследок, нам еще понадобится DH-параметры, которые нужны для Протокола Диффи — Хеллмана, нужно их сгенерировать. Генерация займет некоторое время:

На этом все, теперь достаточно прописать клиенту и серверу пути к этим файлам, и вы сможете установить между ними защищенное соединение.

Источник

How to install OpenSSL in windows 10?

9 Answers 9

I also wanted to create OPEN SSL for Windows 10. An easy way of getting it done without running into a risk of installing unknown software from 3rd party websites and risking entries of viruses, is by using the openssl.exe that comes inside your Git for Windows installation. In my case, I found the open SSL in the following location of Git for Windows Installation.

If you also want instructions on how to use OPENSSL to generate and use Certificates. Here is a write-up on my blog. The step by step instructions first explains how to use Microsoft Windows Default Tool and also OPEN SSL and explains the difference between them.

If you have chocolatey installed you can install openssl via a single command i.e.

In case you have Git installed,

You can install openssl using one single line if you have chocolatey installed

Either set the openssl present in Git as your default openssl and include that into your path in environmental variables (quick way)

I recently needed to document how to get a version of it installed, so I’ve copied my steps here, as the other answers were using different sources from what I recommend, which is Cygwin. I like Cygwin because it is well maintained and provides a wealth of other utilities for Windows. Cygwin also allows you to easily update the versions as needed when vulnerabilities are fixed. Please update your version of OpenSSL often!

Open a Windows Command prompt and check to see if you have OpenSSL installed by entering: openssl version

If you get an error message that the command is NOT recognized, then install OpenSSL by referring to Cygwin following the summary steps below:

Basically, download and run the Cygwin Windows Setup App to install and to update as needed the OpenSSL application:

Источник

Openssl for windows x64 установка

This document describes installation on all supported operating systems (the Unix/Linux family, including macOS), OpenVMS, and Windows).

To install OpenSSL, you will need:

For additional platform specific requirements, solutions to specific issues and other details, please read one of these:

Throughout this document, we use the following conventions.

Any line starting with a dollar sign is a command line.

The dollar sign indicates the shell prompt and is not to be entered as part of the command.

Several words in curly braces separated by pipe characters indicate a mandatory choice, to be replaced with one of the given words. For example, the line

represents one of the following three commands

One or several words in square brackets separated by pipe characters denote an optional choice. It is similar to the mandatory choice, but it can also be omitted entirely.

represents one of the four commands

Mandatory arguments are enclosed in double curly braces. A simple example would be

which is to be understood to use the command type on some file name determined by the user.

Optional Arguments are enclosed in double square brackets.

Quick Installation Guide

If you just want to get OpenSSL installed without bothering too much about the details, here is the short version of how to build and install OpenSSL. If any of the following steps fails, please consult the Installation in Detail section below.

Use the following commands to configure, build and test OpenSSL. The testing is optional, but recommended if you intend to install OpenSSL for production use.

Unix / Linux / macOS

Use the following commands to build OpenSSL:

If you are using Visual Studio, open a Developer Command Prompt and issue the following commands to build OpenSSL.

As mentioned in the Choices section, you need to pick one of the four Configure targets in the first command.

Most likely you will be using the VC-WIN64A target for 64bit Windows binaries (AMD64) or VC-WIN32 for 32bit Windows binaries (X86). The other two options are VC-WIN64I (Intel IA64, Itanium) and VC-CE (Windows CE) are rather uncommon nowadays.

The following commands will install OpenSSL to a default system location.

Danger Zone: even if you are impatient, please read the following two paragraphs carefully before you install OpenSSL.

For security reasons the default system location is by default not writable for unprivileged users. So for the final installation step administrative privileges are required. The default system location and the procedure to obtain administrative privileges depends on the operating system. It is recommended to compile and test OpenSSL with normal user privileges and use administrative privileges only for the final installation step.

On some platforms OpenSSL is preinstalled as part of the Operating System. In this case it is highly recommended not to overwrite the system versions, because other applications or libraries might depend on it. To avoid breaking other applications, install your copy of OpenSSL to a different location which is not in the global search path for system libraries.

Finally, if you plan on using the FIPS module, you need to read the Post-installation Notes further down.

Unix / Linux / macOS

Depending on your distribution, you need to run the following command as root user or prepend sudo to the command:

By default, OpenSSL will be installed to

More precisely, the files will be installed into the subdirectories

depending on the file type, as it is custom on Unix-like operating systems.

Use the following command to install OpenSSL.

By default, OpenSSL will be installed to

If you are using Visual Studio, open the Developer Command Prompt elevated and issue the following command.

The easiest way to elevate the Command Prompt is to press and hold down the both the and key while clicking the menu item in the task menu.

The default installation location is

for native binaries, or

for 32bit binaries on 64bit Windows (WOW64).

Installing to a different location

To install OpenSSL to a different location (for example into your home directory for testing purposes) run Configure as shown in the following examples.

Note: if you do add options to the configuration command, please make sure you’ve read more than just this Quick Start, such as relevant NOTES-* files, the options outline below, as configuration options may change the outcome in otherwise unexpected ways.

Build the OpenSSL libraries to support the API for the specified version. If no-deprecated is also given, don’t build with support for deprecated APIs in or below the specified version number. For example, addding

Cross Compile Prefix

Build OpenSSL with debugging symbols and zero optimization level.

Build OpenSSL without debugging symbols. This is the default.

Directory for OpenSSL configuration files, and also the default certificate and key store. Defaults are:

The top of the installation directory tree. Defaults are:

This is a developer flag that switches on various compiler options recommended for OpenSSL development. It only works when using gcc or clang as the compiler. If you are developing a patch for OpenSSL then it is recommended that you use this option where possible.

The directory for the location of the zlib include file. This option is only necessary if zlib is used and the include file is not already on the system include path.

On Unix: this is the directory containing the zlib library. If not provided the system library path will be used.

On Windows: this is the filename of the zlib library (with or without a path). This flag must be provided if the zlib-dynamic option is not also used. If zlib-dynamic is used then this flag is optional and defaults to ZLIB1 if not provided.

Seeding the Random Generator

A comma separated list of seeding methods which will be tried by OpenSSL in order to obtain random input (a.k.a «entropy») for seeding its cryptographically secure random number generator (CSPRNG). The current seeding methods are:

Use a trusted operating system entropy source. This is the default method if such an entropy source exists.

Use the getrandom(2) or equivalent system call.

Use the first device from the DEVRANDOM list which can be opened to read random bytes. The DEVRANDOM preprocessor constant expands to

on most unix-ish operating systems.

Check for an entropy generating daemon. This source is ignored by the FIPS provider.

Use the RDSEED or RDRAND command if provided by the CPU.

Use librandom (not implemented yet). This source is ignored by the FIPS provider.

Disable automatic seeding. This is the default on some operating systems where no suitable entropy source exists, or no support for it is implemented yet. This option is ignored by the FIPS provider.

For more information, see the section Notes on random number generation at the end of this document.

Setting the FIPS HMAC key

As part of its self-test validation, the FIPS module must verify itself by performing a SHA-256 HMAC computation on itself. The default key is the SHA256 value of «the holy handgrenade of antioch» and is sufficient for meeting the FIPS requirements.

To change the key to a different value, use this flag. The value should be a hex string no more than 64 characters.

Enable and Disable Features

Whether a feature is enabled or disabled by default, depends on the feature. In the following list, always the non-default variant is documented: if feature xxxx is disabled by default then enable-xxxx is documented and if feature xxxx is enabled by default then no-xxxx is documented.

Don’t build the AFALG engine.

This option will be forced on a platform that does not support AFALG.

Build with Kernel TLS support.

This option will enable the use of the Kernel TLS data-path, which can improve performance and allow for the use of sendfile and splice system calls on TLS sockets. The Kernel may use TLS accelerators if any are available on the system. This option will be forced off on systems that do not support the Kernel TLS data-path.

Build with the Address sanitiser.

This is a developer option only. It may not work on all platforms and should never be used in production environments. It will only work when used with gcc or clang and should be used in conjunction with the no-shared option.

Do not build support for Automated Cryptographic Validation Protocol (ACVP) tests.

This is required for FIPS validation purposes. Certain ACVP tests require access to algorithm internals that are not normally accessible. Additional information related to ACVP can be found at https://github.com/usnistgov/ACVP.

Do not use assembler code.

This should be viewed as debugging/troubleshooting option rather than for production use. On some platforms a small amount of assembler code may still be used even with this option.

Do not build support for async operations.

Don’t automatically load all supported ciphers and digests.

Typically OpenSSL will make available all of its supported ciphers and digests. For a statically linked application this may be undesirable if small executable size is an objective. This only affects libcrypto. Ciphers and digests will have to be loaded manually using EVP_add_cipher() and EVP_add_digest() if this option is used. This option will force a non-shared build.

Don’t automatically load all libcrypto/libssl error strings.

Typically OpenSSL will automatically load human readable error strings. For a statically linked application this may be undesirable if small executable size is an objective.

Don’t automatically load the default openssl.cnf file.

Typically OpenSSL will automatically load a system config file which configures default SSL options.

While testing, generate C++ buildtest files that simply check that the public OpenSSL header files are usable standalone with C++.

Build only some minimal set of features. This is a developer option used internally for CI build tests of the project.

Never cache algorithms when they are fetched from a provider. Normally, a provider indicates if the algorithms it supplies can be cached or not. Using this option will reduce run-time memory usage but it also introduces a significant performance penalty. This option is primarily designed to help with detecting incorrect reference counting.

Don’t build the CAPI engine.

This option will be forced if on a platform that does not support CAPI.

Don’t build support for Certificate Management Protocol (CMP) and Certificate Request Message Format (CRMF).

Don’t build support for Cryptographic Message Syntax (CMS).

Don’t build support for SSL/TLS compression.

If this option is enabled (the default), then compression will only work if the zlib or zlib-dynamic options are also chosen.

This now only enables the failed-malloc feature.

This is a no-op; the project uses the compiler’s address/leak sanitizer instead.

Don’t build support for Certificate Transparency (CT).

Don’t build support for datagram based BIOs.

Selecting this option will also force the disabling of DTLS.

Don’t build support for loading Dynamic Shared Objects (DSO)

Build the /dev/crypto engine.

Don’t build the dynamically loaded engines.

This only has an effect in a shared build.

Don’t build support for Elliptic Curves.

Don’t build support for binary Elliptic Curves

Enable support for optimised implementations of some commonly used NIST elliptic curves.

This option is only supported on platforms:

Build support for gathering entropy from the Entropy Gathering Daemon (EGD).

Don’t build support for loading engines.

Don’t compile in any error strings.

Enable building of integration with external test suites.

This is a developer option and may not work on all platforms. The following external test suites are currently supported:

See the file test/README-external.md for further details.

Don’t compile in filename and line number information (e.g. for errors and memory allocation).

Don’t compile the FIPS provider

Don’t perform FIPS module run-time checks related to enforcement of security parameters such as minimum security strength of keys.

Build with support for fuzzing using either libfuzzer or AFL.

These are developer options only. They may not work on all platforms and should never be used in production environments.

See the file fuzz/README.md for further details.

Don’t build support for GOST based ciphersuites.

Note that if this feature is enabled then GOST ciphersuites are only available if the GOST algorithms are also available through loading an externally supplied engine.

Don’t build the legacy provider.

Disabling this also disables the legacy algorithms: MD2 (already disabled by default).

Don’t generate dependencies.

Don’t build any dynamically loadable engines.

Don’t build support for writing multiple records in one go in libssl

Note: this is a different capability to the pipelining functionality.

Don’t build support for the Next Protocol Negotiation (NPN) TLS extension.

Don’t build support for Online Certificate Status Protocol (OCSP).

Don’t build the padlock engine.

Don’t build with support for Position Independent Code.

Don’t pin the shared libraries.

Don’t use POSIX IO capabilities.

Don’t build support for Pre-Shared Key based ciphersuites.

Don’t use hardware RDRAND capabilities.

Don’t build support for RFC3779, «X.509 Extensions for IP Addresses and AS Identifiers».

Build support for Stream Control Transmission Protocol (SCTP).

Do not create shared libraries, only static ones.

Don’t build support for socket BIOs.

Don’t build support for Secure Remote Password (SRP) protocol or SRP based ciphersuites.

Don’t build Secure Real-Time Transport Protocol (SRTP) support.

Exclude SSE2 code paths from 32-bit x86 assembly modules.

Build with the SSL Trace capabilities.

Don’t build the statically linked engines.

This only has an impact when not built «shared».

Don’t use anything from the C header file stdio.h that makes use of the FILE type. Only libcrypto and libssl can be built in this way. Using this option will suppress building the command line applications. Additionally, since the OpenSSL tests also use the command line applications, the tests will also be skipped.

Don’t build test programs or run any tests.

Don’t build with support for multi-threaded applications.

Build with support for multi-threaded applications. Most platforms will enable this by default. However, if on a platform where this is not the case then this will usually require additional system-dependent options!

Build with support for the integrated tracing api.

See manual pages OSSL_trace_set_channel(3) and OSSL_trace_enabled(3) for details.

Don’t build Time Stamping (TS) Authority support.

Build with the Undefined Behaviour sanitiser (UBSAN).

Don’t build with the User Interface (UI) console method

The User Interface console method enables text based console prompts.

Enable additional unit test APIs.

This should not typically be used in production deployments.

Don’t build support for UPLINK interface.

Build support for SSL/TLS ciphers that are considered «weak»

Enabling this includes for example the RC4 based ciphersuites.

Build with support for zlib compression/decompression.

Like the zlib option, but has OpenSSL load the zlib library dynamically when needed.

This is only supported on systems where loading of shared libraries is supported.

In 32-bit x86 builds, use the 80386 instruction set only in assembly modules

The default x86 code is more efficient, but requires at least an 486 processor. Note: This doesn’t affect compiler generated code, so this option needs to be accompanied by a corresponding compiler-specific option.

Don’t build support for negotiating the specified SSL/TLS protocol.

Analogous to no- but in addition do not build the methods for applications to explicitly select individual protocol versions. Note that there is no no-tls1_3-method option because there is no application method for TLSv1.3.

Using individual protocol methods directly is deprecated. Applications should use TLS_method() instead.

Build with support for the specified algorithm.

Build without support for the specified algorithm.

These system specific options will be recognised and passed through to the compiler to allow you to define preprocessor symbols, specify additional libraries, library directories or other compiler options. It might be worth noting that some compilers generate code specifically for processor the compiler currently executes on. This is not necessarily what you might have in mind, since it might be unsuitable for execution on other, typically older, processor. Consult your compiler documentation.

Take note of the Environment Variables documentation below and how these flags interact with those variables.

Take note of the Environment Variables documentation below and how these flags interact with those variables.

These work just like normal environment variable assignments, but are supported on all platforms and are confined to the configuration scripts only. These assignments override the corresponding value in the inherited environment, if there is one.

The following variables are used as » make variables» and can be used as an alternative to giving preprocessor, compiler and linker options directly as configuration. The following variables are supported:

These cannot be mixed with compiling/linking flags given on the command line. In other words, something like this isn’t permitted.

Backward compatibility note:

To be compatible with older configuration scripts, the environment variables are ignored if compiling/linking flags are given on the command line, except for the following:

However, the following will see both set variables:

Reconfigure from earlier data.

Displaying configuration data

For more information, please do:

Installation Steps in Detail

On some platform a config script is available which attempts to guess your operating system (and compiler, if necessary) and calls the Configure Perl script with appropriate target based on its guess. Further options can be supplied to the config script, which will be passed on to the Configure script.

Unix / Linux / macOS

OpenSSL knows about a range of different operating system, hardware and compiler combinations. To see the ones it knows about, run

For the remainder of this text, the Unix form will be used in all examples. Please use the appropriate form for your platform.

Creating your own Configuration

If your system isn’t listed, you will have to create a configuration file named Configurations/<< something >>.conf and add the correct configuration for your system. See the available configs as examples and read Configurations/README.md and Configurations/README-design.md for more information.

The generic configurations cc or gcc should usually work on 32 bit Unix-like systems.

Out of Tree Builds

OpenSSL can be configured to build in a build directory separate from the source code directory. It’s done by placing yourself in some other directory and invoking the configuration commands from there.

Paths can be relative just as well as absolute. Configure will do its best to translate them to relative paths whenever possible.

Build OpenSSL by running:

This will build the OpenSSL libraries ( libcrypto.a and libssl.a on Unix, corresponding on other platforms) and the OpenSSL binary ( openssl ). The libraries will be built in the top-level directory, and the binary will be in the apps/ subdirectory.

If the build fails, take a look at the Build Failures subsection of the Troubleshooting section.

After a successful build, and before installing, the libraries should be tested. Run:

Warning: you MUST run the tests from an unprivileged account (or disable your privileges temporarily if your platform allows it).

See test/README.md for further details how run tests.

See test/README-dev.md for guidelines on adding tests.

If everything tests ok, install OpenSSL with

Note that in order to perform the install step above you need to have appropriate permissions to write to the installation directory.

The above commands will install all the software components in this directory tree under

The installation directory should be appropriately protected to ensure unprivileged users cannot make changes to OpenSSL binaries or files, or install engines. If you already have a pre-installed version of OpenSSL as part of your Operating System it is recommended that you do not overwrite the system version and instead install to somewhere else.

Package builders who want to configure the library for standard locations, but have the package installed somewhere else so that it can easily be packaged, can use

The specified destination directory will be prepended to all installation target paths.

Compatibility issues with previous OpenSSL versions

COMPILING existing applications

Starting with version 1.1.0, OpenSSL hides a number of structures that were previously open. This includes all internal libssl structures and a number of EVP types. Accessor functions have been added to allow controlled access to the structures’ data.

This means that some software needs to be rewritten to adapt to the new ways of doing things. This often amounts to allocating an instance of a structure explicitly where you could previously allocate them on the stack as automatic variables, and using the provided accessor functions where you would previously access a structure’s field directly.

Some APIs have changed as well. However, older APIs have been preserved when possible.

With the default OpenSSL installation comes a FIPS provider module, which needs some post-installation attention, without which it will not be usable. This involves using the following command:

See the openssl-fipsinstall(1) manual for details and examples.

Advanced Build Options

The Configure script generates a Makefile in a format relevant to the specific platform. The Makefiles provide a number of targets that can be used. Not all targets may be available on all platforms. Only the most common targets are described here. Examine the Makefiles themselves for the full list.

Running Selected Tests

Selecting the correct target

The directory contains two README files, which explain the general syntax and design of the configuration files.

If you need further help, try to search the openssl-users mailing list or the GitHub Issues for existing solutions. If you don’t find anything, you can raise an issue to ask a question yourself.

More about our support resources can be found in the SUPPORT file.

Along with a short description of the bug, please provide the complete configure command line and the relevant output including the error message.

Note: To make the output readable, pleace add a ‘code fence’ (three backquotes «` on a separate line) before and after your output:

If the build fails, look carefully at the output. Try to locate and understand the error message. It might be that the compiler is already telling you exactly what you need to do to fix your problem.

There may be reasons for the failure that aren’t problems in OpenSSL itself, for example if the compiler reports missing standard or third party headers.

If the build succeeded previously, but fails after a source or configuration change, it might be helpful to clean the build tree before attempting another build. Use this command:

Assembler error messages can sometimes be sidestepped by using the no-asm configuration option. See also notes.

Compiling parts of OpenSSL with gcc and others with the system compiler will result in unresolved symbols on some systems.

If you are still having problems, try to search the openssl-users mailing list or the GitHub Issues for existing solutions. If you think you encountered an OpenSSL bug, please raise an issue to file a bug report. Please take the time to review the existing issues first; maybe the bug was already reported or has already been fixed.

If some tests fail, look at the output. There may be reasons for the failure that isn’t a problem in OpenSSL itself (like an OS malfunction or a Perl issue).

You may want increased verbosity, that can be accomplished as described in section Test Failures of test/README.md.

You may also want to selectively specify which test(s) to perform. This can be done using the make variable TESTS as described in section Running Selected Tests of test/README.md.

If you find a problem with OpenSSL itself, try removing any compiler optimization flags from the CFLAGS line in the Makefile and run make clean; make or corresponding.

To report a bug please open an issue on GitHub, at https://github.com/openssl/openssl/issues.

Notes on multi-threading

For some systems, the OpenSSL Configure script knows what compiler options are needed to generate a library that is suitable for multi-threaded applications. On these systems, support for multi-threading is enabled by default; use the no-threads option to disable (this should never be necessary).

OpenSSL provides built-in support for two threading models: pthreads (found on most UNIX/Linux systems), and Windows threads. No other threading models are supported. If your platform does not provide pthreads or Windows threads then you should use Configure with the no-threads option.

Notes on shared libraries

For most systems the OpenSSL Configure script knows what is needed to build shared libraries for libcrypto and libssl. On these systems the shared libraries will be created by default. This can be suppressed and only static libraries created by using the no-shared option. On systems where OpenSSL does not know how to build shared libraries the no-shared option will be forced and only static libraries will be created.

Shared libraries are named a little differently on different platforms. One way or another, they all have the major OpenSSL version number as part of the file name, i.e. for OpenSSL 1.1.x, 1.1 is somehow part of the name.

Notes on random number generation

Availability of cryptographically secure random numbers is required for secret key generation. OpenSSL provides several options to seed the internal CSPRNG. If not properly seeded, the internal CSPRNG will refuse to deliver random bytes and a «PRNG not seeded error» will occur.

Notes on assembler modules compilation

Compilation of some code paths in assembler modules might depend on whether the current assembler version supports certain ISA extensions or not. Code paths that use the AES-NI, PCLMULQDQ, SSSE3, and SHA extensions are always assembled. Apart from that, the minimum requirements for the assembler versions are shown in the table below:

instead of the real clang. In which case it doesn’t matter what clang version is used, as it is the version of the GNU assembler that will be checked.

Источник

Вам также может понравиться

Прокат коммерческих авто в москве

Аренда коммерческого авто Когда лучше взять аренду коммерческого транспорта? Представьте, вы хотите перевезти

050

Windows німділігіні индексін немен лшейді

Компьютерді жаңғырту Кез келген құрылғы біраз уақыт өткеннен кейін өзінің мүмкіндігін жоғалта бастайды

045

Windows где лежат логи

Windows где лежат логи Всем привет, тема стать как посмотреть логи windows. Что такое логи думаю знают

048