root
/
openssl
mirror of https://github.com/openssl/openssl.git
1
0
Fork 0
Code Issues Actions 7 Packages Projects Releases Wiki Activity
openssl/Configure

1788 lines
84 KiB
Plaintext
Raw Normal View History

Don't hard-code path to Perl interpreter on shebang line of Configure script. Instead use the usual Shell->Perl transition trick.
1999-02-25 16:48:52 +08:00
:
Fix clearly untested "clever" hack.
1999-02-25 17:43:26 +08:00
eval 'exec perl -S $0 ${1+"$@"}'
Don't hard-code path to Perl interpreter on shebang line of Configure script. Instead use the usual Shell->Perl transition trick.
1999-02-25 16:48:52 +08:00
if $running_under_some_shell;
##
## Configure -- OpenSSL source tree configuration script
##
Add strictness, fix variable substition bugs.
1998-12-29 01:08:48 +08:00
Require Perl 5.
1998-12-29 01:37:17 +08:00
require 5.000;
Add strictness, fix variable substition bugs.
1998-12-29 01:08:48 +08:00
use strict;
Defunct assembler files removed; various cleanups. New Ultrix and Alpha entries submitted by Bernhard Simon <simon@zid.tuwien.ac.at>.
1999-04-19 21:54:11 +08:00
# see INSTALL for instructions.
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
"no-engine" was being ignored, so remove it from the advertised syntax. Also remove some commented-out lines of code that deny CVS its purpose.
2004-06-03 11:34:53 +08:00
my $usage="Usage: Configure [no-<cipher> ...] [-Dxxx] [-lxxx] [-Lxxx] [-fxxx] [-Kxxx] [no-hw-xxx|no-hw] [[no-]threads] [[no-]shared] [[no-]zlib|zlib-dynamic] [no-asm] [no-dso] [no-krb5] [386] [--prefix=DIR] [--openssldir=OPENSSLDIR] [--with-xxx[=vvv]] [--test-sanity] os/compiler[:flags]\n";
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
Document additional Configure flags.
1999-04-19 23:19:58 +08:00
# Options:
Support INSTALL_PREFIX for packagers. Submitted by: Reviewed by: PR:
1999-04-30 05:52:08 +08:00
#
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
# --openssldir install OpenSSL in OPENSSLDIR (Default: DIR/ssl if the
# --prefix option is given; /usr/local/ssl otherwise)
# --prefix prefix for the OpenSSL include, lib and bin directories
# (Default: the OPENSSLDIR directory)
Support INSTALL_PREFIX for packagers. Submitted by: Reviewed by: PR:
1999-04-30 05:52:08 +08:00
#
# --install_prefix Additional prefix for package builders (empty by
# default). This needn't be set in advance, you can
# just as well use "make INSTALL_PREFIX=/whatever install".
#
First tentative impementation of Kerberos 5 cryptos and keys for SSL/TLS. Implemented by Vern Staats <staatsvr@asc.hpc.mil>, further hacked and distributed by Jeffrey Altman <jaltnab@columbia.edu>
2000-12-01 06:53:34 +08:00
# --with-krb5-dir Declare where Kerberos 5 lives. The libraries are expected
# to live in the subdirectory lib/ and the header files in
Add the possibility to specify the use of zlib compression and decompression. It can be set up to link at link time or to load the zlib library at run-time.
2001-07-12 17:11:14 +08:00
# include/. A value is required.
# --with-krb5-lib Declare where the Kerberos 5 libraries live. A value is
# required.
First tentative impementation of Kerberos 5 cryptos and keys for SSL/TLS. Implemented by Vern Staats <staatsvr@asc.hpc.mil>, further hacked and distributed by Jeffrey Altman <jaltnab@columbia.edu>
2000-12-01 06:53:34 +08:00
# (Default: KRB5_DIR/lib)
Add the possibility to specify the use of zlib compression and decompression. It can be set up to link at link time or to load the zlib library at run-time.
2001-07-12 17:11:14 +08:00
# --with-krb5-include Declare where the Kerberos 5 header files live. A
# value is required.
First tentative impementation of Kerberos 5 cryptos and keys for SSL/TLS. Implemented by Vern Staats <staatsvr@asc.hpc.mil>, further hacked and distributed by Jeffrey Altman <jaltnab@columbia.edu>
2000-12-01 06:53:34 +08:00
# (Default: KRB5_DIR/include)
# --with-krb5-flavor Declare what flavor of Kerberos 5 is used. Currently
Add the possibility to specify the use of zlib compression and decompression. It can be set up to link at link time or to load the zlib library at run-time.
2001-07-12 17:11:14 +08:00
# supported values are "MIT" and "Heimdal". A value is required.
First tentative impementation of Kerberos 5 cryptos and keys for SSL/TLS. Implemented by Vern Staats <staatsvr@asc.hpc.mil>, further hacked and distributed by Jeffrey Altman <jaltnab@columbia.edu>
2000-12-01 06:53:34 +08:00
#
I got some reports that some targets have weird dso_schemes. Therefore, I've added a sanity checker. Note that it can be combined with almost any other argument (the other arguments will be completely ignored), with "reconf" as the blatant exception, since it also has the behavior of ignoring all following command line arguments. If --test-sanity and reconf are both used on the command line, the first one wins.
2002-01-27 23:52:37 +08:00
# --test-sanity Make a number of sanity checks on the data in this file.
# This is a debugging tool for OpenSSL developers.
#
Merge the engine branch into the main trunk. All conflicts resolved. At the same time, add VMS support for Rijndael.
2000-10-27 05:07:28 +08:00
# no-hw-xxx do not compile support for specific crypto hardware.
# Generic OpenSSL-style methods relating to this support
# are always compiled but return NULL if the hardware
# support isn't compiled.
# no-hw do not compile support for any crypto hardware.
New Configure options "threads" and "no-threads". For Solaris and Linux, "threads" (with proper compiler options) is the default. Submitted by: Reviewed by: PR:
1999-04-30 00:10:09 +08:00
# [no-]threads [don't] try to create a library that is suitable for
# multithreaded applications (default is "threads" if we
# know how to do it)
Added and corrected documentation for the 'shared' option
2000-08-02 12:19:50 +08:00
# [no-]shared [don't] try to create shared libraries when supported.
Error in comment.
1999-04-19 22:55:56 +08:00
# no-asm do not use assembler
This helps make the DSO stuff more portable; * "no-dso" option available in Configure so that all DSO methods will return NULL, overriding any support the platform might otherwise have built. * dlfcn_no_h config string now available rather than just dlfcn. This is for platforms that have dlfcn.h functions but do not have (or need) the dlfcn.h header file.
2000-04-06 15:09:45 +08:00
# no-dso do not compile in any native shared-library methods. This
# will ensure that all methods just return NULL.
First tentative impementation of Kerberos 5 cryptos and keys for SSL/TLS. Implemented by Vern Staats <staatsvr@asc.hpc.mil>, further hacked and distributed by Jeffrey Altman <jaltnab@columbia.edu>
2000-12-01 06:53:34 +08:00
# no-krb5 do not compile in any KRB5 library or code.
Add the possibility to specify the use of zlib compression and decompression. It can be set up to link at link time or to load the zlib library at run-time.
2001-07-12 17:11:14 +08:00
# [no-]zlib [don't] compile support for zlib compression.
# zlib-dynamic Like "zlib", but the zlib library is expected to be a shared
# library and will be loaded in run-time by the OpenSSL library.
Defunct assembler files removed; various cleanups. New Ultrix and Alpha entries submitted by Bernhard Simon <simon@zid.tuwien.ac.at>.
1999-04-19 21:54:11 +08:00
# 386 generate 80386 code
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
# no-sse2 disables IA-32 SSE2 code, above option implies no-sse2
New Configure option no-<cipher> (rsa, idea, rc5, ...).
1999-04-27 11:19:12 +08:00
# no-<cipher> build without specified algorithm (rsa, idea, rc5, ...)
Update HPUX config, work around HPUX library incompatibility. Submitted by: Anonymous
1999-05-04 19:52:26 +08:00
# -<xxx> +<xxx> compiler options are passed through
This change will cause builds (by default) to not use different STACK structures and functions for each stack type. The previous behaviour can be enabled by configuring with the "-DDEBUG_SAFESTACK" option. This will also cause "make update" (mkdef.pl in particular) to update the libeay.num and ssleay.num symbol tables with the number of extra functions DEBUG_SAFESTACK creates. The way this change works is to accompany each DECLARE_STACK_OF() macro with a set of "#define"d versions of the sk_##type##_*** functions that ensures all the existing "type-safe" stack calls are precompiled into the underlying stack calls. The presence or abscence of the DEBUG_SAFESTACK symbol controls whether this block of "#define"s or the DECLARE_STACK_OF() macro is taking effect. The block of "#define"s is in turn generated and maintained by a perl script (util/mkstack.pl) that encompasses the block with delimiting C comments. This works in a similar way to the auto-generated error codes and, like the other such maintenance utilities, is invoked by the "make update" target. A long (but mundane) commit will follow this with the results of "make update" - this will include all the "#define" blocks for each DECLARE_STACK_OF() statement, along with stripped down libeay.num and ssleay.num files.
2000-06-01 13:13:52 +08:00
#
# DEBUG_SAFESTACK use type-safe stacks to enforce type-safety on stack items
# provided to stack calls. Generates unique stack functions for
# each possible stack type.
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
# DES_PTR use pointer lookup vs arrays in the DES in crypto/des/des_locl.h
# DES_RISC1 use different DES_ENCRYPT macro that helps reduce register
# dependancies but needs to more registers, good for RISC CPU's
# DES_RISC2 A different RISC variant.
# DES_UNROLL unroll the inner DES loop, sometimes helps, somtimes hinders.
# DES_INT use 'int' instead of 'long' for DES_LONG in crypto/des/des.h
# This is used on the DEC Alpha where long is 8 bytes
# and int is 4
# BN_LLONG use the type 'long long' in crypto/bn/bn.h
Import of old SSLeay release: SSLeay 0.9.0b
1998-12-21 18:56:39 +08:00
# MD2_CHAR use 'char' instead of 'int' for MD2_INT in crypto/md2/md2.h
# MD2_LONG use 'long' instead of 'int' for MD2_INT in crypto/md2/md2.h
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
# IDEA_SHORT use 'short' instead of 'int' for IDEA_INT in crypto/idea/idea.h
# IDEA_LONG use 'long' instead of 'int' for IDEA_INT in crypto/idea/idea.h
# RC2_SHORT use 'short' instead of 'int' for RC2_INT in crypto/rc2/rc2.h
# RC2_LONG use 'long' instead of 'int' for RC2_INT in crypto/rc2/rc2.h
# RC4_CHAR use 'char' instead of 'int' for RC4_INT in crypto/rc4/rc4.h
# RC4_LONG use 'long' instead of 'int' for RC4_INT in crypto/rc4/rc4.h
Import of old SSLeay release: SSLeay 0.9.0b
1998-12-21 18:56:39 +08:00
# RC4_INDEX define RC4_INDEX in crypto/rc4/rc4_locl.h. This turns on
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
# array lookups instead of pointer use.
RC4 tune-up featuring 30-40% performance improvement on most RISC platforms. See crypto/rc4/rc4_enc.c for further details.
1999-10-07 20:03:59 +08:00
# RC4_CHUNK enables code that handles data aligned at long (natural CPU
# word) boundary.
# RC4_CHUNK_LL enables code that handles data aligned at long long boundary
# (intended for 64-bit CPUs running 32-bit OS).
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
# BF_PTR use 'pointer arithmatic' for Blowfish (unsafe on Alpha).
Andy Polyakov points out that BF_PTR2 is slower than the generic case.
1999-04-24 05:29:45 +08:00
# BF_PTR2 intel specific version (generic version is more efficient).
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
#
# Following are set automatically by this script
#
Import of old SSLeay release: SSLeay 0.9.0b
1998-12-21 18:56:39 +08:00
# MD5_ASM use some extra md5 assember,
# SHA1_ASM use some extra sha1 assember, must define L_ENDIAN for x86
# RMD160_ASM use some extra ripemd160 assember,
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
# SHA256_ASM sha256_block is implemented in assembler
# SHA512_ASM sha512_block is implemented in assembler
# AES_ASM ASE_[en|de]crypt is implemented in assembler
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
Add strictness, fix variable substition bugs.
1998-12-29 01:08:48 +08:00
my $x86_gcc_des="DES_PTR DES_RISC1 DES_UNROLL";
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
# MD2_CHAR slags pentium pros
Andy Polyakov points out that BF_PTR2 is slower than the generic case.
1999-04-24 05:29:45 +08:00
my $x86_gcc_opts="RC4_INDEX MD2_INT";
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
# MODIFY THESE PARAMETERS IF YOU ARE GOING TO USE THE 'util/speed.sh SCRIPT
# Don't worry about these normally
Add strictness, fix variable substition bugs.
1998-12-29 01:08:48 +08:00
my $tcc="cc";
my $tflags="-fast -Xa";
my $tbn_mul="";
my $tlib="-lnsl -lsocket";
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
#$bits1="SIXTEEN_BIT ";
#$bits2="THIRTY_TWO_BIT ";
Add strictness, fix variable substition bugs.
1998-12-29 01:08:48 +08:00
my $bits1="THIRTY_TWO_BIT ";
my $bits2="SIXTY_FOUR_BIT ";
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
Throw in AES CBC assembler, up to +40% on aes-128-cbc benchmark.
2005-01-18 09:04:41 +08:00
my $x86_elf_asm="x86cpuid-elf.o:asm/bn86-elf.o asm/co86-elf.o:asm/dx86-elf.o asm/yx86-elf.o:asm/ax86-elf.o:asm/bx86-elf.o:asm/mx86-elf.o:asm/sx86-elf.o asm/s512sse2-elf.o:asm/cx86-elf.o:asm/rx86-elf.o:asm/rm86-elf.o:asm/r586-elf.o";
my $x86_coff_asm="x86cpuid-cof.o:asm/bn86-cof.o asm/co86-cof.o:asm/dx86-cof.o asm/yx86-cof.o:asm/ax86-cof.o:asm/bx86-cof.o:asm/mx86-cof.o:asm/sx86-cof.o asm/s512sse2-cof.o:asm/cx86-cof.o:asm/rx86-cof.o:asm/rm86-cof.o:asm/r586-cof.o";
my $x86_out_asm="x86cpuid-out.o:bn86-out.o co86-out.o:dx86-out.o yx86-out.o:ax86-out.o:bx86-out.o:mx86-out.o:sx86-out.o s512sse2-out.o:cx86-out.o:rx86-out.o:rm86-out.o:r586-out.o";
Engage RC4 IA-64 assembler module.
2004-11-26 23:12:17 +08:00
Reserve for AES CBC assembler implementation...
2005-01-18 08:43:32 +08:00
my $ia64_asm=":asm/ia64.o::aes_core.o aes_cbc.o asm/aes-ia64.o:::asm/sha1-ia64.o asm/sha256-ia64.o asm/sha512-ia64.o::asm/rc4-ia64.o::";
Engage RC4 IA-64 assembler module.
2004-11-26 23:12:17 +08:00
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
my $no_asm="::::::::::";
For mips3 and alpha, put the assembler file directives in separate variables and disable the Alpha assembler for now, since it has been shown to fail. The Alpha failure can be shown by adding the following numbers: FFFFFFFFFFFFFF0000FF2E00000000EBFFFFFF0000D1 FFFFFFFFFFFFFFFFFFFFFFFFFFFFFF46FFE0FFFF0000 The result is: 1FFFFFFFFFFFEFF0000FF2E0000000032FFE0FEFF00D1 The result should really be: 1FFFFFFFFFFFFFF0000FF2E0000000032FFE0FEFF00D1
2001-03-27 00:16:12 +08:00
Import of old SSLeay release: SSLeay 0.9.0b
1998-12-21 18:56:39 +08:00
# -DB_ENDIAN slows things down on a sparc for md5, but helps sha1.
# So the md5_locl.h file has an undef B_ENDIAN if sun is defined
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
#config-string $cc : $cflags : $unistd : $thread_cflag : $sys_id : $lflags : $bn_ops : $cpuid_obj : $bn_obj : $des_obj : $aes_obj : $bf_obj : $md5_obj : $sha1_obj : $cast_obj : $rc4_obj : $rmd160_obj : $rc5_obj : $dso_scheme : $shared_target : $shared_cflag : $shared_ldflag : $shared_extension : $ranlib : $arflags
config string comment
2000-01-23 05:26:01 +08:00
Add strictness, fix variable substition bugs.
1998-12-29 01:08:48 +08:00
my %table=(
Changes for QNX: there is no thread support, and the previous configuration only worked with no-asm.
2000-09-06 20:25:58 +08:00
# File 'TABLE' (created by 'make TABLE') contains the data from this list,
# formatted for better readability.
Protect variables from potential misinterpretations, for example a colon which may bite the lesser knowleged...
2000-03-08 07:17:48 +08:00
#"b", "${tcc}:${tflags}::${tlib}:${bits1}:${tbn_mul}::",
#"bl-4c-2c", "${tcc}:${tflags}::${tlib}:${bits1}BN_LLONG RC4_CHAR MD2_CHAR:${tbn_mul}::",
#"bl-4c-ri", "${tcc}:${tflags}::${tlib}:${bits1}BN_LLONG RC4_CHAR RC4_INDEX:${tbn_mul}::",
#"b2-is-ri-dp", "${tcc}:${tflags}::${tlib}:${bits2}IDEA_SHORT RC4_INDEX DES_PTR:${tbn_mul}::",
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
Defunct assembler files removed; various cleanups. New Ultrix and Alpha entries submitted by Bernhard Simon <simon@zid.tuwien.ac.at>.
1999-04-19 21:54:11 +08:00
# Our development configs
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
"purify", "purify gcc:-g -DPURIFY -Wall::(unknown)::-lsocket -lnsl::::",
The general debug target must specify that it doesn't use assembler routines. PR: 105
2002-06-27 17:13:44 +08:00
"debug", "gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DCRYPTO_MDEBUG -DOPENSSL_NO_ASM -ggdb -g2 -Wformat -Wshadow -Wmissing-prototypes -Wmissing-declarations -Werror::(unknown)::-lefence::::",
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
"debug-ben", "gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DCRYPTO_MDEBUG -DPEDANTIC -DDEBUG_SAFESTACK -O2 -pedantic -Wall -Wshadow -Werror -pipe::(unknown):::::asm/bn86-elf.o asm/co86-elf.o",
Clean up EVP macros, rename DES EDE3 modes correctly, temporary support for OpenBSD /dev/crypto (this will be revamped later when the appropriate machinery is available).
2001-07-21 18:24:07 +08:00
"debug-ben-openbsd","gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DCRYPTO_MDEBUG -DPEDANTIC -DDEBUG_SAFESTACK -DOPENSSL_OPENBSD_DEV_CRYPTO -DOPENSSL_NO_ASM -O2 -pedantic -Wall -Wshadow -Werror -pipe::(unknown)::::",
Constification, add config to /dev/crypto.
2002-01-19 00:51:05 +08:00
"debug-ben-openbsd-debug","gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DCRYPTO_MDEBUG -DPEDANTIC -DDEBUG_SAFESTACK -DOPENSSL_OPENBSD_DEV_CRYPTO -DOPENSSL_NO_ASM -g3 -O2 -pedantic -Wall -Wshadow -Werror -pipe::(unknown)::::",
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
"debug-ben-debug", "gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DCRYPTO_MDEBUG -DPEDANTIC -DDEBUG_SAFESTACK -g3 -O2 -pedantic -Wall -Wshadow -Werror -pipe::(unknown)::::::",
"debug-ben-strict", "gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DCRYPTO_MDEBUG -DCONST_STRICT -O2 -Wall -Wshadow -Werror -Wpointer-arith -Wcast-qual -Wwrite-strings -pipe::(unknown)::::::",
"debug-rse","cc:-DTERMIOS -DL_ENDIAN -pipe -O -g -ggdb3 -Wall::(unknown):::BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}",
Oops, wrong line
2001-09-06 01:22:40 +08:00
"debug-bodo", "gcc:-DL_ENDIAN -DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DBIO_PAIR_DEBUG -DPEDANTIC -g -m486 -pedantic -Wshadow -Wall::-D_REENTRANT:::BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}",
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"debug-ulf", "gcc:-DTERMIOS -DL_ENDIAN -march=i486 -Wall -DBN_DEBUG -DBN_DEBUG_RAND -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DCRYPTO_MDEBUG -DOPENSSL_NO_ASM -g -Wformat -Wshadow -Wmissing-prototypes -Wmissing-declarations:::CYGWIN32:::${no_asm}:win32:cygwin-shared:::.dll",
Update debug-steve
2004-10-01 19:35:38 +08:00
"debug-steve", "gcc:-DL_ENDIAN -DREF_CHECK -DCONF_DEBUG -DDEBUG_SAFESTACK -DCRYPTO_MDEBUG_ALL -DPEDANTIC -g -mcpu=i486 -pedantic -Wno-long-long -Wall -Werror -Wshadow -pipe::-D_REENTRANT::-rdynamic -ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn:linux-shared",
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
"debug-steve-linux-pseudo64", "gcc:-DL_ENDIAN -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DDEBUG_SAFESTACK -DCRYPTO_MDEBUG_ALL -DOPENSSL_NO_ASM -g -mcpu=i486 -Wall -Werror -Wshadow -pipe::-D_REENTRANT::-rdynamic -ldl:SIXTY_FOUR_BIT:${no_asm}:dlfcn:linux-shared",
Whooaaaaa, the BN_CTX_DEBUG macro really produces output these days... A little too much for my tests, currently...
2004-04-20 19:53:33 +08:00
"debug-levitte-linux-elf","gcc:-DLEVITTE_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_DEBUG -DBN_DEBUG_RAND -DCRYPTO_MDEBUG -DENGINE_CONF_DEBUG -DL_ENDIAN -DTERMIO -D_POSIX_SOURCE -DPEDANTIC -ggdb -g3 -mcpu=i486 -pedantic -ansi -Wall -Wshadow -Wcast-align -Wstrict-prototypes -Wmissing-prototypes -Wno-long-long -Wundef -Wconversion -pipe::-D_REENTRANT::-ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"debug-levitte-linux-noasm","gcc:-DLEVITTE_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_DEBUG -DBN_DEBUG_RAND -DCRYPTO_MDEBUG -DENGINE_CONF_DEBUG -DOPENSSL_NO_ASM -DL_ENDIAN -DTERMIO -D_POSIX_SOURCE -DPEDANTIC -ggdb -g3 -mcpu=i486 -pedantic -ansi -Wall -Wshadow -Wcast-align -Wstrict-prototypes -Wmissing-prototypes -Wno-long-long -Wundef -Wconversion -pipe::-D_REENTRANT::-ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${no_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Whooaaaaa, the BN_CTX_DEBUG macro really produces output these days... A little too much for my tests, currently...
2004-04-20 19:53:33 +08:00
"debug-levitte-linux-elf-extreme","gcc:-DLEVITTE_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_DEBUG -DBN_DEBUG_RAND -DCRYPTO_MDEBUG -DENGINE_CONF_DEBUG -DL_ENDIAN -DTERMIO -D_POSIX_SOURCE -DPEDANTIC -ggdb -g3 -mcpu=i486 -pedantic -ansi -Wall -W -Wundef -Wshadow -Wcast-align -Wstrict-prototypes -Wmissing-prototypes -Wno-long-long -Wundef -Wconversion -pipe::-D_REENTRANT::-ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"debug-levitte-linux-noasm-extreme","gcc:-DLEVITTE_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_DEBUG -DBN_DEBUG_RAND -DCRYPTO_MDEBUG -DENGINE_CONF_DEBUG -DOPENSSL_NO_ASM -DL_ENDIAN -DTERMIO -D_POSIX_SOURCE -DPEDANTIC -ggdb -g3 -mcpu=i486 -pedantic -ansi -Wall -W -Wundef -Wshadow -Wcast-align -Wstrict-prototypes -Wmissing-prototypes -Wno-long-long -Wundef -Wconversion -pipe::-D_REENTRANT::-ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${no_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"debug-geoff","gcc:-DBN_DEBUG -DBN_DEBUG_RAND -DBN_STRICT -DPURIFY -DOPENSSL_NO_DEPRECATED -DOPENSSL_NO_ASM -DOPENSSL_NO_INLINE_ASM -DL_ENDIAN -DTERMIO -DPEDANTIC -O1 -ggdb2 -Wall -Werror -Wundef -pedantic -Wshadow -Wpointer-arith -Wbad-function-cast -Wcast-align -Wsign-compare -Wmissing-prototypes -Wmissing-declarations -Wno-long-long::-D_REENTRANT::-ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${no_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
"dist", "cc:-O::(unknown)::::::",
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
Avoid GNU C assembler templates under Solaris x86.
2000-02-16 01:31:12 +08:00
# Basic configs that should work on any (32 and less bit) box
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
"gcc", "gcc:-O3::(unknown):::BN_LLONG:::",
"cc", "cc:-O::(unknown)::::::",
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
Changes for VOS, submitted by Paul Green <Paul.Green@stratus.com>. PR: 499
2004-06-29 06:01:37 +08:00
####VOS Configurations
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"vos-gcc","gcc:-b hppa1.1-stratus-vos -O3 -Wall -Wuninitialized -D_POSIX_C_SOURCE=200112L -D_BSD::(unknown):VOS:-Wl,-map:BN_LLONG:${no_asm}:::::.so:",
"debug-vos-gcc","gcc:-b hppa1.1-stratus-vos -O0 -g -Wall -D_POSIX_C_SOURCE=200112L -D_BSD -DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DCRYPTO_MDEBUG::(unknown):VOS:-Wl,-map:BN_LLONG:${no_asm}:::::.so:",
"vos-vcc","vcc:-b i386-stratus-vos -O3 -D_POSIX_C_SOURCE=200112L -D_BSD::(unknown):VOS:-Wl,-map::${no_asm}:::::.so:",
"debug-vos-vcc","vcc:-b i386-stratus-vos -O0 -g -D_POSIX_C_SOURCE=200112L -D_BSD -DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DCRYPTO_MDEBUG::(unknown):VOS:-Wl,-map::${no_asm}:::::.so:",
Changes for VOS, submitted by Paul Green <Paul.Green@stratus.com>. PR: 499
2004-06-29 06:01:37 +08:00
Add support for Sun C on Solaris x86. Contributed by Ben <mouring@etoh.eviladmin.org>
2001-05-03 14:14:47 +08:00
#### Solaris x86 with GNU C setups
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
# -DOPENSSL_NO_INLINE_ASM switches off inline assembler. We have to do it
Avoid GNU C assembler templates under Solaris x86.
2000-02-16 01:31:12 +08:00
# here because whenever GNU C instantiates an assembler template it
# surrounds it with #APP #NO_APP comment pair which (at least Solaris
# 7_x86) /usr/ccs/bin/as fails to assemble with "Illegal mnemonic"
# error message.
Unified targets for ELF assembler modules. Tested on Linux, Solaris and FreeBSD. Goal is to extend support even to SCO5, UnixWare/OpenUnix...
2003-01-04 01:37:53 +08:00
"solaris-x86-gcc","gcc:-O3 -fomit-frame-pointer -m486 -Wall -DL_ENDIAN -DOPENSSL_NO_INLINE_ASM::-D_REENTRANT::-lsocket -lnsl -ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn:solaris-shared:-fPIC:-shared:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Add support for Sun C on Solaris x86. Contributed by Ben <mouring@etoh.eviladmin.org>
2001-05-03 14:14:47 +08:00
#### Solaris x86 with Sun C setups
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"solaris-x86-cc","cc:-fast -O -Xa::-D_REENTRANT::-lsocket -lnsl -ldl:BN_LLONG RC4_CHAR RC4_CHUNK DES_PTR DES_UNROLL BF_PTR:${no_asm}:dlfcn:solaris-shared:-KPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Add support for Sun C on Solaris x86. Contributed by Ben <mouring@etoh.eviladmin.org>
2001-05-03 14:14:47 +08:00
SPARC Solaris config updates. ./config sences whole range of SPARC instruction sets. Do note that it favors Sun C now if both gcc and cc 4.2 or later are present!
1999-07-26 06:25:12 +08:00
#### SPARC Solaris with GNU C setups
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"solaris-sparcv7-gcc","gcc:-O3 -fomit-frame-pointer -Wall -DB_ENDIAN -DBN_DIV2W::-D_REENTRANT::-lsocket -lnsl -ldl:BN_LLONG RC4_CHAR RC4_CHUNK DES_UNROLL BF_PTR:${no_asm}:dlfcn:solaris-shared:-fPIC:-shared:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
"solaris-sparcv8-gcc","gcc:-mv8 -O3 -fomit-frame-pointer -Wall -DB_ENDIAN -DBN_DIV2W::-D_REENTRANT::-lsocket -lnsl -ldl:BN_LLONG RC4_CHAR RC4_CHUNK DES_UNROLL BF_PTR::asm/sparcv8.o:asm/des_enc-sparc.o fcrypt_b.o:::::::::dlfcn:solaris-shared:-fPIC:-shared:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Support for 64-bit Solaris build with GCC 3.0 and later. It should be explicitely noted that 64-bit SPARCv9 ABI is not officially supported by GCC 3.0 (support is scheduled for 3.1 release), but it appears to work, at the very least 'make test' passes...
2001-07-25 23:58:57 +08:00
# -m32 should be safe to add as long as driver recognizes -mcpu=ultrasparc
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
"solaris-sparcv9-gcc","gcc:-m32 -mcpu=ultrasparc -O3 -fomit-frame-pointer -Wall -DB_ENDIAN -DBN_DIV2W::-D_REENTRANT:ULTRASPARC:-lsocket -lnsl -ldl:BN_LLONG RC4_CHAR RC4_CHUNK DES_UNROLL BF_PTR::asm/sparcv8plus.o:asm/des_enc-sparc.o fcrypt_b.o:::asm/md5-sparcv8plus.o::::::dlfcn:solaris-shared:-fPIC:-shared:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"solaris64-sparcv9-gcc","gcc:-m64 -mcpu=ultrasparc -O3 -Wall -DB_ENDIAN::-D_REENTRANT:ULTRASPARC:-lsocket -lnsl -ldl:SIXTY_FOUR_BIT_LONG RC4_CHAR RC4_CHUNK DES_INT DES_PTR DES_RISC1 DES_UNROLL BF_PTR:::asm/des_enc-sparc.o fcrypt_b.o:::asm/md5-sparcv9.o::::::dlfcn:solaris-shared:-fPIC:-m64 -shared:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
SPARC Solaris config updates. ./config sences whole range of SPARC instruction sets. Do note that it favors Sun C now if both gcc and cc 4.2 or later are present!
1999-07-26 06:25:12 +08:00
####
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
"debug-solaris-sparcv8-gcc","gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DCRYPTO_MDEBUG_ALL -O -g -mv8 -Wall -DB_ENDIAN::-D_REENTRANT::-lsocket -lnsl -ldl:BN_LLONG RC4_CHAR RC4_CHUNK DES_UNROLL BF_PTR::asm/sparcv8.o::::::::::dlfcn:solaris-shared:-fPIC:-shared:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"debug-solaris-sparcv9-gcc","gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DCRYPTO_MDEBUG_ALL -DPEDANTIC -O -g -mcpu=ultrasparc -pedantic -ansi -Wall -Wshadow -Wno-long-long -D__EXTENSIONS__ -DB_ENDIAN -DBN_DIV2W::-D_REENTRANT:ULTRASPARC:-lsocket -lnsl -ldl:BN_LLONG RC4_CHAR RC4_CHUNK DES_UNROLL BF_PTR::asm/sparcv8plus.o:asm/des_enc-sparc.o fcrypt_b.o:::asm/md5-sparcv8plus.o::::::dlfcn:solaris-shared:-fPIC:-shared:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
SPARC Solaris config updates. ./config sences whole range of SPARC instruction sets. Do note that it favors Sun C now if both gcc and cc 4.2 or later are present!
1999-07-26 06:25:12 +08:00
#### SPARC Solaris with Sun C setups
# SC4.0 doesn't pass 'make test', upgrade to SC5.0 or SC4.2.
# SC4.2 is ok, better than gcc even on bn as long as you tell it -xarch=v8
Put SC5 warning back in, future changes might trigger the compiler bug again.
1999-06-10 07:29:23 +08:00
# SC5.0 note: Compiler common patch 107357-01 or later is required!
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"solaris-sparcv7-cc","cc:-xO5 -xstrconst -xdepend -Xa -DB_ENDIAN -DBN_DIV2W::-D_REENTRANT::-lsocket -lnsl -ldl:BN_LLONG RC4_CHAR RC4_CHUNK DES_PTR DES_RISC1 DES_UNROLL BF_PTR:${no_asm}:dlfcn:solaris-shared:-KPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
"solaris-sparcv8-cc","cc:-xarch=v8 -xO5 -xstrconst -xdepend -Xa -DB_ENDIAN -DBN_DIV2W::-D_REENTRANT::-lsocket -lnsl -ldl:BN_LLONG RC4_CHAR RC4_CHUNK DES_PTR DES_RISC1 DES_UNROLL BF_PTR::asm/sparcv8.o:asm/des_enc-sparc.o fcrypt_b.o:::::::::dlfcn:solaris-shared:-KPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"solaris-sparcv9-cc","cc:-xtarget=ultra -xarch=v8plus -xO5 -xstrconst -xdepend -Xa -DB_ENDIAN -DBN_DIV2W::-D_REENTRANT:ULTRASPARC:-lsocket -lnsl -ldl:BN_LLONG RC4_CHAR RC4_CHUNK_LL DES_PTR DES_RISC1 DES_UNROLL BF_PTR::asm/sparcv8plus.o:asm/des_enc-sparc.o fcrypt_b.o:::asm/md5-sparcv8plus.o::::::dlfcn:solaris-shared:-KPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"solaris64-sparcv9-cc","cc:-xtarget=ultra -xarch=v9 -xO5 -xstrconst -xdepend -Xa -DB_ENDIAN::-D_REENTRANT:ULTRASPARC:-lsocket -lnsl -ldl:SIXTY_FOUR_BIT_LONG RC4_CHAR RC4_CHUNK DES_INT DES_PTR DES_RISC1 DES_UNROLL BF_PTR:::asm/des_enc-sparc.o fcrypt_b.o:::asm/md5-sparcv9.o::::::dlfcn:solaris-shared:-KPIC:-xarch=v9:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR):/usr/ccs/bin/ar rs",
Add some debug-solaris-...-cc configurations.
1999-09-14 23:05:45 +08:00
####
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
"debug-solaris-sparcv8-cc","cc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DCRYPTO_MDEBUG_ALL -xarch=v8 -g -O -xstrconst -Xa -DB_ENDIAN -DBN_DIV2W::-D_REENTRANT::-lsocket -lnsl -ldl:BN_LLONG RC4_CHAR RC4_CHUNK DES_PTR DES_RISC1 DES_UNROLL BF_PTR::asm/sparcv8.o::::::::::dlfcn:solaris-shared:-KPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"debug-solaris-sparcv9-cc","cc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DCRYPTO_MDEBUG_ALL -xtarget=ultra -xarch=v8plus -g -O -xstrconst -Xa -DB_ENDIAN -DBN_DIV2W::-D_REENTRANT:ULTRASPARC:-lsocket -lnsl -ldl:BN_LLONG RC4_CHAR RC4_CHUNK_LL DES_PTR DES_RISC1 DES_UNROLL BF_PTR::asm/sparcv8plus.o::::asm/md5-sparcv8plus.o::::::dlfcn:solaris-shared:-KPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
#### SunOS configs, assuming sparc for the gcc one.
#"sunos-cc", "cc:-O4 -DNOPROTO -DNOCONST::(unknown):SUNOS::DES_UNROLL:${no_asm}::",
"sunos-gcc","gcc:-O3 -mv8 -Dssize_t=int::(unknown):SUNOS::BN_LLONG RC4_CHAR RC4_CHUNK DES_UNROLL DES_PTR DES_RISC1:${no_asm}::",
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
SGI IRIX config updates.
1999-07-26 04:40:58 +08:00
#### IRIX 5.x configs
# -mips2 flag is added by ./config when appropriate.
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"irix-gcc","gcc:-O3 -DTERMIOS -DB_ENDIAN::(unknown):::BN_LLONG MD2_CHAR RC4_INDEX RC4_CHAR RC4_CHUNK DES_UNROLL DES_RISC2 DES_PTR BF_PTR:${no_asm}:dlfcn:irix-shared:::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"irix-cc", "cc:-O2 -use_readonly_const -DTERMIOS -DB_ENDIAN::(unknown):::BN_LLONG RC4_CHAR RC4_CHUNK DES_PTR DES_RISC2 DES_UNROLL BF_PTR:${no_asm}:dlfcn:irix-shared:::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
SGI IRIX config updates.
1999-07-26 04:40:58 +08:00
#### IRIX 6.x configs
# Only N32 and N64 ABIs are supported. If you need O32 ABI build, invoke
IRIX 6.x shared build fix-up. For reference. Note that both cc and gcc support -Wl flag, but we can't use -Wl,-[not]all with both drivers, because cc rearranges options passed through -Wl. We can't use -Wl,-all,libcrypto.a,-notall with cc either, because it refuses to start with "no input" error.
2004-03-13 05:52:54 +08:00
# './Configure irix-cc -o32' manually.
SGI IRIX config updates.
1999-07-26 04:40:58 +08:00
# -mips4 flag is added by ./config when appropriate.
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
"irix-mips3-gcc","gcc:-mabi=n32 -mmips-as -O3 -DTERMIOS -DB_ENDIAN -DBN_DIV3W::-D_SGI_MP_SOURCE:::MD2_CHAR RC4_INDEX RC4_CHAR RC4_CHUNK_LL DES_UNROLL DES_RISC2 DES_PTR BF_PTR SIXTY_FOUR_BIT::asm/mips3.o::::::::::dlfcn:irix-shared::-mabi=n32:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"irix-mips3-cc", "cc:-n32 -mips3 -O2 -use_readonly_const -DTERMIOS -DB_ENDIAN -DBN_DIV3W::-D_SGI_MP_SOURCE:::DES_PTR RC4_CHAR RC4_CHUNK_LL DES_RISC2 DES_UNROLL BF_PTR SIXTY_FOUR_BIT::asm/mips3.o::::::::::dlfcn:irix-shared::-n32:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
SGI IRIX config updates.
1999-07-26 04:40:58 +08:00
# N64 ABI builds.
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
"irix64-mips4-gcc","gcc:-mabi=64 -mips4 -mmips-as -O3 -DTERMIOS -DB_ENDIAN -DBN_DIV3W::-D_SGI_MP_SOURCE:::RC4_CHAR RC4_CHUNK DES_RISC2 DES_UNROLL SIXTY_FOUR_BIT_LONG::asm/mips3.o::::::::::dlfcn:irix-shared::-mabi=64:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"irix64-mips4-cc", "cc:-64 -mips4 -O2 -use_readonly_const -DTERMIOS -DB_ENDIAN -DBN_DIV3W::-D_SGI_MP_SOURCE:::RC4_CHAR RC4_CHUNK DES_RISC2 DES_UNROLL SIXTY_FOUR_BIT_LONG::asm/mips3.o::::::::::dlfcn:irix-shared::-64:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
HP-UX tune-up: new unified configs, HP C compiler bug workaround.
2000-02-13 07:33:01 +08:00
#### Unified HP-UX ANSI C configs.
# Special notes:
# - Originally we were optimizing at +O4 level. It should be noted
# that the only difference between +O3 and +O4 is global inter-
# procedural analysis. As it has to be performed during the link
# stage the compiler leaves behind certain pseudo-code in lib*.a
# which might be release or even patch level specific. Generating
Avoid GNU C assembler templates under Solaris x86.
2000-02-16 01:31:12 +08:00
# the machine code for and analyzing the *whole* program appears
# to be *extremely* memory demanding while the performance gain is
HP-UX tune-up: new unified configs, HP C compiler bug workaround.
2000-02-13 07:33:01 +08:00
# actually questionable. The situation is intensified by the default
# HP-UX data set size limit (infamous 'maxdsiz' tunable) of 64MB
# which is way too low for +O4. In other words, doesn't +O3 make
# more sense?
Avoid GNU C assembler templates under Solaris x86.
2000-02-16 01:31:12 +08:00
# - Keep in mind that the HP compiler by default generates code
# suitable for execution on the host you're currently compiling at.
# If the toolkit is ment to be used on various PA-RISC processors
Enhanced support for IA-64 Linux and HP-UX (as well as better support for HP-UX in common in ./config). Note that for the moment of this writing none of 64-bit platforms pass bntest. I'm committing this anyway as it's too frustrating to patch snapshots over and over while 0.9.6 is known to work.
2001-07-31 00:42:15 +08:00
# consider './config +DAportable'.
hpux64-parisc2-gcc target added. Once it is verified, ./config should be modified to choose it instead of hpux64-parisc-gcc, which should then be removed. hpux64-parisc-cc is removed already now as redundant [in case you wonder, 64-bit HP-UX ABI *implies* PA-RISC2.0].
2003-11-21 03:10:36 +08:00
# - +DD64 is chosen in favour of +DA2.0W because it's meant to be
HP-UX tune-up: new unified configs, HP C compiler bug workaround.
2000-02-13 07:33:01 +08:00
# compatible with *future* releases.
# - If you run ./Configure hpux-parisc-[g]cc manually don't forget to
# pass -D_REENTRANT on HP-UX 10 and later.
# - -DMD32_XARRAY triggers workaround for compiler bug we ran into in
# 32-bit message digests. (For the moment of this writing) HP C
# doesn't seem to "digest" too many local variables (they make "him"
# chew forever:-). For more details look-up MD32_XARRAY comment in
# crypto/sha/sha_lcl.h.
# <appro@fy.chalmers.se>
#
A couple of corrections concerning HPUX 10 and shared libraries. Contributed by Lutz Jaenicke.
2000-03-25 00:06:22 +08:00
# Since there is mention of this in shlib/hpux10-cc.sh
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"hpux-parisc-cc-o4","cc:-Ae +O4 +ESlit -z -DB_ENDIAN -DBN_DIV2W -DMD32_XARRAY::-D_REENTRANT::-ldld:BN_LLONG DES_PTR DES_UNROLL DES_RISC1:${no_asm}:dl:hpux-shared:+Z:-b:.sl.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"hpux-parisc-gcc","gcc:-O3 -DB_ENDIAN -DBN_DIV2W::-D_REENTRANT::-Wl,+s -ldld:BN_LLONG DES_PTR DES_UNROLL DES_RISC1:${no_asm}:dl:hpux-shared:-fPIC:-shared:.sl.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
"hpux-parisc2-gcc","gcc:-march=2.0 -O3 -DB_ENDIAN -D_REENTRANT::::-Wl,+s -ldld:SIXTY_FOUR_BIT RC4_CHAR RC4_CHUNK DES_PTR DES_UNROLL DES_RISC1::asm/pa-risc2.o::::::::::dl:hpux-shared:-fPIC:-shared:.sl.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"hpux64-parisc2-gcc","gcc:-O3 -DB_ENDIAN -D_REENTRANT::::-ldl:SIXTY_FOUR_BIT_LONG MD2_CHAR RC4_INDEX RC4_CHAR DES_UNROLL DES_RISC1 DES_INT::asm/pa-risc2W.o::::::::::dlfcn:hpux-shared:-fpic:-shared:.sl.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
HP-UX tune-up: new unified configs, HP C compiler bug workaround.
2000-02-13 07:33:01 +08:00
A patch from HP for better performance. Submitted by Kevin Steves <ks@hp.se> 3 months ago...
2000-09-18 04:04:42 +08:00
# More attempts at unified 10.X and 11.X targets for HP C compiler.
#
# Chris Ruemmler <ruemmler@cup.hp.com>
# Kevin Steves <ks@hp.se>
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"hpux-parisc-cc","cc:+O3 +Optrs_strongly_typed -Ae +ESlit -DB_ENDIAN -DBN_DIV2W -DMD32_XARRAY::-D_REENTRANT::-Wl,+s -ldld:MD2_CHAR RC4_INDEX RC4_CHAR DES_UNROLL DES_RISC1 DES_INT:${no_asm}:dl:hpux-shared:+Z:-b:.sl.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"hpux-parisc1_0-cc","cc:+DAportable +O3 +Optrs_strongly_typed -Ae +ESlit -DB_ENDIAN -DMD32_XARRAY::-D_REENTRANT::-Wl,+s -ldld:MD2_CHAR RC4_INDEX RC4_CHAR DES_UNROLL DES_RISC1 DES_INT:${no_asm}:dl:hpux-shared:+Z:-b:.sl.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
"hpux-parisc2-cc","cc:+DA2.0 +DS2.0 +O3 +Optrs_strongly_typed -Ae +ESlit -DB_ENDIAN -DMD32_XARRAY -D_REENTRANT::::-Wl,+s -ldld:SIXTY_FOUR_BIT MD2_CHAR RC4_INDEX RC4_CHAR DES_UNROLL DES_RISC1 DES_INT::asm/pa-risc2.o::::::::::dl:hpux-shared:+Z:-b:.sl.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"hpux64-parisc2-cc","cc:+DD64 +O3 +Optrs_strongly_typed -Ae +ESlit -DB_ENDIAN -DMD32_XARRAY -D_REENTRANT::::-ldl:SIXTY_FOUR_BIT_LONG MD2_CHAR RC4_INDEX RC4_CHAR DES_UNROLL DES_RISC1 DES_INT::asm/pa-risc2W.o::::::::::dlfcn:hpux-shared:+Z:+DD64 -b:.sl.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
A patch from HP for better performance. Submitted by Kevin Steves <ks@hp.se> 3 months ago...
2000-09-18 04:04:42 +08:00
HP/UX PA-RISC 2 targets update.
2004-01-30 06:16:08 +08:00
# HP/UX IA-64 targets
Engage RC4 IA-64 assembler module.
2004-11-26 23:12:17 +08:00
"hpux-ia64-cc","cc:-Ae +DD32 +O2 +Olit=all -z -DB_ENDIAN -D_REENTRANT::::-ldl:SIXTY_FOUR_BIT MD2_CHAR RC4_INDEX DES_UNROLL DES_RISC1 DES_INT:${ia64_asm}:dlfcn:hpux-shared:+Z:+DD32 -b:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
HP/UX PA-RISC 2 targets update.
2004-01-30 06:16:08 +08:00
# Frank Geurts <frank.geurts@nl.abnamro.com> has patiently assisted with
# with debugging of the following config.
Engage RC4 IA-64 assembler module.
2004-11-26 23:12:17 +08:00
"hpux64-ia64-cc","cc:-Ae +DD64 +O3 +Olit=all -z -DB_ENDIAN -D_REENTRANT::::-ldl:SIXTY_FOUR_BIT_LONG MD2_CHAR RC4_INDEX DES_UNROLL DES_RISC1 DES_INT:${ia64_asm}:dlfcn:hpux-shared:+Z:+DD64 -b:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
# GCC builds...
Engage RC4 IA-64 assembler module.
2004-11-26 23:12:17 +08:00
"hpux-ia64-gcc","gcc:-O3 -DB_ENDIAN -D_REENTRANT::::-ldl:SIXTY_FOUR_BIT MD2_CHAR RC4_INDEX DES_UNROLL DES_RISC1 DES_INT::asm/ia64.o::asm/aes-ia64.o:::asm/sha256-ia64.o asm/sha512-ia64.o::asm/rc4-ia64.o:::dlfcn:hpux-shared:-fpic:-shared:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"hpux64-ia64-gcc","gcc:-mlp64 -O3 -DB_ENDIAN -D_REENTRANT::::-ldl:SIXTY_FOUR_BIT_LONG MD2_CHAR RC4_INDEX DES_UNROLL DES_RISC1 DES_INT:${ia64_asm}:dlfcn:hpux-shared:-fpic:-mlp64 -shared:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
# Legacy HPUX 9.X configs...
"hpux-cc", "cc:-DB_ENDIAN -DBN_DIV2W -DMD32_XARRAY -Ae +ESlit +O2 -z::(unknown)::-Wl,+s -ldld:DES_PTR DES_UNROLL DES_RISC1:${no_asm}:dl:hpux-shared:+Z:-b:.sl.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"hpux-gcc", "gcc:-DB_ENDIAN -DBN_DIV2W -O3::(unknown)::-Wl,+s -ldld:DES_PTR DES_UNROLL DES_RISC1:${no_asm}:dl:hpux-shared:-fPIC:-shared:.sl.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Portability patch for HP MPE/iX. Submitted by Mark Bixby <mark_bixby@hp.com>
2000-09-21 13:42:01 +08:00
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
#### HP MPE/iX http://jazz.external.hp.com/src/openssl/
"MPE/iX-gcc", "gcc:-D_ENDIAN -DBN_DIV2W -O3 -D_POSIX_SOURCE -D_SOCKET_SOURCE -I/SYSLOG/PUB::(unknown):MPE:-L/SYSLOG/PUB -lsyslog -lsocket -lcurses:BN_LLONG DES_PTR DES_UNROLL DES_RISC1:::",
Add support for Linux on HP/PA. Submitted by "Bryan W. Headley" <bheadley@interaccess.com>
2002-01-02 18:00:22 +08:00
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
# DEC Alpha OSF/1/Tru64 targets.
Apply the Tru64 patch from Tim Mooney <mooney@dogbert.cc.ndsu.NoDak.edu> His comments are: 1) Changes all references for `True64' to be `Tru64', which is the correct spelling for the OS name. 2) Makes `alpha-cc' be the same as `alpha164-cc', and adds an `alphaold-cc' entry that is the same as the previous `alpha-cc'. The reason is that most people these days are using the newer compiler, so it should be the default. 3) Adds a bit of commentary to Configure, regarding the name changes of the OS over the years, so it's not so confusing to people that haven't been with the OS for a while. 4) Adds an `alpha-cc-rpath' target (which is *not* selected automatically by Configure under any circumstance) that builds an RPATH into the shared libraries. This is explained in the comment in Configure. It's very very useful for people that want it, and people that don't want it just shouldn't choose that target. 5) Adds the `-pthread' flag as the best way to get POSIX thread support from the newer compiler. 6) Updates the Makefile targets, so that when the `alpha164-cc', `alpha-cc', or `alpha-cc-rpath' target is what Configure is set to use, it uses a Makefile target that includes the `-msym' option when building the shared library. This is a performance enhancement. 7) Updates `config' so that if it detects you're running version 4 or 5 of the OS, it automatically selects `alpha-cc', but uses `alphaold-cc' for versions 1-3 of the OS. 8) Updates the comment in opensslv.h, fixing both the OS name typo and adding a reference to IRIX 6.x, since the shared library semantics are virtually identical there.
2001-08-10 23:26:21 +08:00
#
# "What's in a name? That which we call a rose
# By any other word would smell as sweet."
#
# - William Shakespeare, "Romeo & Juliet", Act II, scene II.
#
# There's also an alternate target available (which `config' will never
# select) called alpha-cc-rpath. This target builds an RPATH into the
# shared libraries, which is very convenient on Tru64 since binaries
# linked against that shared library will automatically inherit that RPATH,
# and hence know where to look for the openssl libraries, even if they're in
# an odd place.
#
Import of old SSLeay release: SSLeay 0.9.1b (unreleased)
1998-12-21 19:00:56 +08:00
# For gcc, the following gave a %50 speedup on a 164 over the 'DES_INT' version
Apply the Tru64 patch from Tim Mooney <mooney@dogbert.cc.ndsu.NoDak.edu> His comments are: 1) Changes all references for `True64' to be `Tru64', which is the correct spelling for the OS name. 2) Makes `alpha-cc' be the same as `alpha164-cc', and adds an `alphaold-cc' entry that is the same as the previous `alpha-cc'. The reason is that most people these days are using the newer compiler, so it should be the default. 3) Adds a bit of commentary to Configure, regarding the name changes of the OS over the years, so it's not so confusing to people that haven't been with the OS for a while. 4) Adds an `alpha-cc-rpath' target (which is *not* selected automatically by Configure under any circumstance) that builds an RPATH into the shared libraries. This is explained in the comment in Configure. It's very very useful for people that want it, and people that don't want it just shouldn't choose that target. 5) Adds the `-pthread' flag as the best way to get POSIX thread support from the newer compiler. 6) Updates the Makefile targets, so that when the `alpha164-cc', `alpha-cc', or `alpha-cc-rpath' target is what Configure is set to use, it uses a Makefile target that includes the `-msym' option when building the shared library. This is a performance enhancement. 7) Updates `config' so that if it detects you're running version 4 or 5 of the OS, it automatically selects `alpha-cc', but uses `alphaold-cc' for versions 1-3 of the OS. 8) Updates the comment in opensslv.h, fixing both the OS name typo and adding a reference to IRIX 6.x, since the shared library semantics are virtually identical there.
2001-08-10 23:26:21 +08:00
#
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"osf1-alpha-gcc", "gcc:-O3::(unknown):::SIXTY_FOUR_BIT_LONG RC4_CHUNK DES_UNROLL DES_RISC1:${no_asm}:dlfcn:alpha-osf1-shared:::.so",
Type in Configure and TABLE update.
2004-07-19 00:22:53 +08:00
"ofs1-alpha-cc", "cc:-std1 -tune host -O4 -readonly_strings::(unknown):::SIXTY_FOUR_BIT_LONG RC4_CHUNK:${no_asm}:dlfcn:alpha-osf1-shared:::.so",
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"tru64-alpha-cc", "cc:-std1 -tune host -fast -readonly_strings::-pthread:::SIXTY_FOUR_BIT_LONG RC4_CHUNK:${no_asm}:dlfcn:tru64-shared:::.so",
"alpha-cc-rpath", "cc:-std1 -tune host -fast -readonly_strings::-pthread:::SIXTY_FOUR_BIT_LONG RC4_CHUNK:${no_asm}:dlfcn:tru64-shared-rpath:::.so",
BSD alpha config. Submitted by: Rich Bud <rbud@fortean.com>
1999-05-19 00:04:03 +08:00
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
####
#### Variety of LINUX:-)
####
#### IA-32 targets...
"linux-ia32-icc", "icc:-DL_ENDIAN -DTERMIO -O2 -no_cpprt::-D_REENTRANT::-ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn:linux-shared:-KPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"linux-elf", "gcc:-DL_ENDIAN -DTERMIO -O3 -fomit-frame-pointer -m486 -Wall::-D_REENTRANT::-ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"linux-pentium", "gcc:-DL_ENDIAN -DTERMIO -O3 -fomit-frame-pointer -mcpu=pentium -Wall::-D_REENTRANT::-ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"linux-ppro", "gcc:-DL_ENDIAN -DTERMIO -O3 -fomit-frame-pointer -mcpu=pentiumpro -Wall::-D_REENTRANT::-ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"linux-k6", "gcc:-DL_ENDIAN -DTERMIO -O3 -fomit-frame-pointer -mcpu=k6 -Wall::-D_REENTRANT::-ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"debug-linux-pentium","gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DCRYPTO_MDEBUG -DL_ENDIAN -DTERMIO -g -mcpu=pentium -Wall::-D_REENTRANT::-ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn",
"debug-linux-ppro","gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DCRYPTO_MDEBUG -DL_ENDIAN -DTERMIO -g -mcpu=pentiumpro -Wall::-D_REENTRANT::-ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn",
"debug-linux-elf","gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DCRYPTO_MDEBUG -DL_ENDIAN -DTERMIO -g -m486 -Wall::-D_REENTRANT::-lefence -ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"debug-linux-elf-noefence","gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DCRYPTO_MDEBUG -DL_ENDIAN -DTERMIO -g -m486 -Wall::-D_REENTRANT::-ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"linux-aout", "gcc:-DL_ENDIAN -DTERMIO -O3 -fomit-frame-pointer -m486 -Wall::(unknown):::BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_out_asm}",
#
"linux-mipsel", "gcc:-DL_ENDIAN -DTERMIO -O3 -fomit-frame-pointer -Wall::-D_REENTRANT::-ldl:BN_LLONG RC2_CHAR RC4_INDEX DES_INT DES_UNROLL DES_RISC2:${no_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"linux-mips", "gcc:-DB_ENDIAN -DTERMIO -O3 -fomit-frame-pointer -Wall::-D_REENTRANT::-ldl:BN_LLONG RC2_CHAR RC4_INDEX DES_INT DES_UNROLL DES_RISC2:${no_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
"linux-ppc", "gcc:-DB_ENDIAN -DTERMIO -O3 -Wall::-D_REENTRANT::-ldl:BN_LLONG RC4_CHAR RC4_CHUNK DES_RISC1 DES_UNROLL::asm/linux_ppc32.o::::::::::dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Refine PowerPC platform support.
2004-12-20 21:44:34 +08:00
# -bpowerpc64-linux is transient option, -m64 should be the one to use...
"linux-ppc64", "gcc:-bpowerpc64-linux -DB_ENDIAN -DTERMIO -O3 -Wall::-D_REENTRANT::-ldl:SIXTY_FOUR_BIT_LONG RC4_CHAR RC4_CHUNK DES_RISC1 DES_UNROLL::asm/linux_ppc64.o::::::::::dlfcn:linux-shared:-fPIC:-bpowerpc64-linux:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"linux-m68k", "gcc:-DB_ENDIAN -DTERMIO -O2 -fomit-frame-pointer -Wall::-D_REENTRANT:::BN_LLONG::",
"linux-s390", "gcc:-DB_ENDIAN -DTERMIO -DNO_ASM -O3 -fomit-frame-pointer -Wall::-D_REENTRANT::-ldl:BN_LLONG:${no_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"linux-s390x", "gcc:-DB_ENDIAN -DTERMIO -DNO_ASM -O3 -fomit-frame-pointer -Wall::-D_REENTRANT::-ldl:SIXTY_FOUR_BIT_LONG:${no_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Engage RC4 IA-64 assembler module.
2004-11-26 23:12:17 +08:00
"linux-ia64", "gcc:-DL_ENDIAN -DTERMIO -O3 -Wall::-D_REENTRANT::-ldl:SIXTY_FOUR_BIT_LONG RC4_CHUNK:${ia64_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"linux-ia64-ecc","ecc:-DL_ENDIAN -DTERMIO -O2 -Wall -no_cpprt::-D_REENTRANT::-ldl:SIXTY_FOUR_BIT_LONG RC4_CHUNK:${ia64_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
RC4 tune-up for Intel P4 core, both 32- and 64-bit ones. As it's apparently impossible to compose blended code with would perform satisfactory on all x86 and x86_64 cores, an extra RC4_CHAR code-path is introduced and P4 core is detected at run-time. This way we keep original performance on non-P4 implementations and turbo-charge P4 performance by factor of 2.8x (on 32-bit core).
2004-11-21 18:36:25 +08:00
"linux-x86_64", "gcc:-m64 -DL_ENDIAN -DTERMIO -O3 -Wall -DMD32_REG_T=int::-D_REENTRANT::-ldl:SIXTY_FOUR_BIT_LONG RC4_CHUNK BF_PTR2 DES_INT DES_UNROLL:amd64cpuid.o:asm/x86_64-gcc.o:::::::asm/rc4-amd64.o:::dlfcn:linux-shared:-fPIC:-m64:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"linux-elf-arm","gcc:-DL_ENDIAN -DTERMIO -O3 -fomit-frame-pointer -Wall::-D_REENTRANT::-ldl:BN_LLONG:${no_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"linux-parisc", "gcc:-DB_ENDIAN -DTERMIO -O3 -fomit-frame-pointer -Wall -DBN_DIV2W::-D_REENTRANT:::BN_LLONG RC4_CHAR RC4_CHUNK DES_UNROLL BF_PTR::",
#### SPARC Linux setups
"linux-sparcv7","gcc:-DB_ENDIAN -DTERMIO -O3 -fomit-frame-pointer -Wall::-D_REENTRANT:::BN_LLONG RC4_CHAR RC4_CHUNK DES_UNROLL BF_PTR::",
# Ray Miller <ray.miller@computing-services.oxford.ac.uk> has patiently
# assisted with debugging of following two configs.
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
"linux-sparcv8","gcc:-mv8 -DB_ENDIAN -DTERMIO -O3 -fomit-frame-pointer -Wall -DBN_DIV2W::-D_REENTRANT::-ldl:BN_LLONG RC4_CHAR RC4_CHUNK DES_UNROLL BF_PTR::asm/sparcv8.o:asm/des_enc-sparc.o fcrypt_b.o:::::::::dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
# it's a real mess with -mcpu=ultrasparc option under Linux, but
# -Wa,-Av8plus should do the trick no matter what.
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
"linux-sparcv9","gcc:-m32 -mcpu=ultrasparc -DB_ENDIAN -DTERMIO -O3 -fomit-frame-pointer -Wall -Wa,-Av8plus -DBN_DIV2W::-D_REENTRANT:ULTRASPARC:-ldl:BN_LLONG RC4_CHAR RC4_CHUNK DES_UNROLL BF_PTR::asm/sparcv8plus.o:asm/des_enc-sparc.o fcrypt_b.o:::asm/md5-sparcv8plus.o::::::dlfcn:linux-shared:-fPIC:-m32:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
# GCC 3.1 is a requirement
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
"linux64-sparcv9","gcc:-m64 -mcpu=ultrasparc -DB_ENDIAN -DTERMIO -O3 -fomit-frame-pointer -Wall::-D_REENTRANT:ULTRASPARC:-ldl:SIXTY_FOUR_BIT_LONG RC4_CHAR RC4_CHUNK DES_UNROLL BF_PTR:::asm/des_enc-sparc.o fcrypt_b.o:::asm/md5-sparcv9.o::::::dlfcn:linux-shared:-fPIC:-m64:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Enhanced support for Alpha Linux. See CHANGES for details.
2000-01-03 04:46:58 +08:00
#### Alpha Linux with GNU C and Compaq C setups
Avoid GNU C assembler templates under Solaris x86.
2000-02-16 01:31:12 +08:00
# Special notes:
# - linux-alpha+bwx-gcc is ment to be used from ./config only. If you
# ought to run './Configure linux-alpha+bwx-gcc' manually, do
# complement the command line with -mcpu=ev56, -mcpu=ev6 or whatever
# which is appropriate.
# - If you use ccc keep in mind that -fast implies -arch host and the
# compiler is free to issue instructions which gonna make elder CPU
# choke. If you wish to build "blended" toolkit, add -arch generic
# *after* -fast and invoke './Configure linux-alpha-ccc' manually.
#
# <appro@fy.chalmers.se>
#
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"linux-alpha-gcc","gcc:-O3 -DL_ENDIAN -DTERMIO::-D_REENTRANT::-ldl:SIXTY_FOUR_BIT_LONG RC4_CHUNK DES_RISC1 DES_UNROLL:${no_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"linux-alpha+bwx-gcc","gcc:-O3 -DL_ENDIAN -DTERMIO::-D_REENTRANT::-ldl:SIXTY_FOUR_BIT_LONG RC4_CHAR RC4_CHUNK DES_RISC1 DES_UNROLL:${no_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"linux-alpha-ccc","ccc:-fast -readonly_strings -DL_ENDIAN -DTERMIO::-D_REENTRANT:::SIXTY_FOUR_BIT_LONG RC4_CHUNK DES_INT DES_PTR DES_RISC1 DES_UNROLL:${no_asm}",
"linux-alpha+bwx-ccc","ccc:-fast -readonly_strings -DL_ENDIAN -DTERMIO::-D_REENTRANT:::SIXTY_FOUR_BIT_LONG RC4_CHAR RC4_CHUNK DES_INT DES_PTR DES_RISC1 DES_UNROLL:${no_asm}",
#### *BSD
"NetBSD-sparc", "gcc:-DTERMIOS -O3 -fomit-frame-pointer -mv8 -Wall -DB_ENDIAN::(unknown):::BN_LLONG MD2_CHAR RC4_INDEX DES_UNROLL:${no_asm}:dlfcn:bsd-gcc-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"NetBSD-m68", "gcc:-DTERMIOS -O3 -fomit-frame-pointer -Wall -DB_ENDIAN::(unknown):::BN_LLONG MD2_CHAR RC4_INDEX DES_UNROLL:${no_asm}:dlfcn:bsd-gcc-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"NetBSD-x86", "gcc:-DTERMIOS -O3 -fomit-frame-pointer -m486 -Wall::(unknown):::BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${no_asm}:dlfcn:bsd-gcc-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"FreeBSD-elf", "gcc:-DTERMIOS -DL_ENDIAN -fomit-frame-pointer -O3 -m486 -Wall::-pthread -D_REENTRANT -D_THREAD_SAFE -D_THREADSAFE:::BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn:bsd-gcc-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"FreeBSD-sparc64","gcc:-DB_ENDIAN -DTERMIOS -O3 -fomit-frame-pointer::-pthread -D_REENTRANT -D_THREAD_SAFE -D_THREADSAFE:::SIXTY_FOUR_BIT_LONG DES_INT DES_PTR DES_RISC2 BF_PTR:${no_asm}:dlfcn:bsd-gcc-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Engage RC4 IA-64 assembler module.
2004-11-26 23:12:17 +08:00
"FreeBSD-ia64", "gcc:-DL_ENDIAN -DTERMIOS -O -fomit-frame-pointer::(unknown):::SIXTY_FOUR_BIT_LONG RC4_CHUNK:${ia64_asm}:dlfcn:bsd-gcc-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"FreeBSD-alpha","gcc:-DTERMIOS -O::(unknown):::SIXTY_FOUR_BIT_LONG RC4_CHUNK DES_INT DES_PTR DES_RISC2:${no_asm}:dlfcn:bsd-gcc-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"FreeBSD", "gcc:-DTERMIOS -DL_ENDIAN -fomit-frame-pointer -O3 -m486 -Wall::(unknown):::BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_out_asm}",
Enhanced support for Alpha Linux. See CHANGES for details.
2000-01-03 04:46:58 +08:00
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
# Some OpenBSD from Bob Beck <beck@obtuse.com>
"OpenBSD", "gcc:-DTERMIOS -O3 -fomit-frame-pointer::(unknown):::BN_LLONG RC2_CHAR RC4_INDEX DES_INT DES_UNROLL:${no_asm}:dlfcn:bsd-gcc-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"OpenBSD-alpha", "gcc:-DTERMIOS -O3 -fomit-frame-pointer::(unknown):::SIXTY_FOUR_BIT_LONG DES_INT DES_PTR DES_RISC2:${no_asm}:dlfcn:bsd-gcc-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Proper support for OpenBSD-i386 shared build, including assember modules! "Proper" means "compiles and passes test." Versioning is broken (I think).
2004-08-30 05:36:37 +08:00
"OpenBSD-i386", "gcc:-DL_ENDIAN -DTERMIOS -O3 -fomit-frame-pointer::(unknown):::BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_out_asm}:dlfcn:bsd-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"OpenBSD-m68k", "gcc:-DTERMIOS -O3 -fomit-frame-pointer::(unknown):::BN_LLONG RC2_CHAR RC4_INDEX DES_INT DES_UNROLL:${no_asm}:dlfcn:bsd-gcc-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"OpenBSD-m88k", "gcc:-DTERMIOS -O3 -fomit-frame-pointer::(unknown):::BN_LLONG RC2_CHAR RC4_INDEX DES_INT DES_UNROLL:${no_asm}:dlfcn:bsd-gcc-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"OpenBSD-mips", "gcc:-DL_ENDIAN -DTERMIOS -O3 -fomit-frame-pointer::(unknown):::BN_LLONG RC2_CHAR RC4_INDEX DES_INT DES_UNROLL DES_RISC2:${no_asm}:dlfcn:bsd-gcc-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"OpenBSD-powerpc", "gcc:-DTERMIOS -O3 -fomit-frame-pointer::(unknown):::BN_LLONG RC2_CHAR RC4_INDEX DES_INT DES_UNROLL:${no_asm}:dlfcn:bsd-gcc-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
"OpenBSD-sparc", "gcc:-DTERMIOS -O3 -fomit-frame-pointer::(unknown):::BN_LLONG RC2_CHAR RC4_INDEX DES_INT DES_UNROLL::asm/sparcv8.o:asm/des_enc-sparc.o fcrypt_b.o:::::::::dlfcn:bsd-gcc-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"OpenBSD-sparc64", "gcc:-DB_ENDIAN -DTERMIOS -O3 -fomit-frame-pointer::(unknown):::SIXTY_FOUR_BIT_LONG DES_INT DES_PTR DES_RISC2 BF_PTR:::asm/des_enc-sparc.o fcrypt_b.o:::::::::dlfcn:bsd-gcc-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"OpenBSD-vax", "gcc:-DL_ENDIAN -DTERMIOS -O3 -fomit-frame-pointer::(unknown):::BN_LLONG RC2_CHAR RC4_INDEX DES_INT DES_UNROLL:${no_asm}:dlfcn:bsd-gcc-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"OpenBSD-hppa", "gcc:-DTERMIOS -O3 -fomit-frame-pointer::(unknown):::BN_LLONG RC2_CHAR RC4_INDEX DES_UNROLL:${no_asm}:dlfcn:bsd-gcc-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
On Solaris64, cc needs the flag -xarch=v9 when linking shared libraries. Make a general change to support shared library linking flags in general. Noted by Nick Briggs <briggs@parc.xerox.com>
2002-01-02 18:30:07 +08:00
"bsdi-elf-gcc", "gcc:-DPERL5 -DL_ENDIAN -fomit-frame-pointer -O3 -m486 -Wall::(unknown)::-ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn:bsd-gcc-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
"nextstep", "cc:-O -Wall:<libc.h>:(unknown):::BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:::",
"nextstep3.3", "cc:-O3 -Wall:<libc.h>:(unknown):::BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:::",
Import of old SSLeay release: SSLeay 0.9.0b
1998-12-21 18:56:39 +08:00
# NCR MP-RAS UNIX ver 02.03.01
ncr-scde target needs -lc89 for strcasecmp() and ftime() (Tim Rice, Martin Megele). PR: 450
2003-01-16 15:54:52 +08:00
"ncr-scde","cc:-O6 -Xa -Hoff=BEHAVED -686 -Hwide -Hiw::(unknown)::-lsocket -lnsl -lc89:${x86_gcc_des} ${x86_gcc_opts}:::",
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
# QNX
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
"qnx4", "cc:-DL_ENDIAN -DTERMIO::(unknown):::${x86_gcc_des} ${x86_gcc_opts}:",
Support for QNX (wrat@jump.net (the wharf rat)).
2001-11-30 17:38:57 +08:00
"qnx6", "cc:-DL_ENDIAN -DTERMIOS::(unknown)::-lsocket:${x86_gcc_des} ${x86_gcc_opts}:",
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
#### SCO/Caldera targets.
Just an extra comment.
2003-01-12 23:50:59 +08:00
#
Finalizing asm support for UnixWare, SCO, OpenUnix... Note that I've replaced #if logic around bn_sub_part_words in bn_mul.c. I rely upon OPENSSL_BN_ASM_PART_WORDS being added by ./Configure script. Would it still work on non-Unix platforms?
2003-01-09 16:42:04 +08:00
# Originally we had like unixware-*, unixware-*-pentium, unixware-*-p6, etc.
# Now we only have blended unixware-* as it's the only one used by ./config.
Just an extra comment.
2003-01-12 23:50:59 +08:00
# If you want to optimize for particular microarchitecture, bypass ./config
Finalizing asm support for UnixWare, SCO, OpenUnix... Note that I've replaced #if logic around bn_sub_part_words in bn_mul.c. I rely upon OPENSSL_BN_ASM_PART_WORDS being added by ./Configure script. Would it still work on non-Unix platforms?
2003-01-09 16:42:04 +08:00
# and './Configure unixware-7 -Kpentium_pro' or whatever appropriate.
Just an extra comment.
2003-01-12 23:50:59 +08:00
# Note that not all targets include assembler support. Mostly because of
# lack of motivation to support out-of-date platforms with out-of-date
# compiler drivers and assemblers. Tim Rice <tim@multitalents.net> has
# patiently assisted to debug most of it.
Finalizing asm support for UnixWare, SCO, OpenUnix... Note that I've replaced #if logic around bn_sub_part_words in bn_mul.c. I rely upon OPENSSL_BN_ASM_PART_WORDS being added by ./Configure script. Would it still work on non-Unix platforms?
2003-01-09 16:42:04 +08:00
#
# UnixWare 2.0x fails destest with -O.
Unixware doesn't have strings.h, so we need to declare strcasecmp() differently. Unixware 2 needs to link with libresolv. PR: 148
2002-07-18 15:47:30 +08:00
"unixware-2.0","cc:-DFILIO_H -DNO_STRINGS_H::-Kthread::-lsocket -lnsl -lresolv -lx:${x86_gcc_des} ${x86_gcc_opts}:::",
"unixware-2.1","cc:-O -DFILIO_H::-Kthread::-lsocket -lnsl -lresolv -lx:${x86_gcc_des} ${x86_gcc_opts}:::",
Finalizing asm support for UnixWare, SCO, OpenUnix... Note that I've replaced #if logic around bn_sub_part_words in bn_mul.c. I rely upon OPENSSL_BN_ASM_PART_WORDS being added by ./Configure script. Would it still work on non-Unix platforms?
2003-01-09 16:42:04 +08:00
"unixware-7","cc:-O -DFILIO_H -Kalloca::-Kthread::-lsocket -lnsl:BN_LLONG MD2_CHAR RC4_INDEX ${x86_gcc_des}:${x86_elf_asm}:dlfcn:svr5-shared:-Kpic::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"unixware-7-gcc","gcc:-DL_ENDIAN -DFILIO_H -O3 -fomit-frame-pointer -m486 -Wall::-D_REENTRANT::-lsocket -lnsl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn:gnu-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"OpenUNIX-8","cc:-O -DFILIO_H -Kalloca::-Kthread::-lsocket -lnsl:BN_LLONG MD2_CHAR RC4_INDEX ${x86_gcc_des}:${x86_elf_asm}:dlfcn:svr5-shared:-Kpic::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
It seems like gcc-drivven shared library building on OpenUnix 8 requires -shared rather than -G.
2003-04-01 18:59:15 +08:00
"OpenUNIX-8-gcc","gcc:-O -DFILIO_H -fomit-frame-pointer::-pthread::-lsocket -lnsl:BN_LLONG MD2_CHAR RC4_INDEX ${x86_gcc_des}:${x86_elf_asm}:dlfcn:svr5-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Just an extra comment.
2003-01-12 23:50:59 +08:00
"sco3-gcc", "gcc:-O3 -fomit-frame-pointer -Dssize_t=int -DNO_SYS_UN_H::(unknown)::-lsocket:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:::", # the SCO assembler doesn't seem to like our assembler files ...
# SCO 5 - Ben Laurie <ben@algroup.co.uk> says the -O breaks the SCO cc.
SCO target missed .so suffix.
2003-01-19 05:57:30 +08:00
"sco5-cc", "cc:-belf::(unknown)::-lsocket -lnsl:${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn:svr3-shared:-Kpic::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
"sco5-gcc", "gcc:-O3 -fomit-frame-pointer::(unknown)::-lsocket -lnsl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn:svr3-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Just an extra comment.
2003-01-12 23:50:59 +08:00
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
#### IBM's AIX.
Refine PowerPC platform support.
2004-12-20 21:44:34 +08:00
"aix3-cc", "cc:-O -DB_ENDIAN -qmaxmem=16384::(unknown):AIX::BN_LLONG RC4_CHAR:::",
"aix-gcc", "gcc:-O -DB_ENDIAN::-D_THREAD_SAFE:AIX::BN_LLONG RC4_CHAR::asm/aix_ppc32.o::::::::::dlfcn:",
Improved PowerPC support. Proper ./config support for ppc targets, especially for AIX. But most important BIGNUM assembler implementation submitted by IBM. Submitted by: Peter Waltenberg <pwalten@au1.ibm.com> Reviewed by: appro
2004-04-28 06:05:50 +08:00
# Below targets assume AIX 5. Idea is to effectively disregard $OBJECT_MODE
# at build time. $OBJECT_MODE is respected at ./config stage!
Refine PowerPC platform support.
2004-12-20 21:44:34 +08:00
"aix-cc", "cc:-q32 -O -DB_ENDIAN -qmaxmem=16384::-qthreaded:AIX::BN_LLONG RC4_CHAR::asm/aix_ppc32.o::::::::::dlfcn:aix-shared::-q32:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)::-X 32",
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
"aix64-cc", "cc:-q64 -O -DB_ENDIAN -qmaxmem=16384::(unknown):AIX::SIXTY_FOUR_BIT_LONG RC4_CHAR::asm/aix_ppc64.o::::::::::dlfcn:aix-shared::-q64:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)::-X 64",
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
Import of old SSLeay release: SSLeay 0.9.0b
1998-12-21 18:56:39 +08:00
#
Rename 'cray-t90-cc' into 'cray-j90'. Add to 'config'.
2002-03-16 00:46:41 +08:00
# Cray T90 and similar (SDSC)
Import of old SSLeay release: SSLeay 0.9.0b
1998-12-21 18:56:39 +08:00
# It's Big-endian, but the algorithms work properly when B_ENDIAN is NOT
# defined. The T90 ints and longs are 8 bytes long, and apparently the
# B_ENDIAN code assumes 4 byte ints. Fortunately, the non-B_ENDIAN and
# non L_ENDIAN code aligns the bytes in each word correctly.
#
# The BIT_FIELD_LIMITS define is to avoid two fatal compiler errors:
#'Taking the address of a bit field is not allowed. '
#'An expression with bit field exists as the operand of "sizeof" '
# (written by Wayne Schroeder <schroede@SDSC.EDU>)
Rename 'cray-t90-cc' into 'cray-j90'. Add to 'config'.
2002-03-16 00:46:41 +08:00
#
# j90 is considered the base machine type for unicos machines,
# so this configuration is now called "cray-j90" ...
"cray-j90", "cc: -DBIT_FIELD_LIMITS -DTERMIOS::(unknown):CRAY::SIXTY_FOUR_BIT_LONG DES_INT:::",
Import of old SSLeay release: SSLeay 0.9.0b
1998-12-21 18:56:39 +08:00
Incorporation of RSEs assembled patches
1998-12-22 23:59:57 +08:00
#
# Cray T3E (Research Center Juelich, beckman@acl.lanl.gov)
#
# The BIT_FIELD_LIMITS define was written for the C90 (it seems). I added
# another use. Basically, the problem is that the T3E uses some bit fields
# for some st_addr stuff, and then sizeof and address-of fails
# I could not use the ams/alpha.o option because the Cray assembler, 'cam'
# did not like it.
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
"cray-t3e", "cc: -DBIT_FIELD_LIMITS -DTERMIOS::(unknown):CRAY::SIXTY_FOUR_BIT_LONG RC4_CHUNK DES_INT:::",
Incorporation of RSEs assembled patches
1998-12-22 23:59:57 +08:00
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
# DGUX, 88100.
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
"dgux-R3-gcc", "gcc:-O3 -fomit-frame-pointer::(unknown):::RC4_INDEX DES_UNROLL:::",
Remove a typo in dgux-R4-gcc.
2001-04-03 16:05:03 +08:00
"dgux-R4-gcc", "gcc:-O3 -fomit-frame-pointer::(unknown)::-lnsl -lsocket:RC4_INDEX DES_UNROLL:::",
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
"dgux-R4-x86-gcc", "gcc:-O3 -fomit-frame-pointer -DL_ENDIAN::(unknown)::-lnsl -lsocket:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}",
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
Detect Siemens platforms. Submitted by: Martin Kraemer <Martin.Kraemer@MchP.Siemens.De>
1999-05-21 01:28:19 +08:00
# Sinix/ReliantUNIX RM400
# NOTE: The CDS++ Compiler up to V2.0Bsomething has the IRIX_CC_BUG optimizer problem. Better use -g */
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"ReliantUNIX","cc:-KPIC -g -DTERMIOS -DB_ENDIAN::-Kthread:SNI:-lsocket -lnsl -lc -L/usr/ucblib -lucb:BN_LLONG DES_PTR DES_RISC2 DES_UNROLL BF_PTR:${no_asm}:dlfcn:reliantunix-shared:::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)",
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
"SINIX","cc:-O::(unknown):SNI:-lsocket -lnsl -lc -L/usr/ucblib -lucb:RC4_INDEX RC4_CHAR:::",
"SINIX-N","/usr/ucb/cc:-O2 -misaligned::(unknown)::-lucb:RC4_INDEX RC4_CHAR:::",
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
Support the EBCDIC character set and BS2000/OSD-POSIX (work in progress). Submitted by: Martin Kraemer <Martin.Kraemer@MchP.Siemens.De>
1999-06-05 05:35:58 +08:00
# SIEMENS BS2000/OSD: an EBCDIC-based mainframe
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
"BS2000-OSD","c89:-O -XLLML -XLLMK -XL -DB_ENDIAN -DTERMIOS -DCHARSET_EBCDIC::(unknown)::-lsocket -lnsl:THIRTY_TWO_BIT DES_PTR DES_UNROLL MD2_CHAR RC4_INDEX RC4_CHAR BF_PTR:::",
Support the EBCDIC character set and BS2000/OSD-POSIX (work in progress). Submitted by: Martin Kraemer <Martin.Kraemer@MchP.Siemens.De>
1999-06-05 05:35:58 +08:00
OS/390 support Submitted by: Richard Shapiro <rshapiro@abinitio.com>
2001-11-22 19:09:42 +08:00
# OS/390 Unix an EBCDIC-based Unix system on IBM mainframe
# You need to compile using the c89.sh wrapper in the tools directory, because the
# IBM compiler does not like the -L switch after any object modules.
#
"OS390-Unix","c89.sh:-O -DB_ENDIAN -DCHARSET_EBCDIC -DNO_SYS_PARAM_H -D_ALL_SOURCE::(unknown):::THIRTY_TWO_BIT DES_PTR DES_UNROLL MD2_CHAR RC4_INDEX RC4_CHAR BF_PTR:::",
Win64 placeholder targets. This is merely naming suggestion. As we know Win64 comes in two flavors, IA-64/Itanium and AMD64/Opteron. The suggestion is to refer to former as WIN64I and latter - WIN64A
2004-07-24 03:18:20 +08:00
# Win64 targets, WIN64I denotes IA-64 and WIN64A - AMD64
"VC-WIN64I","cl::::WIN64I::SIXTY_FOUR_BIT EXPORT_VAR_AS_FN:${no_asm}:win32",
"VC-WIN64A","cl::::WIN64A::SIXTY_FOUR_BIT EXPORT_VAR_AS_FN:${no_asm}:win32",
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
Win64 placeholder targets. This is merely naming suggestion. As we know Win64 comes in two flavors, IA-64/Itanium and AMD64/Opteron. The suggestion is to refer to former as WIN64I and latter - WIN64A
2004-07-24 03:18:20 +08:00
# Visual C targets
VC-NT was taken away by mistake, putting it back...
2004-07-24 04:46:41 +08:00
"VC-NT","cl::::WINNT::BN_LLONG RC4_INDEX EXPORT_VAR_AS_FN ${x86_gcc_opts}:${no_asm}:win32",
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"VC-CE","cl::::WINCE::BN_LLONG RC4_INDEX EXPORT_VAR_AS_FN ${x86_gcc_opts}:${no_asm}:win32",
"VC-WIN32","cl::::WIN32::BN_LLONG RC4_INDEX EXPORT_VAR_AS_FN ${x86_gcc_opts}:${no_asm}:win32",
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
"VC-WIN16","cl:::(unknown):WIN16::MD2_CHAR DES_UNROLL DES_PTR RC4_INDEX THIRTY_TWO_BIT:::",
"VC-W31-16","cl:::(unknown):WIN16::BN_LLONG MD2_CHAR DES_UNROLL DES_PTR RC4_INDEX SIXTEEN_BIT:::",
"VC-W31-32","cl::::WIN16::BN_LLONG MD2_CHAR DES_UNROLL DES_PTR RC4_INDEX THIRTY_TWO_BIT:::",
"VC-MSDOS","cl:::(unknown):MSDOS::BN_LLONG MD2_CHAR DES_UNROLL DES_PTR RC4_INDEX SIXTEEN_BIT:::",
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
# Borland C++ 4.5
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"BC-32","bcc32::::WIN32::BN_LLONG DES_PTR RC4_INDEX EXPORT_VAR_AS_FN:${no_asm}:win32",
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
"BC-16","bcc:::(unknown):WIN16::BN_LLONG DES_PTR RC4_INDEX SIXTEEN_BIT:::",
Import of old SSLeay release: SSLeay 0.9.1b (unreleased)
1998-12-21 19:00:56 +08:00
more mingw related cleanups.
2003-02-23 02:00:14 +08:00
# MinGW
Clean-up GAS targets: get rid of "cpp" stuff and replace it with "purified" COFF and a.out targets [similar to ELF targets]. You might notice some rudementary support for shared mingw builds under cygwin. It works (it produces cryptoeay32.dll and ssleay32.dll with everything exported by name), but it's primarily for testing/debugging purposes, at least for now...
2004-08-02 01:33:58 +08:00
"mingw", "gcc:-mno-cygwin -DL_ENDIAN -fomit-frame-pointer -O3 -march=i486 -Wall:::MINGW32:-mno-cygwin -lwsock32 -lgdi32:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_coff_asm}:win32:cygwin-shared:-D_DLL:-shared:.dll",
Add CygWin32 platform information to Configure script. Submitted by: Alan Batie <batie@aahz.jf.intel.com>
1999-01-30 19:50:48 +08:00
Tentatively add support for UWIN, a Unix-like environment on top of Windows. PR: 62
2002-06-14 05:44:23 +08:00
# UWIN
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"UWIN", "cc:-DTERMIOS -DL_ENDIAN -O -Wall:::UWIN::BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${no_asm}:win32",
Tentatively add support for UWIN, a Unix-like environment on top of Windows. PR: 62
2002-06-14 05:44:23 +08:00
At Corinna Vinschen's request, change CygWin32 to Cygwin
2002-02-14 20:28:24 +08:00
# Cygwin
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"Cygwin-pre1.3", "gcc:-DTERMIOS -DL_ENDIAN -fomit-frame-pointer -O3 -m486 -Wall::(unknown):CYGWIN32::BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${no_asm}:win32",
Clean-up GAS targets: get rid of "cpp" stuff and replace it with "purified" COFF and a.out targets [similar to ELF targets]. You might notice some rudementary support for shared mingw builds under cygwin. It works (it produces cryptoeay32.dll and ssleay32.dll with everything exported by name), but it's primarily for testing/debugging purposes, at least for now...
2004-08-02 01:33:58 +08:00
"Cygwin", "gcc:-DTERMIOS -DL_ENDIAN -fomit-frame-pointer -O3 -march=i486 -Wall:::CYGWIN32::BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_coff_asm}:win32:cygwin-shared:-D_DLL:-shared:.dll",
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"debug-Cygwin", "gcc:-DTERMIOS -DL_ENDIAN -march=i486 -Wall -DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DCRYPTO_MDEBUG -DOPENSSL_NO_ASM -g -Wformat -Wshadow -Wmissing-prototypes -Wmissing-declarations -Werror:::CYGWIN32:::${no_asm}:win32:cygwin-shared:::.dll",
fix Cygwin (remove extra colons)
2002-05-25 04:39:31 +08:00
Netware-specific changes, PR: 780 Submitted by: Verdon Walker <VWalker@novell.com> Reviewed by: Richard Levitte
2003-11-28 21:10:58 +08:00
# NetWare from David Ward (dsward@novell.com) - requires MetroWerks NLM development tools
# netware-clib => legacy CLib c-runtime support
"netware-clib", "mwccnlm:::::${x86_gcc_opts}:::",
# netware-libc => LibC/NKS support
"netware-libc", "mwccnlm:::::BN_LLONG ${x86_gcc_opts}:::",
NetWare fixes provided by Verdon Walker for OpenSSL 0.9.8-dev. The changes have been mailed to <crypt@bis.doc.gov> as well. PR: 903
2004-06-28 19:55:28 +08:00
"netware-libc-gcc", "i586-netware-gcc:-nostdinc -I/ndk/libc/include -I/ndk/libc/include/winsock -DL_ENDIAN -DNETWARE_LIBC -DOPENSSL_SYSNAME_NETWARE -DTERMIO -O2 -Wall::::${x86_gcc_opts}:::",
Netware-specific changes, PR: 780 Submitted by: Verdon Walker <VWalker@novell.com> Reviewed by: Richard Levitte
2003-11-28 21:10:58 +08:00
Add support for DJGPP. PR: 75
2002-06-14 04:42:35 +08:00
# DJGPP
DJGPP update. PR: 989 Submitted by: Doug Kaufman
2005-01-04 18:28:38 +08:00
"DJGPP", "gcc:-I/dev/env/WATT_ROOT/inc -DTERMIOS -DL_ENDIAN -fomit-frame-pointer -O2 -Wall -DDEVRANDOM=\"/dev/urandom\\x24\":::MSDOS:-L/dev/env/WATT_ROOT/lib -lwatt:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_out_asm}:",
Add support for DJGPP. PR: 75
2002-06-14 04:42:35 +08:00
Defunct assembler files removed; various cleanups. New Ultrix and Alpha entries submitted by Bernhard Simon <simon@zid.tuwien.ac.at>.
1999-04-19 21:54:11 +08:00
# Ultrix from Bernhard Simon <simon@zid.tuwien.ac.at>
Update -Olimit setting. Submitted by: Bernhard Simon <simon@zid.tuwien.ac.at> Reviewed by: PR:
2002-12-11 02:48:14 +08:00
"ultrix-cc","cc:-std1 -O -Olimit 2500 -DL_ENDIAN::(unknown):::::::",
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
"ultrix-gcc","gcc:-O3 -DL_ENDIAN::(unknown):::::::",
Defunct assembler files removed; various cleanups. New Ultrix and Alpha entries submitted by Bernhard Simon <simon@zid.tuwien.ac.at>.
1999-04-19 21:54:11 +08:00
# K&R C is no longer supported; you need gcc on old Ultrix installations
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
##"ultrix","cc:-O2 -DNOPROTO -DNOCONST -DL_ENDIAN::(unknown):::::::",
Import of old SSLeay release: SSLeay 0.9.1b (unreleased)
1998-12-21 19:00:56 +08:00
Recognise Darwin as well. For now, have it do exactly the same thing as for Rhapsody
2000-12-01 09:11:54 +08:00
##### MacOS X (a.k.a. Rhapsody or Darwin) setup
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"rhapsody-ppc-cc","cc:-O3 -DB_ENDIAN::(unknown):MACOSX_RHAPSODY::BN_LLONG RC4_CHAR RC4_CHUNK DES_UNROLL BF_PTR:${no_asm}::",
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
"darwin-ppc-cc","cc:-O3 -fno-common -DB_ENDIAN::-D_REENTRANT:MACOSX::BN_LLONG RC4_CHAR RC4_CHUNK DES_UNROLL BF_PTR::asm/osx_ppc32.o:::::::::::darwin-shared:-fPIC::.\$(SHLIB_MAJOR).\$(SHLIB_MINOR).dylib",
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"darwin-i386-cc","cc:-O3 -fomit-frame-pointer -fno-common -DB_ENDIAN::-D_REENTRANT:MACOSX::BN_LLONG RC4_CHAR RC4_CHUNK DES_UNROLL BF_PTR:${no_asm}::darwin-shared:-fPIC::.\$(SHLIB_MAJOR).\$(SHLIB_MINOR).dylib",
Support for MacOS X (Rhapsody) is added. Also get rid of volatile qualifier in asm definitions as it prevents compiler from moving the instruction(s) during optimization pass.
2000-02-06 19:15:20 +08:00
Add a configuration target for A/UX. PR: 271
2002-10-11 22:39:03 +08:00
##### A/UX
"aux3-gcc","gcc:-O2 -DTERMIO::(unknown):AUX:-lbsd:RC4_CHAR RC4_CHUNK DES_UNROLL BF_PTR:::",
Add a configuration for Sony News 4. Submitted by NAKAJI Hiroyuki <nakaji@tutrp.tut.ac.jp>
2000-09-14 20:48:48 +08:00
##### Sony NEWS-OS 4.x
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
"newsos4-gcc","gcc:-O -DB_ENDIAN::(unknown):NEWS4:-lmld -liberty:BN_LLONG RC4_CHAR RC4_CHUNK DES_PTR DES_RISC1 DES_UNROLL BF_PTR::::",
Add a configuration for Sony News 4. Submitted by NAKAJI Hiroyuki <nakaji@tutrp.tut.ac.jp>
2000-09-14 20:48:48 +08:00
Add configuration for GNU Hurd.
2001-01-11 20:58:37 +08:00
##### GNU Hurd
zlib default was broken on most platforms. -mcpu=i486 is new as of gcc 2.95, and if you have that, you wouldn't want to optimize for 486 anyway.
2001-09-01 12:18:00 +08:00
"hurd-x86", "gcc:-DL_ENDIAN -DTERMIOS -O3 -fomit-frame-pointer -m486 -Wall::-D_REENTRANT::-ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn:linux-shared:-fPIC",
Add configuration for GNU Hurd.
2001-01-11 20:58:37 +08:00
Changes to have OpenSSL compile on OS/2. Contributed by "Brian Havard" <brianh@kheldar.apana.org.au>
2001-07-02 06:39:46 +08:00
##### OS/2 EMX
"OS2-EMX", "gcc::::::::",
Add the configuration target VxWorks.
2002-02-14 23:37:38 +08:00
##### VxWorks for various targets
"vxworks-ppc405","ccppc:-g -msoft-float -mlongcall -DCPU=PPC405 -I\$(WIND_BASE)/target/h:::VXWORKS:-r:::::",
Add a few more VxWorks targets. Correct misspelled VXWORKS macros. Add VXWORKS identifying macros to e_os2.h. Add required inclusions and mappings for VxWorks in e_os.h. A few small modifications to make OpenSSL build and work on VxWorks. PR: 253, except for the change that was handled in an earlier commit, and a request for easy build of just parts of OpenSSL.
2002-10-09 21:40:48 +08:00
"vxworks-ppc750","ccppc:-ansi -nostdinc -DPPC750 -D_REENTRANT -fvolatile -fno-builtin -fno-for-scope -fsigned-char -Wall -msoft-float -mlongcall -DCPU=PPC604 -I\$(WIND_BASE)/target/h \$(DEBUG_FLAG):::VXWORKS:-r:::::",
"vxworks-ppc750-debug","ccppc:-ansi -nostdinc -DPPC750 -D_REENTRANT -fvolatile -fno-builtin -fno-for-scope -fsigned-char -Wall -msoft-float -mlongcall -DCPU=PPC604 -I\$(WIND_BASE)/target/h -DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DCRYPTO_MDEBUG -DPEDANTIC -DDEBUG_SAFESTACK -DDEBUG -g:::VXWORKS:-r:::::",
Further VxWorks changes from Bob Bradley <bob@chaoticsoftware.com>, this time involving VxWorks on MIPS
2003-09-28 22:06:57 +08:00
"vxworks-ppc860","ccppc:-nostdinc -msoft-float -DCPU=PPC860 -DNO_STRINGS_H -I\$(WIND_BASE)/target/h:::VXWORKS:-r:::::",
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
"vxworks-mipsle","ccmips:-B\$(WIND_BASE)/host/\$(WIND_HOST_TYPE)/lib/gcc-lib/ -DL_ENDIAN -EL -Wl,-EL -mips2 -mno-branch-likely -G 0 -fno-builtin -msoft-float -DCPU=MIPS32 -DMIPSEL -DNO_STRINGS_H -I\$(WIND_BASE)/target/h:::VXWORKS:-r::${no_asm}::::::ranlibmips:",
Add the configuration target VxWorks.
2002-02-14 23:37:38 +08:00
Add Tandem OSS target. PR: 192
2002-11-15 00:30:11 +08:00
##### Compaq Non-Stop Kernel (Tandem)
"tandem-c89","c89:-Ww -D__TANDEM -D_XOPEN_SOURCE -D_XOPEN_SOURCE_EXTENDED=1 -D_TANDEM_SOURCE -DB_ENDIAN::(unknown):::THIRTY_TWO_BIT:::",
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
);
Build-n-link new IA-64 modules on Linux and HP-UX.
2004-07-24 07:27:10 +08:00
my @MK1MF_Builds=qw(VC-WIN64I VC-WIN64A
VC-NT VC-CE VC-WIN32
VC-WIN16 VC-W31-16 VC-W31-32 VC-MSDOS
BC-32 BC-16 OS2-EMX netware-clib netware-libc);
Don't make links on Windoze.
1999-03-07 23:21:08 +08:00
I got some reports that some targets have weird dso_schemes. Therefore, I've added a sanity checker. Note that it can be combined with almost any other argument (the other arguments will be completely ignored), with "reconf" as the blatant exception, since it also has the behavior of ignoring all following command line arguments. If --test-sanity and reconf are both used on the command line, the first one wins.
2002-01-27 23:52:37 +08:00
my $idx = 0;
my $idx_cc = $idx++;
my $idx_cflags = $idx++;
my $idx_unistd = $idx++;
my $idx_thread_cflag = $idx++;
my $idx_sys_id = $idx++;
my $idx_lflags = $idx++;
my $idx_bn_ops = $idx++;
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
my $idx_cpuid_obj = $idx++;
I got some reports that some targets have weird dso_schemes. Therefore, I've added a sanity checker. Note that it can be combined with almost any other argument (the other arguments will be completely ignored), with "reconf" as the blatant exception, since it also has the behavior of ignoring all following command line arguments. If --test-sanity and reconf are both used on the command line, the first one wins.
2002-01-27 23:52:37 +08:00
my $idx_bn_obj = $idx++;
my $idx_des_obj = $idx++;
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
my $idx_aes_obj = $idx++;
I got some reports that some targets have weird dso_schemes. Therefore, I've added a sanity checker. Note that it can be combined with almost any other argument (the other arguments will be completely ignored), with "reconf" as the blatant exception, since it also has the behavior of ignoring all following command line arguments. If --test-sanity and reconf are both used on the command line, the first one wins.
2002-01-27 23:52:37 +08:00
my $idx_bf_obj = $idx++;
my $idx_md5_obj = $idx++;
my $idx_sha1_obj = $idx++;
my $idx_cast_obj = $idx++;
my $idx_rc4_obj = $idx++;
my $idx_rmd160_obj = $idx++;
my $idx_rc5_obj = $idx++;
my $idx_dso_scheme = $idx++;
my $idx_shared_target = $idx++;
my $idx_shared_cflag = $idx++;
my $idx_shared_ldflag = $idx++;
my $idx_shared_extension = $idx++;
my $idx_ranlib = $idx++;
Add aix64-cc, and make sure that ar gets proper flags for 64-bit libraries
2002-08-02 04:28:16 +08:00
my $idx_arflags = $idx++;
I got some reports that some targets have weird dso_schemes. Therefore, I've added a sanity checker. Note that it can be combined with almost any other argument (the other arguments will be completely ignored), with "reconf" as the blatant exception, since it also has the behavior of ignoring all following command line arguments. If --test-sanity and reconf are both used on the command line, the first one wins.
2002-01-27 23:52:37 +08:00
Support INSTALL_PREFIX for packagers. Submitted by: Reviewed by: PR:
1999-04-30 05:52:08 +08:00
my $prefix="";
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
my $openssldir="";
Incorporate some changes that make OpenSSL compilable in CygWin.
2001-04-04 23:50:30 +08:00
my $exe_ext="";
Support INSTALL_PREFIX for packagers. Submitted by: Reviewed by: PR:
1999-04-30 05:52:08 +08:00
my $install_prefix="";
New Configure options "threads" and "no-threads". For Solaris and Linux, "threads" (with proper compiler options) is the default. Submitted by: Reviewed by: PR:
1999-04-30 00:10:09 +08:00
my $no_threads=0;
Redo and enhance the support for building shared libraries. Currently there's support for building under Linux and True64 (using examples from the programming manuals), including versioning that is currently the same as OpenSSL versions but should really be a different series. With this change, it's up to the users to decide if they want shared libraries as well as the static ones. This decision now has to be done at configuration time (well, not really, those who know what they do can still do it the same way as before). The OpenSSL programs (openssl and the test programs) are currently always linked statically, but this may change in the future in a configurable manner. The necessary makefile variables to enable this are in place. Also note that I have done absolutely nothing about the Windows target to get something similar. On the other hand, DLLs are already the default there, but without versioning, and I've no idea what the possibilities for such a thing are there...
2000-07-21 23:08:53 +08:00
my $no_shared=1;
zlib default was broken on most platforms. -mcpu=i486 is new as of gcc 2.95, and if you have that, you wouldn't want to optimize for 486 anyway.
2001-09-01 12:18:00 +08:00
my $zlib=0;
First tentative impementation of Kerberos 5 cryptos and keys for SSL/TLS. Implemented by Vern Staats <staatsvr@asc.hpc.mil>, further hacked and distributed by Jeffrey Altman <jaltnab@columbia.edu>
2000-12-01 06:53:34 +08:00
my $no_krb5=0;
New Configure options "threads" and "no-threads". For Solaris and Linux, "threads" (with proper compiler options) is the default. Submitted by: Reviewed by: PR:
1999-04-30 00:10:09 +08:00
my $threads=0;
Add strictness, fix variable substition bugs.
1998-12-29 01:08:48 +08:00
my $no_asm=0;
This helps make the DSO stuff more portable; * "no-dso" option available in Configure so that all DSO methods will return NULL, overriding any support the platform might otherwise have built. * dlfcn_no_h config string now available rather than just dlfcn. This is for platforms that have dlfcn.h functions but do not have (or need) the dlfcn.h header file.
2000-04-06 15:09:45 +08:00
my $no_dso=0;
New Configure option no-<cipher> (rsa, idea, rc5, ...).
1999-04-27 09:14:46 +08:00
my @skip=();
Add strictness, fix variable substition bugs.
1998-12-29 01:08:48 +08:00
my $Makefile="Makefile.ssl";
my $des_locl="crypto/des/des_locl.h";
my $des ="crypto/des/des.h";
my $bn ="crypto/bn/bn.h";
my $md2 ="crypto/md2/md2.h";
my $rc4 ="crypto/rc4/rc4.h";
my $rc4_locl="crypto/rc4/rc4_locl.h";
my $idea ="crypto/idea/idea.h";
my $rc2 ="crypto/rc2/rc2.h";
my $bf ="crypto/bf/bf_locl.h";
my $bn_asm ="bn_asm.o";
my $des_enc="des_enc.o fcrypt_b.o";
Reserve for AES CBC assembler implementation...
2005-01-18 08:43:32 +08:00
my $aes_enc="aes_core.o aes_cbc.o";
Add strictness, fix variable substition bugs.
1998-12-29 01:08:48 +08:00
my $bf_enc ="bf_enc.o";
my $cast_enc="c_enc.o";
my $rc4_enc="rc4_enc.o";
my $rc5_enc="rc5_enc.o";
my $md5_obj="";
my $sha1_obj="";
my $rmd160_obj="";
New option to generate 80386 code.
1999-03-31 20:38:27 +08:00
my $processor="";
Workaround for solaris64 linking problem (explicit "ar rs" is needed to create a symbol table). Sun patches such as 109147-06 probably fix this problem, but we can easily avoid it.
2001-03-09 21:04:06 +08:00
my $default_ranlib;
New Makefile variables $(RANLIB) and $(PERL).
1999-04-01 20:34:33 +08:00
my $perl;
Certain missing algorithms make some SSL versions or TLS impossible to build.
2001-11-27 19:48:30 +08:00
my $no_ssl2=0;
my $no_ssl3=0;
my $no_tls1=0;
my $no_md5=0;
my $no_sha=0;
my $no_rsa=0;
my $no_dh=0;
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
my $no_sse2=0;
Certain missing algorithms make some SSL versions or TLS impossible to build.
2001-11-27 19:48:30 +08:00
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
&usage if ($#ARGV < 0);
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
my $flags;
my $depflags;
my $openssl_algorithm_defines;
my $openssl_thread_defines;
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
my $openssl_sys_defines="";
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
my $openssl_other_defines;
my $libs;
Handle krb5 libraries separately and make sure only libssl.so depends on it.
2003-02-14 21:12:00 +08:00
my $libkrb5="";
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
my $target;
my $options;
my $symlink;
Add support for DJGPP. PR: 75
2002-06-14 04:42:35 +08:00
my $make_depend=0;
First tentative impementation of Kerberos 5 cryptos and keys for SSL/TLS. Implemented by Vern Staats <staatsvr@asc.hpc.mil>, further hacked and distributed by Jeffrey Altman <jaltnab@columbia.edu>
2000-12-01 06:53:34 +08:00
my %withargs=();
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
my @argvcopy=@ARGV;
my $argvstring="";
my $argv_unprocessed=1;
while($argv_unprocessed)
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
{
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
$flags="";
$depflags="";
$openssl_algorithm_defines="";
$openssl_thread_defines="";
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
$openssl_sys_defines="";
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
$openssl_other_defines="";
$libs="";
$target="";
$options="";
$symlink=1;
$argv_unprocessed=0;
$argvstring=join(' ',@argvcopy);
PROCESS_ARGS:
foreach (@argvcopy)
New Configure option no-<cipher> (rsa, idea, rc5, ...).
1999-04-27 09:14:46 +08:00
{
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
s /^-no-/no-/; # some people just can't read the instructions
I got some reports that some targets have weird dso_schemes. Therefore, I've added a sanity checker. Note that it can be combined with almost any other argument (the other arguments will be completely ignored), with "reconf" as the blatant exception, since it also has the behavior of ignoring all following command line arguments. If --test-sanity and reconf are both used on the command line, the first one wins.
2002-01-27 23:52:37 +08:00
if (/^--test-sanity$/)
{
exit(&test_sanity());
}
elsif (/^no-asm$/)
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
{
$no_asm=1;
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
$openssl_other_defines .= "#define OPENSSL_NO_ASM\n";
automatically use no-mdc2 if no-des is requested.
1999-08-03 06:13:46 +08:00
}
Make the no-err option work properly
2003-02-18 20:14:57 +08:00
elsif (/^no-err$/)
{
$flags .= "-DOPENSSL_NO_ERR ";
$openssl_other_defines .= "#define OPENSSL_NO_ERR\n";
}
Merge the engine branch into the main trunk. All conflicts resolved. At the same time, add VMS support for Rijndael.
2000-10-27 05:07:28 +08:00
elsif (/^no-hw-(.+)$/)
{
my $hw=$1;
$hw =~ tr/[a-z]/[A-Z]/;
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
$openssl_other_defines .= "#define OPENSSL_NO_HW_$hw\n";
Merge the engine branch into the main trunk. All conflicts resolved. At the same time, add VMS support for Rijndael.
2000-10-27 05:07:28 +08:00
}
elsif (/^no-hw$/)
{
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
$openssl_other_defines .= "#define OPENSSL_NO_HW\n";
Merge the engine branch into the main trunk. All conflicts resolved. At the same time, add VMS support for Rijndael.
2000-10-27 05:07:28 +08:00
}
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
elsif (/^no-dso$/)
{ $no_dso=1; }
First tentative impementation of Kerberos 5 cryptos and keys for SSL/TLS. Implemented by Vern Staats <staatsvr@asc.hpc.mil>, further hacked and distributed by Jeffrey Altman <jaltnab@columbia.edu>
2000-12-01 06:53:34 +08:00
elsif (/^no-krb5$/)
{ $no_krb5=1; }
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
elsif (/^no-threads$/)
{ $no_threads=1; }
elsif (/^threads$/)
{ $threads=1; }
elsif (/^no-shared$/)
{ $no_shared=1; }
some people just can't read the instructions
2002-11-06 14:02:05 +08:00
elsif (/^shared$/ || /^-shared$/ || /^--shared$/)
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
{ $no_shared=0; }
Add the possibility to specify the use of zlib compression and decompression. It can be set up to link at link time or to load the zlib library at run-time.
2001-07-12 17:11:14 +08:00
elsif (/^no-zlib$/)
{ $zlib=0; }
elsif (/^zlib$/)
{ $zlib=1; }
elsif (/^zlib-dynamic$/)
{ $zlib=2; }
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
elsif (/^no-symlinks$/)
{ $symlink=0; }
Certain missing algorithms make some SSL versions or TLS impossible to build.
2001-11-27 19:48:30 +08:00
elsif (/^no-ssl$/)
{ $no_ssl2 = $no_ssl3 = 1; }
elsif (/^no-ssl2$/)
{ $no_ssl2 = 1; }
elsif (/^no-ssl3$/)
{ $no_ssl3 = 1; }
elsif (/^no-tls1?$/)
{ $no_tls1 = 1; }
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
elsif (/^no-sse2$/)
{ $no_sse2 = 1; }
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
elsif (/^no-(.+)$/)
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
{
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
my $algo=$1;
push @skip,$algo;
$algo =~ tr/[a-z]/[A-Z]/;
Don't put configuration macro definitions on the command line, we're just fooling ourselves and then screwing up for other applications.
2003-03-21 07:52:41 +08:00
#$flags .= "-DOPENSSL_NO_$algo ";
#$depflags .= "-DOPENSSL_NO_$algo ";
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
$openssl_algorithm_defines .= "#define OPENSSL_NO_$algo\n";
The 'no-rijndael' option must define OPENSSL_NO_AES because that is what we look for in crypto/evp.
2001-09-22 09:40:47 +08:00
if ($algo eq "RIJNDAEL")
{
When AES is skipped because the option 'no-rijndael' was given, make sure it's skipped from SDIRS as well.
2002-11-14 16:20:40 +08:00
push @skip, "aes";
Don't put configuration macro definitions on the command line, we're just fooling ourselves and then screwing up for other applications.
2003-03-21 07:52:41 +08:00
#$flags .= "-DOPENSSL_NO_AES ";
#$depflags .= "-DOPENSSL_NO_AES ";
The 'no-rijndael' option must define OPENSSL_NO_AES because that is what we look for in crypto/evp.
2001-09-22 09:40:47 +08:00
$openssl_algorithm_defines .= "#define OPENSSL_NO_AES\n";
}
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
if ($algo eq "DES")
{
push @skip, "mdc2";
$options .= " no-mdc2";
Don't put configuration macro definitions on the command line, we're just fooling ourselves and then screwing up for other applications.
2003-03-21 07:52:41 +08:00
#$flags .= "-DOPENSSL_NO_MDC2 ";
#$depflags .= "-DOPENSSL_NO_MDC2 ";
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
$openssl_algorithm_defines .= "#define OPENSSL_NO_MDC2\n";
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
}
Add ECDH support. Additional changes: - use EC_GROUP_get_degree() in apps/req.c - add ECDSA and ECDH to apps/speed.c - adds support for EC curves over binary fields to ECDSA - new function EC_KEY_up_ref() in crypto/ec/ec_key.c - reorganize crypto/ecdsa/ecdsatest.c - add engine support for ECDH - fix a few bugs in ECDSA engine support Submitted by: Douglas Stebila <douglas.stebila@sun.com>
2002-08-09 16:43:04 +08:00
if ($algo eq "EC")
{
push @skip, "ecdsa";
push @skip, "ecdh";
$options .= " no-ecdsa";
$options .= " no-ecdh";
$flags .= "-DOPENSSL_NO_ECDSA ";
$flags .= "-DOPENSSL_NO_ECDH ";
$depflags .= "-DOPENSSL_NO_ECDSA ";
$depflags .= "-DOPENSSL_NO_ECDH ";
$openssl_algorithm_defines .= "#define OPENSSL_NO_ECDSA\n";
$openssl_algorithm_defines .= "#define OPENSSL_NO_ECDH\n";
}
if ($algo eq "SHA" || $algo eq "SHA1")
ECDSA support Submitted by: Nils Larsch <nla@trustcenter.de>
2002-02-14 02:21:51 +08:00
{
push @skip, "ecdsa";
$options .= " no-ecdsa";
$flags .= "-DOPENSSL_NO_ECDSA ";
$depflags .= "-DOPENSSL_NO_ECDSA ";
$openssl_algorithm_defines .= "#define OPENSSL_NO_ECDSA\n";
}
Certain missing algorithms make some SSL versions or TLS impossible to build.
2001-11-27 19:48:30 +08:00
if ($algo eq "MD5")
{
$no_md5 = 1;
}
if ($algo eq "SHA")
{
$no_sha = 1;
}
if ($algo eq "RSA")
{
$no_rsa = 1;
}
if ($algo eq "DH")
{
$no_dh = 1;
}
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
}
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
elsif (/^reconfigure/ || /^reconf/)
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
{
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
if (open(IN,"<$Makefile"))
{
while (<IN>)
{
chop;
if (/^CONFIGURE_ARGS=(.*)/)
{
$argvstring=$1;
@argvcopy=split(' ',$argvstring);
die "Incorrect data to reconfigure, please do a normal configuration\n"
if (grep(/^reconf/,@argvcopy));
print "Reconfiguring with: $argvstring\n";
$argv_unprocessed=1;
close(IN);
last PROCESS_ARGS;
}
}
close(IN);
}
die "Insufficient data to reconfigure, please do a normal configuration\n";
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
}
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
elsif (/^386$/)
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
{ $processor=386; $no_sse2=1; }
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
elsif (/^rsaref$/)
Allow DSO flags like -fpic, -fPIC, -KPIC etc. to be specified on the `perl Configure ...' command line. This way one can compile OpenSSL libraries with Position Independent Code (PIC) which is needed for linking it into DSOs.
1999-03-06 22:35:03 +08:00
{
Remove references to RSAref. The glue library is but a memory to fade away now...
2000-11-09 01:51:37 +08:00
# No RSAref support any more since it's not needed.
# The check for the option is there so scripts aren't
# broken
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
}
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
elsif (/^[-+]/)
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
{
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
if (/^-[lL](.*)$/)
{
$libs.=$_." ";
}
elsif (/^-[^-]/ or /^\+/)
{
$flags.=$_." ";
}
elsif (/^--prefix=(.*)$/)
{
$prefix=$1;
}
elsif (/^--openssldir=(.*)$/)
{
$openssldir=$1;
}
elsif (/^--install.prefix=(.*)$/)
{
$install_prefix=$1;
}
First tentative impementation of Kerberos 5 cryptos and keys for SSL/TLS. Implemented by Vern Staats <staatsvr@asc.hpc.mil>, further hacked and distributed by Jeffrey Altman <jaltnab@columbia.edu>
2000-12-01 06:53:34 +08:00
elsif (/^--with-krb5-(dir|lib|include|flavor)=(.*)$/)
{
$withargs{"krb5-".$1}=$2;
}
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
else
{
print STDERR $usage;
exit(1);
}
Allow DSO flags like -fpic, -fPIC, -KPIC etc. to be specified on the `perl Configure ...' command line. This way one can compile OpenSSL libraries with Position Independent Code (PIC) which is needed for linking it into DSOs.
1999-03-06 22:35:03 +08:00
}
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
elsif ($_ =~ /^([^:]+):(.+)$/)
Support INSTALL_PREFIX for packagers. Submitted by: Reviewed by: PR:
1999-04-30 05:52:08 +08:00
{
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
eval "\$table{\$1} = \"$2\""; # allow $xxx constructs in the string
$target=$1;
Support INSTALL_PREFIX for packagers. Submitted by: Reviewed by: PR:
1999-04-30 05:52:08 +08:00
}
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
else
{
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
die "target already defined - $target\n" if ($target ne "");
$target=$_;
}
unless ($_ eq $target) {
if ($options eq "") {
$options = $_;
} else {
$options .= " ".$_;
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
}
Don't confuse mk1mf.pl by combining multiple options into one.
1999-07-09 20:00:34 +08:00
}
}
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
}
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
Certain missing algorithms make some SSL versions or TLS impossible to build.
2001-11-27 19:48:30 +08:00
$no_ssl3=1 if ($no_md5 || $no_sha);
$no_ssl3=1 if ($no_rsa && $no_dh);
$no_ssl2=1 if ($no_md5);
$no_ssl2=1 if ($no_rsa);
$no_tls1=1 if ($no_md5 || $no_sha);
$no_tls1=1 if ($no_dh);
if ($no_ssl2)
{
push @skip,"SSL2";
Don't put configuration macro definitions on the command line, we're just fooling ourselves and then screwing up for other applications.
2003-03-21 07:52:41 +08:00
#$flags .= "-DOPENSSL_NO_SSL2 ";
#$depflags .= "-DOPENSSL_NO_SSL2 ";
Certain missing algorithms make some SSL versions or TLS impossible to build.
2001-11-27 19:48:30 +08:00
$openssl_algorithm_defines .= "#define OPENSSL_NO_SSL2\n";
}
if ($no_ssl3)
{
push @skip,"SSL3";
Don't put configuration macro definitions on the command line, we're just fooling ourselves and then screwing up for other applications.
2003-03-21 07:52:41 +08:00
#$flags .= "-DOPENSSL_NO_SSL3 ";
#$depflags .= "-DOPENSSL_NO_SSL3 ";
Certain missing algorithms make some SSL versions or TLS impossible to build.
2001-11-27 19:48:30 +08:00
$openssl_algorithm_defines .= "#define OPENSSL_NO_SSL3\n";
}
if ($no_tls1)
{
push @skip,"TLS1";
Don't put configuration macro definitions on the command line, we're just fooling ourselves and then screwing up for other applications.
2003-03-21 07:52:41 +08:00
#$flags .= "-DOPENSSL_NO_TLS1 ";
#$depflags .= "-DOPENSSL_NO_TLS1 ";
Certain missing algorithms make some SSL versions or TLS impossible to build.
2001-11-27 19:48:30 +08:00
$openssl_algorithm_defines .= "#define OPENSSL_NO_TLS1\n";
}
Some assembler-related clean-ups.
1999-05-29 07:18:51 +08:00
if ($target eq "TABLE") {
foreach $target (sort keys %table) {
print_table_entry($target);
}
exit 0;
}
More compact Configure usage message.
2000-02-21 08:55:45 +08:00
if ($target eq "LIST") {
foreach (sort keys %table) {
print;
print "\n";
}
exit 0;
}
At Corinna Vinschen's request, change CygWin32 to Cygwin
2002-02-14 20:28:24 +08:00
if ($target =~ m/^CygWin32(-.*)$/) {
$target = "Cygwin".$1;
}
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
print "Configuring for $target\n";
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
&usage if (!defined($table{$target}));
Netware-specific changes, PR: 780 Submitted by: Verdon Walker <VWalker@novell.com> Reviewed by: Richard Levitte
2003-11-28 21:10:58 +08:00
my $IsMK1MF=scalar grep /^$target$/,@MK1MF_Builds;
Fixes so it will compile again under Win32.
1999-04-26 04:57:09 +08:00
Clean-up GAS targets: get rid of "cpp" stuff and replace it with "purified" COFF and a.out targets [similar to ELF targets]. You might notice some rudementary support for shared mingw builds under cygwin. It works (it produces cryptoeay32.dll and ssleay32.dll with everything exported by name), but it's primarily for testing/debugging purposes, at least for now...
2004-08-02 01:33:58 +08:00
$IsMK1MF=1 if ($target eq "mingw" && $^O ne "cygwin");
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
$exe_ext=".exe" if ($target eq "Cygwin" || $target eq "DJGPP" || $target eq "mingw");
$exe_ext=".pm" if ($target =~ /vos/);
Support INSTALL_PREFIX for packagers. Submitted by: Reviewed by: PR:
1999-04-30 05:52:08 +08:00
$openssldir="/usr/local/ssl" if ($openssldir eq "" and $prefix eq "");
$prefix=$openssldir if $prefix eq "";
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
Changes for VOS, submitted by Paul Green <Paul.Green@stratus.com>. PR: 499
2004-06-29 06:01:37 +08:00
$default_ranlib= &which("ranlib") or $default_ranlib="true";
$perl=$ENV{'PERL'} or $perl=&which("perl5") or $perl=&which("perl")
or $perl="perl";
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
chop $openssldir if $openssldir =~ /\/$/;
Support INSTALL_PREFIX for packagers. Submitted by: Reviewed by: PR:
1999-04-30 05:52:08 +08:00
chop $prefix if $prefix =~ /\/$/;
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
Support INSTALL_PREFIX for packagers. Submitted by: Reviewed by: PR:
1999-04-30 05:52:08 +08:00
$openssldir=$prefix . "/ssl" if $openssldir eq "";
Add support for DJGPP. PR: 75
2002-06-14 04:42:35 +08:00
$openssldir=$prefix . "/" . $openssldir if $openssldir !~ /(^\/|^[a-zA-Z]:[\\\/])/;
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
Don't make links on Windoze.
1999-03-07 23:21:08 +08:00
Netware-specific changes, PR: 780 Submitted by: Verdon Walker <VWalker@novell.com> Reviewed by: Richard Levitte
2003-11-28 21:10:58 +08:00
print "IsMK1MF=$IsMK1MF\n";
Don't make links on Windoze.
1999-03-07 23:21:08 +08:00
I got some reports that some targets have weird dso_schemes. Therefore, I've added a sanity checker. Note that it can be combined with almost any other argument (the other arguments will be completely ignored), with "reconf" as the blatant exception, since it also has the behavior of ignoring all following command line arguments. If --test-sanity and reconf are both used on the command line, the first one wins.
2002-01-27 23:52:37 +08:00
my @fields = split(/\s*:\s*/,$table{$target} . ":" x 30 , -1);
my $cc = $fields[$idx_cc];
my $cflags = $fields[$idx_cflags];
my $unistd = $fields[$idx_unistd];
my $thread_cflag = $fields[$idx_thread_cflag];
my $sys_id = $fields[$idx_sys_id];
my $lflags = $fields[$idx_lflags];
my $bn_ops = $fields[$idx_bn_ops];
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
my $cpuid_obj = $fields[$idx_cpuid_obj];
I got some reports that some targets have weird dso_schemes. Therefore, I've added a sanity checker. Note that it can be combined with almost any other argument (the other arguments will be completely ignored), with "reconf" as the blatant exception, since it also has the behavior of ignoring all following command line arguments. If --test-sanity and reconf are both used on the command line, the first one wins.
2002-01-27 23:52:37 +08:00
my $bn_obj = $fields[$idx_bn_obj];
my $des_obj = $fields[$idx_des_obj];
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
my $aes_obj = $fields[$idx_aes_obj];
I got some reports that some targets have weird dso_schemes. Therefore, I've added a sanity checker. Note that it can be combined with almost any other argument (the other arguments will be completely ignored), with "reconf" as the blatant exception, since it also has the behavior of ignoring all following command line arguments. If --test-sanity and reconf are both used on the command line, the first one wins.
2002-01-27 23:52:37 +08:00
my $bf_obj = $fields[$idx_bf_obj];
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
my $md5_obj = $fields[$idx_md5_obj];
my $sha1_obj = $fields[$idx_sha1_obj];
I got some reports that some targets have weird dso_schemes. Therefore, I've added a sanity checker. Note that it can be combined with almost any other argument (the other arguments will be completely ignored), with "reconf" as the blatant exception, since it also has the behavior of ignoring all following command line arguments. If --test-sanity and reconf are both used on the command line, the first one wins.
2002-01-27 23:52:37 +08:00
my $cast_obj = $fields[$idx_cast_obj];
my $rc4_obj = $fields[$idx_rc4_obj];
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
my $rmd160_obj = $fields[$idx_rmd160_obj];
I got some reports that some targets have weird dso_schemes. Therefore, I've added a sanity checker. Note that it can be combined with almost any other argument (the other arguments will be completely ignored), with "reconf" as the blatant exception, since it also has the behavior of ignoring all following command line arguments. If --test-sanity and reconf are both used on the command line, the first one wins.
2002-01-27 23:52:37 +08:00
my $rc5_obj = $fields[$idx_rc5_obj];
my $dso_scheme = $fields[$idx_dso_scheme];
my $shared_target = $fields[$idx_shared_target];
my $shared_cflag = $fields[$idx_shared_cflag];
my $shared_ldflag = $fields[$idx_shared_ldflag];
my $shared_extension = $fields[$idx_shared_extension];
my $ranlib = $fields[$idx_ranlib];
Add aix64-cc, and make sure that ar gets proper flags for 64-bit libraries
2002-08-02 04:28:16 +08:00
my $arflags = $fields[$idx_arflags];
I got some reports that some targets have weird dso_schemes. Therefore, I've added a sanity checker. Note that it can be combined with almost any other argument (the other arguments will be completely ignored), with "reconf" as the blatant exception, since it also has the behavior of ignoring all following command line arguments. If --test-sanity and reconf are both used on the command line, the first one wins.
2002-01-27 23:52:37 +08:00
If the user said 'shared' and we haven't included support for shared libraries, warn him or her. Reminder by Andrew Marlow <apm35@student.open.ac.uk>
2003-01-11 19:40:39 +08:00
my $no_shared_warn=0;
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
$cflags="$flags$cflags" if ($flags ne "");
New Configure options "threads" and "no-threads". For Solaris and Linux, "threads" (with proper compiler options) is the default. Submitted by: Reviewed by: PR:
1999-04-30 00:10:09 +08:00
First tentative impementation of Kerberos 5 cryptos and keys for SSL/TLS. Implemented by Vern Staats <staatsvr@asc.hpc.mil>, further hacked and distributed by Jeffrey Altman <jaltnab@columbia.edu>
2000-12-01 06:53:34 +08:00
# Kerberos settings. The flavor must be provided from outside, either through
# the script "config" or manually.
if ($no_krb5
|| !defined($withargs{"krb5-flavor"})
|| $withargs{"krb5-flavor"} eq "")
{
Don't put configuration macro definitions on the command line, we're just fooling ourselves and then screwing up for other applications.
2003-03-21 07:52:41 +08:00
#$cflags="-DOPENSSL_NO_KRB5 $cflags";
Temporary measure: if no KRB5 is defined, add "no-krb5" to the options. This is so mk1mf.pl can pick it up from Makefile.
2000-12-31 09:41:01 +08:00
$options.=" no-krb5" unless $no_krb5;
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
$openssl_algorithm_defines .= "#define OPENSSL_NO_KRB5\n";
First tentative impementation of Kerberos 5 cryptos and keys for SSL/TLS. Implemented by Vern Staats <staatsvr@asc.hpc.mil>, further hacked and distributed by Jeffrey Altman <jaltnab@columbia.edu>
2000-12-01 06:53:34 +08:00
}
else
{
Patches from Vern Staats <staatsvr@asc.hpc.mil> to get Kerberos 5 in SSL according to RFC 2712. His comment is: This is a patch to openssl-SNAP-20010702 to support Kerberized SSL authentication. I'm expecting to have the full kssl-0.5 kit up on sourceforge by the end of the week. The full kit includes patches for mod-ssl, apache, and a few text clients. The sourceforge URL is http://sourceforge.net/projects/kssl/ . Thanks to a note from Simon Wilkinson I've replaced my KRB5 AP_REQ message with a real KerberosWrapper struct. I think this is fully RFC 2712 compliant now, including support for the optional authenticator field. I also added openssl-style ASN.1 macros for a few Kerberos structs; see crypto/krb5/ if you're interested.
2001-07-10 05:46:58 +08:00
my ($lresolv, $lpath, $lext);
First tentative impementation of Kerberos 5 cryptos and keys for SSL/TLS. Implemented by Vern Staats <staatsvr@asc.hpc.mil>, further hacked and distributed by Jeffrey Altman <jaltnab@columbia.edu>
2000-12-01 06:53:34 +08:00
if ($withargs{"krb5-flavor"} =~ /^[Hh]eimdal$/)
{
Heimdal isn't really supported right now. Say so, and offer a possibility to force the use of Heimdal, and warn if that's used. PR: 346
2002-11-26 17:19:17 +08:00
die "Sorry, Heimdal is currently not supported\n";
}
##### HACK to force use of Heimdal.
##### WARNING: Since we don't really have adequate support for Heimdal,
##### using this will break the build. You'll have to make
##### changes to the source, and if you do, please send
##### patches to openssl-dev@openssl.org
if ($withargs{"krb5-flavor"} =~ /^force-[Hh]eimdal$/)
{
warn "Heimdal isn't really supported. Your build WILL break\n";
I forgot that @ in strings must be escaped in Perl
2002-11-26 23:27:05 +08:00
warn "If you fix the problems, please send a patch to openssl-dev\@openssl.org\n";
First tentative impementation of Kerberos 5 cryptos and keys for SSL/TLS. Implemented by Vern Staats <staatsvr@asc.hpc.mil>, further hacked and distributed by Jeffrey Altman <jaltnab@columbia.edu>
2000-12-01 06:53:34 +08:00
$withargs{"krb5-dir"} = "/usr/heimdal"
if $withargs{"krb5-dir"} eq "";
$withargs{"krb5-lib"} = "-L".$withargs{"krb5-dir"}.
"/lib -lgssapi -lkrb5 -lcom_err"
if $withargs{"krb5-lib"} eq "";
$cflags="-DKRB5_HEIMDAL $cflags";
}
Patches from Vern Staats <staatsvr@asc.hpc.mil> to get Kerberos 5 in SSL according to RFC 2712. His comment is: This is a patch to openssl-SNAP-20010702 to support Kerberized SSL authentication. I'm expecting to have the full kssl-0.5 kit up on sourceforge by the end of the week. The full kit includes patches for mod-ssl, apache, and a few text clients. The sourceforge URL is http://sourceforge.net/projects/kssl/ . Thanks to a note from Simon Wilkinson I've replaced my KRB5 AP_REQ message with a real KerberosWrapper struct. I think this is fully RFC 2712 compliant now, including support for the optional authenticator field. I also added openssl-style ASN.1 macros for a few Kerberos structs; see crypto/krb5/ if you're interested.
2001-07-10 05:46:58 +08:00
if ($withargs{"krb5-flavor"} =~ /^[Mm][Ii][Tt]/)
First tentative impementation of Kerberos 5 cryptos and keys for SSL/TLS. Implemented by Vern Staats <staatsvr@asc.hpc.mil>, further hacked and distributed by Jeffrey Altman <jaltnab@columbia.edu>
2000-12-01 06:53:34 +08:00
{
$withargs{"krb5-dir"} = "/usr/kerberos"
if $withargs{"krb5-dir"} eq "";
$withargs{"krb5-lib"} = "-L".$withargs{"krb5-dir"}.
"/lib -lgssapi_krb5 -lkrb5 -lcom_err -lk5crypto"
if $withargs{"krb5-lib"} eq "";
$cflags="-DKRB5_MIT $cflags";
Patches from Vern Staats <staatsvr@asc.hpc.mil> to get Kerberos 5 in SSL according to RFC 2712. His comment is: This is a patch to openssl-SNAP-20010702 to support Kerberized SSL authentication. I'm expecting to have the full kssl-0.5 kit up on sourceforge by the end of the week. The full kit includes patches for mod-ssl, apache, and a few text clients. The sourceforge URL is http://sourceforge.net/projects/kssl/ . Thanks to a note from Simon Wilkinson I've replaced my KRB5 AP_REQ message with a real KerberosWrapper struct. I think this is fully RFC 2712 compliant now, including support for the optional authenticator field. I also added openssl-style ASN.1 macros for a few Kerberos structs; see crypto/krb5/ if you're interested.
2001-07-10 05:46:58 +08:00
$withargs{"krb5-flavor"} =~ s/^[Mm][Ii][Tt][._-]*//;
if ($withargs{"krb5-flavor"} =~ /^1[._-]*[01]/)
{
$cflags="-DKRB5_MIT_OLD11 $cflags";
}
}
LRESOLV:
foreach $lpath ("/lib", "/usr/lib")
{
foreach $lext ("a", "so")
{
$lresolv = "$lpath/libresolv.$lext";
last LRESOLV if (-r "$lresolv");
$lresolv = "";
}
First tentative impementation of Kerberos 5 cryptos and keys for SSL/TLS. Implemented by Vern Staats <staatsvr@asc.hpc.mil>, further hacked and distributed by Jeffrey Altman <jaltnab@columbia.edu>
2000-12-01 06:53:34 +08:00
}
Patches from Vern Staats <staatsvr@asc.hpc.mil> to get Kerberos 5 in SSL according to RFC 2712. His comment is: This is a patch to openssl-SNAP-20010702 to support Kerberized SSL authentication. I'm expecting to have the full kssl-0.5 kit up on sourceforge by the end of the week. The full kit includes patches for mod-ssl, apache, and a few text clients. The sourceforge URL is http://sourceforge.net/projects/kssl/ . Thanks to a note from Simon Wilkinson I've replaced my KRB5 AP_REQ message with a real KerberosWrapper struct. I think this is fully RFC 2712 compliant now, including support for the optional authenticator field. I also added openssl-style ASN.1 macros for a few Kerberos structs; see crypto/krb5/ if you're interested.
2001-07-10 05:46:58 +08:00
$withargs{"krb5-lib"} .= " -lresolv"
Small tweaks for code consistency.
2002-12-20 05:55:48 +08:00
if ("$lresolv" ne "");
First tentative impementation of Kerberos 5 cryptos and keys for SSL/TLS. Implemented by Vern Staats <staatsvr@asc.hpc.mil>, further hacked and distributed by Jeffrey Altman <jaltnab@columbia.edu>
2000-12-01 06:53:34 +08:00
$withargs{"krb5-include"} = "-I".$withargs{"krb5-dir"}."/include"
Patches from Vern Staats <staatsvr@asc.hpc.mil> to get Kerberos 5 in SSL according to RFC 2712. His comment is: This is a patch to openssl-SNAP-20010702 to support Kerberized SSL authentication. I'm expecting to have the full kssl-0.5 kit up on sourceforge by the end of the week. The full kit includes patches for mod-ssl, apache, and a few text clients. The sourceforge URL is http://sourceforge.net/projects/kssl/ . Thanks to a note from Simon Wilkinson I've replaced my KRB5 AP_REQ message with a real KerberosWrapper struct. I think this is fully RFC 2712 compliant now, including support for the optional authenticator field. I also added openssl-style ASN.1 macros for a few Kerberos structs; see crypto/krb5/ if you're interested.
2001-07-10 05:46:58 +08:00
if $withargs{"krb5-include"} eq "" &&
$withargs{"krb5-dir"} ne "";
First tentative impementation of Kerberos 5 cryptos and keys for SSL/TLS. Implemented by Vern Staats <staatsvr@asc.hpc.mil>, further hacked and distributed by Jeffrey Altman <jaltnab@columbia.edu>
2000-12-01 06:53:34 +08:00
}
This helps make the DSO stuff more portable; * "no-dso" option available in Configure so that all DSO methods will return NULL, overriding any support the platform might otherwise have built. * dlfcn_no_h config string now available rather than just dlfcn. This is for platforms that have dlfcn.h functions but do not have (or need) the dlfcn.h header file.
2000-04-06 15:09:45 +08:00
# The DSO code currently always implements all functions so that no
# applications will have to worry about that from a compilation point
# of view. However, the "method"s may return zero unless that platform
# has support compiled in for them. Currently each method is enabled
# by a define "DSO_<name>" ... we translate the "dso_scheme" config
# string entry into using the following logic;
I forgot about $openssl_other_defines ... should probably do this for consistency. Not sure though whether HAVE_DLFCN_H should be included too? If we go the autoconf route then this probably wouldn't be included.
2000-04-06 19:56:45 +08:00
my $dso_cflags;
This helps make the DSO stuff more portable; * "no-dso" option available in Configure so that all DSO methods will return NULL, overriding any support the platform might otherwise have built. * dlfcn_no_h config string now available rather than just dlfcn. This is for platforms that have dlfcn.h functions but do not have (or need) the dlfcn.h header file.
2000-04-06 15:09:45 +08:00
if (!$no_dso && $dso_scheme ne "")
{
This commit ties the new DSO code (crypto/dso/) into the build for a variety of platforms. A few are missing, and they will be added in eventually, but as this is new stuff, it was better to not break lots of platforms in one go that we can't easily test. The changes to "Configure" should illustrate how to add support to other systems if you feel like having a go. NB: I'll add something shortly to allow you to add "dlfcn.h" support on those platforms that don't have (or need) a dlfcn.h header file. (The symbol for Configure will probably by "dlfcn_no_h"). Thanks to Richard Levitte, who is responsible for the dso_dl.c support, understanding the trickier aspects of the build process, and giving great feedback on everything else. [Don't use this stuff if you're easily offended by changes to the interface or behaviour - it's still work in progress.] PR:
2000-04-05 06:32:19 +08:00
$dso_scheme =~ tr/[a-z]/[A-Z]/;
This helps make the DSO stuff more portable; * "no-dso" option available in Configure so that all DSO methods will return NULL, overriding any support the platform might otherwise have built. * dlfcn_no_h config string now available rather than just dlfcn. This is for platforms that have dlfcn.h functions but do not have (or need) the dlfcn.h header file.
2000-04-06 15:09:45 +08:00
if ($dso_scheme eq "DLFCN")
{
I forgot about $openssl_other_defines ... should probably do this for consistency. Not sure though whether HAVE_DLFCN_H should be included too? If we go the autoconf route then this probably wouldn't be included.
2000-04-06 19:56:45 +08:00
$dso_cflags = "-DDSO_DLFCN -DHAVE_DLFCN_H";
This helps make the DSO stuff more portable; * "no-dso" option available in Configure so that all DSO methods will return NULL, overriding any support the platform might otherwise have built. * dlfcn_no_h config string now available rather than just dlfcn. This is for platforms that have dlfcn.h functions but do not have (or need) the dlfcn.h header file.
2000-04-06 15:09:45 +08:00
}
elsif ($dso_scheme eq "DLFCN_NO_H")
{
I forgot about $openssl_other_defines ... should probably do this for consistency. Not sure though whether HAVE_DLFCN_H should be included too? If we go the autoconf route then this probably wouldn't be included.
2000-04-06 19:56:45 +08:00
$dso_cflags = "-DDSO_DLFCN";
This helps make the DSO stuff more portable; * "no-dso" option available in Configure so that all DSO methods will return NULL, overriding any support the platform might otherwise have built. * dlfcn_no_h config string now available rather than just dlfcn. This is for platforms that have dlfcn.h functions but do not have (or need) the dlfcn.h header file.
2000-04-06 15:09:45 +08:00
}
else
{
I forgot about $openssl_other_defines ... should probably do this for consistency. Not sure though whether HAVE_DLFCN_H should be included too? If we go the autoconf route then this probably wouldn't be included.
2000-04-06 19:56:45 +08:00
$dso_cflags = "-DDSO_$dso_scheme";
This helps make the DSO stuff more portable; * "no-dso" option available in Configure so that all DSO methods will return NULL, overriding any support the platform might otherwise have built. * dlfcn_no_h config string now available rather than just dlfcn. This is for platforms that have dlfcn.h functions but do not have (or need) the dlfcn.h header file.
2000-04-06 15:09:45 +08:00
}
I forgot about $openssl_other_defines ... should probably do this for consistency. Not sure though whether HAVE_DLFCN_H should be included too? If we go the autoconf route then this probably wouldn't be included.
2000-04-06 19:56:45 +08:00
$cflags = "$dso_cflags $cflags";
This helps make the DSO stuff more portable; * "no-dso" option available in Configure so that all DSO methods will return NULL, overriding any support the platform might otherwise have built. * dlfcn_no_h config string now available rather than just dlfcn. This is for platforms that have dlfcn.h functions but do not have (or need) the dlfcn.h header file.
2000-04-06 15:09:45 +08:00
}
This commit ties the new DSO code (crypto/dso/) into the build for a variety of platforms. A few are missing, and they will be added in eventually, but as this is new stuff, it was better to not break lots of platforms in one go that we can't easily test. The changes to "Configure" should illustrate how to add support to other systems if you feel like having a go. NB: I'll add something shortly to allow you to add "dlfcn.h" support on those platforms that don't have (or need) a dlfcn.h header file. (The symbol for Configure will probably by "dlfcn_no_h"). Thanks to Richard Levitte, who is responsible for the dso_dl.c support, understanding the trickier aspects of the build process, and giving great feedback on everything else. [Don't use this stuff if you're easily offended by changes to the interface or behaviour - it's still work in progress.] PR:
2000-04-05 06:32:19 +08:00
New Configure options "threads" and "no-threads". For Solaris and Linux, "threads" (with proper compiler options) is the default. Submitted by: Reviewed by: PR:
1999-04-30 00:10:09 +08:00
my $thread_cflags;
Make Configure add the configuration options that it was copmiled with. Currently, those defines are protected with a OPENSSL_EXCLUDED. That may not be the best strategy, but it will do for now.
2000-02-18 17:11:37 +08:00
my $thread_defines;
New Configure options "threads" and "no-threads". For Solaris and Linux, "threads" (with proper compiler options) is the default. Submitted by: Reviewed by: PR:
1999-04-30 00:10:09 +08:00
if ($thread_cflag ne "(unknown)" && !$no_threads)
{
# If we know how to do it, support threads by default.
$threads = 1;
}
if ($thread_cflag eq "(unknown)")
{
# If the user asked for "threads", hopefully they also provided
# any system-dependent compiler options that are necessary.
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
$thread_cflags="-DOPENSSL_THREADS $cflags" ;
$thread_defines .= "#define OPENSSL_THREADS\n";
New Configure options "threads" and "no-threads". For Solaris and Linux, "threads" (with proper compiler options) is the default. Submitted by: Reviewed by: PR:
1999-04-30 00:10:09 +08:00
}
else
{
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
$thread_cflags="-DOPENSSL_THREADS $thread_cflag $cflags";
$thread_defines .= "#define OPENSSL_THREADS\n";
Don't define platform-dependent preprocessor symbols for OPENSSL_THREAD_DEFINES. "make update".
2000-02-25 02:19:50 +08:00
# my $def;
# foreach $def (split ' ',$thread_cflag)
# {
# if ($def =~ s/^-D// && $def !~ /^_/)
# {
# $thread_defines .= "#define $def\n";
# }
# }
New Configure options "threads" and "no-threads". For Solaris and Linux, "threads" (with proper compiler options) is the default. Submitted by: Reviewed by: PR:
1999-04-30 00:10:09 +08:00
}
Small tweaks for code consistency.
2002-12-20 05:55:48 +08:00
$lflags="$libs$lflags" if ($libs ne "");
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
Import of old SSLeay release: SSLeay 0.9.1b (unreleased)
1998-12-21 19:00:56 +08:00
if ($no_asm)
{
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
$cpuid_obj=$bn_obj=$des_obj=$aes_obj=$bf_obj=$cast_obj=$rc4_obj=$rc5_obj="";
Import of old SSLeay release: SSLeay 0.9.1b (unreleased)
1998-12-21 19:00:56 +08:00
$sha1_obj=$md5_obj=$rmd160_obj="";
}
IA-32 assembler modules (primarily DES) PIC-ification. Idea is to keep shared libraries shared.
2002-12-14 01:56:14 +08:00
if (!$no_shared)
{
$cast_obj=""; # CAST assembler is not PIC
}
New Configure options "threads" and "no-threads". For Solaris and Linux, "threads" (with proper compiler options) is the default. Submitted by: Reviewed by: PR:
1999-04-30 00:10:09 +08:00
if ($threads)
{
Don't put configuration macro definitions on the command line, we're just fooling ourselves and then screwing up for other applications.
2003-03-21 07:52:41 +08:00
#$cflags=$thread_cflags;
Add the possibility to specify the use of zlib compression and decompression. It can be set up to link at link time or to load the zlib library at run-time.
2001-07-12 17:11:14 +08:00
$openssl_thread_defines .= $thread_defines;
}
if ($zlib)
{
$cflags = "-DZLIB $cflags";
$cflags = "-DZLIB_SHARED $cflags" if $zlib == 2;
Add -lz to the ld flags when the user has chosen to link in zlib *statically*. Notified by Doug Kaufman <dkaufman@rahul.net>
2002-08-16 14:15:41 +08:00
$lflags = "$lflags -lz" if $zlib == 1;
New Configure options "threads" and "no-threads". For Solaris and Linux, "threads" (with proper compiler options) is the default. Submitted by: Reviewed by: PR:
1999-04-30 00:10:09 +08:00
}
Avoid loops, and make sure that it's possible to still build shared libraries even if the "shared" configuration option wasn't chosen.
2000-07-25 04:36:46 +08:00
# You will find shlib_mark1 and shlib_mark2 explained in Makefile.org
Rework the system to generate shared libraries: - Make note of the expected extension for the shared libraries and if there is a need for symbolic links from for example libcrypto.so.0 to libcrypto.so.0.9.7. There is extended info in Configure for that. - Make as few rebuilds of the shared libraries as possible. - Still avoid linking the OpenSSL programs with the shared libraries. - When installing, install the shared libraries separately from the static ones.
2000-10-13 23:25:06 +08:00
my $shared_mark = "";
IA-32 assembler modules (primarily DES) PIC-ification. Idea is to keep shared libraries shared.
2002-12-14 01:56:14 +08:00
if ($shared_target eq "")
{
If the user said 'shared' and we haven't included support for shared libraries, warn him or her. Reminder by Andrew Marlow <apm35@student.open.ac.uk>
2003-01-11 19:40:39 +08:00
$no_shared_warn = 1 if !$no_shared;
IA-32 assembler modules (primarily DES) PIC-ification. Idea is to keep shared libraries shared.
2002-12-14 01:56:14 +08:00
$no_shared = 1;
}
if (!$no_shared)
Redo and enhance the support for building shared libraries. Currently there's support for building under Linux and True64 (using examples from the programming manuals), including versioning that is currently the same as OpenSSL versions but should really be a different series. With this change, it's up to the users to decide if they want shared libraries as well as the static ones. This decision now has to be done at configuration time (well, not really, those who know what they do can still do it the same way as before). The OpenSSL programs (openssl and the test programs) are currently always linked statically, but this may change in the future in a configurable manner. The necessary makefile variables to enable this are in place. Also note that I have done absolutely nothing about the Windows target to get something similar. On the other hand, DLLs are already the default there, but without versioning, and I've no idea what the possibilities for such a thing are there...
2000-07-21 23:08:53 +08:00
{
Rework the system to generate shared libraries: - Make note of the expected extension for the shared libraries and if there is a need for symbolic links from for example libcrypto.so.0 to libcrypto.so.0.9.7. There is extended info in Configure for that. - Make as few rebuilds of the shared libraries as possible. - Still avoid linking the OpenSSL programs with the shared libraries. - When installing, install the shared libraries separately from the static ones.
2000-10-13 23:25:06 +08:00
if ($shared_cflag ne "")
{
Complete integration of SPARC assembler DES implementation. Tested on Solaris only. I'll keep my eyes open for Linux and OpenBSD targets.
2003-01-03 07:26:46 +08:00
$cflags = "$shared_cflag -DOPENSSL_PIC $cflags";
Rework the system to generate shared libraries: - Make note of the expected extension for the shared libraries and if there is a need for symbolic links from for example libcrypto.so.0 to libcrypto.so.0.9.7. There is extended info in Configure for that. - Make as few rebuilds of the shared libraries as possible. - Still avoid linking the OpenSSL programs with the shared libraries. - When installing, install the shared libraries separately from the static ones.
2000-10-13 23:25:06 +08:00
}
Make sure that shareable libraries are turned off if we don't know how to make them...
2000-10-09 08:48:30 +08:00
}
Redo and enhance the support for building shared libraries. Currently there's support for building under Linux and True64 (using examples from the programming manuals), including versioning that is currently the same as OpenSSL versions but should really be a different series. With this change, it's up to the users to decide if they want shared libraries as well as the static ones. This decision now has to be done at configuration time (well, not really, those who know what they do can still do it the same way as before). The OpenSSL programs (openssl and the test programs) are currently always linked statically, but this may change in the future in a configurable manner. The necessary makefile variables to enable this are in place. Also note that I have done absolutely nothing about the Windows target to get something similar. On the other hand, DLLs are already the default there, but without versioning, and I've no idea what the possibilities for such a thing are there...
2000-07-21 23:08:53 +08:00
Step 6 of move of engines: rename the macro ENGINE_DYNAMIC_SUPPORT to OPENSSL_NO_DYNAMIC_ENGINE and make sure that gets defined unless shared library support has been specifically requested.
2002-10-12 02:06:08 +08:00
if ($no_shared)
{
$openssl_other_defines.="#define OPENSSL_NO_DYNAMIC_ENGINE\n";
}
For the mkdef.pl to recognise if the "external" engines are inserted into libcrypto, we need the "algorithm" STATIC_ENGINE.
2002-10-25 03:09:03 +08:00
else
{
$openssl_other_defines.="#define OPENSSL_NO_STATIC_ENGINE\n";
}
Step 6 of move of engines: rename the macro ENGINE_DYNAMIC_SUPPORT to OPENSSL_NO_DYNAMIC_ENGINE and make sure that gets defined unless shared library support has been specifically requested.
2002-10-12 02:06:08 +08:00
This is essentially Intel 32-bit compiler tune-up. To start with all available compiler versions generated bogus machine code trying to compile new crypto/des/cfb_enc.c. Secondly, 8th version defines __GNUC__ macro, but fails to compile *some* inline assembler correctly. Note that all versions of icc implement MSC-like _lrot[rl] intrinsic, which is used now instead of offensive asm. Finally, unnecessary linker dependencies are eliminated. Most notably dependency from libirc.a caused trouble at application start-up, if libcrypto.so is linked with -Bsymbolic (which it is).
2004-03-29 05:27:47 +08:00
# Compiler fix-ups
if ($target =~ /icc$/)
{
my($iccver)=`$cc -V 2>&1`;
if ($iccver =~ /Version ([0-9]+)\./) { $iccver=$1; }
else { $iccver=0; }
if ($iccver>=8)
{
# Eliminate unnecessary dependency from libirc.a. This is
# essential for shared library support, as otherwise
# apps/openssl can end up in endless loop upon startup...
$cflags.=" -Dmemcpy=__builtin_memcpy -Dmemset=__builtin_memset";
}
}
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
if ($sys_id ne "")
{
Don't put configuration macro definitions on the command line, we're just fooling ourselves and then screwing up for other applications.
2003-03-21 07:52:41 +08:00
#$cflags="-DOPENSSL_SYSNAME_$sys_id $cflags";
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
$openssl_sys_defines="#define OPENSSL_SYSNAME_$sys_id\n";
}
Workaround for solaris64 linking problem (explicit "ar rs" is needed to create a symbol table). Sun patches such as 109147-06 probably fix this problem, but we can easily avoid it.
2001-03-09 21:04:06 +08:00
if ($ranlib eq "")
{
$ranlib = $default_ranlib;
}
Patch to Configure script. For some reason the BN_ASM part was truncated to the first word which broke (at least) the Linux compile. Hopefully this wont break other platforms.
1999-02-15 06:47:21 +08:00
#my ($bn1)=split(/\s+/,$bn_obj);
#$bn1 = "" unless defined $bn1;
#$bn1=$bn_asm unless ($bn1 =~ /\.o$/);
#$bn_obj="$bn1";
If they ask for 386, keep it as 386 as possible...
2004-08-03 06:41:19 +08:00
$cpuid_obj="" if ($processor==386);
Patch to Configure script. For some reason the BN_ASM part was truncated to the first word which broke (at least) the Linux compile. Hopefully this wont break other platforms.
1999-02-15 06:47:21 +08:00
$bn_obj = $bn_asm unless $bn_obj ne "";
Finalizing asm support for UnixWare, SCO, OpenUnix... Note that I've replaced #if logic around bn_sub_part_words in bn_mul.c. I rely upon OPENSSL_BN_ASM_PART_WORDS being added by ./Configure script. Would it still work on non-Unix platforms?
2003-01-09 16:42:04 +08:00
# bn86* is the only one implementing bn_*_part_words
$cflags.=" -DOPENSSL_BN_ASM_PART_WORDS" if ($bn_obj =~ /bn86/);
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
$cflags.=" -DOPENSSL_IA32_SSE2" if (!$no_sse2 && $bn_obj =~ /bn86/);
Import of old SSLeay release: SSLeay 0.9.1b (unreleased)
1998-12-21 19:00:56 +08:00
Import of old SSLeay release: SSLeay 0.9.0b
1998-12-21 18:56:39 +08:00
$des_obj=$des_enc unless ($des_obj =~ /\.o$/);
$bf_obj=$bf_enc unless ($bf_obj =~ /\.o$/);
$cast_obj=$cast_enc unless ($cast_obj =~ /\.o$/);
$rc4_obj=$rc4_enc unless ($rc4_obj =~ /\.o$/);
$rc5_obj=$rc5_enc unless ($rc5_obj =~ /\.o$/);
if ($sha1_obj =~ /\.o$/)
{
# $sha1_obj=$sha1_enc;
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
$cflags.=" -DSHA1_ASM" if ($sha1_obj =~ /sx86/ || $sha1_obj =~ /sha1/);
$cflags.=" -DSHA256_ASM" if ($sha1_obj =~ /sha256/);
$cflags.=" -DSHA512_ASM" if ($sha1_obj =~ /sha512/);
if ($sha1_obj =~ /x86/)
{ if ($no_sse2)
{ $sha1_obj =~ s/\S*sse2\S+//; }
elsif ($cflags !~ /OPENSSL_IA32_SSE2/)
{ $cflags.=" -DOPENSSL_IA32_SSE2"; }
}
Import of old SSLeay release: SSLeay 0.9.0b
1998-12-21 18:56:39 +08:00
}
if ($md5_obj =~ /\.o$/)
{
# $md5_obj=$md5_enc;
$cflags.=" -DMD5_ASM";
}
if ($rmd160_obj =~ /\.o$/)
{
# $rmd160_obj=$rmd160_enc;
$cflags.=" -DRMD160_ASM";
}
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
if ($aes_obj =~ /\.o$/)
{
$cflags.=" -DAES_ASM";
}
Permit "monolithic" AES assembler implementations, i.e. such which would replace *whole* aes_core.c, not only AES_[de|en]crypt routines.
2005-01-10 00:01:58 +08:00
else {
$aes_obj=$aes_enc;
}
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
Reformat the CFLAG string so it can be made part of a C string. Incidently, this works pretty well on the command line as well. PR: 52
2002-05-31 02:06:35 +08:00
# "Stringify" the C flags string. This permits it to be made part of a string
# and works as well on command lines.
$cflags =~ s/([\\\"])/\\\1/g;
Linux shared libraries.
1999-04-29 00:16:31 +08:00
my $version = "unknown";
Add version info to Win32 DLLs. We might want to edit the strings a bit... Maybe add to 0.9.7 too?
2002-10-05 05:22:47 +08:00
my $version_num = "unknown";
Linux shared libraries.
1999-04-29 00:16:31 +08:00
my $major = "unknown";
my $minor = "unknown";
Redo and enhance the support for building shared libraries. Currently there's support for building under Linux and True64 (using examples from the programming manuals), including versioning that is currently the same as OpenSSL versions but should really be a different series. With this change, it's up to the users to decide if they want shared libraries as well as the static ones. This decision now has to be done at configuration time (well, not really, those who know what they do can still do it the same way as before). The OpenSSL programs (openssl and the test programs) are currently always linked statically, but this may change in the future in a configurable manner. The necessary makefile variables to enable this are in place. Also note that I have done absolutely nothing about the Windows target to get something similar. On the other hand, DLLs are already the default there, but without versioning, and I've no idea what the possibilities for such a thing are there...
2000-07-21 23:08:53 +08:00
my $shlib_version_number = "unknown";
my $shlib_version_history = "unknown";
my $shlib_major = "unknown";
my $shlib_minor = "unknown";
Linux shared libraries.
1999-04-29 00:16:31 +08:00
open(IN,'<crypto/opensslv.h') || die "unable to read opensslv.h:$!\n";
while (<IN>)
{
$version=$1 if /OPENSSL.VERSION.TEXT.*OpenSSL (\S+) /;
Add version info to Win32 DLLs. We might want to edit the strings a bit... Maybe add to 0.9.7 too?
2002-10-05 05:22:47 +08:00
$version_num=$1 if /OPENSSL.VERSION.NUMBER.*0x(\S+)/;
Redo and enhance the support for building shared libraries. Currently there's support for building under Linux and True64 (using examples from the programming manuals), including versioning that is currently the same as OpenSSL versions but should really be a different series. With this change, it's up to the users to decide if they want shared libraries as well as the static ones. This decision now has to be done at configuration time (well, not really, those who know what they do can still do it the same way as before). The OpenSSL programs (openssl and the test programs) are currently always linked statically, but this may change in the future in a configurable manner. The necessary makefile variables to enable this are in place. Also note that I have done absolutely nothing about the Windows target to get something similar. On the other hand, DLLs are already the default there, but without versioning, and I've no idea what the possibilities for such a thing are there...
2000-07-21 23:08:53 +08:00
$shlib_version_number=$1 if /SHLIB_VERSION_NUMBER *"([^"]+)"/;
$shlib_version_history=$1 if /SHLIB_VERSION_HISTORY *"([^"]*)"/;
Linux shared libraries.
1999-04-29 00:16:31 +08:00
}
close(IN);
Redo and enhance the support for building shared libraries. Currently there's support for building under Linux and True64 (using examples from the programming manuals), including versioning that is currently the same as OpenSSL versions but should really be a different series. With this change, it's up to the users to decide if they want shared libraries as well as the static ones. This decision now has to be done at configuration time (well, not really, those who know what they do can still do it the same way as before). The OpenSSL programs (openssl and the test programs) are currently always linked statically, but this may change in the future in a configurable manner. The necessary makefile variables to enable this are in place. Also note that I have done absolutely nothing about the Windows target to get something similar. On the other hand, DLLs are already the default there, but without versioning, and I've no idea what the possibilities for such a thing are there...
2000-07-21 23:08:53 +08:00
if ($shlib_version_history ne "") { $shlib_version_history .= ":"; }
Linux shared libraries.
1999-04-29 00:16:31 +08:00
Detect Siemens platforms. Submitted by: Martin Kraemer <Martin.Kraemer@MchP.Siemens.De>
1999-05-21 01:28:19 +08:00
if ($version =~ /(^[0-9]*)\.([0-9\.]*)/)
Linux shared libraries.
1999-04-29 00:16:31 +08:00
{
$major=$1;
$minor=$2;
}
Redo and enhance the support for building shared libraries. Currently there's support for building under Linux and True64 (using examples from the programming manuals), including versioning that is currently the same as OpenSSL versions but should really be a different series. With this change, it's up to the users to decide if they want shared libraries as well as the static ones. This decision now has to be done at configuration time (well, not really, those who know what they do can still do it the same way as before). The OpenSSL programs (openssl and the test programs) are currently always linked statically, but this may change in the future in a configurable manner. The necessary makefile variables to enable this are in place. Also note that I have done absolutely nothing about the Windows target to get something similar. On the other hand, DLLs are already the default there, but without versioning, and I've no idea what the possibilities for such a thing are there...
2000-07-21 23:08:53 +08:00
if ($shlib_version_number =~ /(^[0-9]*)\.([0-9\.]*)/)
{
$shlib_major=$1;
$shlib_minor=$2;
}
Make Configure work again after eliminating files.
1998-12-29 06:05:44 +08:00
open(IN,'<Makefile.org') || die "unable to read Makefile.org:$!\n";
For those wanting to build for several platforms with the same source directory, making a separate directory tree with lots of symbolic links seems to be the solution. Unfortunatelt, Configure doesn't take appropriate steps to support this solution (as in removing a file that's going to be rewritten). This change corrects that situation. Now I just have to find all other places where there's lack of support for this.
2002-07-16 16:53:32 +08:00
unlink("$Makefile.new") || die "unable to remove old $Makefile.new:$!\n" if -e "$Makefile.new";
open(OUT,">$Makefile.new") || die "unable to create $Makefile.new:$!\n";
Make sure that generated files are labeled as such (except in dofile(), for now)
2000-01-11 21:07:21 +08:00
print OUT "### Generated automatically from Makefile.org by Configure.\n\n";
New Configure option no-<cipher> (rsa, idea, rc5, ...).
1999-04-27 09:14:46 +08:00
my $sdirs=0;
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
while (<IN>)
{
chop;
New Configure option no-<cipher> (rsa, idea, rc5, ...).
1999-04-27 09:14:46 +08:00
$sdirs = 1 if /^SDIRS=/;
if ($sdirs) {
my $dir;
foreach $dir (@skip) {
Certain missing algorithms make some SSL versions or TLS impossible to build.
2001-11-27 19:48:30 +08:00
s/([ ])$dir /\1/;
New Configure option no-<cipher> (rsa, idea, rc5, ...).
1999-04-27 09:14:46 +08:00
}
}
Did not check the last SDIR line for excluded algorithms.
1999-07-16 08:49:51 +08:00
$sdirs = 0 unless /\\$/;
Linux shared libraries.
1999-04-29 00:16:31 +08:00
s/^VERSION=.*/VERSION=$version/;
s/^MAJOR=.*/MAJOR=$major/;
s/^MINOR=.*/MINOR=$minor/;
Redo and enhance the support for building shared libraries. Currently there's support for building under Linux and True64 (using examples from the programming manuals), including versioning that is currently the same as OpenSSL versions but should really be a different series. With this change, it's up to the users to decide if they want shared libraries as well as the static ones. This decision now has to be done at configuration time (well, not really, those who know what they do can still do it the same way as before). The OpenSSL programs (openssl and the test programs) are currently always linked statically, but this may change in the future in a configurable manner. The necessary makefile variables to enable this are in place. Also note that I have done absolutely nothing about the Windows target to get something similar. On the other hand, DLLs are already the default there, but without versioning, and I've no idea what the possibilities for such a thing are there...
2000-07-21 23:08:53 +08:00
s/^SHLIB_VERSION_NUMBER=.*/SHLIB_VERSION_NUMBER=$shlib_version_number/;
s/^SHLIB_VERSION_HISTORY=.*/SHLIB_VERSION_HISTORY=$shlib_version_history/;
s/^SHLIB_MAJOR=.*/SHLIB_MAJOR=$shlib_major/;
s/^SHLIB_MINOR=.*/SHLIB_MINOR=$shlib_minor/;
Rework the system to generate shared libraries: - Make note of the expected extension for the shared libraries and if there is a need for symbolic links from for example libcrypto.so.0 to libcrypto.so.0.9.7. There is extended info in Configure for that. - Make as few rebuilds of the shared libraries as possible. - Still avoid linking the OpenSSL programs with the shared libraries. - When installing, install the shared libraries separately from the static ones.
2000-10-13 23:25:06 +08:00
s/^SHLIB_EXT=.*/SHLIB_EXT=$shared_extension/;
Support INSTALL_PREFIX for packagers. Submitted by: Reviewed by: PR:
1999-04-30 05:52:08 +08:00
s/^INSTALLTOP=.*$/INSTALLTOP=$prefix/;
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
s/^OPENSSLDIR=.*$/OPENSSLDIR=$openssldir/;
Support INSTALL_PREFIX for packagers. Submitted by: Reviewed by: PR:
1999-04-30 05:52:08 +08:00
s/^INSTALL_PREFIX=.*$/INSTALL_PREFIX=$install_prefix/;
Import of old SSLeay release: SSLeay 0.9.1b (unreleased)
1998-12-21 19:00:56 +08:00
s/^PLATFORM=.*$/PLATFORM=$target/;
mk1mf.pl and mkdef.pl read OPTIONS from toplevel Makefile. Configure no longer changes files in place.
1999-05-13 18:28:14 +08:00
s/^OPTIONS=.*$/OPTIONS=$options/;
Allow reconfiguration. This can be useful if some source update requires that you configure again, but you don't want to reenter all those configuration arguments again.
2000-08-17 18:23:45 +08:00
s/^CONFIGURE_ARGS=.*$/CONFIGURE_ARGS=$argvstring/;
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
s/^CC=.*$/CC= $cc/;
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
s/^MAKEDEPPROG=.*$/MAKEDEPPROG= $cc/ if $cc eq "gcc";
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
s/^CFLAG=.*$/CFLAG= $cflags/;
New Configure option no-<cipher> (rsa, idea, rc5, ...).
1999-04-27 09:14:46 +08:00
s/^DEPFLAG=.*$/DEPFLAG= $depflags/;
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
s/^EX_LIBS=.*$/EX_LIBS= $lflags/;
Incorporate some changes that make OpenSSL compilable in CygWin.
2001-04-04 23:50:30 +08:00
s/^EXE_EXT=.*$/EXE_EXT= $exe_ext/;
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
s/^CPUID_OBJ=.*$/CPUID_OBJ= $cpuid_obj/;
Actually use BN when selected!
1999-02-13 03:33:55 +08:00
s/^BN_ASM=.*$/BN_ASM= $bn_obj/;
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
s/^DES_ENC=.*$/DES_ENC= $des_obj/;
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
s/^AES_ASM_OBJ=.*$/AES_ASM_OBJ= $aes_obj/;
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
s/^BF_ENC=.*$/BF_ENC= $bf_obj/;
Import of old SSLeay release: SSLeay 0.9.0b
1998-12-21 18:56:39 +08:00
s/^CAST_ENC=.*$/CAST_ENC= $cast_obj/;
s/^RC4_ENC=.*$/RC4_ENC= $rc4_obj/;
s/^RC5_ENC=.*$/RC5_ENC= $rc5_obj/;
s/^MD5_ASM_OBJ=.*$/MD5_ASM_OBJ= $md5_obj/;
s/^SHA1_ASM_OBJ=.*$/SHA1_ASM_OBJ= $sha1_obj/;
s/^RMD160_ASM_OBJ=.*$/RMD160_ASM_OBJ= $rmd160_obj/;
New option to generate 80386 code.
1999-03-31 20:38:27 +08:00
s/^PROCESSOR=.*/PROCESSOR= $processor/;
New Makefile variables $(RANLIB) and $(PERL).
1999-04-01 20:34:33 +08:00
s/^RANLIB=.*/RANLIB= $ranlib/;
Add aix64-cc, and make sure that ar gets proper flags for 64-bit libraries
2002-08-02 04:28:16 +08:00
s/^ARFLAGS=.*/ARFLAGS= $arflags/;
New Makefile variables $(RANLIB) and $(PERL).
1999-04-01 20:34:33 +08:00
s/^PERL=.*/PERL= $perl/;
First tentative impementation of Kerberos 5 cryptos and keys for SSL/TLS. Implemented by Vern Staats <staatsvr@asc.hpc.mil>, further hacked and distributed by Jeffrey Altman <jaltnab@columbia.edu>
2000-12-01 06:53:34 +08:00
s/^KRB5_INCLUDES=.*/KRB5_INCLUDES=$withargs{"krb5-include"}/;
Handle krb5 libraries separately and make sure only libssl.so depends on it.
2003-02-14 21:12:00 +08:00
s/^LIBKRB5=.*/LIBKRB5=$withargs{"krb5-lib"}/;
Redo and enhance the support for building shared libraries. Currently there's support for building under Linux and True64 (using examples from the programming manuals), including versioning that is currently the same as OpenSSL versions but should really be a different series. With this change, it's up to the users to decide if they want shared libraries as well as the static ones. This decision now has to be done at configuration time (well, not really, those who know what they do can still do it the same way as before). The OpenSSL programs (openssl and the test programs) are currently always linked statically, but this may change in the future in a configurable manner. The necessary makefile variables to enable this are in place. Also note that I have done absolutely nothing about the Windows target to get something similar. On the other hand, DLLs are already the default there, but without versioning, and I've no idea what the possibilities for such a thing are there...
2000-07-21 23:08:53 +08:00
s/^SHLIB_TARGET=.*/SHLIB_TARGET=$shared_target/;
Rework the system to generate shared libraries: - Make note of the expected extension for the shared libraries and if there is a need for symbolic links from for example libcrypto.so.0 to libcrypto.so.0.9.7. There is extended info in Configure for that. - Make as few rebuilds of the shared libraries as possible. - Still avoid linking the OpenSSL programs with the shared libraries. - When installing, install the shared libraries separately from the static ones.
2000-10-13 23:25:06 +08:00
s/^SHLIB_MARK=.*/SHLIB_MARK=$shared_mark/;
s/^SHARED_LIBS=.*/SHARED_LIBS=\$(SHARED_CRYPTO) \$(SHARED_SSL)/ if (!$no_shared);
Add what's needed to get shared libraries on HP-UX. N.B.: This has not been tested at all, that's my next step.
2000-10-22 05:24:11 +08:00
if ($shared_extension ne "" && $shared_extension =~ /^\.s([ol])\.[^\.]*$/)
{
my $sotmp = $1;
Correct darwin shared library support. PR: 39
2002-05-30 20:16:33 +08:00
s/^SHARED_LIBS_LINK_EXTS=.*/SHARED_LIBS_LINK_EXTS=.s$sotmp/;
}
elsif ($shared_extension ne "" && $shared_extension =~ /^\.[^\.]*\.dylib$/)
{
s/^SHARED_LIBS_LINK_EXTS=.*/SHARED_LIBS_LINK_EXTS=.dylib/;
Add what's needed to get shared libraries on HP-UX. N.B.: This has not been tested at all, that's my next step.
2000-10-22 05:24:11 +08:00
}
elsif ($shared_extension ne "" && $shared_extension =~ /^\.s([ol])\.[^\.]*\.[^\.]*$/)
{
my $sotmp = $1;
s/^SHARED_LIBS_LINK_EXTS=.*/SHARED_LIBS_LINK_EXTS=.s$sotmp.\$(SHLIB_MAJOR) .s$sotmp/;
}
Correct darwin shared library support. PR: 39
2002-05-30 20:16:33 +08:00
elsif ($shared_extension ne "" && $shared_extension =~ /^\.[^\.]*\.[^\.]*\.dylib$/)
{
s/^SHARED_LIBS_LINK_EXTS=.*/SHARED_LIBS_LINK_EXTS=.\$(SHLIB_MAJOR).dylib .dylib/;
}
On Solaris64, cc needs the flag -xarch=v9 when linking shared libraries. Make a general change to support shared library linking flags in general. Noted by Nick Briggs <briggs@parc.xerox.com>
2002-01-02 18:30:07 +08:00
s/^SHARED_LDFLAGS=.*/SHARED_LDFLAGS=$shared_ldflag/;
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
print OUT $_."\n";
}
close(IN);
close(OUT);
For those wanting to build for several platforms with the same source directory, making a separate directory tree with lots of symbolic links seems to be the solution. Unfortunatelt, Configure doesn't take appropriate steps to support this solution (as in removing a file that's going to be rewritten). This change corrects that situation. Now I just have to find all other places where there's lack of support for this.
2002-07-16 16:53:32 +08:00
rename($Makefile,"$Makefile.bak") || die "unable to rename $Makefile\n" if -e $Makefile;
rename("$Makefile.new",$Makefile) || die "unable to rename $Makefile.new\n";
Make Configure work again after eliminating files.
1998-12-29 06:05:44 +08:00
Import of old SSLeay release: SSLeay 0.9.0b
1998-12-21 18:56:39 +08:00
print "CC =$cc\n";
print "CFLAG =$cflags\n";
print "EX_LIBS =$lflags\n";
Actually use BN when selected!
1999-02-13 03:33:55 +08:00
print "BN_ASM =$bn_obj\n";
Import of old SSLeay release: SSLeay 0.9.0b
1998-12-21 18:56:39 +08:00
print "DES_ENC =$des_obj\n";
print "BF_ENC =$bf_obj\n";
print "CAST_ENC =$cast_obj\n";
print "RC4_ENC =$rc4_obj\n";
print "RC5_ENC =$rc5_obj\n";
print "MD5_OBJ_ASM =$md5_obj\n";
print "SHA1_OBJ_ASM =$sha1_obj\n";
print "RMD160_OBJ_ASM=$rmd160_obj\n";
New option to generate 80386 code.
1999-03-31 20:38:27 +08:00
print "PROCESSOR =$processor\n";
New Makefile variables $(RANLIB) and $(PERL).
1999-04-01 20:34:33 +08:00
print "RANLIB =$ranlib\n";
Add aix64-cc, and make sure that ar gets proper flags for 64-bit libraries
2002-08-02 04:28:16 +08:00
print "ARFLAGS =$arflags\n";
New Makefile variables $(RANLIB) and $(PERL).
1999-04-01 20:34:33 +08:00
print "PERL =$perl\n";
First tentative impementation of Kerberos 5 cryptos and keys for SSL/TLS. Implemented by Vern Staats <staatsvr@asc.hpc.mil>, further hacked and distributed by Jeffrey Altman <jaltnab@columbia.edu>
2000-12-01 06:53:34 +08:00
print "KRB5_INCLUDES =",$withargs{"krb5-include"},"\n"
if $withargs{"krb5-include"} ne "";
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
Add strictness, fix variable substition bugs.
1998-12-29 01:08:48 +08:00
my $des_ptr=0;
my $des_risc1=0;
my $des_risc2=0;
my $des_unroll=0;
my $bn_ll=0;
my $def_int=2;
my $rc4_int=$def_int;
my $md2_int=$def_int;
my $idea_int=$def_int;
my $rc2_int=$def_int;
my $rc4_idx=0;
RC4 tune-up featuring 30-40% performance improvement on most RISC platforms. See crypto/rc4/rc4_enc.c for further details.
1999-10-07 20:03:59 +08:00
my $rc4_chunk=0;
Add strictness, fix variable substition bugs.
1998-12-29 01:08:48 +08:00
my $bf_ptr=0;
my @type=("char","short","int","long");
my ($b64l,$b64,$b32,$b16,$b8)=(0,0,1,0,0);
Introduce the possibility to access global variables through functions on platform were that's the best way to handle exporting global variables in shared libraries. To enable this functionality, one must configure with "EXPORT_VAR_AS_FN" or defined the C macro "OPENSSL_EXPORT_VAR_AS_FUNCTION" in crypto/opensslconf.h (the latter is normally done by Configure or something similar). To implement a global variable, use the macro OPENSSL_IMPLEMENT_GLOBAL in the source file (foo.c) like this: OPENSSL_IMPLEMENT_GLOBAL(int,foo)=1; OPENSSL_IMPLEMENT_GLOBAL(double,bar); To declare a global variable, use the macros OPENSSL_DECLARE_GLOBAL and OPENSSL_GLOBAL_REF in the header file (foo.h) like this: OPENSSL_DECLARE_GLOBAL(int,foo); #define foo OPENSSL_GLOBAL_REF(foo) OPENSSL_DECLARE_GLOBAL(double,bar); #define bar OPENSSL_GLOBAL_REF(bar) The #defines are very important, and therefore so is including the header file everywere where the defined globals are used. The macro OPENSSL_EXPORT_VAR_AS_FUNCTION also affects the definition of ASN.1 items, but that structure is a bt different. The largest change is in util/mkdef.pl which has been enhanced with better and easier to understand logic to choose which symbols should go into the Windows .def files as well as a number of fixes and code cleanup (among others, algorithm keywords are now sorted lexicographically to avoid constant rewrites).
2001-03-02 18:38:19 +08:00
my $export_var_as_fn=0;
Add strictness, fix variable substition bugs.
1998-12-29 01:08:48 +08:00
my $des_int;
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
foreach (sort split(/\s+/,$bn_ops))
{
$des_ptr=1 if /DES_PTR/;
$des_risc1=1 if /DES_RISC1/;
$des_risc2=1 if /DES_RISC2/;
$des_unroll=1 if /DES_UNROLL/;
$des_int=1 if /DES_INT/;
$bn_ll=1 if /BN_LLONG/;
$rc4_int=0 if /RC4_CHAR/;
$rc4_int=3 if /RC4_LONG/;
$rc4_idx=1 if /RC4_INDEX/;
RC4 tune-up featuring 30-40% performance improvement on most RISC platforms. See crypto/rc4/rc4_enc.c for further details.
1999-10-07 20:03:59 +08:00
$rc4_chunk=1 if /RC4_CHUNK/;
$rc4_chunk=2 if /RC4_CHUNK_LL/;
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
$md2_int=0 if /MD2_CHAR/;
$md2_int=3 if /MD2_LONG/;
$idea_int=1 if /IDEA_SHORT/;
$idea_int=3 if /IDEA_LONG/;
$rc2_int=1 if /RC2_SHORT/;
$rc2_int=3 if /RC2_LONG/;
$bf_ptr=1 if $_ eq "BF_PTR";
$bf_ptr=2 if $_ eq "BF_PTR2";
($b64l,$b64,$b32,$b16,$b8)=(0,1,0,0,0) if /SIXTY_FOUR_BIT/;
Import of old SSLeay release: SSLeay 0.9.0b
1998-12-21 18:56:39 +08:00
($b64l,$b64,$b32,$b16,$b8)=(1,0,0,0,0) if /SIXTY_FOUR_BIT_LONG/;
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
($b64l,$b64,$b32,$b16,$b8)=(0,0,1,0,0) if /THIRTY_TWO_BIT/;
($b64l,$b64,$b32,$b16,$b8)=(0,0,0,1,0) if /SIXTEEN_BIT/;
($b64l,$b64,$b32,$b16,$b8)=(0,0,0,0,1) if /EIGHT_BIT/;
Introduce the possibility to access global variables through functions on platform were that's the best way to handle exporting global variables in shared libraries. To enable this functionality, one must configure with "EXPORT_VAR_AS_FN" or defined the C macro "OPENSSL_EXPORT_VAR_AS_FUNCTION" in crypto/opensslconf.h (the latter is normally done by Configure or something similar). To implement a global variable, use the macro OPENSSL_IMPLEMENT_GLOBAL in the source file (foo.c) like this: OPENSSL_IMPLEMENT_GLOBAL(int,foo)=1; OPENSSL_IMPLEMENT_GLOBAL(double,bar); To declare a global variable, use the macros OPENSSL_DECLARE_GLOBAL and OPENSSL_GLOBAL_REF in the header file (foo.h) like this: OPENSSL_DECLARE_GLOBAL(int,foo); #define foo OPENSSL_GLOBAL_REF(foo) OPENSSL_DECLARE_GLOBAL(double,bar); #define bar OPENSSL_GLOBAL_REF(bar) The #defines are very important, and therefore so is including the header file everywere where the defined globals are used. The macro OPENSSL_EXPORT_VAR_AS_FUNCTION also affects the definition of ASN.1 items, but that structure is a bt different. The largest change is in util/mkdef.pl which has been enhanced with better and easier to understand logic to choose which symbols should go into the Windows .def files as well as a number of fixes and code cleanup (among others, algorithm keywords are now sorted lexicographically to avoid constant rewrites).
2001-03-02 18:38:19 +08:00
$export_var_as_fn=1 if /EXPORT_VAR_AS_FN/;
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
}
Move all autogenerated header file parts to crypto/opensslconf.h.
1999-04-22 01:31:05 +08:00
open(IN,'<crypto/opensslconf.h.in') || die "unable to read crypto/opensslconf.h.in:$!\n";
For those wanting to build for several platforms with the same source directory, making a separate directory tree with lots of symbolic links seems to be the solution. Unfortunatelt, Configure doesn't take appropriate steps to support this solution (as in removing a file that's going to be rewritten). This change corrects that situation. Now I just have to find all other places where there's lack of support for this.
2002-07-16 16:53:32 +08:00
unlink("crypto/opensslconf.h.new") || die "unable to remove old crypto/opensslconf.h.new:$!\n" if -e "crypto/opensslconf.h.new";
open(OUT,'>crypto/opensslconf.h.new') || die "unable to create crypto/opensslconf.h.new:$!\n";
Avoid potential conflicts between #defines in opensslconf.h and defines when compiling applications, and allow applications to select what #defines to enable -- OPENSSL_EXLUCDE_DEFINES enables the "#define NO_whatever" stuff only, which avoids potential severe confusion caused by "#define _REENTRANT" when opensslconf.h is not the first header file #included.
2000-02-18 19:15:32 +08:00
print OUT "/* opensslconf.h */\n";
Make Configure add the configuration options that it was copmiled with. Currently, those defines are protected with a OPENSSL_EXCLUDED. That may not be the best strategy, but it will do for now.
2000-02-18 17:11:37 +08:00
print OUT "/* WARNING: Generated automatically from opensslconf.h.in by Configure. */\n\n";
Avoid potential conflicts between #defines in opensslconf.h and defines when compiling applications, and allow applications to select what #defines to enable -- OPENSSL_EXLUCDE_DEFINES enables the "#define NO_whatever" stuff only, which avoids potential severe confusion caused by "#define _REENTRANT" when opensslconf.h is not the first header file #included.
2000-02-18 19:15:32 +08:00
print OUT "/* OpenSSL was configured with the following options: */\n";
Define the OPENSSL_NO_* macros as NO_* macros for the sake of applications thathaven't yet been changed
2001-02-23 01:36:41 +08:00
my $openssl_algorithm_defines_trans = $openssl_algorithm_defines;
$openssl_algorithm_defines_trans =~ s/^\s*#\s*define\s+OPENSSL_(.*)/# if defined(OPENSSL_$1) \&\& !defined($1)\n# define $1\n# endif/mg;
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
$openssl_algorithm_defines =~ s/^\s*#\s*define\s+(.*)/#ifndef $1\n# define $1\n#endif/mg;
Keep variable names consistent with corresponding pre-processor symbols.
2000-02-19 05:08:37 +08:00
$openssl_algorithm_defines = " /* no ciphers excluded */\n" if $openssl_algorithm_defines eq "";
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
$openssl_thread_defines =~ s/^\s*#\s*define\s+(.*)/#ifndef $1\n# define $1\n#endif/mg;
$openssl_sys_defines =~ s/^\s*#\s*define\s+(.*)/#ifndef $1\n# define $1\n#endif/mg;
$openssl_other_defines =~ s/^\s*#\s*define\s+(.*)/#ifndef $1\n# define $1\n#endif/mg;
print OUT $openssl_sys_defines;
This ghastly hack prevents CVS wars over Kerberos (which is disabled by default).
2001-08-04 02:45:35 +08:00
print OUT "#ifndef OPENSSL_DOING_MAKEDEPEND\n\n";
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
print OUT $openssl_algorithm_defines;
This ghastly hack prevents CVS wars over Kerberos (which is disabled by default).
2001-08-04 02:45:35 +08:00
print OUT "\n#endif /* OPENSSL_DOING_MAKEDEPEND */\n";
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
print OUT $openssl_thread_defines;
print OUT $openssl_other_defines,"\n";
Avoid potential conflicts between #defines in opensslconf.h and defines when compiling applications, and allow applications to select what #defines to enable -- OPENSSL_EXLUCDE_DEFINES enables the "#define NO_whatever" stuff only, which avoids potential severe confusion caused by "#define _REENTRANT" when opensslconf.h is not the first header file #included.
2000-02-18 19:15:32 +08:00
Define the OPENSSL_NO_* macros as NO_* macros for the sake of applications thathaven't yet been changed
2001-02-23 01:36:41 +08:00
print OUT "/* The OPENSSL_NO_* macros are also defined as NO_* if the application\n";
print OUT " asks for it. This is a transient feature that is provided for those\n";
print OUT " who haven't had the time to do the appropriate changes in their\n";
print OUT " applications. */\n";
print OUT "#ifdef OPENSSL_ALGORITHM_DEFINES\n";
print OUT $openssl_algorithm_defines_trans;
print OUT "#endif\n\n";
This is so to say "damage control" for jumbo "cpuid" patch, see http://cvs.openssl.org/chngview?cn=12493. Now all platform should be operational, while SSE2 code pathes get engaged on ELF platforms only.
2004-07-27 06:01:50 +08:00
print OUT "#define OPENSSL_CPUID_OBJ\n\n" if ($cpuid_obj);
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
while (<IN>)
{
Fixed a typo. Submitted by: Reviewed by: PR:
1999-04-25 18:28:26 +08:00
if (/^#define\s+OPENSSLDIR/)
Submitted by: Reviewed by: PR:
1999-04-25 19:01:07 +08:00
{ print OUT "#define OPENSSLDIR \"$openssldir\"\n"; }
This fixes the installation target for dynamic engines, which was trying to install to a different location than it had created. (BTW, VMS will need a matching fix in eng_list.c.) Note, these aren't ssl-specific, so I'm putting "engines/" into the libs directory rather than at the "--prefix" level or inside "ssl/".
2004-06-01 11:18:58 +08:00
elsif (/^#define\s+ENGINESDIR/)
{ print OUT "#define ENGINESDIR \"$prefix/lib/engines\"\n"; }
Introduce the possibility to access global variables through functions on platform were that's the best way to handle exporting global variables in shared libraries. To enable this functionality, one must configure with "EXPORT_VAR_AS_FN" or defined the C macro "OPENSSL_EXPORT_VAR_AS_FUNCTION" in crypto/opensslconf.h (the latter is normally done by Configure or something similar). To implement a global variable, use the macro OPENSSL_IMPLEMENT_GLOBAL in the source file (foo.c) like this: OPENSSL_IMPLEMENT_GLOBAL(int,foo)=1; OPENSSL_IMPLEMENT_GLOBAL(double,bar); To declare a global variable, use the macros OPENSSL_DECLARE_GLOBAL and OPENSSL_GLOBAL_REF in the header file (foo.h) like this: OPENSSL_DECLARE_GLOBAL(int,foo); #define foo OPENSSL_GLOBAL_REF(foo) OPENSSL_DECLARE_GLOBAL(double,bar); #define bar OPENSSL_GLOBAL_REF(bar) The #defines are very important, and therefore so is including the header file everywere where the defined globals are used. The macro OPENSSL_EXPORT_VAR_AS_FUNCTION also affects the definition of ASN.1 items, but that structure is a bt different. The largest change is in util/mkdef.pl which has been enhanced with better and easier to understand logic to choose which symbols should go into the Windows .def files as well as a number of fixes and code cleanup (among others, algorithm keywords are now sorted lexicographically to avoid constant rewrites).
2001-03-02 18:38:19 +08:00
elsif (/^#((define)|(undef))\s+OPENSSL_EXPORT_VAR_AS_FUNCTION/)
{ printf OUT "#undef OPENSSL_EXPORT_VAR_AS_FUNCTION\n"
if $export_var_as_fn;
printf OUT "#%s OPENSSL_EXPORT_VAR_AS_FUNCTION\n",
($export_var_as_fn)?"define":"undef"; }
Some pre-POSIX systems don't have unistd.h (but e.g. lib.c). Allow configuring the name of that header file.
1999-06-09 21:23:38 +08:00
elsif (/^#define\s+OPENSSL_UNISTD/)
{
$unistd = "<unistd.h>" if $unistd eq "";
print OUT "#define OPENSSL_UNISTD $unistd\n";
}
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
elsif (/^#((define)|(undef))\s+SIXTY_FOUR_BIT_LONG/)
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
{ printf OUT "#%s SIXTY_FOUR_BIT_LONG\n",($b64l)?"define":"undef"; }
elsif (/^#((define)|(undef))\s+SIXTY_FOUR_BIT/)
{ printf OUT "#%s SIXTY_FOUR_BIT\n",($b64)?"define":"undef"; }
elsif (/^#((define)|(undef))\s+THIRTY_TWO_BIT/)
{ printf OUT "#%s THIRTY_TWO_BIT\n",($b32)?"define":"undef"; }
elsif (/^#((define)|(undef))\s+SIXTEEN_BIT/)
{ printf OUT "#%s SIXTEEN_BIT\n",($b16)?"define":"undef"; }
elsif (/^#((define)|(undef))\s+EIGHT_BIT/)
{ printf OUT "#%s EIGHT_BIT\n",($b8)?"define":"undef"; }
elsif (/^#((define)|(undef))\s+BN_LLONG\s*$/)
{ printf OUT "#%s BN_LLONG\n",($bn_ll)?"define":"undef"; }
Move all autogenerated header file parts to crypto/opensslconf.h.
1999-04-22 01:31:05 +08:00
elsif (/^\#define\s+DES_LONG\s+.*/)
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
{ printf OUT "#define DES_LONG unsigned %s\n",
($des_int)?'int':'long'; }
Move all autogenerated header file parts to crypto/opensslconf.h.
1999-04-22 01:31:05 +08:00
elsif (/^\#(define|undef)\s+DES_PTR/)
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
{ printf OUT "#%s DES_PTR\n",($des_ptr)?'define':'undef'; }
elsif (/^\#(define|undef)\s+DES_RISC1/)
{ printf OUT "#%s DES_RISC1\n",($des_risc1)?'define':'undef'; }
elsif (/^\#(define|undef)\s+DES_RISC2/)
{ printf OUT "#%s DES_RISC2\n",($des_risc2)?'define':'undef'; }
elsif (/^\#(define|undef)\s+DES_UNROLL/)
{ printf OUT "#%s DES_UNROLL\n",($des_unroll)?'define':'undef'; }
Move all autogenerated header file parts to crypto/opensslconf.h.
1999-04-22 01:31:05 +08:00
elsif (/^#define\s+RC4_INT\s/)
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
{ printf OUT "#define RC4_INT unsigned %s\n",$type[$rc4_int]; }
RC4 tune-up featuring 30-40% performance improvement on most RISC platforms. See crypto/rc4/rc4_enc.c for further details.
1999-10-07 20:03:59 +08:00
elsif (/^#undef\s+RC4_CHUNK/)
{
printf OUT "#undef RC4_CHUNK\n" if $rc4_chunk==0;
printf OUT "#define RC4_CHUNK unsigned long\n" if $rc4_chunk==1;
printf OUT "#define RC4_CHUNK unsigned long long\n" if $rc4_chunk==2;
}
Move all autogenerated header file parts to crypto/opensslconf.h.
1999-04-22 01:31:05 +08:00
elsif (/^#((define)|(undef))\s+RC4_INDEX/)
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
{ printf OUT "#%s RC4_INDEX\n",($rc4_idx)?"define":"undef"; }
SHA-1 cleanups and performance enhancements. Submitted by: Andy Polyakov <appro@fy.chalmers.se>
1999-05-05 08:23:53 +08:00
elsif (/^#(define|undef)\s+I386_ONLY/)
{ printf OUT "#%s I386_ONLY\n", ($processor == 386)?
"define":"undef"; }
Move all autogenerated header file parts to crypto/opensslconf.h.
1999-04-22 01:31:05 +08:00
elsif (/^#define\s+MD2_INT\s/)
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
{ printf OUT "#define MD2_INT unsigned %s\n",$type[$md2_int]; }
Move all autogenerated header file parts to crypto/opensslconf.h.
1999-04-22 01:31:05 +08:00
elsif (/^#define\s+IDEA_INT\s/)
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
{printf OUT "#define IDEA_INT unsigned %s\n",$type[$idea_int];}
Move all autogenerated header file parts to crypto/opensslconf.h.
1999-04-22 01:31:05 +08:00
elsif (/^#define\s+RC2_INT\s/)
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
{printf OUT "#define RC2_INT unsigned %s\n",$type[$rc2_int];}
Move all autogenerated header file parts to crypto/opensslconf.h.
1999-04-22 01:31:05 +08:00
elsif (/^#(define|undef)\s+BF_PTR/)
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
{
printf OUT "#undef BF_PTR\n" if $bf_ptr == 0;
printf OUT "#define BF_PTR\n" if $bf_ptr == 1;
printf OUT "#define BF_PTR2\n" if $bf_ptr == 2;
Move all autogenerated header file parts to crypto/opensslconf.h.
1999-04-22 01:31:05 +08:00
}
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
else
{ print OUT $_; }
}
Close files.
1999-04-30 00:22:55 +08:00
close(IN);
close(OUT);
For those wanting to build for several platforms with the same source directory, making a separate directory tree with lots of symbolic links seems to be the solution. Unfortunatelt, Configure doesn't take appropriate steps to support this solution (as in removing a file that's going to be rewritten). This change corrects that situation. Now I just have to find all other places where there's lack of support for this.
2002-07-16 16:53:32 +08:00
rename("crypto/opensslconf.h","crypto/opensslconf.h.bak") || die "unable to rename crypto/opensslconf.h\n" if -e "crypto/opensslconf.h";
rename("crypto/opensslconf.h.new","crypto/opensslconf.h") || die "unable to rename crypto/opensslconf.h.new\n";
Oops... add other changes this time too.
1999-02-14 07:13:32 +08:00
Some pre-POSIX systems don't have unistd.h (but e.g. lib.c). Allow configuring the name of that header file.
1999-06-09 21:23:38 +08:00
Oops... add other changes this time too.
1999-02-14 07:13:32 +08:00
# Fix the date
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
print "SIXTY_FOUR_BIT_LONG mode\n" if $b64l;
print "SIXTY_FOUR_BIT mode\n" if $b64;
print "THIRTY_TWO_BIT mode\n" if $b32;
print "SIXTEEN_BIT mode\n" if $b16;
print "EIGHT_BIT mode\n" if $b8;
print "DES_PTR used\n" if $des_ptr;
print "DES_RISC1 used\n" if $des_risc1;
print "DES_RISC2 used\n" if $des_risc2;
print "DES_UNROLL used\n" if $des_unroll;
print "DES_INT used\n" if $des_int;
print "BN_LLONG mode\n" if $bn_ll;
print "RC4 uses u$type[$rc4_int]\n" if $rc4_int != $def_int;
print "RC4_INDEX mode\n" if $rc4_idx;
RC4 tune-up featuring 30-40% performance improvement on most RISC platforms. See crypto/rc4/rc4_enc.c for further details.
1999-10-07 20:03:59 +08:00
print "RC4_CHUNK is undefined\n" if $rc4_chunk==0;
print "RC4_CHUNK is unsigned long\n" if $rc4_chunk==1;
print "RC4_CHUNK is unsigned long long\n" if $rc4_chunk==2;
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
print "MD2 uses u$type[$md2_int]\n" if $md2_int != $def_int;
print "IDEA uses u$type[$idea_int]\n" if $idea_int != $def_int;
print "RC2 uses u$type[$rc2_int]\n" if $rc2_int != $def_int;
print "BF_PTR used\n" if $bf_ptr == 1;
print "BF_PTR2 used\n" if $bf_ptr == 2;
Always make links.
1999-03-07 22:05:36 +08:00
Netware-specific changes, PR: 780 Submitted by: Verdon Walker <VWalker@novell.com> Reviewed by: Richard Levitte
2003-11-28 21:10:58 +08:00
if($IsMK1MF) {
avoid -DPLATFORM=\"...\" and -DCFLAGS=\"...\" command lines, use new file buildinf.h instead.
1999-07-22 04:49:15 +08:00
open (OUT,">crypto/buildinf.h") || die "Can't open buildinf.h";
printf OUT <<EOF;
Always use buildinf.h, which now includes the mk1mfinf.h data. Using different files caused problems because the dependencies in the Makefiles produced by mk1mf.pl were for the standard case, i.e. mentioned buildinf.h and not mk1mfinf.h.
1999-07-29 20:57:23 +08:00
#ifndef MK1MF_BUILD
/* auto-generated by Configure for crypto/cversion.c:
* for Unix builds, crypto/Makefile.ssl generates functional definitions;
* Windows builds (and other mk1mf builds) compile cversion.c with
* -DMK1MF_BUILD and use definitions added to this file by util/mk1mf.pl. */
#error "Windows builds (PLATFORM=$target) use mk1mf.pl-created Makefiles"
#endif
Provide CFLAGS and PLATFORM information on Windows platforms.
1999-07-09 22:04:57 +08:00
EOF
Move all autogenerated header file parts to crypto/opensslconf.h.
1999-04-22 01:31:05 +08:00
close(OUT);
Fixes so it will compile again under Win32.
1999-04-26 04:57:09 +08:00
} else {
Add support for DJGPP. PR: 75
2002-06-14 04:42:35 +08:00
my $make_command = "make -f Makefile.ssl PERL=\'$perl\'";
my $make_targets = "";
$make_targets .= " links" if $symlink;
$make_targets .= " depend" if $depflags ne "" && $make_depend;
Make dummy tests to make it easier to physically remove unwanted algorithms. This should complete PR 75
2002-07-23 21:31:04 +08:00
$make_targets .= " gentests" if $symlink;
Add support for DJGPP. PR: 75
2002-06-14 04:42:35 +08:00
(system $make_command.$make_targets) == 0 or exit $?
if $make_targets ne "";
Set #!... path to Perl in apps/der_chop automatically.
1999-05-26 06:31:49 +08:00
if ( $perl =~ m@^/@) {
Fix c_rehash script, add -fingerprint option to crl.
2000-05-18 08:33:00 +08:00
&dofile("tools/c_rehash",$perl,'^#!/', '#!%s','^my \$dir;$', 'my $dir = "' . $openssldir . '";');
Rename CA.pl to CA.pl.in (no actual changes), and let Configure set the #! line with the path to Perl. Submitted by: Peter Jones
2000-01-05 09:31:22 +08:00
&dofile("apps/CA.pl",$perl,'^#!/', '#!%s');
Set #!... path to Perl in apps/der_chop automatically.
1999-05-26 06:31:49 +08:00
} else {
# No path for Perl known ...
Fix c_rehash script, add -fingerprint option to crl.
2000-05-18 08:33:00 +08:00
&dofile("tools/c_rehash",'/usr/local/bin/perl','^#!/', '#!%s','^my \$dir;$', 'my $dir = "' . $openssldir . '";');
Rename CA.pl to CA.pl.in (no actual changes), and let Configure set the #! line with the path to Perl. Submitted by: Peter Jones
2000-01-05 09:31:22 +08:00
&dofile("apps/CA.pl",'/usr/local/bin/perl','^#!/', '#!%s');
Add support for DJGPP. PR: 75
2002-06-14 04:42:35 +08:00
}
if ($depflags ne "" && !$make_depend) {
print <<EOF;
Since you've disabled at least one algorithm, you need to do the following
before building:
make depend
EOF
}
Move all autogenerated header file parts to crypto/opensslconf.h.
1999-04-22 01:31:05 +08:00
}
Add version info to Win32 DLLs. We might want to edit the strings a bit... Maybe add to 0.9.7 too?
2002-10-05 05:22:47 +08:00
# create the ms/version32.rc file if needed
Netware-specific changes, PR: 780 Submitted by: Verdon Walker <VWalker@novell.com> Reviewed by: Richard Levitte
2003-11-28 21:10:58 +08:00
if ($IsMK1MF) {
Add version info to Win32 DLLs. We might want to edit the strings a bit... Maybe add to 0.9.7 too?
2002-10-05 05:22:47 +08:00
my ($v1, $v2, $v3, $v4);
if ($version_num =~ /(^[0-9a-f]{1})([0-9a-f]{2})([0-9a-f]{2})([0-9a-f]{2})/i) {
$v1=hex $1;
$v2=hex $2;
$v3=hex $3;
$v4=hex $4;
}
open (OUT,">ms/version32.rc") || die "Can't open ms/version32.rc";
print OUT <<EOF;
#include <winver.h>
LANGUAGE 0x09,0x01
1 VERSIONINFO
FILEVERSION $v1,$v2,$v3,$v4
PRODUCTVERSION $v1,$v2,$v3,$v4
FILEFLAGSMASK 0x3fL
#ifdef _DEBUG
FILEFLAGS 0x01L
#else
FILEFLAGS 0x00L
#endif
FILEOS VOS__WINDOWS32
FILETYPE VFT_DLL
FILESUBTYPE 0x0L
BEGIN
BLOCK "StringFileInfo"
BEGIN
BLOCK "040904b0"
BEGIN
// Required:
VALUE "CompanyName", "The OpenSSL Project, http://www.openssl.org/\\0"
VALUE "FileDescription", "OpenSSL Shared Library\\0"
VALUE "FileVersion", "$version\\0"
#if defined(CRYPTO)
VALUE "InternalName", "libeay32\\0"
VALUE "OriginalFilename", "libeay32.dll\\0"
#elif defined(SSL)
VALUE "InternalName", "ssleay32\\0"
VALUE "OriginalFilename", "ssleay32.dll\\0"
#endif
VALUE "ProductName", "The OpenSSL Toolkit\\0"
VALUE "ProductVersion", "$version\\0"
// Optional:
//VALUE "Comments", "\\0"
VALUE "LegalCopyright", "Copyright <20> 1998-2002 The OpenSSL Project. Copyright <20> 1995-1998 Eric A. Young, Tim J. Hudson. All rights reserved.\\0"
//VALUE "LegalTrademarks", "\\0"
//VALUE "PrivateBuild", "\\0"
//VALUE "SpecialBuild", "\\0"
END
END
BLOCK "VarFileInfo"
BEGIN
VALUE "Translation", 0x409, 0x4b0
END
END
EOF
close(OUT);
}
Re-enable message about transition <foo.h> => <openssl/foo.h> because various programs are not updated that often and hence still expect header files names without the openssl/ prefix.
1999-09-10 04:21:10 +08:00
print <<EOF;
Print a reassuring message when Configure is done.
2000-02-01 09:35:29 +08:00
Configured for $target.
Re-enable message about transition <foo.h> => <openssl/foo.h> because various programs are not updated that often and hence still expect header files names without the openssl/ prefix.
1999-09-10 04:21:10 +08:00
EOF
New Configure options "threads" and "no-threads". For Solaris and Linux, "threads" (with proper compiler options) is the default. Submitted by: Reviewed by: PR:
1999-04-30 00:10:09 +08:00
print <<\EOF if (!$no_threads && !$threads);
The library could not be configured for supporting multi-threaded
applications as the compiler options required on this system are not known.
The Mingw32 configuration entry was missing one colon (found by "Zot O'Connor" <zot@ZotConsulting.com>). Also, Windows probably does not need the warning about multi-threading.
1999-06-27 05:25:01 +08:00
See file INSTALL for details if you need multi-threading.
Change #include filenames from <foo.h> to <openssl.h>. Submitted by: Reviewed by: PR:
1999-04-24 06:13:45 +08:00
EOF
If the user said 'shared' and we haven't included support for shared libraries, warn him or her. Reminder by Andrew Marlow <apm35@student.open.ac.uk>
2003-01-11 19:40:39 +08:00
print <<\EOF if ($no_shared_warn);
You gave the option 'shared'. Normally, that would give you shared libraries.
Unfortunately, the OpenSSL configuration doesn't include shared library support
for this platform yet, so it will pretend you gave the option 'no-shared'. If
you can inform the developpers (openssl-dev\@openssl.org) how to support shared
add something to the '$no_shared_warn' text
2003-01-13 22:58:49 +08:00
libraries on this platform, they will at least look at it and try their best
(but please first make sure you have tried with a current version of OpenSSL).
If the user said 'shared' and we haven't included support for shared libraries, warn him or her. Reminder by Andrew Marlow <apm35@student.open.ac.uk>
2003-01-11 19:40:39 +08:00
EOF
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
exit(0);
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
sub usage
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
{
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
print STDERR $usage;
More compact Configure usage message.
2000-02-21 08:55:45 +08:00
print STDERR "\npick os/compiler from:\n";
Add strictness, fix variable substition bugs.
1998-12-29 01:08:48 +08:00
my $j=0;
Minor tweaks to keep Perl 5.001 happy.
1999-02-08 02:22:15 +08:00
my $i;
More compact Configure usage message.
2000-02-21 08:55:45 +08:00
my $k=0;
Minor tweaks to keep Perl 5.001 happy.
1999-02-08 02:22:15 +08:00
foreach $i (sort keys %table)
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
{
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
next if $i =~ /^debug/;
More compact Configure usage message.
2000-02-21 08:55:45 +08:00
$k += length($i) + 1;
if ($k > 78)
{
print STDERR "\n";
$k=length($i);
}
print STDERR $i . " ";
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
}
foreach $i (sort keys %table)
{
next if $i !~ /^debug/;
More compact Configure usage message.
2000-02-21 08:55:45 +08:00
$k += length($i) + 1;
if ($k > 78)
{
print STDERR "\n";
$k=length($i);
}
print STDERR $i . " ";
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
}
More compact Configure usage message.
2000-02-21 08:55:45 +08:00
print STDERR "\n\nNOTE: If in doubt, on Unix-ish systems use './config'.\n";
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
exit(1);
Import of old SSLeay release: SSLeay 0.8.1b
1998-12-21 18:52:47 +08:00
}
New Makefile variables $(RANLIB) and $(PERL).
1999-04-01 20:34:33 +08:00
sub which
{
my($name)=@_;
my $path;
foreach $path (split /:/, $ENV{PATH})
{
Changes for VOS, submitted by Paul Green <Paul.Green@stratus.com>. PR: 499
2004-06-29 06:01:37 +08:00
if (-f "$path/$name$exe_ext" and -x _)
New Makefile variables $(RANLIB) and $(PERL).
1999-04-01 20:34:33 +08:00
{
Changes for VOS, submitted by Paul Green <Paul.Green@stratus.com>. PR: 499
2004-06-29 06:01:37 +08:00
return "$path/$name$exe_ext" unless ($name eq "perl" and
system("$path/$name$exe_ext -e " . '\'exit($]<5.0);\''));
New Makefile variables $(RANLIB) and $(PERL).
1999-04-01 20:34:33 +08:00
}
}
}
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
sub dofile
{
my $f; my $p; my %m; my @a; my $k; my $ff;
($f,$p,%m)=@_;
mk1mf.pl and mkdef.pl read OPTIONS from toplevel Makefile. Configure no longer changes files in place.
1999-05-13 18:28:14 +08:00
open(IN,"<$f.in") || open(IN,"<$f") || die "unable to open $f:$!\n";
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
@a=<IN>;
close(IN);
foreach $k (keys %m)
{
grep(/$k/ && ($_=sprintf($m{$k}."\n",$p)),@a);
}
cleaning up a little
2000-03-13 07:27:14 +08:00
open(OUT,">$f.new") || die "unable to open $f.new:$!\n";
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
print OUT @a;
close(OUT);
cleaning up a little
2000-03-13 07:27:14 +08:00
rename($f,"$f.bak") || die "unable to rename $f\n" if -e $f;
rename("$f.new",$f) || die "unable to rename $f.new\n";
New Configure option --openssldir to replace util/ssldir.pl.
1999-04-25 06:59:36 +08:00
}
Some assembler-related clean-ups.
1999-05-29 07:18:51 +08:00
sub print_table_entry
{
my $target = shift;
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
(my $cc,my $cflags,my $unistd,my $thread_cflag,my $sys_id,my $lflags,
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
my $bn_ops,my $cpuid_obj,my $bn_obj,my $des_obj,my $aes_obj, my $bf_obj,
I had completely forgotten that print_table_entry also needed to read the two "shared" variables on it's own...
2000-08-02 11:43:29 +08:00
my $md5_obj,my $sha1_obj,my $cast_obj,my $rc4_obj,my $rmd160_obj,
Rework the system to generate shared libraries: - Make note of the expected extension for the shared libraries and if there is a need for symbolic links from for example libcrypto.so.0 to libcrypto.so.0.9.7. There is extended info in Configure for that. - Make as few rebuilds of the shared libraries as possible. - Still avoid linking the OpenSSL programs with the shared libraries. - When installing, install the shared libraries separately from the static ones.
2000-10-13 23:25:06 +08:00
my $rc5_obj,my $dso_scheme,my $shared_target,my $shared_cflag,
Add aix64-cc, and make sure that ar gets proper flags for 64-bit libraries
2002-08-02 04:28:16 +08:00
my $shared_ldflag,my $shared_extension,my $ranlib,my $arflags)=
Rework the system to generate shared libraries: - Make note of the expected extension for the shared libraries and if there is a need for symbolic links from for example libcrypto.so.0 to libcrypto.so.0.9.7. There is extended info in Configure for that. - Make as few rebuilds of the shared libraries as possible. - Still avoid linking the OpenSSL programs with the shared libraries. - When installing, install the shared libraries separately from the static ones.
2000-10-13 23:25:06 +08:00
split(/\s*:\s*/,$table{$target} . ":" x 30 , -1);
Some assembler-related clean-ups.
1999-05-29 07:18:51 +08:00
print <<EOF
*** $target
\$cc = $cc
\$cflags = $cflags
Some pre-POSIX systems don't have unistd.h (but e.g. lib.c). Allow configuring the name of that header file.
1999-06-09 21:23:38 +08:00
\$unistd = $unistd
Some assembler-related clean-ups.
1999-05-29 07:18:51 +08:00
\$thread_cflag = $thread_cflag
Make all configuration macros available for application by making sure they are available in opensslconf.h, by giving them names starting with "OPENSSL_" to avoid conflicts with other packages and by making sure e_os2.h will cover all platform-specific cases together with opensslconf.h. I've checked fairly well that nothing breaks with this (apart from external software that will adapt if they have used something like NO_KRB5), but I can't guarantee it completely, so a review of this change would be a good thing.
2001-02-20 00:06:34 +08:00
\$sys_id = $sys_id
Some assembler-related clean-ups.
1999-05-29 07:18:51 +08:00
\$lflags = $lflags
\$bn_ops = $bn_ops
Add framework for yet another assembler module dubbed "cpuid." Idea is to have a placeholder to small routines, which can be written only in assembler. In IA-32 case this includes processor capability identification and access to Time-Stamp Counter. As discussed earlier OPENSSL_ia32cap is introduced to control recently added SSE2 code pathes (see docs/crypto/OPENSSL_ia32cap.pod). For the moment the code is operational on ELF platforms only. I haven't checked it yet, but I have all reasons to believe that Windows build should fail to link too. I'll be looking into it shortly...
2004-07-27 04:18:55 +08:00
\$cpuid_obj = $cpuid_obj
Some assembler-related clean-ups.
1999-05-29 07:18:51 +08:00
\$bn_obj = $bn_obj
\$des_obj = $des_obj
Add anchors for AES, SHA-256/-512 assembler modules and SSE2 code pathes. I also used this opportunity to clean up some out-of-date targets and re-group targets by OS.
2004-07-19 00:19:34 +08:00
\$aes_obj = $aes_obj
Some assembler-related clean-ups.
1999-05-29 07:18:51 +08:00
\$bf_obj = $bf_obj
\$md5_obj = $md5_obj
\$sha1_obj = $sha1_obj
\$cast_obj = $cast_obj
\$rc4_obj = $rc4_obj
\$rmd160_obj = $rmd160_obj
\$rc5_obj = $rc5_obj
This commit ties the new DSO code (crypto/dso/) into the build for a variety of platforms. A few are missing, and they will be added in eventually, but as this is new stuff, it was better to not break lots of platforms in one go that we can't easily test. The changes to "Configure" should illustrate how to add support to other systems if you feel like having a go. NB: I'll add something shortly to allow you to add "dlfcn.h" support on those platforms that don't have (or need) a dlfcn.h header file. (The symbol for Configure will probably by "dlfcn_no_h"). Thanks to Richard Levitte, who is responsible for the dso_dl.c support, understanding the trickier aspects of the build process, and giving great feedback on everything else. [Don't use this stuff if you're easily offended by changes to the interface or behaviour - it's still work in progress.] PR:
2000-04-05 06:32:19 +08:00
\$dso_scheme = $dso_scheme
Redo and enhance the support for building shared libraries. Currently there's support for building under Linux and True64 (using examples from the programming manuals), including versioning that is currently the same as OpenSSL versions but should really be a different series. With this change, it's up to the users to decide if they want shared libraries as well as the static ones. This decision now has to be done at configuration time (well, not really, those who know what they do can still do it the same way as before). The OpenSSL programs (openssl and the test programs) are currently always linked statically, but this may change in the future in a configurable manner. The necessary makefile variables to enable this are in place. Also note that I have done absolutely nothing about the Windows target to get something similar. On the other hand, DLLs are already the default there, but without versioning, and I've no idea what the possibilities for such a thing are there...
2000-07-21 23:08:53 +08:00
\$shared_target= $shared_target
\$shared_cflag = $shared_cflag
fix 'Configure TABLE' output
2002-01-04 21:30:05 +08:00
\$shared_ldflag = $shared_ldflag
Rework the system to generate shared libraries: - Make note of the expected extension for the shared libraries and if there is a need for symbolic links from for example libcrypto.so.0 to libcrypto.so.0.9.7. There is extended info in Configure for that. - Make as few rebuilds of the shared libraries as possible. - Still avoid linking the OpenSSL programs with the shared libraries. - When installing, install the shared libraries separately from the static ones.
2000-10-13 23:25:06 +08:00
\$shared_extension = $shared_extension
Workaround for solaris64 linking problem (explicit "ar rs" is needed to create a symbol table). Sun patches such as 109147-06 probably fix this problem, but we can easily avoid it.
2001-03-09 21:04:06 +08:00
\$ranlib = $ranlib
Add aix64-cc, and make sure that ar gets proper flags for 64-bit libraries
2002-08-02 04:28:16 +08:00
\$arflags = $arflags
Some assembler-related clean-ups.
1999-05-29 07:18:51 +08:00
EOF
}
I got some reports that some targets have weird dso_schemes. Therefore, I've added a sanity checker. Note that it can be combined with almost any other argument (the other arguments will be completely ignored), with "reconf" as the blatant exception, since it also has the behavior of ignoring all following command line arguments. If --test-sanity and reconf are both used on the command line, the first one wins.
2002-01-27 23:52:37 +08:00
sub test_sanity
{
my $errorcnt = 0;
print STDERR "=" x 70, "\n";
print STDERR "=== SANITY TESTING!\n";
print STDERR "=== No configuration will be done, all other arguments will be ignored!\n";
print STDERR "=" x 70, "\n";
foreach $target (sort keys %table)
{
@fields = split(/\s*:\s*/,$table{$target} . ":" x 30 , -1);
Detect one-step shifts of the dso_scheme.
2002-01-27 23:58:34 +08:00
if ($fields[$idx_dso_scheme-1] =~ /^(dl|dlfcn|win32|vms)$/)
{
$errorcnt++;
print STDERR "SANITY ERROR: '$target' has the dso_scheme [$idx_dso_scheme] values\n";
print STDERR " in the previous field\n";
}
elsif ($fields[$idx_dso_scheme+1] =~ /^(dl|dlfcn|win32|vms)$/)
{
$errorcnt++;
print STDERR "SANITY ERROR: '$target' has the dso_scheme [$idx_dso_scheme] values\n";
print STDERR " in the following field\n";
}
elsif ($fields[$idx_dso_scheme] !~ /^(dl|dlfcn|win32|vms|)$/)
I got some reports that some targets have weird dso_schemes. Therefore, I've added a sanity checker. Note that it can be combined with almost any other argument (the other arguments will be completely ignored), with "reconf" as the blatant exception, since it also has the behavior of ignoring all following command line arguments. If --test-sanity and reconf are both used on the command line, the first one wins.
2002-01-27 23:52:37 +08:00
{
$errorcnt++;
print STDERR "SANITY ERROR: '$target' has the dso_scheme [$idx_dso_scheme] field = ",$fields[$idx_dso_scheme],"\n";
print STDERR " valid values are 'dl', 'dlfcn', 'win32' and 'vms'\n";
}
}
Correct the number of colons for the targets aix43-cc, aix43-gcc, alpha-cc, alpha-cc-rpath, alpha-gcc, alpha164-cc and alphaold-cc.
2002-01-28 00:03:40 +08:00
print STDERR "No sanity errors detected!\n" if $errorcnt == 0;
I got some reports that some targets have weird dso_schemes. Therefore, I've added a sanity checker. Note that it can be combined with almost any other argument (the other arguments will be completely ignored), with "reconf" as the blatant exception, since it also has the behavior of ignoring all following command line arguments. If --test-sanity and reconf are both used on the command line, the first one wins.
2002-01-27 23:52:37 +08:00
return $errorcnt;
}