aboutsummaryrefslogtreecommitdiffstats
path: root/configure.pl
Commit message (Collapse)AuthorAgeFilesLines
* Add basic support for Doxygen, including a make targetlloyd2008-10-131-0/+6
|
* Add a new switch to configure.pl --with-tr1-implementation whichlloyd2008-10-121-51/+84
| | | | | | | | | | | | | | | | | | | | | | | | accepts options 'boost' and 'system'. Now GF(p) math (and indirectly, ECDSA) will be enabled if --with-tr1=boost or --with-tr1=system is passed at build time to enable a shared_ptr implementation. Modules can now specify that they require TR1, in which case they will only be autoloaded if a TR1 implementation was set (they can choose which one to use by checking the feature macros from build.h) The gfpmath module was set to load only on request. That has changed to load automatically (but will only do so if a TR1 impl is set as described above). CVC has also been marked as requiring TR1. (ECDSA/ECDH are not, since they do not use tr1 shared_ptr directly) Update and cleanup help output. Do not print the list of modules in --help anymore (too long); you can still get the list (in an easier to parse format) --module-info. Reorganize the help text so the more useful options are described closer to the top. Fix the --with-endian and --with-unaligned-mem options: they were being accepted but ignored.
* Reenable /proc/cpuinfo CPU detectionlloyd2008-10-111-1/+1
|
* Comment out or remove some of the louder/less useful trace messageslloyd2008-10-111-6/+3
|
* Only run uname if Perl's -f operator returns true forlloyd2008-10-111-34/+52
| | | | | | | | | | | | | | | | '/bin/uname', '/usr/bin/uname', or '/bin/sh' - the existence of these files gives us some hope that exec'ing uname will work, and avoids an error on Windows builds. Only run uname once (uname -a), collecting it into $$config{'uname'} (previously configure.pl might run it as many as three times, uname -s, uname -p, and uname -m). Only submodels were being searched longest to shortest. Modify guess_cpu_from_this to also search arch names and aliases in this manner. This allows Config{'archname'} of 'x86_64-linux' to be correctly detected as x86-64 (was being detected as x86, since before the search was ordered by the Perl hash order).
* Set solib version, readme, log for 1.7.17 sometime in the mysterious future.lloyd2008-10-091-1/+1
|
* Fix infinite loop if no close tag found in info.txt filelloyd2008-10-081-0/+2
|
* Bump soname to 1.7.16lloyd2008-10-081-1/+1
|
* New /proc/cpuinfo matching logic suggested by Derek Scherger. Attemptlloyd2008-10-071-10/+14
| | | | | longest match first instead of random Perl hash order. Seems to work very well in practice.
* Make configuration quieterlloyd2008-10-071-2/+2
|
* Don't generate the makefile until after we have processed build.h.inlloyd2008-10-011-4/+8
| | | | | | | | | | Markus Wanner noted on the mailing list that build.h was not being installed. That is because it was not included in the HEADERS variable in the Makefile, and build.h did not yet exist when configure.pl scanned the source and include directories. Rearrange the order of operations in configure.pl a bit so first build.h is generated and then the makefile is.
* Search submodels before submodel alises when trying to figure out /proc/cpuinfolloyd2008-10-011-6/+6
|
* Make autodetection of CPU type based on /proc/cpuinfo a bit smarter.lloyd2008-10-011-27/+47
|
* Add back --modules as an alias for --enable-modules (finger memory)lloyd2008-09-301-0/+1
|
* s/--enable-module-sets/--use-module-set/ - otherwise one has to type all of ↵lloyd2008-09-301-1/+1
| | | | --enable-modules for it to be recognized
* Update soname to 1.7.15, also update readme and loglloyd2008-09-301-1/+1
|
* Improve configure.pl's --help output. s/--module-set/--enable-module-set/lloyd2008-09-301-26/+22
|
* Queitlloyd2008-09-301-1/+1
|
* Silently swallow/save --exec-prefix and most of the autoconf --.*dir= optionslloyd2008-09-301-42/+61
| | | | | (--bindir, --libdir, --docdir already supported, add --includdir=, --localstatedir=, etc)
* Rename more options for autoconf compatability.lloyd2008-09-301-34/+38
|
* First step in configure.pl options rename for better autoconf/automake compat.lloyd2008-09-301-9/+9
| | | | | | | Suggestions from Zack Weinberg on the monotone-devel list. Rename --endian to --with-endian Rename --unaligned-mem to --with-unaligned-mem
* Remove dead code from guess_from_from_thislloyd2008-09-301-32/+9
|
* Guess CPU from /proc/cpuinfo or uname by looking first in the knownlloyd2008-09-301-20/+34
| | | | names for CPUs we pulled from the text files.
* Do not redirect 2> to /dev/null when exec'ing uname. It causeslloyd2008-09-301-3/+3
| | | | problems on MSYS. And doesn't have any real advantage that I can see...
* Target i686 if uname is i686lloyd2008-09-301-1/+1
|
* propagate from branch 'net.randombit.botan' (head ↵lloyd2008-09-301-0/+1
|\ | | | | | | | | | | fd327b29aa542e0ad5ff6d37d8392321670f0369) to branch 'net.randombit.botan.modularized' (head 3f8d05493d4b192243fdc8a7f518ed1013c3be54)
| * Better support for MinGW/MSYS, sent by Zbigniew Zagorskilloyd2008-09-301-0/+1
| |
* | Remove needless optionslloyd2008-09-301-8/+2
| |
* | Remove noisy print statementlloyd2008-09-301-1/+0
| |
* | If a module specifies <module_name>:<file> in its <add> block, includelloyd2008-09-301-4/+18
| | | | | | | | | | | | that file from another module (without loading it). This was needed since mp_asm64 and mp_ia32_msvc do not implement both mp_asm.h and mp_asmi.h like the other mp_ modules - instead they pull the mp_generic version in.
* | Also display non-enabled modules in configure outputlloyd2008-09-301-8/+20
| |
* | Rewrite part of the module loading code in configure.pl, now dependencieslloyd2008-09-301-137/+119
| | | | | | | | | | | | seem to be handled correctly (er, at least mostly), and more importantly the asm MPI modules are detected and used correctly (at least on x86-64 and x86).
* | Start trying to get asm BigInt working. Currently does not. Just checkinglloyd2008-09-291-33/+68
| | | | | | | | | | | | this in and sync'ing before I take a break. However the info.txts are now cleaned up and more or less accurate now. But configure.pl is not tracking dependencies properly.
* | Cleanuplloyd2008-09-291-2/+1
| |
* | Merge the 3 mlocks (ml_unix, ml_win32, stub mlock.cpp) into a single mlock.cpplloyd2008-09-291-19/+29
| | | | | | | | | | | | in utils. Support OS feature macros, eg BOTAN_TARGET_OS_HAS_POSIX_MLOCK (how very autoconf)
* | Fix build.h modules listlloyd2008-09-291-1/+1
| |
* | Remove the misc dir:lloyd2008-09-291-1/+3
| | | | | | | | | | | | | | | | | | | | Moved XS, Boost Python, and SWIG wrappers to new toplevel directory 'wrappers' Moved NIST X.509 test suite into checks directory Move the build information used by configure.pl to src/build-data Move scripts directory to doc (for lack of a better spot)
* | In build_cmds, push onto a list and join the strings at the endlloyd2008-09-291-6/+5
| |
* | Better dependency handlinglloyd2008-09-291-22/+34
| |
* | Fix display of loaded moduleslloyd2008-09-291-5/+9
| |
* | Quieterlloyd2008-09-291-1/+1
| |
* | Don't load deps of a module that we won't loadlloyd2008-09-291-6/+6
| |
* | Version bump to 1.7.14lloyd2008-09-291-1/+1
| |
* | Make asm implementations distinctly named objects, for instance MD5_IA32,lloyd2008-09-291-23/+37
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | rather than silently replacing the C++ versions. Instead they are silently replaced (currently, at least) at the lookup level: we switch off the set of feature macros set to choose the best implementation in the current build configuration. So you can have (and benchmark) MD5 and MD5_IA32 directly against each other in the same program with no hassles, but if you ask for "MD5", you'll get maybe an MD5 or maybe MD5_IA32. Also make the canonical asm names (which aren't guarded by C++ namespaces) of the form botan_<algo>_<arch>_<func> as in botan_sha160_ia32_compress, to avoid namespace collisions. This change has another bonus that it should in many cases be possible to derive the asm specializations directly from the original implementation, saving some code (and of course logically SHA_160_IA32 is a SHA_160, just one with a faster implementation of the compression function, so this seems reasonable anyway).
* | propagate from branch 'net.randombit.botan' (head ↵lloyd2008-09-291-0/+2
|\| | | | | | | | | | | ca7d7fc1ae6b55c5328c9cf1ec1cafd1daadedd4) to branch 'net.randombit.botan.modularized' (head 614263a9742a0c554e4093620147f6e156264d41)
| * Oops, continue to look in /proc/cpuinfo rather than my home dirlloyd2008-09-281-1/+1
| |
| * Add support for Pentium-M autodetection and GCC builds. Bug report bylloyd2008-09-281-1/+3
| | | | | | | | Derek Scherger
* | Rename all modinfo.txt files to info.txt, since they are all (none) oflloyd2008-09-291-1/+1
| | | | | | | | | | them modules now. In any case there is no distinction so info.txt seems better.
* | Move what is left of the uncategorized library to 'core'. There is stilllloyd2008-09-291-14/+7
| | | | | | | | | | a lot of public key stuff in here that needs to be extracted however, and probably 2-3 other modules worth of stuff to split off (engines, etc)
* | More tracing. Print enabled modules in a block at the endlloyd2008-09-281-2/+8
| |
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-22 07:17:55 +0000