same time, make sure that g++ is never installed alongside a higher version
of libstdc++, so that stuff will always link to the correct library.
feedback espie@ and kurt@, ok kurt@
Bulk builders suffer without this, and we already have a sligt
idea what's going wrong (new thread receives a signal before
__tfork returns and then it fails in sigreturn).
I hope to find some time to work on a regression test for this
problem (which isn't related at all to ghc).
- enable thread model posix; bump estdc++ major
- fix manpages: install cpp.1 and gcov.1 with an 'e' prefix, and install
another copy of gcc.1 as eg++.1 instead of an effectively empty page.
feedback and ok kurt@
The Gambit Scheme system is a complete, portable, efficient and reliable
implementation of the Scheme programming language.
The Gambit-C system conforms to the R4RS, R5RS and IEEE Scheme
standards. The full numeric tower is implemented, including: infinite
precision integers (bignums), rationals, inexact reals (floating point
numbers), and complex numbers.
ok sthen@
Railo is a high-performance open-source CFML engine. It has good
compatibility with other CFML engines. Most sites can be ported to
Railo with very little effort and all of the major CFML frameworks
support Railo out of the box.
Railo will very probably run your sites faster than any other CFML
engine, allowing you to handle more load and to provide a better
experience for end users.
While here add @option no-default-conflict + @conflict
spidermonkey->=1.9.2 in preparation for devel/spidermonkey import.
MAINTAINER timeout, ok/prodding ajacoutot@
Also, sync a bit with base gcc:
- disable unsafe builtins
- disable -Wpointer-sign in -Wall
- move -fstrict-aliasing to -O3, disable -fstrict-overflow and
-fdelete-null-pointer-checks
- enable -Wsystem-headers
- default to march=i486 on i386
- fix crtsavres on powerpc
- sync sparc64 config
- implement -Wvariable-decl, -Wstack-larger-than-N (alpha, amd64, i386, sparc,
sparc64), -Wtrampolines, -ftrampolines
- default to -fno-ident
- fix -rdynamic by passing -export-dynamic to ld(1)
some bits from Brad, others from me.
tested by myself and jsg@, adastrap now hosted by sthen@ (thanks).
ok jsg@ sthen@
Go is an open source programming environment that makes it easy to build
simple, reliable, and efficient software.
With assistance from sthen@
ok sthen@
Unfortunately, we can't completely remove or skip include/execvpe.h
and cbits/execvpe.c, because they also contain pPrPr_disableITimers(),
which is called from the Haskell code.
- Switch back to GhcThreaded=YES (default). It's stupid, it's not
required by GHC (IIRC it was required a long time ago back when
ghci used libreadline), but it may help finding rthread bugs.
Until now, I did run four complete builds of GHC and all Haskell
ports with dpb without any problems.
routines from Perl6-like code. The key feature of NQP is that it's
designed to be a very small compiler (as compared with, say, perl6
or Rakudo) and is focused on being a high-level way to create
compilers and libraries for virtual machines (such as the Parrot
Virtual Machine). Unlike a full-fledged implementation of Perl 6,
NQP strives to have as small a runtime footprint as it can, while
still providing a Perl 6 object model and regular expression engine
for the virtual machine.
ok landry@
- backporting a fix for the ppc backend from upstream
- telling libgo configure not to try to link to libffi_convenience on ppc
issue brought up and fix tested by ajacoutot@, "go ahead" kettenis@
This is a devel version but 0.16.0 is just around the corner and it will
ease dealing with the big batch of g* updates (glib, gtk, gnome-* ...)
that are due in the upcoming weeks.
itself using the non-threaded rts. This doesn't fix the real problem,
but it makes it possible to build hs-* ports without filling up
dpb's build slots with stuck ghc processes.
ok naddy@ espie@
Note: I'll re-run the testsuite with this diff and hope to get some
reasonable results that may help us to identify the real problem.
function name to __stack_smash_handler. Only supported for "c-family"
languages at the moment.
- enable it per default on archs where FRAME_GROWS_DOWNWARD
- remove obsolete comment about adastrap needing a specific libc
- -java actually needs a runtime dep on -main for -lgcc
ok espie@
fix for CVE-2011-4885 included in PHP 5.3.9
Note that the Suhosin extension (which is enabled by default in our packages)
mitigates against this problem
ok robert@ ajacoutot@
This solves some nasty 5.0 -> current update problems.
Up to 5.0, lots of php-using packages were depending on php-mysql-5.2.
So the dependencies would update to php-mysql-5.2 from current.
BUT the php-using packages are now depending on 5.3 -> internal conflict !
With this change, things work again. After the update, you end up with
php-mysql-5.2 AND php-mysql-5.3, you can then try pkg_delete -a to weed
out "unneeded" php*-5.2
okay sthen@, ajacoutot@
- deal with ldconfig shitz being run at fake install everywhere
- fix sprintf/strcpy usage in libestdc++ so we don't get a warning every time
we compile a C++ program
Everything brought up by espie@.
Stack overflow, only affects a very specific and unusual non-default
configuration: transparent cookie encryption enabled, HTTP response
splitting protection disabled, *and* a PHP application must allow
unfiltered user input to header() in order to be affected.
Good write-up at http://seclists.org/fulldisclosure/2012/Jan/295
ok robert@
CONFIG_SITE even if the FSF says no.
Also, finally get rid of the stage3 build/comparison, since apart from
taking build time, it's not really THAT useful unless you're specifically
debugging gcc.
libc.so from OpenBSD 5.0, and tweaks the ada scripts to LD_LIBRARY_PATH
the bundled library.
That way, we should be able to get away by providing a new bootstrap ways
less often (preferably every release, but things won't break if we do it
even LESS often).
- uncomment a couple of lines in the default php-fpm.conf so you
don't have to make any changes before it will start up
- add missing crypt_blowfish.h header to PLIST-main
ok robert@
matching the requirements of haskell-platform-2011.4.0.0.
Moving the xhtml library back to a separate port (www/hs-xhtml)
would be nice but it causes too much headache (like dependency
cycles with devel/haddock).
(from OpenBSD 5.0-release), thanks ajacoutot@ for hosting the distfile.
- check that the bootstrap egcc can actually run; the reason for failure
due to missing libs is now clear (previously it was a cryptic "C compiler
cannot create executables" type message during configure).
Suspending multithreaded programs built with ghc (including ghc
itself) should just work[tm] now. (Except for the bootstrapping
compiler which of course still uses the old code)
- fix conflict with speech-dispatcher
- fix gen-versions target
- i386 ethread compat
update to R14B04 from dlg@, rest by piotr sikora and me
tested with rabbitmq and ejabberd on amd64; sparc64 is still happy too
maintainer timed-out
Also include the compiler standard libraries under compiler-libs/, as this
is the convention adopted by other packaging (Debian, FreeBSD), and is used
by various utility software such as the Lwt interactive toplevel and some
of the graphical IDEs.
CoffeeScript is a little language that compiles into JavaScript.
Underneath all those awkward braces and semicolons, JavaScript has
always had a gorgeous object model at its heart. CoffeeScript is an
attempt to expose the good parts of JavaScript in a simple way.
The golden rule of CoffeeScript is: "It's just JavaScript". The code
compiles one-to-one into the equivalent JS, and there is no
interpretation at runtime. You can use any existing JavaScript library
seamlessly from CoffeeScript (and vice-versa). The compiled output is
readable and pretty-printed, passes through JavaScript Lint without
warnings, will work in every JavaScript implementation, and tends to run
as fast or faster than the equivalent handwritten JavaScript.
Feedback and OK jasper@
without any modifications to PATH. To run an ruby program that is
installed in jruby's bin directory (which no longer needs to be in the
PATH), use jruby -S program.
I don't know why it doesn't find it, but it looks like the tmp inst
is supposed to get it...
so work around until eventually robert fixes this mess for real...
builtins.c:
Allow MD backend to prevent the optimization of a bcopy() or memmove() of
size 1 (the size being known at compile-time) into an inline mempcpy()
expansion, which will in turn expand into a byte load and store operation.
This expansion loses precious address alignment information at some point
(because everybody knows that you can read a byte from any address, right?),
and this loses bigtime on strict alignment platforms which lack the ability
to accesse bytes directly, such as alpha (unless compiling with -mbwx and
runnning on a BWX-capable cpu).
config/alpha:
Require alignment of local arrays on word boundaries, and enable
the builtins.c `one-byte memcpy' workaround.
ok espie@
is not set, but only for gem, extconf, and setup based ports. Other
ports could be using the lang/ruby module and still have a standard
make-based regression test suite.
anywhere else. The only thing not supported yet is tracing using ptrace(2).
- clean up Makefile and restructure a bit
- VMEM_WARNING when -java is built
- some minor PLIST tweaks
ok espie@
rspec 1 using rspec and rspec 2 using rspec2. Additionally, add
support for ruby and testrb, calling the appropriate binary for the
ruby implementation.
Start checking sanity of MODRUBY_REGRESS entry, using a fatal error
if it is defined and doesn't contain a recognized word.
Instead of RAKE_REGRESS_TARGET and RSPEC_REGRESS_TARGET, just use
MODRUBY_REGRESS_TARGET for all cases. On ruby 1.9, modifying the
environment to always look in the current directory for libraries
(the ruby 1.8 behavior), since many ports depend on that for regress.
Allow ports to set their own regress environment and current directory
using MODRUBY_REGRESS_ENV and MODRUBY_REGRESS_DIR.
which should allow building on arches without enough memory to
generate the ri files. Also, set PKG_ARCH=* for the ri_docs
subpackage, so that such arches can still use the documentation,
idea from sthen@. Original idea for PSUEDO_FLAVOR from phessler@.
OK sthen@
The spec task requires network access via git, which is not allowed.
Since git isn't needed anymore for regress, remove it from
REGRESS_DEPENDS. The spec task is just commented out in case somone
wants to run it manually.
Set JAVA_HOME and PATH and disable forking so java doesn't crash when
running the test and spec tasks. Don't install the jruby-launcher gem
when running the test and spec tasks, as it fails.
Regress depend on git when running the specs, as it needs to download
current rubyspecs for the spec task.
Delete the jruby-complete.jar file and just use the jruby.jar file
when running the specs, as otherwise you get additional test failures
and warnings.
Switch to SUBST_CMD while here now that has been fixed. Also, bump
the memory limits by 50M as 384M doesn't seem to be enough in all
cases.
Verilator is the fastest free Verilog HDL simulator, and beats most
commercial simulators. It compiles synthesizable Verilog (not test-bench
code!), plus some PSL, SystemVerilog and Synthesis assertions into C++
or SystemC code. It is designed for large projects where fast simulation
performance is of primary concern, and is especially well suited to
generate executable models of CPUs for embedded software design teams.
ok aja@
upstream decided to move to separate modules instead
of one monolithic jar file. packaging all these standalone modules is not going
to work, so please use devel/leiningen instead to fetch individual modules.
that no longer create .require_paths files. No RUN_DEPENDS
are being added as older ruby versions can use gems without
.require_paths files (we never had a ruby 1.9.1 package).
This is being done to avoid errors by porters who are creating/updating
gem ports without installing new ruby versions.
Make jruby binary not segfault if PATH isn't set correctly.
Remove rubygems hacks, since they are no longer necessary now
that the versions of rubygems for the other ruby interpreters
are > 1.3.7.
No longer ship the ri documentation files with JRuby. JRuby
ships with a broken ri binary anyway (requires the rdoc gem to
work), and almost all users of JRuby have MRI ruby installed
and can use its ri binary.
Make jruby binary not segfault if PATH isn't set correctly.
Remove rubygems hacks, since they are no longer necessary now
that the versions of rubygems for the other ruby interpreters
are > 1.3.7.
No longer ship the ri documentation files with JRuby. JRuby
ships with a broken ri binary anyway (requires the rdoc gem to
work), and almost all users of JRuby have MRI ruby installed
and can use its ri binary.
a few new features. A few of our patches were accepted upstream.
Add an -ri_docs multipackage to hold the ri documentation files. These
aren't used very often, slow down installation significantly, and bloat
the package.
This also fixes the conflict marker for the -tk multipackage.
bootstrapper).
Bump, because some of the `library hashes' changed.
Unfortunately, this also affects a couple of other haskell ports,
which get different `library hashes' now and will be reported as
broken by ghc-pkg check.
People using snapshot packages should get seamless updates when the
next set of snapshots is available. People building from ports and
*not* using dpb should use the out-of-date-script to get a subdirlist
of haskell ports which should be rebuilt.
I could also bump the affected ports, but I'm not sure wether it's
worth the trouble.
of one monolithic jar file. packaging all these standalone modules is not going
to work, so please use devel/leiningen instead to fetch individual modules.
clojure-contrib will be removed soon
also annotate gamin for bootstrap property, and mark haddock,no_deps as
a "real flavor", since it's only used for bootstrap anyways AND significantly
changes the build.
rework pkg-readme to conform to new template, update man.conf patch,
move man dir to Tcl's lib dir, properly preprocess man pages.
Bump major due to function signature change in tcllib.
ok sebastiap@, sthen@
Boo is a new object oriented statically typed programming language for
the Common Language Infrastructure with a Python inspired sysntax and a
special focus on language and compiler extensibility.
from ryan boggs (MAINTAINER) with some tweaks by me
ok aja@
