CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: tobi...@cvs.openbsd.org 2013/06/25 00:07:46 Modified files: x11/i3 : Makefile Added files: x11/i3/patches : patch-libi3_ipc_recv_message_c Log message: Fix alignment problem. Makes i3 work on sparc64. OK dcoppa@
Re: bulk results with gcc 4.8.0
I'm adding this diff to the adastrap update I'm working on and will send a complete patch soonish. On Fri, Jun 21, 2013 at 11:05:08AM +0200, Pascal Stumpf wrote: On Sat, 08 Jun 2013 22:52:29 +0200, Pascal Stumpf wrote: On Sat, 8 Jun 2013 22:20:44 +0200, Juan Francisco Cantero Hurtado wrote: On Sat, Jun 08, 2013 at 08:44:15PM +0200, Pascal Stumpf wrote: On Sat, 8 Jun 2013 14:51:45 +0100, Stuart Henderson wrote: (req'd by pascal; cc'ing ports@ in case of interest interest). Here are results from a bulk with MODGCC4_VERSION?=4.8 (this is with 4.8.0, I didn't have 4.8.1 when it started). math/galculator: picked up libquadmath at build time, failed when dpb junked it. I didn't spot any other instances of this in the tree. www/mozilla-firefox: www/seamonkey: c++: error: unrecognized command line option '-rdynamic' Does this happen during link time? egcc -dumpspecs suggests that rdynamic is correctly translated to -export-dynamic. I had the same problem with racket and gcc 4.7 tree months ago. https://github.com/plt/racket/commit/292c81a826f9ebba435f464fce375bd51a118355 D'oh, now I see the problem; diff for 4.8: ping; updated diff: Index: Makefile === RCS file: /home/pascal/cvs/ports/lang/gcc/4.8/Makefile,v retrieving revision 1.8 diff -u -p -r1.8 Makefile --- Makefile 18 Jun 2013 20:24:29 - 1.8 +++ Makefile 21 Jun 2013 09:04:25 - @@ -5,7 +5,12 @@ ONLY_FOR_ARCHS = alpha amd64 i386 powerp V = 4.8.1 FULL_VERSION = $V FULL_PKGVERSION = $V -REVISION-main = 0 +REVISION-main = 1 +REVISION-c++ = 0 +REVISION-java = 0 +REVISION-f95 = 0 +REVISION-objc = 0 +REVISION-ada = 0 ADASTRAP_LIBC-i386 = 68.2 ADASTRAP_LIBM-i386 = 8.0 Index: patches/patch-gcc_config_openbsd_opt === RCS file: patches/patch-gcc_config_openbsd_opt diff -N patches/patch-gcc_config_openbsd_opt --- /dev/null 1 Jan 1970 00:00:00 - +++ patches/patch-gcc_config_openbsd_opt 8 Jun 2013 20:50:00 - @@ -0,0 +1,11 @@ +$OpenBSD$ +--- gcc/config/openbsd.opt.orig Sat Jun 8 22:49:21 2013 gcc/config/openbsd.opt Sat Jun 8 22:49:33 2013 +@@ -32,4 +32,7 @@ Driver + pthread + Driver + ++rdynamic ++Driver ++ + ; This comment is to ensure we retain the blank line above.
Re: gcc/4.7
On Tue, Jun 18, 2013 at 10:39:14PM +0200, Marc Espie wrote: there's talk about killing that port, as it serves no useful purpose, doesn't provide more arches than 4.8, isn't more functional than 4.8, and uses resources (brain-power to keep it alive and cpu time to build somewhat useless packages). It's not even useful for ada-boostrap, as the path can go 4.6-4.8 directly... Anyone sees a reason not to let it go ? As the original complainer, I would be happy to see it go into the attic. Pascal?
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: tobi...@cvs.openbsd.org 2013/06/17 02:44:29 Modified files: geo/openbsd-developers: Makefile geo/openbsd-developers/files: OpenBSD Log message: Putin and Obama agree this is where I live...
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: tobi...@cvs.openbsd.org 2013/05/03 17:13:49 Modified files: lang/gcc/4.6 : Makefile lang/gcc/4.6/patches: patch-gcc_ada_adaint_c lang/gcc/4.7 : Makefile lang/gcc/4.7/patches: patch-Makefile_in patch-gcc_ada_adaint_c Log message: Sync 4.8 changes to 4.7 and 4.6 - Absolute path for STRIP - Enable ncpu detection in gnatlib - Force sequential make install target
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: tobi...@cvs.openbsd.org 2013/05/03 17:23:26 Modified files: lang/gcc/4.6/patches: patch-gcc_ada_Make-generated_in lang/gcc/4.7/patches: patch-gcc_ada_Make-generated_in lang/gcc/4.8/patches: patch-gcc_ada_Make-generated_in Log message: Take another stab at debugging the sporadic build failures in the gnat build tools. Split compiling and linking/binding into two parts and add some ls in between to spot the disappearing object file... Reverts previous attempt running gnatmake with -j1 which turned out to be insufficient. ok pascal@ sthen@
Re: CVS: cvs.openbsd.org: ports
On Fri, May 03, 2013 at 05:13:49PM -0600, Tobias Ulmer wrote: CVSROOT: /cvs Module name: ports Changes by: tobi...@cvs.openbsd.org 2013/05/03 17:13:49 Modified files: lang/gcc/4.6 : Makefile lang/gcc/4.6/patches: patch-gcc_ada_adaint_c lang/gcc/4.7 : Makefile lang/gcc/4.7/patches: patch-Makefile_in patch-gcc_ada_adaint_c Log message: Sync 4.8 changes to 4.7 and 4.6 - Absolute path for STRIP - Enable ncpu detection in gnatlib - Force sequential make install target Forgot to mention, this was OK pascal@ sthen@
Re: lang/gcc: buildtools debugging; sync with 4.8
On Mon, Apr 15, 2013 at 01:34:21AM +0200, Tobias Ulmer wrote: So we have a problem that gnat buildtools sometimes fail to build (except on my machines, grr). Parallel build issues have pretty much been ruled out. The question is now whether egcc fails silently while producing object code or gnatlink/gnatbind are doing something funny. Since egcc/gnat1 gets called a lot during the build, I am leaning towards the later. The diff below first compiles, then binds/links the buildtools with some ls(1) thrown in between so we can see whether the object exists at one point or not. Second part of this diff syncs minor changes from 4.8 to 4.7 and 4.6: * Absolute path to strip, reduces the gnat package size by 50% * Cpu core detection * -j1 for the install target in 4.7, same as the rest I plan to commit these separate, single diff is only for testing. Comments? OK? Ping. I would like to get this in before the T64 change, reduces diskspace requirements for adastrap quite a bit.
Re: UPDATE: www/netsurf
On Tue, Apr 30, 2013 at 10:33:02AM +0100, Stuart Henderson wrote: On 2013/04/29 20:56, Anthony J. Bentley wrote: Here is an updated patch that handles default LANG and USE_CCACHE=No. better, thanks - OK as far as I can test - but as it's probably the best of the browsers which is likely to work on unusual arch, it would be really nice to have a test report on loongson/sparc64 or running it locally on macppc if anyone has one handy .. (remote doesn't work on macppc, but mine is headless so I can't test it locally..) Works on sparc64, but has the same issue as before (most image formats are shifted to red). See old screenshot here: http://www.tmux.org/~tobiasu/tmp/2013-04-17-210925_1280x1024_scrot.png Obviously not endian clean. I'm using it anyway, it's way better than firefox 3.6 for my purpose.
lang/gcc: buildtools debugging; sync with 4.8
So we have a problem that gnat buildtools sometimes fail to build
(except on my machines, grr). Parallel build issues have pretty much
been ruled out.
The question is now whether egcc fails silently while producing
object code or gnatlink/gnatbind are doing something funny.
Since egcc/gnat1 gets called a lot during the build, I am leaning
towards the later.
The diff below first compiles, then binds/links the buildtools with
some ls(1) thrown in between so we can see whether the object exists at
one point or not.
Second part of this diff syncs minor changes from 4.8 to 4.7 and 4.6:
* Absolute path to strip, reduces the gnat package size by 50%
* Cpu core detection
* -j1 for the install target in 4.7, same as the rest
I plan to commit these separate, single diff is only for testing.
Comments? OK?
Index: 4.6/Makefile
===
RCS file: /home/vcs/cvs/openbsd/ports/lang/gcc/4.6/Makefile,v
retrieving revision 1.51
diff -u -p -r1.51 Makefile
--- 4.6/Makefile21 Mar 2013 08:46:32 - 1.51
+++ 4.6/Makefile14 Apr 2013 22:32:06 -
@@ -30,7 +30,7 @@ REVISION-estdc = 10
REVISION-f95 = 14
REVISION-java = 15
REVISION-objc = 14
-REVISION-ada = 15
+REVISION-ada = 16
REVISION-go = 14
SHARED_LIBS = estdc++ 14.0 \
@@ -116,6 +116,7 @@ EXTRA_ENV += am_cv_func_iconv=no JAR=no
EXTRA_ENV += ac_cv_prog_M4=/usr/bin/m4
EXTRA_ENV += ac_cv_header_magic_h=no
EXTRA_ENV += lt_cv_path_SED=/usr/bin/sed
+EXTRA_ENV += STRIP=/usr/bin/strip
MAKE_FLAGS += ac_cv_path_mkdir=/bin/mkdir
CONFIGURE_ENV += ${EXTRA_ENV}
Index: 4.6/patches/patch-gcc_ada_Make-generated_in
===
RCS file:
/home/vcs/cvs/openbsd/ports/lang/gcc/4.6/patches/patch-gcc_ada_Make-generated_in,v
retrieving revision 1.2
diff -u -p -r1.2 patch-gcc_ada_Make-generated_in
--- 4.6/patches/patch-gcc_ada_Make-generated_in 11 Apr 2013 08:34:13 -
1.2
+++ 4.6/patches/patch-gcc_ada_Make-generated_in 14 Apr 2013 22:32:06 -
@@ -1,26 +1,26 @@
$OpenBSD: patch-gcc_ada_Make-generated_in,v 1.2 2013/04/11 08:34:13 pascal Exp
$
--- gcc/ada/Make-generated.in.orig Sun Nov 13 15:25:58 2011
-+++ gcc/ada/Make-generated.in Thu Apr 11 09:55:36 2013
gcc/ada/Make-generated.in Sun Apr 14 04:12:00 2013
@@ -27,26 +27,26 @@ $(ADA_GEN_SUBDIR)/treeprs.ads : $(ADA_GEN_SUBDIR)/tree
-$(MKDIR) $(ADA_GEN_SUBDIR)/bldtools/treeprs
$(RM) $(addprefix $(ADA_GEN_SUBDIR)/bldtools/treeprs/,$(notdir $^))
$(CP) $^ $(ADA_GEN_SUBDIR)/bldtools/treeprs
- (cd $(ADA_GEN_SUBDIR)/bldtools/treeprs; gnatmake -q xtreeprs ;
./xtreeprs ../../treeprs.ads )
-+ (cd $(ADA_GEN_SUBDIR)/bldtools/treeprs; gnatmake -j1 xtreeprs ;
./xtreeprs ../../treeprs.ads )
++ (cd $(ADA_GEN_SUBDIR)/bldtools/treeprs; ls -la; gnatmake -c xtreeprs;
ls -la; gnatmake xtreeprs; ls -la; ./xtreeprs ../../treeprs.ads )
$(ADA_GEN_SUBDIR)/einfo.h : $(ADA_GEN_SUBDIR)/einfo.ads
$(ADA_GEN_SUBDIR)/einfo.adb $(ADA_GEN_SUBDIR)/xeinfo.adb
$(ADA_GEN_SUBDIR)/ceinfo.adb
-$(MKDIR) $(ADA_GEN_SUBDIR)/bldtools/einfo
$(RM) $(addprefix $(ADA_GEN_SUBDIR)/bldtools/einfo/,$(notdir $^))
$(CP) $^ $(ADA_GEN_SUBDIR)/bldtools/einfo
- (cd $(ADA_GEN_SUBDIR)/bldtools/einfo; gnatmake -q xeinfo ; ./xeinfo
../../einfo.h )
-+ (cd $(ADA_GEN_SUBDIR)/bldtools/einfo; gnatmake -j1 xeinfo ; ./xeinfo
../../einfo.h )
++ (cd $(ADA_GEN_SUBDIR)/bldtools/einfo; ls -la; gnatmake -c xeinfo; ls
-la; gnatmake xeinfo; ls -la; ./xeinfo ../../einfo.h )
$(ADA_GEN_SUBDIR)/sinfo.h : $(ADA_GEN_SUBDIR)/sinfo.ads
$(ADA_GEN_SUBDIR)/sinfo.adb $(ADA_GEN_SUBDIR)/xsinfo.adb
$(ADA_GEN_SUBDIR)/csinfo.adb
-$(MKDIR) $(ADA_GEN_SUBDIR)/bldtools/sinfo
$(RM) $(addprefix $(ADA_GEN_SUBDIR)/bldtools/sinfo/,$(notdir $^))
$(CP) $^ $(ADA_GEN_SUBDIR)/bldtools/sinfo
- (cd $(ADA_GEN_SUBDIR)/bldtools/sinfo; gnatmake -q xsinfo ; ./xsinfo
../../sinfo.h )
-+ (cd $(ADA_GEN_SUBDIR)/bldtools/sinfo; gnatmake -j1 xsinfo ; ./xsinfo
../../sinfo.h )
++ (cd $(ADA_GEN_SUBDIR)/bldtools/sinfo; ls -la; gnatmake -c xsinfo; ls
-la; gnatmake xsinfo; ls -la; ./xsinfo ../../sinfo.h )
$(ADA_GEN_SUBDIR)/snames.h $(ADA_GEN_SUBDIR)/snames.ads
$(ADA_GEN_SUBDIR)/snames.adb : $(ADA_GEN_SUBDIR)/stamp-snames ; @true
$(ADA_GEN_SUBDIR)/stamp-snames : $(ADA_GEN_SUBDIR)/snames.ads-tmpl
$(ADA_GEN_SUBDIR)/snames.adb-tmpl $(ADA_GEN_SUBDIR)/snames.h-tmpl
$(ADA_GEN_SUBDIR)/xsnamest.adb $(ADA_GEN_SUBDIR)/xutil.ads
$(ADA_GEN_SUBDIR)/xutil.adb
@@ -28,7 +28,7 @@ $OpenBSD: patch-gcc_ada_Make-generated_i
$(RM) $(addprefix $(ADA_GEN_SUBDIR)/bldtools/snamest/,$(notdir $^))
$(CP) $^ $(ADA_GEN_SUBDIR)/bldtools/snamest
- (cd $(ADA_GEN_SUBDIR)/bldtools/snamest; gnatmake -q xsnamest ;
./xsnamest )
-+ (cd $(ADA_GEN_SUBDIR)/bldtools/snamest; gnatmake -j1 xsnamest ;
./xsnamest )
++ (cd
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: tobi...@cvs.openbsd.org 2013/04/12 04:01:47 Modified files: lang/gcc/4.8 : Makefile distinfo lang/gcc/4.8/patches: patch-gcc_ada_adaint_c patch-gcc_ada_gcc-interface_Makefile_in lang/gcc/4.8/pkg: PLIST-ada Added files: lang/gcc/4.8/patches: patch-gcc_ada_system-openbsd-sparcv9_ads lang/gcc/4.8/pkg: PFRAG.I386-ada Log message: Add Ada support for sparc64. Use SJLJ exceptions instead of ZCX because gcc produces broken code for the later. Enable ncpu detection in the runtime for all archs. ok pascal@
Re: CVS: cvs.openbsd.org: ports
On Thu, Apr 11, 2013 at 02:34:14AM -0600, Pascal Stumpf wrote: CVSROOT: /cvs Module name: ports Changes by: pas...@cvs.openbsd.org 2013/04/11 02:34:13 Modified files: lang/gcc/4.8/patches: patch-gcc_ada_Make-generated_in lang/gcc/4.6/patches: patch-gcc_ada_Make-generated_in lang/gcc/4.7/patches: patch-gcc_ada_Make-generated_in Log message: Attempt to fix longstanding ada build failure when building in parallel by forcing gnatmake to use -j1. I'm pretty sure this is a noop. Our gnatmake can't yet detect multiple CPUs because we lack support for it in adaint.c (Patch coming soon). But sure, lets try for a while... ok sthen@
lang/gcc/4.8 sparc64 gnat support
Add sparc64 to the list of supported Ada archs
* Add ncpu detection, this also affects the other archs thus REVISION bump.
* Use SJLJ (setjump/longjump) exceptions because ZCX generates broken
code on sparc64 (exceptions are thrown, but not caught).
* Disable PCH support, which caused hangs on sparc64, from Pascal
* a-numaux.adb, for those who wonder, contains special sauce for
i386/amd64 - other architectures get their math functions via -lm,
thus a-numaux.adb is not needed. Instead of introducing yet another
PLIST I've chosen to fake it and supply an empty body.
Comments? OK?
Index: Makefile
===
RCS file: /home/vcs/cvs/openbsd/ports/lang/gcc/4.8/Makefile,v
retrieving revision 1.1.1.1
diff -u -p -u -p -r1.1.1.1 Makefile
--- Makefile8 Apr 2013 08:20:19 - 1.1.1.1
+++ Makefile11 Apr 2013 20:46:58 -
@@ -12,6 +12,9 @@ ADASTRAP-i386 = adastrap-i386-$V-1.tar.x
ADASTRAP_LIBC-amd64 = 67.0
ADASTRAP_LIBM-amd64 = 8.0
ADASTRAP-amd64 = adastrap-amd64-$V-1.tar.xz
+ADASTRAP_LIBC-sparc64 = 67.0
+ADASTRAP_LIBM-sparc64 = 8.0
+ADASTRAP-sparc64 = adastrap-sparc64-$V-1.tar.xz
PKGNAME-main = gcc-${FULL_PKGVERSION}
PKGNAME-c++ = g++-${FULL_PKGVERSION}
@@ -24,6 +27,8 @@ PKGNAME-ada = gnat-${FULL_PKGVERSION}
#PKGNAME-go = gccgo-${FULL_PKGVERSION}
PKGSPEC-main = gcc-=4.8,4.9
+REVISION-ada = 0
+
SHARED_LIBS = estdc++ 16.0 \
gfortran5.0 \
gcj 3.0 \
@@ -41,7 +46,7 @@ SHARED_LIBS = estdc++ 16.0 \
PSEUDO_FLAVORS = no_c++ no_f95 no_objc no_java no_ada no_go full
FLAVOR ?=
-ONLY_FOR_ARCHS-ada = amd64 i386
+ONLY_FOR_ARCHS-ada = amd64 i386 sparc64
ONLY_FOR_ARCHS-java = amd64 i386 powerpc sparc64
MULTI_PACKAGES = -main -f95 -objc -c++ -estdc -java -ada
@@ -59,7 +64,8 @@ DISTNAME =gcc-${FULL_VERSION}
DISTFILES =${DISTNAME}.tar.bz2
# there's no ecj-4.8.jar
ECJ = ecj-4.5.jar
-SUPDISTFILES = ${ADASTRAP-amd64}:0 ${ADASTRAP-i386}:0 ${ECJ}:1
+SUPDISTFILES = ${ADASTRAP-amd64}:0 ${ADASTRAP-i386}:0 ${ECJ}:1 \
+ ${ADASTRAP-sparc64}:0
EXTRACT_ONLY = ${DISTNAME}.tar.bz2
BUILD_DEPENDS += devel/bison \
@@ -136,7 +142,8 @@ CONFIGURE_ARGS += \
--enable-threads=posix \
--enable-wchar_t \
--with-gmp=${LOCALBASE} \
- --enable-languages=${LANGS}
+ --enable-languages=${LANGS} \
+ --disable-libstdcxx-pch
# This is needed, as internal cpp is no longer compatible with the
# visible beast
@@ -241,9 +248,12 @@ post-extract:
.if ${BUILD_PACKAGES:M-java}
cp ${FULLDISTDIR}/${ECJ} ${WRKSRC}/ecj.jar
.endif
+ cp ${FILESDIR}/system-openbsd-sparcv9.ads ${WRKSRC}/gcc/ada/
post-install:
.if ${BUILD_PACKAGES:M-ada}
+ # add a-numaux.adb to all archs to simplify packaging
+ touch ${PREFIX}/lib/gcc/${CONFIG}/$V/adainclude/a-numaux.adb
chown -R ${SHAREOWN}:${SHAREGRP} ${PREFIX}/lib/gcc/*/$V/adainclude
chown -R ${SHAREOWN}:${SHAREGRP} ${PREFIX}/lib/gcc/*/$V/adalib
.endif
Index: distinfo
===
RCS file: /home/vcs/cvs/openbsd/ports/lang/gcc/4.8/distinfo,v
retrieving revision 1.1.1.1
diff -u -p -u -p -r1.1.1.1 distinfo
--- distinfo8 Apr 2013 08:20:19 - 1.1.1.1
+++ distinfo11 Apr 2013 20:46:58 -
@@ -1,8 +1,10 @@
SHA256 (gcc/adastrap-amd64-4.8.0-1.tar.xz) =
8zqPYBYcNX913KeXrRlg37VHLuPGnssCCob4Fc51hNg=
SHA256 (gcc/adastrap-i386-4.8.0-1.tar.xz) =
HoPDSgf0fCHyyKknoxThnFgxWZ0T3mdSiXBomEEH5m4=
+SHA256 (gcc/adastrap-sparc64-4.8.0-1.tar.xz) =
g9RJkaXJXtumPRctcUSH83xKXrd8BDLyB2MV6r3EgHs=
SHA256 (gcc/ecj-4.5.jar) = mP0Sjx03TZ5C/Z1INr3SScbVEevGwN8X+8G535bD14E=
SHA256 (gcc/gcc-4.8.0.tar.bz2) = sDf+UTK3HsrS6nFB7JIpK10yQnv5D9kM3kMrHVq6zCw=
SIZE (gcc/adastrap-amd64-4.8.0-1.tar.xz) = 31996808
SIZE (gcc/adastrap-i386-4.8.0-1.tar.xz) = 30038792
+SIZE (gcc/adastrap-sparc64-4.8.0-1.tar.xz) = 27002840
SIZE (gcc/ecj-4.5.jar) = 1470676
SIZE (gcc/gcc-4.8.0.tar.bz2) = 86663646
Index: files/system-openbsd-sparcv9.ads
===
RCS file: files/system-openbsd-sparcv9.ads
diff -N files/system-openbsd-sparcv9.ads
--- /dev/null 1 Jan 1970 00:00:00 -
+++ files/system-openbsd-sparcv9.ads11 Apr 2013 20:46:58 -
@@ -0,0 +1,142 @@
+--
+-- --
+--GNAT RUN-TIME COMPONENTS --
+-- --
+-- S Y S T E M--
+-- --
+-- S p e c --
+-- (OpenBSD/sparc64 Version)
Re: lang/gcc/4.8 sparc64 gnat support
On Thu, Apr 11, 2013 at 11:25:15PM +0200, Pascal Stumpf wrote:
On Thu, 11 Apr 2013 22:51:30 +0200, Tobias Ulmer wrote:
Add sparc64 to the list of supported Ada archs
How about this?
Index: Makefile
===
RCS file: /home/vcs/cvs/openbsd/ports/lang/gcc/4.8/Makefile,v
retrieving revision 1.1.1.1
diff -u -p -u -p -r1.1.1.1 Makefile
--- Makefile8 Apr 2013 08:20:19 - 1.1.1.1
+++ Makefile11 Apr 2013 22:09:54 -
@@ -12,6 +12,9 @@ ADASTRAP-i386 = adastrap-i386-$V-1.tar.x
ADASTRAP_LIBC-amd64 = 67.0
ADASTRAP_LIBM-amd64 = 8.0
ADASTRAP-amd64 = adastrap-amd64-$V-1.tar.xz
+ADASTRAP_LIBC-sparc64 = 67.0
+ADASTRAP_LIBM-sparc64 = 8.0
+ADASTRAP-sparc64 = adastrap-sparc64-$V-0.tar.xz
PKGNAME-main = gcc-${FULL_PKGVERSION}
PKGNAME-c++ = g++-${FULL_PKGVERSION}
@@ -24,6 +27,8 @@ PKGNAME-ada = gnat-${FULL_PKGVERSION}
#PKGNAME-go = gccgo-${FULL_PKGVERSION}
PKGSPEC-main = gcc-=4.8,4.9
+REVISION-ada = 0
+
SHARED_LIBS = estdc++ 16.0 \
gfortran5.0 \
gcj 3.0 \
@@ -41,7 +46,7 @@ SHARED_LIBS = estdc++ 16.0 \
PSEUDO_FLAVORS = no_c++ no_f95 no_objc no_java no_ada no_go full
FLAVOR ?=
-ONLY_FOR_ARCHS-ada = amd64 i386
+ONLY_FOR_ARCHS-ada = amd64 i386 sparc64
ONLY_FOR_ARCHS-java = amd64 i386 powerpc sparc64
MULTI_PACKAGES = -main -f95 -objc -c++ -estdc -java -ada
@@ -59,7 +64,8 @@ DISTNAME =gcc-${FULL_VERSION}
DISTFILES =${DISTNAME}.tar.bz2
# there's no ecj-4.8.jar
ECJ = ecj-4.5.jar
-SUPDISTFILES = ${ADASTRAP-amd64}:0 ${ADASTRAP-i386}:0 ${ECJ}:1
+SUPDISTFILES = ${ADASTRAP-amd64}:0 ${ADASTRAP-i386}:0 ${ECJ}:1 \
+ ${ADASTRAP-sparc64}:0
EXTRACT_ONLY = ${DISTNAME}.tar.bz2
BUILD_DEPENDS += devel/bison \
@@ -136,7 +142,8 @@ CONFIGURE_ARGS += \
--enable-threads=posix \
--enable-wchar_t \
--with-gmp=${LOCALBASE} \
- --enable-languages=${LANGS}
+ --enable-languages=${LANGS} \
+ --disable-libstdcxx-pch
# This is needed, as internal cpp is no longer compatible with the
# visible beast
Index: distinfo
===
RCS file: /home/vcs/cvs/openbsd/ports/lang/gcc/4.8/distinfo,v
retrieving revision 1.1.1.1
diff -u -p -u -p -r1.1.1.1 distinfo
--- distinfo8 Apr 2013 08:20:19 - 1.1.1.1
+++ distinfo11 Apr 2013 22:09:54 -
@@ -1,8 +1,10 @@
SHA256 (gcc/adastrap-amd64-4.8.0-1.tar.xz) =
8zqPYBYcNX913KeXrRlg37VHLuPGnssCCob4Fc51hNg=
SHA256 (gcc/adastrap-i386-4.8.0-1.tar.xz) =
HoPDSgf0fCHyyKknoxThnFgxWZ0T3mdSiXBomEEH5m4=
+SHA256 (gcc/adastrap-sparc64-4.8.0-0.tar.xz) =
g9RJkaXJXtumPRctcUSH83xKXrd8BDLyB2MV6r3EgHs=
SHA256 (gcc/ecj-4.5.jar) = mP0Sjx03TZ5C/Z1INr3SScbVEevGwN8X+8G535bD14E=
SHA256 (gcc/gcc-4.8.0.tar.bz2) = sDf+UTK3HsrS6nFB7JIpK10yQnv5D9kM3kMrHVq6zCw=
SIZE (gcc/adastrap-amd64-4.8.0-1.tar.xz) = 31996808
SIZE (gcc/adastrap-i386-4.8.0-1.tar.xz) = 30038792
+SIZE (gcc/adastrap-sparc64-4.8.0-0.tar.xz) = 27002840
SIZE (gcc/ecj-4.5.jar) = 1470676
SIZE (gcc/gcc-4.8.0.tar.bz2) = 86663646
Index: patches/patch-gcc_ada_adaint_c
===
RCS file:
/home/vcs/cvs/openbsd/ports/lang/gcc/4.8/patches/patch-gcc_ada_adaint_c,v
retrieving revision 1.1.1.1
diff -u -p -u -p -r1.1.1.1 patch-gcc_ada_adaint_c
--- patches/patch-gcc_ada_adaint_c 8 Apr 2013 08:20:24 - 1.1.1.1
+++ patches/patch-gcc_ada_adaint_c 11 Apr 2013 22:09:54 -
@@ -1,7 +1,17 @@
$OpenBSD: patch-gcc_ada_adaint_c,v 1.1.1.1 2013/04/08 08:20:24 pascal Exp $
gcc/ada/adaint.c.orig Thu Jan 3 11:09:24 2013
-+++ gcc/ada/adaint.c Sat Jan 12 13:27:07 2013
-@@ -3855,3 +3855,9 @@ void __gnat_cpu_set (int cpu, size_t count ATTRIBUTE_U
+--- gcc/ada/adaint.c.orig Wed Feb 6 13:01:20 2013
gcc/ada/adaint.c Wed Apr 10 01:42:27 2013
+@@ -2467,7 +2467,8 @@ __gnat_number_of_cpus (void)
+ {
+ int cores = 1;
+
+-#if defined (linux) || defined (sun) || defined (AIX) || defined (__APPLE__)
++#if defined (linux) || defined (sun) || defined (AIX) || defined (__APPLE__)
||\
++ defined (__OpenBSD__)
+ cores = (int) sysconf (_SC_NPROCESSORS_ONLN);
+
+ #elif defined (__hpux__)
+@@ -3855,3 +3856,9 @@ void __gnat_cpu_set (int cpu, size_t count ATTRIBUTE_U
#ifdef __cplusplus
}
#endif
Index: patches/patch-gcc_ada_gcc-interface_Makefile_in
===
RCS file:
/home/vcs/cvs/openbsd/ports/lang/gcc/4.8/patches/patch-gcc_ada_gcc-interface_Makefile_in,v
retrieving revision 1.1.1.1
diff -u -p -u -p -r1.1.1.1 patch-gcc_ada_gcc-interface_Makefile_in
--- patches/patch-gcc_ada_gcc-interface_Makefile_in 8 Apr 2013 08:20:24
- 1.1.1.1
+++ patches/patch-gcc_ada_gcc-interface_Makefile_in 11 Apr 2013 22:09:54
-
@@ -1,7 +1,7 @@
$OpenBSD: patch-gcc_ada_gcc-interface_Makefile_in,v 1.1.1.1 2013/04/08
08:20:24 pascal
Re: new: gcc/4.8
On Tue, Mar 26, 2013 at 04:09:33PM +0100, Pascal Stumpf wrote:
On Tue, 26 Mar 2013 15:47:45 +0100, Pascal Stumpf wrote:
So here's the final version of GCC 4.8.0. Tested by me on amd64, i386,
powerpc and by jsg some time ago on hppa. sparc64 hasn't been tested
yet, so it's disabled for now. Needless to say, it would be cool if
someone could try that out, as well as hppa again and maybe alpha. I
haven't merged the mips64(el) backend patches yet, but I'll do that
together with bcallah@ as soon as binutils 2.17 is ready on those archs.
-java works, and I've merged kurt@'s patches from 4.6. The resulting
gcj can compile classpath and pdftk. Also added support for -java on
powerpc.
Disabled -faggressive-loop-optimizations. It breaks SPEC.
The compiler is now written in C++. This means that adastrap must now
also contain the c++ part of GCC.
PIC/no-PIC bullshit is no more (yay).
I hope plists are correct. Updating this port without a working
update-plist is a real PITA.
Aand of course I just found a small problem: move ${CONFIG]-e[gc]++ to
PLIST-c++.
I've so far only looked at the Ada stuff, which is OK by me (lets commit
this, nothing depends on it and it's not going to get better outside the
tree)
Changes to the tarball:
- Bundle libm as in 4.7
- Remove USE_LIBTOOL, default now.
- Point to strip, which failed before because our install(1) is silly.
This decreases adastrap size by ~15MB (!!)
- FULLPKGNAME otherwise it blows up as soon as REVISION is set.
- Install with -j1 as done with others, parallel building seems fine
with 6 and 12 cores here.
- Drop messing with the -O flags, it seems a little cargo-cultish by now
and I could observe no negative effect.
- -lpthread as done elsewhere
I'm not really happy about the e prefix to the gnat tools, but that
seems the direction they're going. Dropping that would require to
bring back a lot of patches which is also annoying, so.. such is life.
diff --git a/lang/gcc/4.8/Makefile b/lang/gcc/4.8/Makefile
index d345ea6..4685341 100644
--- a/lang/gcc/4.8/Makefile
+++ b/lang/gcc/4.8/Makefile
@@ -6,10 +6,12 @@ V = 4.8.0
FULL_VERSION = $V
FULL_PKGVERSION = $V
-ADASTRAP_LIBC-i386 = 66.2
-ADASTRAP-i386 = adastrap-i386-$V-0.tar.xz
-ADASTRAP_LIBC-amd64 = 66.2
-ADASTRAP-amd64 = adastrap-amd64-$V-0.tar.xz
+ADASTRAP_LIBC-i386 = 67.0
+ADASTRAP_LIBM-i386 = 8.0
+ADASTRAP-i386 = adastrap-i386-$V-1.tar.xz
+ADASTRAP_LIBC-amd64 = 67.0
+ADASTRAP_LIBM-amd64 = 8.0
+ADASTRAP-amd64 = adastrap-amd64-$V-1.tar.xz
PKGNAME-main = gcc-${FULL_PKGVERSION}
PKGNAME-c++ = g++-${FULL_PKGVERSION}
@@ -60,8 +62,6 @@ ECJ = ecj-4.5.jar
SUPDISTFILES = ${ADASTRAP-amd64}:0 ${ADASTRAP-i386}:0 ${ECJ}:1
EXTRACT_ONLY = ${DISTNAME}.tar.bz2
-USE_LIBTOOL = Yes
-
BUILD_DEPENDS += devel/bison \
devel/libexecinfo
@@ -110,6 +110,7 @@ MODGNU_CONFIG_GUESS_DIRS = ${WRKSRC} ${WRKSRC}/gcc
EXTRA_ENV += am_cv_func_iconv=no JAR=no
EXTRA_ENV += ac_cv_prog_M4=/usr/bin/m4
EXTRA_ENV += ac_cv_header_magic_h=no
+EXTRA_ENV += STRIP=/usr/bin/strip
MAKE_FLAGS += ac_cv_path_mkdir=/bin/mkdir
CONFIGURE_ENV += ${EXTRA_ENV}
@@ -253,22 +254,35 @@ post-install:
.include bsd.port.mk
-ADA_PACKAGE = ${PACKAGE_REPOSITORY}/${MACHINE_ARCH}/all/${PKGNAME-ada}.tgz
-GCC_PACKAGE = ${PACKAGE_REPOSITORY}/${MACHINE_ARCH}/all/${PKGNAME-main}.tgz
-GXX_PACKAGE = ${PACKAGE_REPOSITORY}/${MACHINE_ARCH}/all/${PKGNAME-c++}.tgz
+ADA_PACKAGE = ${PACKAGE_REPOSITORY}/${MACHINE_ARCH}/all/${FULLPKGNAME-ada}.tgz
+GCC_PACKAGE = ${PACKAGE_REPOSITORY}/${MACHINE_ARCH}/all/${FULLPKGNAME-main}.tgz
+GXX_PACKAGE = ${PACKAGE_REPOSITORY}/${MACHINE_ARCH}/all/${FULLPKGNAME-c++}.tgz
adastrap: ${ADA_PACKAGE} ${GCC_PACKAGE} ${GXX_PACKAGE}
-.if !defined(ADASTRAP_LIBC-${MACHINE_ARCH})
- @echo 12 Error: you must specify a libc to bundle
+.if !defined(ADASTRAP_LIBC-${MACHINE_ARCH}) || \
+ !defined(ADASTRAP_LIBM-${MACHINE_ARCH})
+ @echo 12 Error: you must specify libc and libm to bundle
@exit 1
.else
mkdir -p ${WRKDIR}/prepare/bootstrap/system-libs
cp /usr/lib/libc.so.${ADASTRAP_LIBC-${MACHINE_ARCH}} \
${WRKDIR}/prepare/bootstrap/system-libs
+ cp /usr/lib/libm.so.${ADASTRAP_LIBM-${MACHINE_ARCH}} \
+ ${WRKDIR}/prepare/bootstrap/system-libs
cd ${WRKDIR}/prepare/bootstrap tar zxf ${ADA_PACKAGE}
cd ${WRKDIR}/prepare/bootstrap tar zxf ${GCC_PACKAGE}
cd ${WRKDIR}/prepare/bootstrap tar zxf ${GXX_PACKAGE}
- rm -rf ${WRKDIR}/prepare/bootstrap/{+*,info,man}
+ rm -rf ${WRKDIR}/prepare/bootstrap/{+*,info,man,share}
+ if ! ldd ${WRKDIR}/prepare/bootstrap/bin/egcc | \
+ fgrep libc.so.${ADASTRAP_LIBC-${MACHINE_ARCH}} /dev/null; then
\
+ echo 12 Error: adastrap egcc linked against unbundled libc;
\
+ exit 1; \
+ fi
+ if ! ldd ${WRKDIR}/prepare/bootstrap/bin/egcc | \
+ fgrep
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: tobi...@cvs.openbsd.org 2013/04/03 22:21:14 Modified files: lang/gcc/4.7 : Makefile distinfo Log message: Update adastrap tarball due to libm major bump. Bundle libm with adastrap as we do for libc since it's cheap to do and can save us some work in the future. ok pascal@
Re: [NEW] py-pika python AMQP client library
On Mon, Feb 11, 2013 at 01:02:23PM +0100, Markus Bergkvist wrote: On 01/24/13 08:52, Markus Bergkvist wrote: On 01/23/13 15:39, Markus Bergkvist wrote: Pika is a pure-Python implementation of the AMQP 0-9-1 protocol that tries to stay fairly independent of the underlying network support library. Pika was developed primarily for use with RabbitMQ, but should also work with other AMQP 0-9-1 brokers. Forgot to mention, tested on amd64. Comments? Anyone else interested in this? Ports is currently locked for release. Wait for the unlock mail and remind us again, if necessary.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: tobi...@cvs.openbsd.org 2013/02/04 14:23:40 Modified files: lang/gcc/4.2 : Makefile lang/gcc/4.2/patches: patch-gcc_config_gcc Added files: lang/gcc/4.2/patches: patch-gcc_config_sh_lib1funcs_asm patch-gcc_config_sh_openbsd_h patch-gcc_config_sh_sh-protos_h patch-gcc_config_sh_sh_c patch-gcc_config_sh_sh_h patch-gcc_config_sh_sh_md patch-gcc_config_sh_sh_opt patch-gcc_config_sh_t-sh Log message: Add our SH backend. ok espie@ pascal@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: tobi...@cvs.openbsd.org 2013/01/29 05:40:06 Modified files: devel/gmp : Makefile Added files: devel/gmp/patches: patch-mpn_sh_add_n_s patch-mpn_sh_sub_n_s Log message: Fix symbol names of assembly routines for SH. ok naddy@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: tobi...@cvs.openbsd.org 2013/01/27 08:00:23 Modified files: lang/gcc/4.2 : Makefile lang/gcc/4.2/patches: patch-gcc_config_gcc Added files: lang/gcc/4.2/patches: patch-gcc_config_arm_arm_c patch-gcc_config_arm_openbsd_h Log message: Add our ARM backend to prepare for later gnat cross-compile adventures. Rein damit pascal@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: tobi...@cvs.openbsd.org 2013/01/27 18:03:05 Modified files: lang/gcc/4.2 : Makefile lang/gcc/4.6 : Makefile Log message: Remove sparc, m68k, and vax from the list of working architectures. Their respective gcc backend ports are unfinished and broken. ok pascal@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: tobi...@cvs.openbsd.org 2013/01/19 19:23:40 Modified files: lang/gcc/4.6 : Makefile lang/gcc/4.6/patches: patch-Makefile_in Log message: Fix parallel make install by forcing gmake to only run 1 job. It's not pretty and we would like to have a better fix, but for now it should suffice. ok naddy@, sthen@
Re: NEW: audio/mikmod
On Sat, Jan 12, 2013 at 12:45:57AM -0500, Brian Callahan wrote: Hi ports -- Attached is a tarball for mikmod. Built with sthen's audio/libmikmod patch sent yesterday. Patch work done by tobiasu. OK? Tested on i386, amd64, sparc64 and hppa. I'm OK with this, but since I did the patches, you might want to get a second opinion. ~Brian
Re: SDL doesn't works on sparc64
On Fri, Jan 11, 2013 at 09:42:13AM +0100, David Coppa wrote: On Thu, Jan 10, 2013 at 4:45 PM, contezero con...@gmail.com wrote: Hello Jacob Meuser, I have tried to use SDL library on my SunBlade 2500, but I only get a core dump. There're some (maybe) useful pointers here: http://article.gmane.org/gmane.comp.lib.sdl/5095/ http://article.gmane.org/gmane.comp.lib.sdl/5173/ Ciao, David Not really. As contezero correctly identifies, the problem is that these cards only expose a 7bit framebuffer, that's how the prom sets them up I presume. We don't know how to change that. Either teach SDL about the 7bit framebuffer, or somehow pretend that that it is a 8 bit framebuffer and ignore the broken image. I don't know which is easier. The proper solution would be to figure out how the damn cards work and take a closer look at these binaries: http://hub.opensolaris.org/bin/view/downloads/sparc_graphics
Re: audio/mikmod: fix bug causing ri-li crash
On Thu, Jan 10, 2013 at 02:00:08AM +0100, Stefan Sperling wrote:
On Thu, Jan 10, 2013 at 12:37:54AM +, Stuart Henderson wrote:
On 2012/12/31 01:50, Stefan Sperling wrote:
Perhaps mikmod should be updated but I'd rather just fix this for now.
updated to 3.2.0 here, some of the patches were rolled in, and I've
added a comment in patches for the remaining ones where it's not obvious.
Diff is fine with me. Unfortunately this update breaks sdl-mixer :(
You need this. The new version does segfault with mikmod, currently
debugging...
Index: files/drv_sndio.c
===
RCS file: /home/vcs/cvs/openbsd/ports/audio/libmikmod/files/drv_sndio.c,v
retrieving revision 1.1
diff -u -p -r1.1 drv_sndio.c
--- files/drv_sndio.c 11 Jan 2010 04:25:26 - 1.1
+++ files/drv_sndio.c 10 Jan 2013 01:46:05 -
@@ -65,7 +65,7 @@ static BOOL Sndio_Init(void)
return 1;
}
- if (!(audiobuffer = (SBYTE *)_mm_malloc(fragsize)))
+ if (!(audiobuffer = (SBYTE *)MikMod_malloc(fragsize)))
return 1;
sio_initpar(par);
@@ -106,7 +106,7 @@ static BOOL Sndio_Init(void)
static void Sndio_Exit(void)
{
VC_Exit();
- _mm_free(audiobuffer);
+ MikMod_free(audiobuffer);
if (hdl) {
sio_close(hdl);
hdl = NULL;
Re: audio/mikmod: fix bug causing ri-li crash
Take two:
Fix MDRIVER initialisation, add help text. Sndio_PlayStop must call
sio_stop(), otherwise mikmod playback stops after one tune.
Tested with mikmod using sndio/wav drivers on amd64/sparc64. I did not
check port depends.
With the fix below, ok tobiasu@
Index: files/drv_sndio.c
===
RCS file: /home/vcs/cvs/openbsd/ports/audio/libmikmod/files/drv_sndio.c,v
retrieving revision 1.1
diff -u -p -r1.1 drv_sndio.c
--- files/drv_sndio.c 11 Jan 2010 04:25:26 - 1.1
+++ files/drv_sndio.c 10 Jan 2013 04:51:03 -
@@ -65,7 +65,7 @@ static BOOL Sndio_Init(void)
return 1;
}
- if (!(audiobuffer = (SBYTE *)_mm_malloc(fragsize)))
+ if (!(audiobuffer = (SBYTE *)MikMod_malloc(fragsize)))
return 1;
sio_initpar(par);
@@ -106,7 +106,7 @@ static BOOL Sndio_Init(void)
static void Sndio_Exit(void)
{
VC_Exit();
- _mm_free(audiobuffer);
+ MikMod_free(audiobuffer);
if (hdl) {
sio_close(hdl);
hdl = NULL;
@@ -139,9 +139,8 @@ static BOOL Sndio_PlayStart(void)
static void Sndio_PlayStop(void)
{
- //sio_stop(hdl);
-
VC_PlayStop();
+ sio_stop(hdl);
}
MIKMODAPI MDRIVER drv_sndio = {
@@ -150,7 +149,7 @@ MIKMODAPI MDRIVER drv_sndio = {
sndio audio driver v1.0,
0, 255,
audio,
-
+buffer:r:7,17,12:Audio buffer log2 size\n,
Sndio_CommandLine,
Sndio_IsThere,
VC_SampleLoad,
Re: UPDATE: devel/cscope
On Thu, Jan 03, 2013 at 03:34:58PM +0100, Benoit Lecocq (blq) wrote:
Hi,
This diff updates cscope to the latest release 15.8a.
Comments ? OK ?
Tested on sparc64, standalone and in vim. Seems fine.
ok tobiasu@
Cheers,
benoit
Index: Makefile
===
RCS file: /cvs/ports/devel/cscope/Makefile,v
retrieving revision 1.15
diff -u -r1.15 Makefile
--- Makefile 15 Nov 2010 19:45:57 - 1.15
+++ Makefile 3 Jan 2013 14:29:58 -
@@ -2,7 +2,7 @@
COMMENT= code browsing program
-DISTNAME=cscope-15.7a
+DISTNAME=cscope-15.8a
CATEGORIES= devel
HOMEPAGE=http://cscope.sourceforge.net/
@@ -18,10 +18,7 @@
MASTER_SITES=${MASTER_SITE_SOURCEFORGE:=cscope/}
-EXTRACT_SUFX = .tar.bz2
-
CONFIGURE_STYLE= gnu
-USE_GROFF = Yes
post-install:
@cd ${WRKSRC}; ${INSTALL_SCRIPT} contrib/xcscope/cscope-indexer \
Index: distinfo
===
RCS file: /cvs/ports/devel/cscope/distinfo,v
retrieving revision 1.6
diff -u -r1.6 distinfo
--- distinfo 26 May 2009 15:44:54 - 1.6
+++ distinfo 3 Jan 2013 14:29:58 -
@@ -1,5 +1,2 @@
-MD5 (cscope-15.7a.tar.bz2) = 2kOYdiKs6MNrvxTBWjUOwQ==
-RMD160 (cscope-15.7a.tar.bz2) = i8kE9jbo5upFQnOAlRg0dZEYWPk=
-SHA1 (cscope-15.7a.tar.bz2) = 9jSGlOVEN2mt2FH5f9OTZek9xHQ=
-SHA256 (cscope-15.7a.tar.bz2) = AmOLy6eQvIsFYvKNvnieYXlEFQedlKZ278KH0Y3JYDc=
-SIZE (cscope-15.7a.tar.bz2) = 361214
+SHA256 (cscope-15.8a.tar.gz) = 63NqxA1avr6PpGggx6jszIoXlmqaX3A3U2e3cXeHTR4=
+SIZE (cscope-15.8a.tar.gz) = 601232
Re: UPDATE: games/openttd
On Thu, Dec 20, 2012 at 09:19:41PM -0700, Anthony J. Bentley wrote: Hi, Here is an update to openttd-1.2.3. Tested on i386 and extensively on amd64. ok? Works for me (amd64). ok tobiasu
Re: Remove www/firefox35 ?
On Sat, Nov 24, 2012 at 12:03:09PM +0100, Landry Breuil wrote: Hi, i'd like to remove www/firefox35 from the ports tree : - it's been EOL'ed upstream since 18 months, and thus receives no security updates. I have no idea if there are users of this port still around. - it was originally kept in the tree because it was the only version working with the java plugin from devel/jdk/1.6, but the java plugin is known for an enormous amount of security vulns those days, and i think 1.6 doesnt receive security updates regarding this (kurt ?) Firefox blocks it by default, see http://blog.mozilla.org/addons/2012/04/02/blocking-java/ and https://addons.mozilla.org/en-US/firefox/blocked/p80. I havent personally tested that. - those days we have www/icedtea-web, which apparently works not so badly, and is maintained. So, unless someone steps up with a valid usecase (ie 'my java applet used by thousands of users work in the old crusty insecure java plugin and not icedtea-web'), i'd like to remove www/firefox35. Opinions ? Yays ? Landry (of course, www/firefox36, even if EOL'ed upstream too, stays for sparc64.. sigh.) I was just about to complain because of sparc64 ;) Since we still have ff 3.6, I'm all for removing 3.5.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: tobi...@cvs.openbsd.org 2012/11/08 17:58:26 Modified files: lang/gcc/4.7 : distinfo Log message: Oops, revert adastrap-i386-4.7.2-0.tar.xz checksum back to previous value This slipped in due to testing adastrap on i386. No bump required since the package doesn't build :( OK sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: tobi...@cvs.openbsd.org 2012/11/05 17:57:43 Modified files: lang/gcc/4.2 : Makefile distinfo lang/gcc/4.6 : Makefile distinfo lang/gcc/4.7 : Makefile distinfo lang/gcc/4.7/patches: patch-gcc_ada_gcc-interface_Makefile_in Log message: Add Ada support on amd64 and sparc64 Also unbreak and improve adastrap generation Cleanup bootstrap libc handling ok pascal@
lang/gcc - gnat for amd64 and sparc64
I think this is ready for inclusion. Nobody complained during my
holiday, and I'm quite happy with it.
There's more work to be done in order to build GtkAda and others, but
I'm looking for OKs on this diff first:
(note: PARALLEL_INSTALL=No, the gnat part can't handle it)
Index: 4.2/Makefile
===
RCS file: /home/vcs/cvs/openbsd/ports/lang/gcc/4.2/Makefile,v
retrieving revision 1.86
diff -u -p -r1.86 Makefile
--- 4.2/Makefile7 Oct 2012 20:09:44 - 1.86
+++ 4.2/Makefile4 Nov 2012 19:43:18 -
@@ -7,10 +7,13 @@ DPB_PROPERTIES = parallel
V = 4.2.4
FULL_VERSION = $V
FULL_PKGVERSION = $V
-BOOTSTRAP_GEN = 5
-# XXX adastrap bundles libc.so.60.1 (from OpenBSD 5.0) to build on i386
-LIBC_VERSION = 60.1
-ADASTRAP = adastrap-i386-$V-${BOOTSTRAP_GEN}.tgz
+
+ADASTRAP_LIBC-i386 = 60.1
+ADASTRAP-i386 = adastrap-i386-$V-5.tgz
+ADASTRAP_LIBC-amd64 = 66.0
+ADASTRAP-amd64 = adastrap-amd64-$V-1.tgz
+ADASTRAP_LIBC-sparc64 = 66.0
+ADASTRAP-sparc64 = adastrap-sparc64-$V-1.tgz
PKGNAME-main = gcc-${FULL_PKGVERSION}
PKGNAME-c++ = g++-${FULL_PKGVERSION}
@@ -20,13 +23,13 @@ PKGNAME-java = gcj-${FULL_PKGVERSION}
PKGNAME-objc = gobjc-${FULL_PKGVERSION}
PKGNAME-ada = gnat-${FULL_PKGVERSION}
-REVISION-main = 13
+REVISION-main = 14
REVISION-c++ = 17
REVISION-estdc = 5
REVISION-f95 = 15
REVISION-java = 15
REVISION-objc = 14
-REVISION-ada = 11
+REVISION-ada = 12
EPOCH = 0
SHARED_LIBS = estdc++ 13.0 \
@@ -48,7 +51,7 @@ USE_GROFF = Yes
PSEUDO_FLAVORS = no_c++ no_f95 no_objc no_java no_ada full
FLAVOR ?=
-ONLY_FOR_ARCHS-ada = i386
+ONLY_FOR_ARCHS-ada = i386 amd64 sparc64
ONLY_FOR_ARCHS-java = amd64 sparc64 i386
MULTI_PACKAGES = -main -f95 -objc -java -c++ -estdc -ada
@@ -61,7 +64,7 @@ BUILD_PACKAGES := ${BUILD_PACKAGES:N-est
.endif
DISTFILES = ${DISTNAME}.tar.bz2
-SUPDISTFILES = ${ADASTRAP}:0
+SUPDISTFILES = ${ADASTRAP-i386}:0 ${ADASTRAP-amd64}:0 ${ADASTRAP-sparc64}:0
BUILD_DEPENDS += devel/autoconf/2.59
BUILD_DEPENDS += devel/metaauto
@@ -101,7 +104,7 @@ CONFIGURE_ARGS += --enable-libgcj
.endif
.if ${BUILD_PACKAGES:M-ada}
LANGS := ${LANGS},ada
-DISTFILES += ${ADASTRAP}:0
+DISTFILES += ${ADASTRAP-${MACHINE_ARCH}}:0
CONFIGURE_ENV = ADAC=${WRKDIR}/bin/egcc GNATBIND=${WRKDIR}/bin/gnatbind
MAKE_ENV = ADAC=${WRKDIR}/bin/egcc GNATBIND=${WRKDIR}/bin/gnatbind
CC = ${WRKDIR}/bin/egcc
@@ -119,9 +122,6 @@ post-extract:
.for l v in ${SHARED_LIBS}
echo LIB$l_LTVERSION = -version-info ${v:S/./:/}
${WRKSRC}/libversions
.endfor
-.if ${BUILD_PACKAGES:M-ada}
- ${WRKDIR}/bin/egcc -v # if this fails, you must extract suitable
libraries from the relevant OpenBSD release or use the no_ada FLAVOR.
-.endif
post-install:
.if ${BUILD_PACKAGES:M-ada}
@@ -159,8 +159,8 @@ CONFIGURE_ARGS += \
# visible beast
CONFIGURE_ARGS += --enable-cpp
-CONFIGURE_ARGS += --with-gnu-as
-CONFIGURE_ARGS += --with-gnu-ld
+CONFIGURE_ARGS += --with-gnu-as
+CONFIGURE_ARGS += --with-gnu-ld
CONFIGURE_ARGS += ${CONFIGURE_SHARED}
post-patch:
@@ -208,6 +208,7 @@ RUN_DEPENDS-java = archivers/unzip \
archivers/zip=2.3p0
RUN_DEPENDS-c++ = ${RUN_DEPENDS}
libstdc++-=4.2,4.3|libstdc++-=4.2v0,4.3v0:${BUILD_PKGPATH},-estdc
+RUN_DEPENDS-ada = ${FULLPKGNAME-main}:${BUILD_PKGPATH}
WANTLIB-estdc =m
WANTLIB-java = c m pthread z
@@ -216,19 +217,26 @@ LIB_DEPENDS-f95 = devel/gmp=4.2 \
WANTLIB-f95 = ${WANTLIB} gmp mpfr
.include bsd.port.mk
-ADA_PACKAGE = ${PACKAGE_REPOSITORY}/i386/all/${PKGNAME-ada}${PKG_SUFX}
-GCC_PACKAGE = ${PACKAGE_REPOSITORY}/i386/all/${PKGNAME-main}${PKG_SUFX}
+ADA_PACKAGE = ${PACKAGE_REPOSITORY}/${MACHINE_ARCH}/all/${FULLPKGNAME-ada}.tgz
+GCC_PACKAGE = ${PACKAGE_REPOSITORY}/${MACHINE_ARCH}/all/${FULLPKGNAME-main}.tgz
adastrap: ${ADA_PACKAGE} ${GCC_PACKAGE}
-.if !defined(LIBC_VERSION)
+.if !defined(ADASTRAP_LIBC-${MACHINE_ARCH})
@echo 12 Error: you must specify a libc to bundle
@exit 1
.else
mkdir -p ${WRKDIR}/prepare/bootstrap/system-libs
- cp /usr/lib/libc.so.${LIBC_VERSION}
${WRKDIR}/prepare/bootstrap/system-libs
+ cp /usr/lib/libc.so.${ADASTRAP_LIBC-${MACHINE_ARCH}} \
+ ${WRKDIR}/prepare/bootstrap/system-libs
cd ${WRKDIR}/prepare/bootstrap tar zxf ${ADA_PACKAGE}
cd ${WRKDIR}/prepare/bootstrap tar zxf ${GCC_PACKAGE}
rm -rf ${WRKDIR}/prepare/bootstrap/{+*,info,man}
- cd ${WRKDIR}/prepare tar zcf ${FULLDISTDIR}/${ADASTRAP} bootstrap
+ if ! ldd ${WRKDIR}/prepare/bootstrap/bin/egcc | \
+ fgrep libc.so.${ADASTRAP_LIBC-${MACHINE_ARCH}} /dev/null; then
\
+ echo 12 Error: adastrap egcc linked against wrong libc; \
+ exit 1; \
+ fi
+ cd ${WRKDIR}/prepare \
+ tar zcf ${FULLDISTDIR}/${ADASTRAP-${MACHINE_ARCH}} bootstrap
.endif
Index: 4.2/distinfo
Re: lang/gcc: add Ada support on amd64
On Thu, Sep 27, 2012 at 01:02:07AM +0200, Tobias Ulmer wrote:
sparc64 is building native as we speak. Fingers crossed...
WIP diff now includes sparc64. Powerpc segfaults straight away
and requires more investigation.
Index: 4.2/Makefile
===
RCS file: /home/vcs/cvs/openbsd/ports/lang/gcc/4.2/Makefile,v
retrieving revision 1.85
diff -u -p -r1.85 Makefile
--- 4.2/Makefile28 Sep 2012 15:02:28 - 1.85
+++ 4.2/Makefile4 Oct 2012 00:58:22 -
@@ -7,10 +7,13 @@ DPB_PROPERTIES = parallel
V = 4.2.4
FULL_VERSION = $V
FULL_PKGVERSION = $V
-BOOTSTRAP_GEN = 5
-# XXX adastrap bundles libc.so.60.1 (from OpenBSD 5.0) to build on i386
-LIBC_VERSION = 60.1
-ADASTRAP = adastrap-i386-$V-${BOOTSTRAP_GEN}.tgz
+
+ADASTRAP_LIBC-i386 = 60.1
+ADASTRAP-i386 = adastrap-i386-$V-5.tgz
+ADASTRAP_LIBC-amd64 = 66.0
+ADASTRAP-amd64 = adastrap-amd64-$V-1.tgz
+ADASTRAP_LIBC-sparc64 = 66.0
+ADASTRAP-sparc64 = adastrap-sparc64-$V-1.tgz
PKGNAME-main = gcc-${FULL_PKGVERSION}
PKGNAME-c++ = g++-${FULL_PKGVERSION}
@@ -20,7 +23,7 @@ PKGNAME-java = gcj-${FULL_PKGVERSION}
PKGNAME-objc = gobjc-${FULL_PKGVERSION}
PKGNAME-ada = gnat-${FULL_PKGVERSION}
-REVISION-main = 12
+REVISION-main = 14
REVISION-c++ = 16
REVISION-estdc = 4
REVISION-f95 = 14
@@ -48,7 +51,7 @@ USE_GROFF = Yes
PSEUDO_FLAVORS = no_c++ no_f95 no_objc no_java no_ada full
FLAVOR ?=
-ONLY_FOR_ARCHS-ada = i386
+ONLY_FOR_ARCHS-ada = i386 amd64 sparc64
ONLY_FOR_ARCHS-java = amd64 sparc64 i386
MULTI_PACKAGES = -main -f95 -objc -java -c++ -estdc -ada
@@ -61,7 +64,7 @@ BUILD_PACKAGES := ${BUILD_PACKAGES:N-est
.endif
DISTFILES = ${DISTNAME}.tar.bz2
-SUPDISTFILES = ${ADASTRAP}:0
+SUPDISTFILES = ${ADASTRAP-i386}:0 ${ADASTRAP-amd64}:0 ${ADASTRAP-sparc64}:0
BUILD_DEPENDS += devel/autoconf/2.59
BUILD_DEPENDS += devel/metaauto
@@ -101,7 +104,7 @@ CONFIGURE_ARGS += --enable-libgcj
.endif
.if ${BUILD_PACKAGES:M-ada}
LANGS := ${LANGS},ada
-DISTFILES += ${ADASTRAP}:0
+DISTFILES += ${ADASTRAP-${MACHINE_ARCH}}:0
CONFIGURE_ENV = ADAC=${WRKDIR}/bin/egcc GNATBIND=${WRKDIR}/bin/gnatbind
MAKE_ENV = ADAC=${WRKDIR}/bin/egcc GNATBIND=${WRKDIR}/bin/gnatbind
CC = ${WRKDIR}/bin/egcc
@@ -119,9 +122,6 @@ post-extract:
.for l v in ${SHARED_LIBS}
echo LIB$l_LTVERSION = -version-info ${v:S/./:/}
${WRKSRC}/libversions
.endfor
-.if ${BUILD_PACKAGES:M-ada}
- ${WRKDIR}/bin/egcc -v # if this fails, you must extract suitable
libraries from the relevant OpenBSD release or use the no_ada FLAVOR.
-.endif
post-install:
.if ${BUILD_PACKAGES:M-ada}
@@ -159,8 +159,8 @@ CONFIGURE_ARGS += \
# visible beast
CONFIGURE_ARGS += --enable-cpp
-CONFIGURE_ARGS += --with-gnu-as
-CONFIGURE_ARGS += --with-gnu-ld
+CONFIGURE_ARGS += --with-gnu-as
+CONFIGURE_ARGS += --with-gnu-ld
CONFIGURE_ARGS += ${CONFIGURE_SHARED}
post-patch:
@@ -208,6 +208,7 @@ RUN_DEPENDS-java = archivers/unzip \
archivers/zip=2.3p0
RUN_DEPENDS-c++ = ${RUN_DEPENDS}
libstdc++-=4.2,4.3|libstdc++-=4.2v0,4.3v0:${BUILD_PKGPATH},-estdc
+RUN_DEPENDS-ada = ${FULLPKGNAME-main}:${BUILD_PKGPATH}
WANTLIB-estdc =m
WANTLIB-java = c m pthread z
@@ -216,19 +217,26 @@ LIB_DEPENDS-f95 = devel/gmp=4.2 \
WANTLIB-f95 = ${WANTLIB} gmp mpfr
.include bsd.port.mk
-ADA_PACKAGE = ${PACKAGE_REPOSITORY}/i386/all/${PKGNAME-ada}${PKG_SUFX}
-GCC_PACKAGE = ${PACKAGE_REPOSITORY}/i386/all/${PKGNAME-main}${PKG_SUFX}
+ADA_PACKAGE = ${PACKAGE_REPOSITORY}/${MACHINE_ARCH}/all/${FULLPKGNAME-ada}.tgz
+GCC_PACKAGE = ${PACKAGE_REPOSITORY}/${MACHINE_ARCH}/all/${FULLPKGNAME-main}.tgz
adastrap: ${ADA_PACKAGE} ${GCC_PACKAGE}
-.if !defined(LIBC_VERSION)
+.if !defined(ADASTRAP_LIBC-${MACHINE_ARCH})
@echo 12 Error: you must specify a libc to bundle
@exit 1
.else
mkdir -p ${WRKDIR}/prepare/bootstrap/system-libs
- cp /usr/lib/libc.so.${LIBC_VERSION}
${WRKDIR}/prepare/bootstrap/system-libs
+ cp /usr/lib/libc.so.${ADASTRAP_LIBC-${MACHINE_ARCH}} \
+ ${WRKDIR}/prepare/bootstrap/system-libs
cd ${WRKDIR}/prepare/bootstrap tar zxf ${ADA_PACKAGE}
cd ${WRKDIR}/prepare/bootstrap tar zxf ${GCC_PACKAGE}
rm -rf ${WRKDIR}/prepare/bootstrap/{+*,info,man}
- cd ${WRKDIR}/prepare tar zcf ${FULLDISTDIR}/${ADASTRAP} bootstrap
+ if ! ldd ${WRKDIR}/prepare/bootstrap/bin/egcc | \
+ grep libc.so.${ADASTRAP_LIBC-${MACHINE_ARCH}} /dev/null; then \
+ echo 12 Error: adastrap egcc linked against wrong libc; \
+ exit 1; \
+ fi
+ cd ${WRKDIR}/prepare \
+ tar zcf ${FULLDISTDIR}/${ADASTRAP-${MACHINE_ARCH}} bootstrap
.endif
Index: 4.2/distinfo
===
RCS file: /home/vcs/cvs/openbsd/ports/lang/gcc/4.2/distinfo,v
retrieving revision 1.13
diff -u -p -r1.13 distinfo
--- 4.2/distinfo23 Jan
Re: lang/gcc: add Ada support on amd64
On Thu, Sep 27, 2012 at 01:38:49PM +, John Long wrote: On Thu, Sep 27, 2012 at 12:33:12AM +0200, Marc Espie wrote: On Wed, Sep 26, 2012 at 07:42:07PM +0200, Tobias Ulmer wrote: Add amd64 support to gnat 4.2 - 4.7. snipped Question: would you have the patience to start doing that on other targets ? It would be nice if ada were supported on our major platforms, and not just i386/amd64... Yes! That would be awesome! Especially if someone could get it going on mipsel64... was glad to see his message about sparc64, maybe that will give me incentive to un-mothball one of my sparc boxes and put 5.2 on it when it comes out. I do not have access to a loongson machine, so it's doubtful I can help there. Also, there's no gcc in ports for mips64 or mips64el, that needs to be done first... Feel free to help ;-) Thanks Tobias! /jl
lang/gcc: add Ada support on amd64
Add amd64 support to gnat 4.2 - 4.7.
general:
* renamed LIBC_VERSION to ADASTRAP_LIBC-arch which IMO is clearer
* dropped BOOTSTRAP_GEN since there is only a single use
* unified ADASTRAP variable block across all gcc4 ports
* made RUN_DEPENDS for gnat very strict, otherwise installing gnat does
not update gcc, resulting in an error. (This may be a general problem
with these ports, someone should check if the global RUN_DEPENDS
still works as intended)
* Repaired the adastrap target.
gcc-4.6 and gcc-4.7:
* Removed the libc symlink hack. This had no effect, since the binaries
are linked against the supplied libc, not version 61.0. Probably an
artifact from bootstrapping once.
gcc-4.7:
* Adapt gnat configuration done for i386 to amd64 as well. While there,
change -pthread to -lpthread for both architectures.
The required bootstrap files are temporarily at:
http://www.tmux.org/~tobiasu/tmp/adastrap-amd64-4.2.4-1.tgz
http://www.tmux.org/~tobiasu/tmp/adastrap-amd64-4.6.3-0.tar.xz
http://www.tmux.org/~tobiasu/tmp/adastrap-amd64-4.7.1-0.tar.xz
You can also find the diffs for each gcc port in the same directory.
All adastrap tarballs are at least five generations away from the
initial bootstrap compiler. To build your own, run:
while ! heat_death_of_the_universe; do
make clean=all
make package
make adastrap
make makesum
done
Index: 4.2/Makefile
===
RCS file: /home/vcs/cvs/openbsd/ports/lang/gcc/4.2/Makefile,v
retrieving revision 1.83
diff -u -p -r1.83 Makefile
--- 4.2/Makefile1 Sep 2012 00:00:32 - 1.83
+++ 4.2/Makefile26 Sep 2012 16:45:23 -
@@ -6,10 +6,11 @@ ONLY_FOR_ARCHS = alpha i386 m68k sparc s
V = 4.2.4
FULL_VERSION = $V
FULL_PKGVERSION = $V
-BOOTSTRAP_GEN = 5
-# XXX adastrap bundles libc.so.60.1 (from OpenBSD 5.0) to build on i386
-LIBC_VERSION = 60.1
-ADASTRAP = adastrap-i386-$V-${BOOTSTRAP_GEN}.tgz
+
+ADASTRAP_LIBC-i386 = 60.1
+ADASTRAP-i386 = adastrap-i386-$V-5.tgz
+ADASTRAP_LIBC-amd64 = 66.0
+ADASTRAP-amd64 = adastrap-amd64-$V-1.tgz
PKGNAME-main = gcc-${FULL_PKGVERSION}
PKGNAME-c++ = g++-${FULL_PKGVERSION}
@@ -19,7 +20,7 @@ PKGNAME-java = gcj-${FULL_PKGVERSION}
PKGNAME-objc = gobjc-${FULL_PKGVERSION}
PKGNAME-ada = gnat-${FULL_PKGVERSION}
-REVISION-main = 12
+REVISION-main = 13
REVISION-c++ = 16
REVISION-estdc = 4
REVISION-f95 = 14
@@ -47,7 +48,7 @@ USE_GROFF = Yes
PSEUDO_FLAVORS = no_c++ no_f95 no_objc no_java no_ada full
FLAVOR ?=
-ONLY_FOR_ARCHS-ada = i386
+ONLY_FOR_ARCHS-ada = i386 amd64
ONLY_FOR_ARCHS-java = amd64 sparc64 i386
MULTI_PACKAGES = -main -f95 -objc -java -c++ -estdc -ada
@@ -60,7 +61,7 @@ BUILD_PACKAGES := ${BUILD_PACKAGES:N-est
.endif
DISTFILES = ${DISTNAME}.tar.bz2
-SUPDISTFILES = ${ADASTRAP}:0
+SUPDISTFILES = ${ADASTRAP-i386}:0 ${ADASTRAP-amd64}:0
BUILD_DEPENDS += devel/autoconf/2.59
BUILD_DEPENDS += devel/metaauto
@@ -101,7 +102,7 @@ CONFIGURE_ARGS += --enable-libgcj
.endif
.if ${BUILD_PACKAGES:M-ada}
LANGS := ${LANGS},ada
-DISTFILES += ${ADASTRAP}:0
+DISTFILES += ${ADASTRAP-${MACHINE_ARCH}}:0
CONFIGURE_ENV = ADAC=${WRKDIR}/bin/egcc GNATBIND=${WRKDIR}/bin/gnatbind
MAKE_ENV = ADAC=${WRKDIR}/bin/egcc GNATBIND=${WRKDIR}/bin/gnatbind
CC = ${WRKDIR}/bin/egcc
@@ -159,8 +160,8 @@ CONFIGURE_ARGS += \
# visible beast
CONFIGURE_ARGS += --enable-cpp
-CONFIGURE_ARGS += --with-gnu-as
-CONFIGURE_ARGS += --with-gnu-ld
+CONFIGURE_ARGS += --with-gnu-as
+CONFIGURE_ARGS += --with-gnu-ld
CONFIGURE_ARGS += ${CONFIGURE_SHARED}
post-patch:
@@ -208,6 +209,7 @@ RUN_DEPENDS-java = archivers/unzip \
archivers/zip=2.3p0
RUN_DEPENDS-c++ = ${RUN_DEPENDS}
libstdc++-=4.2,4.3|libstdc++-=4.2v0,4.3v0:${BUILD_PKGPATH},-estdc
+RUN_DEPENDS-ada = ${FULLPKGNAME-main}:${BUILD_PKGPATH}
WANTLIB-estdc =m
WANTLIB-java = c m pthread z
@@ -216,19 +218,21 @@ LIB_DEPENDS-f95 = devel/gmp=4.2 \
WANTLIB-f95 = ${WANTLIB} gmp mpfr
.include bsd.port.mk
-ADA_PACKAGE = ${PACKAGE_REPOSITORY}/i386/all/${PKGNAME-ada}${PKG_SUFX}
-GCC_PACKAGE = ${PACKAGE_REPOSITORY}/i386/all/${PKGNAME-main}${PKG_SUFX}
+ADA_PACKAGE = ${PACKAGE_REPOSITORY}/${MACHINE_ARCH}/all/${FULLPKGNAME-ada}.tgz
+GCC_PACKAGE = ${PACKAGE_REPOSITORY}/${MACHINE_ARCH}/all/${FULLPKGNAME-main}.tgz
adastrap: ${ADA_PACKAGE} ${GCC_PACKAGE}
-.if !defined(LIBC_VERSION)
+.if !defined(ADASTRAP_LIBC-${MACHINE_ARCH})
@echo 12 Error: you must specify a libc to bundle
@exit 1
.else
mkdir -p ${WRKDIR}/prepare/bootstrap/system-libs
- cp /usr/lib/libc.so.${LIBC_VERSION}
${WRKDIR}/prepare/bootstrap/system-libs
+ cp /usr/lib/libc.so.${ADASTRAP_LIBC-${MACHINE_ARCH}} \
+ ${WRKDIR}/prepare/bootstrap/system-libs
cd ${WRKDIR}/prepare/bootstrap tar zxf ${ADA_PACKAGE}
cd ${WRKDIR}/prepare/bootstrap tar zxf ${GCC_PACKAGE}
rm -rf
Re: lang/gcc: add Ada support on amd64
On Thu, Sep 27, 2012 at 12:33:12AM +0200, Marc Espie wrote: On Wed, Sep 26, 2012 at 07:42:07PM +0200, Tobias Ulmer wrote: Add amd64 support to gnat 4.2 - 4.7. general: * renamed LIBC_VERSION to ADASTRAP_LIBC-arch which IMO is clearer * dropped BOOTSTRAP_GEN since there is only a single use * unified ADASTRAP variable block across all gcc4 ports * made RUN_DEPENDS for gnat very strict, otherwise installing gnat does not update gcc, resulting in an error. (This may be a general problem with these ports, someone should check if the global RUN_DEPENDS still works as intended) * Repaired the adastrap target. gcc-4.6 and gcc-4.7: * Removed the libc symlink hack. This had no effect, since the binaries are linked against the supplied libc, not version 61.0. Probably an artifact from bootstrapping once. gcc-4.7: * Adapt gnat configuration done for i386 to amd64 as well. While there, change -pthread to -lpthread for both architectures. The required bootstrap files are temporarily at: http://www.tmux.org/~tobiasu/tmp/adastrap-amd64-4.2.4-1.tgz http://www.tmux.org/~tobiasu/tmp/adastrap-amd64-4.6.3-0.tar.xz http://www.tmux.org/~tobiasu/tmp/adastrap-amd64-4.7.1-0.tar.xz You can also find the diffs for each gcc port in the same directory. All adastrap tarballs are at least five generations away from the initial bootstrap compiler. To build your own, run: while ! heat_death_of_the_universe; do make clean=all make package make adastrap make makesum done Cool. I'll be sure to have a look in the not so distant future. bootstrapping ada tends to be a pain. Question: would you have the patience to start doing that on other targets ? It would be nice if ada were supported on our major platforms, and not just i386/amd64... sparc64 is building native as we speak. Fingers crossed...
Re: update: graphics/freeglut
On Sun, Sep 09, 2012 at 09:39:12AM +0200, Landry Breuil wrote:
On Sun, Sep 09, 2012 at 02:02:08AM +0200, Tobias Ulmer wrote:
Ping. Anyone brave enough to put this patch and the fixes into a bulk
build?
Send a full diff and i'll put it into the next.
Landry
Index: graphics/freeglut/Makefile
===
RCS file: /home/vcs/cvs/openbsd/ports/graphics/freeglut/Makefile,v
retrieving revision 1.6
diff -u -p -r1.6 Makefile
--- graphics/freeglut/Makefile 1 May 2012 16:10:07 - 1.6
+++ graphics/freeglut/Makefile 9 Sep 2012 10:00:14 -
@@ -1,9 +1,8 @@
# $OpenBSD: Makefile,v 1.6 2012/05/01 16:10:07 pascal Exp $
COMMENT = open source alternative to glut
-DISTNAME = freeglut-2.6.0
-REVISION = 1
-SHARED_LIBS = glut 4.1
+DISTNAME = freeglut-2.8.0
+SHARED_LIBS = glut 5.0
CATEGORIES = graphics
HOMEPAGE = http://freeglut.sourceforge.net/
MAINTAINER = Tobias Ulmer tobi...@tmux.org
@@ -14,7 +13,8 @@ PERMIT_PACKAGE_FTP = Yes
PERMIT_DISTFILES_CDROM =Yes
PERMIT_DISTFILES_FTP = Yes
-WANTLIB = GL X11 Xext Xi Xxf86vm m xcb Xdamage Xfixes drm stdc++
+WANTLIB = GL X11 Xext Xxf86vm m xcb Xdamage Xfixes drm stdc++
+WANTLIB += usbhid
MASTER_SITES = ${MASTER_SITE_SOURCEFORGE:=freeglut/}
Index: graphics/freeglut/distinfo
===
RCS file: /home/vcs/cvs/openbsd/ports/graphics/freeglut/distinfo,v
retrieving revision 1.2
diff -u -p -r1.2 distinfo
--- graphics/freeglut/distinfo 21 Mar 2010 12:45:21 - 1.2
+++ graphics/freeglut/distinfo 9 Sep 2012 10:00:14 -
@@ -1,5 +1,2 @@
-MD5 (freeglut-2.6.0.tar.gz) = OfDy3onzmVKdK5gRiAgiGA==
-RMD160 (freeglut-2.6.0.tar.gz) = Q72xptmURe4JNwkw7GvA+DEOf88=
-SHA1 (freeglut-2.6.0.tar.gz) = aDBsRIbBPQBaTk1UA14MCxvcIgs=
-SHA256 (freeglut-2.6.0.tar.gz) = YVtB7clsQF7mu36aKUJkIUDD5Xwd2MaOlooX3ftZ6Y8=
-SIZE (freeglut-2.6.0.tar.gz) = 1124186
+SHA256 (freeglut-2.8.0.tar.gz) = GXKT/4hqvmE7yetKdi2RYbDJ5ks+jmE+18XjU5dPugU=
+SIZE (freeglut-2.8.0.tar.gz) = 979459
Index: graphics/freeglut/patches/patch-Makefile_in
===
RCS file: graphics/freeglut/patches/patch-Makefile_in
diff -N graphics/freeglut/patches/patch-Makefile_in
--- /dev/null 1 Jan 1970 00:00:00 -
+++ graphics/freeglut/patches/patch-Makefile_in 9 Sep 2012 10:00:14 -
@@ -0,0 +1,12 @@
+$OpenBSD$
+--- Makefile.in.orig Thu Aug 30 08:08:49 2012
Makefile.inThu Aug 30 08:09:12 2012
+@@ -230,7 +230,7 @@ target_alias = @target_alias@
+ top_build_prefix = @top_build_prefix@
+ top_builddir = @top_builddir@
+ top_srcdir = @top_srcdir@
+-SUBDIRS = src include progs doc
++SUBDIRS = src include #progs doc
+ AUTOMAKE_OPTIONS = gnu
+ EXTRA_DIST = \
+ FrequentlyAskedQuestions \
Index: graphics/freeglut/patches/patch-include_GL_freeglut_ext_h
===
RCS file:
/home/vcs/cvs/openbsd/ports/graphics/freeglut/patches/patch-include_GL_freeglut_ext_h,v
retrieving revision 1.1
diff -u -p -r1.1 patch-include_GL_freeglut_ext_h
--- graphics/freeglut/patches/patch-include_GL_freeglut_ext_h 1 May 2012
16:10:07 - 1.1
+++ graphics/freeglut/patches/patch-include_GL_freeglut_ext_h 9 Sep 2012
10:00:14 -
@@ -1,7 +1,7 @@
$OpenBSD: patch-include_GL_freeglut_ext_h,v 1.1 2012/05/01 16:10:07 pascal Exp
$
include/GL/freeglut_ext.h.orig Mon Apr 30 20:39:58 2012
-+++ include/GL/freeglut_ext.h Mon Apr 30 20:40:07 2012
-@@ -163,7 +163,7 @@ FGAPI voidFGAPIENTRY glutSolidCylinder( GLdouble r
+--- include/GL/freeglut_ext.h.orig Sun Sep 4 21:38:32 2011
include/GL/freeglut_ext.h Thu Aug 30 07:10:04 2012
+@@ -170,7 +170,7 @@ FGAPI voidFGAPIENTRY glutSolidCylinder( GLdouble r
/*
* Extension functions, see freeglut_ext.c
*/
Index: graphics/hs-GLUT/Makefile
===
RCS file: /home/vcs/cvs/openbsd/ports/graphics/hs-GLUT/Makefile,v
retrieving revision 1.13
diff -u -p -r1.13 Makefile
--- graphics/hs-GLUT/Makefile 2 Sep 2012 21:21:05 - 1.13
+++ graphics/hs-GLUT/Makefile 9 Sep 2012 10:00:15 -
@@ -3,7 +3,7 @@
COMMENT = Haskell binding for the OpenGL Utility Toolkit
DISTNAME = GLUT-2.1.2.1
-REVISION = 9
+REVISION = 10
CATEGORIES = graphics devel
HOMEPAGE = http://www.haskell.org/HOpenGL/
@@ -28,6 +28,6 @@ BUILD_DEPENDS = ${RUN_DEPENDS}
MODGHC_SETUP_CONF_ENV =CPPFLAGS=-I${LOCALBASE}/include \
- LDFLAGS=-L${LOCALBASE}/lib
+ LDFLAGS=-L${LOCALBASE}/lib -L${X11BASE}/lib -lXrandr
.include bsd.port.mk
Index: graphics/hugin/Makefile
Re: update: graphics/freeglut
Ping. Anyone brave enough to put this patch and the fixes into a bulk build?
fix: games/glsfcave (freeglut update)
Fix fallout from freeglut update Index: Makefile === RCS file: /home/vcs/cvs/openbsd/ports/games/glsfcave/Makefile,v retrieving revision 1.6 diff -u -p -r1.6 Makefile --- Makefile16 Nov 2010 11:23:27 - 1.6 +++ Makefile31 Aug 2012 23:50:07 - @@ -5,7 +5,7 @@ COMMENT=control a ribbon to avoid hitti V= 1.0 DISTNAME= glsfcave-$V-mitl PKGNAME= glsfcave-$V -REVISION = 2 +REVISION = 3 CATEGORIES=games EXTRACT_SUFX= .tar.bz2 @@ -20,7 +20,7 @@ PERMIT_DISTFILES_FTP= Yes PERMIT_PACKAGE_CDROM= Yes PERMIT_PACKAGE_FTP=Yes -WANTLIB= GL GLU X11 Xext Xi Xmu c m pthread-stubs xcb glut=4 +WANTLIB= GL GLU X11 Xext Xi Xmu c m pthread-stubs xcb glut=4 Xrandr LIB_DEPENDS= graphics/freeglut Index: patches/patch-Makefile === RCS file: /home/vcs/cvs/openbsd/ports/games/glsfcave/patches/patch-Makefile,v retrieving revision 1.1.1.1 diff -u -p -r1.1.1.1 patch-Makefile --- patches/patch-Makefile 19 Mar 2008 20:22:09 - 1.1.1.1 +++ patches/patch-Makefile 31 Aug 2012 23:50:07 - @@ -9,7 +9,7 @@ $OpenBSD: patch-Makefile,v 1.1.1.1 2008/ #LDFLAGS=-lglut -lGLU -lGL -lXmu -lXi -lXext -lX11 -lgen ## FreeBSD/Linux -#LDFLAGS=-L/usr/X11R6/lib -lglut -lGLU -lGL -lXmu -lXi -lXext -lX11 -+LDFLAGS+=-lglut -lGLU -lGL -lXmu -lXi -lXext -lX11 -lm ++LDFLAGS+=-lglut -lGLU -lGL -lXmu -lXi -lXrandr -lXext -lX11 -lm glsfcave: main.o $(CC) $(CFLAGS) -o glsfcave main.o $(LDFLAGS)
fix: graphics/hs-GLUT (freeglut update)
Fix fallout from freeglut update
Index: Makefile
===
RCS file: /home/vcs/cvs/openbsd/ports/graphics/hs-GLUT/Makefile,v
retrieving revision 1.12
diff -u -p -r1.12 Makefile
--- Makefile16 Sep 2011 10:02:37 - 1.12
+++ Makefile31 Aug 2012 23:29:13 -
@@ -3,7 +3,7 @@
COMMENT = Haskell binding for the OpenGL Utility Toolkit
DISTNAME = GLUT-2.1.2.1
-REVISION = 8
+REVISION = 9
CATEGORIES = graphics devel
HOMEPAGE = http://www.haskell.org/HOpenGL/
@@ -27,6 +27,6 @@ BUILD_DEPENDS = ${RUN_DEPENDS}
MODGHC_SETUP_CONF_ENV =CPPFLAGS=-I${LOCALBASE}/include \
- LDFLAGS=-L${LOCALBASE}/lib
+ LDFLAGS=-L${LOCALBASE}/lib -L${X11BASE}/lib -lXrandr
.include bsd.port.mk
fix: lang/parrot (freeglut update)
Fix fallout from freeglut update
Index: Makefile
===
RCS file: /home/vcs/cvs/openbsd/ports/lang/parrot/Makefile,v
retrieving revision 1.23
diff -u -p -r1.23 Makefile
--- Makefile13 Jul 2012 09:46:05 - 1.23
+++ Makefile31 Aug 2012 23:26:06 -
@@ -6,6 +6,7 @@ COMMENT=virtual machine designed for in
V= 4.5.0
DISTNAME= parrot-$V
+REVISION= 0
CATEGORIES=lang perl6
SHARED_LIBS= parrot 6.0
@@ -22,7 +23,7 @@ PERMIT_DISTFILES_CDROM= Yes
PERMIT_DISTFILES_FTP= Yes
WANTLIB += GL GLU c glut gmp icudata icuuc m ncurses pthread
-WANTLIB += readline stdc++ util z ffi
+WANTLIB += readline stdc++ util z ffi Xi Xrandr
BUILD_DEPENDS= graphics/glew \
devel/pcre
Index: patches/patch-config_auto_opengl_pm
===
RCS file: patches/patch-config_auto_opengl_pm
diff -N patches/patch-config_auto_opengl_pm
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-config_auto_opengl_pm 31 Aug 2012 23:26:06 -
@@ -0,0 +1,12 @@
+$OpenBSD$
+--- config/auto/opengl.pm.orig Fri Aug 31 15:43:50 2012
config/auto/opengl.pm Fri Aug 31 15:44:11 2012
+@@ -177,7 +177,7 @@ sub runstep {
+ win32_gcc = '-lglut32 -lglu32 -lopengl32',
+ win32_nongcc= 'opengl32.lib glu32.lib glut32.lib',
+ darwin = '-framework OpenGL -framework GLUT',
+-default = '-lglut -lGLU -lGL',
++default = '-lglut -lGLU -lGL -lXi -lXrandr',
+ } );
+
+ $conf-cc_gen('config/auto/opengl/opengl_c.in');
fix: games/supertuxkart (freeglut update)
Fix fallout from freeglut update
Index: Makefile
===
RCS file: /home/vcs/cvs/openbsd/ports/games/supertuxkart/Makefile,v
retrieving revision 1.7
diff -u -p -r1.7 Makefile
--- Makefile8 May 2011 20:14:54 - 1.7
+++ Makefile31 Aug 2012 23:46:24 -
@@ -5,6 +5,7 @@ COMMENT = kart racing game with OSS mas
BASENAME = supertuxkart-0.7
DISTNAME = ${BASENAME}-src
PKGNAME = ${BASENAME}
+REVISION = 0
CATEGORIES = games
@@ -19,7 +20,7 @@ PERMIT_DISTFILES_CDROM =Yes
PERMIT_DISTFILES_FTP = Yes
WANTLIB = GL GLU Irrlicht SDL X11 Xi Xmu c fribidi m ogg openal \
- pthread stdc++ vorbis vorbisfile glut
+ pthread stdc++ vorbis vorbisfile glut Xrandr
MASTER_SITES = ${MASTER_SITE_SOURCEFORGE:=supertuxkart/SuperTuxKart/}
EXTRACT_SUFX = .tar.bz2
@@ -43,7 +44,8 @@ CONFIGURE_ARGS = --with-libintl-prefix=$
CONFIGURE_ENV =CPPFLAGS=-I${LOCALBASE}/include
-I${X11BASE}/include \
LDFLAGS=-L${LOCALBASE}/lib -L${X11BASE}/lib -liconv \
-lGL -lGLU -lIrrlicht -lSDL -lXi -lXmu -lX11 \
- -lglut -lm -logg -pthread -lvorbis -lvorbisfile
+ -lXrandr -lglut -lm -logg -pthread -lvorbis \
+ -lvorbisfile
WRKDIST = ${WRKDIR}/${BASENAME}
update: graphics/freeglut
Update for freeglut, triggered by a rough diff from Azwaw OUSADOU.
Some symbols have been added, but there is also a size change - Major
version bump
Instead of fixing the demo programs and documentation, disable both
since they are not considered for installation anyways.
Tested with crack-attack on amd64, compile tested on i386 and sparc64.
Index: Makefile
===
RCS file: /home/vcs/cvs/openbsd/ports/graphics/freeglut/Makefile,v
retrieving revision 1.6
diff -u -p -r1.6 Makefile
--- Makefile1 May 2012 16:10:07 - 1.6
+++ Makefile30 Aug 2012 12:42:43 -
@@ -1,9 +1,8 @@
# $OpenBSD: Makefile,v 1.6 2012/05/01 16:10:07 pascal Exp $
COMMENT = open source alternative to glut
-DISTNAME = freeglut-2.6.0
-REVISION = 1
-SHARED_LIBS = glut 4.1
+DISTNAME = freeglut-2.8.0
+SHARED_LIBS = glut 5.0
CATEGORIES = graphics
HOMEPAGE = http://freeglut.sourceforge.net/
MAINTAINER = Tobias Ulmer tobi...@tmux.org
@@ -14,7 +13,8 @@ PERMIT_PACKAGE_FTP = Yes
PERMIT_DISTFILES_CDROM =Yes
PERMIT_DISTFILES_FTP = Yes
-WANTLIB = GL X11 Xext Xi Xxf86vm m xcb Xdamage Xfixes drm stdc++
+WANTLIB = GL X11 Xext Xxf86vm m xcb Xdamage Xfixes drm stdc++
+WANTLIB += usbhid
MASTER_SITES = ${MASTER_SITE_SOURCEFORGE:=freeglut/}
Index: distinfo
===
RCS file: /home/vcs/cvs/openbsd/ports/graphics/freeglut/distinfo,v
retrieving revision 1.2
diff -u -p -r1.2 distinfo
--- distinfo21 Mar 2010 12:45:21 - 1.2
+++ distinfo30 Aug 2012 12:42:43 -
@@ -1,5 +1,2 @@
-MD5 (freeglut-2.6.0.tar.gz) = OfDy3onzmVKdK5gRiAgiGA==
-RMD160 (freeglut-2.6.0.tar.gz) = Q72xptmURe4JNwkw7GvA+DEOf88=
-SHA1 (freeglut-2.6.0.tar.gz) = aDBsRIbBPQBaTk1UA14MCxvcIgs=
-SHA256 (freeglut-2.6.0.tar.gz) = YVtB7clsQF7mu36aKUJkIUDD5Xwd2MaOlooX3ftZ6Y8=
-SIZE (freeglut-2.6.0.tar.gz) = 1124186
+SHA256 (freeglut-2.8.0.tar.gz) = GXKT/4hqvmE7yetKdi2RYbDJ5ks+jmE+18XjU5dPugU=
+SIZE (freeglut-2.8.0.tar.gz) = 979459
Index: patches/patch-Makefile_in
===
RCS file: patches/patch-Makefile_in
diff -N patches/patch-Makefile_in
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-Makefile_in 30 Aug 2012 12:42:43 -
@@ -0,0 +1,12 @@
+$OpenBSD$
+--- Makefile.in.orig Thu Aug 30 08:08:49 2012
Makefile.inThu Aug 30 08:09:12 2012
+@@ -230,7 +230,7 @@ target_alias = @target_alias@
+ top_build_prefix = @top_build_prefix@
+ top_builddir = @top_builddir@
+ top_srcdir = @top_srcdir@
+-SUBDIRS = src include progs doc
++SUBDIRS = src include #progs doc
+ AUTOMAKE_OPTIONS = gnu
+ EXTRA_DIST = \
+ FrequentlyAskedQuestions \
Index: patches/patch-include_GL_freeglut_ext_h
===
RCS file:
/home/vcs/cvs/openbsd/ports/graphics/freeglut/patches/patch-include_GL_freeglut_ext_h,v
retrieving revision 1.1
diff -u -p -r1.1 patch-include_GL_freeglut_ext_h
--- patches/patch-include_GL_freeglut_ext_h 1 May 2012 16:10:07 -
1.1
+++ patches/patch-include_GL_freeglut_ext_h 30 Aug 2012 12:42:43 -
@@ -1,7 +1,7 @@
$OpenBSD: patch-include_GL_freeglut_ext_h,v 1.1 2012/05/01 16:10:07 pascal Exp
$
include/GL/freeglut_ext.h.orig Mon Apr 30 20:39:58 2012
-+++ include/GL/freeglut_ext.h Mon Apr 30 20:40:07 2012
-@@ -163,7 +163,7 @@ FGAPI voidFGAPIENTRY glutSolidCylinder( GLdouble r
+--- include/GL/freeglut_ext.h.orig Sun Sep 4 21:38:32 2011
include/GL/freeglut_ext.h Thu Aug 30 07:10:04 2012
+@@ -170,7 +170,7 @@ FGAPI voidFGAPIENTRY glutSolidCylinder( GLdouble r
/*
* Extension functions, see freeglut_ext.c
*/
fix: graphics/hugin (due to freeglut update)
hugin is missing Xrandr after my earlier freeglut update.
This should fix it:
Index: Makefile
===
RCS file: /home/vcs/cvs/openbsd/ports/graphics/hugin/Makefile,v
retrieving revision 1.27
diff -u -p -r1.27 Makefile
--- Makefile29 Mar 2012 13:38:31 - 1.27
+++ Makefile30 Aug 2012 19:13:58 -
@@ -6,7 +6,7 @@ SHARED_ONLY = Yes
COMMENT = panorama image stitcher
DISTNAME = hugin-2011.0.0
-REVISION= 0
+REVISION= 1
EXTRACT_SUFX = .tar.bz2
# not needed for releases, but makes RCs/betas easier to work with
Index: patches/patch-CMakeModules_FindGLUT_cmake
===
RCS file: patches/patch-CMakeModules_FindGLUT_cmake
diff -N patches/patch-CMakeModules_FindGLUT_cmake
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-CMakeModules_FindGLUT_cmake 30 Aug 2012 19:13:58 -
@@ -0,0 +1,28 @@
+$OpenBSD$
+--- CMakeModules/FindGLUT.cmake.orig Thu Aug 30 19:53:26 2012
CMakeModules/FindGLUT.cmakeThu Aug 30 20:10:13 2012
+@@ -54,6 +54,10 @@ ELSE (WIN32)
+ FIND_LIBRARY( GLUT_Xi_LIBRARY Xi
+ /usr/openwin/lib
+ )
++
++FIND_LIBRARY( GLUT_Xrandr_LIBRARY Xrandr
++ /usr/openwin/lib
++ )
+
+ FIND_LIBRARY( GLUT_Xmu_LIBRARY Xmu
+ /usr/openwin/lib
+@@ -72,6 +76,7 @@ IF(GLUT_INCLUDE_DIR)
+ ${GLUT_glut_LIBRARY}
+ ${GLUT_Xmu_LIBRARY}
+ ${GLUT_Xi_LIBRARY}
++ ${GLUT_Xrandr_LIBRARY}
+ ${GLUT_cocoa_LIBRARY}
+ )
+ SET( GLUT_FOUND YES )
+@@ -94,4 +99,5 @@ MARK_AS_ADVANCED(
+ GLUT_glut_LIBRARY
+ GLUT_Xmu_LIBRARY
+ GLUT_Xi_LIBRARY
++ GLUT_Xrandr_LIBRARY
+ )
Re: Update: net/csup
On Sat, May 19, 2012 at 05:14:05PM +, Christian Weisgerber wrote: No real changes. I'm late to the party, but wanted to say that it segfaults even quicker than before. I've tried to debug this in the past, but it somehow stops completely in the weeds, gdb is of no real help. I did play around with gcc instrumentation in order to figure out which function crashed where and what the various threads are doing, and found something that looked like a use after free in the attribute(?) code. I also remember compiling it without -DNDEBUG, which promptly made it trigger an assert(). Anyway, how do you use it? This segfault is so obvious that I suspect I do something wrong here... I'm not saying this should be backed out, btw. The old version was just as broken for me. $ uname -a OpenBSD radon.tmux.org 5.1 GENERIC.MP#294 amd64 $ cat csup.conf *default host=cvsup.de.FreeBSD.org *default prefix=/home/vcs/cvs/freebsd *default base=/home/vcs/cvs/freebsd *default release=cvs delete use-rel-suffix compress cvs-all $ csup csup.conf Connected to 82.98.105.214 Updating collection cvs-all/cvs Append to CVSROOT/checkoutlist Append to CVSROOT/commitinfo Append to CVSROOT/config Append to CVSROOT/cvswrappers Append to CVSROOT/editinfo Append to CVSROOT/loginfo Append to CVSROOT/loglinks.sh Append to CVSROOT/mergemodules.sh Append to CVSROOT/modules Edit CVSROOT/modules,v Segmentation fault (core dumped) $ gdb `which csup` csup.core GNU gdb 6.3 Copyright 2004 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type show copying to see the conditions. There is absolutely no warranty for GDB. Type show warranty for details. This GDB was configured as amd64-unknown-openbsd5.1...(no debugging symbols found) Core was generated by `csup'. Program terminated with signal 11, Segmentation fault. Reading symbols from /usr/lib/libpthread.so.15.2...done. Loaded symbols for /usr/lib/libpthread.so.15.2 Reading symbols from /usr/lib/libz.so.4.1...done. Loaded symbols for /usr/lib/libz.so.4.1 Reading symbols from /usr/lib/libcrypto.so.20.1...done. Loaded symbols for /usr/lib/libcrypto.so.20.1 Symbols already loaded for /usr/lib/libpthread.so.15.2 Reading symbols from /usr/lib/libc.so.64.1...done. Loaded symbols for /usr/lib/libc.so.64.1 Reading symbols from /usr/libexec/ld.so...done. Loaded symbols for /usr/libexec/ld.so #0 0x004069f6 in ?? () (gdb) bt #0 0x004069f6 in ?? () #1 0x0040bd8c in fnmatch () #2 0x00416574 in fnmatch () #3 0x000200af068e in _rthread_start (v=Variable v is not available. ) at /usr/src/lib/librthread/rthread.c:113 #4 0x00020062d07e in __tfork_thread () from /usr/lib/libc.so.64.1 #5 0x7f997fc07fbf7fbe in ?? () #6 0x7e4a7e497e487fdd in ?? () #7 0x7e4e7e4d7e4c7fa3 in ?? () #8 0x7fe87e517e507e4f in ?? () #9 0x7fec7feb7fea7fe9 in ?? () #10 0x7ff07fef7fee7fed in ?? () #11 0x7ff47ff37ff27ff1 in ?? () #12 0x7ff87ff77ff67ff5 in ?? () #13 0x7ffc7ffb7ffa7ff9 in ?? () #14 0x80007fff7ffe7ffd in ?? () #15 0x7e6e7e6d80028001 in ?? () #16 0x7fe57fe47e707e6f in ?? () #17 0x80047a267e747e73 in ?? () #18 0x7a2b7a2a7a297a28 in ?? () #19 0x7a2f7a2e7a2d7a2c in ?? () #20 0x7f3c7f3b7f3a7f39 in ?? () #21 0x1b2201be7198 in ?? () #22 0x8010800f7f7d in ?? () #23 0x8014801380128011 in ?? () #24 0x8018801780168015 in ?? () #25 0x801c801b801a8019 in ?? () #26 0x8020801f801e801d in ?? () #27 0x8024802380228021 in ?? () #28 0x8028802780268025 in ?? () #29 0x7f9b803380038029 in ?? () #30 0x803e803d803c7f38 in ?? () #31 0x7a3c7f3d803f in ?? () #32 0x7fda7fd97fd8 in ?? () #33 0x7fe01e46 in ?? () #34 0x7fa267a97fe1 in ?? () #35 0x69fd8008 in ?? () ---Type return to continue, or q return to quit---q Quit (gdb) q $
new: font/gohufont
Attached is my current favourite terminal font. It's pretty similar to Terminus, but offers a 11px version that includes bold glyphs. Give it a try :) cat pkg/DESCR Gohufont is a monospace bitmap font well suited for programming and terminal use. It is intended to be very legible and offers very discernable glyphs for all characters, including signs and symbols. gohufont.tar.gz Description: application/tar-gz
Re: new: font/gohufont
On Sun, May 13, 2012 at 10:15:19PM +0200, Juan Francisco Cantero Hurtado wrote: On Sun, May 13, 2012 at 05:34:25PM +0200, Tobias Ulmer wrote: Attached is my current favourite terminal font. It's pretty similar to Terminus, but offers a 11px version that includes bold glyphs. Give it a try :) cat pkg/DESCR Gohufont is a monospace bitmap font well suited for programming and terminal use. It is intended to be very legible and offers very discernable glyphs for all characters, including signs and symbols. Hi. The homepage shows two versions: 11px and 14px. I'm testing the font on terminator but I only can use the 11px version. When I select 14px, the size of the font doesn't change. Hi, I don't know about terminator, but both fonts are included. The website buttons only change the screenshots, but the tarball has both versions. There may be some confusion here because pt and px are different things and freetype probably adds its own magic somewhere. You can test with xfontsel, select pxlsz to verify both sizes are available. Tobias -- Juan Francisco Cantero Hurtado http://juanfra.info
DEBUG does not disable stripping in some ports
Setting DEBUG not only sets additional CFLAGS, it also disables
stripping during the 'fake' stage for ports that use BSD make as build
system.
Curiously this does not work for ports that have a custom 'do-install'
target (or friends) and use INSTALL_PROGRAM.
After pulling my hair a bit, it became obvious that _SUDOMAKESYS
filters out most environment variables, including DEBUG (usage begins
in bsd.port.mk:2527, implementation is in pkgpath.mk:115).
I can see two options to fix this: Either extend the default sudoers to
pass DEBUG/INSTALL_STRIP(?). Or pass MAKE_ENV like the standard install
target does. It seems to do the right thing in my light testing. I've
also modified _SUDOMAKE for symmetry.
devel/csup is a port that ignores DEBUG partially.
x11/spectrwm as an example where it works.
Comments?
Index: pkgpath.mk
===
RCS file: /home/vcs/cvs/openbsd/ports/infrastructure/mk/pkgpath.mk,v
retrieving revision 1.44
diff -u -p -r1.44 pkgpath.mk
--- pkgpath.mk 17 Feb 2012 07:40:35 - 1.44
+++ pkgpath.mk 5 May 2012 15:31:22 -
@@ -110,9 +110,9 @@ _cache_fragment = \
HTMLIFY = sed -e 's//\amp;/g' -e 's//\gt;/g' -e 's//\lt;/g'
_MAKE = cd ${.CURDIR} PKGPATH=${PKGPATH} exec ${MAKE}
-_SUDOMAKE = cd ${.CURDIR} PKGPATH=${PKGPATH} exec ${SUDO} ${MAKE}
+_SUDOMAKE = cd ${.CURDIR} exec ${SUDO} ${SETENV} ${MAKE_ENV}
PKGPATH=${PKGPATH} ${MAKE}
_MAKESYS = cd ${.CURDIR} PKGPATH=${PKGPATH} exec ${_SYSTRACE_CMD} ${MAKE}
-_SUDOMAKESYS = cd ${.CURDIR} PKGPATH=${PKGPATH} exec ${SUDO}
${_SYSTRACE_CMD} ${MAKE}
+_SUDOMAKESYS = cd ${.CURDIR} exec ${SUDO} ${SETENV} ${MAKE_ENV}
PKGPATH=${PKGPATH} ${_SYSTRACE_CMD} ${MAKE}
REPORT_PROBLEM_LOGFILE ?=
.if !empty(REPORT_PROBLEM_LOGFILE)
Re: Updates for musicpd ports (libmpdclient, mpc, mpd, ncmpc)
On Fri, Mar 02, 2012 at 01:31:23PM +0100, David Coppa wrote: The following diff updates: libmpdclient 2.6 - 2.7 mpc 0.21 - 0.22 mpd 0.16.5 - 0.16.7 ncmpc 0.19 - 0.20 (I've also fixed some crap into its configure) Looks good to me. Tested on sparc64 with rthreads. Thanks!
Re: NEW: Nagelfar
On Mon, Jan 30, 2012 at 05:33:03PM -0500, Stuart Cassoff wrote: On 11/29/11 10:53, Stuart Cassoff wrote: Now at version 1.12. Stu Hmmm? Any go on this? http://marc.info/?l=openbsd-portsm=132762660906987w=2
bugfix: x11/rxvt-unicode
Hi guys, rxvt-unicode picks up some startup notification library in the latest amd64 snaps. Not nice. Here's a patch, please test it :) Index: Makefile === RCS file: /home/vcs/cvs/openbsd/ports/x11/rxvt-unicode/Makefile,v retrieving revision 1.16 diff -u -p -r1.16 Makefile --- Makefile22 Dec 2011 11:29:33 - 1.16 +++ Makefile4 Jan 2012 13:27:55 - @@ -34,7 +34,8 @@ CONFIGURE_ARGS = --enable-256-color \ --enable-perl \ --enable-pixbuf \ --with-term=rxvt-256color \ - --disable-afterimage + --disable-afterimage \ + --disable-startup-notification # missing locale support CONFIGURE_ARGS += --disable-xim
Re: mpd-0.16.5
On Thu, Oct 13, 2011 at 05:00:56PM +0200, David Coppa wrote: Straightforward update to mpd-0.16.5. Finally got around testing this, on sparc64 no less (see bugs@). Thanks for the work David, OK by me if you fix the configure patch comment. Cheers, David Index: patches/patch-configure === RCS file: /cvs/ports/audio/mpd/patches/patch-configure,v retrieving revision 1.13 diff -u -p -r1.13 patch-configure --- patches/patch-configure 5 Apr 2011 19:55:38 - 1.13 +++ patches/patch-configure 13 Oct 2011 14:53:47 - @@ -1,19 +1,16 @@ -$OpenBSD: patch-configure,v 1.13 2011/04/05 19:55:38 landry Exp $ - -Don't reset CPPFLAGS...stupid configure. - Might want to keep Jaspers rage comment there ;)
update: audio/libmpdclient
Latest version of libmpdclient. Tested on sparc64. API additions only.
Index: Makefile
===
RCS file: /home/vcs/cvs/openbsd/ports/audio/libmpdclient/Makefile,v
retrieving revision 1.5
diff -u -p -r1.5 Makefile
--- Makefile5 Apr 2011 20:09:32 - 1.5
+++ Makefile29 Oct 2011 23:03:09 -
@@ -2,8 +2,8 @@
COMMENT= library for interfacing MPD
-DISTNAME= libmpdclient-2.4
-SHARED_LIBS= mpdclient 1.0 # 2.1
+DISTNAME= libmpdclient-2.6
+SHARED_LIBS= mpdclient 1.1 # 2.5
CATEGORIES=audio
HOMEPAGE= http://mpd.wikia.com/wiki/ClientLib:libmpdclient
Index: distinfo
===
RCS file: /home/vcs/cvs/openbsd/ports/audio/libmpdclient/distinfo,v
retrieving revision 1.2
diff -u -p -r1.2 distinfo
--- distinfo5 Apr 2011 20:09:32 - 1.2
+++ distinfo29 Oct 2011 23:03:09 -
@@ -1,5 +1,5 @@
-MD5 (libmpdclient-2.4.tar.gz) = 76aWTET6fwFIGh+DZcH2yw==
-RMD160 (libmpdclient-2.4.tar.gz) = hB11lNdmzrCSrPBjsJ3AHZWb+pU=
-SHA1 (libmpdclient-2.4.tar.gz) = dMrq82AkFdz2JxOLNaCc5HOni74=
-SHA256 (libmpdclient-2.4.tar.gz) = 4PdKo0GhjoqpKCYDTOgAqcQH5TIorOBKPE1V55pWmQE=
-SIZE (libmpdclient-2.4.tar.gz) = 334720
+MD5 (libmpdclient-2.6.tar.gz) = qzh/Uy/QvZlTD3fw9b/PFA==
+RMD160 (libmpdclient-2.6.tar.gz) = lCmE6y8eH21Z/+3W8AxRWSbgFFo=
+SHA1 (libmpdclient-2.6.tar.gz) = Ob3PU5n7mONLi2EQBXtk9NGUUzM=
+SHA256 (libmpdclient-2.6.tar.gz) = evpUdow1fbuWyIMlmd4+8BeXTXwDlJnC3oucWHUj4Go=
+SIZE (libmpdclient-2.6.tar.gz) = 357982
Index: patches/patch-src_idle_c
===
RCS file:
/home/vcs/cvs/openbsd/ports/audio/libmpdclient/patches/patch-src_idle_c,v
retrieving revision 1.1
diff -u -p -r1.1 patch-src_idle_c
--- patches/patch-src_idle_c6 Sep 2010 11:40:39 - 1.1
+++ patches/patch-src_idle_c29 Oct 2011 23:03:09 -
@@ -2,9 +2,9 @@ $OpenBSD: patch-src_idle_c,v 1.1 2010/09
Fix build on gcc2.
src/idle.c.origSun Sep 5 22:39:01 2010
-+++ src/idle.c Sun Sep 5 22:40:02 2010
-@@ -57,7 +57,8 @@ static const char *const idle_names[] = {
+--- src/idle.c.origMon Oct 24 04:43:05 2011
src/idle.c Sun Oct 30 00:18:24 2011
+@@ -60,7 +60,8 @@ static const char *const idle_names[] = {
const char *
mpd_idle_name(enum mpd_idle idle)
{
@@ -14,7 +14,7 @@ Fix build on gcc2.
if (idle == (enum mpd_idle)(1 i))
return idle_names[i];
-@@ -67,9 +68,10 @@ mpd_idle_name(enum mpd_idle idle)
+@@ -70,9 +71,10 @@ mpd_idle_name(enum mpd_idle idle)
enum mpd_idle
mpd_idle_name_parse(const char *name)
{
@@ -26,7 +26,7 @@ Fix build on gcc2.
if (strcmp(name, idle_names[i]) == 0)
return 1 i;
-@@ -141,6 +143,7 @@ mpd_send_idle(struct mpd_connection *connection)
+@@ -144,6 +146,7 @@ mpd_send_idle(struct mpd_connection *connection)
bool
mpd_send_idle_mask(struct mpd_connection *connection, enum mpd_idle mask)
{
@@ -34,7 +34,7 @@ Fix build on gcc2.
/* this buffer is large enough even for the full mask */
char buffer[128] = idle;
-@@ -149,7 +152,7 @@ mpd_send_idle_mask(struct mpd_connection *connection,
+@@ -152,7 +155,7 @@ mpd_send_idle_mask(struct mpd_connection *connection,
if (mpd_error_is_defined(connection-error))
return false;
Index: pkg/PLIST
===
RCS file: /home/vcs/cvs/openbsd/ports/audio/libmpdclient/pkg/PLIST,v
retrieving revision 1.2
diff -u -p -r1.2 PLIST
--- pkg/PLIST 5 Apr 2011 20:09:32 - 1.2
+++ pkg/PLIST 29 Oct 2011 23:03:09 -
@@ -13,6 +13,7 @@ include/mpd/entity.h
include/mpd/error.h
include/mpd/idle.h
include/mpd/list.h
+include/mpd/message.h
include/mpd/mixer.h
include/mpd/output.h
include/mpd/pair.h
Re: [maintainer update] games/openttd 1.1.0 - 1.1.3
worked fine on amd64
Re: UPDATE: chicken
On Sat, Sep 24, 2011 at 06:00:21PM +0300, Timo Myyrä wrote: The update to chicken port is almost done. There seems to some issue with gmake though. It seems to print make database in addition to other tasks. Something seems to pass '-p' option to gmake but I haven't been able locate what and where. This is because you don't properly quote the variables you pass via MAKE_FLAGS Building chicken manually doesn't print the database. It only happens when using ports. I first thought that the patch to change '-lpthread' to '-pthread' would somehow be related but that doesn't seem to be the case. It's not necessary to patch Makefile.bsd, you can overwrite LIBRARIES with MAKE_FLAGS as well. Any help fixing the issue would be welcome. Port can be found at openbsd-wip git repository: https://github.com/jasperla/openbsd-wip/tree/master/lang/chicken Timo Two other things I remember from manually compiling: - echo #define C_STACK_GROWS_DOWNWARD 1 $@ I doubt this is true for all architectures. - # OpenBSD base still uses GCC 3.3.5 which does not support -fwrapv - ifeq ($(SYSTEM), OpenBSD) This is questionable today and causes chicken to omit -fwrapv, which may not be desirable.
update: devel/tig
Quick update to tig, survived my initial testing on amd64. The test-graph binary, which I guess is for internal testing, was removed from the tig install target. If anyone needs it, let me know. Index: Makefile === RCS file: /home/vcs/cvs/openbsd/ports/devel/tig/Makefile,v retrieving revision 1.7 diff -u -p -r1.7 Makefile --- Makefile16 Sep 2011 09:24:58 - 1.7 +++ Makefile24 Sep 2011 18:11:14 - @@ -1,8 +1,7 @@ # $OpenBSD: Makefile,v 1.7 2011/09/16 09:24:58 espie Exp $ COMMENT = ncurses GIT repository browser -DISTNAME = tig-0.17 -REVISION= 0 +DISTNAME = tig-0.18 CATEGORIES = devel HOMEPAGE = http://jonas.nitro.dk/tig/ MAINTAINER = Tobias Ulmer tobi...@tmux.org Index: distinfo === RCS file: /home/vcs/cvs/openbsd/ports/devel/tig/distinfo,v retrieving revision 1.4 diff -u -p -r1.4 distinfo --- distinfo13 Apr 2011 08:18:00 - 1.4 +++ distinfo24 Sep 2011 18:11:14 - @@ -1,5 +1,5 @@ -MD5 (tig-0.17.tar.gz) = 83M0MZlCLFlRh3bbRI3sDg== -RMD160 (tig-0.17.tar.gz) = MDwJIOfyuLi7dNZc8Ls+kUUB7NQ= -SHA1 (tig-0.17.tar.gz) = pI6dCNlvi0m2XPhmAhEsfI4aSAQ= -SHA256 (tig-0.17.tar.gz) = RNnJ1N7GiRHg9AJ1IYIGA5+EO8Oa93cYd35wCOlolPs= -SIZE (tig-0.17.tar.gz) = 168918 +MD5 (tig-0.18.tar.gz) = T6njPF2qdrb+0R4GhAU1bw== +RMD160 (tig-0.18.tar.gz) = pZwE03OrFNsNsOO34110tGv7O+4= +SHA1 (tig-0.18.tar.gz) = DdojTLbTQM/w51ETGxCzIaEirRo= +SHA256 (tig-0.18.tar.gz) = y3IgkrbUVKFBVpIDoVrhDCKq/QRaeEBiVNAY1+kc0jg= +SIZE (tig-0.18.tar.gz) = 173205 Index: pkg/PLIST === RCS file: /home/vcs/cvs/openbsd/ports/devel/tig/pkg/PLIST,v retrieving revision 1.2 diff -u -p -r1.2 PLIST --- pkg/PLIST 13 Apr 2011 08:18:00 - 1.2 +++ pkg/PLIST 24 Sep 2011 18:11:14 - @@ -1,5 +1,4 @@ @comment $OpenBSD: PLIST,v 1.2 2011/04/13 08:18:00 landry Exp $ -@bin bin/test-graph @bin bin/tig @man man/man1/tig.1 @man man/man5/tigrc.5
editors/emacs21 build problem
I'm running a bulk build (dpb) on sparc64 and emacs21 now hangs waiting for input: tail /usr/ports/logs/sparc64/packages/emacs-21.4p16.log Checking /home/ports/pobj/emacs-21.4/emacs-21.4/leim/quail/hanja-jis.el ... Checking /home/ports/pobj/emacs-21.4/emacs-21.4/leim/quail/hangul3.el ... Checking /home/ports/pobj/emacs-21.4/emacs-21.4/leim/quail/hangul.el ... Checking /home/ports/pobj/emacs-21.4/emacs-21.4/leim/quail/greek.el ... Checking /home/ports/pobj/emacs-21.4/emacs-21.4/leim/quail/ethiopic.el ... Checking /home/ports/pobj/emacs-21.4/emacs-21.4/leim/quail/devanagari.el ... Checking /home/ports/pobj/emacs-21.4/emacs-21.4/leim/quail/czech.el ... Checking /home/ports/pobj/emacs-21.4/emacs-21.4/leim/quail/cyrillic.el ... Checking /home/ports/pobj/emacs-21.4/emacs-21.4/leim/quail/cyril-jis.el ... Select coding system (default utf-8): Has anyone seen that before? I have LC_CTYPE=en_US.UTF-8 in my environment, that might make a difference.
Re: maintainer wanted
On Wed, Sep 07, 2011 at 09:32:09AM -0500, Amit Kulkarni wrote: Anyone interested in maintaining one or more of the following ports? archivers/unrar I will take unrar Can this do anything that p7zip-rar can not? net/dlint this appears to be dead upstream
fix: sysutils/xstatbar
Hi Antoine,
xstatbar hangs on startup eating cycles with cmpci(4). I'm not sure if
this is a fault of the driver or what is going on exactly.
Anyway, if devinfo.type != AUDIO_MIXER_CLASS is true, xstatbar will never
make progress and loop forever. Fix it by trying the next device.
This patch does not make the mixer appear on cmpci(4), but xstatbar works
fine otherwise.
Unrelated: ${HOMEPAGE} seems to be down, temporary?
Index: Makefile
===
RCS file: /home/vcs/cvs/openbsd/ports/sysutils/xstatbar/Makefile,v
retrieving revision 1.8
diff -u -p -r1.8 Makefile
--- Makefile18 Jan 2011 19:04:22 - 1.8
+++ Makefile1 Aug 2011 11:41:25 -
@@ -5,6 +5,7 @@ ONLY_FOR_ARCHS= ${APM_ARCHS}
COMMENT= simple system monitor bar
DISTNAME= xstatbar-0.5
+REVISION= 0
CATEGORIES=sysutils x11
Index: patches/patch-stats_c
===
RCS file: patches/patch-stats_c
diff -N patches/patch-stats_c
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-stats_c 1 Aug 2011 11:41:25 -
@@ -0,0 +1,15 @@
+$OpenBSD$
+--- stats.c.orig Mon Aug 1 09:48:58 2011
stats.cMon Aug 1 09:50:09 2011
+@@ -101,8 +101,10 @@ volume_init()
+devinfo.index = 0;
+while (ioctl(volume.dev_fd, AUDIO_MIXER_DEVINFO, devinfo) = 0) {
+
+- if (devinfo.type != AUDIO_MIXER_CLASS)
++ if (devinfo.type != AUDIO_MIXER_CLASS) {
++ devinfo.index++;
+ continue;
++ }
+
+ if (strncmp(devinfo.label.name, AudioCoutputs, MAX_AUDIO_DEV_LEN) == 0)
+ oclass_idx = devinfo.index;
Re: UPDATE: tig
Survived cursory testing on amd64, diff looks good. Ok by me. Tobias
Re: UPDATE: audio/mpd
On Mon, Apr 04, 2011 at 10:09:43PM +0200, Landry Breuil wrote: On Tue, Apr 05, 2011 at 12:01:18AM +0530, Martin Pieuchot wrote: Hi Tobias, Here is an updated version of your diff for mpd-0.16.2 . Previously discussed issues (http://marc.info/?l=openbsd-portsm=130114470610474w=2) seem to have been fixed. I've tested the libao and the httpd output without any problem on amd64. The manpage is also well displayed with mandoc. Fwiw, this works fine for me on ppc with libao/httpd. Landry I don't have interwebs at home for a few weeks and can't test it. If it looks good to you, get it in. There's also an update to mpc/ncmpc/libmpdclient from before the ports lock still waiting. At the moment I have neither time nor net to re-check them. Tobias
Re: fix: www/mozilla-firefox opentype support on sparc64
On Sat, Mar 19, 2011 at 04:48:27PM +0100, Landry Breuil wrote: On Sat, Mar 19, 2011 at 12:49:44AM +0100, Tobias Ulmer wrote: On Fri, Mar 18, 2011 at 11:51:14PM +0100, Landry Breuil wrote: On Fri, Mar 18, 2011 at 11:09:14PM +0100, Tobias Ulmer wrote: Firefox crashes on sparc64 when trying to display OpenType fonts. The ots code appears to be from google and can be found in firefox, chromium and probably some other places like webkit. Is there an upstream bug report for that particular issue ? A crash stacktrace ? Where does the patch come from ? Upstream bugtracker would be here, there is no bug report on this issue: https://code.google.com/p/chromium/issues/list?can=2q=OTS I have removed the debug builds and core files already, so no backtrace. Since there is no attribution for the patch, you can safely assume that I wrote and tested it ;-) I can only be bothered to report it upstream if it (or something better) gets commited... I try to do the procedure the other way round, report upstream first, see if they assess the issue, and then fix the port. https://bugzilla.mozilla.org/show_bug.cgi?id=643137 Upstream ticket here: https://code.google.com/p/chromium/issues/detail?id=76882 You mean nobody told you that eg midori crashes within 3 seconds of seeing any javascript? Shocking! :D It blows up in the javascript core. Yes it's the next thing I want to get fixed. My concern atm with sparc64 is getting ffx4 to run on it... Feel free to send a port in my direction.. Landry
fix: www/mozilla-firefox opentype support on sparc64
Firefox crashes on sparc64 when trying to display OpenType fonts. The
ots code appears to be from google and can be found in firefox,
chromium and probably some other places like webkit.
I've so far only tested FF since webkit has bigger issues on sparc64
than just some font stuff.
This site (http://thebrowsereview.com/category/css3/) can be used to
reproduce the issue.
Index: Makefile
===
RCS file: /srv/boron/data/vcs/cvs/openbsd/ports/www/mozilla-firefox/Makefile,v
retrieving revision 1.168
diff -u -p -r1.168 Makefile
--- Makefile10 Mar 2011 14:34:18 - 1.168
+++ Makefile17 Mar 2011 23:27:55 -
@@ -7,7 +7,7 @@ MOZILLA_VERSION = 3.6.14
MOZILLA_BRANCH = 1.9.2
MOZILLA_PROJECT = mozilla-firefox
MOZILLA_CODENAME = browser
-REVISION = 0
+REVISION = 1
SO_VERSION = 22.2
# NOTE: Must bump minor version if any shlib's are removed from the
Index: patches/patch-gfx_ots_include_opentype-sanitiser_h
===
RCS file: patches/patch-gfx_ots_include_opentype-sanitiser_h
diff -N patches/patch-gfx_ots_include_opentype-sanitiser_h
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-gfx_ots_include_opentype-sanitiser_h 17 Mar 2011 23:32:21
-
@@ -0,0 +1,19 @@
+$OpenBSD$
+
+Fix strict alignment architectures
+
+--- gfx/ots/include/opentype-sanitiser.h.orig Fri Mar 18 00:02:27 2011
gfx/ots/include/opentype-sanitiser.h Fri Mar 18 00:02:35 2011
+@@ -64,8 +64,10 @@ class OTSStream {
+ }
+
+ while (length = 4) {
+- chksum_ += ntohl(*reinterpret_castconst uint32_t*(
+- reinterpret_castconst uint8_t*(data) + offset));
++ uint32_t tmp;
++ std::memcpy(tmp, reinterpret_castconst uint8_t *(data) + offset,
++sizeof(uint32_t));
++ chksum_ += ntohl(tmp);
+ length -= 4;
+ offset += 4;
+ }
Re: fix: www/mozilla-firefox opentype support on sparc64
On Fri, Mar 18, 2011 at 11:51:14PM +0100, Landry Breuil wrote: On Fri, Mar 18, 2011 at 11:09:14PM +0100, Tobias Ulmer wrote: Firefox crashes on sparc64 when trying to display OpenType fonts. The ots code appears to be from google and can be found in firefox, chromium and probably some other places like webkit. Is there an upstream bug report for that particular issue ? A crash stacktrace ? Where does the patch come from ? Upstream bugtracker would be here, there is no bug report on this issue: https://code.google.com/p/chromium/issues/list?can=2q=OTS I have removed the debug builds and core files already, so no backtrace. Since there is no attribution for the patch, you can safely assume that I wrote and tested it ;-) I can only be bothered to report it upstream if it (or something better) gets commited... I've so far only tested FF since webkit has bigger issues on sparc64 than just some font stuff. Great. Since i didn't hear about the 'bigger issues on sparc64', i suppose you're fixing them and directly sending the fixes upstream ? :) You mean nobody told you that eg midori crashes within 3 seconds of seeing any javascript? Shocking! :D It blows up in the javascript core. Yes it's the next thing I want to get fixed. Landry
Re: [new] gtk+3-3.0.2
On Wed, Mar 09, 2011 at 06:25:03PM +0100, Antoine Jacoutot wrote: GTK+ is a multi-platform toolkit for creating graphical user interfaces. Offering a complete set of widgets, GTK+ is suitable for projects ranging from small one-off projects to complete application suites. This port can co-exist without any issue with gtk+2. It passes all regress tests and gtk3-demo seems happy so far with all widgets. I'd appreciate as much testings as possible on all arches (don't mind the Failed to load module ... messages, these are normal since we only have these modules for gtk+2 for now). Trying to run gtk3-demo on sparc64 over a forwarded X11 connection results in this: blade:tobiasu$ gtk3-demo --sync (gtk3-demo:14353): Gdk-WARNING **: The program 'gtk3-demo' received an X Window System error. This probably reflects a bug in the program. The error was 'BadRequest (invalid request code or no such operation)'. (Details: serial 69 error_code 1 request_code 142 minor_code 47) (Note to programmers: normally, X errors are reported asynchronously; that is, you will receive the error a while after causing it. To debug your program, run it with the --sync command line option to change this behavior. You can then get a meaningful backtrace from your debugger if you break on the gdk_x_error() function.) blade:tobiasu$ echo $DISPLAY localhost:10.0 Thanks. -- Antoine
Re: [new] gtk+3-3.0.2
On Thu, Mar 10, 2011 at 09:03:57AM +0100, Antoine Jacoutot wrote: Do you use ssh -Y ? Yes Does gtk2 hexibit the same? No, both gtk-demo and midori work (unless js is involved, but that's another bug hunt) Also when possible please try on a real display, I wouldn't bet this is specific to gtk+3. crashes right away: (gdb) run Starting program: /usr/local/bin/gtk3-demo Program received signal SIGBUS, Bus error. 0x0002096c0298 in copy_classes () from /usr/X11R6/lib/libXi.so.11.0 (gdb) bt full #0 0x0002096c0298 in copy_classes () from /usr/X11R6/lib/libXi.so.11.0 No symbol table info available. #1 0x0002096c2c8c in XIQueryDevice () from /usr/X11R6/lib/libXi.so.11.0 No symbol table info available. #2 0x000210c955f4 in gdk_drag_action_get_type () from /usr/local/lib/libgdk-3.so.0.0 No symbol table info available. #3 0x00020cad96a4 in g_object_newv () from /usr/local/lib/libgobject-2.0.so.2800.0 No symbol table info available. #4 0x00020cada2ec in g_object_new_valist () from /usr/local/lib/libgobject-2.0.so.2800.0 No symbol table info available. #5 0x00020cada44c in g_object_new () from /usr/local/lib/libgobject-2.0.so.2800.0 No symbol table info available. #6 0x000210c92e10 in gdk_drag_action_get_type () from /usr/local/lib/libgdk-3.so.0.0 No symbol table info available. #7 0x000210c97170 in gdk_drag_action_get_type () from /usr/local/lib/libgdk-3.so.0.0 No symbol table info available. #8 0x000210c95a10 in gdk_drag_action_get_type () from /usr/local/lib/libgdk-3.so.0.0 No symbol table info available. #9 0x000210c74518 in gdk_display_manager_open_display () from /usr/local/lib/libgdk-3.so.0.0 No symbol table info available. #10 0x000203882ea4 in gtk_init_check () from /usr/local/lib/libgtk-3.so.0.0 No symbol table info available. #11 0x000203882edc in gtk_init () from /usr/local/lib/libgtk-3.so.0.0 No symbol table info available. #12 0x001236bc in main () No symbol table info available. again, gtk-demo works on a real display However keep in mind that this is only a 7(?)bit framebuffer, stupid wildcat graphics... Thanks. -- Antoine
Re: [update] Adsuck 2.1
On Thu, Mar 10, 2011 at 07:01:12PM +, Stuart Henderson wrote: On 2011/03/10 15:50, Gonzalo L. R. wrote: Hi porters, I made the diff for the new update for Adsuck. Ok with me, but what does the maintainer say? I think he recently unsubscribed from ports@ ;-)
bugfix: net/libtorrent
This fixes a reproducible segmentation fault in rtorrent when you're
downloading a large amount of p^Wlinux distros. Can be triggered more
easily by decreasing the file descriptor limit. See description of the
fix in the patch.
Index: Makefile
===
RCS file: /srv/boron/data/vcs/cvs/openbsd/ports/net/libtorrent/Makefile,v
retrieving revision 1.25
diff -u -p -r1.25 Makefile
--- Makefile26 Nov 2010 15:00:45 - 1.25
+++ Makefile29 Jan 2011 00:24:49 -
@@ -3,6 +3,7 @@
COMMENT= BitTorrent library written in C++
DISTNAME= libtorrent-0.12.6
+REVISION= 0
SHARED_LIBS += torrent 17.0 # .11.6
CATEGORIES=net devel
Index: patches/patch-src_torrent_poll_kqueue_cc
===
RCS file:
/srv/boron/data/vcs/cvs/openbsd/ports/net/libtorrent/patches/patch-src_torrent_poll_kqueue_cc,v
retrieving revision 1.1
diff -u -p -r1.1 patch-src_torrent_poll_kqueue_cc
--- patches/patch-src_torrent_poll_kqueue_cc25 Nov 2008 16:21:28 -
1.1
+++ patches/patch-src_torrent_poll_kqueue_cc29 Jan 2011 00:21:01 -
@@ -1,6 +1,14 @@
$OpenBSD: patch-src_torrent_poll_kqueue_cc,v 1.1 2008/11/25 16:21:28 naddy Exp
$
src/torrent/poll_kqueue.cc.origTue Oct 28 04:48:48 2008
-+++ src/torrent/poll_kqueue.cc Sun Nov 23 21:17:59 2008
+
+The number of events is used to index an array of maxOpenSockets
+(file descriptors) elements. If there are less than fds than maxEvents,
+this will cause a segfault.
+
+Limit the number of events to the number of fds with an upper limit of
+maxEvents.
+
+--- src/torrent/poll_kqueue.cc.origThu Nov 12 09:03:58 2009
src/torrent/poll_kqueue.cc Sat Jan 29 01:04:59 2011
@@ -37,6 +37,7 @@
#include config.h
@@ -9,3 +17,12 @@ $OpenBSD: patch-src_torrent_poll_kqueue_
#include algorithm
#include unistd.h
+@@ -115,7 +116,7 @@ PollKQueue::create(int maxOpenSockets) {
+
+ PollKQueue::PollKQueue(int fd, int maxEvents, int maxOpenSockets) :
+ m_fd(fd),
+- m_maxEvents(maxEvents),
++ m_maxEvents((maxOpenSockets maxEvents) ? maxOpenSockets : maxEvents),
+ m_waitingEvents(0),
+ m_changedEvents(0),
+ m_stdinEvent(NULL) {
Re: update: audio/mpd
On Tue, Jan 18, 2011 at 11:46:49AM +0100, Jasper Lievisse Adriaanse wrote: On Sun, Jan 16, 2011 at 02:40:28PM -0800, Bryan Linton wrote: On 2011-01-16 05:31:18, Tobias Ulmer tobi...@tmux.org wrote: Update mpd to version 0.16.1 This thing has about 17 billion features, so please test. Worksforme on i386/amd64 Echoing worksforme on i386. mpd no longer segfaults when I accidently type mpd instead of mpc. It also used to crash if I moved a new song under the one currently playing (so that it would be next in the queue) while mpd was crossfading into the old song. I'm happy to report that this crash is also gone. audio/sonata and audio/ncmpc work fine with the new version too. -- Bryan This would be a post-lock thing IMHO. I disagree :) MPD is used by a lot of people, with loads of different audio formats and server options. So possibly breaking it with a major update is bad. I havn't heard of any regressions yet and there have been a number of testers. -- Cheers, Jasper Capable, generous men do not create victims, they nurture them.
Re: NEW calibre
On Sun, Jan 16, 2011 at 11:31:00AM +0100, Peter Hessler wrote:
Hi. I just got an ebook reader, and decided I needed to manage my
ebooks. So, here is a port for calibre.
There are still notes and TODO items left before it can be committed,
but right now it works for me for basic management.
My notes from the Makefile
# not compiling pyc files yet
# XXX - fails to build if qt3 is installed!!!
# detects and uses qt3 before qt4
# XXX - rss feeds are broken
# doesn't install .desktop etc files (needs xdg)
# not using libpodofo yet
It currently does not do the fancy HW management, since I do not have a
compatible device. I don't intend to do that myself, but others are
more than welcome to test and submit fixes.
--
i'm living so far beyond my income that we may almost be said to be
living apart.
-- e. e. cummings
I needed this diff to make it build. calibre (the binary) should be
fixed to call python2.6 instead of python.
If I give it a pdf, it just sits there, chews cycles, stops after a while and
nothing happens.
Maybe because libpodofo is missing?
It tries to run itself during install. Not having X on the build box
might have been an issue..
diff --git a/Makefile b/Makefile
index 624bbe8..215d9fa 100644
--- a/Makefile
+++ b/Makefile
@@ -25,14 +25,16 @@ MODULES = lang/python
LIB_DEPENDS= graphics/ImageMagick \
print/poppler,-qt4 \
graphics/libwmf \
- devel/chmlib
-BUILD_DEPENDS =${RUN_DEPENDS}
+ devel/chmlib \
+ textproc/icu4c
RUN_DEPENDS = www/py-mechanize \
textproc/py-cssutils \
devel/py-dateutil \
textproc/py-lxml=2.2.8 \
x11/py-qt4=4.8.2p1 \
- www/py-beautifulsoup
+ www/py-beautifulsoup \
+ graphics/py-Imaging
+BUILD_DEPENDS =${RUN_DEPENDS}
#RUN_DEPENDS= devel/xdg-utils \
# dns/py-dnspython \
update: audio/libmpdclient
Update of libmpdclient to version 2.4.
According to the changelog some more functions have been added but
none changed. safety bump anyway, can't hurt.
Tested on amd64/i386
Index: Makefile
===
RCS file: /srv/boron/data/vcs/cvs/openbsd/ports/audio/libmpdclient/Makefile,v
retrieving revision 1.4
diff -u -p -r1.4 Makefile
--- Makefile6 Sep 2010 11:40:39 - 1.4
+++ Makefile16 Jan 2011 02:04:43 -
@@ -2,9 +2,8 @@
COMMENT= library for interfacing MPD
-DISTNAME= libmpdclient-2.1
-REVISION= 1
-SHARED_LIBS= mpdclient 0.0 # 2.0
+DISTNAME= libmpdclient-2.4
+SHARED_LIBS= mpdclient 1.0 # 2.1
CATEGORIES=audio
HOMEPAGE= http://mpd.wikia.com/wiki/ClientLib:libmpdclient
Index: distinfo
===
RCS file: /srv/boron/data/vcs/cvs/openbsd/ports/audio/libmpdclient/distinfo,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 distinfo
--- distinfo5 Apr 2010 18:59:01 - 1.1.1.1
+++ distinfo16 Jan 2011 01:56:59 -
@@ -1,5 +1,5 @@
-MD5 (libmpdclient-2.1.tar.gz) = 4WDbRCkxOtd6V8MCi/dnLA==
-RMD160 (libmpdclient-2.1.tar.gz) = Vrmpk19xEgMcEiU1pOU612z4TQ0=
-SHA1 (libmpdclient-2.1.tar.gz) = 86rdkRMraOFzj25uwwrkwGlAAxk=
-SHA256 (libmpdclient-2.1.tar.gz) = GNe+ecOeCcPgMO90Ua8iF3gUMMySZdqLi67WGy8LagM=
-SIZE (libmpdclient-2.1.tar.gz) = 331508
+MD5 (libmpdclient-2.4.tar.gz) = 76aWTET6fwFIGh+DZcH2yw==
+RMD160 (libmpdclient-2.4.tar.gz) = hB11lNdmzrCSrPBjsJ3AHZWb+pU=
+SHA1 (libmpdclient-2.4.tar.gz) = dMrq82AkFdz2JxOLNaCc5HOni74=
+SHA256 (libmpdclient-2.4.tar.gz) = 4PdKo0GhjoqpKCYDTOgAqcQH5TIorOBKPE1V55pWmQE=
+SIZE (libmpdclient-2.4.tar.gz) = 334720
Index: patches/patch-src_song_c
===
RCS file:
/srv/boron/data/vcs/cvs/openbsd/ports/audio/libmpdclient/patches/patch-src_song_c,v
retrieving revision 1.1
diff -u -p -r1.1 patch-src_song_c
--- patches/patch-src_song_c6 Sep 2010 11:40:39 - 1.1
+++ patches/patch-src_song_c16 Jan 2011 01:57:38 -
@@ -2,17 +2,17 @@ $OpenBSD: patch-src_song_c,v 1.1 2010/09
Fix build on gcc2.
src/song.c.origMon Sep 6 02:24:51 2010
-+++ src/song.c Mon Sep 6 03:26:07 2010
-@@ -86,6 +86,7 @@ static struct mpd_song *
+--- src/song.c.origTue Jan 4 03:17:22 2011
src/song.c Sun Jan 16 02:57:01 2011
+@@ -101,6 +101,7 @@ static struct mpd_song *
mpd_song_new(const char *uri)
{
struct mpd_song *song;
+ unsigned i;
assert(uri != NULL);
-
-@@ -100,7 +101,7 @@ mpd_song_new(const char *uri)
+ assert(mpd_verify_uri(uri));
+@@ -116,7 +117,7 @@ mpd_song_new(const char *uri)
return NULL;
}
@@ -21,7 +21,7 @@ Fix build on gcc2.
song-tags[i].value = NULL;
song-duration = 0;
-@@ -116,12 +117,15 @@ mpd_song_new(const char *uri)
+@@ -134,12 +135,15 @@ mpd_song_new(const char *uri)
}
void mpd_song_free(struct mpd_song *song) {
@@ -39,7 +39,7 @@ Fix build on gcc2.
if (tag-value == NULL)
continue;
-@@ -151,6 +155,7 @@ struct mpd_song *
+@@ -169,6 +173,7 @@ struct mpd_song *
mpd_song_dup(const struct mpd_song *song)
{
struct mpd_song *ret;
@@ -47,7 +47,7 @@ Fix build on gcc2.
bool success;
assert(song != NULL);
-@@ -160,7 +165,7 @@ mpd_song_dup(const struct mpd_song *song)
+@@ -178,7 +183,7 @@ mpd_song_dup(const struct mpd_song *song)
/* out of memory */
return NULL;
Index: pkg/PLIST
===
RCS file: /srv/boron/data/vcs/cvs/openbsd/ports/audio/libmpdclient/pkg/PLIST,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 PLIST
--- pkg/PLIST 5 Apr 2010 18:59:01 - 1.1.1.1
+++ pkg/PLIST 16 Jan 2011 02:05:56 -
@@ -26,6 +26,7 @@ include/mpd/recv.h
include/mpd/response.h
include/mpd/search.h
include/mpd/send.h
+include/mpd/settings.h
include/mpd/song.h
include/mpd/stats.h
include/mpd/status.h
update: audio/mpc
mpc update to the latest version 0.20
Manpage looks better with mandoc
Tested on i386/amd64
Index: Makefile
===
RCS file: /srv/boron/data/vcs/cvs/openbsd/ports/audio/mpc/Makefile,v
retrieving revision 1.9
diff -u -p -r1.9 Makefile
--- Makefile15 Nov 2010 00:22:47 - 1.9
+++ Makefile17 Jan 2011 00:16:12 -
@@ -2,7 +2,7 @@
COMMENT= command line client for mpd
-DISTNAME= mpc-0.19
+DISTNAME= mpc-0.20
CATEGORIES=audio
HOMEPAGE= http://mpd.wikia.com/wiki/Client:Mpc
@@ -25,7 +25,6 @@ MODULES= converters/libiconv
CONFIGURE_STYLE= gnu
CONFIGURE_ARGS=ac_cv_func_iconv=yes
CONFIGURE_ENV= LDFLAGS=-L${LOCALBASE}/lib -liconv
-USE_GROFF =Yes
post-install:
${INSTALL_DATA_DIR} ${PREFIX}/share/examples/mpc
Index: distinfo
===
RCS file: /srv/boron/data/vcs/cvs/openbsd/ports/audio/mpc/distinfo,v
retrieving revision 1.5
diff -u -p -r1.5 distinfo
--- distinfo26 May 2010 13:59:37 - 1.5
+++ distinfo17 Jan 2011 00:08:39 -
@@ -1,5 +1,5 @@
-MD5 (mpc-0.19.tar.gz) = CK9Fmhg4jDXdgd6PnxU0yQ==
-RMD160 (mpc-0.19.tar.gz) = rIkd9Hxr3j2pUrIERfWfY+wcIM4=
-SHA1 (mpc-0.19.tar.gz) = XYkvMpJsux/2FcIMiz+/iUOlmZc=
-SHA256 (mpc-0.19.tar.gz) = nVGvnNz31olFEgkWX5VWUztBDjT2iClyR7lNf6cO3kM=
-SIZE (mpc-0.19.tar.gz) = 117232
+MD5 (mpc-0.20.tar.gz) = 6U1x/DhpkU4SrhT47gKGTA==
+RMD160 (mpc-0.20.tar.gz) = Hc7ojtFfzxP94QjbBEtVm+EnuNI=
+SHA1 (mpc-0.20.tar.gz) = jHPTgohDK1FxV900eo503CcWI+M=
+SHA256 (mpc-0.20.tar.gz) = XcAqZYiPKwU2SzISb+1v8eZjvG/G3k0SBxw7hcvZPwQ=
+SIZE (mpc-0.20.tar.gz) = 117871
Index: patches/patch-Makefile_in
===
RCS file:
/srv/boron/data/vcs/cvs/openbsd/ports/audio/mpc/patches/patch-Makefile_in,v
retrieving revision 1.1
diff -u -p -r1.1 patch-Makefile_in
--- patches/patch-Makefile_in 26 May 2010 13:59:37 - 1.1
+++ patches/patch-Makefile_in 17 Jan 2011 00:15:45 -
@@ -1,7 +1,7 @@
$OpenBSD: patch-Makefile_in,v 1.1 2010/05/26 13:59:37 jasper Exp $
Makefile.in.orig Sat May 22 00:39:05 2010
-+++ Makefile.inSat May 22 00:39:32 2010
-@@ -218,9 +218,7 @@ top_srcdir = @top_srcdir@
+--- Makefile.in.orig Tue Jan 4 03:31:41 2011
Makefile.inMon Jan 17 01:11:40 2011
+@@ -204,9 +204,7 @@ top_srcdir = @top_srcdir@
ACLOCAL_AMFLAGS = -I m4
AUTOMAKE_OPTIONS = foreign 1.10 dist-bzip2
man_MANS = doc/mpc.1
update: audio/ncmpc
Update ncmpc to the latest version 0.18
Manpage looks better with mandoc.
Uses ncursesw now, so your humppa song titles should be readable.
Did I get the EPOCH stuff right? ;-)
Tested briefly on amd64/i386, cursor/screen corruption seems to have
gone away
Index: Makefile
===
RCS file: /srv/boron/data/vcs/cvs/openbsd/ports/audio/ncmpc/Makefile,v
retrieving revision 1.19
diff -u -p -r1.19 Makefile
--- Makefile15 Nov 2010 00:22:47 - 1.19
+++ Makefile17 Jan 2011 00:40:15 -
@@ -2,9 +2,8 @@
COMMENT= curses based frontend for mpd
-DISTNAME= ncmpc-0.17
-EPOCH= 0
-REVISION= 0
+DISTNAME= ncmpc-0.18
+EPOCH= 1
CATEGORIES=audio
HOMEPAGE= http://mpd.wikia.com/wiki/Client:Ncmpc
@@ -17,7 +16,7 @@ PERMIT_PACKAGE_FTP= Yes
PERMIT_DISTFILES_CDROM=Yes
PERMIT_DISTFILES_FTP= Yes
-WANTLIB= c glib-2.0 mpdclient ncurses
+WANTLIB= c glib-2.0 mpdclient ncursesw
LIB_DEPENDS= devel/glib2 \
audio/libmpdclient
MODULES= devel/gettext
@@ -27,7 +26,6 @@ MASTER_SITES= ${MASTER_SITE_SOURCEFORGE
CONFIGURE_STYLE= gnu
CONFIGURE_ENV= CPPFLAGS=-I${LOCALBASE}/include \
LDFLAGS=-L${LOCALBASE}/lib
-USE_GROFF =Yes
post-install:
${INSTALL_DATA_DIR} ${PREFIX}/share/examples/ncmpc
Index: distinfo
===
RCS file: /srv/boron/data/vcs/cvs/openbsd/ports/audio/ncmpc/distinfo,v
retrieving revision 1.5
diff -u -p -r1.5 distinfo
--- distinfo19 Aug 2010 08:08:24 - 1.5
+++ distinfo17 Jan 2011 00:33:34 -
@@ -1,5 +1,5 @@
-MD5 (ncmpc-0.17.tar.gz) = iroZZ1ovpz7hgcSeDQEYgg==
-RMD160 (ncmpc-0.17.tar.gz) = ykpuHim74Mf+3eHFVRJuvRanUB0=
-SHA1 (ncmpc-0.17.tar.gz) = MfRArE+dm1sWC8y+N6fZQzYKQXo=
-SHA256 (ncmpc-0.17.tar.gz) = cixS/YKFla+sJ4JgNnAabx3ztDEICS9EaDYnUdz0I8E=
-SIZE (ncmpc-0.17.tar.gz) = 336303
+MD5 (ncmpc-0.18.tar.gz) = mfEcpxvjMuArF4fSxNlAEQ==
+RMD160 (ncmpc-0.18.tar.gz) = x2O2bJJgc6PZt+zT65dQREVyOAE=
+SHA1 (ncmpc-0.18.tar.gz) = MT6G7ICV/E9wtACPsToilvtvgks=
+SHA256 (ncmpc-0.18.tar.gz) = NaHNzxvEfwJhAk1gBX21ia/dSw7E80RJ9H479alY6aw=
+SIZE (ncmpc-0.18.tar.gz) = 345000
Index: pkg/PLIST
===
RCS file: /srv/boron/data/vcs/cvs/openbsd/ports/audio/ncmpc/pkg/PLIST,v
retrieving revision 1.4
diff -u -p -r1.4 PLIST
--- pkg/PLIST 19 Aug 2010 08:08:24 - 1.4
+++ pkg/PLIST 17 Jan 2011 00:37:38 -
@@ -19,6 +19,7 @@ share/locale/fr/LC_MESSAGES/ncmpc.mo
share/locale/gl/LC_MESSAGES/ncmpc.mo
share/locale/he/LC_MESSAGES/ncmpc.mo
share/locale/hu/LC_MESSAGES/ncmpc.mo
+share/locale/it/LC_MESSAGES/ncmpc.mo
share/locale/ko/LC_MESSAGES/ncmpc.mo
share/locale/nb/LC_MESSAGES/ncmpc.mo
share/locale/nl/LC_MESSAGES/ncmpc.mo
Re: update: audio/ncmpc
On Mon, Jan 17, 2011 at 01:34:35AM +, Mikolaj Kucharski wrote:
On Mon, Jan 17, 2011 at 02:05:29AM +0100, Tobias Ulmer wrote:
Did I get the EPOCH stuff right? ;-)
...
-DISTNAME= ncmpc-0.17
-EPOCH= 0
-REVISION= 0
+DISTNAME= ncmpc-0.18
+EPOCH= 1
No. Version 0.18 goes correctly up from 0.17p0, so there is no need to
bump EPOCH. EPOCH should stay at zero.
EPOCH never goes away or goes to lower value. Always need to go up, but
*only* in certain (hopefuly rare) situations. Here is example when would
go up with EPOCH:
If by any chance in a week you need to revert ncmpc update back to 0.17
(from 0.18) then you need to tell pkg tools that from now on 0.17 is
newer version than 0.18 (in terms of OpenBSD ports and packages) so you
bump EPOCH. On a everyday basis that usually does not happend.
EPOCH is bumped always when version goes backward, so pkg tools know that
package needs to be updated anyway, even if the version is lower. That
is why EPOCH is for.
Hope makes sense.
Thanks, that does make sense :)
--
best regards
q#
Fixed port:
Index: Makefile
===
RCS file: /srv/boron/data/vcs/cvs/openbsd/ports/audio/ncmpc/Makefile,v
retrieving revision 1.19
diff -u -p -r1.19 Makefile
--- Makefile15 Nov 2010 00:22:47 - 1.19
+++ Makefile17 Jan 2011 01:41:52 -
@@ -2,9 +2,8 @@
COMMENT= curses based frontend for mpd
-DISTNAME= ncmpc-0.17
+DISTNAME= ncmpc-0.18
EPOCH= 0
-REVISION= 0
CATEGORIES=audio
HOMEPAGE= http://mpd.wikia.com/wiki/Client:Ncmpc
@@ -17,7 +16,7 @@ PERMIT_PACKAGE_FTP= Yes
PERMIT_DISTFILES_CDROM=Yes
PERMIT_DISTFILES_FTP= Yes
-WANTLIB= c glib-2.0 mpdclient ncurses
+WANTLIB= c glib-2.0 mpdclient ncursesw
LIB_DEPENDS= devel/glib2 \
audio/libmpdclient
MODULES= devel/gettext
@@ -27,7 +26,6 @@ MASTER_SITES= ${MASTER_SITE_SOURCEFORGE
CONFIGURE_STYLE= gnu
CONFIGURE_ENV= CPPFLAGS=-I${LOCALBASE}/include \
LDFLAGS=-L${LOCALBASE}/lib
-USE_GROFF =Yes
post-install:
${INSTALL_DATA_DIR} ${PREFIX}/share/examples/ncmpc
Index: distinfo
===
RCS file: /srv/boron/data/vcs/cvs/openbsd/ports/audio/ncmpc/distinfo,v
retrieving revision 1.5
diff -u -p -r1.5 distinfo
--- distinfo19 Aug 2010 08:08:24 - 1.5
+++ distinfo17 Jan 2011 00:43:40 -
@@ -1,5 +1,5 @@
-MD5 (ncmpc-0.17.tar.gz) = iroZZ1ovpz7hgcSeDQEYgg==
-RMD160 (ncmpc-0.17.tar.gz) = ykpuHim74Mf+3eHFVRJuvRanUB0=
-SHA1 (ncmpc-0.17.tar.gz) = MfRArE+dm1sWC8y+N6fZQzYKQXo=
-SHA256 (ncmpc-0.17.tar.gz) = cixS/YKFla+sJ4JgNnAabx3ztDEICS9EaDYnUdz0I8E=
-SIZE (ncmpc-0.17.tar.gz) = 336303
+MD5 (ncmpc-0.18.tar.gz) = mfEcpxvjMuArF4fSxNlAEQ==
+RMD160 (ncmpc-0.18.tar.gz) = x2O2bJJgc6PZt+zT65dQREVyOAE=
+SHA1 (ncmpc-0.18.tar.gz) = MT6G7ICV/E9wtACPsToilvtvgks=
+SHA256 (ncmpc-0.18.tar.gz) = NaHNzxvEfwJhAk1gBX21ia/dSw7E80RJ9H479alY6aw=
+SIZE (ncmpc-0.18.tar.gz) = 345000
Index: pkg/PLIST
===
RCS file: /srv/boron/data/vcs/cvs/openbsd/ports/audio/ncmpc/pkg/PLIST,v
retrieving revision 1.4
diff -u -p -r1.4 PLIST
--- pkg/PLIST 19 Aug 2010 08:08:24 - 1.4
+++ pkg/PLIST 17 Jan 2011 00:43:40 -
@@ -19,6 +19,7 @@ share/locale/fr/LC_MESSAGES/ncmpc.mo
share/locale/gl/LC_MESSAGES/ncmpc.mo
share/locale/he/LC_MESSAGES/ncmpc.mo
share/locale/hu/LC_MESSAGES/ncmpc.mo
+share/locale/it/LC_MESSAGES/ncmpc.mo
share/locale/ko/LC_MESSAGES/ncmpc.mo
share/locale/nb/LC_MESSAGES/ncmpc.mo
share/locale/nl/LC_MESSAGES/ncmpc.mo
update: devel/tig
Update of tig from Martin Pieuchot (thanks!).
Minor tweaks and testing on amd64/i386 by myself.
Index: Makefile
===
RCS file: /srv/boron/data/vcs/cvs/openbsd/ports/devel/tig/Makefile,v
retrieving revision 1.4
diff -u -p -r1.4 Makefile
--- Makefile15 Nov 2010 19:46:09 - 1.4
+++ Makefile15 Jan 2011 18:41:28 -
@@ -1,7 +1,7 @@
# $OpenBSD: Makefile,v 1.4 2010/11/15 19:46:09 espie Exp $
COMMENT = ncurses GIT repository browser
-DISTNAME = tig-0.16
+DISTNAME = tig-0.16.2
CATEGORIES = devel
HOMEPAGE = http://jonas.nitro.dk/tig/
MAINTAINER = Tobias Ulmer tobi...@tmux.org
@@ -12,14 +12,13 @@ PERMIT_PACKAGE_FTP =Yes
PERMIT_DISTFILES_CDROM = Yes
PERMIT_DISTFILES_FTP = Yes
-WANTLIB = c ncurses
+WANTLIB = c ncursesw
MODULES = converters/libiconv
RUN_DEPENDS = devel/git
-MASTER_SITES = ${HOMEPAGE}/releases/
+MASTER_SITES = ${HOMEPAGE}releases/
USE_GMAKE =Yes
-USE_GROFF =Yes
CONFIGURE_STYLE = gnu
CONFIGURE_ARGS += --with-libiconv=${LOCALBASE} ASCIIDOC=false \
XMLTO=false DOCBOOK2PDF=false
Index: distinfo
===
RCS file: /srv/boron/data/vcs/cvs/openbsd/ports/devel/tig/distinfo,v
retrieving revision 1.2
diff -u -p -r1.2 distinfo
--- distinfo22 Aug 2010 17:05:19 - 1.2
+++ distinfo15 Jan 2011 16:31:38 -
@@ -1,5 +1,5 @@
-MD5 (tig-0.16.tar.gz) = aEVy2TAz1sv8Xucc/+ApNQ==
-RMD160 (tig-0.16.tar.gz) = qvAbDaPnKSH8Z2XEKNP6E5yVZRk=
-SHA1 (tig-0.16.tar.gz) = xF8wMH7ixG9vdKYL+84yT7oV7CY=
-SHA256 (tig-0.16.tar.gz) = PY7FSIc0r/GxxJdmITg7gCkpI8FoZ2mkxNwYS8hU85g=
-SIZE (tig-0.16.tar.gz) = 161519
+MD5 (tig-0.16.2.tar.gz) = 1ytdNDfbxTjqb2bHSYjXXg==
+RMD160 (tig-0.16.2.tar.gz) = LHmHXVyGZAJbtWp7y0DjQ2SsVKQ=
+SHA1 (tig-0.16.2.tar.gz) = d366EaT1On3Ms2Qf7aRNbv+2oC4=
+SHA256 (tig-0.16.2.tar.gz) = fTKQZB3LFXlJGgybN5pbaJcHaqqEWCb7NwJRDwdAbqI=
+SIZE (tig-0.16.2.tar.gz) = 162809
Index: patches/patch-configure
===
RCS file: patches/patch-configure
diff -N patches/patch-configure
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-configure 15 Jan 2011 18:58:47 -
@@ -0,0 +1,15 @@
+--- configure.orig Tue Sep 21 14:44:10 2010
configure Sat Jan 15 19:58:35 2011
+@@ -3203,10 +3203,10 @@ done
+
+ for ac_header in ncursesw/ncurses.h
+ do :
+- ac_fn_c_check_header_mongrel $LINENO ncursesw/ncurses.h
ac_cv_header_ncursesw_ncurses_h $ac_includes_default
++ ac_fn_c_check_header_mongrel $LINENO ncurses.h
ac_cv_header_ncursesw_ncurses_h $ac_includes_default
+ if test x$ac_cv_header_ncursesw_ncurses_h = xyes; then :
+ cat confdefs.h _ACEOF
+-#define HAVE_NCURSESW_NCURSES_H 1
++/* #define HAVE_NCURSESW_NCURSES_H 1 */
+ _ACEOF
+ { $as_echo $as_me:${as_lineno-$LINENO}: checking for library containing
initscr 5
+ $as_echo_n checking for library containing initscr... 6; }
Re: update: devel/tig
On Sat, Jan 15, 2011 at 08:49:14PM +0100, Tobias Ulmer wrote: Update of tig from Martin Pieuchot (thanks!). I forgot to quote Martin: Here is a simple patch for updating tig to 0.16.2. I've removed USE_GROFF because the man page is ok with mandoc and I've add a patch to detect our wchar ncurses lib. Tested here on macppc and amd64. Minor tweaks and testing on amd64/i386 by myself.
update: audio/mpd
Update mpd to version 0.16.1
This thing has about 17 billion features, so please test.
Worksforme on i386/amd64
Index: Makefile
===
RCS file: /srv/boron/data/vcs/cvs/openbsd/ports/audio/mpd/Makefile,v
retrieving revision 1.36
diff -u -p -r1.36 Makefile
--- Makefile27 Dec 2010 14:50:22 - 1.36
+++ Makefile16 Jan 2011 03:41:16 -
@@ -1,8 +1,7 @@
# $OpenBSD: Makefile,v 1.36 2010/12/27 14:50:22 ajacoutot Exp $
COMMENT = Music Player Daemon
-DISTNAME = mpd-0.15.12
-REVISION = 5
+DISTNAME = mpd-0.16.1
CATEGORIES = audio
HOMEPAGE = http://www.musicpd.org/
MAINTAINER = Tobias Ulmer tobi...@tmux.org
@@ -14,7 +13,7 @@ PERMIT_DISTFILES_CDROM =Yes
PERMIT_DISTFILES_FTP = Yes
WANTLIB = c m pthread ogg z stdc++ sndio curl glib-2.0 gthread-2.0
-WANTLIB += audiofile ao samplerate FLAC id3tag mad mikmod
+WANTLIB += audiofile ao samplerate FLAC id3tag mad mikmod sndfile
MASTER_SITES = ${MASTER_SITE_SOURCEFORGE:=musicpd/}
@@ -51,7 +50,19 @@ CONFIGURE_ARGS +=--enable-ao \
--disable-modplug \
--disable-sidplay \
--disable-ffmpeg \
- --enable-mpc
+ --enable-mpc \
+ --disable-libwrap \
+ --enable-lsr \
+ --enable-audiofile \
+ --disable-gme \
+ --enable-mad \
+ --disable-mpg123 \
+ --enable-sndfile \
+ --disable-twolame-encoder
+
+# we lack SO_PASSCRED
+# CONFIGURE_ARGS +=mpd_cv_have_struct_ucred=yes
+
# OggFLAC and tremor tests, initgroups(3)
CONFIGURE_ENV =LDFLAGS=-L${LOCALBASE}/lib \
CPPFLAGS=-I${LOCALBASE}/include -D_BSD_SOURCE
@@ -76,7 +87,9 @@ LIB_DEPENDS +=audio/libvorbis \
net/libshout \
audio/faad=2.6.1p0 \
audio/libmpcdec
-CONFIGURE_ARGS += --enable-aac
+CONFIGURE_ARGS += --enable-aac \
+ --enable-shout \
+ --enable-vorbis-encoder
.endif
USE_LIBTOOL = Yes
Index: distinfo
===
RCS file: /srv/boron/data/vcs/cvs/openbsd/ports/audio/mpd/distinfo,v
retrieving revision 1.10
diff -u -p -r1.10 distinfo
--- distinfo23 Aug 2010 14:46:55 - 1.10
+++ distinfo16 Jan 2011 02:08:33 -
@@ -1,5 +1,5 @@
-MD5 (mpd-0.15.12.tar.gz) = W9YcISTUOIsTmLMvLHFDig==
-RMD160 (mpd-0.15.12.tar.gz) = QUgnNCADCYPdKuKiN6qiOc9kO9M=
-SHA1 (mpd-0.15.12.tar.gz) = fZObgk2YmjG+NmA/jVA+N3RJxj8=
-SHA256 (mpd-0.15.12.tar.gz) = p2oa1OAxLL2EwJ9S0/eLpcEvLrmRRSSJsEsXlkrftBc=
-SIZE (mpd-0.15.12.tar.gz) = 483928
+MD5 (mpd-0.16.1.tar.gz) = 0wuOoqxislfjcj85qDSp+w==
+RMD160 (mpd-0.16.1.tar.gz) = hNcbIi3ggN7Kg4J5Vj4udWnJiC4=
+SHA1 (mpd-0.16.1.tar.gz) = f7Y6jTyfa6fS5cupBWgyWGU8iD0=
+SHA256 (mpd-0.16.1.tar.gz) = KqU4YO1YKH7PuExCRi+9iTNVJeEqN2D7hcwI7DY26NE=
+SIZE (mpd-0.16.1.tar.gz) = 617328
Index: patches/patch-Makefile_in
===
RCS file:
/srv/boron/data/vcs/cvs/openbsd/ports/audio/mpd/patches/patch-Makefile_in,v
retrieving revision 1.2
diff -u -p -r1.2 patch-Makefile_in
--- patches/patch-Makefile_in 23 Aug 2010 14:46:55 - 1.2
+++ patches/patch-Makefile_in 16 Jan 2011 03:25:52 -
@@ -1,12 +1,12 @@
$OpenBSD: patch-Makefile_in,v 1.2 2010/08/23 14:46:55 jasper Exp $
Makefile.in.orig Wed Jul 21 02:42:27 2010
-+++ Makefile.inSun Aug 22 18:45:03 2010
-@@ -1447,7 +1447,7 @@ SPARSE_CPPFLAGS = $(DEFAULT_INCLUDES) -I$(shell $(CC)
+--- Makefile.in.orig Mon Jan 10 02:58:36 2011
Makefile.inSun Jan 16 04:25:26 2011
+@@ -2241,7 +2241,7 @@ SPARSE_CPPFLAGS = $(DEFAULT_INCLUDES) -I$(shell $(CC)
# Documentation
#
man_MANS = doc/mpd.1 doc/mpd.conf.5
-doc_DATA = AUTHORS COPYING NEWS README UPGRADING doc/mpdconf.example
+doc_DATA = AUTHORS COPYING NEWS README UPGRADING
- DOCBOOK_FILES = doc/protocol.xml doc/user.xml doc/developer.xml
doc/sticker.xml
+ DOCBOOK_FILES = doc/protocol.xml doc/user.xml doc/developer.xml
@ENABLE_DOCUMENTATION_TRUE@protocoldir = $(docdir)/protocol
@ENABLE_DOCUMENTATION_TRUE@protocol_DATA = $(wildcard doc/protocol/*.html)
Index: patches/patch-configure
===
RCS file:
/srv/boron/data/vcs/cvs/openbsd/ports/audio/mpd/patches/patch-configure,v
retrieving revision 1.12
diff -u -p -r1.12 patch-configure
--- patches/patch-configure 23 Aug 2010 14:46:55 - 1.12
+++ patches/patch-configure 16 Jan 2011 02:45:26 -
@@ -2,18 +2,9 @@ $OpenBSD: patch-configure,v 1.12 2010/08
Don't reset
Re: new: audio/sidplay2
NACK for the moment, there's a hidden libtool dep I need to investigate first. Tobias On Sun, Dec 19, 2010 at 07:18:02PM +0100, Tobias Ulmer wrote: Hai, attached is a sndio-enabled port for sidplay2, the successor of sidplay. This player claims to be cycle accurate and supports many more SIDs than sidplay1 - at the cost of more CPU cycles. There's a tiny problem with libresid-builder from libsidplay2. It can't be found with the lib-depends-check mechanism. No idea how to solve this. Should do no harm at all. Tested on i386 and amd64 Test material can be found at http://www.hvsc.de/ Tobias
new: audio/sidplay2
Hai, attached is a sndio-enabled port for sidplay2, the successor of sidplay. This player claims to be cycle accurate and supports many more SIDs than sidplay1 - at the cost of more CPU cycles. There's a tiny problem with libresid-builder from libsidplay2. It can't be found with the lib-depends-check mechanism. No idea how to solve this. Should do no harm at all. Tested on i386 and amd64 Test material can be found at http://www.hvsc.de/ Tobias libsidplay2.tar.gz Description: application/tar-gz sidplay2.tar.gz Description: application/tar-gz
Re: CVS: cvs.openbsd.org: ports
Ohhh... shiny! Another thing to remove from my ever increasing TODO list. Thanks guys :) On Wed, Dec 01, 2010 at 02:06:53AM -0700, Jasper Lievisse Adriaanse wrote: CVSROOT: /cvs Module name: ports Changes by: jas...@cvs.openbsd.org 2010/12/01 02:06:53 Log message: Import mspgcc4, the toolchain for MSP430 MCUs. based on gnu gcc, binutils, gdb with additional patches for msp430 support. One will need an upcoming umodem.c commit by jakemsr@, and an upcoming mspdebug port in order to actually program the devices though.
devel/metaauto return code
Hi Marc,
metaauto should IMHO return an error code instead of failing silently.
Index: Makefile
===
RCS file: /srv/boron/data/vcs/cvs/openbsd/ports/devel/metaauto/Makefile,v
retrieving revision 1.14
diff -u -p -r1.14 Makefile
--- Makefile31 Mar 2008 08:47:25 - 1.14
+++ Makefile25 Oct 2010 21:16:39 -
@@ -2,7 +2,7 @@
COMMENT= wrapper for gnu auto*
-VERSION= 0.9
+VERSION= 1.0
DISTNAME= metaauto-${VERSION}
CATEGORIES=devel
DISTFILES= pkg-config-0.21.tar.gz
Index: files/meta.in
===
RCS file: /srv/boron/data/vcs/cvs/openbsd/ports/devel/metaauto/files/meta.in,v
retrieving revision 1.5
diff -u -p -r1.5 meta.in
--- files/meta.in 31 Mar 2008 08:47:25 - 1.5
+++ files/meta.in 25 Oct 2010 21:16:39 -
@@ -26,9 +26,10 @@
case x...@prog_version in
x) # figure out manually what to do
- echo Provide an @PROG_VERSION environment variable, please
+ echo Provide an @PROG_VERSION environment variable, please 2
+ exit 127
;;
-x*)
+x*)
exec @LOCALBASE@/bin/@scr...@-$@PROG_VERSION $@
;;
esac
Re: UPDATE: warzone2100 2.3.5
On Wed, Oct 13, 2010 at 06:38:15PM -0400, Brad wrote: Here is an update to Warzone 2100 2.3.5. I'm looking for someone with an i386 system to test this out. Tried it on amd64 with the intention to look at the lp64 bugs. Anyway, it doesn't even pass configure now because it uses /usr/bin/flex instead of gflex. I've tracked this down to CONFIG_SITE forcing flex. Setting CONFIG_SITE=/dev/null makes it go away, but this is probably not what you want... The patch also needs to be regenerated without USE_X11. Btw the version in the tree right now is broken in the middle of the campaign, there are missing game data files (This can be repaired, you have hunt around and repack the zip file containing the game date) It would be nice to get the update in...
Re: games/supertuxkart: use freeglut
On Sat, Oct 09, 2010 at 11:53:14PM +0200, Pascal Stumpf wrote: On Saturday 09 October 2010 16:33:38 Matthias Kilian wrote: Is there any reason why this shouldn't use freeglut (instead of glut)? Hmm. Is there a consensus to always use either of the two? Freeglut does seem like the better solution to me, but since glut and freeglut conflict, I think we should move all ports to using freeglut and then remove glut entirely. Freeglut Maintainer here: Nothing should depend on glut. If it doesn't work with freeglut, it was probably written in 1993 and never touched since. So please, switch it over :) Afair there is no port that depends on glut other than supertuxcart. All ports were switched over to freeglut after it was imported. I would love to see glut bite the dust. Getting rid of the PITA that is glut was the reason for porting freeglut (not that there was much to do). Tobias
Re: mpd skipping a lot on my thinkpad
On Fri, Oct 08, 2010 at 10:03:15PM +, Jacob Meuser wrote: [...] I committed some changes to cmpci today. it shouldn't go out of sync due to delayed interrupt notifications anymore. let me know if the problem persists. It seems to work fine so far, but I had to catch some sleep. The pciide messages are normal, the disks are in standby and sometimes they don't wake up quickly enough. /var/log/messages: Oct 9 04:12:33 boron /bsd: wd4a: device timeout writing fsbn 13840 of 13840-13855 (wd4 bn 13904; cn 0 tn 220 sn 44), retrying Oct 9 04:12:33 boron /bsd: cmpci_intr: DMA0 hwpos=3 swpos=0 Oct 9 04:12:33 boron /bsd: cmpci_intr: DMA0 hwpos=3 swpos=1 Oct 9 04:12:33 boron /bsd: cmpci_intr: DMA0 hwpos=3 swpos=2 Oct 9 04:12:33 boron /bsd: cmpci_intr: DMA0 hwpos=0 swpos=4 Oct 9 04:12:33 boron /bsd: wd4: soft error (corrected) Oct 9 04:22:59 boron /bsd: wd4(pciide1:1:0): timeout Oct 9 04:22:59 boron /bsd: type: ata Oct 9 04:22:59 boron /bsd: c_bcount: 32768 Oct 9 04:22:59 boron /bsd: c_skip: 0 Oct 9 04:22:59 boron /bsd: pciide1:1:0: bus-master DMA error: missing interrupt, status=0x60 Oct 9 04:22:59 boron /bsd: wd4: transfer error, downgrading to Ultra-DMA mode 4 Oct 9 04:22:59 boron /bsd: wd4(pciide1:1:0): using PIO mode 4, Ultra-DMA mode 4 Oct 9 04:22:59 boron /bsd: wd5(pciide1:1:1): using PIO mode 4, Ultra-DMA mode 4 Oct 9 04:22:59 boron /bsd: wd4a: device timeout writing fsbn 380298192 of 380298192-380298255 (wd4 bn 380298256; cn 23672 tn 120 sn 16), retrying Oct 9 04:22:59 boron /bsd: cmpci_intr: DMA0 hwpos=2 swpos=1 Oct 9 04:22:59 boron /bsd: cmpci_intr: DMA1 hwpos=2 swpos=1 Oct 9 04:23:00 boron /bsd: wd4: soft error (corrected) Oct 9 05:00:01 boron newsyslog[24384]: logfile turned over Oct 9 05:00:01 boron newsyslog[24384]: logfile turned over Oct 9 05:00:01 boron syslogd: restart Oct 9 05:27:19 boron /bsd: cmpci_intr: DMA0 hwpos=0 swpos=4 Oct 9 05:27:19 boron /bsd: cmpci_intr: DMA1 hwpos=0 swpos=4 Oct 9 05:28:30 boron /bsd: cmpci_intr: DMA0 hwpos=0 swpos=4 Oct 9 05:28:30 boron /bsd: cmpci_intr: DMA1 hwpos=0 swpos=4 Oct 9 05:37:09 boron /bsd: cmpci_intr: DMA0 hwpos=1 swpos=0 Oct 9 05:37:09 boron /bsd: cmpci_intr: DMA1 hwpos=1 swpos=0 Oct 9 06:05:41 boron /bsd: cmpci_intr: DMA0 hwpos=0 swpos=4 Oct 9 06:05:41 boron /bsd: cmpci_intr: DMA1 hwpos=0 swpos=4 Oct 9 06:43:20 boron /bsd: cmpci_intr: DMA0 hwpos=4 swpos=3 Oct 9 06:43:20 boron /bsd: cmpci_intr: DMA1 hwpos=4 swpos=3 Oct 9 07:02:03 boron /bsd: cmpci_intr: DMA1 hwpos=4 swpos=3 Oct 9 07:02:03 boron /bsd: cmpci_intr: DMA0 hwpos=0 swpos=4 Oct 9 07:48:38 boron /bsd: cmpci_intr: DMA0 hwpos=0 swpos=4 Oct 9 07:48:38 boron /bsd: cmpci_intr: DMA1 hwpos=0 swpos=4 Oct 9 08:27:58 boron /bsd: cmpci_intr: DMA0 hwpos=0 swpos=2 Oct 9 08:27:58 boron /bsd: cmpci_intr: DMA0 hwpos=0 swpos=3 Oct 9 08:27:58 boron /bsd: cmpci_intr: DMA0 hwpos=0 swpos=4 Oct 9 08:27:58 boron /bsd: cmpci_intr: DMA1 hwpos=0 swpos=2 Oct 9 08:27:58 boron /bsd: cmpci_intr: DMA1 hwpos=0 swpos=3 Oct 9 08:27:58 boron /bsd: cmpci_intr: DMA1 hwpos=0 swpos=4 Oct 9 08:30:17 boron /bsd: cmpci_intr: DMA0 hwpos=4 swpos=2 Oct 9 08:30:17 boron /bsd: cmpci_intr: DMA0 hwpos=4 swpos=3 Oct 9 08:30:17 boron /bsd: cmpci_intr: DMA1 hwpos=4 swpos=2 Oct 9 08:30:17 boron /bsd: cmpci_intr: DMA1 hwpos=4 swpos=3 Oct 9 08:34:23 boron /bsd: cmpci_intr: DMA0 hwpos=1 swpos=4 Oct 9 08:34:23 boron /bsd: cmpci_intr: DMA0 hwpos=1 swpos=0 Oct 9 08:34:23 boron /bsd: cmpci_intr: DMA1 hwpos=1 swpos=4 Oct 9 08:34:23 boron /bsd: cmpci_intr: DMA1 hwpos=1 swpos=0 Oct 9 08:37:02 boron /bsd: cmpci_intr: DMA0 hwpos=1 swpos=0 Oct 9 08:37:03 boron /bsd: cmpci_intr: DMA1 hwpos=1 swpos=0 Oct 9 08:50:57 boron /bsd: cmpci_intr: DMA0 hwpos=1 swpos=3 Oct 9 08:50:57 boron /bsd: cmpci_intr: DMA0 hwpos=1 swpos=4 Oct 9 08:50:57 boron /bsd: cmpci_intr: DMA0 hwpos=1 swpos=0 Oct 9 08:50:57 boron /bsd: cmpci_intr: DMA1 hwpos=1 swpos=3 Oct 9 08:50:57 boron /bsd: cmpci_intr: DMA1 hwpos=1 swpos=4 Oct 9 08:50:57 boron /bsd: cmpci_intr: DMA1 hwpos=1 swpos=0 Oct 9 09:07:46 boron /bsd: cmpci_intr: DMA0 hwpos=1 swpos=0 Oct 9 09:07:46 boron /bsd: cmpci_intr: DMA1 hwpos=1 swpos=0 Oct 9 10:35:23 boron /bsd: cmpci_intr: DMA0 hwpos=1 swpos=0 Oct 9 10:35:23 boron /bsd: cmpci_intr: DMA1 hwpos=1 swpos=0 -- jake...@sdf.lonestar.org SDF Public Access UNIX System - http://sdf.lonestar.org
Re: mpd skipping a lot on my thinkpad
On Tue, Oct 05, 2010 at 05:48:08PM +, Jacob Meuser wrote: On Tue, Oct 05, 2010 at 04:27:59PM +0200, Landry Breuil wrote: On Tue, Oct 05, 2010 at 02:15:29PM +, Jacob Meuser wrote: On Mon, Oct 04, 2010 at 05:11:47PM +, Jacob Meuser wrote: On Mon, Oct 04, 2010 at 05:57:50PM +0200, Tobias Ulmer wrote: I'm against adding any cargo cult stuff like this and would rather see someone who knows aucat and the audio stuff figure out why it skips and clicks like mad. I doubt it's mpds fault. But that's probably not going to happen... Oh well. Commit it if you really have to. this is the first I heard of mpd skipping with auich. skipping was reported with azalia, but since I haven't heard anything about that since the interrupt issue was dealt with, I figured that's what was causing it. anyway, I have no idea how to use mpd, and don't really care enough to figure it out on my own. there are many mpd users here, afaik. so, I pkg_add'd mpd and mpc ... $ mpd log: problem opening log file /var/log/mpd/mpd.log (config line 37) for writing Abort trap (core dumped) $ oh come on. pkg_info -M mpd will show you that by default it's supposed to be launched by root, and /etc/mpd.conf has shitload of comments. no, it tells you how to launch it from rc.local. it does not say mpd will crash and burn if you start it as a regular user without changing the config file. Hey Jacob, cool down a bit ;) The core dump comes from the fact that it uses glib error message routines that are intended for debugging. Thanks to David, that has been fixed upstream recently. Another stupid thing you will likely run into is that it doesn't create some files on its own. touch them... Now, mpd is a daemon and the default config reflects that. You can of course change it, but that's not how upstream distributes it. I don't like the Debian way of doing things, and I'm probably not alone here... Anyway, I don't know what Davids exact problem is. I can only describe mine. I have a cmpci(4) which usually works great: cmpci0 at pci0 dev 16 function 0 C-Media Electronics CMI8738/C3DX Audio rev 0x10: apic 2 int 19 (irq 11) Under heavy IO + crypto softraid, it sometimes skips. That's ok by me, I don't expect a 2x500mhz machine to run cvs up and backups and nfs etc while providing excellent audio. What is new since a couple of months (guessing 2ish), it goes into an endless clicking noise intermixed with music and doesn't recover from that even if the machine is idle again. It's not that easy to reproduce, I have to do quite a few things in parallel to get it there, and often it doesn't happen at all. mpc stop sleep 2 mpc play fixes it for a short time under load. There is a noticable delay between the end of clicking and the end of music (~.3 sec or so) Afair play.errors increases during the clicking time. What makes me think it's not mpd is that it also happens with mplayer. The difference is that mplayer consumes 1% cpu compared to 10% mpd. It just happens less often. I can't really pin it down, and I didn't want to write a bug report which basically consists of wild guessing. I'll try without aucat for a while to check if that has any influence. Davids problem sounds different to mine, could well be they are unrelated. Tobias And yes, mpd's error handling is totally awful. MAINTAINER is aware of it. Landry and so why should I care, really? if it doesn't work as a regular user try running it as root? talk about oh come on ... -- jake...@sdf.lonestar.org SDF Public Access UNIX System - http://sdf.lonestar.org
Re: mpd skipping a lot on my thinkpad
I'm against adding any cargo cult stuff like this and would rather see someone who knows aucat and the audio stuff figure out why it skips and clicks like mad. I doubt it's mpds fault. But that's probably not going to happen... Oh well. Commit it if you really have to. On Mon, Oct 04, 2010 at 01:21:57PM +0100, Stuart Henderson wrote: On 2010/10/04 14:07, Jasper Lievisse Adriaanse wrote: On Mon, Oct 04, 2010 at 02:00:06PM +0200, David Coppa wrote: On Mon, 04 Oct 2010, Jasper Lievisse Adriaanse wrote: On Mon, Oct 04, 2010 at 10:35:11AM +0200, David Coppa wrote: Since I've spent quite a lot of time to figure out why mpd was skipping so badly on my X41, I'd like to have the following diff committed (for posterity ;)). That was because auich(4) has a sampling rate of 48000Hz instead of the default 44100 most cards have (infact I run aucat with -r 48000). Comments or suggestions are welcome... this varies depending on the codec; my older thinkpad supports a wider range of sample rates on auich. No, please put that blurb in the mpd.conf patch, instead of poluting MESSAGE. What about this? No, that's confusing, please put Defaults to: between the two audio_output_format lines and this is OK with me. that's not the default; default is to perform no sample rate conversion: .B audio_output_format sample_rate:bits:channels This specifies the sample rate, bits per sample, and number of channels of audio that is sent to each audio output. Note that audio outputs may specify their own audio format which will be used for actual output to the audio device. An example is 44100:16:2 for 44100Hz, 16 bits, and 2 channels. The default is to use the audio format of the input file. ++# If mpd skips a lot during playback, adjust this setting to reflect the ++# sound card's sampling rate. ++# For example: this text is good, ++# audio_output_format 48000:16:2 + # + #audio_output_format44100:16:2 I would either just leave the 44K1 example, or just change it to the 48K example, rather than leave both in.
Re: UPDATE, WIP: openttd 1.0.3
On Wed, Sep 01, 2010 at 12:47:44AM -0400, Daniel Dickman wrote: Hi Anton (and Michael), Here is update for the OpenTTD port to the latest stable version. Ports for OpenGFX, OpenMSX, and OpenSFX are included. Now OpenTTD could be played without the non-free stuff. I've reordered the port a bit and made some subdirs (or should it be done with multi packages?). Also I had to alter the MESSAGE, so check it for readable and valid English. I had a different way of updating this port to 1.0.3 that might be a bit simpler since it keeps everything in one package instead of splitting things up (which I don't think is needed). *poke* *poke* I would love to see this port updated :) Two hours of serious testing revealed that the music doesn't work, but sound was OK. Everything else looked good.
Re: bulk rebuild of previously installed ports
On Fri, Sep 03, 2010 at 09:23:48PM -0400, as...@spamcop.net wrote: I notice that port executables seem to get significantly stale quicker than I thought. At least stuff stops fully working and maybe I should try to rebuild the executables. Now I do not know of a way for me to automatically rebuild ports that I have previously installed and only those ports. And I am not inclined to do a true bulk build. Now I tend to favor python and it would seem there is enough information to write a program that would support that kind of limited automatic rebuild. Sort of thinking I might be able to look at port tree make files and see where the executable would have gone and see if the executable is actually there. Keep it simple and generate a script with a lot cd's and make's in it and then execute it. Does this seem like it would be of use to more than just myself? Any thoughts about the right design and algorithm? Or is there already an obvious to a non-luser way to do this? Usual disclaimer: Use packages, not ports. You probably want to look into the (poorly documented) SUBDIRLIST feature. From memory: L=$(mktemp) pkg_info -a -P -q $L make FORCE_UPDATE=Yes FETCH_PACKAGES=Yes SUBDIRLIST=$L update It sometimes blows up and you have to fix things manually, but hey... Tobias
update: devel/tig
Tested on amd64 fora while...
Index: Makefile
===
RCS file: /vcs/cvs/openbsd/ports/devel/tig/Makefile,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 Makefile
--- Makefile24 Oct 2008 22:49:21 - 1.1.1.1
+++ Makefile22 Aug 2010 16:09:22 -
@@ -1,7 +1,7 @@
# $OpenBSD: Makefile,v 1.1.1.1 2008/10/24 22:49:21 bernd Exp $
COMMENT = ncurses GIT repository browser
-DISTNAME = tig-0.12.1
+DISTNAME = tig-0.16
CATEGORIES = devel
HOMEPAGE = http://jonas.nitro.dk/tig/
MAINTAINER = Tobias Ulmer tobi...@tmux.org
@@ -20,7 +20,8 @@ MASTER_SITES =${HOMEPAGE}/releases/
USE_GMAKE =Yes
CONFIGURE_STYLE = gnu
-CONFIGURE_ARGS += --with-libiconv=${LOCALBASE}
+CONFIGURE_ARGS += --with-libiconv=${LOCALBASE} ASCIIDOC=false \
+ XMLTO=false DOCBOOK2PDF=false
NO_REGRESS = Yes
post-install:
Index: distinfo
===
RCS file: /vcs/cvs/openbsd/ports/devel/tig/distinfo,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 distinfo
--- distinfo24 Oct 2008 22:49:21 - 1.1.1.1
+++ distinfo22 Aug 2010 16:09:22 -
@@ -1,5 +1,5 @@
-MD5 (tig-0.12.1.tar.gz) = Rfh6gj/9AVNnIIz+aAvLnw==
-RMD160 (tig-0.12.1.tar.gz) = q4VIclkEOQFfzYcdYTNFJIEeSSA=
-SHA1 (tig-0.12.1.tar.gz) = uGeiLiSOPAFEiep8M1Z+TKiF2Io=
-SHA256 (tig-0.12.1.tar.gz) = jOdGKgw4gl8wdXwq+lV+vIqiPY0X446oPc3Hu4Fa81M=
-SIZE (tig-0.12.1.tar.gz) = 121286
+MD5 (tig-0.16.tar.gz) = aEVy2TAz1sv8Xucc/+ApNQ==
+RMD160 (tig-0.16.tar.gz) = qvAbDaPnKSH8Z2XEKNP6E5yVZRk=
+SHA1 (tig-0.16.tar.gz) = xF8wMH7ixG9vdKYL+84yT7oV7CY=
+SHA256 (tig-0.16.tar.gz) = PY7FSIc0r/GxxJdmITg7gCkpI8FoZ2mkxNwYS8hU85g=
+SIZE (tig-0.16.tar.gz) = 161519
Index: patches/patch-tig_c
===
RCS file: patches/patch-tig_c
diff -N patches/patch-tig_c
--- patches/patch-tig_c 24 Oct 2008 22:49:21 - 1.1.1.1
+++ /dev/null 1 Jan 1970 00:00:00 -
@@ -1,13 +0,0 @@
-$OpenBSD: patch-tig_c,v 1.1.1.1 2008/10/24 22:49:21 bernd Exp $
tig.c.orig Wed Oct 8 14:44:57 2008
-+++ tig.c Fri Oct 24 21:40:33 2008
-@@ -43,7 +43,9 @@
- #include iconv.h
-
- /* ncurses(3): Must be defined to have extended wide-character functions. */
-+/* disable because of wchar type conflicts and we don't have them anyway
- #define _XOPEN_SOURCE_EXTENDED
-+*/
-
- #ifdef HAVE_NCURSESW_NCURSES_H
- #include ncursesw/ncurses.h
update: audio/mpd
Minor update from 0.15.9 to 0.15.12
Works for me on i386 and amd64. More tests welcome...
Index: Makefile
===
RCS file: /vcs/cvs/openbsd/ports/audio/mpd/Makefile,v
retrieving revision 1.23
diff -u -p -r1.23 Makefile
--- Makefile3 Jun 2010 16:06:07 - 1.23
+++ Makefile22 Aug 2010 22:17:17 -
@@ -1,8 +1,7 @@
# $OpenBSD: Makefile,v 1.23 2010/06/03 16:06:07 pea Exp $
COMMENT = Music Player Daemon
-DISTNAME = mpd-0.15.9
-PKGNAME = ${DISTNAME}p1
+DISTNAME = mpd-0.15.12
CATEGORIES = audio
HOMEPAGE = http://www.musicpd.org/
MAINTAINER = Tobias Ulmer tobi...@tmux.org
@@ -28,7 +27,7 @@ LIB_DEPENDS = audiofile::devel/libaudio
mad::audio/libmad \
mikmod::audio/libmikmod \
curl::net/curl \
- glib-2.0,gthread-2.0::devel/glib2 \
+ glib-2.0,gthread-2.0::devel/glib2
FLAVORS = tremor
FLAVOR ?=
Index: distinfo
===
RCS file: /vcs/cvs/openbsd/ports/audio/mpd/distinfo,v
retrieving revision 1.9
diff -u -p -r1.9 distinfo
--- distinfo9 Apr 2010 11:39:33 - 1.9
+++ distinfo22 Aug 2010 16:54:43 -
@@ -1,5 +1,5 @@
-MD5 (mpd-0.15.9.tar.gz) = LL0ieicmUv3qzkKhoivwyw==
-RMD160 (mpd-0.15.9.tar.gz) = Nf36W4HDq6I6KvTVi26cX5ep+Mo=
-SHA1 (mpd-0.15.9.tar.gz) = 5RGDouU/68bBXbGTlHfBJezoD8k=
-SHA256 (mpd-0.15.9.tar.gz) = /jmP7yP0l8SS0vLJoSDrvCHMFfOAjGfT8XIoFOzV0pY=
-SIZE (mpd-0.15.9.tar.gz) = 482909
+MD5 (mpd-0.15.12.tar.gz) = W9YcISTUOIsTmLMvLHFDig==
+RMD160 (mpd-0.15.12.tar.gz) = QUgnNCADCYPdKuKiN6qiOc9kO9M=
+SHA1 (mpd-0.15.12.tar.gz) = fZObgk2YmjG+NmA/jVA+N3RJxj8=
+SHA256 (mpd-0.15.12.tar.gz) = p2oa1OAxLL2EwJ9S0/eLpcEvLrmRRSSJsEsXlkrftBc=
+SIZE (mpd-0.15.12.tar.gz) = 483928
Index: patches/patch-Makefile_in
===
RCS file: /vcs/cvs/openbsd/ports/audio/mpd/patches/patch-Makefile_in,v
retrieving revision 1.1
diff -u -p -r1.1 patch-Makefile_in
--- patches/patch-Makefile_in 12 Jan 2010 18:58:26 - 1.1
+++ patches/patch-Makefile_in 22 Aug 2010 16:54:43 -
@@ -1,7 +1,7 @@
$OpenBSD: patch-Makefile_in,v 1.1 2010/01/12 18:58:26 wcmaier Exp $
Makefile.in.orig Wed Nov 18 20:48:14 2009
-+++ Makefile.inThu Dec 17 16:11:52 2009
-@@ -1525,7 +1525,7 @@ SPARSE_CPPFLAGS = $(DEFAULT_INCLUDES) -I$(shell $(CC)
+--- Makefile.in.orig Wed Jul 21 02:42:27 2010
Makefile.inSun Aug 22 18:45:03 2010
+@@ -1447,7 +1447,7 @@ SPARSE_CPPFLAGS = $(DEFAULT_INCLUDES) -I$(shell $(CC)
# Documentation
#
man_MANS = doc/mpd.1 doc/mpd.conf.5
Index: patches/patch-configure
===
RCS file: /vcs/cvs/openbsd/ports/audio/mpd/patches/patch-configure,v
retrieving revision 1.11
diff -u -p -r1.11 patch-configure
--- patches/patch-configure 22 May 2010 11:10:09 - 1.11
+++ patches/patch-configure 22 Aug 2010 16:54:43 -
@@ -2,9 +2,9 @@ $OpenBSD: patch-configure,v 1.11 2010/05
Don't reset CPPFLAGS...stupid configure.
configure.orig Mon Mar 22 01:24:53 2010
-+++ configure Fri May 21 16:51:51 2010
-@@ -8979,7 +8979,7 @@ if test x$enable_mpc = xyes; then
+--- configure.orig Wed Jul 21 02:42:27 2010
configure Sun Aug 22 18:45:03 2010
+@@ -8990,7 +8990,7 @@ if test x$enable_mpc = xyes; then
oldcppflags=$CPPFLAGS
CFLAGS=$CFLAGS $MPD_CFLAGS $MPCDEC_CFLAGS -I.
LIBS=$LIBS $MPD_LIBS $MPCDEC_LIBS
@@ -13,7 +13,7 @@ Don't reset CPPFLAGS...stupid configure.
ac_fn_c_check_header_mongrel $LINENO mpc/mpcdec.h
ac_cv_header_mpc_mpcdec_h $ac_includes_default
if test x$ac_cv_header_mpc_mpcdec_h = xyes; then :
old_mpcdec=no
-@@ -9203,7 +9203,7 @@ if test x$enable_aac = xyes; then
+@@ -9214,7 +9214,7 @@ if test x$enable_aac = xyes; then
oldcppflags=$CPPFLAGS
CFLAGS=$CFLAGS $MPD_CFLAGS $FAAD_CFLAGS -I.
LIBS=$LIBS $MPD_LIBS $FAAD_LIBS
@@ -22,7 +22,7 @@ Don't reset CPPFLAGS...stupid configure.
ac_fn_c_check_header_mongrel $LINENO faad.h ac_cv_header_faad_h
$ac_includes_default
if test x$ac_cv_header_faad_h = xyes; then :
-@@ -9457,7 +9457,7 @@ if test x$enable_aac = xyes; then
+@@ -9468,7 +9468,7 @@ if test x$enable_aac = xyes; then
oldcppflags=$CPPFLAGS
CFLAGS=$CFLAGS $MPD_CFLAGS $FAAD_CFLAGS -Werror
LIBS=$LIBS $MPD_LIBS $FAAD_LIBS
@@ -31,7 +31,7 @@ Don't reset CPPFLAGS...stupid configure.
{ $as_echo $as_me:${as_lineno-$LINENO}: checking for broken libfaad
headers 5
$as_echo_n checking for broken libfaad headers... 6; }
-@@ -9503,7 +9503,7 @@ if test x$enable_aac = xyes; then
+@@ -9514,7 +9514,7 @@ if test x$enable_aac = xyes; then
oldcppflags=$CPPFLAGS
CFLAGS=$CFLAGS $FAAD_CFLAGS
Re: mpd core dumps with bad argument
On Wed, Aug 11, 2010 at 09:44:57PM +0200, Jiri B. wrote: Hello, mpd core dumps if executed with bad arguments. It doesn't affect running mpd, still playing :) It's because the mpd developers havn't read the manual of the library they are using: http://library.gnome.org/devel/glib/stable/glib-Message-Logging.html#g-error I'm certainly not going to patch this. Tobias jirib # mpd -kill ** ERROR **: option parsing failed: Unknown option -kill aborting... Abort trap (core dumped) # rm *.core # mpd -create-db ** ERROR **: option parsing failed: Unknown option -create-db aborting... Abort trap (core dumped) # rm *.core # mpd -verbose ** ERROR **: option parsing failed: Unknown option -verbose aborting... Abort trap (core dumped) # rm *.core # mpd --foobar ** ERROR **: option parsing failed: Unknown option --foobar aborting... Abort trap (core dumped) # rm *.core
Re: any suggestions on how to remove installed BUILD_DEPENDS ?
On Sun, May 23, 2010 at 03:20:32AM +0200, Robert wrote:
Hi,
i have have a system where i build and use ports, no dedicated build
system.
Over time the BUILD_DEPENDS start to accumulate.
Because of the smallish ssd and slow net-connection, i'd like to
remove those periodically.
Is there an easy way to remove the packages installed as BUILD_DEPENDS ?
So far is see two options:
First there is sqlports, ... (Just a more elegant way than parsing the
Makefile's)
Secondly a loop over
pkg_delete ('pkg_info -t' minus 'pkg_info -m' )
until there is nothing left.
Any other ideas? Missing something obvious?
Build in a chroot
~roberth
update: audio/mpc
Update mpc to 0.19. It's using libmpdclient now.
Based on a diff done by oga@, thanks :)
Tested on i386
Index: Makefile
===
RCS file: /vcs/cvs/openbsd/ports/audio/mpc/Makefile,v
retrieving revision 1.5
diff -u -p -r1.5 Makefile
--- Makefile2 Apr 2009 05:18:15 - 1.5
+++ Makefile21 May 2010 23:13:35 -
@@ -2,11 +2,10 @@
COMMENT= command line client for mpd
-DISTNAME= mpc-0.12.1
-PKGNAME= ${DISTNAME}p0
+DISTNAME= mpc-0.19
CATEGORIES=audio
-HOMEPAGE= http://www.musicpd.org/mpc.shtml
+HOMEPAGE= http://mpd.wikia.com/wiki/Client:Mpc
MAINTAINER=Tobias Ulmer tobi...@tmux.org
@@ -17,18 +16,24 @@ PERMIT_DISTFILES_CDROM= Yes
PERMIT_DISTFILES_FTP= Yes
WANTLIB= c
+LIB_DEPENDS= mpdclient::audio/libmpdclient
-MASTER_SITES= http://www.musicpd.org/uploads/files/
+MASTER_SITES= ${MASTER_SITE_SOURCEFORGE:=musicpd/}
MODULES= converters/libiconv
CONFIGURE_STYLE= gnu
-CONFIGURE_ARGS=--with-iconv=${LOCALBASE}
+CONFIGURE_ARGS=ac_cv_func_iconv=yes
+CONFIGURE_ENV= LDFLAGS=-L${LOCALBASE}/lib -liconv
post-install:
${INSTALL_DATA_DIR} ${PREFIX}/share/examples/mpc
- ${INSTALL_SCRIPT} ${WRKSRC}/doc/m3u-handler.sh \
- ${WRKSRC}/doc/pls-handler.sh ${PREFIX}/share/examples/mpc
- ${INSTALL_DATA} ${WRKSRC}/doc/mpc-bashrc ${PREFIX}/share/examples/mpc
+ ${INSTALL_SCRIPT} \
+ ${WRKSRC}/doc/mpd-m3u-handler.sh \
+ ${WRKSRC}/doc/mpd-pls-handler.sh \
+ ${WRKSRC}/doc/mppledit \
+ ${PREFIX}/share/examples/mpc
+ ${INSTALL_DATA} ${WRKSRC}/doc/mpc-completion.bash \
+ ${PREFIX}/share/examples/mpc
.include bsd.port.mk
Index: distinfo
===
RCS file: /vcs/cvs/openbsd/ports/audio/mpc/distinfo,v
retrieving revision 1.4
diff -u -p -r1.4 distinfo
--- distinfo26 May 2007 22:44:36 - 1.4
+++ distinfo21 May 2010 23:13:35 -
@@ -1,5 +1,5 @@
-MD5 (mpc-0.12.1.tar.gz) = C+jMm6KC9GdjHuCxuKDElA==
-RMD160 (mpc-0.12.1.tar.gz) = NQs+X7oXjakP8XqJQ+TS0Aki1dg=
-SHA1 (mpc-0.12.1.tar.gz) = IJluzrvbMfZiUw5/XdhLvz7a40g=
-SHA256 (mpc-0.12.1.tar.gz) = kxfwQMcajQ1WXjP72EAOoPGVCqZqKXvvMyy6qYMM3gU=
-SIZE (mpc-0.12.1.tar.gz) = 113341
+MD5 (mpc-0.19.tar.gz) = CK9Fmhg4jDXdgd6PnxU0yQ==
+RMD160 (mpc-0.19.tar.gz) = rIkd9Hxr3j2pUrIERfWfY+wcIM4=
+SHA1 (mpc-0.19.tar.gz) = XYkvMpJsux/2FcIMiz+/iUOlmZc=
+SHA256 (mpc-0.19.tar.gz) = nVGvnNz31olFEgkWX5VWUztBDjT2iClyR7lNf6cO3kM=
+SIZE (mpc-0.19.tar.gz) = 117232
Index: patches/patch-Makefile_in
===
RCS file: patches/patch-Makefile_in
diff -N patches/patch-Makefile_in
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-Makefile_in 21 May 2010 23:13:36 -
@@ -0,0 +1,14 @@
+$OpenBSD$
+--- Makefile.in.orig Sat May 22 00:39:05 2010
Makefile.inSat May 22 00:39:32 2010
+@@ -218,9 +218,7 @@ top_srcdir = @top_srcdir@
+ ACLOCAL_AMFLAGS = -I m4
+ AUTOMAKE_OPTIONS = foreign 1.10 dist-bzip2
+ man_MANS = doc/mpc.1
+-doc_DATA = AUTHORS COPYING NEWS README \
+- doc/mpd-m3u-handler.sh doc/mpd-pls-handler.sh doc/mppledit \
+- doc/mpc-completion.bash
++doc_DATA = AUTHORS COPYING NEWS README
+
+ EXTRA_DIST = $(man_MANS) $(doc_DATA)
+ src_mpc_headers = \
Index: patches/patch-doc_Makefile_in
===
RCS file: patches/patch-doc_Makefile_in
diff -N patches/patch-doc_Makefile_in
--- patches/patch-doc_Makefile_in 26 May 2007 22:44:36 - 1.3
+++ /dev/null 1 Jan 1970 00:00:00 -
@@ -1,11 +0,0 @@
doc/Makefile.in.orig Wed Mar 21 19:22:57 2007
-+++ doc/Makefile.inSat May 26 15:59:01 2007
-@@ -79,7 +79,7 @@ install_sh = @install_sh@
- man_MANS = mpc.1
- pkgdata_DATA =
- docdir = $(prefix)/share/doc/$(PACKAGE)
--doc_DATA = mpc-bashrc m3u-handler.sh pls-handler.sh mppledit
-+#doc_DATA = mpc-bashrc m3u-handler.sh pls-handler.sh mppledit
- EXTRA_DIST = $(pkgdata_DATA) $(man_MANS) $(doc_DATA)
- subdir = doc
- mkinstalldirs = $(SHELL) $(top_srcdir)/mkinstalldirs
Index: patches/patch-src_command_c
===
RCS file: patches/patch-src_command_c
diff -N patches/patch-src_command_c
--- patches/patch-src_command_c 2 Apr 2009 05:18:15 - 1.3
+++ /dev/null 1 Jan 1970 00:00:00 -
@@ -1,23 +0,0 @@
-$OpenBSD: patch-src_command_c,v 1.3 2009/04/02 05:18:15 jasper Exp $
-
-Fix compilation with GCC2.
-
src/command.c.orig Tue Mar 31 15:19:10 2009
-+++ src/command.c Tue Mar 31 15:21:48 2009
-@@ -362,6 +362,7 @@ int cmd_play ( int argc, char ** argv, mpd_Connection
- {
- int song;
- int i
Re: FIX: ncmpc shuffle
Hi Edd, thanks for debugging it :) Port builds, update works, shuffle works - OK by me
Re: new audio/{lib}sidplay2
On Sun, Apr 18, 2010 at 08:22:04PM +, Jacob Meuser wrote: please port this to sio_open(3). Working on it. One issue I'm running into is mono (1 pchan). With aucat running, it plays on one channel only. Without it, I get two channel mono audio?! How do I fix that? The libresid-builder problem still remains, any clues are welcome...
new audio/{lib}sidplay2
Hi there, attached is a port of sidplay2 and libsidplay2. Both ports use PKGNAME to align with sidplay and libsidplay, pkgsrc, FreeBSD ports and various Linux distros. There's a problem with the port that I don't know how to deal with: libsidplay2 contains a shared library (libresid-builder.so.0.0) in a non-standard location. The ports infrastructure detects it anyway and deals with it nicely. However when using said library in sidplay2, it complains that it can't find resid-builder in the verifying specs stage. I've looked into bsd.port.mk and it appears to loop over /usr/lib, /usr/local/lib and /usr/X11R6/lib to check if a specified library exists. There is no way to specify a different location where it should look as well. So... sidplay2 either doesn't build at all or complains when you run lib-depends-check. I've chosen the later and commented resid-builder for now... What's the proper way to deal with this problem? Port tested on i386. Test material can be found at http://www.hvsc.de/ libsidplay2.tar.gz Description: application/tar-gz sidplay2.tar.gz Description: application/tar-gz
Re: tendra
On Fri, Apr 09, 2010 at 02:22:26AM -0400, Ted Unangst wrote: time to die? Yes please. The ancient release is useless. For the new tendra compiler collection, a completely different port has to be made. People interested should get http://svn.tendra.org/trunk/ and join #tendra/freenode. Afaik does work (for a small number of work) on OpenBSD/i386. For those who wonder: No, this is not a gcc replacement, don't bother.
Re: update audio/faad
On Tue, Apr 06, 2010 at 04:46:03PM +0200, Jasper Lievisse Adriaanse wrote:
what's the deal with this? it broke something or it was just forgotten
about?
The faad port lacks header files required to use libmp4ff. The only user
is audacious-plugins. It works there because audacious-plugins has its
own copy of mp4ff and links against this port but uses its own header
files.
Things that would need to be done:
- commit patch below
- teach audacious-plugins to use the proper includes
(- AFAIR a-p picks up libshout if installed, could be fixed in one go)
- mpd likely requires LIB_DEPENDS adjusting
There's a mpd update sitting in my inbox. I could look at these issues
and prepare patches if you help.
cheers,
jasper
On Sun, Sep 06, 2009 at 03:32:42AM +0200, Tobias Ulmer wrote:
This installs mp4ff headers required for a new mpd version.
I've already made such a patch some time ago, but there have been some
tears because it broke audacious-plugins. This happend because faad
doesn't install mp4ff properly, and instead of fixing it, everyone
thinks it's a good idea to ship their own bastardized version of faad.
audacious-plugins builds with this change now, but please test!
Tobias
Index: Makefile
===
RCS file: /nfs/titan/data0/cvs/openbsd/ports/audio/faad/Makefile,v
retrieving revision 1.39
diff -u -p -r1.39 Makefile
--- Makefile10 Aug 2009 06:29:52 - 1.39
+++ Makefile6 Sep 2009 00:58:26 -
@@ -4,7 +4,7 @@ COMMENT-main= MPEG-2 and MPEG-4 AAC deco
COMMENT-xmms= XMMS input plugin for AAC files
V= 2.6.1
DISTNAME= faad2-${V}
-PKGNAME-main= faad-${V}
+PKGNAME-main= faad-${V}p0
PKGNAME-xmms= faad-xmms-${V}p0
CATEGORIES=audio
MASTER_SITES= ${MASTER_SITE_SOURCEFORGE:=faac/}
@@ -66,5 +66,7 @@ post-configure:
post-install:
@${INSTALL_DATA} ${WRKSRC}/faadconfig.h ${PREFIX}/include
+ @${INSTALL_DATA} ${WRKSRC}/common/mp4ff/mp4ff.h ${PREFIX}/include
+ @${INSTALL_DATA} ${WRKSRC}/common/mp4ff/mp4ff_int_types.h
${PREFIX}/include
.include bsd.port.mk
Index: pkg/PLIST-main
===
RCS file: /nfs/titan/data0/cvs/openbsd/ports/audio/faad/pkg/PLIST-main,v
retrieving revision 1.2
diff -u -p -r1.2 PLIST-main
--- pkg/PLIST-main 15 Sep 2008 21:58:19 - 1.2
+++ pkg/PLIST-main 6 Sep 2009 01:17:15 -
@@ -6,6 +6,8 @@
@bin bin/faad
include/faad.h
include/faadconfig.h
+include/mp4ff.h
+include/mp4ff_int_types.h
include/neaacdec.h
lib/libfaad.a
lib/libfaad.la
--
Stay Hungry. Stay Foolish.
Re: new: audio/libmpdclient
On Sun, Mar 21, 2010 at 10:07:32PM +0100, Landry Breuil wrote: [..] Gut! For the record, i'm ok with this once imports are re-allowed. Landry Final version (I hope ;-)). Fixed implicit declaration of memset spotted by Landry. libmpdclient.tar.gz Description: application/tar-gz
Re: update: audio/ncmpc
[..]
Fwiw, works fine @i386, but fails to build @macppc for a reason why
puzzles me :
src/screen_client.c:60: error: conflicting types for `screen_auth'
src/screen_client.h:28: error: previous declaration of `screen_auth'
*** Error code 1
This issue has been resolved. Diff below
Can't you provide diffs to apply in audio/ncmpc without -pX, like
everyone :) ? I know git is cool to use, but everyone has its habits..
Landry
Index: Makefile
===
RCS file: /export/tfs/cvs/openbsd/ports/audio/ncmpc/Makefile,v
retrieving revision 1.11
diff -u -p -r1.11 Makefile
--- Makefile13 Feb 2010 13:17:03 - 1.11
+++ Makefile5 Apr 2010 11:57:10 -
@@ -2,44 +2,33 @@
COMMENT= curses based frontend for mpd
-DISTNAME= ncmpc-20070917
-PKGNAME= ${DISTNAME}p1
+DISTNAME= ncmpc-0.16.1
+PKGNAME= ${DISTNAME}v0
CATEGORIES=audio
-HOMEPAGE= http://hem.bredband.net/kaw/ncmpc/
+HOMEPAGE= http://mpd.wikia.com/wiki/Client:Ncmpc
MAINTAINER=Tobias Ulmer tobi...@tmux.org
-# GPL
+# GPLv2
PERMIT_PACKAGE_CDROM= Yes
PERMIT_PACKAGE_FTP=Yes
PERMIT_DISTFILES_CDROM=Yes
PERMIT_DISTFILES_FTP= Yes
WANTLIB= c ncurses
-LIB_DEPENDS= glib-2.0::devel/glib2
-
-USE_LIBTOOL= Yes
-
-BUILD_DEPENDS= ${MODGNU_AUTOCONF_DEPENDS} \
- ${MODGNU_AUTOMAKE_DEPENDS}
-
+LIB_DEPENDS= glib-2.0::devel/glib2 mpdclient::audio/libmpdclient
MODULES= devel/gettext
-MASTER_SITES= http://www.phil.uu.nl/~lievisse/distfiles/
+MASTER_SITES= ${MASTER_SITE_SOURCEFORGE:=musicpd/}
CONFIGURE_STYLE= gnu
-AUTOMAKE_VERSION= 1.8
-AUTOCONF_VERSION= 2.61
-CONFIGURE_ENV= CPPFLAGS=-I${LOCALBASE}/include \
- LDFLAGS=-L${LOCALBASE}/lib
-
-CFLAGS+= -ftrampolines
-
-WRKDIST= ${WRKDIR}/ncmpc
-
-pre-configure:
- cd ${WRKSRC} env AUTOCONF_VERSION=${AUTOCONF_VERSION} \
- AUTOMAKE_VERSION=${AUTOMAKE_VERSION} ./autogen.sh
+CONFIGURE_ENV= CPPFLAGS=-I${LOCALBASE}/include \
+ LDFLAGS=-L${LOCALBASE}/lib
+
+post-install:
+ ${INSTALL_DATA_DIR} ${PREFIX}/share/examples/ncmpc
+ mv ${PREFIX}/share/doc/ncmpc/{config.sample,keys.sample,ncmpc.lirc} \
+ ${PREFIX}/share/examples/ncmpc/
.include bsd.port.mk
Index: distinfo
===
RCS file: /export/tfs/cvs/openbsd/ports/audio/ncmpc/distinfo,v
retrieving revision 1.3
diff -u -p -r1.3 distinfo
--- distinfo19 Sep 2007 19:15:12 - 1.3
+++ distinfo5 Apr 2010 11:57:10 -
@@ -1,5 +1,5 @@
-MD5 (ncmpc-20070917.tar.gz) = OqGlggazd/isBhJYbw/oBg==
-RMD160 (ncmpc-20070917.tar.gz) = wIIALvJO/u0mkGUzj+YuhRySxoQ=
-SHA1 (ncmpc-20070917.tar.gz) = GhgwMJZ4bsQEsb7Cu3jc0xQ+8Pg=
-SHA256 (ncmpc-20070917.tar.gz) = kM/BD6vSt/KT64PQ2HEw8RGkgMG0nrxzcrl0w7lHzaA=
-SIZE (ncmpc-20070917.tar.gz) = 247962
+MD5 (ncmpc-0.16.1.tar.gz) = NahCXVZwjIYx5Hcabsu5Wg==
+RMD160 (ncmpc-0.16.1.tar.gz) = KuRO+r/4JkXhv9tgDiEzWK420Ug=
+SHA1 (ncmpc-0.16.1.tar.gz) = 58cxX+ZUH8LYzoJ+5arZF6RZfvo=
+SHA256 (ncmpc-0.16.1.tar.gz) = H4lLoXsPq22q5B4wGI+x8yrABZ7Yj/9uRxGr8nstrl8=
+SIZE (ncmpc-0.16.1.tar.gz) = 326156
Index: patches/patch-doc_Makefile_am
===
RCS file: patches/patch-doc_Makefile_am
diff -N patches/patch-doc_Makefile_am
--- patches/patch-doc_Makefile_am 19 Sep 2007 19:15:12 - 1.1
+++ /dev/null 1 Jan 1970 00:00:00 -
@@ -1,11 +0,0 @@
-$OpenBSD: patch-doc_Makefile_am,v 1.1 2007/09/19 19:15:12 jasper Exp $
doc/Makefile.am.orig Wed Sep 19 19:34:17 2007
-+++ doc/Makefile.amWed Sep 19 19:34:30 2007
-@@ -3,6 +3,6 @@
- #
-
- man_MANS = ncmpc.1
--docdir = $(prefix)/share/doc/$(PACKAGE)
-+docdir = $(prefix)/share/examples/$(PACKAGE)
- doc_DATA = config.sample keys.sample ncmpc.lirc
- EXTRA_DIST = $(man_MANS) $(doc_DATA)
Index: patches/patch-src_colors_c
===
RCS file: patches/patch-src_colors_c
diff -N patches/patch-src_colors_c
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-src_colors_c 5 Apr 2010 11:57:10 -
@@ -0,0 +1,23 @@
+$OpenBSD$
+Fix screen corruption in ncmpc. Colors and attributes can not be mixed.
+Colors are in the low 8 bit, mask them in the appropriate places.
+--- src/colors.c.orig Sat Mar 20 11:00:32 2010
src/colors.c Sat Mar 20 11:02:03 2010
+@@ -86,7 +86,7 @@ colors_update_pair(enum color id)
+
+ assert(id 0 id COLOR_END);
+
+- fg = colors[id].color;
++ fg = colors[id].color 0xff;
+ bg = colors[COLOR_BACKGROUND].color;
+
+ /* If color == COLOR_NONE (negative),
+@@ -273,7 +273,7 @@ colors_use(WINDOW
