CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/10/05 09:46:43 Modified files: sysutils/neofetch: Makefile security/ccrypt: Makefile Log message: neofetch, ccrypt: drop maintainership
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/09/19 13:09:39 Modified files: www/webkitgtk4 : Makefile Log message: webkitgtk4: use the "lonesome" DPB_PROPERTIES on macppc No objections from aja@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/09/05 07:26:20 Modified files: games/corewars : Makefile Log message: corewars: unbreak with ld.lld on powerpc powerpc does not inline sqrt(3) unlike other lld archs, so it actually uses libm and as such the `-lm' linker flag is needed. ok gkoehler@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/08/28 03:16:56 Modified files: sysutils/gdmap : Makefile Log message: gdmap: unbreak with ld.lld on powerpc powerpc does not inline sqrt(3) unlike other lld archs, so it actually uses libm and as such the `-lm' linker flag is needed. While here, refresh WANTLIB. ok gkoehler@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/08/28 03:07:11 Modified files: games/moonlander: Makefile games/moonlander/patches: patch-Makefile Log message: moonlander: unbreak with ld.lld on powerpc powerpc does not inline sqrt(3) unlike other lld archs, so it actually uses libm and as such the `-lm' linker flag is needed. While here, refresh WANTLIB and add FIX_EXTRACT_PERMISSIONS. tweak and ok gkoehler@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/08/28 03:03:46 Modified files: japanese/kanjipad: Makefile Log message: kanjipad: unbreak with ld.lld on powerpc powerpc does not inline sqrt(3) unlike other lld archs, so it actually uses libm and as such the `-lm' linker flag is needed. While here, refresh WANTLIB. ok gkoehler@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/08/24 14:32:52 Modified files: archivers/innoextract: Makefile Removed files: archivers/innoextract/patches: patch-cmake_BuildType_cmake Log message: innoextract: unbreak on clang+ld.bfd archs Innoextract used to fail to build on these archs due to undefined references at linking time, because of the `--as-needed' linker flag. There was a powerpc specific fix, but since then mips64 can also build innoextract. Extend the fix by disabling optimizations on all clang+ld.bfd archs. This only impacts mips64 now that macppc moved to ld.lld, but may be useful to sparc64 in the future if it moves to clang. Build tested on mips64 by Yifei Zhan < openbsd () zhan ! science >, and macppc by me (with ld.lld). There is no impact on other archs. No objection from thfr@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/08/04 03:45:30 Modified files: audio/abcde: Makefile audio/p5-MusicBrainz-DiscID: Makefile devel/p5-IO-Pipely: Makefile devel/p5-IO-Prompt-Tiny: Makefile devel/p5-MooseX-InsideOut: Makefile devel/p5-POE-Component-Syndicator: Makefile devel/p5-Test-Compile: Makefile devel/p5-Test-FailWarnings: Makefile devel/p5-Test-utf8: Makefile net/p5-Net-IP-Minimal: Makefile net/p5-POE-Component-Resolver: Makefile sysutils/screenfetch: Makefile www/p5-WebService-MusicBrainz: Makefile Log message: Drop maintainership on various ports, bump REVISION accordingly.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/08/03 05:15:44 Modified files: telephony/linphone/bctoolbox: Makefile telephony/linphone/bctoolbox/patches: patch-CMakeLists_txt Log message: bctoolbox: unbreak some consumers' build on ld.bfd archs Upstream's cmake magic checks for execinfo.h, but there is no test to check if the backtrace(3) function family requires extra linker flags. Replace the header check by testing if backtrace(3) exists, and if it does, add the required linker flag. This fixes telephony/linphone/belr build on ld.bfd archs, at least. Tested on macppc and amd64. « ok if it works for you » landry@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/08/02 13:13:28 Modified files: games/cataclysm-dda: Makefile Log message: cataclysm-dda: unbreak on powerpc Use the `-relax' linker flag to fix relocation errors.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/08/01 05:04:35 Modified files: lang/gforth: Makefile Log message: gforth: unbreak on powerpc With base-clang it either runs out of memory with i386 patches or -O1. Using -O0 makes preforth segfault. Use ports-gcc instead. OK jasper@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/07/31 09:12:14 Modified files: devel/libffi : Makefile Added files: devel/libffi/patches: patch-src_powerpc_sysv_S Log message: libffi: add a fix for the powerpc assembly code to improve runtime gkoehler@ emitted the hypothesis that libffi was partly broken on powerpc. Later i found out that upstream has already a patch for this, reducing greatly the number of test failures. This will allow to build lang/gforth, and lang/guile2 passes now its ffi test has well. Testing for possible regression on powerpc64 by gkoehler@. OK jasper@ gkoehler@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/07/30 16:50:19 Modified files: audio/grip : Makefile distinfo Log message: grip: update to 4.2.2 Changelog: https://sf.net/p/grip/code/ci/master/tree/ChangeLog OK jca@ (on a previous diff with a fix for unsigned char archs that naddy@ committed in the meantime)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/07/28 17:22:33 Modified files: devel/clang-tools-extra: Makefile Log message: clang-tools-extra: remove BROKEN on powerpc There is no more reasons to keep it broken due to wasted bulk time. Rust has taken out many ports, Qt4 has been removed and guile2 has now a bootstrap that allows a fast build. Originally proposed by Brad and gkoehler@ long ago. OK rsadowski@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/07/25 16:53:09 Modified files: multimedia/atomicparsley: Makefile distinfo multimedia/atomicparsley/patches: patch-src_CDtoc_cpp Added files: multimedia/atomicparsley/patches: patch-src_AtomicParsley_h patch-src_parsley_cpp Log message: atomicparsley: update to 20210715 Changelog: https://github.com/wez/atomicparsley/compare/0.9.6...20210715.151551.e7ad03a Also add patches to ensure the runtime is fine on unsigned char archs. OK sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/07/25 03:27:36 Modified files: mail/trojita : Makefile Added files: mail/trojita/patches: patch-src_Gui_Spinner_h Log message: trojita: fix the spinner behaviour on unsigned char archs, from upstream. OK tb@ and Caspar Schutijser (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/07/24 13:35:41 Modified files: games/qqwing : Makefile Added files: games/qqwing/patches: patch-main_cpp Log message: qqwing: fix an infinite loop on unsigned char archs Using the `--solve` option on powerpc made qqwing freeze. A looped getchar(3) call never reached EOF, since the return value was stored as a char, that is unsigned on powerpc and arm*, instead of int. OK jca@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/07/24 13:31:49 ports/games/qqwing/patches Update of /cvs/ports/games/qqwing/patches In directory cvs.openbsd.org:/tmp/cvs-serv60453/patches Log Message: Directory /cvs/ports/games/qqwing/patches added to the repository
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/07/24 13:26:54 Modified files: converters/ish : Makefile converters/ish/patches: patch-decode_c Log message: ish: fix an infinite loop on unsigned char archs Decoding using the -lf option on powerpc made ish freeze. A looped fgetc(3) call never reached EOF, since the return value was stored as a char, that is unsigned on powerpc and arm*, instead of int. OK jca@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/07/24 03:31:03 Modified files: lang/guile2: Makefile distinfo Log message: guile2: provide a valid bootstrap for 32 bit big endian archs This allows to reduce the building time by a factor of 6 on powerpc. A target has been provided to create your own bootstrap (similarly to what lang/gcc does), and could be extended for archs that have no bootstrap (64 bit big endian notably) later. EXTRACT_ONLY tweak and OK gkoehler@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/07/13 02:37:57 Modified files: editors/emacs : Makefile Log message: emacs: fix the build on powerpc A recent commit brought base libexecinfo support for emacs, but it appears it makes 'temacs' segfault during the build on powerpc. Disable the libexecinfo support, only on this arch, to fix the build. comment tweak and OK jca@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/07/10 05:49:34 Modified files: x11/mplayer: Makefile Log message: mplayer: fix the runtime on powerpc G3s Brad sent me a diff to rearrange CONFIGURE_ARGS for powerpc. Runtime testing on a G3 machine led to SIGILLs. The runtime cpu detection configure flag enables the '-maltivec' compiler flag by default, leading to the execution of altivec instructions unconditionally at runtime. They are not supported on G3s, so fix that issue by disabling the runtime cpu detection feature. OK Brad (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/07/06 13:33:21 Modified files: www/snownews : Makefile distinfo www/snownews/patches: patch-Config_mk_in patch-Makefile patch-configure www/snownews/pkg: DESCR PLIST Added files: www/snownews/patches: patch-man_Module_mk patch-po_Module_mk Removed files: www/snownews/patches: patch-netio_c www/snownews/pkg: README Log message: snownews: update to 1.8 Changelog: https://github.com/msharov/snownews/releases Major highlights: - Atom feeds are now supported, and so are HTTPS and UTF-8 display - The feeds list is now in the OPML format, configuration files have moved to ~/.config/snownews and cache files to ~/.local/share/snownews. The migration will be automatically done during the first run, with the notable exception of the 'exec:ftp -Vo-' invocations you may have used for HTTPS; they are not removed but still work. Extra testing (thanks!) by: nathanael Tim Meunier OK bcallah@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/07/03 09:00:57 Modified files: games/hyperrogue: Makefile distinfo games/hyperrogue/pkg: PLIST Added files: games/hyperrogue/patches: patch-Makefile Removed files: games/hyperrogue/patches: patch-Makefile_am Log message: hyperrogue: update to 12.0b Changelog: https://github.com/zenorogue/hyperrogue/blob/master/changelog.txt#L3997 tweak and OK bcallah@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/06/28 15:34:19 Modified files: misc/uniutils : Makefile Added files: misc/uniutils/patches: patch-unireverse_c Log message: uniutils: unbreak unireverse's runtime on unsigned char arches A looped getc(3) call in unireverse never reached EOF on these arches because the return value was stored in a char, instead of the by-the-book type, int. This fixes an infinite loop at runtime. OK jca@, maintainer timeout
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/06/19 14:37:52 Modified files: devel/sdl2 : Makefile devel/sdl : Makefile Log message: sdl, sdl2: disable AltiVec on powerpc SDL contains AltiVec instructions, whose support is not properly detected at runtime. Disable AltiVec on powerpc to fix SIGILLs on macppc G3s, while keeping it for powerpc64. « Later, one might try to enable altivec again by moving the altivec blitter to a separate file, and compiling only that file with -maltivec (as xenocara/lib/pixman does now). » - gkoehler@ Reported by @w...@empty.cafe. OK gkoehler@, no news from thfr@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/06/15 13:37:41 Modified files: net/tdlib : Makefile Log message: tdlib: big endian architectures aren't supported, don't build there OK sthen@ and Omar Polo (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/06/10 15:51:45 Modified files: graphics/mtpaint: Makefile graphics/mtpaint/pkg: DESCR Log message: mtpaint: move to gtk+3 and remove llvm-10 -fno-common workaround OK sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/06/09 09:56:32 Modified files: games/godot: Makefile distinfo games/godot/patches: patch-platform_x11_detect_py Added files: games/godot/patches: patch-thirdparty_embree_common_math_math_h patch-thirdparty_embree_common_sys_sysinfo_cpp Log message: godot: update to 3.3.2, mark BROKEN-powerpc Changelog: https://github.com/godotengine/godot/blob/3.3.2-stable/CHANGELOG.md >From Omar Polo (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/06/07 01:25:58 Modified files: x11/gnome/seahorse: Makefile Log message: seahorse: unbreak on ld.bfd arches There were undefined references errors during the build. Remove the `--as-needed' linker flag (meson's default) for these arches. OK aja@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/05/22 13:13:18 Modified files: x11/bemenu : Makefile distinfo x11/bemenu/patches: patch-GNUmakefile x11/bemenu/pkg : PLIST Log message: bemenu: update to 0.6.0, from Dimitri Karamazov (maintainer) OK bcallah@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/05/19 02:39:16 Modified files: devel/arm-none-eabi/gcc-linaro: Makefile Log message: arm-none-eabi/gcc-linaro: remove BROKEN-powerpc, the two flavors build fine there
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/05/16 13:15:23 Modified files: graphics/krita : Makefile Log message: krita: remove BROKEN-powerpc, it won't build there because rust is in its build dependencies chain.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/05/16 13:09:53 Modified files: geo/qgis : Makefile Log message: qgis: remove BROKEN-powerpc, it won't build there because rust is in its build dependencies chain.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/05/11 14:41:30 Modified files: games/gnushogi : Makefile Log message: gnushogi: remove BROKEN-powerpc, it builds and works fine with clang-11
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/05/11 09:33:59 Modified files: x11/qt5/qtwebkit: Makefile Added files: x11/qt5/qtwebkit/patches: patch-Source_WTF_wtf_Atomics_cpp Log message: qtwebkit: unbreak with clang 11 on powerpc Disable some code assuming that 8 byte __sync_* operators are not available on that arch and as such trying to redefine builtin functions. OK jca@, gkoehler@ and rsadowski@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/04/29 12:27:57 Modified files: devel/p5-Devel-Hide: Makefile distinfo Log message: p5-Devel-Hide: update to 0.0014, drop maintainership Changelog: https://metacpan.org/changes/distribution/Devel-Hide OK sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/04/12 05:22:33 Modified files: comms/sigrok/libsigrok: Makefile Added files: comms/sigrok/libsigrok/patches: patch-src_backend_c Log message: libsigrok: use libhidapi's hid_init() instead of our usbhid(3)'s one Because our usbhid(3) defines hid_init(3), the libhidapi's one has been renamed to hidapi_hid_init(). Sigrok expects the libhidapi one. This led to build failures on ld.bfd arches, and probably runtime errors on other arches. As such, redefine the hid_init() call to hidapi_hid_init(). OK sthen@, kmos@ (who build tested on sparc64) and "seems ok to me" bentley@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/04/12 04:53:55 Modified files: comms/libhidapi: Makefile Log message: libhidapi: add a comment mentioning that any consumer should be patched to use hidapi_hid_init() instead of hid_init(3), to avoid conflict with base usbhid(3). OK sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/04/03 13:31:41 Modified files: x11/gtk+4 : Makefile Log message: gtk+4: unbreak on ld.bfd arches Fix undefined references by disabling "-Wl,--as-needed" on these arches. OK aja@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/04/01 03:39:33 Modified files: www/webkitgtk4 : Makefile Added files: www/webkitgtk4/patches: patch-Source_WebKit_WebProcess_WebPage_CoordinatedGraphics_LayerTreeHost_h Removed files: www/webkitgtk4/patches: powerpc-patch-Source_JavaScriptCore_llint_LowLevelInterpreter32_64_asm Log message: webkitgtk4: unbreak on powerpc - remove powerpc specific patch (upstreamed) - the ENABLE_GRAPHICS_CONTEXT_GL option is now USE_OPENGL_OR_ES - add a patch that fixes the build when USE_OPENGL_OR_ES is set to OFF Javascript runtime on powerpc is mostly broken like it used to be before 2.30. OK aja@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/03/28 15:11:05 Modified files: games/an : Makefile distinfo games/an/pkg : PLIST Log message: an: update to 1.2 OK kirby@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/03/27 03:13:39 Modified files: games/egoboo : Makefile games/egoboo/pkg: DESCR Added files: games/egoboo/patches: patch-src_game_md2_c Log message: egoboo: unbreak on big endian archs There was a typo in the big endian specific code, leading to an undefined reference at build time. While here, remove trailing whitespaces in DESCR. OK kmos@ (who also tested on sparc64, thanks!)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/20 18:19:28 Modified files: graphics/openvdb: Makefile Log message: openvdb: don't build on powerpc This can't be built on powerpc because devel/tbb does not support 64-bits atomics on ppc32 and forcing them does not help (as seen previously with geo/osrm-backend). Spotted in the current macppc bulk.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/20 13:15:29 Modified files: shells/ksh93 : Makefile Log message: ksh93: mark BROKEN on powerpc libast fails to build on hashalloc.c/tokscan.c, as seen on powerpc64. While here make the BROKEN-powerpc64 line portcheck(1) compliant (trailing whitespace, length > 80 chars).
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/18 15:01:17 Modified files: x11/oroborus : Makefile distinfo x11/oroborus/pkg: PLIST Added files: x11/oroborus/patches: patch-src_hints_c patch-src_keyboard_c patch-src_keyboard_h patch-src_oroborus_c Log message: oroborus: update to 2.0.20, fix the build with -fno-common Also tested by Ryan Freeman, who had a similar diff. "fine with me" sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/18 14:57:10 ports/x11/oroborus/patches Update of /cvs/ports/x11/oroborus/patches In directory cvs.openbsd.org:/tmp/cvs-serv90475/patches Log Message: Directory /cvs/ports/x11/oroborus/patches added to the repository
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/13 13:49:31 Modified files: sysutils/wmcube: Makefile distinfo sysutils/wmcube/patches: patch-wmcube_wmcube_c sysutils/wmcube/pkg: PLIST Added files: sysutils/wmcube/patches: patch-wmgeneral_wmgeneral_c patch-wmgeneral_wmgeneral_h Removed files: sysutils/wmcube/patches: patch-wmcube_Makefile_OPENBSD patch-wmgeneral_list_h Log message: wmcube: update to 1.02, with -fno-common fix This update adds color schemes, smooth fonts, more logos, a manual page and various fixes. OK sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/10 15:51:28 Modified files: net/echoping : Makefile net/echoping/pkg: DESCR Log message: echoping: unbreak on powerpc clang does not inline sqrt(3) on powerpc, leading to undefined references, it requires libm (it has been applied to all archs). Thanks to gkoehler and naddy for providing better explanations about the situation! While here, remove a trailing whitespace in DESCR. OK gkoehler@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/08 16:16:49 Modified files: games/typespeed: Makefile games/typespeed/patches: patch-src_file_c patch-src_misc_c patch-src_typespeed_c Added files: games/typespeed/patches: patch-src_menu_c patch-src_network_c patch-src_typespeed_h patch-testsuite_t_loadwords_c Log message: typespeed: fix the build with -fno-common While here, fix COMMENT casing (from sthen@, thanks!) OK sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/07 15:55:38 Modified files: x11/skippy : Makefile x11/skippy/patches: patch-Makefile patch-wm_c patch-wm_h Log message: skippy: fix the build with -fno-common While here, refresh WANTLIB, MASTER_SITES (thanks to sthen@ for hosting it) and don't build with the '-g' CFLAG. OK naddy@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/05 08:47:02 Modified files: games/wmtictactoe: Makefile games/wmtictactoe/pkg: PLIST Added files: games/wmtictactoe/patches: patch-wmgeneral_wmgeneral_c patch-wmgeneral_wmgeneral_h Log message: wmtictactoe: fix the build with -fno-common While here, provide HOMEPAGE and MASTER_SITES, add FIX_EXTRACT_PERMISSIONS and refresh WANTLIB/PLIST. OK sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/05 08:42:21 Modified files: x11/wmtime : Makefile distinfo x11/wmtime/patches: patch-wmtime_wmtime_c x11/wmtime/pkg : PLIST Removed files: x11/wmtime/patches: patch-wmgeneral_list_h patch-wmtime_Makefile Log message: wmtime: update to 1.4; fix the build with -fno-common wmtime now uses libdockapp. OK sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/05 08:39:14 Modified files: audio/wmmp3: Makefile audio/wmmp3/patches: patch-main_c Added files: audio/wmmp3/patches: patch-main_h patch-mpg123ctl_h patch-wmgeneral_c patch-wmgeneral_h Log message: wmmp3: fix the build with -fno-common While here, fix hardcoded paths and add missing includes. OK sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/05 08:32:06 Modified files: games/xmj : Makefile distinfo games/xmj/pkg : PLIST Removed files: games/xmj/patches: patch-Imakefile patch-event_c Log message: xmj: update to 1.16; fix the build with -fno-common 20 years of changes: https://mahjong.julianbradfield.org/news.html Note that xmj moved to gtk+2. While here, build with -std=gnu99; this should avoid a build failure on base-gcc archs because of initial declaractions in 'for' loops. OK sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/04 06:59:25 Modified files: x11/wmtz : Makefile distinfo x11/wmtz/patches: patch-wmtz_Makefile patch-wmtz_wmtz_c Removed files: x11/wmtz/patches: patch-wmgeneral_list_h Log message: wmtz: update to 0.7.20150816 This update allows wmtz to use libdockapp, reducing the amount of patches needed, and fixes the build with -fno-common. https tweak and OK sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/04 06:55:17 Modified files: astro/wmmoonclock: Makefile distinfo astro/wmmoonclock/patches: patch-wmMoonClock_c astro/wmmoonclock/pkg: PLIST Removed files: astro/wmmoonclock/patches: patch-Makefile Log message: wmmoonclock: update to 1.30; fix the build with -fno-common OK sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/04 06:51:20 Modified files: astro/wmglobe : Makefile Log message: wmglobe: fix the build with -fno-common OK sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/04 06:50:09 Modified files: x11/wmweather : Makefile Added files: x11/wmweather/patches: patch-src_wmgeneral_c patch-src_wmgeneral_h Log message: wmweather: fix the build with -fno-common OK sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/04 06:48:53 Modified files: x11/wmitime: Makefile distinfo x11/wmitime/patches: patch-Makefile patch-wmitime_c x11/wmitime/pkg: PLIST Removed files: x11/wmitime/patches: patch-wmgeneral_list_h Log message: wmitime: update to 0.5; fix the build with -fno-common tweaks and OK sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/04 04:38:33 Modified files: x11/wmmenu : Makefile distinfo x11/wmmenu/patches: patch-Makefile patch-example_defaults patch-utils_c Added files: x11/wmmenu/patches: patch-buttonbar_c patch-events_c patch-pixmaps_c patch-wmmenu_c Log message: wmmenu: update to 1.3 OK sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/04 04:35:56 Modified files: x11/wmdate : Makefile x11/wmdate/patches: patch-Imakefile patch-wmdate_c Log message: wmdate: fix with the new libdockapp update OK sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/04 04:34:54 Modified files: net/wmnetload : Makefile net/wmnetload/patches: patch-configure Added files: net/wmnetload/patches: patch-src_wmnetload_c Log message: wmnetload: fix with the new libdockapp update OK sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/04 04:33:24 Modified files: devel/libdockapp: Makefile distinfo devel/libdockapp/patches: patch-src_Makefile_in devel/libdockapp/pkg: PLIST Log message: libdockapp: update to 0.7.3 tweaks and OK sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/02 15:41:40 Modified files: graphics/mtpaint: Makefile distinfo Log message: mtpaint: update to 3.50, fix the build with -fno-common Changelog: http://mtpaint.sourceforge.net/news.html Upstream brought back `-fcommon', but only for clang11 and gcc10, meanwhile force `-fcommon' for any compiler. OK naddy@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/02 15:12:13 Modified files: sysutils/wmapm : Makefile sysutils/wmapm/patches: patch-wmapm_c Added files: sysutils/wmapm/patches: patch-wmgeneral_wmgeneral_c patch-wmgeneral_wmgeneral_h Log message: wmapm: fix the build with -fno-common OK naddy@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/02 15:10:38 Modified files: x11/wmclockmon : Makefile x11/wmclockmon/patches: patch-wmclockmon-config_edit_c x11/wmclockmon/pkg: DESCR Added files: x11/wmclockmon/patches: patch-wmclockmon-cal_main_c patch-wmclockmon-cal_main_h patch-wmclockmon-cal_mainwindow_h patch-wmclockmon-config_actions_c patch-wmclockmon-config_edit_h patch-wmclockmon-config_mainwindow_h patch-wmclockmon-config_tools_c patch-wmclockmon-config_variables_h Log message: wmclockmon: fix the build with -fno-common and remove a trailing whitespace in DESCR. OK naddy@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/02 15:05:42 Modified files: x11/wmcalclock : Makefile x11/wmcalclock/patches: patch-wmCalClock_c Added files: x11/wmcalclock/patches: patch-Src_xutils_c patch-Src_xutils_h Log message: wmcalclock: fix the build with -fno-common OK naddy@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/02/02 05:28:50 Modified files: textproc/p5-Template-Tiny: Makefile Log message: p5-Template-Tiny: requires FIX_EXTRACT_PERMISSIONS
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/01/31 18:22:57 Modified files: misc/deco : Makefile misc/deco/patches: patch-scr_c Added files: misc/deco/patches: patch-env_h patch-tty_c Log message: deco: fix the build with -fno-common OK kn@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/01/31 18:21:36 Modified files: net/ncftp : Makefile Added files: net/ncftp/patches: patch-sh_util_gpshare_c Log message: ncftp: fix the build with -fno-common OK naddy@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/01/31 09:45:21 Modified files: games/ace : Makefile Added files: games/ace/patches: patch-lib_table_c Log message: ace: fix the build with -fno-common OK jsg@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/01/31 09:42:51 Modified files: games/xblast : Makefile Added files: games/xblast/patches: patch-cfg_control_h patch-network_h Log message: xblast: fix the build with -fno-common OK bcallah@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/01/31 09:40:35 Modified files: games/xinvaders: Makefile games/xinvaders/patches: patch-vaders_h Log message: xinvaders: fix the build with -fno-common OK naddy@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/01/30 08:37:06 Modified files: devel/fmt : Makefile Added files: devel/fmt/patches: patch-CMakeLists_txt Log message: devel/fmt: fix the build on clang+ld.bfd archs. fmt needs to be built without -Wl,--as-needed on these archs, add a AS_NEEDED option to do so. This fixes the build on powerpc and mips64. Thanks to jj@ for testing on mips64, where there is still an int128 related test failure. OK jca@ and Ashton Fagg (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/01/25 12:09:16 Modified files: security/fwbuilder: Makefile distinfo security/fwbuilder/pkg: PLIST Added files: security/fwbuilder/patches: patch-CMakeLists_txt patch-cmake_VERSION_cmake Removed files: security/fwbuilder/patches: patch-autogen_sh Log message: fwbuilder: update to 6.0.0-rc1 This comes from a new upstream, that focused on moving fwbuilder from Qt4 to Qt5. There is no other functional changes to be expected by this update. OK rsadowski@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/01/23 07:05:51 Modified files: devel/liboil : Makefile devel/liboil/patches: patch-m4_as-host-defines_m4 Log message: liboil: extend the powerpc64 fix to powerpc This fixes liboil's runtime on this platform; building with -fno-integrated-as lead to sigabrt at runtime. While here remove the tentative arm fix, that we introduced with the initial powerpc fix, it does not fix the build. Tested again by jasper@ on powerpc64, thanks!
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/01/16 15:51:21 Modified files: security/ccrypt: Makefile Log message: ccrypt: add PORTROACH to skip MacOS version reporting.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/01/10 19:17:59 Modified files: www/netsurf/browser: Makefile Added files: www/netsurf/browser/patches: patch-frontends_gtk_res_netsurf-gtk_desktop Log message: netsurf/browser: make the .desktop file actually call netsurf Since the transition of netsurf to gtk3, the binary name is 'netsurf-gtk3', but the .desktop file still uses 'netsurf-gtk'. Correct the 'Exec' entry of the .desktop file accordingly. Already reported before by Andinus , thanks! OK kn@ and bentley@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/01/07 15:29:58 Modified files: net/lagrange : Makefile Log message: lagrange: disable SSE4.1 support The check for SSE4.1 support is only done at build time, so the runtime was leading to SIGILL on machines with old amd64 processors where this instruction set is not implemented. This issue most likely hit i386 as well. tweak and OK sthen@, OK solene@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2021/01/04 10:43:02 Modified files: games/scummvm : Makefile Log message: scummvm needs a '-Wl,--relax' ldflag to build on powerpc OK jca@ and jsg@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/12/31 13:16:54 Modified files: graphics/qiv : Makefile distinfo graphics/qiv/patches: patch-Makefile patch-qiv_1 Added files: graphics/qiv/patches: patch-options_c patch-utils_c Log message: qiv: update to 2.3.2 Changelog: https://spiegl.de/qiv/changes.shtml While here, move HOMEPAGE to https, and add some patches to silence warnings emitted by clang-10 (upstreamed). OK pamela@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/12/31 13:05:33 Modified files: misc/lfm : Makefile distinfo misc/lfm/patches: patch-lfm_compress_py patch-setup_py misc/lfm/pkg : PLIST Added files: misc/lfm/patches: patch-lfm_actions_py patch-lfm_common_py Removed files: misc/lfm/patches: patch-lfm___init___py Log message: lfm: update to 3.1 This update moves lfm from Python 2 to 3, and has various backward incompatible changes; see https://inigo.katxi.org/devel/lfm/#upgrading-from-2-x-to-3-x Thanks to sthen@ and bket@ for their feedback. OK bket@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/12/29 03:47:01 Modified files: x11/icewm : Makefile distinfo x11/icewm/patches: patch-lib_Makefile_in patch-src_icesh_cc x11/icewm/pkg : PLIST Log message: icewm: update to 2.0.0 Changelog: https://github.com/ice-wm/icewm/releases/tag/2.0.0 Update highlight: image rendering is now dealt by imlib2 instead of gdk-pixbuf-xlib (that has been deprecated by GNOME). >From Timo Myyrä, with tweaks from bket@ and me. OK bket@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/12/25 17:25:19 Modified files: www/sogo : Makefile Removed files: www/sogo/patches: gcc-patch-SoObjects_Appointments_GNUmakefile Log message: sogo: unbreak on powerpc An additional Objective-C related flag has been added during the gcc era through a patch, exclusively for powerpc. That flag is not supported by clang. maintainer timeout
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/12/16 07:39:32 Modified files: geo/spatialite/gui: Makefile Log message: geo/spatialiate/gui: unbreak on ld.bfd archs The libpq configure check failed because library flags are not sanitised before the test, uselessly pulling X11 libraries. Work around that issue by providing an extra linker flag for ld.bfd archs. OK landry@ and kmos@ (who tested on sparc64, thanks!)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/12/15 12:38:22 Added files: archivers/innoextract/patches: patch-cmake_BuildType_cmake Log message: innoextract: unbreak on powerpc The -Wl,--as-needed linker flag causes the std::unique_ptr configure test to fail due to undefined reference errors. A fallback exists, using std::auto_ptr, but it has been removed from the C++17 standard, used by upstream. Remove the use of -Wl,--as-needed on powerpc, so we use the same code on all archs, instead of reenabling std::auto_ptr by force. OK thfr@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/12/15 12:33:43 ports/archivers/innoextract/patches Update of /cvs/ports/archivers/innoextract/patches In directory cvs.openbsd.org:/tmp/cvs-serv36535/patches Log Message: Directory /cvs/ports/archivers/innoextract/patches added to the repository
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/12/01 14:54:58 Modified files: emulators/spike: Makefile Added files: emulators/spike/patches: patch-fesvr_syscall_cc Log message: spike: big endian and base-gcc architectures fixes - big endian archs: correct a type mismatch in a template function call, that caused a build failure. - base-gcc archs: Add a COMPILER line since it requires ports-gcc (C++11). - powerpc: The simulation generates internal exceptions, mark it BROKEN. OK kmos@ (who tested on sparc64, thanks!) and jasper@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/11/25 14:19:51 Modified files: x11/gnome/gucharmap/patches: patch-meson_build Log message: gucharmap: unbreak on ld.bfd archs The `-Bsymbolic-functions' linker flag is not supported by ld.bfd and broke a configure test. Emit a simple message instead of an assertion error, and use that flag only if the linker is capable. Built and run tested on powerpc. "sure" jasper@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/11/11 14:27:03 Modified files: www/webkitgtk4 : Makefile Added files: www/webkitgtk4/patches: patch-Source_WebCore_platform_graphics_x11_PlatformDisplayX11_cpp powerpc-patch-Source_JavaScriptCore_llint_LowLevelInterpreter32_64_asm Log message: webkitgtk4: powerpc-only fixes 1) Fix javascript crashes at runtime Take a patch from Daniel Kolesa @ Void-PPC to avoid segfaults in consumers as soon as javascript is used. Keep it powerpc-only for now, since it has not yet been reviewed by upstream and can have consequences on other 32-bit archs. 2) Fix colors Colors were off as soon as javascript is involved. Disabling context GL fixes the issue (but hurts performances). Another patch was needed because of a missing include when this option is disabled. Thanks to sthen and aja for hints! OK aja@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/11/11 06:37:58 Modified files: multimedia/gstreamer1/plugins-good: Makefile Log message: gstreamer1/plugins-good: reenable on !rust archs (but without pulse support) pulseaudio has rust in its chain of dependencies since spidermonkey78. Disable the -pulse subpackage on !rust archs, and work around a meson issue about pulse autodetection not being honoured, so these plugins can be built there and used by www/webkitgtk4 notably. Tested on powerpc. improvements (thanks!) and no objection from aja@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/11/11 06:14:49 Modified files: emulators/frodo: Makefile Log message: frodo: unbreak with clang on powerpc CodeWarrior styled asm inline functions are used for powerpc optimisation, guarded by "__POWERPC__" that base-clang defines, but not base-gcc. As such, undefine __POWERPC__ to revert the situation as it was when powerpc was a gcc arch. Runtime tests by Anthony Richardby (thanks!) and me. OK sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/11/09 14:48:25 Modified files: x11/agar : Makefile.inc x11/agar/agar/patches: patch-core_byteswap_h patch-core_cpuinfo_c Log message: agar: powerpc fixes: remove BROKEN, fix assembly code, add runtime AltiVec detection (improvements hinted by naddy@, thanks!) OK naddy@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/11/01 16:05:14 Modified files: www/otter-browser: Makefile Log message: otter-browser: move to QtWebEngine I asked for QtWebKit in Otter to be kept when QtWebEngine was introduced, because it was the only fully featured browser that really worked on macppc. Since the last QtWebKit update, it was crashing when running javascript. And now the rust-powered spidermonkey78 is in its chain of dependencies, so it's not built anymore on macppc anyway. OK rsadoswki@, maintainer timeout
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/10/21 09:57:00 Modified files: x11/gtk+3 : Makefile x11/gtk+4 : Makefile Log message: gtk+{3,4}: a revision bump was needed on !rust archs after the last commit for the -cups subpackage. Found the hard way in the macppc bulk.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/10/20 11:13:10 Modified files: games/fs2open : Makefile Added files: games/fs2open/patches: patch-code_model_modelread_cpp Log message: fs2open: fix big endian build, limit to hardware OpenGL>=3.2-capable archs Add an upstream fix for big endian archs, and ONLY_FOR_ARCHS to limit the build to architectures with (required) OpenGL>=3.2 hardware acceleration, excepted powerpc64, that is capable but has no drm(4) support for now. Runtime tested by and OK thfr@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/10/20 10:30:27 Modified files: x11/qt5/qtbase : Makefile Log message: qt5/qtbase: does not need audio/pulseaudio, reenable on !rust archs Remove pulseaudio from BUILD_DEPENDS; there is no reference in the source code and no configure check anymore. This reenables most of x11/qt5/* on !rust archs, since pulseaudio needs the rust-powered spidermonkey78. Was already hinted by adr on a now closely related issue. Tested on macppc and amd64. OK kn@ and rsadowski@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/10/19 03:10:17 Modified files: x11/gtk+4 : Makefile Log message: gtk+4: unlock the build on non-rust archs Change a conditional so only rust archs build the -cups subpackage with colord support, since it requires the rust-powered spidermonkey78 via polkit. Tested on macppc (no rust) and amd64 (rust). "sure, if it does the same" aja@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/10/19 03:02:15 Modified files: x11/gtk+3 : Makefile Log message: gtk+3: unlock the build on non-rust archs Change a conditional so only rust archs build the -cups subpackage with colord support, since it requires the rust-powered spidermonkey78 via polkit. Tested on macppc (no rust) and amd64 (rust). "sure, if it does the same" aja@ (maintainer)