CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: bent...@cvs.openbsd.org 2018/10/27 23:59:19 Modified files: net/mutella: Makefile Added files: net/mutella/patches: patch-mutella_basicstruct_h patch-mutella_conversions_h patch-mutella_tstring_h Log message: Include string.h when using string functions. Hopefully this will fix the build on sparc64.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: afre...@cvs.openbsd.org 2018/10/27 21:11:58 Modified files: devel/p5-Moo : Makefile distinfo Log message: Update devel/p5-Moo to 2.003004 OK Sergey Bronnikov (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: afre...@cvs.openbsd.org 2018/10/27 21:09:16 Modified files: devel/p5-strictures: Makefile distinfo Log message: Update devel/p5-strictures to 2.05 OK Sergey Bronnikov (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: afre...@cvs.openbsd.org 2018/10/27 21:04:25 Modified files: textproc/p5-POE-Filter-XML: Makefile distinfo textproc/p5-POE-Filter-XML/pkg: PLIST Log message: Update textproc/p5-POE-Filter-XML to 1.140700 >From Charlene Wendling
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: afre...@cvs.openbsd.org 2018/10/27 21:00:53 Modified files: devel : Makefile net: Makefile Log message: +devel/p5-Test-FatalWarnings +net/p5-Net-IP-Minimal
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: afre...@cvs.openbsd.org 2018/10/27 20:57:14 Log message: Add net/p5-Net-IP-Minimal This Perl module is a minimal subset of Net::IP with no dependencies. It can verify and guess valid IPv4/6 addresses. From MAINTAINER Charlene Wendling Status: Vendor Tag: afresh1 Release Tags: afresh1_20181027 N ports/net/p5-Net-IP-Minimal/Makefile N ports/net/p5-Net-IP-Minimal/distinfo N ports/net/p5-Net-IP-Minimal/pkg/DESCR N ports/net/p5-Net-IP-Minimal/pkg/PLIST No conflicts created by this import
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: afre...@cvs.openbsd.org 2018/10/27 20:51:40 Log message: Add devel/p5-Test-FailWarnings This module converts warnings to Test::More fail() calls. It is designed to be used with done_testing, when you don't need to know the test count in advance. From MAINTAINER Charlene Wendling Status: Vendor Tag: afresh1 Release Tags: afresh1_20181027 N ports/devel/p5-Test-FailWarnings/Makefile N ports/devel/p5-Test-FailWarnings/distinfo N ports/devel/p5-Test-FailWarnings/pkg/DESCR N ports/devel/p5-Test-FailWarnings/pkg/PLIST No conflicts created by this import
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: afre...@cvs.openbsd.org 2018/10/27 20:47:33 Modified files: devel/p5-POE-Test-Loops: Makefile distinfo Log message: Update devel/p5-POE-Test-Loops to 1.360 Also change HOMEPAGE as suggested by Charlene >From Charlene Wendling
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: jtur...@cvs.openbsd.org 2018/10/27 18:42:37 Modified files: sysutils/login_duo: Makefile Log message: Remove MODULES = lang/python, which is clearly not needed and was most likely included when tests worked. No functional change to end pkg. Pointed out by danj@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: bcal...@cvs.openbsd.org 2018/10/27 18:28:03 Modified files: games/freeserf : Makefile distinfo games/freeserf/patches: patch-CMakeLists_txt patch-src_CMakeLists_txt Log message: Update to freeserf-0.1.1.20181027, which is mostly a small code refactoring
NEW: graphics/dibuja
Hi ports -- Attached is a new port, graphics/dibuja. Dibuja is a simple to use paint program. --- pkg/DESCR: Dibuja is a simple to use paint program like Paintbrush for Mac or classic MS Paint. It allows you to easily create drawings and edit images, add arrows, text, and effects to your drawings and pictures, or crop and resize your images. --- Works well for me on amd64. OK? ~Brian dibuja.tgz Description: Binary data
sparc64 bulk build report
bulk build on sparc64-1.ports.openbsd.org started on Wed Oct 24 11:40:56 MDT 2018 finished at Sat Oct 27 16:50:50 MDT 2018 lasted 03D22h09m done with kern.version=OpenBSD 6.4-current (GENERIC) #12: Tue Oct 23 19:42:19 MDT 2018 built packages:4882 Oct 24:927 Oct 25:3020 Oct 26:66 Oct 27:868 critical path missing pkgs: http://build-failures.rhaalovely.net//sparc64/2018-10-24/summary.log build failures: 47 http://build-failures.rhaalovely.net//sparc64/2018-10-24/archivers/zipios.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/audio/id3lib.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/audio/libofa.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/cad/gnucap.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/cad/necpp.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/converters/lastools.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/converters/pflogx.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/databases/db/v4,no_java,no_tcl.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/devel/atlas.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/devel/codeworker.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/devel/commoncpp.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/devel/xtensa-elf/gcc.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/games/dungeon-crawl.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/games/einstein.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/games/hex-a-hop.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/games/prboom-plus.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/games/xevil.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/graphics/animorph.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/graphics/cal3d.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/graphics/imv.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/graphics/png2ico.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/japanese/mecab,-ipadic.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/lang/apl.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/lang/freehdl.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/lang/python/2.7.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/lang/python/3.6.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/lang/racket-minimal.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/math/gbc.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/math/netcdf.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/misc/cdrdao,audio.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/multimedia/streamdvd.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/net/gloox.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/net/mutella.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/net/p5-Net-SSH-Perl.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/net/xprobe.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/net/zeromq.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/productivity/devtodo.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/security/sslscan,openssl.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/sysutils/atf-allwinner.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/sysutils/colortail.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/telephony/pjsua,-main.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/textproc/aspell/core.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/www/dansguardian.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/x11/blackbox.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/x11/treewm.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/x11/waimea.log http://build-failures.rhaalovely.net//sparc64/2018-10-24/x11/wmi.log recurrent failures failures/devel/xtensa-elf/gcc.log failures/games/prboom-plus.log failures/lang/apl.log failures/lang/racket-minimal.log failures/math/gbc.log failures/net/p5-Net-SSH-Perl.log failures/net/zeromq.log failures/security/sslscan,openssl.log failures/summary.log new failures +++ ls-failures Sat Oct 27 16:51:08 2018 +failures/archivers/zipios.log +failures/audio/id3lib.log +failures/audio/libofa.log +failures/cad/gnucap.log +failures/cad/necpp.log +failures/converters/lastools.log +failures/converters/pflogx.log +failures/databases/db/v4,no_java,no_tcl.log +failures/devel/atlas.log +failures/devel/codeworker.log +failures/devel/commoncpp.log +failures/games/dungeon-crawl.log +failures/games/einstein.log +failures/games/hex-a-hop.log +failures/games/xevil.log +failures/graphics/animorph.log +failures/graphics/cal3d.log +failures/graphics/imv.log +failures/graphics/png2ico.log +failures/japanese/mecab,-ipadic.log
remove py-keyczar
On Thu, 18 Oct 2018 19:57:55 -0500, Edward Lopez-Acosta wrote: > Looks like only ansible requires py-keyczar, so once this is updated > should the port be removed Anyone is against its removal? rpe (maintainer) is ok with it.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: p...@cvs.openbsd.org2018/10/27 15:27:36 Modified files: sysutils/logstash: Makefile distinfo sysutils/logstash/pkg: PLIST Log message: Update logstash 6.4.1 -> 6.4.2 ok jasper@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: p...@cvs.openbsd.org2018/10/27 15:26:08 Modified files: www/kibana : Makefile distinfo Log message: Update kibana 6.4.1 -> 6.4.2 ok jasper@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: p...@cvs.openbsd.org2018/10/27 15:24:42 Modified files: textproc/elasticsearch: Makefile distinfo Log message: Update elasticsearch 6.4.1 -> 6.4.2 ok jasper@
NEW: math/bcal
Hi ports -- Attached is a new port, math/bcal. Byte CALculator is a storage and general-purpose calculator. --- pkg/DESCR: bcal (Byte CALculator) is a REPL CLI utility for storage expressions, unit conversions or address calculations. If you can't calculate the hex address offset for (512 - 16) MiB, or the value when the 43rd bit of a 64-bit address is set mentally, bcal is for you. It has a bc mode for general-purpose numerical calculations. Features: * evaluate arithmetic expressions involving storage units * perform general purpose calculations (using bc) * convert to IEC/SI standard data storage units * interactive mode with the last valid result stored for reuse * show the address in bytes * show address as LBA:OFFSET * convert CHS to LBA and vice versa * base conversion to binary, decimal and hex * custom sector size, max heads/cylinder and max sectors/track --- This is using the GH_* variables for now but I got upstream to begin releasing real tarballs starting next release. Also, the test program is a Python 3 script. I really don't want to invoke MODULES=lang/python for something that only uses Python for testing though, so that's why the TDEPs look the way they do. OK? ~Brian bcal.tgz Description: Binary data
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2018/10/27 10:59:36 Modified files: x11/gnome/tracker-miners: Makefile x11/gnome/grilo: Makefile audio/rhythmbox: Makefile Log message: Sync WANTLIB after recent gmime30 update.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2018/10/27 10:59:07 Modified files: mail/gmime30 : Makefile distinfo mail/gmime30/pkg: PLIST Added files: mail/gmime30/patches: patch-tests_test-mime_c Log message: Update to gmime30-3.2.1.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2018/10/27 10:58:35 ports/mail/gmime30/patches Update of /cvs/ports/mail/gmime30/patches In directory cvs.openbsd.org:/tmp/cvs-serv25846/patches Log Message: Directory /cvs/ports/mail/gmime30/patches added to the repository
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2018/10/27 10:57:53 Modified files: sysutils/terraform/provider-alicloud: Makefile distinfo sysutils/terraform/provider-aws: Makefile distinfo sysutils/terraform/provider-vault: Makefile distinfo Log message: Update terraform providers to their latest release.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2018/10/27 10:45:24 Modified files: sysutils : Makefile Log message: +nomad
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2018/10/27 10:44:55 Log message: Import nomad-0.8.6 Nomad is a flexible, enterprise-grade cluster scheduler designed to easily integrate into existing workflows. Nomad can run a diverse workload of micro-service, batch, containerized and non-containerized applications. Nomad is easy to operate and scale and integrates seamlessly with Consul and Vault. Support for Mhz computation on OpenBSD is being worked on. Until it's committed, it must be manually configured using 'cpu_total_compute'. ok robert@ Status: Vendor Tag: ajacoutot Release Tags: ajacoutot_20181027 N ports/sysutils/nomad/Makefile N ports/sysutils/nomad/distinfo N ports/sysutils/nomad/pkg/DESCR N ports/sysutils/nomad/pkg/nomad.rc N ports/sysutils/nomad/pkg/PLIST N ports/sysutils/nomad/patches/patch-vendor_github_com_mitchellh_go-ps_process_openbsd_go N ports/sysutils/nomad/patches/patch-client_allocdir_fs_openbsd_go N ports/sysutils/nomad/patches/patch-vendor_github_com_fsouza_go-dockerclient_external_github_com_docker_docker_pkg_system_stat_openbsd_go N ports/sysutils/nomad/files/nomad.hcl No conflicts created by this import
Possible fix for p5-Net-SSH-Perl on sparc64
Hi ports,
net/p5-Net-SSH-Perl can't be built currently on sparc64. It uses
OpenSSH's code to implement the chacha20-poly1305 algorithm, with a
slight difference, apparently to silence a warning. I also learnt thanks
to danj@ and landry@ that it was already found out.
The patch i propose here builds and works on sparc64 and amd64, and
simply reverts poly1305.c as it is in OpenSSH, but there are 2
concerns:
1) Security implications, as the context is different than OpenSSH
itself.
2) It "reintroduces" the warning, and it appears it's overriding
Perl's own definition of the U8TO32_LE macro:
src/chacha/poly1305.c:14:1: warning: "U8TO32_LE" redefined
In file included from /usr/libdata/perl5/sparc64-openbsd/CORE/hv.h:631,
from /usr/libdata/perl5/sparc64-openbsd/CORE/perl.h:
3912, from Perl.xs:3:
/usr/libdata/perl5/sparc64-openbsd/CORE/hv_func.h:128:1: warning: this
is the location of the previous definition
I've met no issues with the sole consumer, Net::SFTP. Math::Prime::Util
is the only module i've found out that uses the same macro with the
additional ifndef, but we don't ship it, so i wasn't able to find a
precedent.
Comments are welcome!
Charlène.
Index: Makefile
===
RCS file: /cvs/ports/net/p5-Net-SSH-Perl/Makefile,v
retrieving revision 1.21
diff -u -p -u -p -r1.21 Makefile
--- Makefile10 Sep 2018 18:37:09 - 1.21
+++ Makefile25 Oct 2018 19:41:00 -
@@ -4,6 +4,7 @@ COMMENT = perl implementation of the SSH
MODULES = cpan
DISTNAME = Net-SSH-Perl-2.14
+REVISION = 0
CATEGORIES = net security
# perl
Index: patches/patch-src_chacha_poly1305_c
===
RCS file: patches/patch-src_chacha_poly1305_c
diff -N patches/patch-src_chacha_poly1305_c
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-src_chacha_poly1305_c 25 Oct 2018 19:41:00
- @@ -0,0 +1,20 @@
+$OpenBSD$
+Sync with vanilla OpenSSH code (/usr/src/usr.bin/ssh/poly1305.c)
+Fix broken sparc64 build
+Index: src/chacha/poly1305.c
+--- src/chacha/poly1305.c.orig
src/chacha/poly1305.c
+@@ -11,13 +11,11 @@
+
+ #define mul32x32_64(a,b) ((uint64_t)(a) * (b))
+
+-#ifndef U8TO32_LE
+ #define U8TO32_LE(p) \
+ (((uint32_t)((p)[0])) | \
+((uint32_t)((p)[1]) << 8) | \
+((uint32_t)((p)[2]) << 16) | \
+((uint32_t)((p)[3]) << 24))
+-#endif
+
+ #define U32TO8_LE(p, v) \
+ do { \
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2018/10/27 10:35:58 Modified files: sysutils/terraform/terraform: Makefile distinfo Log message: Update to terraform-0.11.10.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2018/10/27 10:26:05 Modified files: www/py-bokeh : Makefile distinfo www/py-bokeh/pkg: PLIST Log message: Update to py-bokeh-1.0.0.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2018/10/27 10:16:11 Modified files: sysutils/amazon-ssm-agent: Makefile distinfo sysutils/amazon-ssm-agent/patches: patch-agent_ssm_service_go Log message: Update to amazon-ssm-agent-2.3.193.0.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2018/10/27 10:13:18 Modified files: sysutils/amazon-ecs-cli: Makefile distinfo Log message: Update to ecs-cli-1.10.0.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2018/10/27 10:12:26 Modified files: security/libgcrypt: Makefile distinfo Log message: Update to libgcrypt-1.8.4.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2018/10/27 10:09:37 Modified files: graphics/gegl04: Makefile distinfo Log message: Update to gegl04-0.4.12.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2018/10/27 10:05:54 Modified files: sysutils/awscli: Makefile distinfo Log message: Update to awscli-1.16.43.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2018/10/27 10:05:42 Modified files: net/py-botocore: Makefile distinfo Log message: Update to py-botocore-1.12.33.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2018/10/27 10:05:29 Modified files: net/py-boto3 : Makefile distinfo Log message: Update to py-boto3-1.9.33.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: juan...@cvs.openbsd.org 2018/10/27 10:00:02 Modified files: graphics/pdf2djvu: Makefile distinfo Log message: Update to pdf2djvu 0.9.11.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: juan...@cvs.openbsd.org 2018/10/27 09:58:32 Modified files: lang/racket-minimal: Makefile distinfo lang/racket-minimal/pkg: PLIST Removed files: lang/racket-minimal/patches: patch-src_racket_sconfig_h Log message: Update to racket 7.1.
Re: NEW: py-tacacs_plus (tacacs+ client written in python)
On Thu, 25 Oct 2018 12:59:22 +0200, Jan Vlach wrote: > Hello ports@ > > this is py-tacacs_plus python module / standalone tacacs+ client. It > could speak with the new tacacs+ power based on shrubbery.net's > implementation. > > basic usage/syntax is here: > https://github.com/ansible/tacacs_plus/ I quickly looked at it. Can you please: - remove the license link - stop using GH stuff and use MODPY_PI - vertically align stuff I think providing a py3 flavour would be a good thing. Cheers, Daniel
Re: amd64 bulk build failures (2018-10-25) - glabels / libtool
Sebastien Marie: > So ld.lld uses RUNPATH instead of RPATH, and as libtool only uses RPATH > to infer no standard directories... it fail to found the library. Great analysis! > The following diff on libtool permits to productivity/glabels to build. ok naddy@ -- Christian "naddy" Weisgerber na...@mips.inka.de
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2018/10/27 08:03:23 Modified files: x11/e17/e : Makefile x11/e17/e/pkg : PLIST Log message: Put sample files under share/examples/enlightenment
Re: [NEW] ibus-hangul v1.5.1
On 2018-10-11 23:02, Enric Morales wrote: [...] Attached you'll find the necessary build files. [...] I have been using this port as my main IME for the last weeks and it's been working fine, but I'd like to get some feedback on this port and know about the possibility of it being committed. Thanks in advance.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: jas...@cvs.openbsd.org 2018/10/27 07:28:06 Modified files: graphics/evince: Makefile distinfo graphics/evince/patches: patch-libdocument_ev-document-misc_c Log message: update to evince-3.28.5
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: jas...@cvs.openbsd.org 2018/10/27 07:22:08 Modified files: sysutils/beats/filebeat: Makefile distinfo sysutils/beats/filebeat/pkg: PLIST sysutils/beats/heartbeat: Makefile distinfo sysutils/beats/heartbeat/pkg: PLIST sysutils/beats/metricbeat: Makefile distinfo sysutils/beats/metricbeat/pkg: PLIST sysutils/beats/packetbeat: Makefile distinfo sysutils/beats/packetbeat/pkg: PLIST Removed files: sysutils/beats/metricbeat/patches: patch-vendor_github_com_elastic_gosigar_sigar_openbsd_go Log message: update to beats-6.4.2 OK pvk@ sebastia@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: lan...@cvs.openbsd.org 2018/10/27 06:40:13 Modified files: x11/xfce4/xfce4-clipman: Makefile distinfo Log message: Update to xfce4-clipman 1.4.3
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: lan...@cvs.openbsd.org 2018/10/27 06:36:52 Modified files: x11/xfce4/xfce4-notifyd: Makefile distinfo x11/xfce4/xfce4-notifyd/pkg: PLIST Log message: Update to xfce4-notifyd 0.4.3
Re: amd64 bulk build failures (2018-10-25) - glabels / libtool
On Sat, Oct 27, 2018 at 09:29:21AM +0200, Sebastien Marie wrote:
> On Fri, Oct 26, 2018 at 05:24:26PM +0200, Christian Weisgerber wrote:
> >
> > Remaining failures from the switch to lld:
> >
> > mail/evolution-rss edbus-private not found
> > net/telepathy/folks edbus-private not found
> > productivity/glabelsedbus-private not found
>
> I took a look at glabels, and I am assuming others ports have similar
> problem.
>
> The error log is:
>
> /usr/bin/libtool --tag=CC --mode=link cc ...
> warning: could not find a edbus-private library
> Link error: edbus-private not found!
> at /usr/libdata/perl5/LT/Library.pm line 137.
> LT::Library::resolve_library(LT::Library=HASH(0x126c6fa79898),
> ARRAY(0x126c06720070), 1, undef, "LT::Program") called at
> /usr/libdata/perl5/LT/Mode/Link/Program.pm line 84
> LT::Linker::Program::link(LT::Linker::Program=HASH(0x126c6fa76868),
> LT::Program=HASH(0x126c329be310), ARRAY(0x126c0aee3cd0),
> LT::OSConfig=HASH(0x126c975caac0), ARRAY(0x126c06720070),
> LT::Library::Stash=HASH(0x126c06720028), ARRAY(0x126c0aee3f88),
> ARRAY(0x126c10a22058), ...) called at
> /usr/libdata/perl5/LT/Mode/Link/Program.pm line 28
> LT::Program::link(LT::Program=HASH(0x126c329be310),
> ARRAY(0x126c0aee3cd0), LT::OSConfig=HASH(0x126c975caac0),
> ARRAY(0x126c06720070), LT::Library::Stash=HASH(0x126c06720028),
> ARRAY(0x126c0aee3f88), ARRAY(0x126c10a22058),
> LT::Parser=HASH(0x126c329be598), ...) called at
> /usr/libdata/perl5/LT/Mode/Link.pm line 235
> LT::Mode::Link::run(LT::Mode::Link=HASH(0x126c329beb50),
> ARRAY(0x126c0aee3cd0), LT::Options=HASH(0x126c0aee3b20),
> LT::OSConfig=HASH(0x126c975caac0)) called at /usr/bin/libtool line 428
>
> it is libtool that generate an error because it didn't found
> edbus-private library.
>
> This particular library comes from databases/evolution-data-server :
>
> $ pkg_locate edbus-private
> evolution-data-server-3.28.5:databases/evolution-data-server:/usr/local/lib/evolution-data-server/libedbus-private.so
>
>
> So with ld.lld, libtool is unable to found the library.
>
> I am able to reproduce the problem with just the following command line:
>
> $ libtool --tag=CC --mode=link cc -v -o test -lebook-1.2
> warning: could not find a edbus-private library
> Link error: edbus-private not found!
> ...
>
>
> For what I understood, libtool uses objdump -p libebook-1.2.so to found
> the path dependencies (I have ktraced libtool).
>
> On pre-ld.lld system, I have the following output:
>
> $ objdump -p /usr/local/lib/libebook-1.2.so.21.0
> ...
> Dynamic Section:
> NEEDED libedata-book-1.2.so.18.0
> ...
> NEEDED libedbus-private.so
> ...
> RPATH /usr/local/lib/evolution-data-server
> ...
>
> But on post-ld.lld, the output is slightly different:
>
> $ objdump -p /usr/local/lib/libebook-1.2.so.21.0
> ...
> Dynamic Section:
> RUNPATH /usr/local/lib/evolution-data-server
> NEEDED libedata-book-1.2.so
> ...
> NEEDED libedbus-private.so
> ...
>
>
> So ld.lld uses RUNPATH instead of RPATH, and as libtool only uses RPATH
> to infer no standard directories... it fail to found the library.
>
> The following diff on libtool permits to productivity/glabels to build.
>
>
> Index: LT/Library.pm
> ===
> RCS file: /cvs/src/usr.bin/libtool/LT/Library.pm,v
> retrieving revision 1.12
> diff -u -p -r1.12 Library.pm
> --- LT/Library.pm 21 Sep 2015 08:49:06 - 1.12
> +++ LT/Library.pm 27 Oct 2018 07:26:13 -
> @@ -191,7 +191,7 @@ sub inspect
> return @deps;
> }
>
> -# give the list of RPATH directories
> +# give the list of RPATH/RUNPATH directories
> sub findrpaths
> {
> my $self = shift;
> @@ -203,16 +203,20 @@ sub findrpaths
> say "warning: library was specified that could not be found:
> $self->{key}";
> return;
> }
> - tsay {"inspecting $filename for non standard RPATH..."};
> + tsay {"inspecting $filename for non standard RPATH/RUNPATH..."};
> open(my $fh, '-|', "objdump", "-p", "--", $filename);
> while (<$fh>) {
> if (m/RPATH\s+(.*)$/) {
> @dirs = split(":", $1);
> last;
> }
> + if (m/RUNPATH\s+(.*)$/) {
> + @dirs = split(":", $1);
> + last;
> + }
> }
> tsay {"found ", (@dirs == 0) ? 'none ' : '',
> - "RPATH for $filename\n@dirs"};
> + "RPATH/RUNPATH for $filename\n@dirs"};
> return @dirs;
> }
I would refactor it to use
if (m/R(?:UN)?PATH\s+(.*)$/) {
instead, but apart from that, obviously okay.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2018/10/27 03:27:56 Modified files: databases/mariadb: Makefile databases/mariadb/pkg: PLIST-server PLIST-tests Log message: - Enable the Mroonga storage engine. - Enable XML support for the CONNECT storage engine. from Brad (maintainer)
Re: amd64 bulk build failures (2018-10-25) - glabels / libtool
On 2018 Oct 27 (Sat) at 09:48:53 +0200 (+0200), Landry Breuil wrote: :On Sat, Oct 27, 2018 at 09:29:21AM +0200, Sebastien Marie wrote: :> On Fri, Oct 26, 2018 at 05:24:26PM +0200, Christian Weisgerber wrote: :> > :> > Remaining failures from the switch to lld: :> > :> > mail/evolution-rss edbus-private not found :> > net/telepathy/folks edbus-private not found :> > productivity/glabelsedbus-private not found :> :> I took a look at glabels, and I am assuming others ports have similar :> problem. :> :> So ld.lld uses RUNPATH instead of RPATH, and as libtool only uses RPATH :> to infer no standard directories... it fail to found the library. :> :> The following diff on libtool permits to productivity/glabels to build. : :The analysis feels 100% correct, and the diff reads fine. As it will :only reinstating a behaviour that was present with ld.bfd, imo it cant :cause breakage, and might even fix other lld-related breakage.. :will put it on my (slow, 48h) builder, but i guess ajacoutot will beat :me to it :) : :Landry : from my reading of http://www.sco.com/developers/gabi/2000-07-17/ch5.dynamic.html#shobj_dependencies and the ld.lld man page, this is correct. OK -- Weiler's Law: Nothing is impossible for the man who doesn't have to do it himself.
Re: UPDATE: x11/qt5
On Fri, Oct 26, 2018 at 05:19:15PM +0200, Rafael Sadowski wrote: > Update Qt to the latest 5.9 TLS. I removed almost all libressl patches > and the backports patches from upstream. > > Can any of the libressl fallas look over it? According to the tickets, > everything's closed. > > I know it's crazy these days. Maybe someone will find a free bulk build > slot for the diff below. What could go wrong ? Will put it along semarie's libtool diff.
Re: amd64 bulk build failures (2018-10-25) - glabels / libtool
On Sat, Oct 27, 2018 at 09:29:21AM +0200, Sebastien Marie wrote: > On Fri, Oct 26, 2018 at 05:24:26PM +0200, Christian Weisgerber wrote: > > > > Remaining failures from the switch to lld: > > > > mail/evolution-rss edbus-private not found > > net/telepathy/folks edbus-private not found > > productivity/glabelsedbus-private not found > > I took a look at glabels, and I am assuming others ports have similar > problem. > > So ld.lld uses RUNPATH instead of RPATH, and as libtool only uses RPATH > to infer no standard directories... it fail to found the library. > > The following diff on libtool permits to productivity/glabels to build. The analysis feels 100% correct, and the diff reads fine. As it will only reinstating a behaviour that was present with ld.bfd, imo it cant cause breakage, and might even fix other lld-related breakage.. will put it on my (slow, 48h) builder, but i guess ajacoutot will beat me to it :) Landry
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: bent...@cvs.openbsd.org 2018/10/27 01:47:57 Modified files: fonts/ja-kanjistrokeorders-ttf: Makefile distinfo fonts/ja-kanjistrokeorders-ttf/pkg: PLIST Log message: Update to ja-kanjistrokeorders-ttf-4.002. Changelog: http://www.nihilist.org.uk/readme_en_v4.002.txt
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: bent...@cvs.openbsd.org 2018/10/27 01:32:57 Modified files: net/megatools : Makefile Added files: net/megatools/patches: patch-Makefile_in Log message: Build in C99 mode. Should fix the build on GCC architectures.
Re: amd64 bulk build failures (2018-10-25) - glabels / libtool
On Fri, Oct 26, 2018 at 05:24:26PM +0200, Christian Weisgerber wrote:
>
> Remaining failures from the switch to lld:
>
> mail/evolution-rss edbus-private not found
> net/telepathy/folks edbus-private not found
> productivity/glabelsedbus-private not found
I took a look at glabels, and I am assuming others ports have similar
problem.
The error log is:
/usr/bin/libtool --tag=CC --mode=link cc ...
warning: could not find a edbus-private library
Link error: edbus-private not found!
at /usr/libdata/perl5/LT/Library.pm line 137.
LT::Library::resolve_library(LT::Library=HASH(0x126c6fa79898),
ARRAY(0x126c06720070), 1, undef, "LT::Program") called at
/usr/libdata/perl5/LT/Mode/Link/Program.pm line 84
LT::Linker::Program::link(LT::Linker::Program=HASH(0x126c6fa76868),
LT::Program=HASH(0x126c329be310), ARRAY(0x126c0aee3cd0),
LT::OSConfig=HASH(0x126c975caac0), ARRAY(0x126c06720070),
LT::Library::Stash=HASH(0x126c06720028), ARRAY(0x126c0aee3f88),
ARRAY(0x126c10a22058), ...) called at
/usr/libdata/perl5/LT/Mode/Link/Program.pm line 28
LT::Program::link(LT::Program=HASH(0x126c329be310),
ARRAY(0x126c0aee3cd0), LT::OSConfig=HASH(0x126c975caac0),
ARRAY(0x126c06720070), LT::Library::Stash=HASH(0x126c06720028),
ARRAY(0x126c0aee3f88), ARRAY(0x126c10a22058), LT::Parser=HASH(0x126c329be598),
...) called at /usr/libdata/perl5/LT/Mode/Link.pm line 235
LT::Mode::Link::run(LT::Mode::Link=HASH(0x126c329beb50),
ARRAY(0x126c0aee3cd0), LT::Options=HASH(0x126c0aee3b20),
LT::OSConfig=HASH(0x126c975caac0)) called at /usr/bin/libtool line 428
it is libtool that generate an error because it didn't found
edbus-private library.
This particular library comes from databases/evolution-data-server :
$ pkg_locate edbus-private
evolution-data-server-3.28.5:databases/evolution-data-server:/usr/local/lib/evolution-data-server/libedbus-private.so
So with ld.lld, libtool is unable to found the library.
I am able to reproduce the problem with just the following command line:
$ libtool --tag=CC --mode=link cc -v -o test -lebook-1.2
warning: could not find a edbus-private library
Link error: edbus-private not found!
...
For what I understood, libtool uses objdump -p libebook-1.2.so to found
the path dependencies (I have ktraced libtool).
On pre-ld.lld system, I have the following output:
$ objdump -p /usr/local/lib/libebook-1.2.so.21.0
...
Dynamic Section:
NEEDED libedata-book-1.2.so.18.0
...
NEEDED libedbus-private.so
...
RPATH /usr/local/lib/evolution-data-server
...
But on post-ld.lld, the output is slightly different:
$ objdump -p /usr/local/lib/libebook-1.2.so.21.0
...
Dynamic Section:
RUNPATH /usr/local/lib/evolution-data-server
NEEDED libedata-book-1.2.so
...
NEEDED libedbus-private.so
...
So ld.lld uses RUNPATH instead of RPATH, and as libtool only uses RPATH
to infer no standard directories... it fail to found the library.
The following diff on libtool permits to productivity/glabels to build.
Index: LT/Library.pm
===
RCS file: /cvs/src/usr.bin/libtool/LT/Library.pm,v
retrieving revision 1.12
diff -u -p -r1.12 Library.pm
--- LT/Library.pm 21 Sep 2015 08:49:06 - 1.12
+++ LT/Library.pm 27 Oct 2018 07:26:13 -
@@ -191,7 +191,7 @@ sub inspect
return @deps;
}
-# give the list of RPATH directories
+# give the list of RPATH/RUNPATH directories
sub findrpaths
{
my $self = shift;
@@ -203,16 +203,20 @@ sub findrpaths
say "warning: library was specified that could not be found:
$self->{key}";
return;
}
- tsay {"inspecting $filename for non standard RPATH..."};
+ tsay {"inspecting $filename for non standard RPATH/RUNPATH..."};
open(my $fh, '-|', "objdump", "-p", "--", $filename);
while (<$fh>) {
if (m/RPATH\s+(.*)$/) {
@dirs = split(":", $1);
last;
}
+ if (m/RUNPATH\s+(.*)$/) {
+ @dirs = split(":", $1);
+ last;
+ }
}
tsay {"found ", (@dirs == 0) ? 'none ' : '',
- "RPATH for $filename\n@dirs"};
+ "RPATH/RUNPATH for $filename\n@dirs"};
return @dirs;
}
Thanks.
--
Sebastien Marie
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: bent...@cvs.openbsd.org 2018/10/27 01:22:47 Modified files: games/valyriatear: Makefile games/valyriatear/pkg: PLIST devel/quirks : Makefile devel/quirks/files: Quirks.pm Log message: Rename valyriatear package to lowercase. The package name used to be lowercase, but inadvertently gained capitals with the 1.0.0 update. ok bcallah@ (MAINTAINER)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: bent...@cvs.openbsd.org 2018/10/27 01:15:11 Modified files: games : Makefile Log message: +pioneer
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: bent...@cvs.openbsd.org 2018/10/27 01:14:27 Log message: Import pioneer-20180203, a game of lonely space and adventure. Pioneer is a space adventure game set in the Milky Way galaxy at the turn of the 31st century. The game is open-ended, and you are free to explore the millions of star systems in the game. You can land on planets, slingshot past gas giants, and burn yourself to a crisp flying between binary star systems. You can try your hand at piracy, make your fortune trading between systems, or do missions for the various factions fighting for power, freedom or self-determination. tweaks + ok kirby@ Status: Vendor Tag: bentley Release Tags: bentley_20181027 N ports/games/pioneer/Makefile N ports/games/pioneer/distinfo N ports/games/pioneer/patches/patch-configure_ac N ports/games/pioneer/patches/patch-contrib_imgui_imgui_draw_cpp N ports/games/pioneer/pkg/DESCR N ports/games/pioneer/pkg/PLIST No conflicts created by this import
Re: [UPDATE] sysutils/glide 0.13.1 -> 0.13.2
ping On 10/2/18 10:15 AM, fredl wrote: Hey, attached diff updates sysutils/glide from 0.13.1 to 0.13.2 release notes: https://github.com/Masterminds/glide/releases/tag/v0.13.2
