CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/08/27 17:12:20 Modified files: security/botan2: Makefile security/botan2/pkg: PFRAG.x86 PLIST Log message: botan2: unbreak on !x86 archs The latest botan2 version provides a new processor_rng feature, but the feature is unavailable on most archs, breaking the packaging at least on aarch64, macppc and sparc64 due to a missing header. Add that header to the x86 PFRAG. OK kmos@ (who tested on sparc64), bluhm@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/08/25 17:56:31 Modified files: biology/bioperl: Makefile distinfo biology/bioperl/pkg: PLIST Removed files: biology/bioperl/patches: patch-Bio_Tools_Alignment_Consed_pm patch-Bio_Tools_SiRNA_Ruleset_tuschl_pm patch-maintenance_symlink_script_pl Log message: bioperl: update to 1.7.7 Beware, bioperl has been split in several submodules, please check an exhaustive list of what it implies at: https://metacpan.org/changes/distribution/BioPerl#L71 OK afresh1@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/08/25 17:54:02 Modified files: biology/p5-Bio-ASN1-EntrezGene: Makefile distinfo biology/p5-Bio-ASN1-EntrezGene/pkg: PLIST Removed files: biology/p5-Bio-ASN1-EntrezGene/patches: patch-lib_Bio_ASN1_EntrezGene_pm patch-lib_Bio_ASN1_Sequence_pm Log message: p5-Bio-ASN1-EntrezGene: update to 1.73 Changelog: https://metacpan.org/changes/distribution/Bio-ASN1-EntrezGene @conflict tweak from sthen OK afresh1@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/08/25 17:44:13 Modified files: biology: Makefile Log message: +p5-Bio-DB-EMBL +p5-Bio-DB-NCBIHelper +p5-Bio-Variation +p5-Bio-Cluster
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/08/25 17:43:06 Log message: Import biology/p5-Bio-Cluster This interface is the basic structure for a cluster of bioperl objects. @conflict tweak from sthen OK afresh1@ Status: Vendor Tag: cwen Release Tags: cwen_20200826 N ports/biology/p5-Bio-Cluster/Makefile N ports/biology/p5-Bio-Cluster/distinfo N ports/biology/p5-Bio-Cluster/pkg/DESCR N ports/biology/p5-Bio-Cluster/pkg/PLIST No conflicts created by this import
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/08/25 17:41:26 Log message: Import biology/p5-Bio-Variation Bio::Variation name space contains modules to store sequence variation information as differences between the reference sequence and changes sequences. Also included are classes to write out and recrete objects from EMBL-like flat files and XML. Lastly, there are simple classes to calculate values for sequence change objects. @conflict tweak from sthen OK afresh1@ Status: Vendor Tag: cwen Release Tags: cwen_20200826 N ports/biology/p5-Bio-Variation/Makefile N ports/biology/p5-Bio-Variation/distinfo N ports/biology/p5-Bio-Variation/pkg/DESCR N ports/biology/p5-Bio-Variation/pkg/PLIST No conflicts created by this import
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/08/25 17:39:36 Log message: Import biology/p5-Bio-DB-NCBIHelper This module provides a single place to setup some common methods for querying NCBI web databases. @conflict tweak from sthen OK afresh1@ Status: Vendor Tag: cwen Release Tags: cwen_20200826 N ports/biology/p5-Bio-DB-NCBIHelper/Makefile N ports/biology/p5-Bio-DB-NCBIHelper/distinfo N ports/biology/p5-Bio-DB-NCBIHelper/pkg/DESCR N ports/biology/p5-Bio-DB-NCBIHelper/pkg/PLIST No conflicts created by this import
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/08/25 17:37:52 Log message: Import biology/p5-Bio-DB-EMBL This module allows the dynamic retrieval of sequence objects Bio::Seq from the EMBL database using the dbfetch script at EBI: https://www.ebi.ac.uk/Tools/dbfetch/dbfetch @conflict tweak from sthen OK afresh1@ Status: Vendor Tag: cwen Release Tags: cwen_20200826 N ports/biology/p5-Bio-DB-EMBL/Makefile N ports/biology/p5-Bio-DB-EMBL/distinfo N ports/biology/p5-Bio-DB-EMBL/pkg/DESCR N ports/biology/p5-Bio-DB-EMBL/pkg/PLIST No conflicts created by this import
Re: 回复: [Update] devel/p5-Moose : Update to 2.2013
Hi,
I imported devel/p5-Devel-OverloadInfo.
- devel/p5-MooseX-Clone: i was suspicious about
"BUILD_DEPENDS=${RUN_DEPENDS}", but this avoids nasty warnings.
OK cwen@
- devel/p5-MooseX-Getopt:
Can't locate Test/Needs.pm in @INC (you may need to install the
Test::Needs module)
- devel/p5-Test-Needs needs to be added to TEST_DEPENDS
- "FIX_EXTRACT_PERMISSIONS = Yes" needs to be set
- devel/p5-Test-Warnings as a TEST_DEPENDS is only needed for
author tests, so it's okay to not add it since we don't run them
OK cwen@ with that fixed.
- devel/p5-Moose (tested on amd64 only): Module::Runtime::Conflicts
is missing, but that's only for distribution stuff we don't care
about in a non-cpan context. OK cwen@
I tested direct consumers for each of them, i see no regression
from them.
On Sun, 23 Aug 2020 08:12:37 +
wen heping wrote:
> Here is the revised patch:
>
> i) Add devel/p5-CPAN-Meta-Check to p5-Moose's TEST_DEPENDS
> ii) Add devel/p5-MooseX-StrictConstructor to p5-MooseX-Getopt's
> TEST_DEPENDS iii) Add CONFIGURE_STYLE = modbuild tiny to
> p5-MooseX-Getopt
>
>It need two new ports: p5-Test-CleanNamespaces had been imported,
> but p5-Devel-OverloadInfo is waiting for one more OK to be imported.
>
>
> Cheers !
> wen
>
> 发件人: Andrew Hewus Fresh
> 发送时间: 2020年7月31日 12:16
> 收件人: wen heping
> 抄送: ports@openbsd.org
> 主题: Re: [Update] devel/p5-Moose : Update to 2.2013
>
> This needs "CONFIGURE_STYLE = modbuild tiny" and a TEST_DEPENDS on
> devel/p5-MooseX-StrictConstructor
>
> Although I notice that tests fail because p5-MooseX-ConfigFromFile is
> missing a RUN_DEPENDS, so look for a change to that incoming.
>
> Other than that OK afresh1@
>
> On Sun, Jul 26, 2020 at 01:28:19PM +, wen heping wrote:
> > Index: devel/p5-MooseX-Getopt/Makefile
> > ===
> > RCS file: /cvs/ports/devel/p5-MooseX-Getopt/Makefile,v
> > retrieving revision 1.13
> > diff -u -p -r1.13 Makefile
> > --- devel/p5-MooseX-Getopt/Makefile 3 Jul 2020 21:45:02
> > - 1.13 +++ devel/p5-MooseX-Getopt/Makefile 26 Jul 2020
> > 13:11:06 - @@ -4,16 +4,16 @@ COMMENT=Moose role for
> > processing comm
> >
> > MODULES= cpan
> > PKG_ARCH=*
> > -DISTNAME=MooseX-Getopt-0.62
> > +DISTNAME=MooseX-Getopt-0.74
> > CATEGORIES= devel
> > -REVISION=1
> >
> > # perl
> > PERMIT_PACKAGE= Yes
> >
> > -RUN_DEPENDS= devel/p5-Getopt-Long-Descriptive \
> > +RUN_DEPENDS= devel/p5-Getopt-Long-Descriptive>=0.088 \
> >devel/p5-Moose \
> > - devel/p5-MooseX-Role-Parameterized>=1.02
> > + devel/p5-MooseX-Role-Parameterized>=1.01 \
> > + devel/p5-Try-Tiny
> > TEST_DEPENDS=devel/p5-Class-Load \
> >devel/p5-Devel-GlobalDestruction \
> >devel/p5-Module-Runtime \
> > Index: devel/p5-MooseX-Getopt/distinfo
> > ===
> > RCS file: /cvs/ports/devel/p5-MooseX-Getopt/distinfo,v
> > retrieving revision 1.3
> > diff -u -p -r1.3 distinfo
> > --- devel/p5-MooseX-Getopt/distinfo 22 Mar 2014 14:03:20
> > - 1.3 +++ devel/p5-MooseX-Getopt/distinfo 26 Jul 2020
> > 13:11:06 - @@ -1,2 +1,2 @@
> > -SHA256 (MooseX-Getopt-0.62.tar.gz) =
> > y+i+aOxe1hGZYPVhVcui8F1RuBPSmNDLEjllnsPY5xA= -SIZE
> > (MooseX-Getopt-0.62.tar.gz) = 48289 +SHA256
> > (MooseX-Getopt-0.74.tar.gz) =
> > HeDfO0Mevp81Y730Vp6s1+B+hlqDl/KpkNDLV9TLLCQ= +SIZE
> > (MooseX-Getopt-0.74.tar.gz) = 58129
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/08/23 07:28:15 Modified files: devel : Makefile Log message: +p5-Devel-OverloadInfo
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/08/23 07:27:13 Log message: Import devel/p5-Devel-OverloadInfo Devel::OverloadInfo returns information about overloaded operators for a given class (or object), including where in the inheritance hierarchy the overloads are declared and where the code implementing it is. From Wen Heping who takes maintainer Tweak and OK afresh1@ Status: Vendor Tag: cwen Release Tags: cwen_20200823 N ports/devel/p5-Devel-OverloadInfo/Makefile N ports/devel/p5-Devel-OverloadInfo/distinfo N ports/devel/p5-Devel-OverloadInfo/pkg/DESCR N ports/devel/p5-Devel-OverloadInfo/pkg/PLIST No conflicts created by this import
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/08/23 05:46:37 Modified files: devel/geany: Makefile Added files: devel/geany/patches: patch-scintilla_src_RunStyles_cxx Log message: geany: fix the build on powerpc (and probably i386) There were various undefined references errors at linking time on these archs, define the missing symbols also on all 32-bit archs (with hints from sthen@, thanks!). While here, move HOMEPAGE to https.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/08/23 05:20:33 Modified files: games/gnushogi : Makefile Log message: gnushogi: mark BROKEN on powerpc Since the switch to clang-10, gnushogi itself hangs at runtime during the build.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/08/23 04:56:39 Modified files: emulators/advancemame: Makefile Log message: advancemame: mark BROKEN on powerpc Since the switch to clang-10, the build hangs indefinitely.
Re: [NEW] www/p5-Catalyst-Controller-ActionRole
On Sun, 17 May 2020 17:05:44 -0700 Andrew Hewus Fresh wrote: > On Sun, May 03, 2020 at 01:38:41AM +, wen heping wrote: > > Hi, ports@: > > > > Here is a patch to create new port > > www/p5-Catalyst-Controller-ActionRole. Although it is marked > > DEPRECATED upstream, it is needed by > > www/p5-Catalyst-ActionRole-ACL, which is needed by the update of > > devel/catalyst. It build well and pass all tests on amd64-current > > system. > > OK afresh1@ if someone could import or provide a second OK cwen@, but the devel/p5-Moose TEST_DEPENDS is redundant. > > > > Regards, > > wen > > > > > > > > -- > andrew - http://afresh1.com > > Hey! It compiles! Ship it! >
Re: [NEW] www/p5-URI-ws
On Fri, 21 Aug 2020 07:54:20 + wen heping wrote: > Hi, ports@: > >Here is a patch to create new port www/p5-URI-ws, which is required > by the update of p5-Catalyst-Runtime. >It build well and pass all tests on amd64-current system. > > Cheers ! > wen OK cwen@ if someone is willing to import it.
Re: 回复: [NEW] www/p5-CGI-Struct
On Fri, 21 Aug 2020 00:19:44 + wen heping wrote: > ping ... I've found the XS version (i had to tweak WANTLIB), so i committed the 2 of them. Thanks! > > 发件人: Andrew Hewus Fresh > 发送时间: 2020年8月16日 4:19 > 收件人: wen heping > 抄送: ports@openbsd.org > 主题: Re: [NEW] www/p5-CGI-Struct > > On Mon, Jul 27, 2020 at 01:12:01PM +, wen heping wrote: > > Hi, ports@: > > > >Here is a patch to create new port www/p5-CGI-Struct, > > which is required by the update of p5-Catalyst-Runtime. > >It build well and pass all tests on amd64-current system. > > I said OK to both this and p5-CGI-Struct-XS that were in the same > message back in May, but apparently that mail got lost somewhere along > the way as I don't see it on MARC or nabble (but I don't see the > original on MARC either). > > http://openbsd-archive.7691.n7.nabble.com/NEW-www-p5-CGI-Struct-td395846.html > > In any case, OK to this or on both. > > l8rZ, > -- > andrew - http://afresh1.com > > Hey! It compiles! Ship it!
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/08/21 05:17:00 Modified files: www: Makefile Log message: +p5-CGI-Struct +p5-CGI-Struct-XS
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/08/21 05:15:31 Log message: Import www/p5-CGI-Struct CGI::Struct lets you transform CGI data keys that look like perl data structures into actual perl data structures. From Wen Heping OK afresh1@ Status: Vendor Tag: cwen Release Tags: cwen_20200821 N ports/www/p5-CGI-Struct/Makefile N ports/www/p5-CGI-Struct/distinfo N ports/www/p5-CGI-Struct/pkg/DESCR N ports/www/p5-CGI-Struct/pkg/PLIST No conflicts created by this import
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/08/21 05:14:00 Log message: Import www/p5-CGI-Struct-XS This module is XS implementation of CGI::Struct. It's fully compatible with CGI::Struct, except for error messages. CGI::Struct::XS is 3-15(5-25 with dclone disabled) times faster than original module. From Wen Heping , with a WANTLIB tweak from me. OK afresh1@ Status: Vendor Tag: cwen Release Tags: cwen_20200821 N ports/www/p5-CGI-Struct-XS/Makefile N ports/www/p5-CGI-Struct-XS/distinfo N ports/www/p5-CGI-Struct-XS/pkg/DESCR N ports/www/p5-CGI-Struct-XS/pkg/PLIST No conflicts created by this import
[update] biology/bioperl (and related ports)
Hi,
Our current bioperl related ports are old, and we had to patch
them already to deal with Perl changes. So i'm proposing an update here
to 1.7.7 to avoid dealing with emergency fixes at a large scale.
These updates and new ports will not totally replace what bioperl used
to provide, see [0] for an exhaustive list of what it implies.
There are backward incompatibilities, and a lot of inter dependencies,
so you need all these ports at once, since bioperl has been split in
many modules. This also means that conflict markers had to be sprinkled
around, as hinted by update-plist(1).
Updates are in the diff below, new ports are attached.
biology/bioperl => PASS
biology/p5-Bio-ASN1-EntrezGene => PASS
biology/p5-Bio-Cluster => NEW, PASS
biology/p5-Bio-Variation=> NEW, PASS
biology/p5-Bio-DB-NCBIHelper=> NEW, PASS, NETWORK
biology/p5-Bio-DB-EMBL => NEW, PASS, NETWORK
(NETWORK ones requires network for testing if you use PORTS_PRIVSEP)
There are no other ports depending on bioperl.
Comments/feedback are welcome,
Charlène.
[0] https://metacpan.org/changes/distribution/BioPerl#L71
Index: bioperl/Makefile
===
RCS file: /cvs/ports/biology/bioperl/Makefile,v
retrieving revision 1.21
diff -u -p -r1.21 Makefile
--- bioperl/Makefile12 Jul 2019 20:43:43 - 1.21
+++ bioperl/Makefile20 Aug 2020 10:11:28 -
@@ -2,68 +2,51 @@
COMMENT= perl tools for bioinformatics
-DISTNAME= BioPerl-1.6.924
-REVISION= 1
+DISTNAME= BioPerl-1.7.7
PKGNAME= ${DISTNAME:L}# No p5-prefix here.
CATEGORIES=biology
HOMEPAGE= https://bioperl.org/
-# perl
+# Perl
PERMIT_PACKAGE=Yes
-CPAN_AUTHOR= CJFIELDS
-
MODULES= cpan
-PKG_ARCH= *
+
+CPAN_AUTHOR= CJFIELDS
MODCPAN_EXAMPLES= Yes
MODCPAN_EXAMPLES_DIR= ${PREFIX}/share/examples/bioperl/
-CONFIGURE_STYLE= modbuild
+RUN_DEPENDS= devel/p5-Data-Stag \
+ devel/p5-Error \
+ devel/p5-Graph \
+ devel/p5-IO-String \
+ devel/p5-IO-stringy \
+ devel/p5-IPC-Run \
+ devel/p5-List-MoreUtils \
+ devel/p5-Module-Build \
+ devel/p5-Set-Scalar \
+ devel/p5-Test-Most \
+ devel/p5-Test-RequiresInternet \
+ devel/p5-YAML \
+ textproc/p5-Libxml \
+ textproc/p5-XML-DOM \
+ textproc/p5-XML-DOM-XPath \
+ textproc/p5-XML-LibXML \
+ textproc/p5-XML-SAX \
+ textproc/p5-XML-SAX-Base \
+ textproc/p5-XML-SAX-Writer \
+ textproc/p5-XML-Twig \
+ textproc/p5-XML-Writer \
+ www/p5-HTTP-Message \
+ www/p5-URI \
+ www/p5-libwww
+TEST_DEPENDS= devel/p5-Test-Memory-Cycle \
+ devel/p5-Test-Weaken
-BUILD_DEPENDS= ${RUN_DEPENDS}
-RUN_DEPENDS= biology/AcePerl \
- databases/p5-DBD-mysql \
- converters/p5-Convert-Binary-C \
- devel/p5-Algorithm-Munkres \
- devel/p5-Array-Compare \
- devel/p5-Class-AutoClass \
- devel/p5-Clone \
- devel/p5-Data-Stag \
- devel/p5-Graph \
- devel/p5-IO-String \
- devel/p5-IO-stringy \
- devel/p5-Set-Scalar \
- devel/p5-Test-Exception \
- devel/p5-Tree-DAG_Node \
- graphics/p5-GD \
- graphics/p5-GD-SVG \
- graphics/p5-SVG-Graph \
- math/p5-GraphViz \
- misc/p5-Spreadsheet-ParseExcel \
- misc/p5-Spreadsheet-WriteExcel \
- textproc/p5-XML-DOM \
- textproc/p5-Libxml \
- textproc/p5-XML-DOM-XPath \
- textproc/p5-XML-Parser \
- textproc/p5-XML-SAX-Writer \
- textproc/p5-XML-Simple \
- textproc/p5-XML-Twig \
- textproc/p5-XML-Writer \
- www/p5-CGI \
- www/p5-HTML-Parser \
- www/p5-SOAP-Lite \
- www/p5-URI \
- www/p5-libwww
-TEST_DEPENDS= devel/p5-Test-Most \
- www/p5-HTML-TableExtract
-
-MAKE_ENV +=TEST_POD=Yes
-
-# Optional (not ported) dependencies:
-# Postscript-TextBlock
-# Tie-Cacher
-CONFIGURE_ENV+=PERL_MM_USE_DEFAULT=Yes
+MAKE_ENV +=TEST_POD=Yes
+
+PKG_ARCH= *
.include
Index: bioperl/distinfo
===
RCS file:
Re: 回复: 回复: [NEW] www/p5-Catalyst-Action-REST
It's imported, thanks! As usual i forgot to mention it's coming from you in the commit message, sorry. I'm looking forward for the Catalyst updates :) On Wed, 19 Aug 2020 13:37:20 + wen heping wrote: > ping again , which is required by the update of p5-Catalyst-* ports. > > 发件人: owner-po...@openbsd.org 代表 wen > heping 发送时间: 2020年8月13日 16:15 > 收件人: afre...@openbsd.org > 抄送: ports@openbsd.org > 主题: 回复: 回复: [NEW] www/p5-Catalyst-Action-REST > > ping ... > > 发件人: owner-po...@openbsd.org 代表 wen > heping 发送时间: 2020年7月28日 19:36 > 收件人: afre...@openbsd.org > 抄送: ports@openbsd.org > 主题: 回复: 回复: [NEW] www/p5-Catalyst-Action-REST > > I rework the update of p5-Catalyst-* ports, this new port > is required, would someone have a look of it ? > > Thanks ! > wen > > 发件人: wen heping > 发送时间: 2020年5月2日 7:23 > 收件人: afre...@openbsd.org > 抄送: ports@openbsd.org > 主题: 回复: [NEW] www/p5-Catalyst-Action-REST > > ping ... > > 发件人: Andrew Hewus Fresh > 发送时间: 2020年1月27日 5:05 > 收件人: wen heping > 抄送: ports@openbsd.org > 主题: Re: [NEW] www/p5-Catalyst-Action-REST > > On Fri, Dec 20, 2019 at 06:37:07AM +, wen heping wrote: > > Hi, ports@: > > > >Here is a patch to create www/p5-Catalyst-Action-REST, > > which is required by the future update of www/p5-Catalyst-*. > > It build well and pass all tests on amd64-current system. > > > > Comments? OK? > > wen > > OK afresh1@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/08/19 08:09:04 Modified files: www: Makefile Log message: +p5-Catalyst-Action-REST
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/08/19 08:06:51 Log message: Import www/p5-Catalyst-Action-REST This Action handles doing automatic method dispatching for REST requests. It takes a normal Catalyst action, and changes the dispatch to append an underscore and method name. First it will try dispatching to an action with the generated name, and failing that it will try to dispatch to a regular method. OK afresh1@ Status: Vendor Tag: cwen Release Tags: cwen_20200819 N ports/www/p5-Catalyst-Action-REST/Makefile N ports/www/p5-Catalyst-Action-REST/distinfo N ports/www/p5-Catalyst-Action-REST/pkg/DESCR N ports/www/p5-Catalyst-Action-REST/pkg/PLIST No conflicts created by this import
[macppc] Mark emulators/advancemame BROKEN
Hi, As seen on the current macppc bulk: > emulators/advancemame(build) [55003] on macppc-1 4% frozen for 32 > HOURS! I would like to see it marked BROKEN, since no usual suspect is present here (and we don't distribute the package anyway). I'm sharing the log [0] if someone is interested. OK? Charlène. [0] https://bin.charlenew.xyz/advancemame.log Index: Makefile === RCS file: /cvs/ports/emulators/advancemame/Makefile,v retrieving revision 1.17 diff -u -p -u -p -r1.17 Makefile --- Makefile15 Apr 2020 13:31:42 - 1.17 +++ Makefile18 Aug 2020 17:59:28 - @@ -1,5 +1,7 @@ # $OpenBSD: Makefile,v 1.17 2020/04/15 13:31:42 fcambus Exp $ +BROKEN-powerpc = clang-10 hangs indefinitely + COMMENT = MAME port with advanced video support for monitors and TVs V =3.9
Re: [macppc] Unbreak devel/geany
On Thu, 11 Jun 2020 15:15:34 +0200
Charlene Wendling wrote:
> Hi,
>
> > http://build-failures.rhaalovely.net/powerpc/2020-05-09/devel/geany.log
>
> It's a wild guess, but i think it wants a template with a long
> argument because on macppc:
>
> $ c++ -dM -E - < /dev/null | grep PTRDIFF_MAX
> #define __PTRDIFF_MAX__ 2147483647L
> $ c++ -dM -E - < /dev/null | grep INT_MAX
> #define __INT_MAX__ 2147483647
>
> Either way, the below diff fixes the build on macppc, the runtime is
> fine [0] as well. While here i've moved HOMEPAGE to https.
>
> [...}
>
> Charlène.
>
>
> [0] https://bsd.network/@julianaito/104325481078218916
>
>
Since i386 is impacted as well, here is a variant heavily hinted by
sthen@. This builds fine on macppc.
Index: Makefile
===
RCS file: /cvs/ports/devel/geany/Makefile,v
retrieving revision 1.63
diff -u -p -u -p -r1.63 Makefile
--- Makefile16 Oct 2019 06:51:42 - 1.63
+++ Makefile18 Aug 2020 17:49:13 -
@@ -3,11 +3,12 @@
COMMENT= small and lightweight IDE
DISTNAME = geany-1.36
+REVISION = 0
SHARED_LIBS += geany 0.0 # 0.0
CATEGORIES=devel
-HOMEPAGE= http://www.geany.org/
+HOMEPAGE= https://www.geany.org/
MAINTAINER=Victor Kukshiev
Index: patches/patch-scintilla_src_RunStyles_cxx
===
RCS file: patches/patch-scintilla_src_RunStyles_cxx
diff -N patches/patch-scintilla_src_RunStyles_cxx
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-scintilla_src_RunStyles_cxx 18 Aug 2020 17:49:13 -
@@ -0,0 +1,18 @@
+$OpenBSD$
+
+i386/powerpc fix for
+"undefined reference to Scintilla::RunStyles::RunStyles()"
+
+Index: scintilla/src/RunStyles.cxx
+--- scintilla/src/RunStyles.cxx.orig
scintilla/src/RunStyles.cxx
+@@ -308,7 +308,8 @@ void RunStyles::Check() const {
+
+ template class Scintilla::RunStyles;
+ template class Scintilla::RunStyles;
+-#if (PTRDIFF_MAX != INT_MAX) || PLAT_HAIKU
++#if (PTRDIFF_MAX != INT_MAX) || PLAT_HAIKU || \
++( defined(__OpenBSD__) && defined(_ILP32) )
+ template class Scintilla::RunStyles;
+ template class Scintilla::RunStyles;
+ #endif
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/08/18 11:39:30 Modified files: games/godot: Makefile Log message: godot: unbreak on powerpc and i386 Don't build with `-Werror'; this "fixes" the "large atomic operation may incur significant performance penalty" warning upgraded to error seen on these archs Thanks to Omar Polo build testing on i386. OK thfr@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/08/17 14:19:49 Modified files: sysutils/neofetch: Makefile distinfo sysutils/neofetch/patches: patch-neofetch Log message: neofetch: update to 7.1.0 Changelog: https://github.com/dylanaraps/neofetch/releases/tag/7.1.0 Also, add a patch to fix the cpu temperature and add ksmn(4) and adt(4) support OK bket@
Re: [macppc, i386] Unbreak games/godot
On Mon, 17 Aug 2020 20:18:55 +0200 Omar Polo wrote: > > Thomas Frohwein writes: > > > On Thu, Aug 13, 2020 at 10:09:41PM +0200, Charlene Wendling wrote: > >> Hi, > >> > >> Godot does not build in the current macppc bulk with, as seen on > >> i386: > >> > >> > ./core/safe_refcount.h:139:7: error: large atomic operation may > >> > incur significant performance penalty > >> > [-Werror,-Watomic-alignment] > >> > >> This warning is far from being a rare sight on macppc, with no > >> noticeable impact. > >> > >> This is happening because development C/CXXFLAGS are provided with > >> "dev=yes". This build option overrides the "werror" one [0], so > >> "werror=no" won't do the job. > >> > >> This builds fine on macppc. > > > > Still builds fine on amd64. I'm ok thfr@ with this diff. Curious if > > this fixes i386, too. > > Out of curiosity, I tried this patch on a i686 I have at home. It > compiles fine, but I cannot run it because my "video card driver does > not support any of the supported OpenGL versions." Thanks for the feedback! Godot requires OpenGL>=2; your card only supports 1.4. > nadeshiko$ dmesg | grep intel > inteldrm0 at pci0 dev 2 function 0 "Intel 82945GM Video" rev 0x03 > drm0 at inteldrm0 > intagp0 at inteldrm0 > inteldrm0: apic 1 int 16, I945GM, gen 3 > inteldrm0: 1024x768, 32bpp > wsdisplay0 at inteldrm0 mux 1: console (std, vt100 emulation), using > wskbd0 >
[update] sysutils/neofetch 7.0.0 -> 7.1.0
Hi,
Here is an update for neofetch to 7.1.0. There are various improvements
with this version [0], noticeably this includes the OpenBSD flavor used.
Port-wise, it's a simple update, but i added improvements that are yet
to be committed [1]:
- fix the cpu temperature and add ksmn(4) and adt(4) support
- make glxinfo(1) call faster; you probably won't notice
it on modern machines
It runs fine on all my machines.
OK?
Charlène.
[0] https://github.com/dylanaraps/neofetch/releases/tag/7.1.0
[1] https://github.com/dylanaraps/neofetch/pull/1494
Index: Makefile
===
RCS file: /cvs/ports/sysutils/neofetch/Makefile,v
retrieving revision 1.8
diff -u -p -u -p -r1.8 Makefile
--- Makefile14 May 2020 17:10:20 - 1.8
+++ Makefile16 Aug 2020 16:43:21 -
@@ -5,7 +5,7 @@ CATEGORIES =sysutils misc
GH_ACCOUNT = dylanaraps
GH_PROJECT = neofetch
-GH_TAGNAME = 7.0.0
+GH_TAGNAME = 7.1.0
MAINTAINER = Charlene Wendling
Index: distinfo
===
RCS file: /cvs/ports/sysutils/neofetch/distinfo,v
retrieving revision 1.4
diff -u -p -u -p -r1.4 distinfo
--- distinfo14 May 2020 17:10:20 - 1.4
+++ distinfo16 Aug 2020 16:43:21 -
@@ -1,2 +1,2 @@
-SHA256 (neofetch-7.0.0.tar.gz) = jGvSF89tNPwfPcuw6LETdlW8E/uyEWUnPbsqe84NMTA=
-SIZE (neofetch-7.0.0.tar.gz) = 90892
+SHA256 (neofetch-7.1.0.tar.gz) = WKlea3FOQe/IBOyjiaIjMJFpst7zXlf6k0SCprR8J+c=
+SIZE (neofetch-7.1.0.tar.gz) = 95400
Index: patches/patch-neofetch
===
RCS file: /cvs/ports/sysutils/neofetch/patches/patch-neofetch,v
retrieving revision 1.6
diff -u -p -u -p -r1.6 patch-neofetch
--- patches/patch-neofetch 14 May 2020 17:10:20 - 1.6
+++ patches/patch-neofetch 16 Aug 2020 16:43:21 -
@@ -1,20 +1,26 @@
$OpenBSD: patch-neofetch,v 1.6 2020/05/14 17:10:20 cwen Exp $
-From https://github.com/dylanaraps/neofetch/pull/1426
-Can be removed >7.0.0
+From https://github.com/dylanaraps/neofetch/pull/1494
Index: neofetch
--- neofetch.orig
+++ neofetch
-@@ -1032,6 +1032,11 @@ get_distro() {
- *) distro="Guix System $(guix system -V | awk
'NR==1{printf $5}')"
- esac
+@@ -2196,8 +2196,7 @@ get_cpu() {
+ ;;
+ "OpenBSD"* | "Bitrig"*)
+ deg="$(sysctl hw.sensors | \
+- awk -F '=| degC' '/lm0.temp|cpu0.temp/ {print $2;
exit}')"
+-deg="${deg/00/0}"
++awk -F'=|degC' '/(ksmn|adt|lm|cpu)0.temp0/
{printf("%2.1f", $2); exit}')"
+ ;;
+ esac
+ ;;
+@@ -2520,7 +2519,7 @@ get_gpu() {
+ ;;
-+# Display whether using '-current' or '-release' on OpenBSD.
-+elif [[ $kernel_name = OpenBSD ]] ; then
-+read -ra kernel_info <<< "$(sysctl -n kern.version)"
-+distro=${kernel_info[*]:0:2}
-+
- else
- for release_file in /etc/*-release; do
- distro+=$(< "$release_file")
+ *)
+-gpu="$(glxinfo | grep -F 'OpenGL renderer string')"
++gpu="$(glxinfo -B | grep -F 'OpenGL renderer string')"
+ gpu="${gpu/OpenGL renderer string: }"
+ ;;
+ esac
[macppc, i386] Unbreak games/godot
Hi,
Godot does not build in the current macppc bulk with, as seen on i386:
> ./core/safe_refcount.h:139:7: error: large atomic operation may incur
> significant performance penalty [-Werror,-Watomic-alignment]
This warning is far from being a rare sight on macppc, with no
noticeable impact.
This is happening because development C/CXXFLAGS are provided with
"dev=yes". This build option overrides the "werror" one [0], so
"werror=no" won't do the job.
This builds fine on macppc.
Comments/feedback are welcome,
Charlène.
[0] https://github.com/godotengine/godot/blob/master/SConstruct#L134
Index: Makefile
===
RCS file: /cvs/ports/games/godot/Makefile,v
retrieving revision 1.13
diff -u -p -u -p -r1.13 Makefile
--- Makefile3 Aug 2020 18:41:08 - 1.13
+++ Makefile13 Aug 2020 20:06:13 -
@@ -1,6 +1,5 @@
# $OpenBSD: Makefile,v 1.13 2020/08/03 18:41:08 thfr Exp $
-BROKEN-i386 = core/safe_refcount.h error: misaligned atomic operation may
incur significant performance penalty
COMMENT = 2D and 3D game engine
V =3.2.2
@@ -50,11 +49,13 @@ MODSCONS_FLAGS =CC="${CC}" \
builtin_pcre2=no \
builtin_zlib=no \
builtin_zstd=no \
- dev=yes \
platform=x11 \
progress=no \
pulseaudio=no \
- target=release_debug
+ target=release_debug \
+ verbose=yes \
+ warnings=extra \
+ werror=no
LIB_DEPENDS = archivers/zstd \
audio/libvorbis \
audio/musepack \
Re: i386/LLVM 10
On Sat, 08 Aug 2020 23:16:09 +0200
Jeremie Courreges-Anglas wrote:
>
> < sthen> ld.lld:
> error: /pobj/webkitgtk4-2.28.4/build-i386/lib/libjavascriptcoregtk-4.0.so:
> undefined reference to WTF::Optional
> JSC::parse(char const*)
>
> cwen did hit this too on powerpc when it was switched to clang,
> I propose to extend the workaround to all archs. Disclaimer: diff not
> tested on i386, no build regression on amd64.
I've been able to build webkitgtk4 without issues on powerpc with that
diff.
OK cwen@ if i386 survives to this :)
> Index: Makefile
> ===
> RCS file: /cvs/ports/www/webkitgtk4/Makefile,v
> retrieving revision 1.125
> diff -u -p -r1.125 Makefile
> --- Makefile 7 Aug 2020 14:30:10 - 1.125
> +++ Makefile 8 Aug 2020 21:09:58 -
> @@ -16,7 +16,7 @@ COMMENT = GTK+ port of the WebKit
> rende V = 2.28.4
> DISTNAME = webkitgtk-${V}
> PKGNAME =webkitgtk4-${V}
> -REVISION = 0
> +REVISION = 1
> EXTRACT_SUFX = .tar.xz
> API =4.0
> SUBST_VARS = API
> Index: patches/patch-Source_JavaScriptCore_runtime_Options_cpp
> ===
> RCS
> file:
> /cvs/ports/www/webkitgtk4/patches/patch-Source_JavaScriptCore_runtime_Options_cpp,v
> retrieving revision 1.3 diff -u -p -r1.3
> patch-Source_JavaScriptCore_runtime_Options_cpp
> --- patches/patch-Source_JavaScriptCore_runtime_Options_cpp 18
> Apr 2020 16:14:28 - 1.3 +++
> patches/patch-Source_JavaScriptCore_runtime_Options_cpp 8 Aug
> 2020 21:09:58 - @@ -1,6 +1,6 @@ $OpenBSD:
> patch-Source_JavaScriptCore_runtime_Options_cpp,v 1.3 2020/04/18
> 16:14:28 cwen Exp $ -powerpc fix for:
> +powerpc and i386 fix for:
> undefined reference to `WTF::Optional
> JSC::parse(char const*)'
>
> @@ -12,7 +12,7 @@ Index: Source/JavaScriptCore/runtime/Opt
> }
>
> -#if CPU(ADDRESS64) || OS(DARWIN)
> -+#if CPU(ADDRESS64) || OS(DARWIN) || ( defined(__OpenBSD__) &&
> defined(__powerpc__) ) ++#if CPU(ADDRESS64) || OS(DARWIN) ||
> OS(OPENBSD) template<>
> Optional parse(const char* string)
> {
> @@ -21,7 +21,7 @@ Index: Source/JavaScriptCore/runtime/Opt
> return WTF::nullopt;
> }
> -#endif // CPU(ADDRESS64) || OS(DARWIN)
> -+#endif // CPU(ADDRESS64) || OS(DARWIN) || ( defined(__OpenBSD__) &&
> defined(__powerpc__) ) ++#endif // CPU(ADDRESS64) || OS(DARWIN) ||
> OS(OPENBSD)
> template<>
> Optional parse(const char* string)
>
>
> --
> jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF DDCC 0DFA 74AE
> 1524 E7EE
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/08/05 15:07:04 Modified files: emulators/nono : Makefile Log message: nono: don't build on big endian archs, BE support is not finished yet OK bcallah@ kmos@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/08/02 08:06:14 Modified files: devel/dyncall : Makefile devel/dyncall/patches: patch-dyncall_dyncall_call_ppc32_S patch-dyncallback_dyncall_callback_ppc32_S Log message: dyncall: fix the build on powerpc without '-fno-integrated-as' Drop the use of '.machine ppc' in assembly code, since it's not supported by clang's integrated assembler and causes no ill effects to do so (hint from bcallah@ a few monthes ago, thanks!). "Fine with me" jasper@
[macppc] Unbreak devel/dyncall
Hi,
In the current macppc bulk, devel/dyncall fails with:
> /usr/obj/ports/dyncall-1.1/dyncall-1.1/dyncall/dyncall_call_ppc32.S:
> 36:11: error: unrecognized machine type .machine ppc
This is happening since the powerpc{32,64} fixes [0] that solves one
part of the issues that made me use '-fno-integrated-as' on powerpc,
but there was still this one.
As pointed out by bcallah@ out of list back then, we can drop this
instruction without bad consequences.
With the below diff, it builds fine on macppc.
Comments/feedback are welcome,
Charlène.
[0] https://marc.info/?l=openbsd-ports-cvs=159550568211055=2
Index: Makefile
===
RCS file: /cvs/ports/devel/dyncall/Makefile,v
retrieving revision 1.3
diff -u -p -u -p -r1.3 Makefile
--- Makefile23 Jul 2020 12:01:19 - 1.3
+++ Makefile2 Aug 2020 12:48:47 -
@@ -3,7 +3,7 @@
COMMENT= arch-independent interface for dynamic calls of C functions
V= 1.1
-REVISION= 0
+REVISION= 1
DISTNAME= dyncall-${V}
CATEGORIES=devel
Index: patches/patch-dyncall_dyncall_call_ppc32_S
===
RCS file: /cvs/ports/devel/dyncall/patches/patch-dyncall_dyncall_call_ppc32_S,v
retrieving revision 1.1
diff -u -p -u -p -r1.1 patch-dyncall_dyncall_call_ppc32_S
--- patches/patch-dyncall_dyncall_call_ppc32_S 23 Jul 2020 12:01:19 -
1.1
+++ patches/patch-dyncall_dyncall_call_ppc32_S 2 Aug 2020 12:48:47 -
@@ -1,11 +1,22 @@
$OpenBSD: patch-dyncall_dyncall_call_ppc32_S,v 1.1 2020/07/23 12:01:19 jasper
Exp $
+Hunk #1: 'machine ppc' is not supported by clang's integrated assembler
+ on powerpc
+
https://dyncall.org/pub/dyncall/dyncall/rev/d982a00c2177
Index: dyncall/dyncall_call_ppc32.S
--- dyncall/dyncall_call_ppc32.S.orig
+++ dyncall/dyncall_call_ppc32.S
-@@ -100,7 +100,7 @@ ENTRY_C(dcCall_ppc32_darwin)
+@@ -33,7 +33,6 @@
+
+ */
+
+- .machine ppc
+ .text
+
+ /*
+@@ -100,7 +99,7 @@ ENTRY_C(dcCall_ppc32_darwin)
srwi r5,r5,2/* r5 = size in words */
@@ -14,7 +25,7 @@ Index: dyncall/dyncall_call_ppc32.S
beq cr0,.osx_done
mtctr r5/* copy loop */
-@@ -214,7 +214,7 @@ ENTRY_C(dcCall_ppc32_sysv)
+@@ -214,7 +213,7 @@ ENTRY_C(dcCall_ppc32_sysv)
srwi r5,r5,2/* r5 = size in words */
Index: patches/patch-dyncallback_dyncall_callback_ppc32_S
===
RCS file:
/cvs/ports/devel/dyncall/patches/patch-dyncallback_dyncall_callback_ppc32_S,v
retrieving revision 1.1
diff -u -p -u -p -r1.1 patch-dyncallback_dyncall_callback_ppc32_S
--- patches/patch-dyncallback_dyncall_callback_ppc32_S 23 Jul 2020 12:01:19
- 1.1
+++ patches/patch-dyncallback_dyncall_callback_ppc32_S 2 Aug 2020 12:48:47
-
@@ -1,11 +1,22 @@
$OpenBSD: patch-dyncallback_dyncall_callback_ppc32_S,v 1.1 2020/07/23 12:01:19
jasper Exp $
+Hunk #1: 'machine ppc' is not supported by clang's integrated assembler
+ on powerpc
+
https://dyncall.org/pub/dyncall/dyncall/rev/d982a00c2177
Index: dyncallback/dyncall_callback_ppc32.S
--- dyncallback/dyncall_callback_ppc32.S.orig
+++ dyncallback/dyncall_callback_ppc32.S
-@@ -123,9 +123,9 @@ Input:
+@@ -25,7 +25,6 @@
+
+ #include "../portasm/portasm-ppc.S"
+
+-.machine ppc
+ .text
+ .align 2
+
+@@ -123,9 +122,9 @@ Input:
mtctr r12
bctrl
/* check result type */
[BE_ARCHS] don't build emulators/nono
Hi,
> http://build-failures.rhaalovely.net/powerpc/2020-07-15/emulators/nono.log
> http://build-failures.rhaalovely.net/mips64/2020-07-11/emulators/nono.log
(it fails on sparc64 because ports-gcc is not used)
$WRKSRC/vm/bitmap.h includes $WRKSRC/lib/device.h, where HLB and HLW
are only defined on little endian archs. $WRKSRC/vm/bitmap.cpp
expect them and has no fallback code.
As such, big endian support is not complete, so i'm proposing to
mark it NOT_FOR_ARCHS.
Comments/feedback are welcome,
Charlène.
Index: Makefile
===
RCS file: /cvs/ports/emulators/nono/Makefile,v
retrieving revision 1.3
diff -u -p -u -p -r1.3 Makefile
--- Makefile22 Jul 2020 14:53:59 - 1.3
+++ Makefile31 Jul 2020 20:04:14 -
@@ -1,5 +1,8 @@
# $OpenBSD: Makefile,v 1.3 2020/07/22 14:53:59 gonzalo Exp $
+# Big endian support is not complete as of 0.1.0
+NOT_FOR_ARCHS= ${BE_ARCHS}
+
COMMENT= LUNA-I emulator
DISTNAME= nono-0.1.0
Re: [sparc64/ld.bfd] Fix build for graphics/mscgen
Hi,
On Sun, 26 Jul 2020 15:30:15 -0400
Kurt Mosiejczuk wrote:
> On Sat, Jul 25, 2020 at 03:57:06PM -0400, Kurt Mosiejczuk wrote:
> > The following fixes the build of graphics/mscgen for sparc64 and
> > presumably other ld.bfd architectures.
>
> > ok?
>
> Brad Smith pointed out I should be using X11BASE instead of hardcoding
> the path. Here's a version that does it right.
>
> ok?
You may want this instead, in any case OK cwen@
> --Kurt
>
Index: Makefile
===
RCS file: /cvs/ports/graphics/mscgen/Makefile,v
retrieving revision 1.8
diff -u -p -u -p -r1.8 Makefile
--- Makefile12 Jul 2019 20:47:04 - 1.8
+++ Makefile26 Jul 2020 19:49:16 -
@@ -27,6 +27,12 @@ FAKE_FLAGS= exampledir="${PREFIX}/share
WRKDIST= ${WRKDIR}/${DISTNAME:S/-src//}
+# Fix undefined references to freetype and fontconfig symbols
+.include
+.if !${PROPERTIES:Mlld}
+CONFIGURE_ENV += LDFLAGS="${LDFLAGS} -L${X11BASE}/lib"
+.endif
+
pre-configure:
@sed -i 's,/bin/bash,/bin/sh,g' ${WRKSRC}/test/renderercheck.sh
Re: NEW: x11/elementary/screenshot
Hi, On Sun, 05 Jul 2020 17:51:48 + Brian Callahan wrote: > Hi ports -- > > Now that I have a decently well working elementary OS desktop running > (a picture is here [0] for the curious), I need to start pushing all > these ports. I won't flood the lists, but it is 40+ new ports and the > revival of some deleted ports. > > First up is elementary-screenshot, a screenshot tool. > > --- > pkg/DESCR: > Screenshot tool designed for elementary OS. > --- > > Works well on amd64. > > OK? > > ~Brian > > [0] https://bsd.network/web/statuses/104462236599534937 It builds fine on macppc by the way. I got this on amd64: --8<-- (io.elementary.screenshot-tool:53366): CRITICAL **: 21:59:16.902: Settings.vala:87: GDBus.Error:org.freedesktop.DBus.Error.ServiceUnknown: The name org.freedesktop.Accounts was not provided by any .service files --8<-- sysutils/accountsservice is needed. But then i hit: --8<-- The name org.gnome.Shell.Screenshot was not provided by any .service files -->8-- x11/gnome/gnome-shell is needed, and i guess it should be used under gnome-shell, since i still meet the issue with it installed without being run. I can't use gnome-shell with my radeondrm because the display is then garbled (drm issue). I can't cross test quickly on my macppc radeondrm where gnome-shell should be built in the current bulk. It would be nice for a third party who has an intel or radeondrm that works to test ;) Charlène.
Re: NEW: x11/elementary/calculator
Hi, On Tue, 30 Jun 2020 17:07:50 + Brian Callahan wrote: > Hi ports -- > > Attached is a new port, x11/elementary/calculator. Calculator is the > elementary OS calculator. > > --- > pkg/DESCR: > Calculator is a calculator app designed for elementary OS. > --- > > It's a basic calculator. Works on amd64. > > OK? > > ~Brian It builds and works fine for me on amd64 and macppc. OK cwen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/07/03 16:36:10 Modified files: audio/p5-MP3-Tag: Makefile distinfo Log message: p5-MP3-Tag: update to 1.15 This only includes a Perl 5.32 fix. While here reorder the Makefile per Makefile.template. OK sthen@
Re: [update/perl 5.32 fix] audio/p5-MP3-Tag 1.15
On Thu, 2 Jul 2020 00:11:25 +0100 Stuart Henderson wrote: > On 2020/07/01 23:51, Charlene Wendling wrote: > > Hi, > > > > This is a quick update for MP3::Tag to 1.15, fixing unescaped left > > braces that will be fatal with Perl >= 5.32. That's the only change > > [0], oh and now the module is distributed in zip form :-/ > > odd, what possible reason do they have for the zip?! ILYAZ moved to zip files for all their modules, i don't know why as well, maybe they got tired by permissions issues?
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/07/01 16:40:38 Modified files: devel/clang-tools-extra: Makefile Log message: clang-tools-extra: don't build on powerpc clang-tools-extra has no consumers anymore on powerpc, and takes 25 hours to build on the macppc bulk cluster. Don't waste time building it. OK rsadowski@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/07/01 16:24:08 Modified files: devel/kdevelop : Makefile Removed files: devel/kdevelop/patches: patch-plugins_welcomepage_CMakeLists_txt Log message: kdevelop: move from QtWebKit to QtWebEngine Also, remove the ports-gcc fix because no gcc archs can and will use QtWebEngine (it's x86 and arm64 only) OK rsadowski@ (maintainer)
[update/perl 5.32 fix] audio/p5-MP3-Tag 1.15
Hi, This is a quick update for MP3::Tag to 1.15, fixing unescaped left braces that will be fatal with Perl >= 5.32. That's the only change [0], oh and now the module is distributed in zip form :-/ I've tested audio/pacpl though, it works fine. I also totally rewrote the Makefile with proper ordering and spacing while here. Comments/feedback are welcome, Charlène. [0] https://metacpan.org/diff/file?target=ILYAZ/MP3-Tag-1.15/=ILYAZ%2FMP3-Tag-1.14 Index: Makefile === RCS file: /cvs/ports/audio/p5-MP3-Tag/Makefile,v retrieving revision 1.13 diff -u -p -u -p -r1.13 Makefile --- Makefile12 Jul 2019 20:43:38 - 1.13 +++ Makefile1 Jul 2020 21:14:56 - @@ -1,14 +1,17 @@ # $OpenBSD: Makefile,v 1.13 2019/07/12 20:43:38 sthen Exp $ -COMMENT= read tags of MP3 audio files +COMMENT = read tags of MP3 audio files -MODULES= cpan -PKG_ARCH= * -DISTNAME= MP3-Tag-1.14 -CATEGORIES=audio -FIX_EXTRACT_PERMISSIONS=Yes +DISTNAME = MP3-Tag-1.15 +EXTRACT_SUFX = .zip -# perl -PERMIT_PACKAGE=Yes +CATEGORIES = audio + +# Perl +PERMIT_PACKAGE = Yes + +MODULES = cpan + +PKG_ARCH = * .include Index: distinfo === RCS file: /cvs/ports/audio/p5-MP3-Tag/distinfo,v retrieving revision 1.7 diff -u -p -u -p -r1.7 distinfo --- distinfo28 Apr 2017 22:20:45 - 1.7 +++ distinfo1 Jul 2020 21:14:56 - @@ -1,2 +1,2 @@ -SHA256 (MP3-Tag-1.14.tar.gz) = PsOS8nPOOCKO4tflP3Yi9vkKnDHFB55snzVPsEqYwlk= -SIZE (MP3-Tag-1.14.tar.gz) = 279630 +SHA256 (MP3-Tag-1.15.zip) = qqxI9GN+3KQI/Xk4G8C/8PmxG9jh6U3gWdrpkzZfVtE= +SIZE (MP3-Tag-1.15.zip) = 321983
devel/kdevelop: use QtWebEngine, don't build devel/clang-tools-extra on macppc
Hi,
The below diff moves KDevelop from QtWebKit to QtWebEngine (x86 + arm64
only), this change implies 2 things:
- it can't be built on !QtWebEngine archs, so remove the ports-gcc fix
because no gcc archs can and will use QtWebEngine
- re-enable QtWebEngine detection in the port Makefile; since it takes
precedence over QtWebKit we don't need to disable it
And as a consequence clang-tools-extra has no consumers anymore on
!QtWebEngine archs, since qt-creator already moved to QtWebEngine:
$ show-reverse-deps devel/clang-tools-extra
devel/kdevelop
devel/kdevelop,
devel/qt-creator
As such, i'm proposing to not build clang-tools-extra on macppc, to
reduce wasted time. It saves 25 build hours (almost the same as
lang/gcc/8 for comparison), on top of the 10 hours needed to build
KDevelop!
As far as testing goes, build/install/testing have been run in an empty
amd64 chroot for good measure.
Comments/feedback are welcome,
Charlène.
Index: Makefile
===
RCS file: /cvs/ports/devel/kdevelop/Makefile,v
retrieving revision 1.37
diff -u -p -u -p -r1.37 Makefile
--- Makefile3 Jun 2020 05:20:54 - 1.37
+++ Makefile1 Jul 2020 17:00:30 -
@@ -6,6 +6,7 @@ USE_WXNEEDED = Yes
COMMENT = IDE for C, C++, Python, QML/JavaScript and PHP
VERSION = 5.5.2
+REVISION= 0
DISTNAME = kdevelop-${VERSION}
HOMEPAGE = https://www.kdevelop.org/
@@ -44,10 +45,11 @@ WANTLIB += KF5ThreadWeaver KF5WidgetsAdd
WANTLIB += Kasten4Controllers Kasten4Core Kasten4Gui Kasten4Okteta2Controllers
WANTLIB += Kasten4Okteta2Core Kasten4Okteta2Gui Okteta3Core Okteta3Gui
WANTLIB += Qt5Concurrent Qt5Core Qt5DBus Qt5Gui Qt5Help Qt5Network
-WANTLIB += Qt5Qml Qt5Quick Qt5QuickWidgets Qt5Sql Qt5Test Qt5WebKit
-WANTLIB += Qt5WebKitWidgets Qt5Widgets Qt5Xml apr-1 aprutil-1
-WANTLIB += c clang komparediff2 m svn_client-1 svn_fs-1 svn_ra-1
-WANTLIB += svn_repos-1 svn_subr-1 svn_wc-1
+WANTLIB += Qt5Positioning Qt5PrintSupport Qt5Qml Qt5Quick Qt5QuickWidgets
+WANTLIB += Qt5Sql Qt5Test Qt5WebChannel Qt5WebEngineCore Qt5WebEngineWidgets
+WANTLIB += Qt5Widgets Qt5Xml apr-1 aprutil-1 c clang komparediff2
+WANTLIB += m svn_client-1 svn_fs-1 svn_ra-1 svn_repos-1 svn_subr-1
+WANTLIB += svn_wc-1
MASTER_SITES = ${MASTER_SITE_KDE:=stable/kdevelop/${VERSION}/src/}
@@ -103,9 +105,7 @@ LIB_DEPENDS = devel/kf5/kconfig \
x11/kde-applications/libkomparediff2 \
x11/kde-applications/okteta>=0.26.1v0 \
x11/qt5/qttools \
- x11/qt5/qtwebkit
-
-CONFIGURE_ARGS += -DCMAKE_DISABLE_FIND_PACKAGE_Qt5WebEngineWidgets=ON
+ x11/qt5/qtwebengine
TEST_IS_INTERACTIVE = X11
Index: patches/patch-plugins_welcomepage_CMakeLists_txt
===
RCS file: patches/patch-plugins_welcomepage_CMakeLists_txt
diff -N patches/patch-plugins_welcomepage_CMakeLists_txt
--- patches/patch-plugins_welcomepage_CMakeLists_txt15 Nov 2019 14:30:20
- 1.1
+++ /dev/null 1 Jan 1970 00:00:00 -
@@ -1,17 +0,0 @@
-$OpenBSD: patch-plugins_welcomepage_CMakeLists_txt,v 1.1 2019/11/15 14:30:20
cwen Exp $
-
-Drop useless fix used for gcc-5, because it breaks the build with
-ports-gcc>=8.3.0.
-
-Index: plugins/welcomepage/CMakeLists.txt
plugins/welcomepage/CMakeLists.txt.orig
-+++ plugins/welcomepage/CMakeLists.txt
-@@ -28,8 +28,3 @@ target_link_libraries(kdevwelcomepage
- Qt5::QuickWidgets
- KF5::Declarative
- )
--# see https://bugs.launchpad.net/ubuntu/+source/gcc-5/+bug/1568899
--if (UNIX AND CMAKE_CXX_COMPILER_ID STREQUAL "GNU")
--message(WARNING "Applying workaround for
https://bugs.launchpad.net/ubuntu/+source/gcc-5/+bug/1568899;)
--target_link_libraries(kdevwelcomepage gcc_s gcc)
--endif()
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/06/30 19:28:18 Modified files: audio/pacpl: Makefile audio/pacpl/patches: patch-pacpl Added files: audio/pacpl/patches: patch-pacpl_conf Log message: pacpl: use gnudb.org instead of the dead freedb.org tweaks and OK from kn@ and bket@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/06/30 19:18:21 Modified files: audio/p5-cddb : Makefile Added files: audio/p5-cddb/patches: patch-lib_CDDB_pm Log message: p5-cddb: use gnudb.org instead of the dead freedb.org, add proper test depends tweaks and OK from kn@ and bket@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/06/30 19:09:00 Modified files: audio/p5-CDDB_get: Makefile Added files: audio/p5-CDDB_get/patches: patch-CDDB_get_pm Log message: p5-CDDB_get: use gnudb.org instead of the dead freedb.org, and move homepage to https. comment tweak and OK kn@, OK bket@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/06/30 19:00:43 Modified files: audio/abcde: Makefile Added files: audio/abcde/patches: patch-abcde patch-abcde_conf Log message: abcde: use gnudb.org instead of the dead freedb.org OK bket@ kn@
Re: Ports still using freedb
Hi,
OK? (diffs below)
On Mon, 29 Jun 2020 23:59:45 +0200
Christian Weisgerber wrote:
> The freedb.org database of compact track listings has shut down.
> Any program functionality that tries to fetch from or submit data
> to freedb.org is broken now.
>
> An alternative service is available at gnudb.org, see
> https://www.gnudb.org/
>
> Below is a list of ports that still appear to use freedb.org in
> code and need to be fixed. I've cc'ed any maintainers.
>
> audio/aqualung jer...@openbsd.org
> audio/audacious/plugins b...@openbsd.org
> audio/cantata ports@openbsd.org
> audio/cd-discid ports@openbsd.org
> audio/cmus ports@openbsd.org
> audio/deadbeef ports@openbsd.org
> audio/disc-coverports@openbsd.org
> audio/easytag ports@openbsd.org
> audio/grip ports@openbsd.org
> audio/libcdaudioports@openbsd.org
> audio/libcddb ports@openbsd.org
> audio/libcdio ports@openbsd.org
> audio/libdiscid ports@openbsd.org
> audio/libmp3spltports@openbsd.org
> audio/mp3cddb ports@openbsd.org
> audio/mp3splt ports@openbsd.org
> audio/puddletag ports@openbsd.org
> audio/py-cddb ports@openbsd.org
> audio/vlorb ports@openbsd.org
> devel/kf5/kio rsadow...@openbsd.org
> misc/cdrdao ports@openbsd.org
> misc/mc rob...@openbsd.org
> misc/xcdroast ports@openbsd.org
> multimedia/xine-lib b...@comstyle.com
> net/mldonkeyports@openbsd.org
> print/texlive/texmf e...@openbsd.org
> sysutils/cdrtools ports@openbsd.org
> www/surfraw ben...@openbsd.org
> x11/gnome/vfs2 jas...@openbsd.org,
> ajacou...@openbsd.org x11/gnustep/cdplayer
> sebas...@openbsd.org x11/kde/base3 ports@openbsd.org
> x11/kde-applications/libkcddb rsadow...@openbsd.org
> x11/mplayer b...@comstyle.com
>
> audio/abcde c...@openbsd.org
>
Musicbrainz is used by default, but some users having old config files
may still use cddb. Upstream is aware of the situation but no patches
yet.
Index: Makefile
===
RCS file: /cvs/ports/audio/abcde/Makefile,v
retrieving revision 1.30
diff -u -p -u -p -r1.30 Makefile
--- Makefile12 Jul 2019 20:43:31 - 1.30
+++ Makefile29 Jun 2020 22:39:59 -
@@ -3,6 +3,7 @@
COMMENT= command-line utility to rip and encode audio cds
DISTNAME= abcde-2.9.3
+REVISION= 0
EPOCH= 0
CATEGORIES=audio
Index: patches/patch-abcde
===
RCS file: patches/patch-abcde
diff -N patches/patch-abcde
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-abcde 29 Jun 2020 22:39:59 -
@@ -0,0 +1,16 @@
+$OpenBSD$
+
+Index: abcde
+--- abcde.orig
abcde
+@@ -3770,8 +3770,8 @@ post_encode ()
+ # CDDB
+ # Currently three supported options ("musicbrainz", "cddb" for freedb.org and
"cdtext")
+ CDDBMETHOD=musicbrainz
+-CDDBURL="http://freedb.freedb.org/~cddb/cddb.cgi;
+-CDDBSUBMIT=freedb-sub...@freedb.org
++CDDBURL="https://gnudb.gnudb.org/~cddb/cddb.cgi;
++CDDBSUBMIT=sub...@gnudb.org
+ CDDBPROTO=6
+ HELLOINFO="$(whoami)@$(hostname)"
+ CDDBCOPYLOCAL="n"
Index: patches/patch-abcde_conf
===
RCS file: patches/patch-abcde_conf
diff -N patches/patch-abcde_conf
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-abcde_conf29 Jun 2020 22:39:59 -
@@ -0,0 +1,23 @@
+$OpenBSD$
+
+Index: abcde.conf
+--- abcde.conf.orig
abcde.conf
+@@ -15,7 +15,7 @@
+ # If you wish to use a different CDDB server, edit this line.
+ # If you just wanted to use a proxy server, just set your http_proxy
+ # environment variable - wget will use it correctly.
+-#CDDBURL="http://freedb.freedb.org/~cddb/cddb.cgi;
++#CDDBURL="https://gnudb.gnudb.org/~cddb/cddb.cgi;
+
+ # The CDDB protocol level.
+ # Right now 5 is latin1 output and 6 is UTF8 encoding.
+@@ -27,7 +27,7 @@
+ #HELLOINFO="`whoami`@`hostname`"
+
+ # This controls the email address CDDB changes are submitted to.
+-#CDDBSUBMIT=freedb-sub...@freedb.org
++#CDDBSUBMIT=sub...@gnudb.org
+
+ # The following options control whether or not fetched CDDB entries
+ # are cached locally in $CDDBLOCALDIR
>
> audio/p5-CDDB_get ports@openbsd.org
>
Tested for real - the test suite is useless
Index: Makefile
===
RCS file: /cvs/ports/audio/p5-CDDB_get/Makefile,v
retrieving revision 1.16
diff -u -p -u -p -r1.16 Makefile
--- Makefile12 Jul 2019 20:43:38 -
Re: powerpc bulk build report
Hi, On Mon, 29 Jun 2020 11:45:14 -0600 (MDT) c...@openbsd.org wrote: > Bulk build on macppc-0.ports.openbsd.org > > Started : Sat Jun 13 07:45:16 MDT 2020 > Finished: Mon Jun 29 11:45:00 MDT 2020 > Duration: 16 Days 4 hours 0 minutes > > Built using OpenBSD 6.7-current (GENERIC) #735: Fri Jun 12 08:20:27 > MDT 2020 > > Built 9822 packages This bulk has been shorter and with less packages because it has been started after kde4 has been mostly unhooked, but before Rafael imported dozens of new kf5 ports. Some GNUstep-related software that used to be broken have been built again, runtime should be wrong [0]; this explains some failures below. Maybe^W i'll try again with my fix for x11/gnustep/make. MATE is working since last bulk [1]. GNOME still lacks gnome-session and gnome-shell, now that spidermonkey68 is rust-free it may be available again soon :) > http://build-failures.rhaalovely.net/powerpc/2020-06-13/devel/geany.log I've a fix for it [2], despite my assumptions being wrong. Charlène. [0] https://marc.info/?l=openbsd-ports=158722515929053=2 [1] https://bsd.network/web/statuses/104336811837102326 [2] https://marc.info/?l=openbsd-ports=159188140009019=2 > Critical path missing pkgs: > http://build-failures.rhaalovely.net/powerpc/2020-06-13/summary.log > > Build failures: 15 > http://build-failures.rhaalovely.net/powerpc/2020-06-13/emulators/frodo.log > http://build-failures.rhaalovely.net/powerpc/2020-06-13/games/hyperrogue.log > http://build-failures.rhaalovely.net/powerpc/2020-06-13/games/valyriatear.log > http://build-failures.rhaalovely.net/powerpc/2020-06-13/lang/gforth.log > http://build-failures.rhaalovely.net/powerpc/2020-06-13/lang/squeak/vm.log > http://build-failures.rhaalovely.net/powerpc/2020-06-13/shells/ksh93.log > http://build-failures.rhaalovely.net/powerpc/2020-06-13/sysutils/libvirt.log > http://build-failures.rhaalovely.net/powerpc/2020-06-13/www/sope.log > http://build-failures.rhaalovely.net/powerpc/2020-06-13/x11/agar/agar.log > http://build-failures.rhaalovely.net/powerpc/2020-06-13/x11/e17/elementary.log > http://build-failures.rhaalovely.net/powerpc/2020-06-13/x11/gnustep/dbuskit.log > http://build-failures.rhaalovely.net/powerpc/2020-06-13/x11/gnustep/gui.log > http://build-failures.rhaalovely.net/powerpc/2020-06-13/x11/gnustep/performance.log > http://build-failures.rhaalovely.net/powerpc/2020-06-13/x11/kde4/pimlibs.log
Re: x11/sisctrl removable
On Fri, 26 Jun 2020 11:13:50 +0200 Solene Rapenne wrote: > I propose to remove x11/sisctrl because sis driver was deleted a > few months ago from xenocara. > > No port depend on it. > OK cwen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/06/26 03:41:40 Modified files: devel/codeblocks: Makefile distinfo devel/codeblocks/patches: patch-configure patch-src_base_tinyxml_Makefile_in patch-src_plugins_compilergcc_compilergcc_cpp devel/codeblocks/pkg: PLIST Added files: devel/codeblocks/patches: patch-src_plugins_contrib_dragscroll_dragscrollcfg_h Log message: codeblocks: update to 20.03 changelog: http://www.codeblocks.org/downloads/binaries/changelog Also, change COMPILER reason, add a new patch to fix the build on !x86 archs and remove a trailing whitespace in Makefile. >From Josh Grosse, with some tweaks from me. OK Josh Grosse (maintainer)
Re: NEW: x11/elementary/dock
On Fri, 26 Jun 2020 01:54:42 + Brian Callahan wrote: > Hi ports -- > > Attached is a new port, x11/elementary/dock. This is the elementary OS > dock app. The binary is named plank. > > --- > pkg/DESCR: > Plank is meant to be the simplest dock on the planet. The goal is to > provide just what a dock needs and absolutely nothing more. It is, > however, a library which can be extended to create other dock programs > with more advanced features. > --- > > Needs the bamf port sent earlier. Works well on amd64. Note that the > CPU and battery docklets do not appear to work, but I believe I can > get those working at some point. > > OK? > > ~Brian It builds and works fine on amd64 and macppc. It adds mupdf, but unlike mpv it has no "drag file there to open it" dialog. Well, we still miss some parts of the elementary puzzle anyway, so go ahead :) OK cwen@
Re: NEW: devel/bamf
Hi, On Fri, 26 Jun 2020 00:18:01 + Brian Callahan wrote: > Hi ports -- > > Attached is a new port, devel/bamf. Bamf is an application matching > library. > > --- > pkg/DESCR: > Bamf matches application windows to desktop files. It removes the > headache of applications matching into a simple DBus daemon and C > wrapper library. It currently features application matching at amazing > levels of accuracy (covering nearly every corner case). > --- > > This is needed for the elementary OS dock application. Builds ok on > amd64 and sparc64. The elementary OS dock app works fine on amd64 so > I assume this runs fine too. > > OK? > > ~Brian port-lib-depends-check complains about this on my macppc and amd64 machines: bamf-0.5.4(devel/bamf): Missing: png.18 from png-1.6.37 (/usr/local/libexec/bamf/bamfdaemon) Extra: XRes.5 gthread-2.0.4200 png16.18 WANTLIB += png It's weird because i thought we fixed such png WANTLIB issues a while ago. It's still packaging though, so once the situation is cleared up it's fine with me.
Re: NEW: x11/elementary/code
On Thu, 25 Jun 2020 18:05:30 + Brian Callahan wrote: > Hi ports -- > > Attached is a new port, x11/elementary/code. Code is the elementary OS > code editor. Requires the editors/editorconfig-core-c port I just > sent. > > Works well on amd64. > > OK? > > ~Brian It builds and works fine on amd64 and macppc. OK cwen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/06/25 15:13:17 Modified files: devel/spidermonkey68: Makefile devel/spidermonkey68/patches: patch-moz_configure Added files: devel/spidermonkey68/patches: patch-js_moz_configure Log message: spidermonkey68: don't use rust and llvm; they are unneeded It is part of Firefox, so rust and llvm are checked out at configure time, but the javascript engine uses only C/C++. Add patches to remove that check, allowing spidermonkey to be built on !rust archs. From John Paul Adrian Glaubitz for Debian. OK jasper@ (maintainer)
Re: NEW: editors/editorconfig-core-c
On Thu, 25 Jun 2020 17:53:14 + Brian Callahan wrote: > Hi ports -- > > Attached is a new port, editors/editorconfig-core-c. > Editorconfig-core-c is a library for IDE consistency. > > --- > pkg/DESCR: > EditorConfig helps developers define and maintain consistent coding > styles between different editors and IDEs. > > This package provides the core library and editorconfig(1) executable > both written in C for use by text editor plugins supporting > EditorConfig parsing. > --- > > This is needed for a port of the elementary OS code editor. > > OK? > > ~Brian This tries to pick up doxygen during the build, so the port system is trying to install manpages, that are not generated on my machines where doxygen is not available. I'm attaching a new tarball with documentation disabled and an updated PLIST. With that fixed, OK cwen@ editorconfig-core-c.tgz Description: Binary data
Re: NEW: x11/elementary/terminal
On Thu, 25 Jun 2020 16:06:51 + Brian Callahan wrote: > > On Thursday, June 25, 2020 11:59 AM, Brian Callahan > wrote: > > > Hi ports -- > > > > Attached is a new port, x11/elementary/terminal. Terminal is the > > elementary OS terminal emulator. > > > > --- > > > > pkg/DESCR: > > Terminal is a super lightweight, beautiful, and simple terminal. > > Comes with sane defaults, browser-class tabs, sudo paste > > protection, smart copy/paste, and little to no configuration. > > > > --- > > > > Works well on my amd64 machine. > > > > OK? It builds and works fine on amd64 and macppc; note that when i click on the search icon, it generates a sigbus, but it seems it's because i don't run dbus in my X sessions. OK cwen@ > > ~Brian > > Now with MODULES=devel/dconf, as informed by portcheck. > > ~Brian
Re: NEW: x11/elementary/granite
Hi, On Thu, 25 Jun 2020 14:50:15 + Brian Callahan wrote: > Hi ports -- > > Attached is a new port, x11/elementary/granite. Also includes related > infrastructure to have an x11/elementary subdir, which will be filled > with ports from the elementary OS project. > > Granite is the graphical support library for elementary OS apps. > > --- > pkg/DESCR: > Granite is a companion library for GTK+ and GLib. Among other things, > it provides complex widgets and convenience functions designed for > use in apps built for elementary OS. > --- > > I have a number of elementary OS apps running on amd64 which will be > sent after granite goes in. > > OK? > > ~Brian It builds fine on amd64 and macppc. OK cwen@
Re: [update] devel/codeblocks
Hi,
On Fri, 5 Jun 2020 19:44:06 -0400
Josh Grosse wrote:
> Update from 17.12 to 20.03. Builds and runs
> on amd64. Changelog:
> http://www.codeblocks.org/downloads/binaries/changelog
>
As far as x86 is concerned it's good, but it won't build on !x86 because
some plugin uses the int3 asm instruction.
I made the following changes to your diff:
- patch out the int3 instruction, fix stolen from Debian
- refresh patches
- change COMPILER reason
- fix a trailing whitespace in Makefile
This builds and works fine on amd64 and macppc.
Charlène.
Index: Makefile
===
RCS file: /cvs/ports/devel/codeblocks/Makefile,v
retrieving revision 1.55
diff -u -p -u -p -r1.55 Makefile
--- Makefile12 Jul 2019 20:44:06 - 1.55
+++ Makefile25 Jun 2020 13:58:00 -
@@ -3,9 +3,8 @@
BROKEN-alpha= sqvm.cpp:394:1: internal compiler error: Segmentation fault
COMMENT= Code::Blocks C/C++ IDE
-V= 17.12
-REVISION= 6
-DISTNAME= codeblocks_${V}
+V= 20.03
+DISTNAME= codeblocks-${V}
PKGNAME= codeblocks-${V}
EXTRACT_SUFX= .tar.xz
@@ -34,8 +33,8 @@ PERMIT_PACKAGE= Yes
WANTLIB += ${COMPILER_LIBCXX} ICE SDL2 SM X11 Xcomposite Xcursor
WANTLIB += Xdamage Xext Xfixes Xi Xinerama Xrandr Xrender Xxf86vm
-WANTLIB += atk-1.0 atk-bridge-2.0 atspi boost_system bz2 c cairo
-WANTLIB += cairo-gobject dbus-1 epoxy expat ffi fontconfig freetype
+WANTLIB += atk-1.0 atk-bridge-2.0 boost_system bz2 c cairo
+WANTLIB += cairo-gobject epoxy expat ffi fontconfig freetype
WANTLIB += fribidi gamin-1 gdk-3 gdk_pixbuf-2.0 gio-2.0 glib-2.0
WANTLIB += gmodule-2.0 gobject-2.0 graphite2 gthread-2.0 gtk-3
WANTLIB += harfbuzz hunspell-1.6 iconv intl jpeg m mspack notify
@@ -45,6 +44,7 @@ WANTLIB += wx_baseu_xml-3.0 wx_gtk3u_adv
WANTLIB += wx_gtk3u_core-3.0 wx_gtk3u_html-3.0 wx_gtk3u_propgrid-3.0
WANTLIB += wx_gtk3u_qa-3.0 wx_gtk3u_richtext-3.0 wx_gtk3u_xrc-3.0
WANTLIB += xcb xcb-render xcb-shm z
+WANTLIB += Xss lzma zstd
MASTER_SITES= ${MASTER_SITE_SOURCEFORGE:=codeblocks/Sources/${V}/}
@@ -60,7 +60,7 @@ RUN_DEPENDS= devel/desktop-file-utils \
misc/shared-mime-info \
x11/gtk+3,-guic
-# XXX :0: internal compiler error: Segmentation fault
+# C++11
COMPILER = base-clang ports-gcc
AUTOCONF_VERSION= 2.69
@@ -80,7 +80,7 @@ CONFIGURE_ARGS += --with-boost-libdir=${
MAKE_ENV= AUTOMAKE_VERSION=${AUTOMAKE_VERSION} \
AUTOCONF_VERSION=${AUTOCONF_VERSION}
-pre-patch:
+pre-patch:
cd ${WRKSRC}; \
env \
AUTOMAKE_VERSION=${AUTOMAKE_VERSION} \
Index: distinfo
===
RCS file: /cvs/ports/devel/codeblocks/distinfo,v
retrieving revision 1.6
diff -u -p -u -p -r1.6 distinfo
--- distinfo6 Feb 2018 15:28:51 - 1.6
+++ distinfo25 Jun 2020 13:58:00 -
@@ -1,2 +1,2 @@
-SHA256 (codeblocks_17.12.tar.xz) = E4gaCnJ2lpToLlMbjngU1R+/H6EixzxQBOGGVg+8V+A=
-SIZE (codeblocks_17.12.tar.xz) = 15725976
+SHA256 (codeblocks-20.03.tar.xz) = Fe6z4orqBU4fOLDH9GcbTU0RFv0F9jwHqpWpHbieqsU=
+SIZE (codeblocks-20.03.tar.xz) = 11799264
Index: patches/patch-configure
===
RCS file: /cvs/ports/devel/codeblocks/patches/patch-configure,v
retrieving revision 1.6
diff -u -p -u -p -r1.6 patch-configure
--- patches/patch-configure 6 Feb 2018 15:28:51 - 1.6
+++ patches/patch-configure 25 Jun 2020 13:58:00 -
@@ -5,7 +5,7 @@ undefined reference to `free'
Index: configure
--- configure.orig
+++ configure
-@@ -16887,7 +16887,6 @@ if ${ax_cv_check_ldflags___Wl___no_undefined+:} false;
+@@ -16911,7 +16911,6 @@ if ${ax_cv_check_ldflags___Wl___no_undefined+:} false;
else
ax_check_save_flags=$LDFLAGS
@@ -13,7 +13,7 @@ Index: configure
cat confdefs.h - <<_ACEOF >conftest.$ac_ext
/* end confdefs.h. */
-@@ -16911,7 +16910,6 @@ fi
+@@ -16935,7 +16934,6 @@ fi
{ $as_echo "$as_me:${as_lineno-$LINENO}: result:
$ax_cv_check_ldflags___Wl___no_undefined" >&5
$as_echo "$ax_cv_check_ldflags___Wl___no_undefined" >&6; }
if test "x$ax_cv_check_ldflags___Wl___no_undefined" = xyes; then :
Index: patches/patch-src_base_tinyxml_Makefile_in
===
RCS file:
/cvs/ports/devel/codeblocks/patches/patch-src_base_tinyxml_Makefile_in,v
retrieving revision 1.4
diff -u -p -u -p -r1.4 patch-src_base_tinyxml_Makefile_in
--- patches/patch-src_base_tinyxml_Makefile_in 6 Feb 2018 15:28:51 -
1.4
+++ patches/patch-src_base_tinyxml_Makefile_in 25 Jun 2020 13:58:00 -
@@ -6,7 +6,7 @@ Fix build when tinyxml is installed.
Index: src/base/tinyxml/Makefile.in
--- src/base/tinyxml/Makefile.in.orig
+++ src/base/tinyxml/Makefile.in
-@@ -416,10 +416,10 @@ top_builddir = @top_builddir@
+@@ -423,10 +423,10 @@ top_builddir = @top_builddir@
top_srcdir
Re: [all archs] devel/spidermonkey68 can be built without rust and llvm
On Thu, 25 Jun 2020 08:38:44 +0200
Jasper Lievisse Adriaanse wrote:
>
> > On 25 Jun 2020, at 00:28, Charlene Wendling
> > wrote:
> >
> > Hi,
> >
> > I was trying to build a port depending on spidermonkey68 on my
> > macppc machine and found out it "requires" rust. After seeing how
> > other downstreams deal with it, i've found out that Debian actually
> > does not use rust for spidermonkey. For all archs [0].
> >
> > What i'm proposing here is a proof of concept that allows building
> > spidermonkey68 on !rust archs, like macppc [1], but i think really
> > we should drop rust and llvm for all archs.
> >
> > Comments/feedback are welcome,
> >
> > Charlène.
> >
> > [0] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=959144
> > [1] https://bin.charlenew.xyz/spidermonkey_norust.log
>
> Hi,
>
> Since it doesn’t change anything (as in, nothing ends up using it) I
> see no point in keeping the dependencies around for a subset of
> architectures. The cost of maintaining these patches is already there
> anyway. Go ahead and just drop these dependencies on all
> architectures.
>
> Cheers,
> Jasper
Here it is, i added a safety REVISION bump as well but i don't think
it's needed.
Index: Makefile
===
RCS file: /cvs/ports/devel/spidermonkey68/Makefile,v
retrieving revision 1.4
diff -u -p -u -p -r1.4 Makefile
--- Makefile24 Jun 2020 08:57:02 - 1.4
+++ Makefile25 Jun 2020 09:05:38 -
@@ -20,7 +20,7 @@ V = ${MOZILLA_VERSION}.9.0
DISTNAME = firefox-${V}esr.source
EXTRACT_SUFX = .tar.bz2
PKGNAME = spidermonkey${MOZILLA_VERSION}-${V}
-REVISION = 0
+REVISION = 1
SHARED_LIBS = mozjs-68 0.0
@@ -54,9 +54,7 @@ LIB_DEPENDS = devel/nspr \
DEBUG_PACKAGES = ${BUILD_PACKAGES}
AUTOCONF_VERSION= 2.13
-BUILD_DEPENDS += devel/autoconf/${AUTOCONF_VERSION} \
- devel/llvm \
- lang/rust
+BUILD_DEPENDS += devel/autoconf/${AUTOCONF_VERSION}
SEPARATE_BUILD = Yes
WRKDIST = ${WRKDIR}/firefox-${V}
@@ -77,9 +75,7 @@ CONFIGURE_ARGS = --disable-debug \
--with-system-zlib \
--with-system-icu \
--with-system-nspr \
- --enable-unaligned-private-values \
- --with-libclang-path=${LOCALBASE}/lib \
- --with-clang-path=${LOCALBASE}/bin/clang
+ --enable-unaligned-private-values
# Build failure on mips64(,el}. Related bug on Debian:
# https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=908486
Index: patches/patch-js_moz_configure
===
RCS file: patches/patch-js_moz_configure
diff -N patches/patch-js_moz_configure
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-js_moz_configure 25 Jun 2020 09:05:38 -
@@ -0,0 +1,31 @@
+$OpenBSD$
+
+Remove unused LLVM and Rust build dependencies. Since the Javascript engine is
+normally part of Firefox, its build system has dependencies on the LLVM and
+Rust toolchains. This limits the number of architectures which mozjs68 can be
+built on. It turns out, however, that neither LLVM nor Rust are used when
+mozjs68 is being built and these build dependencies are therefore not
+necessary.
+
+This patch removes them and allows mozjs68 to be built on any
+architecture.
+
+Author: John Paul Adrian Glaubitz
+
+From: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=959144
+
+Index: js/moz.configure
+--- js/moz.configure.orig
js/moz.configure
+@@ -18,11 +18,6 @@ def building_js(build_project):
+ option(env='JS_STANDALONE', default=building_js,
+help='Reserved for internal use')
+
+-include('../build/moz.configure/rust.configure',
+-when='--enable-compile-environment')
+-include('../build/moz.configure/bindgen.configure',
+-when='--enable-compile-environment')
+-
+ @depends('JS_STANDALONE')
+ def js_standalone(value):
+ if value:
Index: patches/patch-moz_configure
===
RCS file: /cvs/ports/devel/spidermonkey68/patches/patch-moz_configure,v
retrieving revision 1.1.1.1
diff -u -p -u -p -r1.1.1.1 patch-moz_configure
--- patches/patch-moz_configure 14 May 2020 14:54:11 - 1.1.1.1
+++ patches/patch-moz_configure 25 Jun 2020 09:05:38 -
@@ -1,5 +1,9 @@
$OpenBSD: patch-moz_configure,v 1.1.1.1 2020/05/14 14:54:11 jasper Exp $
+- Fix awk detection
+- Remove unused LLVM and Rust build dependencies, from:
+ https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=959144
+
Index: moz.configure
--- moz.configure.orig
+++ moz.configure
@@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/06/24 16:50:41 Modified files: graphics/openscenegraph: Makefile Log message: openscenegraph: do not build on powerpc It takes 17 hours to build openscenegraph on the macppc bulk cluster, but no final consumers are built on this arch. Don't waste time building it. OK kn@, pascal@ (maintainer)
[all archs] devel/spidermonkey68 can be built without rust and llvm
Hi,
I was trying to build a port depending on spidermonkey68 on my macppc
machine and found out it "requires" rust. After seeing how other
downstreams deal with it, i've found out that Debian actually does not
use rust for spidermonkey. For all archs [0].
What i'm proposing here is a proof of concept that allows building
spidermonkey68 on !rust archs, like macppc [1], but i think really we
should drop rust and llvm for all archs.
Comments/feedback are welcome,
Charlène.
[0] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=959144
[1] https://bin.charlenew.xyz/spidermonkey_norust.log
Index: Makefile
===
RCS file: /cvs/ports/devel/spidermonkey68/Makefile,v
retrieving revision 1.4
diff -u -p -u -p -r1.4 Makefile
--- Makefile24 Jun 2020 08:57:02 - 1.4
+++ Makefile24 Jun 2020 22:18:37 -
@@ -54,9 +54,7 @@ LIB_DEPENDS = devel/nspr \
DEBUG_PACKAGES = ${BUILD_PACKAGES}
AUTOCONF_VERSION= 2.13
-BUILD_DEPENDS += devel/autoconf/${AUTOCONF_VERSION} \
- devel/llvm \
- lang/rust
+BUILD_DEPENDS += devel/autoconf/${AUTOCONF_VERSION}
SEPARATE_BUILD = Yes
WRKDIST = ${WRKDIR}/firefox-${V}
@@ -77,9 +75,7 @@ CONFIGURE_ARGS = --disable-debug \
--with-system-zlib \
--with-system-icu \
--with-system-nspr \
- --enable-unaligned-private-values \
- --with-libclang-path=${LOCALBASE}/lib \
- --with-clang-path=${LOCALBASE}/bin/clang
+ --enable-unaligned-private-values
# Build failure on mips64(,el}. Related bug on Debian:
# https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=908486
@@ -99,6 +95,16 @@ MAKE_ENV = SHELL=/bin/sh \
SO_VERSION=${SO_VERSION}
NO_TEST = Yes
+
+.include
+.if ${PROPERTIES:Mrust}
+BUILD_DEPENDS += devel/llvm \
+ lang/rust
+CONFIGURE_ARGS += --with-libclang-path=${LOCALBASE}/lib \
+ --with-clang-path=${LOCALBASE}/bin/clang
+.else
+PATCH_LIST += patch-* patch_norust-*
+.endif
pre-configure:
cd ${WRKSRC} && env -i AUTOCONF_VERSION=${AUTOCONF_VERSION} autoconf
Index: patches/patch_norust-js_moz_configure
===
RCS file: patches/patch_norust-js_moz_configure
diff -N patches/patch_norust-js_moz_configure
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch_norust-js_moz_configure 24 Jun 2020 22:18:37 -
@@ -0,0 +1,31 @@
+$OpenBSD$
+
+Remove unused LLVM and Rust build dependencies. Since the Javascript engine is
+normally part of Firefox, its build system has dependencies on the LLVM and
+Rust toolchains. This limits the number of architectures which mozjs68 can be
+built on. It turns out, however, that neither LLVM nor Rust are used when
+mozjs68 is being built and these build dependencies are therefore not
+necessary.
+
+This patch removes them and allows mozjs68 to be built on any
+architecture.
+
+Author: John Paul Adrian Glaubitz
+
+From: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=959144
+
+Index: js/moz.configure
+--- js/moz.configure.orig
js/moz.configure
+@@ -18,11 +18,6 @@ def building_js(build_project):
+ option(env='JS_STANDALONE', default=building_js,
+help='Reserved for internal use')
+
+-include('../build/moz.configure/rust.configure',
+-when='--enable-compile-environment')
+-include('../build/moz.configure/bindgen.configure',
+-when='--enable-compile-environment')
+-
+ @depends('JS_STANDALONE')
+ def js_standalone(value):
+ if value:
Index: patches/patch_norust-moz_configure
===
RCS file: patches/patch_norust-moz_configure
diff -N patches/patch_norust-moz_configure
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch_norust-moz_configure 24 Jun 2020 22:18:37 -
@@ -0,0 +1,56 @@
+$OpenBSD$
+
+Remove unused LLVM and Rust build dependencies. Since the Javascript engine is
+normally part of Firefox, its build system has dependencies on the LLVM and
+Rust toolchains. This limits the number of architectures which mozjs68 can be
+built on. It turns out, however, that neither LLVM nor Rust are used when
+mozjs68 is being built and these build dependencies are therefore not
+necessary.
+
+This patch removes them and allows mozjs68 to be built on any
+architecture.
+
+Author: John Paul Adrian Glaubitz
+
+From: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=959144
+
+Index: moz.configure
+--- moz.configure.orig
moz.configure
+@@ -598,36 +598,6 @@ set_config('MAKENSISU_FLAGS', nsis_flags)
+
+ check_prog('7Z', ('7z', '7za'), allow_missing=True, when=target_is_windows)
+
+-
+-@depends(host_c_compiler, c_compiler, bindgen_config_paths)
Re: [macppc] Don't build graphics/openscenegraph
On Wed, 24 Jun 2020 21:02:11 +0200 Klemens Nanni wrote: > On Wed, Jun 24, 2020 at 08:39:09PM +0200, Charlene Wendling wrote: > > It takes 17 hours to build graphics/openscenegraph on the macppc > > bulk cluster, the same as x11/qt5/qtbase for example. > > > > It appears that according that all consumers are not built on > > macppc. As i'm trying to reduce wasted time i would like to mark it > > NOT_FOR_ARCHS, similarly to what we've done already with cad/oce. > $ show-reverse-deps graphics/openscenegraph > games/flightgear/base > games/openmw > games/speeddreams > games/speeddreams,-main > graphics/simgear > > openmw and speeddreams are marked with ONLY_FOR_ARCHS excluding maccpc > but flightgear and simgear are not: what is it that makes those ports > not building on macppc? flightgear has also ONLY_FOR_ARCHS in Makefile.inc. About simgear, flightgear is the only depends: $ show-reverse-deps graphics/simgear games/flightgear/base > Assuming that none of those are in fact built, disabling > openscenegraph seems quite sane to me, so OK kn >
[macppc] Don't build graphics/openscenegraph
Hi, It takes 17 hours to build graphics/openscenegraph on the macppc bulk cluster, the same as x11/qt5/qtbase for example. It appears that according that all consumers are not built on macppc. As i'm trying to reduce wasted time i would like to mark it NOT_FOR_ARCHS, similarly to what we've done already with cad/oce. OK? Charlène. Index: Makefile === RCS file: /cvs/ports/graphics/openscenegraph/Makefile,v retrieving revision 1.21 diff -u -p -u -p -r1.21 Makefile --- Makefile14 May 2020 12:35:31 - 1.21 +++ Makefile24 Jun 2020 18:25:38 - @@ -1,5 +1,8 @@ # $OpenBSD: Makefile,v 1.21 2020/05/14 12:35:31 ajacoutot Exp $ +# No consumers on macppc, saves 17 build hours in bulks +NOT_FOR_ARCHS =powerpc + PORTROACH =limitw:1,even COMMENT = 3D graphics toolkit
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/06/24 11:56:43 Modified files: x11/gnome/photos: Makefile Log message: x11/gnome/photos: remove BROKEN-powerpc It was marked BROKEN due to undefined references to 64-bit __sync primitives, but x11/gnome/libdazzle now uses a fallback that does not use them on powerpc, allowing Photos to be built.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/06/24 11:30:00 Modified files: misc/osinfo: Makefile Makefile.inc misc/osinfo/libosinfo: Makefile misc/osinfo/osinfo-db: Makefile misc/osinfo/osinfo-db-tools: Makefile Added files: misc/osinfo/libosinfo/patches: patch-meson_build Log message: osinfo/*: move HOMEPAGE to https, fix the build on clang+ld.bfd archs The generated binaries from libosinfo used to include _Unwind_* symbols, they were unresolved on clang+ld.bfd archs, with possible consequences at runtime on lld archs. Drop '-fexceptions', so we don't generate such symbols. Also remove a trailing whitespace in osinfo's Makefile. Thanks to sthen and aja for hints! OK aja@ (maintainer)
Re: UPDATE x11/icewm-1.6.6
Hi,
On Sun, 14 Jun 2020 11:15:17 +0200
Bjorn Ketelaars wrote:
> On Sun 24/05/2020 14:03, Bjorn Ketelaars wrote:
> > On Wed 08/04/2020 10:21, Bjorn Ketelaars wrote:
> > > Diff below brings icewm to 1.6.5, which is a maintenance release.
> > > Changes: https://github.com/bbidulock/icewm/blob/1.6.5/ChangeLog.
> > >
> > > Run tested on amd64.
> > >
> > > Comments/OK?
> >
> > I was contacted off list by Greg Jones who found an issue with the
> > taskbar that has been fixed upstream [0]. For now cherrypick this
> > fix. New diff enclosed.
> >
> > Comments/OK?
>
> Upstream of icewm released a new version (1.6.6), which fixes the
> taskbar. Overview on changes can be found at
> https://github.com/ice-wm/icewm/releases/tag/1.6.6.
>
> Comments/OK?
It builds and works fine on macppc, including icewm-menu-xrandr.
OK cwen@
> diff --git Makefile Makefile
> index a0f97493575..febd21b6cba 100644
> --- Makefile
> +++ Makefile
> @@ -3,7 +3,7 @@
> # unconditionally uses machine/apmvar.h and related
> # structures/defines
> ONLY_FOR_ARCHS= ${APM_ARCHS}
>
> -V= 1.6.4
> +V= 1.6.6
> COMMENT= small and fast window manager
> DISTNAME=icewm-${V}
> EXTRACT_SUFX=.tar.lz
> diff --git distinfo distinfo
> index 54fa162eb04..f50c9813398 100644
> --- distinfo
> +++ distinfo
> @@ -1,2 +1,2 @@
> -SHA256 (icewm-1.6.4.tar.lz) =
> c83MIQl7lPPYNSLi5od8up8G6SwYI8oU4KoyowztDKs= -SIZE
> (icewm-1.6.4.tar.lz) = 1684348 +SHA256 (icewm-1.6.6.tar.lz) =
> d/US4LzcnG7I4aEMgiuCOFSCZbq8cN/k63a0Ug1gwBw= +SIZE
> (icewm-1.6.6.tar.lz) = 1684639 diff --git
> patches/patch-doc_icewm_adoc patches/patch-doc_icewm_adoc index
> f013af21948..8bb9a04dfec 100644
> --- patches/patch-doc_icewm_adoc
> +++ patches/patch-doc_icewm_adoc
> @@ -3,7 +3,7 @@ $OpenBSD: patch-doc_icewm_adoc,v 1.1 2019/03/14
> 13:48:23 bcallah Exp $ Index: doc/icewm.adoc
> --- doc/icewm.adoc.orig
> +++ doc/icewm.adoc
> -@@ -395,10 +395,10 @@ $ICEWM_PRIVCFG; $XDG_CONFIG_HOME/icewm;
> $HOME/.icewm +@@ -396,10 +396,10 @@ $ICEWM_PRIVCFG;
> $XDG_CONFIG_HOME/icewm; $HOME/.icewm The first of these which is
> defined and existent is used to search for the user's personal
> customization.
> @@ -16,7 +16,7 @@ Index: doc/icewm.adoc
> the compiled-in default directory with default files
>
> The directories are searched in the above order, so any file located
> -@@ -407,8 +407,8 @@ creating the same directory hierarchy under
> `$HOME/.ic +@@ -408,8 +408,8 @@ creating the same directory hierarchy
> under `$HOME/.ic
> To customize icewm yourself, you could create the `$HOME/.icewm`
> directory and copy the files that you wish to modify, like
> diff --git patches/patch-lib_Makefile_in patches/patch-lib_Makefile_in
> index 9840ac6de6e..e08cc79fb0b 100644
> --- patches/patch-lib_Makefile_in
> +++ patches/patch-lib_Makefile_in
> @@ -3,7 +3,7 @@ $OpenBSD: patch-lib_Makefile_in,v 1.5 2019/11/09
> 14:06:33 fcambus Exp $ Index: lib/Makefile.in
> --- lib/Makefile.in.orig
> +++ lib/Makefile.in
> -@@ -362,7 +362,7 @@ target_vendor = @target_vendor@
> +@@ -365,7 +365,7 @@ target_vendor = @target_vendor@
> top_build_prefix = @top_build_prefix@
> top_builddir = @top_builddir@
> top_srcdir = @top_srcdir@
> diff --git patches/patch-man_icewm-preferences_pod
> patches/patch-man_icewm-preferences_pod index 5ca305f477d..
> 3fe1a64ae85 100644
> --- patches/patch-man_icewm-preferences_pod
> +++ patches/patch-man_icewm-preferences_pod
> @@ -3,7 +3,7 @@ $OpenBSD: patch-man_icewm-preferences_pod,v 1.5
> 2020/02/01 19:48:20 bket Exp $ Index: man/icewm-preferences.pod
> --- man/icewm-preferences.pod.orig
> +++ man/icewm-preferences.pod
> -@@ -982,7 +982,7 @@ B to allow B for
> B to allow
> B for B Combine all CPUs to one.
>
> diff --git patches/patch-src_default_h patches/patch-src_default_h
> index 84d5e4eea12..bab7dee4fa8 100644
> --- patches/patch-src_default_h
> +++ patches/patch-src_default_h
> @@ -2,7 +2,7 @@ $OpenBSD: patch-src_default_h,v 1.18 2020/02/01
> 19:48:20 bket Exp $ Index: src/default.h
> --- src/default.h.orig
> +++ src/default.h
> -@@ -38,7 +38,7 @@ XIV(bool, showTaskBar,
> true) +@@ -37,7 +37,7 @@ XIV(bool,
> showTaskBar, true) XIV(bool,
> taskBarAtTop, false) XIV(bool,
> taskBarKeepBelow, false) XIV(bool,
> taskBarShowClock, true) @@ -11,7 +11,7 @@ Index:
> src/default.h XIV(bool, taskBarShowApmAuto, true)
> XIV(bool, taskBarShowApmTime, true)
> XIV(bool, taskBarShowApmGraph, true)
> -@@ -173,10 +173,10 @@ XIV(int,
> netWorkAreaBehaviour, 0) +@@ -172,10 +172,10 @@
> XIV(int, netWorkAreaBehaviour, 0)
> XSV(const char *, acpiIgnoreBatteries, 0)
> XSV(const char *, mailBoxPath, 0)
> @@ -25,7 +25,7 @@ Index: src/default.h
> XSV(const char *, clockCommand, "xclock -name icewm
> -title Clock") XSV(const
Re: [clang+ld.bfd/all archs] Unbreak misc/osinfo/libosinfo, https HOMEPAGE
On Wed, 24 Jun 2020 00:51:14 +0200
Antoine Jacoutot wrote:
> On Tue, Jun 23, 2020 at 05:51:09PM +0200, Charlene Wendling wrote:
> > On Mon, 22 Jun 2020 10:26:31 +0100
> > Stuart Henderson wrote:
> >
> > > I haven't checked but the libraries are probably really needed at
> > > runtime (at least if it crashes at runtime, that's where the
> > > unwind functions are most likely to be used) on clang LLD arches
> > > too. The difference is that LLD will still link if there are
> > > missing symbols and crash at runtime if the functions are called,
> > > ld.bfd will check at link time.
> > >
> > > I would make it conditional on just clang rather than the linker
> > > too.
> > >
> > > --
> > > Sent from a phone, apologies for poor formatting.
> > > On 22 June 2020 09:43:51 Charlene Wendling
> > > wrote:
> > >
> > > > On Mon, 15 Jun 2020 21:12:48 +0100
> > > > Stuart Henderson wrote:
> > > >
> > > >> On 2020/06/15 21:46, Charlene Wendling wrote:
> > > >> > Hi,
> > > >> >
> > > >> > libosinfo's build fails on my macppc machine with:
> > > >> >
> > > >> > > /usr/ports/pobj/libosinfo-1.8.0/build-powerpc/osinfo/libosinfo-1.0.so.
> > > >> > > 1.5: undefined reference to `_Unwind_GetIP'
> > > >> > > /usr/ports/pobj/libosinfo-1.8.0/build-powerpc/osinfo/libosinfo-1.0.so.
> > > >> > > 1.5: undefined reference to `_Unwind_GetRegionStart'
> > > >> > (etc.)
> > > >> >
> > > >> > I don't know why macppc is the only one needing that, but
> > > >> > linking against libc++abi and pthread works, tests have only
> > > >> > 1 failure, as seen on amd64.
> > > >>
> > > >> btw, same on amd64 if you set USE_LLD=No.
> > > >
> > > > I missed the fact that mips64 had the same issue, so the diff
> > > > is now targeting clang+lb.bfd archs.
> > > >
> > > >> > While here i've moved HOMEPAGE to https.
> > > >> >
> > > >> > Comments/feedback are welcome,
> > > >> >
> > > >> > Charlène.
> > > >> >
> >
> > The _Unwind_* symbols seem to come from
> > osinfo_loader_process_user_path, i've tried to isolate the
> > situation in C and vala without success. I may be wrong about this
> > assumption.
> >
> > The only other case where it's needed is for lang/luajit, so i've
> > looked at the cvs log, and there was no clear reason invoked by
> > espie back then as well.
> >
> > So here is a diff that adds the WANTLIBs and LDFLAGS for all
> > base-clang archs. I've updated the comment in case someone may
> > think it's unneeded.
> >
> > Charlène.
>
> Did you try patching out -fexceptions or -fasynchronous-unwind-tables
> from meson.build?
It does work without -fexceptions or by providing -fno-exceptions as a
cflag [0]. As you mention patching, i've patched it away, there are no
more _Unwind_* symbols in the resulting binaries.
Charlène.
[0] https://bin.charlenew.xyz/libosinfo.log
Index: Makefile.inc
===
RCS file: /cvs/ports/misc/osinfo/Makefile.inc,v
retrieving revision 1.11
diff -u -p -u -p -r1.11 Makefile.inc
--- Makefile.inc31 May 2020 10:36:21 - 1.11
+++ Makefile.inc24 Jun 2020 14:59:09 -
@@ -6,7 +6,7 @@ CATEGORIES= misc devel
MAINTAINER=Antoine Jacoutot
-HOMEPAGE= http://libosinfo.org/
+HOMEPAGE= https://libosinfo.org/
# GPLv2+ - LGPLv2+
PERMIT_PACKAGE=Yes
Index: libosinfo/Makefile
===
RCS file: /cvs/ports/misc/osinfo/libosinfo/Makefile,v
retrieving revision 1.12
diff -u -p -u -p -r1.12 Makefile
--- libosinfo/Makefile 9 Jun 2020 07:14:44 - 1.12
+++ libosinfo/Makefile 24 Jun 2020 14:59:09 -
@@ -3,7 +3,7 @@
COMMENT= library for managing information about operating systems
DISTNAME= libosinfo-${V}
-REVISION= 0
+REVISION= 1
SHARED_LIBS += osinfo-1.01.5 # 0.1008.0
Index: libosinfo/patches/patch-meson_build
===
RCS file: libosinfo/patches/patch-meson_build
diff -N libosinfo/patches/patch-meson_build
--- /dev/null 1 Jan 1970 00:00:00 -
+++ libosinfo/patches/patch-meson_build 24 Jun 2020 14:59:09 -
@@ -0,0 +1,16 @@
+$Open
[clang+ld.bfd/all archs] Unbreak misc/osinfo/libosinfo, https HOMEPAGE
On Mon, 22 Jun 2020 10:26:31 +0100
Stuart Henderson wrote:
> I haven't checked but the libraries are probably really needed at
> runtime (at least if it crashes at runtime, that's where the unwind
> functions are most likely to be used) on clang LLD arches too. The
> difference is that LLD will still link if there are missing symbols
> and crash at runtime if the functions are called, ld.bfd will check
> at link time.
>
> I would make it conditional on just clang rather than the linker too.
>
> --
> Sent from a phone, apologies for poor formatting.
> On 22 June 2020 09:43:51 Charlene Wendling
> wrote:
>
> > On Mon, 15 Jun 2020 21:12:48 +0100
> > Stuart Henderson wrote:
> >
> >> On 2020/06/15 21:46, Charlene Wendling wrote:
> >> > Hi,
> >> >
> >> > libosinfo's build fails on my macppc machine with:
> >> >
> >> > > /usr/ports/pobj/libosinfo-1.8.0/build-powerpc/osinfo/libosinfo-1.0.so.
> >> > > 1.5: undefined reference to `_Unwind_GetIP'
> >> > > /usr/ports/pobj/libosinfo-1.8.0/build-powerpc/osinfo/libosinfo-1.0.so.
> >> > > 1.5: undefined reference to `_Unwind_GetRegionStart'
> >> > (etc.)
> >> >
> >> > I don't know why macppc is the only one needing that, but linking
> >> > against libc++abi and pthread works, tests have only 1 failure,
> >> > as seen on amd64.
> >>
> >> btw, same on amd64 if you set USE_LLD=No.
> >
> > I missed the fact that mips64 had the same issue, so the diff is now
> > targeting clang+lb.bfd archs.
> >
> >> > While here i've moved HOMEPAGE to https.
> >> >
> >> > Comments/feedback are welcome,
> >> >
> >> > Charlène.
> >> >
The _Unwind_* symbols seem to come from osinfo_loader_process_user_path,
i've tried to isolate the situation in C and vala without success. I may
be wrong about this assumption.
The only other case where it's needed is for lang/luajit, so i've
looked at the cvs log, and there was no clear reason invoked by espie
back then as well.
So here is a diff that adds the WANTLIBs and LDFLAGS for all base-clang
archs. I've updated the comment in case someone may think it's unneeded.
Charlène.
Index: Makefile.inc
===
RCS file: /cvs/ports/misc/osinfo/Makefile.inc,v
retrieving revision 1.11
diff -u -p -u -p -r1.11 Makefile.inc
--- Makefile.inc31 May 2020 10:36:21 - 1.11
+++ Makefile.inc23 Jun 2020 15:39:20 -
@@ -6,7 +6,7 @@ CATEGORIES= misc devel
MAINTAINER=Antoine Jacoutot
-HOMEPAGE= http://libosinfo.org/
+HOMEPAGE= https://libosinfo.org/
# GPLv2+ - LGPLv2+
PERMIT_PACKAGE=Yes
Index: libosinfo/Makefile
===
RCS file: /cvs/ports/misc/osinfo/libosinfo/Makefile,v
retrieving revision 1.12
diff -u -p -u -p -r1.12 Makefile
--- libosinfo/Makefile 9 Jun 2020 07:14:44 - 1.12
+++ libosinfo/Makefile 23 Jun 2020 15:39:20 -
@@ -3,7 +3,7 @@
COMMENT= library for managing information about operating systems
DISTNAME= libosinfo-${V}
-REVISION= 0
+REVISION= 1
SHARED_LIBS += osinfo-1.01.5 # 0.1008.0
@@ -29,5 +29,13 @@ TEST_DEPENDS=misc/osinfo/osinfo-db
CONFIGURE_ARGS=
-Dwith-pci-ids-path=${LOCALBASE}/share/hwdata/pnp.ids \
-Dwith-usb-ids-path=${LOCALBASE}/share/hwdata/usb.ids
+
+.include
+# Fix undefined reference errors to _Unwind_* on clang+ld.bfd archs;
+# it may cause runtime issues on lld archs as well.
+.if ${PROPERTIES:Mclang}
+CONFIGURE_ENV += LDFLAGS="${LDFLAGS} -lc++abi -lpthread"
+WANTLIB += c++abi pthread
+.endif
.include
Index: osinfo-db/Makefile
===
RCS file: /cvs/ports/misc/osinfo/osinfo-db/Makefile,v
retrieving revision 1.33
diff -u -p -u -p -r1.33 Makefile
--- osinfo-db/Makefile 31 May 2020 10:36:36 - 1.33
+++ osinfo-db/Makefile 23 Jun 2020 15:39:20 -
@@ -3,6 +3,7 @@
COMMENT= osinfo database files
V= 20200529
+REVISION= 0
DISTNAME= osinfo-db-${V}
EXTRACT_SUFX= .tar.xz
Index: osinfo-db-tools/Makefile
===
RCS file: /cvs/ports/misc/osinfo/osinfo-db-tools/Makefile,v
retrieving revision 1.11
diff -u -p -u -p -r1.11 Makefile
--- osinfo-db-tools/Makefile31 May 2020 10:36:21 - 1.11
+++ osinfo-db-tools/Makefile23 Jun 2020 15:39:21 -
@@ -3,6 +3,7 @@
COMMENT= tools to manage the libosinfo database
DISTNAME= osinfo-db-tools-${V}
+REVISION= 0
WANTLIB += archive c gio-2.0 glib-2.0 gobject-2.0 intl json-glib-1.0
WANTLIB += soup-2.4 xml2
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/06/22 03:43:45 Modified files: x11/agar : Makefile.inc x11/agar/agar : Makefile x11/agar/agar/pkg: PLIST x11/agar/test : Makefile Log message: agar: move HOMEPAGE to https, update plist, mark BROKEN-powerpc On top of another fix for some asm code on powerpc, naddy@ proposed me to fix the AltiVec runtime detection that broke the build. I've found out that running agartest makes my 2 powerpc machines freeze, with or without AltiVec. « I won't object to that » - naddy@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/06/22 03:00:00 Modified files: x11/gnome/libdazzle: Makefile Added files: x11/gnome/libdazzle/patches: patch-src_util_dzl-counter_c patch-src_util_dzl-counter_h Log message: libdazzle: fix some consumers' build on powerpc libdazzle uses 64-bit __sync_* primitives that are not supported on powerpc, use the fallback code on this arch instead (merged upstream).
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/06/22 02:48:43 ports/x11/gnome/libdazzle/patches Update of /cvs/ports/x11/gnome/libdazzle/patches In directory cvs.openbsd.org:/tmp/cvs-serv67989/patches Log Message: Directory /cvs/ports/x11/gnome/libdazzle/patches added to the repository
Re: [macppc] Unbreak misc/osinfo/libosinfo, https HOMEPAGE
On Mon, 15 Jun 2020 21:12:48 +0100
Stuart Henderson wrote:
> On 2020/06/15 21:46, Charlene Wendling wrote:
> > Hi,
> >
> > libosinfo's build fails on my macppc machine with:
> >
> > > /usr/ports/pobj/libosinfo-1.8.0/build-powerpc/osinfo/libosinfo-1.0.so.
> > > 1.5: undefined reference to `_Unwind_GetIP'
> > > /usr/ports/pobj/libosinfo-1.8.0/build-powerpc/osinfo/libosinfo-1.0.so.
> > > 1.5: undefined reference to `_Unwind_GetRegionStart'
> > (etc.)
> >
> > I don't know why macppc is the only one needing that, but linking
> > against libc++abi and pthread works, tests have only 1 failure, as
> > seen on amd64.
>
> btw, same on amd64 if you set USE_LLD=No.
I missed the fact that mips64 had the same issue, so the diff is now
targeting clang+lb.bfd archs.
> > While here i've moved HOMEPAGE to https.
> >
> > Comments/feedback are welcome,
> >
> > Charlène.
> >
Index: Makefile.inc
===
RCS file: /cvs/ports/misc/osinfo/Makefile.inc,v
retrieving revision 1.11
diff -u -p -u -p -r1.11 Makefile.inc
--- Makefile.inc31 May 2020 10:36:21 - 1.11
+++ Makefile.inc20 Jun 2020 19:29:23 -
@@ -6,7 +6,7 @@ CATEGORIES= misc devel
MAINTAINER=Antoine Jacoutot
-HOMEPAGE= http://libosinfo.org/
+HOMEPAGE= https://libosinfo.org/
# GPLv2+ - LGPLv2+
PERMIT_PACKAGE=Yes
Index: libosinfo/Makefile
===
RCS file: /cvs/ports/misc/osinfo/libosinfo/Makefile,v
retrieving revision 1.12
diff -u -p -u -p -r1.12 Makefile
--- libosinfo/Makefile 9 Jun 2020 07:14:44 - 1.12
+++ libosinfo/Makefile 20 Jun 2020 19:29:23 -
@@ -3,7 +3,7 @@
COMMENT= library for managing information about operating systems
DISTNAME= libosinfo-${V}
-REVISION= 0
+REVISION= 1
SHARED_LIBS += osinfo-1.01.5 # 0.1008.0
@@ -29,5 +29,12 @@ TEST_DEPENDS=misc/osinfo/osinfo-db
CONFIGURE_ARGS=
-Dwith-pci-ids-path=${LOCALBASE}/share/hwdata/pnp.ids \
-Dwith-usb-ids-path=${LOCALBASE}/share/hwdata/usb.ids
+
+.include
+# Fix undefined reference errors to _Unwind_*
+.if ${PROPERTIES:Mclang} && !${PROPERTIES:Mlld}
+CONFIGURE_ENV += LDFLAGS="${LDFLAGS} -lc++abi -lpthread"
+WANTLIB += c++abi pthread
+.endif
.include
Index: osinfo-db/Makefile
===
RCS file: /cvs/ports/misc/osinfo/osinfo-db/Makefile,v
retrieving revision 1.33
diff -u -p -u -p -r1.33 Makefile
--- osinfo-db/Makefile 31 May 2020 10:36:36 - 1.33
+++ osinfo-db/Makefile 20 Jun 2020 19:29:23 -
@@ -3,6 +3,7 @@
COMMENT= osinfo database files
V= 20200529
+REVISION= 0
DISTNAME= osinfo-db-${V}
EXTRACT_SUFX= .tar.xz
Index: osinfo-db-tools/Makefile
===
RCS file: /cvs/ports/misc/osinfo/osinfo-db-tools/Makefile,v
retrieving revision 1.11
diff -u -p -u -p -r1.11 Makefile
--- osinfo-db-tools/Makefile31 May 2020 10:36:21 - 1.11
+++ osinfo-db-tools/Makefile20 Jun 2020 19:29:23 -
@@ -3,6 +3,7 @@
COMMENT= tools to manage the libosinfo database
DISTNAME= osinfo-db-tools-${V}
+REVISION= 0
WANTLIB += archive c gio-2.0 glib-2.0 gobject-2.0 intl json-glib-1.0
WANTLIB += soup-2.4 xml2
Re: [macppc] Unbreak x11/agar/agar
Hi,
On Tue, 16 Jun 2020 22:52:33 +0200
Christian Weisgerber wrote:
> Charlene Wendling:
>
> > > http://build-failures.rhaalovely.net/powerpc/2020-05-26/x11/agar/agar.log
> >
> > `rlwimi' needs a 32-bit integer to work with, taken from another
> > upstream for the same code [0].
>
> That's the correct fix.
>
> (Although I have to wonder whether that handcrafted optimization
> is even worth it, or whether the compiler will produce equivalent
> code from the generic C code anyway.)
>
> > Later it breaks with:
> >
> > > cpuinfo.c:289:23: error: use of undeclared identifier
> > > 'IllegalInsn' cpuinfo.c:290:14: error: use of undeclared
> > > identifier 'jmpbuf'
>
> You could fix the #ifdef maze so this attempt at run-time Altivec
> detection builds.
>
> Even better, you could add code to query machdep.altivec via sysctl(3)
> on OpenBSD.
>
> --
> Christian "naddy" Weisgerber
> na...@mips.inka.de
>
I did that in the attached diff. It appears that running agartest
freezes my PowerBook and trashed my filesystems without a trace
as soon as i start a benchmark, something i did not do originally.
I tried on my Mac mini G4, same result, but the machine is still
usable. With or without AltiVec that is.
I'm gone assessing damages^W^W reinstalling OpenBSD on the PowerBook
now, so i won't be able to test stuff before some days.
As such, the below diff moves HOMEPAGE to https, and at least updates
PLIST. I prefer to mark it BROKEN-powerpc.
Charlène.
Index: Makefile.inc
===
RCS file: /cvs/ports/x11/agar/Makefile.inc,v
retrieving revision 1.2
diff -u -p -u -p -r1.2 Makefile.inc
--- Makefile.inc13 Jul 2019 10:59:27 - 1.2
+++ Makefile.inc18 Jun 2020 15:11:36 -
@@ -1,11 +1,13 @@
# $OpenBSD: Makefile.inc,v 1.2 2019/07/13 10:59:27 sthen Exp $
+BROKEN-powerpc = agartest causes system freezes
+
V ?= 1.5.0
DISTNAME ?=agar-$V
CATEGORIES ?= x11
-HOMEPAGE ?=http://libagar.org/
+HOMEPAGE ?=https://libagar.org/
# BSD
PERMIT_PACKAGE ?= Yes
Index: agar/Makefile
===
RCS file: /cvs/ports/x11/agar/agar/Makefile,v
retrieving revision 1.5
diff -u -p -u -p -r1.5 Makefile
--- agar/Makefile 23 Oct 2017 17:11:03 - 1.5
+++ agar/Makefile 18 Jun 2020 15:11:36 -
@@ -3,7 +3,7 @@
COMMENT = cross-platform widget toolkit
PKGNAME = agar-$V
-REVISION = 2
+REVISION = 3
SHARED_LIBS += ag_au 0.0 # 5.0
SHARED_LIBS += ag_core 0.0 # 5.0
Index: agar/pkg/PLIST
===
RCS file: /cvs/ports/x11/agar/agar/pkg/PLIST,v
retrieving revision 1.1.1.1
diff -u -p -u -p -r1.1.1.1 PLIST
--- agar/pkg/PLIST 10 Oct 2016 09:33:17 - 1.1.1.1
+++ agar/pkg/PLIST 18 Jun 2020 15:11:36 -
@@ -406,22 +406,22 @@ include/agar/agar/vg/vg_text.h
include/agar/agar/vg/vg_tool.h
include/agar/agar/vg/vg_tools.h
include/agar/agar/vg/vg_view.h
-lib/libag_au.a
+@static-lib lib/libag_au.a
lib/libag_au.la
@lib lib/libag_au.so.${LIBag_au_VERSION}
-lib/libag_core.a
+@static-lib lib/libag_core.a
lib/libag_core.la
@lib lib/libag_core.so.${LIBag_core_VERSION}
-lib/libag_dev.a
+@static-lib lib/libag_dev.a
lib/libag_dev.la
@lib lib/libag_dev.so.${LIBag_dev_VERSION}
-lib/libag_gui.a
+@static-lib lib/libag_gui.a
lib/libag_gui.la
@lib lib/libag_gui.so.${LIBag_gui_VERSION}
-lib/libag_math.a
+@static-lib lib/libag_math.a
lib/libag_math.la
@lib lib/libag_math.so.${LIBag_math_VERSION}
-lib/libag_vg.a
+@static-lib lib/libag_vg.a
lib/libag_vg.la
@lib lib/libag_vg.so.${LIBag_vg_VERSION}
@man man/man3/AG_Anim.3
Index: test/Makefile
===
RCS file: /cvs/ports/x11/agar/test/Makefile,v
retrieving revision 1.1.1.1
diff -u -p -u -p -r1.1.1.1 Makefile
--- test/Makefile 10 Oct 2016 09:33:17 - 1.1.1.1
+++ test/Makefile 18 Jun 2020 15:11:36 -
@@ -3,6 +3,7 @@
COMMENT = interactive test suite for agar
PKGNAME = agartest-$V
+REVISION = 0
WANTLIB += GL SDL X11 Xinerama ag_core ag_dev ag_gui ag_math c
WANTLIB += db fontconfig freetype jpeg m png pthread z
agar.ppc.diff
Description: Binary data
[macppc] x11/gnome/photos: remove BROKEN-powerpc
Hi, To actually build Photos on macppc, you need 2 other diffs waiting for OKs as well: - libdazzle [0] (the BROKEN reason was due to it) - libosinfo [1] Comments/feedback are welcome, Charlène. [0] https://marc.info/?l=openbsd-ports=159199473212267=2 [1] https://marc.info/?l=openbsd-ports=159225047517839=2 Index: Makefile === RCS file: /cvs/ports/x11/gnome/photos/Makefile,v retrieving revision 1.80 diff -u -p -u -p -r1.80 Makefile --- Makefile21 May 2020 10:07:12 - 1.80 +++ Makefile15 Jun 2020 19:42:34 - @@ -1,7 +1,5 @@ # $OpenBSD: Makefile,v 1.80 2020/05/21 10:07:12 ajacoutot Exp $ -BROKEN-powerpc=undefined reference to __sync_add_and_fetch_8 - COMMENT= access, organize and share your photos on GNOME GNOME_PROJECT= gnome-photos
[macppc] Unbreak misc/osinfo/libosinfo, https HOMEPAGE
Hi,
libosinfo's build fails on my macppc machine with:
> /usr/ports/pobj/libosinfo-1.8.0/build-powerpc/osinfo/libosinfo-1.0.so.1.5:
> undefined reference to `_Unwind_GetIP'
> /usr/ports/pobj/libosinfo-1.8.0/build-powerpc/osinfo/libosinfo-1.0.so.1.5:
> undefined reference to `_Unwind_GetRegionStart'
(etc.)
I don't know why macppc is the only one needing that, but linking
against libc++abi and pthread works, tests have only 1 failure, as
seen on amd64.
While here i've moved HOMEPAGE to https.
Comments/feedback are welcome,
Charlène.
Index: Makefile.inc
===
RCS file: /cvs/ports/misc/osinfo/Makefile.inc,v
retrieving revision 1.11
diff -u -p -u -p -r1.11 Makefile.inc
--- Makefile.inc31 May 2020 10:36:21 - 1.11
+++ Makefile.inc15 Jun 2020 14:11:38 -
@@ -6,7 +6,7 @@ CATEGORIES= misc devel
MAINTAINER=Antoine Jacoutot
-HOMEPAGE= http://libosinfo.org/
+HOMEPAGE= https://libosinfo.org/
# GPLv2+ - LGPLv2+
PERMIT_PACKAGE=Yes
Index: libosinfo/Makefile
===
RCS file: /cvs/ports/misc/osinfo/libosinfo/Makefile,v
retrieving revision 1.12
diff -u -p -u -p -r1.12 Makefile
--- libosinfo/Makefile 9 Jun 2020 07:14:44 - 1.12
+++ libosinfo/Makefile 15 Jun 2020 14:11:38 -
@@ -3,7 +3,7 @@
COMMENT= library for managing information about operating systems
DISTNAME= libosinfo-${V}
-REVISION= 0
+REVISION= 1
SHARED_LIBS += osinfo-1.01.5 # 0.1008.0
@@ -29,5 +29,11 @@ TEST_DEPENDS=misc/osinfo/osinfo-db
CONFIGURE_ARGS=
-Dwith-pci-ids-path=${LOCALBASE}/share/hwdata/pnp.ids \
-Dwith-usb-ids-path=${LOCALBASE}/share/hwdata/usb.ids
+
+# Fix undefined reference errors to _Unwind_*
+.if ${MACHINE_ARCH:Mpowerpc}
+CONFIGURE_ENV += LDFLAGS="${LDFLAGS} -lc++abi -lpthread"
+WANTLIB += c++abi pthread
+.endif
.include
Index: osinfo-db/Makefile
===
RCS file: /cvs/ports/misc/osinfo/osinfo-db/Makefile,v
retrieving revision 1.33
diff -u -p -u -p -r1.33 Makefile
--- osinfo-db/Makefile 31 May 2020 10:36:36 - 1.33
+++ osinfo-db/Makefile 15 Jun 2020 14:11:38 -
@@ -3,6 +3,7 @@
COMMENT= osinfo database files
V= 20200529
+REVISION= 0
DISTNAME= osinfo-db-${V}
EXTRACT_SUFX= .tar.xz
Index: osinfo-db-tools/Makefile
===
RCS file: /cvs/ports/misc/osinfo/osinfo-db-tools/Makefile,v
retrieving revision 1.11
diff -u -p -u -p -r1.11 Makefile
--- osinfo-db-tools/Makefile31 May 2020 10:36:21 - 1.11
+++ osinfo-db-tools/Makefile15 Jun 2020 14:11:38 -
@@ -3,6 +3,7 @@
COMMENT= tools to manage the libosinfo database
DISTNAME= osinfo-db-tools-${V}
+REVISION= 0
WANTLIB += archive c gio-2.0 glib-2.0 gobject-2.0 intl json-glib-1.0
WANTLIB += soup-2.4 xml2
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/06/15 05:48:04 Modified files: devel/p5-Test-Compile: Makefile distinfo Log message: p5-Test-Compile: update to 2.4.0 Changelog: https://metacpan.org/changes/distribution/Test-Compile OK afresh1@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/06/15 05:43:22 Modified files: devel/p5-Devel-Hide: Makefile distinfo Log message: p5-Devel-Hide: update to 0.0013, change maintainer email address Changelog: https://metacpan.org/changes/distribution/Devel-Hide OK afresh1@
[BE_ARCHS] Fix sysutils/smartmontools runtime
Hi, As i took a peek at misc@, i've noticed that there was a smartmontools fix for big endian archs [0] proposed by Marek on upstream's GitHub [1]. Note that i'm only backporting that, the switch to the new interface for OpenBSD requires way more patching. It builds and works fine on macppc (M.2 SSD with PATA adapter), and i tested it on amd64 as well. Comments/feedback are welcome, Charlène. [0] https://marc.info/?l=openbsd-misc=159138099215533=2 [1] https://github.com/smartmontools/smartmontools/pull/56 Index: Makefile === RCS file: /cvs/ports/sysutils/smartmontools/Makefile,v retrieving revision 1.43 diff -u -p -u -p -r1.43 Makefile --- Makefile19 Feb 2020 17:46:52 - 1.43 +++ Makefile14 Jun 2020 14:13:59 - @@ -4,6 +4,7 @@ COMMENT=control and monitor storage sy # XXX at update time check whether C++11 is actually needed DISTNAME= smartmontools-7.1 +REVISION= 0 CATEGORIES=sysutils HOMEPAGE= https://www.smartmontools.org/ Index: patches/patch-os_openbsd_cpp === RCS file: patches/patch-os_openbsd_cpp diff -N patches/patch-os_openbsd_cpp --- /dev/null 1 Jan 1970 00:00:00 - +++ patches/patch-os_openbsd_cpp14 Jun 2020 14:13:59 - @@ -0,0 +1,123 @@ +$OpenBSD$ + +From 61c49a66d6ee9df516fb9757c7fcce18ea508f6a Mon Sep 17 00:00:00 2001 +From: Marek Benc +Date: Thu, 4 Jun 2020 16:49:48 +0200 +Subject: [PATCH] Fix runtime problems on big-engian OpenBSD platforms. +Can be removed for smarmontools>7.1 + +Index: os_openbsd.cpp +--- os_openbsd.cpp.orig os_openbsd.cpp +@@ -184,7 +184,7 @@ ata_command_interface(int fd, smart_command_set comman + req.command = ATAPI_SMART; + req.databuf = (caddr_t) inbuf; + req.datalen = sizeof(inbuf); +-req.cylinder = htole16(WDSMART_CYL); ++req.cylinder = WDSMART_CYL; + req.timeout = 1000; + copydata = 1; + break; +@@ -194,7 +194,7 @@ ata_command_interface(int fd, smart_command_set comman + req.command = ATAPI_SMART; + req.databuf = (caddr_t) inbuf; + req.datalen = sizeof(inbuf); +-req.cylinder = htole16(WDSMART_CYL); ++req.cylinder = WDSMART_CYL; + req.timeout = 1000; + copydata = 1; + break; +@@ -204,7 +204,7 @@ ata_command_interface(int fd, smart_command_set comman + req.command = ATAPI_SMART; + req.databuf = (caddr_t) inbuf; + req.datalen = sizeof(inbuf); +-req.cylinder = htole16(WDSMART_CYL); ++req.cylinder = WDSMART_CYL; + req.sec_num = select; + req.sec_count = 1; + req.timeout = 1000; +@@ -217,7 +217,7 @@ ata_command_interface(int fd, smart_command_set comman + req.command = ATAPI_SMART; + req.databuf = (caddr_t) inbuf; + req.datalen = sizeof(inbuf); +-req.cylinder = htole16(WDSMART_CYL); ++req.cylinder = WDSMART_CYL; + req.sec_num = select; + req.sec_count = 1; + req.timeout = 1000; +@@ -242,14 +242,14 @@ ata_command_interface(int fd, smart_command_set comman + req.flags = ATACMD_READ; + req.features = ATA_SMART_ENABLE; + req.command = ATAPI_SMART; +-req.cylinder = htole16(WDSMART_CYL); ++req.cylinder = WDSMART_CYL; + req.timeout = 1000; + break; + case DISABLE: + req.flags = ATACMD_READ; + req.features = ATA_SMART_DISABLE; + req.command = ATAPI_SMART; +-req.cylinder = htole16(WDSMART_CYL); ++req.cylinder = WDSMART_CYL; + req.timeout = 1000; + break; + case AUTO_OFFLINE: +@@ -259,7 +259,7 @@ ata_command_interface(int fd, smart_command_set comman + req.command = ATAPI_SMART; + req.databuf = (caddr_t) inbuf; + req.datalen = sizeof(inbuf); +-req.cylinder = htole16(WDSMART_CYL); ++req.cylinder = WDSMART_CYL; + req.sec_num = select; + req.sec_count = 1; + req.timeout = 1000; +@@ -268,7 +268,7 @@ ata_command_interface(int fd, smart_command_set comman + req.flags = ATACMD_READ; + req.features = ATA_SMART_AUTOSAVE;/* XXX missing from wdcreg.h */ + req.command = ATAPI_SMART; +-req.cylinder = htole16(WDSMART_CYL); ++req.cylinder = WDSMART_CYL; + req.sec_count = 0xf1; + /* to enable autosave */ + req.timeout = 1000; +@@ -280,7 +280,7 @@ ata_command_interface(int fd, smart_command_set comman + req.command = ATAPI_SMART; + req.databuf = (caddr_t) inbuf; + req.datalen = sizeof(inbuf); +-req.cylinder = htole16(WDSMART_CYL); ++req.cylinder = WDSMART_CYL; + req.sec_num = select; + req.sec_count = 1; + req.timeout = 1000; +@@ -291,7 +291,7 @@ ata_command_interface(int fd, smart_command_set comman + req.flags = ATACMD_READ; + req.features = ATA_SMART_STATUS; + req.command = ATAPI_SMART; +-req.cylinder = htole16(WDSMART_CYL); ++req.cylinder = WDSMART_CYL; + req.timeout = 1000; + break; + case CHECK_POWER_MODE: +@@ -316,11 +316,11
[macppc] Unbreak x11/agar/agar
Hi,
> http://build-failures.rhaalovely.net/powerpc/2020-05-26/x11/agar/agar.log
`rlwimi' needs a 32-bit integer to work with, taken from another
upstream for the same code [0].
Later it breaks with:
> cpuinfo.c:289:23: error: use of undeclared identifier 'IllegalInsn'
> cpuinfo.c:290:14: error: use of undeclared identifier 'jmpbuf'
It did not occur in this ifdef hell back in the gcc days, because unlike
clang, it does not define __ppc__. Dropping AltiVec support fixes the
issue.
With the below diff it builds on macppc, x11/agar/test runs fine.
Comments/feedback are welcome,
Charlène.
[0] https://github.com/DCurrent/openbor/pull/188
Index: Makefile
===
RCS file: /cvs/ports/x11/agar/agar/Makefile,v
retrieving revision 1.5
diff -u -p -u -p -r1.5 Makefile
--- Makefile23 Oct 2017 17:11:03 - 1.5
+++ Makefile14 Jun 2020 06:31:49 -
@@ -3,7 +3,7 @@
COMMENT = cross-platform widget toolkit
PKGNAME = agar-$V
-REVISION = 2
+REVISION = 3
SHARED_LIBS += ag_au 0.0 # 5.0
SHARED_LIBS += ag_core 0.0 # 5.0
@@ -28,6 +28,7 @@ LIB_DEPENDS = audio/libsndfile \
CONFIGURE_ARGS += --with-gl=${X11BASE} \
--with-x=${X11BASE} \
--with-portaudio=${LOCALBASE} \
- --with-sndfile=${LOCALBASE}
+ --with-sndfile=${LOCALBASE} \
+ --with-altivec=no
.include
Index: patches/patch-core_byteswap_h
===
RCS file: patches/patch-core_byteswap_h
diff -N patches/patch-core_byteswap_h
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-core_byteswap_h 14 Jun 2020 06:31:49 -
@@ -0,0 +1,23 @@
+$OpenBSD$
+
+powerpc fix for:
+unsupported inline asm: input with type 'int' matching output with type
+'u_int16_t' (aka 'unsigned short')
+
+Index: core/byteswap.h
+--- core/byteswap.h.orig
core/byteswap.h
+@@ -37,11 +37,11 @@ AG_Swap16(Uint16 x)
+ static __inline__ Uint16
+ AG_Swap16(Uint16 x)
+ {
+- Uint16 rv;
++ Uint32 rv;
+ __asm__("rlwimi %0,%2,8,16,23" :
+ "=" (rv) :
+ "0" (x >> 8), "r" (x));
+- return (rv);
++ return (Uint16)(rv);
+ }
+ #else
+ static __inline__ Uint16
Index: pkg/PLIST
===
RCS file: /cvs/ports/x11/agar/agar/pkg/PLIST,v
retrieving revision 1.1.1.1
diff -u -p -u -p -r1.1.1.1 PLIST
--- pkg/PLIST 10 Oct 2016 09:33:17 - 1.1.1.1
+++ pkg/PLIST 14 Jun 2020 06:31:49 -
@@ -34,7 +34,6 @@ include/agar/agar/config/ag_debug_gui.h
include/agar/agar/config/ag_legacy.h
include/agar/agar/config/ag_network.h
include/agar/agar/config/ag_threads.h
-include/agar/agar/config/altivec_cflags.h
include/agar/agar/config/bindir.h
include/agar/agar/config/clock_cflags.h
include/agar/agar/config/clock_libs.h
@@ -406,22 +405,22 @@ include/agar/agar/vg/vg_text.h
include/agar/agar/vg/vg_tool.h
include/agar/agar/vg/vg_tools.h
include/agar/agar/vg/vg_view.h
-lib/libag_au.a
+@static-lib lib/libag_au.a
lib/libag_au.la
@lib lib/libag_au.so.${LIBag_au_VERSION}
-lib/libag_core.a
+@static-lib lib/libag_core.a
lib/libag_core.la
@lib lib/libag_core.so.${LIBag_core_VERSION}
-lib/libag_dev.a
+@static-lib lib/libag_dev.a
lib/libag_dev.la
@lib lib/libag_dev.so.${LIBag_dev_VERSION}
-lib/libag_gui.a
+@static-lib lib/libag_gui.a
lib/libag_gui.la
@lib lib/libag_gui.so.${LIBag_gui_VERSION}
-lib/libag_math.a
+@static-lib lib/libag_math.a
lib/libag_math.la
@lib lib/libag_math.so.${LIBag_math_VERSION}
-lib/libag_vg.a
+@static-lib lib/libag_vg.a
lib/libag_vg.la
@lib lib/libag_vg.so.${LIBag_vg_VERSION}
@man man/man3/AG_Anim.3
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/06/13 11:38:49 Modified files: security/aircrack-ng: Makefile security/aircrack-ng/pkg: PLIST Added files: security/aircrack-ng/patches: patch-build_m4_aircrack_ng_simd_m4 Removed files: security/aircrack-ng/pkg: PFRAG.ppc Log message: aircrack-ng: fix the build on powerpc aircrack-ng uses an old version of John the Ripper, the whole Altivec acceleration code does not build since the move to clang/compiler_rt; it fails with ports-gcc as well. John codebase massively changed in this regard since then, making backports too massive. Remove AltiVec support and associated PFRAG, with a patch ensuring that AltiVec and POWER8 are *really* disabled at build time (from upstream). OK benoit@ (maintainer)
Re: [macppc] Unbreak security/aircrack-ng, drop simd
Ping.
On Sat, 6 Jun 2020 04:22:26 +0200
Charlene Wendling wrote:
>
> I've removed the .orig diff. Sorry for the noise.
>
> On Sat, 6 Jun 2020 02:01:27 +0200
> Charlene Wendling wrote:
>
> > Hi,
> >
> > > http://build-failures.rhaalovely.net/powerpc/last/security/aircrack-ng.log
> >
> > The faulty code is based on a vintage version of John The Ripper.
> >
> > It would need a 'long long' here, but then it moves the error in
> > other parts of the code, and so it would need a lot of patches as
> > upstream John reworked all that part.
> >
> > As such i'm proposing to simply remove the AltiVec support so it
> > builds, and to add a fix that allows really to disable AltiVec as
> > build flags were no honoured.
> >
> > Comments/feedback are welcome,
> >
> > Charlène.
>
Index: Makefile
===
RCS file: /cvs/ports/security/aircrack-ng/Makefile,v
retrieving revision 1.29
diff -u -p -u -p -r1.29 Makefile
--- Makefile6 Dec 2019 14:29:40 - 1.29
+++ Makefile6 Jun 2020 02:19:55 -
@@ -52,10 +52,10 @@ PKG_ARGS+= -Dx86=0
#.else
PKG_ARGS+= -Darm=0
#.endif
+# XXX AltiVec optimizations can't be built with clang, and POWER8 ones
+# are useless on powerpc
.if ${MACHINE_ARCH} == "powerpc"
-PKG_ARGS+= -Dppc=1
-.else
-PKG_ARGS+= -Dppc=0
+MAKE_ENV+= ALTIVEC=false POWER8=false
.endif
Index: patches/patch-build_m4_aircrack_ng_simd_m4
===
RCS file: patches/patch-build_m4_aircrack_ng_simd_m4
diff -N patches/patch-build_m4_aircrack_ng_simd_m4
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-build_m4_aircrack_ng_simd_m4 6 Jun 2020 02:19:55 -
@@ -0,0 +1,25 @@
+$OpenBSD$
+
+Ensure that disabling AltiVec and POWER8 instructions with MAKE_ENV is honoured
+even if the build machine supports them, from:
+https://github.com/aircrack-ng/aircrack-ng/issues/1941
+
+Index: build/m4/aircrack_ng_simd.m4
+--- build/m4/aircrack_ng_simd.m4.orig
build/m4/aircrack_ng_simd.m4
+@@ -132,7 +132,6 @@ then
+ AX_CHECK_COMPILE_FLAG([-maltivec], [
+ AX_APPEND_FLAG(-maltivec, [ppc_altivec_[]_AC_LANG_ABBREV[]flags])
+ AC_SUBST(ppc_altivec_[]_AC_LANG_ABBREV[]flags)
+-ALTIVEC_FOUND=1
+ ])
+
+ AX_CHECK_COMPILE_FLAG([-mabi=altivec], [
+@@ -148,7 +147,6 @@ then
+ AX_CHECK_COMPILE_FLAG([-mpower8-vector], [
+ AX_APPEND_FLAG(-mpower8-vector,
[ppc_altivec_[]_AC_LANG_ABBREV[]flags])
+ AC_SUBST(ppc_altivec_[]_AC_LANG_ABBREV[]flags)
+-POWER8_FOUND=1
+ ])
+ fi
+
Index: pkg/PFRAG.ppc
===
RCS file: pkg/PFRAG.ppc
diff -N pkg/PFRAG.ppc
--- pkg/PFRAG.ppc 12 Aug 2019 14:40:04 - 1.2
+++ /dev/null 1 Jan 1970 00:00:00 -
@@ -1,4 +0,0 @@
-@comment $OpenBSD: PFRAG.ppc,v 1.2 2019/08/12 14:40:04 cwen Exp $
-lib/libaircrack-crypto-ppc-altivec-1.3.0.so
-lib/libaircrack-crypto-ppc-altivec.la
-lib/libaircrack-crypto-ppc-altivec.so
Index: pkg/PLIST
===
RCS file: /cvs/ports/security/aircrack-ng/pkg/PLIST,v
retrieving revision 1.4
diff -u -p -u -p -r1.4 PLIST
--- pkg/PLIST 23 Feb 2019 22:23:25 - 1.4
+++ pkg/PLIST 6 Jun 2020 02:19:55 -
@@ -10,7 +10,6 @@
@bin bin/wpaclean
lib/libaircrack-crypto-1.3.0.so
%%x86%%
-%%ppc%%
%%arm%%
lib/libaircrack-crypto.la
lib/libaircrack-crypto.so
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/06/13 06:07:22 Modified files: lang/gprolog : Makefile Log message: gprolog: mark BROKEN-powerpc OK sthen@
[macppc] Mark lang/gprolog BROKEN
> http://build-failures.rhaalovely.net/powerpc/2020-05-26/lang/gprolog.log It has been broken since at least one year and a half. I can't fix it, and clearly nobody has provided a fix since then. I'm proposing to mark it BROKEN. OK? Charlène. Index: Makefile === RCS file: /cvs/ports/lang/gprolog/Makefile,v retrieving revision 1.28 diff -u -p -u -p -r1.28 Makefile --- Makefile12 Jul 2019 20:47:18 - 1.28 +++ Makefile13 Jun 2020 10:57:32 - @@ -1,6 +1,7 @@ # $OpenBSD: Makefile,v 1.28 2019/07/12 20:47:18 sthen Exp $ BROKEN-i386= various errors from gplc during build (segv, "fatal error", etc at various different points) +BROKEN-powerpc=ld.so: pl2wam: relocation failed ONLY_FOR_ARCHS=amd64 i386 powerpc sparc64
[ld.bfd] Don't build lang/squeak/*
Hi,
> http://build-failures.rhaalovely.net/mips64/2020-05-28/lang/squeak/vm.log
(BROKEN for the same reason on sparc64)
> http://build-failures.rhaalovely.net/powerpc/2020-05-26/lang/squeak/vm.log
(I've attached a diff with fixes that leads to the same situation on
macppc, it may be useful if one day powerpc moves to lld)
It breaks with undefined references to its own internal plugins, and i
don't see why it should.
kmos@ marked the vm BROKEN-sparc64, but not the other subports, this
diff changes that, so we don't build ports that we can't use.
Comments/feedback are welcome,
Charlène.
Index: Makefile.inc
===
RCS file: /cvs/ports/lang/squeak/Makefile.inc,v
retrieving revision 1.7
diff -u -p -u -p -r1.7 Makefile.inc
--- Makefile.inc12 Jul 2019 20:47:23 - 1.7
+++ Makefile.inc13 Jun 2020 10:43:30 -
@@ -1,5 +1,8 @@
# $OpenBSD: Makefile.inc,v 1.7 2019/07/12 20:47:23 sthen Exp $
+# Undefined references to internal plugins with ld.bfd
+ONLY_FOR_ARCHS = ${LLD_ARCHS}
+
CATEGORIES = lang
MAINTAINER = Marc Espie
Index: vm/Makefile
===
RCS file: /cvs/ports/lang/squeak/vm/Makefile,v
retrieving revision 1.30
diff -u -p -u -p -r1.30 Makefile
--- vm/Makefile 11 Feb 2020 13:44:10 - 1.30
+++ vm/Makefile 13 Jun 2020 10:43:30 -
@@ -1,7 +1,5 @@
# $OpenBSD: Makefile,v 1.30 2020/02/11 13:44:10 kmos Exp $
-BROKEN-sparc64 = Many undefined references when linking
-
COMMENT = smalltalk system, vm
SRCV = 1
squeak_powerpc.diff
Description: Binary data
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/06/12 16:40:41 Modified files: editors/qscintilla: Makefile editors/qscintilla/patches: patch-Qt4Qt5_qscintilla_pro Added files: editors/qscintilla/patches: patch-src_RunStyles_cpp Log message: qscintilla: fix some consumers build on powerpc cad/openscad and databases/sqlitebrowser builds were failing with some undefined references to "Scintilla::RunStyles"; use ptrdiff_t variants on that arch. While here, regen a patch. OK rsadowski@
Re: powerpc bulk build report
On Fri, 12 Jun 2020 16:04:22 -0600 (MDT) c...@openbsd.org wrote: > Bulk build on macppc-0.ports.openbsd.org > > Started : Tue May 26 11:12:39 MDT 2020 > Finished: Fri Jun 12 16:04:08 MDT 2020 > Duration: 17 Days 4 hours 51 minutes ^ We swapped some monsters that failed at runtime for GNOME/MATE stuff :) > Built using OpenBSD 6.7-current (GENERIC) #722: Mon May 25 18:07:09 > MDT 2020 > > Built 9939 packages > http://build-failures.rhaalovely.net/powerpc/2020-05-26/cad/openscad.log > http://build-failures.rhaalovely.net/powerpc/2020-05-26/databases/sqlitebrowser.log I'll commit the fix just after that mail > http://build-failures.rhaalovely.net/powerpc/2020-05-26/devel/geany.log https://marc.info/?l=openbsd-ports=159188140009019=2 > http://build-failures.rhaalovely.net/powerpc/2020-05-26/mail/kopano/core.log Fixed, won't build on ld.bfd archs again > http://build-failures.rhaalovely.net/powerpc/2020-05-26/security/aircrack-ng.log https://marc.info/?l=openbsd-ports=159141021925207=2 > http://build-failures.rhaalovely.net/powerpc/2020-05-26/x11/gnome/builder.log Fixed, but it requires also that uncommitted fix for libdazzle: https://marc.info/?l=openbsd-ports=159199473212267=2 > http://build-failures.rhaalovely.net/powerpc/2020-05-26/x11/gnome/eog.log Rust only, it wasn't and won't be available, fixed. > http://build-failures.rhaalovely.net/powerpc/2020-05-26/x11/qt5/qtwebengine.log Fixed, added NOT_FOR_ARCHS
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/06/12 16:09:09 Modified files: x11/gnome/builder: Makefile Log message: gnome-builder: fix the build with ld.bfd Don't use "-Wl,--as-needed" (the meson default) with ld.bfd, it causes undefined references errors. Allow building with ld.bfd on i386 where it's broken with ld.lld, we'll see the result in the next bulk. Tested on powerpc -- this arch needs a yet-to-be-committed x11/gnome/libdazzle fix to build it there. OK sthen@, jca@ and aja@ (maintainer)
[macppc] x11/gnome/libdazzle: don't use unsupported 64-bit __sync_* primitives
Hi,
I've proposed a fix for x11/gnome/gnome-builder for ld.bfd archs and
possibly i386 (offlist), but on macppc it still failed due to undefined
references to 64-bit __sync_* primitives.
libdazzle uses 64-bit __sync_* primitives that are not supported on
macppc. There is no __atomic_* fallback, but there is a non-atomic
one, that was unused. I've proposed to upstream to use it on
!linux ppc32 generally [0], and it has been merged.
With that, gnome-builder will build and run fine on macppc.
Comments/feedback are welcome,
Charlène.
[0] https://gitlab.gnome.org/GNOME/libdazzle/-/merge_requests/44
Index: Makefile
===
RCS file: /cvs/ports/x11/gnome/libdazzle/Makefile,v
retrieving revision 1.17
diff -u -p -u -p -r1.17 Makefile
--- Makefile14 May 2020 14:59:03 - 1.17
+++ Makefile12 Jun 2020 16:39:14 -
@@ -4,6 +4,7 @@ COMMENT=companion library to GObject a
GNOME_PROJECT= libdazzle
GNOME_VERSION= 3.36.0
+REVISION= 0
SHARED_LIBS += dazzle-1.01.2 # 0
Index: patches/patch-src_util_dzl-counter_c
===
RCS file: patches/patch-src_util_dzl-counter_c
diff -N patches/patch-src_util_dzl-counter_c
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-src_util_dzl-counter_c12 Jun 2020 16:39:14 -
@@ -0,0 +1,20 @@
+$OpenBSD$
+
+Don't use unsupported 64-bit __sync primitives on powerpc, from:
+https://gitlab.gnome.org/GNOME/libdazzle/-/merge_requests/44
+
+Index: src/util/dzl-counter.c
+--- src/util/dzl-counter.c.orig
src/util/dzl-counter.c
+@@ -52,7 +52,11 @@ G_DEFINE_BOXED_TYPE (DzlCounterArena, dzl_counter_aren
+ #define CELLS_PER_GROUP(ncpu) \
+ (((sizeof (CounterInfo) * COUNTERS_PER_GROUP) + \
+ (sizeof(DzlCounterValue) * (ncpu))) / DATA_CELL_SIZE)
++#ifdef DZL_COUNTER_REQUIRES_ATOMIC
+ #define DZL_MEMORY_BARRIER __sync_synchronize()
++#else
++#define DZL_MEMORY_BARRIER 0
++#endif
+
+ typedef struct
+ {
Index: patches/patch-src_util_dzl-counter_h
===
RCS file: patches/patch-src_util_dzl-counter_h
diff -N patches/patch-src_util_dzl-counter_h
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-src_util_dzl-counter_h12 Jun 2020 16:39:14 -
@@ -0,0 +1,18 @@
+$OpenBSD$
+
+Don't use unsupported 64-bit __sync primitives on powerpc, from:
+https://gitlab.gnome.org/GNOME/libdazzle/-/merge_requests/44
+
+Index: src/util/dzl-counter.h
+--- src/util/dzl-counter.h.orig
src/util/dzl-counter.h
+@@ -163,6 +163,9 @@ G_BEGIN_DECLS
+ # define dzl_get_current_cpu() dzl_get_current_cpu_rdtscp()
+ #elif defined(__linux__)
+ # define dzl_get_current_cpu() dzl_get_current_cpu_call()
++#elif defined(__powerpc__) && !defined(__powerpc64__)
++# define dzl_get_current_cpu() 0
++# undef DZL_COUNTER_REQUIRES_ATOMIC
+ #else
+ # define dzl_get_current_cpu() 0
+ # define DZL_COUNTER_REQUIRES_ATOMIC 1
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/06/12 09:54:49 Modified files: mail/kopano/core: Makefile Log message: kopano: don't build on ld.bfd archs 'USE_LLD=Yes' does not prevent building the port on ld.bfd archs, where it fails to build. Using ld.bfd causes undefined references errors. Use ONLY_FOR_ARCHS, and remove BROKEN-sparc64 accordingly. "sure" robert@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/06/12 04:27:52 Modified files: x11/gnome/eog : Makefile Log message: eog: don't build on !rust archs It requires librsvg>=2.44.0 that is rust-only. OK semarie@, aja@ (maintainer)
[macppc] editors/qscintilla: unbreak some consumers
Hi,
> http://build-failures.rhaalovely.net/powerpc/last/databases/sqlitebrowser.log
> http://build-failures.rhaalovely.net/powerpc/last/cad/openscad.log
See my geany mail [0] since the issue is exactly the same.
The below diff fixes the build of openscad and sqlitebrowser on macppc,
the runtime is fine as well. openscad suffers from what i suppose is a
QtSVG vs macppc OpenGL direct rendering issue (widgets are purple), but
unlike Krita that had it as well, the software is totally usable.
Comments/feedback are welcome,
Charlène.
[0] https://marc.info/?l=openbsd-ports=159188140009019=2
Index: Makefile
===
RCS file: /cvs/ports/editors/qscintilla/Makefile,v
retrieving revision 1.27
diff -u -p -u -p -r1.27 Makefile
--- Makefile29 Feb 2020 08:19:59 - 1.27
+++ Makefile11 Jun 2020 18:22:17 -
@@ -6,7 +6,7 @@ COMMENT = Qt port of the Scintilla sourc
VERSION = 2.11.4
DISTNAME = QScintilla-${VERSION}
PKGNAME = ${DISTNAME:L}
-REVISION = 0
+REVISION = 1
CATEGORIES = editors x11
Index: patches/patch-Qt4Qt5_qscintilla_pro
===
RCS
file: /cvs/ports/editors/qscintilla/patches/patch-Qt4Qt5_qscintilla_pro,v
retrieving revision 1.5 diff -u -p -u -p -r1.5
patch-Qt4Qt5_qscintilla_pro
--- patches/patch-Qt4Qt5_qscintilla_pro 24 Jul 2019 08:49:43
- 1.5 +++ patches/patch-Qt4Qt5_qscintilla_pro 11 Jun
2020 18:22:17 - @@ -2,7 +2,7 @@ $OpenBSD:
patch-Qt4Qt5_qscintilla_pro,v Index: Qt4Qt5/qscintilla.pro
--- Qt4Qt5/qscintilla.pro.orig
+++ Qt4Qt5/qscintilla.pro
-@@ -68,7 +68,7 @@ unix:!macx {
+@@ -70,7 +70,7 @@ unix:!macx {
# Scintilla namespace rather than pollute the global namespace.
#DEFINES += SCI_NAMESPACE
Index: patches/patch-src_RunStyles_cpp
===
RCS file: patches/patch-src_RunStyles_cpp
diff -N patches/patch-src_RunStyles_cpp
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-src_RunStyles_cpp 11 Jun 2020 18:22:17 -
@@ -0,0 +1,18 @@
+$OpenBSD$
+
+powerpc consumers fix for
+"undefined reference to Scintilla::RunStyles::RunStyles()"
+
+Index: src/RunStyles.cpp
+--- src/RunStyles.cpp.orig
src/RunStyles.cpp
+@@ -308,7 +308,8 @@ void RunStyles::Check() const {
+
+ template class Scintilla::RunStyles;
+ template class Scintilla::RunStyles;
+-#if (PTRDIFF_MAX != INT_MAX) || PLAT_HAIKU
++#if (PTRDIFF_MAX != INT_MAX) || PLAT_HAIKU || \
++( defined(__OpenBSD__) && defined(__powerpc__) )
+ template class Scintilla::RunStyles;
+ template class Scintilla::RunStyles;
+ #endif
[macppc] Unbreak devel/geany
Hi,
> http://build-failures.rhaalovely.net/powerpc/2020-05-09/devel/geany.log
It's a wild guess, but i think it wants a template with a long argument
because on macppc:
$ c++ -dM -E - < /dev/null | grep PTRDIFF_MAX
#define __PTRDIFF_MAX__ 2147483647L
$ c++ -dM -E - < /dev/null | grep INT_MAX
#define __INT_MAX__ 2147483647
Either way, the below diff fixes the build on macppc, the runtime is
fine [0] as well. While here i've moved HOMEPAGE to https.
I should be able to fix the broken ports depending on other scintilla
flavors pretty soon :]
Comments/feedback are welcome,
Charlène.
[0] https://bsd.network/@julianaito/104325481078218916
Index: Makefile
===
RCS file: /cvs/ports/devel/geany/Makefile,v
retrieving revision 1.63
diff -u -p -u -p -r1.63 Makefile
--- Makefile16 Oct 2019 06:51:42 - 1.63
+++ Makefile11 Jun 2020 13:13:47 -
@@ -3,11 +3,12 @@
COMMENT= small and lightweight IDE
DISTNAME = geany-1.36
+REVISION = 0
SHARED_LIBS += geany 0.0 # 0.0
CATEGORIES=devel
-HOMEPAGE= http://www.geany.org/
+HOMEPAGE= https://www.geany.org/
MAINTAINER=Victor Kukshiev
Index: patches/patch-scintilla_src_RunStyles_cxx
===
RCS file: patches/patch-scintilla_src_RunStyles_cxx
diff -N patches/patch-scintilla_src_RunStyles_cxx
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-scintilla_src_RunStyles_cxx 11 Jun 2020 13:13:47 -
@@ -0,0 +1,18 @@
+$OpenBSD$
+
+powerpc fix for
+"undefined reference to Scintilla::RunStyles::RunStyles()"
+
+Index: scintilla/src/RunStyles.cxx
+--- scintilla/src/RunStyles.cxx.orig
scintilla/src/RunStyles.cxx
+@@ -308,7 +308,8 @@ void RunStyles::Check() const {
+
+ template class Scintilla::RunStyles;
+ template class Scintilla::RunStyles;
+-#if (PTRDIFF_MAX != INT_MAX) || PLAT_HAIKU
++#if (PTRDIFF_MAX != INT_MAX) || PLAT_HAIKU || \
++( defined(__OpenBSD__) && defined(__powerpc__) )
+ template class Scintilla::RunStyles;
+ template class Scintilla::RunStyles;
+ #endif
x11/gnome/eog can't be built on !rust archs
Hi,
I've spotted this in the current macppc bulk while building eog:
> eog-3.36.2/meson.build:198:2: ERROR: Invalid version of dependency,
> need 'librsvg-2.0' ['>= 2.44.0'] found '2.40.21'.
Since newer versions require rust, i'm proposing to not trying to build
eog on !rust archs.
Comments/feedback are welcome,
Charlène.
Index: Makefile
===
RCS file: /cvs/ports/x11/gnome/eog/Makefile,v
retrieving revision 1.156
diff -u -p -u -p -r1.156 Makefile
--- Makefile14 May 2020 15:11:18 - 1.156
+++ Makefile11 Jun 2020 11:32:28 -
@@ -1,5 +1,8 @@
# $OpenBSD: Makefile,v 1.156 2020/05/14 15:11:18 jasper Exp $
+# Need librsvg>=2.44.0 that is rust only
+ONLY_FOR_ARCHS=${RUST_ARCHS}
+
COMMENT= Eye of GNOME, image viewing and cataloging program
GNOME_PROJECT= eog
[update] devel/p5-Devel-Hide 0.0013
Hi, Here is an update for Devel::Hide. The changelog can be found there [0]. Port-wise it's a simple update, i've just changed my maintainer email address while here. I've tested the direct depends [1] and they still pass. I've run tested databases/ports-readme-dancer successfully so openports.pl won't blow up :) Comments/feedback are welcome, Charlène. [0] https://metacpan.org/changes/distribution/Devel-Hide [1] https://bin.charlenew.xyz/p5-Devel-Hide.tgz Index: Makefile === RCS file: /cvs/ports/devel/p5-Devel-Hide/Makefile,v retrieving revision 1.2 diff -u -p -u -p -r1.2 Makefile --- Makefile12 Jul 2019 20:44:51 - 1.2 +++ Makefile10 Jun 2020 10:36:36 - @@ -2,10 +2,10 @@ COMMENT = forces the unavailability of specified Perl modules -DISTNAME = Devel-Hide-0.0010 +DISTNAME = Devel-Hide-0.0013 CATEGORIES = devel -MAINTAINER = Charlene Wendling +MAINTAINER = Charlene Wendling MODULES = cpan PKG_ARCH = * Index: distinfo === RCS file: /cvs/ports/devel/p5-Devel-Hide/distinfo,v retrieving revision 1.1.1.1 diff -u -p -u -p -r1.1.1.1 distinfo --- distinfo5 Jan 2019 10:50:38 - 1.1.1.1 +++ distinfo10 Jun 2020 10:36:36 - @@ -1,2 +1,2 @@ -SHA256 (Devel-Hide-0.0010.tar.gz) = dW/vpg/r8djbyFFCa/sIRSDbRfdF01yOhiOUXW3aXoI= -SIZE (Devel-Hide-0.0010.tar.gz) = 7225 +SHA256 (Devel-Hide-0.0013.tar.gz) = b8O7z08CU6blFoKWyQmTjBPIhpI6eZWslp5Hrfzwfss= +SIZE (Devel-Hide-0.0013.tar.gz) = 9787
Re: [Update] www/p5-Mojo : Update to 8.52
Hi, On Wed, 10 Jun 2020 01:25:27 + wen heping wrote: > Hi, > > Here is a patch for www/p5-Mojo to update to 8.52. > It build well and pass all tests but one fail. One pod_coverage > tests fail, it would pass after we update p5-EV. It is a small > problem. > > 11 ports depends on p5-Mojo, all build and pass test with this > patch. It works fine here, OK cwen@ > Regards, > wen
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2020/06/09 11:09:24 Modified files: cad/kicad : Makefile cad/kicad-share: Makefile.inc Log message: kicad-share/*: add powerpc to NOT_FOR_ARCHS Kicad being broken on powerpc, it's useless to build these ports there. Add a reminder in cad/kicad/Makefile to sync cad/kicad-share/Makefile.inc with archs where kicad is not built. input and OK sthen@, OK kirby@
[macppc] Don't build cad/kicad-share
Hi, Since kicad is marked BROKEN on powerpc, it's useless to build kicad-share ports. OK ? Charlène. Index: Makefile.inc === RCS file: /cvs/ports/cad/kicad-share/Makefile.inc,v retrieving revision 1.4 diff -u -p -u -p -r1.4 Makefile.inc --- Makefile.inc10 Dec 2019 23:09:48 - 1.4 +++ Makefile.inc9 Jun 2020 07:37:50 - @@ -1,5 +1,8 @@ # $OpenBSD: Makefile.inc,v 1.4 2019/12/10 23:09:48 sthen Exp $ +# XXX Sync with archs where kicad is broken +NOT_FOR_ARCHS =powerpc + PKG_ARCH ?=* V ?= 5.1.5
