Re: Upgrading Perl... Somebody just shoot me and put me out of my misery!
On 2013-11-25 08:04, Matthias Andree wrote:
Am 25.11.2013 07:53, schrieb olli hauer:
On 2013-11-25 07:40, Matthias Andree wrote:
Am 23.11.2013 12:20, schrieb Mark Martinec:
On Friday 22 November 2013 21:40:07 Ronald F. Guilmette wrote:
Now, one last little thing...
The note in the UPDATING file dated 20131120 gives essentially the same
instructions as the one dated 20131023, *however* it also contains this:
1) Change the option in lang/perl5.16:
make -C /usr/ports/lang/perl5.16 config
HUH?? I don't understand this at all. What exactly is the option that
we are changing here? And what does it matter to anything?
It would be Nice if this were entierly less opaque.
$ man ports
[...]
config Configure OPTIONS for this port using dialog4ports(1).
And what does it matter to anything?
Gives you a choice to re-think your existing/chosen port options.
For example, a new default is now THREADS, but you may not like
it, as it somewhat increases the memory usage and requires
to rebuild all perl modules.
Which shows an interesting facet of this whole tedious process:
We're doing a lousy job of explaining the options to unsavvy users, and
we're also doing a lousy job of tracking options. Perhaps we should
just slash down the options and go more for build the default - it
also reduces testing complexity and would give for a more uniform ports
experience for everyone (packages use default options anyways).
${opt}_DESC is limited, but help/explanation can be given in pkg-help.
Granted, but last time I checked I did not have a Help button on
dialog4ports. Either none of the ports I've seen offering options offer
pkg-help, or dialog4ports needs to be told to feed pkg-help through $PAGER.
Only if the file pkg-help exists a hint will be displayed on the bottom if
you call `make config'.
make config -C www/apache24
Detailed help is available hit F1 or ^E to view it
F1 will terminate the config dialog if called by a putty ssh session but ^E
has always worked for me.
--
olli
___
freebsd-ports@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: [QAT] r334788: 1x depend (depend_package in databases/rrdtool), 3x leftovers
Hi All, I'm not sure how to deal with /home/$USER leftovers (i.e. while package install a new user/group is created and while package deinstalled /home/$USER becomes a leftover, but we display a message that a user should be removed manually): [...] phase 7: make package === Building package for slurm-2.6.4 Deleting slurm-2.6.4 Deinstallation has been requested for the following 1 packages: slurm-2.6.4 The deinstallation will free 51 MB [1/1] Deleting slurm-2.6.4...== You should manually remove the slurm user. done === Checking filesystem state list of extra files and directories in / (not present before this port was installed but present after it was deinstalled) 305288 drwxr-xr-x3 root wheel 64 Nov 24 20:32 home 346190 drwxr-xr-x2 slurmslurm 0 Nov 24 20:32 home/slurm [...] Thanks! -- WBR, Boris Samorodov (bsam) FreeBSD Committer, http://www.FreeBSD.org The Power To Serve ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
[QAT] r334818: 2x leftovers, 2x success
- Update to new upstream release 2.1.17, resetting PORTREVISION. - Remove patch-Makefile.in, which has been integrated by upstream. - Announcement: https://mail.python.org/pipermail/mailman-announce/2013-November/000186.html - Poudriere 3.0.13 ticket filed about misreported orphans in testport: https://fossil.etoilebsd.net/poudriere/tktview/e8d957a27f8ce8b6255ed655d031e6d05b02492c - Build ID: 20131125075801-29233 Job owner: mand...@freebsd.org Buildtime: 63 minutes Enddate: Mon, 25 Nov 2013 09:00:31 GMT Revision: r334818 Repository: https://svnweb.freebsd.org/ports?view=revisionrevision=334818 - Port:mail/mailman 2.1.17 Buildgroup: 8.4-QAT/amd64 Buildstatus: SUCCESS Log: https://qat.redports.org//~mand...@freebsd.org/20131125075801-29233-231392/mailman-2.1.17.log Buildgroup: 8.4-QAT/i386 Buildstatus: LEFTOVERS Log: https://qat.redports.org//~mand...@freebsd.org/20131125075801-29233-231393/mailman-2.1.17.log Buildgroup: 9.2-QAT/amd64 Buildstatus: LEFTOVERS Log: https://qat.redports.org//~mand...@freebsd.org/20131125075801-29233-231394/mailman-2.1.17.log Buildgroup: 9.2-QAT/i386 Buildstatus: SUCCESS Log: https://qat.redports.org//~mand...@freebsd.org/20131125075801-29233-231395/mailman-2.1.17.log -- Buildarchive URL: https://qat.redports.org/buildarchive/20131125075801-29233 redports https://qat.redports.org/ ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
[pkgng] wildcard architecture: howto?
Hi All, while building a port I get: - pkg-static: DEVELOPER_MODE: Notice: arch freebsd:11:x86:64 -- no architecture specific files found: could this package use a wildcard architecture? - What should I add to Makefile to achieve this? Thanks! -- WBR, Boris Samorodov (bsam) FreeBSD Committer, http://www.FreeBSD.org The Power To Serve ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: portmaster refuses to use pkgng with local packages
Am 24.11.2013 12:43, schrieb Axel Rau: While trying ports-mgmt/poudriere in my ezjail/portmaster environment, I learned: poudriere can't run at secure level 1, because it loads linux.ko and uses chflags. Regarding moving to pkgng, what are the replacements to portaudit / jailaudit? # pkg audit -F Regards, STefan ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: poudriere behave-alike for
On Mon, 25 Nov 2013 08:40:54 +0100 John Marino dragonfly...@marino.st wrote: On 11/25/2013 02:15, Christopher J. Ruwe wrote: I think my question is slightly off-topic, but I think freebsd-ports@ may be the best of many not so good fits: I need to build packages for Solaris and SmartOS. My first choice would be ports, which unfortunately are not very well suited to cross-building. Instead I use, as many people, pkgsrc. I would like to leverage pkgsrc with something like poudriere, especially as I have ZFS and zones in Solaris/SmartOS. I found in a message on the DragonFlyBSD list http://leaf.dragonflybsd.org/mailarchive/users/2013-01/msg8.html a mention of poudriere being used on DragonFly/pkgsrc. I was involved in that referenced email. The first point to make is that currently ports is *not* an option for solaris or SmartOS, regardless of its ability to cross compile. Point #2 is that I want to try to bring ports to the solaris-alike family in the future (aka sunports), but work on this hasn't started yet, and adapting solaris will be a lot more work than adapting DragonFly was (and believe me DF was *A LOT* of work. Point #3 is that if I were still heavily involved in pkgsrc, I would probably create a branch of poudriere that supported pkgsrc. It is something I would recommend highly to the pkgsrc community. However, it suffers greatly from Not invented Here syndrome, so most consider (without proper evaluation) that pkgsrc tools are more or less equivalent. The fact is that they are not. Does anybody know of the state of this piece of software? The git repos I can find on google are stale links. As etoilebsd is referenced in the mail from DragonFly, I chose to ask here first. There is no poudriere-for-pkgsrc. The current poudriere branches are here: https://fossil.etoilebsd.net/poudriere/brlist For pkgsrc your choices are: http://pkgsrc.se/pkgtools/distbb http://pkgsrc.se/pkgtools/pbulk Here's a recent post about setting up pbulk: http://mail-index.netbsd.org/pkgsrc-users/2013/11/09/msg018881.html In general its poorly documented and difficult to set up parallel building. The script above is yet another attempt to reduce the complexity but I don't think either pbulk or distcc have nearly the polish or features that poudriere has. But take that with a grain of salt because I haven't used either in a long time. One more thing: SmartOS not only uses pkgsrc officially, they have a full builder farm that makes a full set of packages quarterly packages available. It also works on other illumos platforms. The best approach is just use their work. Another tutorial how to set up bulk build: http://www.perkin.org.uk/posts/distributed-chrooted-pkgsrc-bulk-builds.html info about packages already built: http://www.perkin.org.uk/posts/whats-new-in-pkgsrc-2013Q2.html You might want to check out the reset of www.perkin.org.uk for interesting posts. John ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org Thanks to both and Shane for your kind answers. I know of Jonathan Perkin's work and I also know of an approach roughly resembling poudriere ideas named 'pk' (github.com/mamash/pk), which however seems do have been abandoned by Joyent. While it would be well be possible to use Joyent's binaries, I am a huge fan of the ability of ports-like systems to be tuned. I am also a huge fan of the poudriere approach of building everything in clean jails, especially after being nastily bitten by implicit dependencies (i.e., such deps as introduced by autoconf runs instead of being declared in the port). I had the hopes that issue had come up sufficiently often that somebody already did a poudriere port or build a functional clone or however to call that. I do not really know whether I can port poudriere capability- and capacity-wise. However, many thanks, your answers have helped me a lot in getting a better picture. Cheers, -- Christopher TZ: GMT + 1h GnuPG/GPG: 0xE8DE2C14 FreeBSD 9.2-STABLE #1 r256184: Thu Oct 10 19:12:54 CEST 2013 c...@dijkstra.cruwe.de:/usr/obj/usr/home/cjr/media/src/freebsd/base/stable/9/sys/GEN_WDTRACE Punctuation matters: Lets eat Grandma. or Lets eat, Grandma. - Punctuation saves lives. A panda eats shoots and leaves. or A panda eats, shoots, and leaves. - Punctuation teaches proper biology. With sufficient thrust, pigs fly just fine. However, this is not necessarily a good idea. It is hard to be sure where they are going to land, and it could be dangerous sitting under them as they fly overhead. (RFC 1925) ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to
Re: [pkgng] wildcard architecture: howto?
On 11/25/13 09:32, Boris Samorodov wrote: Hi All, while building a port I get: - pkg-static: DEVELOPER_MODE: Notice: arch freebsd:11:x86:64 -- no architecture specific files found: could this package use a wildcard architecture? - What should I add to Makefile to achieve this? Thanks! Alas, there is no support for this in bsd.ports.mk et al yet. pkg(8) supports this functionality just fine, so you can make arch independent packages by writing your own +MANIFEST and running 'pkg create' directly -- eg. like in des' blog: http://blog.des.no/2013/07/ This is pkg(8) being prepared for future enhancements to the pkgrepos: not storing multiple essentially identical copies of packages where we don't need to. But that's a work in progress still. Cheers, Matthew signature.asc Description: OpenPGP digital signature
Re: poudriere behave-alike for
On 11/25/2013 10:47, Christopher J. Ruwe wrote: While it would be well be possible to use Joyent's binaries, I am a huge fan of the ability of ports-like systems to be tuned. If by tuned you mean I want to have a non-default set of options, I think pbulk and distbb support that. I am also a huge fan of the poudriere approach of building everything in clean jails, especially after being nastily bitten by implicit dependencies (i.e., such deps as introduced by autoconf runs instead of being declared in the port). pbulk and distbb also employ the clean jail approach. I had the hopes that issue had come up sufficiently often that somebody already did a poudriere port or build a functional clone or however to call that. I do not really know whether I can port poudriere capability- and capacity-wise. It's not trivial, but poudriere is just a bunch of shell scripts (right now). But add on a new implemention of jail.sh per platform, plus changing pkgng out for pkgsrc/pkgin and it's even a bigger job than the dragonfly branch was. But no, you should assume there is no poudriere clone available for pkgsrc at the moment. However, many thanks, your answers have helped me a lot in getting a better picture. np, John ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
FreeBSD ports you maintain which are out of date
Dear port maintainer, The portscout new distfile checker has detected that one or more of your ports appears to be out of date. Please take the opportunity to check each of the ports listed below, and if possible and appropriate, submit/commit an update. If any ports have already been updated, you can safely ignore the entry. You will not be e-mailed again for any of the port/version combinations below. Full details can be found at the following URL: http://portscout.freebsd.org/po...@freebsd.org.html Port| Current version | New version +-+ devel/ocaml-extlib | 1.5.4 | 1.6.0 +-+ If any of the above results are invalid, please check the following page for details on how to improve portscout's detection and selection of distfiles on a per-port basis: http://portscout.freebsd.org/info/portscout-portconfig.txt If wish to stop receiving portscout reminders, please contact portsc...@freebsd.org Thanks. ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Current unassigned ports problem reports
(Note: an HTML version of this report is available at http://www.freebsd.org/cgi/query-pr-summary.cgi?category=ports .) The following is a listing of current problems submitted by FreeBSD users. These represent problem reports covering all versions including experimental development code and obsolete releases. S Tracker Resp. Description o ports/184273maintainer update: mail/dovecot2 f ports/184264audio/sox: /usr/local/man/man7/soxeffect.7.gz invalid f ports/184257graphics/openimageio fails to build o ports/184256[maintainer-update] sysutils/froxlor: Update to 0.9.30 o ports/184248finance/qhacc: Update to version 4.2.1 o ports/184247games/flying: Support CXX properly o ports/184246emulators/mame: Update to version 0.151 o ports/184245biology/blat: Update to version 35 o ports/184244games/jfk: Fix build with clang o ports/184243multimedia/oggvideotools: Fix build on -current o ports/184242biology/avida: Fix build with clang o ports/184241graphics/qslim: Fix build on -current o ports/184240math/tvmet: Fix build with clang o ports/184239devel/libcwd: Fix build on -current o ports/184238biology/lagan: Fix build o ports/184237japanese/skkfep: Update MASTER_SITES and WWW line o ports/184236graphics/rayshade: Fix build on -current o ports/184235net/pmf: Fix build on -current o ports/184234databases/gtksql: Fix build on -current o ports/184233graphics/rgbpaint: Support CC properly o ports/184232graphics/png2ico: Fix build o ports/184231games/tuxpuck: Support CC properly o ports/184230devel/ace: Update to version 6.2.3 o ports/184229devel/regexx: Fix build o ports/184228devel/rudiments: Update to version 0.44.1 o ports/184227games/xspacewarp: Fix build with clang o ports/184226print/xmbibtex: Fix build with clang o ports/184225editors/leo: Update to version 4.11 o ports/184224science/gromacs: Update to version 4.6.4 o ports/184223x11/roxterm: Fix build on -current o ports/184222sysutils/eiciel: Update to version 0.9.8.2 o ports/184221games/nighthawk: Support CC/CFLAGS properly o ports/184220games/freedoko: Fix build on -current o ports/184219biology/lamarc: Fix build on -current o ports/184211[NEW PORT] net-im/pybitmessage: A Bitmessage Client fo o ports/184210[FIX] multimedia/tstools requires MAKE_JOBS_UNSAFE=yes o ports/184208fix japanese/plain2 build failure o ports/184206fix audio/x11amp build failure o ports/184205building japanese/eijiro-fpw fails o ports/184203[NEW PORT] net/motsognir: an ANSI C gopher server f ports/184194security/bro fails to build (in the 'package' phase) u o ports/184183[PATCH] security/pam_yubico: update to 2.14 and fix ld f ports/184177[patch] audio/teamspeak3-server f ports/184161multimedia/gpac-mp4box missing dependency f ports/184158[patch] security/chntpw: install other useful binaries o ports/184157[patch] net/ssvnc: correct options and work with stagi o ports/184151audio/last.fm 1.5.4.26862 doesn't build on FreeBSD 10. f ports/184150Patch to Allow irc/unreal to Run When Compiled with Cl f ports/184143[patch] security/lynis incorrect permissions for data o ports/184134editors/slime is outdated. o ports/184132multimedia/dvd-slideshow: cleanups and fixes to make 0 f ports/184131Mk/Scripts/qa.sh: support spaces in filenames o ports/184125[maintainer-update] Fix textproc/dictem build o ports/184123[maintainer update] audio/teamspeak3-server to v3.0.10 o ports/184119[MAINTAINER] devel/gdb: Fix the build with gcc 4.8 f ports/184111[UPDATE] emulators/desmume to v0.9.9 o ports/184103maintainer update: mail/dovecot2 o ports/184102[MAINTAINER-UPDATE] databases/spatialite_gui 1.5.0 - o ports/184101[MAINTAINER-UPDATE] graphics/libgaiagraphics 0.4b - 0 o ports/184099[MAINTAINER-UPDATE] databases/spatialite 4.1.1_1 o ports/184093[patch] sysutils/javaservicewrapper Update to latest v o ports/184079net-mgmt/collectd5 plugin notify_email not copied o ports/184074fix port: lang/polyml o ports/184072fix build: audio/xmms-timidity o ports/184069update port: devel/bam o ports/184050[maintainer update] www/free-sa-devel: update to 2.0.0 o ports/184043maintainer update: Upgrade net-mgmt/bsnmptools to vers f
Re: [pkgng] wildcard architecture: howto?
25.11.2013 13:52, Matthew Seaman пишет: On 11/25/13 09:32, Boris Samorodov wrote: Hi All, while building a port I get: - pkg-static: DEVELOPER_MODE: Notice: arch freebsd:11:x86:64 -- no architecture specific files found: could this package use a wildcard architecture? - What should I add to Makefile to achieve this? Thanks! Alas, there is no support for this in bsd.ports.mk et al yet. OK. pkg(8) supports this functionality just fine, so you can make arch independent packages by writing your own +MANIFEST and running 'pkg create' directly -- eg. like in des' blog: http://blog.des.no/2013/07/ Thanks. This is pkg(8) being prepared for future enhancements to the pkgrepos: not storing multiple essentially identical copies of packages where we don't need to. But that's a work in progress still. Well, shouln't then this warning be silenced up to date when this function is realised? Just a note... -- WBR, Boris Samorodov (bsam) FreeBSD Committer, http://www.FreeBSD.org The Power To Serve ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
audio/clementine-player build failure
Any news on audio/clementine-player? It does not build on 10-BETA3, with either clang or gcc. PR: http://www.freebsd.org/cgi/query-pr.cgi?pr=182107 -- Marko Cupać ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
[QAT] r334841: 1x leftovers, 3x success
- Update to 3.3.1.2 Approved by:pawel / wg (mentors, implicit) - Build ID: 20131125113800-57189 Job owner: nemy...@freebsd.org Buildtime: 39 minutes Enddate: Mon, 25 Nov 2013 12:16:37 GMT Revision: r334841 Repository: https://svnweb.freebsd.org/ports?view=revisionrevision=334841 - Port:games/rocksndiamonds 3.3.1.2 Buildgroup: 8.4-QAT/amd64 Buildstatus: LEFTOVERS Log: https://qat.redports.org//~nemy...@freebsd.org/20131125113800-57189-231484/rocksndiamonds-3.3.1.2.log Buildgroup: 8.4-QAT/i386 Buildstatus: SUCCESS Log: https://qat.redports.org//~nemy...@freebsd.org/20131125113800-57189-231485/rocksndiamonds-3.3.1.2.log Buildgroup: 9.2-QAT/amd64 Buildstatus: SUCCESS Log: https://qat.redports.org//~nemy...@freebsd.org/20131125113800-57189-231486/rocksndiamonds-3.3.1.2.log Buildgroup: 9.2-QAT/i386 Buildstatus: SUCCESS Log: https://qat.redports.org//~nemy...@freebsd.org/20131125113800-57189-231487/rocksndiamonds-3.3.1.2.log -- Buildarchive URL: https://qat.redports.org/buildarchive/20131125113800-57189 redports https://qat.redports.org/ ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: [QAT] r334788: 1x depend (depend_package in databases/rrdtool), 3x leftovers
The files in /home/$USER (and /var) should be preserved. Deinstalls may be part of an upgrade process, vs permanent removal, and some of these files contain usage statistics needed by sysadmins for reporting. I would suggest the following deinstall message: You may remove /home/$USER if you are no longer using SLURM. Thanks, JB On 11/25/13 2:56 AM, Boris Samorodov wrote: Hi All, I'm not sure how to deal with /home/$USER leftovers (i.e. while package install a new user/group is created and while package deinstalled /home/$USER becomes a leftover, but we display a message that a user should be removed manually): [...] phase 7: make package === Building package for slurm-2.6.4 Deleting slurm-2.6.4 Deinstallation has been requested for the following 1 packages: slurm-2.6.4 The deinstallation will free 51 MB [1/1] Deleting slurm-2.6.4...== You should manually remove the slurm user. done === Checking filesystem state list of extra files and directories in / (not present before this port was installed but present after it was deinstalled) 305288 drwxr-xr-x3 root wheel 64 Nov 24 20:32 home 346190 drwxr-xr-x2 slurmslurm 0 Nov 24 20:32 home/slurm [...] Thanks! -- ~~~ Jason W. Bacon jwba...@tds.net Circumstances don't make a man: They reveal him. ~~~ ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
[QAT] r334852: 4x leftovers
Fix build. Reported by:Leo Vandewoestijne - Build ID: 20131125142201-10072 Job owner: m...@freebsd.org Buildtime: 5 minutes Enddate: Mon, 25 Nov 2013 14:27:25 GMT Revision: r334852 Repository: https://svnweb.freebsd.org/ports?view=revisionrevision=334852 - Port:dns/yadifa 1.0.3_1 Buildgroup: 8.4-QAT/amd64 Buildstatus: LEFTOVERS Log: https://qat.redports.org//~m...@freebsd.org/20131125142201-10072-231524/yadifa-1.0.3_1.log Buildgroup: 8.4-QAT/i386 Buildstatus: LEFTOVERS Log: https://qat.redports.org//~m...@freebsd.org/20131125142201-10072-231525/yadifa-1.0.3_1.log Buildgroup: 9.2-QAT/amd64 Buildstatus: LEFTOVERS Log: https://qat.redports.org//~m...@freebsd.org/20131125142201-10072-231526/yadifa-1.0.3_1.log Buildgroup: 9.2-QAT/i386 Buildstatus: LEFTOVERS Log: https://qat.redports.org//~m...@freebsd.org/20131125142201-10072-231527/yadifa-1.0.3_1.log -- Buildarchive URL: https://qat.redports.org/buildarchive/20131125142201-10072 redports https://qat.redports.org/ ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Getting to know your portmgr@ -- Martin Wilke
In our ongoing series on getting to know your portmgr@, we talk to Martin Wilke, the high quality automated PR and commit bot. http://blogs.freebsdish.org/portmgr/2013/11/25/getting-to-know-your-portmgr-martin-wilke/ ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: Porting a software which uses INP_GPIO?
Alexander Leidinger alexan...@leidinger.net writes:
On Thu, 21 Nov 2013 16:21:20 -0500
Lowell Gilbert freebsd-ports-lo...@be-well.ilk.org wrote:
Alexander Leidinger alexan...@leidinger.net writes:
I try to compile a software on FreeBSD which wants to use INP_GPIO,
OUTP_GPIO and some oder *GPIO* things.
A quick googling shows me some raspberry pi sites. Is this something
linux-specific (so that I can forget this software on FreeBSD as
long as we don't gain something similar)?
Searching for gpio in names of ports didn't show a hit and in the
basesystem includes I can't find it either.
GPIO is a way to do pin assignments for a chip package at run-time. I
use it on embedded platforms all the time, but it isn't normally
available on a PC. There's a gpioctl(1) that should be able to set the
a pin for input or output, as those flags indicate, or
programmatically I guess it would be GPIO_PIN_INPUT or
GPIO_PIN_OUTPUT in /usr/include/sys/gpio.h but again, you need to
have the hardware for it.
I have the hardware. Currently it is accessed from an old Laptop with
the Windows-binary of the program. I would like to replace the Laptop
and use a FreeBSD version of the program.
The code in question is:
---snip---
const int banks[4]={18,23,24,25};
[...]
for(i=0;i4;i++)
{
INP_GPIO(banks[i]);
OUT_GPIO(banks[i]);
if(i==bank)
{
GPIO_SET = 1 banks[i]; // enable bank
}
else
{
GPIO_CLR = 1 banks[i];// disable bank
}
}
---snip---
When looking at sys/gpio.h, I have no idea how I shall translate the
above into something FreeBSD understands. Could you please explain how
the above translates into FreeBSD-gpio-speak?
Not offhand, no; that would take some work, and would require some
knowledge of the hardware devices installed, and preferably their wiring
configurations as well. It would be a reasonable guess that INP_GPIO is
equivalent to a GPIOSETCONFIG with a state structure that (at least)
includes GPIO_PIN_INPUT in its flags.
But the code seems to be operating on whole banks at once, no just
individual pins.
___
freebsd-ports@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
[QAT] r334858: 4x leftovers
Add PKGNAMESUFFIX=-hpc to have a non-conflict package name with net/slurm. Submitted by: Jason Bacon jwba...@tds.net (maintainer, via e-mail) - Build ID: 20131125155601-60469 Job owner: b...@freebsd.org Buildtime: 16 minutes Enddate: Mon, 25 Nov 2013 16:11:49 GMT Revision: r334858 Repository: https://svnweb.freebsd.org/ports?view=revisionrevision=334858 - Port:sysutils/slurm-hpc 2.6.4 Buildgroup: 8.4-QAT/amd64 Buildstatus: LEFTOVERS Log: https://qat.redports.org//~b...@freebsd.org/20131125155601-60469-231548/slurm-hpc-2.6.4.log Buildgroup: 8.4-QAT/i386 Buildstatus: LEFTOVERS Log: https://qat.redports.org//~b...@freebsd.org/20131125155601-60469-231549/slurm-hpc-2.6.4.log Buildgroup: 9.2-QAT/amd64 Buildstatus: LEFTOVERS Log: https://qat.redports.org//~b...@freebsd.org/20131125155601-60469-231550/slurm-hpc-2.6.4.log Buildgroup: 9.2-QAT/i386 Buildstatus: LEFTOVERS Log: https://qat.redports.org//~b...@freebsd.org/20131125155601-60469-231551/slurm-hpc-2.6.4.log -- Buildarchive URL: https://qat.redports.org/buildarchive/20131125155601-60469 redports https://qat.redports.org/ ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
[QAT] r334864: 2x leftovers, 2x depend (configure_error in misc/help2man)
Don't clobber config files. Also adds stage support. PR: ports/171669 - Build ID: 20131125164800-28566 Job owner: ad...@freebsd.org Buildtime: 18 minutes Enddate: Mon, 25 Nov 2013 17:05:45 GMT Revision: r334864 Repository: https://svnweb.freebsd.org/ports?view=revisionrevision=334864 - Port:www/hastymail2 1.1_3,1 Buildgroup: 8.4-QAT/amd64 Buildstatus: LEFTOVERS Log: https://qat.redports.org//~ad...@freebsd.org/20131125164800-28566-231568/hastymail2-1.1_3,1.log Buildgroup: 8.4-QAT/i386 Buildstatus: LEFTOVERS Log: https://qat.redports.org//~ad...@freebsd.org/20131125164800-28566-231569/hastymail2-1.1_3,1.log Buildgroup: 9.2-QAT/amd64 Buildstatus: DEPEND (CONFIGURE_ERROR IN MISC/HELP2MAN) Log: https://qat.redports.org//~ad...@freebsd.org/20131125164800-28566-231570/help2man-1.43.3_1.log Buildgroup: 9.2-QAT/i386 Buildstatus: DEPEND (CONFIGURE_ERROR IN MISC/HELP2MAN) Log: https://qat.redports.org//~ad...@freebsd.org/20131125164800-28566-231571/help2man-1.43.3_1.log -- Buildarchive URL: https://qat.redports.org/buildarchive/20131125164800-28566 redports https://qat.redports.org/ ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: [QAT] r334788: 1x depend (depend_package in databases/rrdtool), 3x leftovers
(reformatted for consistency sake) 25.11.2013 18:26, Jason bacon пишет: On 11/25/13 2:56 AM, Boris Samorodov wrote: I'm not sure how to deal with /home/$USER leftovers (i.e. while package install a new user/group is created and while package deinstalled /home/$USER becomes a leftover, but we display a message that a user should be removed manually): [...] phase 7: make package === Building package for slurm-2.6.4 Deleting slurm-2.6.4 Deinstallation has been requested for the following 1 packages: slurm-2.6.4 The deinstallation will free 51 MB [1/1] Deleting slurm-2.6.4...== You should manually remove the slurm user. done === Checking filesystem state list of extra files and directories in / (not present before this port was installed but present after it was deinstalled) 305288 drwxr-xr-x3 root wheel 64 Nov 24 20:32 home 346190 drwxr-xr-x2 slurmslurm 0 Nov 24 20:32 home/slurm [...] The files in /home/$USER (and /var) should be preserved. Deinstalls may be part of an upgrade process, vs permanent removal, and some of these files contain usage statistics needed by sysadmins for reporting. I would suggest the following deinstall message: You may remove /home/$USER if you are no longer using SLURM. Yep, that is the question: why did QAT suggests manuall removing _and_ exit with error code? -- WBR, Boris Samorodov (bsam) FreeBSD Committer, http://www.FreeBSD.org The Power To Serve ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
[QAT] r334859: 4x leftovers
- Convert to staging. - Take maintainership. - Build ID: 20131125163001-58900 Job owner: m...@freebsd.org Buildtime: 105 minutes Enddate: Mon, 25 Nov 2013 18:15:16 GMT Revision: r334859 Repository: https://svnweb.freebsd.org/ports?view=revisionrevision=334859 - Port:dns/maradns 1.4.12 Buildgroup: 8.4-QAT/amd64 Buildstatus: LEFTOVERS Log: https://qat.redports.org//~m...@freebsd.org/20131125163001-58900-231552/maradns-1.4.12.log Buildgroup: 8.4-QAT/i386 Buildstatus: LEFTOVERS Log: https://qat.redports.org//~m...@freebsd.org/20131125163001-58900-231553/maradns-1.4.12.log Buildgroup: 9.2-QAT/amd64 Buildstatus: LEFTOVERS Log: https://qat.redports.org//~m...@freebsd.org/20131125163001-58900-231554/maradns-1.4.12.log Buildgroup: 9.2-QAT/i386 Buildstatus: LEFTOVERS Log: https://qat.redports.org//~m...@freebsd.org/20131125163001-58900-231555/maradns-1.4.12.log -- Buildarchive URL: https://qat.redports.org/buildarchive/20131125163001-58900 redports https://qat.redports.org/ ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: [QAT] r334788: 1x depend (depend_package in databases/rrdtool), 3x leftovers
Am 25.11.2013 18:30 schrieb Boris Samorodov b...@passap.ru: (reformatted for consistency sake) 25.11.2013 18:26, Jason bacon пишет: On 11/25/13 2:56 AM, Boris Samorodov wrote: I'm not sure how to deal with /home/$USER leftovers (i.e. while package install a new user/group is created and while package deinstalled /home/$USER becomes a leftover, but we display a message that a user should be removed manually): [...] phase 7: make package === Building package for slurm-2.6.4 Deleting slurm-2.6.4 Deinstallation has been requested for the following 1 packages: slurm-2.6.4 The deinstallation will free 51 MB [1/1] Deleting slurm-2.6.4...== You should manually remove the slurm user. done === Checking filesystem state list of extra files and directories in / (not present before this port was installed but present after it was deinstalled) 305288 drwxr-xr-x3 root wheel 64 Nov 24 20:32 home 346190 drwxr-xr-x2 slurmslurm 0 Nov 24 20:32 home/slurm [...] The files in /home/$USER (and /var) should be preserved. Deinstalls may be part of an upgrade process, vs permanent removal, and some of these files contain usage statistics needed by sysadmins for reporting. I would suggest the following deinstall message: You may remove /home/$USER if you are no longer using SLURM. Yep, that is the question: why did QAT suggests manuall removing _and_ exit with error code? With my QAT hat on I consider this a false positive. You can safely ignore that complain from QAT. ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: [QAT] r334788: 1x depend (depend_package in databases/rrdtool), 3x leftovers
25.11.2013 22:44, Bernhard Fröhlich пишет: Am 25.11.2013 18:30 schrieb Boris Samorodov b...@passap.ru: (reformatted for consistency sake) 25.11.2013 18:26, Jason bacon пишет: On 11/25/13 2:56 AM, Boris Samorodov wrote: I'm not sure how to deal with /home/$USER leftovers (i.e. while package install a new user/group is created and while package deinstalled /home/$USER becomes a leftover, but we display a message that a user should be removed manually): [...] phase 7: make package === Building package for slurm-2.6.4 Deleting slurm-2.6.4 Deinstallation has been requested for the following 1 packages: slurm-2.6.4 The deinstallation will free 51 MB [1/1] Deleting slurm-2.6.4...== You should manually remove the slurm user. done === Checking filesystem state list of extra files and directories in / (not present before this port was installed but present after it was deinstalled) 305288 drwxr-xr-x3 root wheel 64 Nov 24 20:32 home 346190 drwxr-xr-x2 slurmslurm 0 Nov 24 20:32 home/slurm [...] The files in /home/$USER (and /var) should be preserved. Deinstalls may be part of an upgrade process, vs permanent removal, and some of these files contain usage statistics needed by sysadmins for reporting. I would suggest the following deinstall message: You may remove /home/$USER if you are no longer using SLURM. Yep, that is the question: why did QAT suggests manuall removing _and_ exit with error code? With my QAT hat on I consider this a false positive. You can safely ignore that complain from QAT. Noted, thanks! -- WBR, Boris Samorodov (bsam) FreeBSD Committer, http://www.FreeBSD.org The Power To Serve ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: FreeBSD Port: py27-pytz-2013.8,1
Fabien Debuire wrote: Hello, Since the update to the 2013.8,1 I'm not able to install the port. I'm running freebsd 9.2 svn with ports up-to-date. The error I have is: --- Installing the new version via the port === Staging for py27-pytz-2013.8,1 === py27-pytz-2013.8,1 depends on file: /usr/local/bin/easy_install-2.7 - found === py27-pytz-2013.8,1 depends on file: /usr/local/bin/python2.7 - found === py27-pytz-2013.8,1 depends on executable: python - found === Generating temporary packing list usage: setup.py [global_opts] cmd1 [cmd1_opts] [cmd2 [cmd2_opts] ...] or: setup.py --help [cmd1 cmd2 ...] or: setup.py --help-commands or: setup.py cmd --help error: option --user not recognized *** [do-install] Error code 1 Stop in /usr/ports/devel/py-pytz. *** [reinstall] Error code 1 Stop in /usr/ports/devel/py-pytz. ** Command failed [exit code 1]: /usr/bin/script -qa /tmp/portupgrade20131116-48239-bu0vrp env UPGRADE_TOOL=portupgrade UPGRADE_PORT=py27-pytz-2013d UPGRADE_PORT_VER=2013d make reinstall --- Restoring the old version 362 385 Similiar Problem here: --- Installing the new version via the port === Staging for py27-boto-2.16.0 === py27-boto-2.16.0 depends on file: /usr/local/bin/easy_install-2.7 - found === py27-boto-2.16.0 depends on file: /usr/local/bin/python2.7 - found === py27-boto-2.16.0 depends on executable: python - found === py27-boto-2.16.0 depends on executable: python2 - found === Generating temporary packing list usage: setup.py [global_opts] cmd1 [cmd1_opts] [cmd2 [cmd2_opts] ...] or: setup.py --help [cmd1 cmd2 ...] or: setup.py --help-commands or: setup.py cmd --help error: option --user not recognized *** [do-install] Error code 1 Stop in /usr/ports/devel/py-boto. *** [reinstall] Error code 1 Stop in /usr/ports/devel/py-boto. ** Command failed [exit code 1]: /usr/bin/script -qa /tmp/portupgrade20131125-87632-1nb1pg7 env UPGRADE_TOOL=portupgrade UPGRADE_PORT=py27-boto-2.14.0 UPGRADE_PORT_VER=2.14.0 make BATCH=yes reinstall --- Restoring the old version 215 239 ** Fix the installation problem and try again. ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: Upgrading Perl... Somebody just shoot me and put me out of my misery!
On Fri, 22 Nov 2013 00:25:26 -0800 Ronald F. Guilmette wrote: Anyway, the specific broken port I'm trying to fix at the moment is net/p5-Socket, which is failing thusly: I had to revert the switch to threaded perl because of this, dns/p5-Net-DNS, net/p5-IP-Country and converters/p5-Encode-Detect. ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: Upgrading Perl... Somebody just shoot me and put me out of my misery!
On Mon, 25 Nov 2013, Matthias Andree wrote: Am 23.11.2013 12:20, schrieb Mark Martinec: On Friday 22 November 2013 21:40:07 Ronald F. Guilmette wrote: Now, one last little thing... The note in the UPDATING file dated 20131120 gives essentially the same instructions as the one dated 20131023, *however* it also contains this: 1) Change the option in lang/perl5.16: make -C /usr/ports/lang/perl5.16 config HUH?? I don't understand this at all. What exactly is the option that we are changing here? And what does it matter to anything? It would be Nice if this were entierly less opaque. $ man ports [...] config Configure OPTIONS for this port using dialog4ports(1). And what does it matter to anything? Gives you a choice to re-think your existing/chosen port options. For example, a new default is now THREADS, but you may not like it, as it somewhat increases the memory usage and requires to rebuild all perl modules. Which shows an interesting facet of this whole tedious process: We're doing a lousy job of explaining the options to unsavvy users, and we're also doing a lousy job of tracking options. I don't know if I'd say it that way, but many users really have a hard time interpreting UPDATING. Perhaps we should just slash down the options and go more for build the default - it also reduces testing complexity and would give for a more uniform ports experience for everyone (packages use default options anyways). That's supposed to be changing. I would even go that far to propose killing some common options such as NLS DOCS EXAMPLES and replace them by a make globcalconfig that sets them system-wide through make.conf, so that we don't need to set/reset them each and every time a port changes options, nor even offer them. That is an excellent idea! This would also act as documentation of what global options are available. ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: Upgrading Perl... Somebody just shoot me and put me out of my misery!
On Mon, Nov 25, 2013 at 10:10:20PM +, RW wrote: On Fri, 22 Nov 2013 00:25:26 -0800 Ronald F. Guilmette wrote: Anyway, the specific broken port I'm trying to fix at the moment is net/p5-Socket, which is failing thusly: I had to revert the switch to threaded perl because of this, dns/p5-Net-DNS, net/p5-IP-Country and converters/p5-Encode-Detect. For those, I ended up forcibly de-installing them, then building them. Seemed to go OK (after I applied that hammer). Peace, david -- David H. Wolfskill da...@catwhisker.org Taliban: Evil cowards with guns afraid of truth from a 14-year old girl. See http://www.catwhisker.org/~david/publickey.gpg for my public key. pgp4QIDDZUCjS.pgp Description: PGP signature
Re: FreeBSD Port: py27-pytz-2013.8,1
Hi, [2013/11/25 Helmut Schneider jumpe...@gmx.de] Similiar Problem here: --- Installing the new version via the port === Staging for py27-boto-2.16.0 [...] error: option --user not recognized Output of the following commands? # cat /etc/make.conf # easy_install-2.7 --version # (cd /usr/ports/devel/py-boto make -n) What options/command do you use to run portupgrade? What happens if you update the port with the following command? (cd /usr/ports/devel/py-boto make deinstall install) Thanks. -- Nicola Vitale ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
sysutils/parallel maintainer: looking for guidance
The latest version of parallel (20131122) has added a Please use the below reference to cite parallel notice that prints a few lines of output to stderr when it runs. User action is required to stop this notice from appearing upon every invocation (by running parallel --bibtex once). My concern is that anyone currently running automated scripts using parallel may be caught off guard by this, should I keep it anyway, or comment it out in a patch? Maybe a pkg-message, or a note in UPDATING, or is it even worth worrying about? ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
[QAT] r334895: 4x leftovers
- Fix the usage of 'python' to get rid of the implicit lang/python dependency - Reset maintainer Approved by:Bruno Schwander br...@tinkerbox.org - Build ID: 20131125202000-44968 Job owner: m...@freebsd.org Buildtime: 5 hours Enddate: Tue, 26 Nov 2013 01:44:10 GMT Revision: r334895 Repository: https://svnweb.freebsd.org/ports?view=revisionrevision=334895 - Port:games/childsplay 0.90.2_2 Buildgroup: 8.4-QAT/amd64 Buildstatus: LEFTOVERS Log: https://qat.redports.org//~m...@freebsd.org/20131125202000-44968-231732/childsplay-0.90.2_2.log Buildgroup: 8.4-QAT/i386 Buildstatus: LEFTOVERS Log: https://qat.redports.org//~m...@freebsd.org/20131125202000-44968-231733/childsplay-0.90.2_2.log Buildgroup: 9.2-QAT/amd64 Buildstatus: LEFTOVERS Log: https://qat.redports.org//~m...@freebsd.org/20131125202000-44968-231734/childsplay-0.90.2_2.log Buildgroup: 9.2-QAT/i386 Buildstatus: LEFTOVERS Log: https://qat.redports.org//~m...@freebsd.org/20131125202000-44968-231735/childsplay-0.90.2_2.log -- Buildarchive URL: https://qat.redports.org/buildarchive/20131125202000-44968 redports https://qat.redports.org/ ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Making WITH_DEBUG less magical
Hi all,
At the moment WITH_DEBUG is treated specially by the ports system. I
propose to turn the old WITH_DEBUG into a port option 'DEBUG' which is
understood by the port system as a whole (similar to DOCS, etc.)
This change also changes what DEBUG does: it only appends
DEBUG_{CFLAGS,CPPFLAGS,CXXFLAGS} to the appropriate variable. It does
not any longer strip off -O*.
My first attempt is below. One known problem is that requires the
option DEBUG to be in OPTIONS_DEFINE. I am not sure how to change this
(or if we want to change it). I am also unsure if the code changes
should live above or below the inclusion of Mk/bsd.options.mk.
Documentation changes are intentionally ignored at the moment.
Thoughts?
Index: Mk/bsd.port.mk
===
--- Mk/bsd.port.mk (revision 334916)
+++ Mk/bsd.port.mk (working copy)
@@ -1311,17 +1311,6 @@ MAKE_ENV+= TMPDIR=${TMPDIR}
CONFIGURE_ENV+= TMPDIR=${TMPDIR}
.endif # defined(TMPDIR)
-.if defined(WITH_DEBUG_PORTS)
-.if ${WITH_DEBUG_PORTS:M${PKGORIGIN}}
-WITH_DEBUG= yes
-.endif
-.endif
-
-# Reset value from bsd.own.mk.
-.if defined(WITH_DEBUG) !defined(WITHOUT_DEBUG)
-STRIP= #none
-.endif
-
.include ${PORTSDIR}/Mk/bsd.options.mk
# Start of pre-makefile section.
@@ -1603,11 +1592,25 @@ CFLAGS:= ${CFLAGS:C/${_CPUCFLAGS}//}
.endif
.endif
-.if defined(WITH_DEBUG) !defined(WITHOUT_DEBUG)
+.if defined(WITH_DEBUG_PORTS)
+.if ${WITH_DEBUG_PORTS:M${PKGORIGIN}}
+WITH_DEBUG= yes
+.endif
+.endif
+
+DEBUG_CFLAGS?= -g
+DEBUG_CXXFLAGS?= -g
+DEBUG_CPPFLAGS?= -g
+
+.if defined(PORT_OPTIONS)
+.if ${PORT_OPTIONS:MDEBUG} || defined(WITH_DEBUG)
+STRIP= #none
STRIP_CMD= ${TRUE}
-DEBUG_FLAGS?= -g
-CFLAGS:= ${CFLAGS:N-O*:N-fno-strict*} ${DEBUG_FLAGS}
+CFLAGS+= ${DEBUG_CFLAGS}
+CXXFLAGS+= ${DEBUG_CXXFLAGS}
+CPPFLAGS+= ${DEBUG_CPPFLAGS}
.endif
+.endif
.if defined(WITH_SSP) || defined(WITH_SSP_PORTS)
.include ${PORTSDIR}/Mk/bsd.ssp.mk
--
Eitan Adler
___
freebsd-ports@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: Making WITH_DEBUG less magical
On 26/11/2013 4:14 PM, Eitan Adler wrote:
Hi all,
At the moment WITH_DEBUG is treated specially by the ports system. I
propose to turn the old WITH_DEBUG into a port option 'DEBUG' which is
understood by the port system as a whole (similar to DOCS, etc.)
This change also changes what DEBUG does: it only appends
DEBUG_{CFLAGS,CPPFLAGS,CXXFLAGS} to the appropriate variable. It does
not any longer strip off -O*.
My first attempt is below. One known problem is that requires the
option DEBUG to be in OPTIONS_DEFINE. I am not sure how to change this
(or if we want to change it). I am also unsure if the code changes
should live above or below the inclusion of Mk/bsd.options.mk.
Documentation changes are intentionally ignored at the moment.
Thoughts?
Index: Mk/bsd.port.mk
===
--- Mk/bsd.port.mk (revision 334916)
+++ Mk/bsd.port.mk (working copy)
@@ -1311,17 +1311,6 @@ MAKE_ENV+= TMPDIR=${TMPDIR}
CONFIGURE_ENV+= TMPDIR=${TMPDIR}
.endif # defined(TMPDIR)
-.if defined(WITH_DEBUG_PORTS)
-.if ${WITH_DEBUG_PORTS:M${PKGORIGIN}}
-WITH_DEBUG= yes
-.endif
-.endif
-
-# Reset value from bsd.own.mk.
-.if defined(WITH_DEBUG) !defined(WITHOUT_DEBUG)
-STRIP= #none
-.endif
-
.include ${PORTSDIR}/Mk/bsd.options.mk
# Start of pre-makefile section.
@@ -1603,11 +1592,25 @@ CFLAGS:= ${CFLAGS:C/${_CPUCFLAGS}//}
.endif
.endif
-.if defined(WITH_DEBUG) !defined(WITHOUT_DEBUG)
+.if defined(WITH_DEBUG_PORTS)
+.if ${WITH_DEBUG_PORTS:M${PKGORIGIN}}
+WITH_DEBUG= yes
+.endif
+.endif
+
+DEBUG_CFLAGS?= -g
+DEBUG_CXXFLAGS?= -g
+DEBUG_CPPFLAGS?= -g
+
+.if defined(PORT_OPTIONS)
+.if ${PORT_OPTIONS:MDEBUG} || defined(WITH_DEBUG)
+STRIP= #none
STRIP_CMD= ${TRUE}
-DEBUG_FLAGS?= -g
-CFLAGS:= ${CFLAGS:N-O*:N-fno-strict*} ${DEBUG_FLAGS}
+CFLAGS+= ${DEBUG_CFLAGS}
+CXXFLAGS+= ${DEBUG_CXXFLAGS}
+CPPFLAGS+= ${DEBUG_CPPFLAGS}
.endif
+.endif
.if defined(WITH_SSP) || defined(WITH_SSP_PORTS)
.include ${PORTSDIR}/Mk/bsd.ssp.mk
Thanks Eitan :) Its much easier for me to see/grok in your code how we
can evolve and get to:
- DEBUG sets good global defaults (if any)
- DEBUG is user-configurable (make.conf *and/or* make config)
- A port can define DEBUG as an OPTION
- A port can extend DEBUG with additional flags, defines, tools, etc
- b.p.m provides common boilerplate (where it is valuable) to:
-- reduce code duplication in port Makefiles
-- ensure consistent use of DEBUG
-- eg: CONFIGURE_ARGS+=--enable-debug (if GNU_CONFIGURE)
-- eg: CONFIGURE_ARGS+=--disable-strip (if GNU_CONFIGURE)
-- eg: User messaging (if necessary and useful)
-- eg: im sure many others
On the above goals (not a code review), your changes look great :]
Thanks again
--
koobs
___
freebsd-ports@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
gvfs on FreeBSD 10 fails to build
Hello all, I just updated my FreeBSD 9 Stable machine to FreeBSD 10 Stable. I am now in the proces of rebuilding all ports. First I noticed that libiconv was not needed anymore from ports so I deleted that port. But on devel/gvfs I get an error. I am stuck with the following port. grep: /usr/local/lib/libiconv.la: No such file or directory sed: /usr/local/lib/libiconv.la: No such file or directory libtool: link: `/usr/local/lib/libiconv.la' is not a valid libtool archive gmake[4]: *** [libgvfscommon-dnssd.la] Error 1 gmake[4]: Leaving directory `/usr/ports/devel/gvfs/work/gvfs-1.12.3/common' gmake[3]: *** [all-recursive] Error 1 gmake[3]: Leaving directory `/usr/ports/devel/gvfs/work/gvfs-1.12.3' gmake[2]: *** [all] Error 2 gmake[2]: Leaving directory `/usr/ports/devel/gvfs/work/gvfs-1.12.3' *** Error code 1 Stop. make[1]: stopped in /usr/ports/devel/gvfs *** Error code 1 Stop. make: stopped in /usr/ports/devel/gvfs regards Johan ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
