>>> Building on exopi-2 under devel/p5-EV BDEPENDS = [devel/p5-Canary-Stability] DIST = [devel/p5-EV:EV-4.34.tar.gz] FULLPKGNAME = p5-EV-4.34 RDEPENDS = [devel/p5-common-sense] (Junk lock obtained for exopi-2 at 1713770044.23) >>> Running depends in devel/p5-EV at 1713770044.27 /usr/sbin/pkg_add -aI -Drepair p5-Canary-Stability-2013p0 was: /usr/sbin/pkg_add -aI -Drepair p5-Canary-Stability-2013p0 /usr/sbin/pkg_add -aI -Drepair p5-Canary-Stability-2013p0 >>> Running show-prepare-results in devel/p5-EV at 1713770046.13 ===> devel/p5-EV ===> Building from scratch p5-EV-4.34 ===> p5-EV-4.34 depends on: p5-Canary-Stability-* -> p5-Canary-Stability-2013p0 ===> Verifying specs: c perl pthread ===> found c.100.0 perl.23.0 pthread.27.1 p5-Canary-Stability-2013p0 (Junk lock released for exopi-2 at 1713770047.67) distfiles size=220047 >>> Running build in devel/p5-EV at 1713770047.73 ===> devel/p5-EV ===> Checking files for p5-EV-4.34 `/exopi-cvs/ports/distfiles/EV-4.34.tar.gz' is up to date. >> (SHA256) EV-4.34.tar.gz: OK ===> Extracting for p5-EV-4.34 ===> Patching for p5-EV-4.34 ===> Compiler link: clang -> /usr/bin/clang ===> Compiler link: clang++ -> /usr/bin/clang++ ===> Compiler link: cc -> /usr/bin/cc ===> Compiler link: c++ -> /usr/bin/c++ ===> Generating configure for p5-EV-4.34 ===> Configuring for p5-EV-4.34 *** *** Canary::Stability COMPATIBILITY AND SUPPORT CHECK *** ================================================= *** *** Hi! *** *** I do my best to provide predictable and reliable software. *** *** However, in recent releases, P5P (who maintain perl) have been *** introducing regressions that are sometimes subtle and at other times *** catastrophic, often for personal preferences with little or no concern *** for existing code, most notably CPAN. *** *** For this reason, it has become very hard for me to maintain the level *** of reliability and support I have committed myself to in the past, at *** least with some perl versions: I simply can't keep up working around new *** bugs or gratituous incompatibilities, and in turn you might suffer from *** unanticipated problems. *** *** Therefore I have introduced a support and compatibility check, the results *** of which follow below, together with a FAQ and some recommendations. *** *** This check is just to let you know that there might be a risk, so you can *** make judgement calls on how to proceed - it will not keep the module from *** installing or working. *** *** The stability canary says: (nothing, it was driven away by harsh weather) *** *** It seems you are running perl version 5.036003, likely the "official" or *** "standard" version. While there is nothing wrong with doing that, *** standard perl versions 5.022 and up are not supported by EV. *** While this might be fatal, it might also be all right - if you run into *** problems, you might want to downgrade your perl or switch to the *** stability branch. *** *** If everything works fine, you can ignore this message. *** *** *** Stability canary mini-FAQ: *** *** Do I need to do anything? *** With luck, no. While some distributions are known to fail *** already, most should probably work. This message is here *** to alert you that your perl is not supported by EV, *** and if things go wrong, you either need to downgrade, or *** sidegrade to the stability variant of your perl version, *** or simply live with the consequences. *** *** What is this canary thing? *** It's purpose is to check support status of EV with *** respect to your perl version. *** *** What is this "stability branch"? *** It's a branch or fork of the official perl, by schmorp, to *** improve stability and compatibility with existing modules. *** *** How can I skip this prompt on automated installs? *** Set PERL_CANARY_STABILITY_NOPROMPT=1 in your environment. *** More info is in the Canary::Stability manpage. *** *** Long version of this FAQ: http://stableperl.schmorp.de/faq.html *** Stability Branch homepage: http://stableperl.schmorp.de/ *** Continue anyways? [y] y *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** Welcome to EV configuration. If you are in a hurry, just press return here and hope for the best. The defaults should usually do. Skip further questions and use defaults (y/n)? [y] y *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** POSIX optionally offers support for a monotonic clock source. EV can take advantage of this clock source to detect time jumps more reliably. Unfortunately, some systems are bound to be broken, so you can disable this here: you can completely disable the detection and use of the monotonic clock by answering 'n' here. Support for this clock type will otherwise be autodetected at both compile- and runtime. (this setting currently affects the use of nanosleep over select as well). Enable optional support for CLOCK_MONOTONIC (y/n)? [y] y *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** POSIX optionally offers support for a (potentially) high-resolution realtime clock interface. In a good implementation, using it is faster than the normal method of using gettimeofday. Unfortunately, this option is also bound to be broken on some systems, and current EV versions do not actually call gettimeofday very often, so it defaults to no. Prefer clock_gettime (CLOCK_REALTIME) over gettimeofday (y/n)? [n] n *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** EV can use various backends with various portability issues. The select backend is the most portable and makes for a good fallback, but it can be limited to a low number of file descriptors and/or might not compile. If you have problems with compiling ev_select.c, you might try to play around with disabling it here, or forcing it to use the fd_set provided by your OS, via the next question. I highly recommend keeping it in. Enable select backend (y/n)? [y] y *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** The select backend can operate in two modes. One uses the system-provided fd_set and is usually limited to 1024 file descriptors (64 on windows), the other requires your header files to define NFDBITS and declare a suitable fd_mask type. If you run into problems compiling ev_select.c, you can try forcing the use of the system fd_set here. Force use of system fd_set for select backend (y/n)? [n] n *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** The second very portable backend is poll(2). It does not exist on windows and various versions of Mac OS X (and on the other versions it simply doesn't work), but works basically everywhere else. It is recommended to use the default here unless you run into compilation problems in ev_poll.c. Enable poll backend (y/n)? [y] y *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** Select and poll make it hard to write efficient servers, especially if the number of active connections is much lower than the watched ones. GNU/Linux systems have a more scalable method called "epoll", which EV can use. For this to work, both your kernel and glibc have to support epoll, but if you can compile it, the detection will be done at runtime, and EV will safely fall back to using select when epoll isn't available. If unsure, accept the default. Enable epoll backend (y/n)? [n] n *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** Linux 4.18 introduced another event polling interface, this time using the Linux AIO API. While this API is far superior to epoll and almost rivals kqueue, it also suffers from the same issues as kqueue typically does: only a subset of file types are supported (as of 4.19, I have seen eventfd, pipes, sockets files and some devices, but no ttys). It also is subject arbitrary system-wide limits imposed on it. Therefore, this backend is not used by default, even when it is compiled in, and you have to request it explicitly, e.g. with LIBEV_FLAGS=64. If unsure, accept the default. Enable linux aio backend (y/n)? [n] n *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** Linux 4.19 introduced another event polling interface, "io_uring". While this API is far superior to epoll and almost rivals linuxaio, it also suffers from the same issues as kqueue typically does: only a subset of file types are supported (as of 5.2). It is also very buggy still, and most importantly, very very slow for most workloads. Therefore, this backend is not used by default, even when it is compiled in, and you have to request it explicitly, e.g. with LIBEV_FLAGS=128. If unsure, accept the default. Enable linux io_uring backend (y/n)? [n] n *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** EV can take advantage of kqueue on many BSD systems. Support for kqueue will be detected at runtime, with a safe fallback to other methods when it cannot be used. Note that kqueue is subtly broken on most operating systems, so by default it won't be used on many platforms, but you can still create your own event loop with kqueue backend if you ask specifically for it. Here is what we know: NetBSD: partially working in at least 3.1 and later. Yeah! :) FreeBSD: broken on at least 6.2-STABLE, spotty in later versions, sockets *likely* work, ptys definitely don't. OpenBSD: reports indicate that it likely doesn't work (similar problems as on FreeBSD). OS X: completely, utterly broken on at least <= 10.6. Enable kqueue backend (y/n)? [y] y *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** Similarly to the kqueue backend above, EV can take advantage of the solaris 10 event port interface. Support for event ports will be detected at runtime, with a safe fallback to other methods when it cannot be used. Enable event port backend (y/n)? [n] n *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** EV needs the functions pthread_atfork and clock_gettime. On most systems you need some special libraries for this (such as -lrt and -lpthread). You can specify additional libraries to provide these calls (and any other required by EV) now, or accept the default. On GNU/Linux systems, EV uses the LSB 3.1 __register_atfork function to avoid the dependency on libpthread, and directly uses the clock_gettime syscall to avoid a dependency on librt. Extra libraries for pthread_atfork and clock_gettime? [-lpthread -lrt ] -lpthread -lrt *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** A backend of a different kind is the Linux inotify(7) interface, which can be used to speed up (and reduce resource consumption) of stat watchers. If you have the include file and libc support for it, it is usually a good idea to enable it, as kernel availability is detected at runtime. Enable inotify support (y/n)? [n] n *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** Another useful bit of functionality is the Linux eventfd, which is useful for faster signal handling (don't care) and intra-thread communications (more relevant). Kernel support for this will be probed at runtime, but your libc must contain the necessary wrapper. Glibc 2.7 and later should have this wrapper. Enable linux eventfd support (y/n)? [n] n *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** Another sometimes useful bit of functionality is the Linux signalfd, which is useful for faster signal handling (don't care). Kernel support for this will be probed at runtime, but your libc must contain the necessary wrapper. Glibc 2.7 and later should have this wrapper. Enable linux signalfd support (y/n)? [n] n *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** Linux kernels can notify userspace about realtime clock timejumps using timerfd. Libev by default will try to take advantage of this if possible. You can completely disable the detection and use of timerfd for this purpose by answering 'n' here. Support for timerfd will otherwise be autodetected at both compile- and runtime. Enable optional support for timerfd to detect timejumps (y/n)? [y] y *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** Libev contains numerous internal assert() invocations to check for consistency and user errors. These are normally enabled, but most perl builds disable this error reporting mechanism by default. You can re-enable these asserts here. Enabling them might help you catch programming bugs earlier, but might cause a small slowdown. Also, failures will be reported by aboritng your program, instead of throwing a perl exception. If unsure, enable this if you only use this perl installation for development, and leave it off for use in production environments. Make sure assertions are enabled? (y/n)? [n] n *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** Very rarely, people want to tweak EV even more, e.g. to exclude or include certain watcher types or backends. This can be done by adding extra -D options here, or via the EV_EXTRA_DEFS environment variable. For example, if you run into compilation problems because of missing memory fences (or you just want extra performance), you can tell EV to not support smp and threads via -DEV_NO_THREADS. Most people would just press enter. Any extra -D options? [] Warning: prerequisite common::sense 0 not found. *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** Checking if your kit is complete... Looks good Warning (mostly harmless): No library found for -lrt Generating a Unix-style Makefile Writing Makefile for EV Writing MYMETA.yml and MYMETA.json ===> Building for p5-EV-4.34 cp libev/ev.h blib/lib/EV/ev.h cp EV/EVAPI.h blib/lib/EV/EVAPI.h cp libev/ev.pod blib/lib/EV/libev.pod cp EV.pm blib/lib/EV.pm cp EV/MakeMaker.pm blib/lib/EV/MakeMaker.pm Running Mkbootstrap for EV () chmod 644 "EV.bs" "/usr/bin/perl" -MExtUtils::Command::MM -e 'cp_nonempty' -- EV.bs blib/arch/auto/EV/EV.bs 644 "/usr/bin/perl" "/usr/libdata/perl5/ExtUtils/xsubpp" -typemap '/usr/libdata/perl5/ExtUtils/typemap' -typemap '/exopi-obj/pobj/p5-EV-4.34/EV-4.34/typemap' EV.xs > EV.xsc mv EV.xsc EV.c cc -c -Ilibev -DNO_LOCALE_NUMERIC -DNO_LOCALE_COLLATE -fno-strict-aliasing -fno-delete-null-pointer-checks -pipe -fstack-protector-strong -I/usr/local/include -O2 -DVERSION=\"4.34\" -DXS_VERSION=\"4.34\" -DPIC -fpic "-I/usr/libdata/perl5/amd64-openbsd/CORE" -DEV_USE_REALTIME=0 -DEV_USE_SELECT=1 -DEV_USE_POLL=1 -DEV_USE_EPOLL=0 -DEV_USE_LINUXAIO=0 -DEV_USE_IOURING=0 -DEV_USE_KQUEUE=1 -DEV_USE_PORT=0 -DEV_USE_INOTIFY=0 -DEV_USE_EVENTFD=0 -DEV_USE_SIGNALFD=0 EV.c rm -f blib/arch/auto/EV/EV.so LD_RUN_PATH="/usr/lib" cc -shared -fpic -fstack-protector-strong -L/usr/local/lib EV.o -o blib/arch/auto/EV/EV.so -lpthread chmod 755 blib/arch/auto/EV/EV.so Manifying 3 pod documents >>> Running package in devel/p5-EV at 1713770058.79 ===> devel/p5-EV ===> Faking installation for p5-EV-4.34 "/usr/bin/perl" -MExtUtils::Command::MM -e 'cp_nonempty' -- EV.bs blib/arch/auto/EV/EV.bs 644 Manifying 3 pod documents Files found in blib/arch: installing files in blib/lib into architecture dependent library tree Installing /exopi-obj/pobj/p5-EV-4.34/fake-amd64/usr/local/libdata/perl5/site_perl/amd64-openbsd/auto/EV/EV.so Installing /exopi-obj/pobj/p5-EV-4.34/fake-amd64/usr/local/libdata/perl5/site_perl/amd64-openbsd/EV.pm Installing /exopi-obj/pobj/p5-EV-4.34/fake-amd64/usr/local/libdata/perl5/site_perl/amd64-openbsd/EV/ev.h Installing /exopi-obj/pobj/p5-EV-4.34/fake-amd64/usr/local/libdata/perl5/site_perl/amd64-openbsd/EV/EVAPI.h Installing /exopi-obj/pobj/p5-EV-4.34/fake-amd64/usr/local/libdata/perl5/site_perl/amd64-openbsd/EV/libev.pod Installing /exopi-obj/pobj/p5-EV-4.34/fake-amd64/usr/local/libdata/perl5/site_perl/amd64-openbsd/EV/MakeMaker.pm Installing /exopi-obj/pobj/p5-EV-4.34/fake-amd64/usr/local/man/man3p/EV.3p Installing /exopi-obj/pobj/p5-EV-4.34/fake-amd64/usr/local/man/man3p/EV::MakeMaker.3p Installing /exopi-obj/pobj/p5-EV-4.34/fake-amd64/usr/local/man/man3p/EV::libev.3p Appending installation info to /exopi-obj/pobj/p5-EV-4.34/fake-amd64/usr/./libdata/perl5/amd64-openbsd/perllocal.pod ===> Building package for p5-EV-4.34 Create /exopi-cvs/ports/packages/amd64/all/p5-EV-4.34.tgz Creating package p5-EV-4.34 reading plist| checking dependencies| checking dependencies|devel/p5-common-sense checksumming| checksumming| | 0% checksumming|*** | 4% checksumming|***** | 9% checksumming|******** | 13% checksumming|*********** | 17% checksumming|************* | 22% checksumming|**************** | 26% checksumming|******************* | 30% checksumming|********************* | 35% checksumming|************************ | 39% checksumming|*************************** | 43% checksumming|***************************** | 48% checksumming|******************************** | 52% checksumming|********************************** | 57% checksumming|************************************* | 61% checksumming|**************************************** | 65% checksumming|****************************************** | 70% checksumming|********************************************* | 74% checksumming|************************************************ | 78% checksumming|************************************************** | 83% checksumming|***************************************************** | 87% checksumming|******************************************************** | 91% checksumming|********************************************************** | 96% checksumming|*************************************************************|100% archiving| archiving| | 0% archiving|************* | 20% archiving|**************** | 25% archiving|****************** | 29% archiving|************************************ | 57% archiving|***************************************** | 64% archiving|*************************************************************** | 98% archiving|*************************************************************** | 99% archiving|****************************************************************|100% Link to /exopi-cvs/ports/packages/amd64/ftp/p5-EV-4.34.tgz >>> Running clean in devel/p5-EV at 1713770064.77 ===> devel/p5-EV ===> Cleaning for p5-EV-4.34 >>> Ended at 1713770065.61 max_stuck=1.25/depends=1.78/show-prepare-results=1.66/build=10.13/package=6.72/clean=1.09