=> Bootstrap dependency digest>=20010302: found digest-20180917 ===> Skipping vulnerability checks. WARNING: No /var/db/pkg/pkg-vulnerabilities file found. WARNING: To fix run: `/usr/sbin/pkg_admin -K /var/db/pkg fetch-pkg-vulnerabilities'. => Checksum SHA1 OK for ghc-7.6.3-boot-x86_64-unknown-netbsd.tar.xz => Checksum RMD160 OK for ghc-7.6.3-boot-x86_64-unknown-netbsd.tar.xz => Checksum SHA512 OK for ghc-7.6.3-boot-x86_64-unknown-netbsd.tar.xz => Checksum SHA1 OK for ghc-7.6.3-src.tar.bz2 => Checksum RMD160 OK for ghc-7.6.3-src.tar.bz2 => Checksum SHA512 OK for ghc-7.6.3-src.tar.bz2 ===> Installing dependencies for ghc-7.6.3nb12 ========================================================================== The following variables will affect the build process of this package, ghc-7.6.3nb12. Their current value is shown below: * CURSES_DEFAULT = curses Based on these variables, the following variables have been set: * CURSES_TYPE = curses You may want to abort the process now with CTRL-C and change their value before continuing. Be sure to run `/usr/bin/make clean' after the changes. ========================================================================== => Tool dependency libtool-base>=2.4.2nb9: found libtool-base-2.4.6nb2 => Tool dependency autoconf>=2.50: found autoconf-2.69nb8 => Tool dependency gmake>=3.81: found gmake-4.2.1nb1 => Tool dependency checkperms>=1.1: found checkperms-1.12 => Build dependency libxslt-[0-9]*: found libxslt-1.1.32nb1 => Build dependency docbook-xsl-[0-9]*: found docbook-xsl-1.79.1nb3 => Build dependency cwrappers>=20150314: found cwrappers-20180325 => Full dependency perl>=5.0: found perl-5.28.1 => Full dependency gmp>=5.0.1: found gmp-6.1.2 ===> Overriding tools for ghc-7.6.3nb12 ===> Extracting for ghc-7.6.3nb12 ===> Patching for ghc-7.6.3nb12 => Applying pkgsrc patches for ghc-7.6.3nb12 => Verifying /data/pkgsrc/lang/ghc7/patches/patch-Makefile => Applying pkgsrc patch /data/pkgsrc/lang/ghc7/patches/patch-Makefile Hmm... Looks like a unified diff to me... The text leading up to this was: -------------------------- |$NetBSD: patch-Makefile,v 1.1 2013/12/12 12:47:17 obache Exp $ | |Required by patches/patch-ghc.mk | | |--- Makefile.orig 2013-02-14 02:53:07.000000000 +0000 |+++ Makefile -------------------------- Patching file Makefile using Plan A... Hunk #1 succeeded at 75. done => Verifying /data/pkgsrc/lang/ghc7/patches/patch-ghc.mk => Applying pkgsrc patch /data/pkgsrc/lang/ghc7/patches/patch-ghc.mk Hmm... Looks like a unified diff to me... The text leading up to this was: -------------------------- |$NetBSD: patch-ghc.mk,v 1.1 2013/12/12 12:47:17 obache Exp $ | |We want bootstrapping bindists to be as small as possible, and bzip2 |is not enough: | | % du -sh ghc-*.tar* | 291M ghc-7.6.2-boot-powerpc-apple-darwin.tar | 60M ghc-7.6.2-boot-powerpc-apple-darwin.tar.bz2 | 32M ghc-7.6.2-boot-powerpc-apple-darwin.tar.xz | | |--- ghc.mk.orig 2013-02-14 02:24:35.000000000 +0000 |+++ ghc.mk -------------------------- Patching file ghc.mk using Plan A... Hunk #1 succeeded at 1009. done => Verifying /data/pkgsrc/lang/ghc7/patches/patch-libffi_ghc.mk => Applying pkgsrc patch /data/pkgsrc/lang/ghc7/patches/patch-libffi_ghc.mk Hmm... Looks like a unified diff to me... The text leading up to this was: -------------------------- |$NetBSD: patch-libffi_ghc.mk,v 1.1 2016/05/07 10:02:06 joerg Exp $ | |--- libffi/ghc.mk.orig 2015-09-22 19:10:58.000000000 +0000 |+++ libffi/ghc.mk -------------------------- Patching file libffi/ghc.mk using Plan A... Hunk #1 succeeded at 31. Hunk #2 succeeded at 85. Hunk #3 succeeded at 103. done => Verifying /data/pkgsrc/lang/ghc7/patches/patch-libraries_base_configure.ac => Applying pkgsrc patch /data/pkgsrc/lang/ghc7/patches/patch-libraries_base_configure.ac Hmm... Looks like a unified diff to me... The text leading up to this was: -------------------------- |$NetBSD: patch-libraries_base_configure.ac,v 1.2 2015/02/16 07:04:58 pho Exp $ | |Upstream Trac #10093: |https://ghc.haskell.org/trac/ghc/ticket/10093 | |Do not clobber CPPFLAGS nor LDFLAGS: | |When we are using pkgsrc converters/libiconv, we have libiconv.so in |${PREFIX}/lib and "-Wl,-R${PREFIX}/lib" in LDFLAGS. In this case |FP_SEARCH_LIBS_PROTO(iconv) appends "-liconv" to $LIBS so it will be |linked to any conftest executables that follow, including one which |will be generated by AC_CHECK_SIZEOF(). | |The problem is that if libraries listed in $LIBS are in a non-standard |path while rpath flags are missing (due to LDFLAGS being clobbered in |this case), conftest executables cannot run even if they can be |linked. And if anything goes wrong during AC_CHECK_SIZEOF(T), it |considers sizeof(T) as 0 unless T is known to be an existing type: | | ... | checking for library containing iconv... -liconv | checking for library containing locale_charset... none required | checking size of struct MD5Context... 0 | ... | |This means SIZEOF_STRUCT_MD5CONTEXT gets defined to 0, |GHC.Fingerprint.fingerprintData allocates 0 bytes on the heap, |MD5Init/Update/Final corrupts the heap and then Bad Things will |happen. | |--- libraries/base/configure.ac.orig 2013-04-18 21:30:14.000000000 +0000 |+++ libraries/base/configure.ac -------------------------- Patching file libraries/base/configure.ac using Plan A... Hunk #1 succeeded at 70. Hunk #2 succeeded at 183. done => Verifying /data/pkgsrc/lang/ghc7/patches/patch-libraries_integer-gmp_configure.ac => Applying pkgsrc patch /data/pkgsrc/lang/ghc7/patches/patch-libraries_integer-gmp_configure.ac Hmm... Looks like a unified diff to me... The text leading up to this was: -------------------------- |$NetBSD: patch-libraries_integer-gmp_configure.ac,v 1.1 2013/12/12 12:47:17 obache Exp $ | |In Darwin, integer-gmp prefers GMP.framework when it happens to be |installed somewhere in the system, even though it doesn't come with |Darwin. Let it stop that. (Already fixed in the upstream.) | | |--- libraries/integer-gmp/configure.ac.orig 2013-01-27 13:36:41.000000000 +0000 |+++ libraries/integer-gmp/configure.ac -------------------------- Patching file libraries/integer-gmp/configure.ac using Plan A... Hunk #1 succeeded at 45. done => Verifying /data/pkgsrc/lang/ghc7/patches/patch-libraries_integer-simple_GHC_Integer.hs => Applying pkgsrc patch /data/pkgsrc/lang/ghc7/patches/patch-libraries_integer-simple_GHC_Integer.hs Hmm... Looks like a unified diff to me... The text leading up to this was: -------------------------- |$NetBSD: patch-libraries_integer-simple_GHC_Integer.hs,v 1.1 2013/12/12 12:47:17 obache Exp $ | |Missing functions: already fixed in the upstream. | | |--- libraries/integer-simple/GHC/Integer.hs.orig 2013-02-13 18:50:32.000000000 +0000 |+++ libraries/integer-simple/GHC/Integer.hs -------------------------- Patching file libraries/integer-simple/GHC/Integer.hs using Plan A... Hunk #1 succeeded at 27. done => Verifying /data/pkgsrc/lang/ghc7/patches/patch-libraries_integer-simple_GHC_Integer_Type.hs => Applying pkgsrc patch /data/pkgsrc/lang/ghc7/patches/patch-libraries_integer-simple_GHC_Integer_Type.hs Hmm... Looks like a unified diff to me... The text leading up to this was: -------------------------- |$NetBSD: patch-libraries_integer-simple_GHC_Integer_Type.hs,v 1.1 2013/12/12 12:47:17 obache Exp $ | |Missing functions: already fixed in the upstream. | | |--- libraries/integer-simple/GHC/Integer/Type.hs.orig 2013-02-13 18:45:55.000000000 +0000 |+++ libraries/integer-simple/GHC/Integer/Type.hs -------------------------- Patching file libraries/integer-simple/GHC/Integer/Type.hs using Plan A... Hunk #1 succeeded at 370. done => Verifying /data/pkgsrc/lang/ghc7/patches/patch-libraries_time_Data_Time_Clock_CTimeval.hs => Applying pkgsrc patch /data/pkgsrc/lang/ghc7/patches/patch-libraries_time_Data_Time_Clock_CTimeval.hs Hmm... Looks like a unified diff to me... The text leading up to this was: -------------------------- |$NetBSD: patch-libraries_time_Data_Time_Clock_CTimeval.hs,v 1.1 2015/02/13 19:28:40 pho Exp $ | |Suppress linker warnings about compatibility syscall wrappers by using |"capi" instead of "ccall". In Haskell FFI, "ccall" is actually an |interface to C ABI rather than C API. That is, GHC generates direct |references to the symbol even if it's actually defined as a cpp macro |or something like that, because GHC knows nothing about those macros |in foreign headers. Hence the following warnings: | | .../libHStime-1.4.0.1.a(CTimeval.o): In function `s1Kp_info': | (.text+0x2f): warning: warning: reference to compatibility | gettimeofday(); include to generate correct reference | |In other words, you can safely use "ccall" only when you are sure the |symbol you want to import is actually a symbol in the ABI sense, which |is not always the case for the POSIX API. | |--- libraries/time/Data/Time/Clock/CTimeval.hs.orig 2015-02-13 15:42:45.000000000 +0000 |+++ libraries/time/Data/Time/Clock/CTimeval.hs -------------------------- Patching file libraries/time/Data/Time/Clock/CTimeval.hs using Plan A... Hunk #1 succeeded at 1. Hunk #2 succeeded at 21. done => Verifying /data/pkgsrc/lang/ghc7/patches/patch-libraries_unix_System_Posix_Files.hsc => Applying pkgsrc patch /data/pkgsrc/lang/ghc7/patches/patch-libraries_unix_System_Posix_Files.hsc Hmm... Looks like a unified diff to me... The text leading up to this was: -------------------------- |$NetBSD: patch-libraries_unix_System_Posix_Files.hsc,v 1.1 2014/05/21 16:07:59 jperkin Exp $ | |Add more errnos. | |--- libraries/unix/System/Posix/Files.hsc.orig 2013-04-18 21:36:40.000000000 +0000 |+++ libraries/unix/System/Posix/Files.hsc -------------------------- Patching file libraries/unix/System/Posix/Files.hsc using Plan A... Hunk #1 succeeded at 171. done => Verifying /data/pkgsrc/lang/ghc7/patches/patch-libraries_unix_System_Posix_Signals.hsc => Applying pkgsrc patch /data/pkgsrc/lang/ghc7/patches/patch-libraries_unix_System_Posix_Signals.hsc Hmm... Looks like a unified diff to me... The text leading up to this was: -------------------------- |$NetBSD: patch-libraries_unix_System_Posix_Signals.hsc,v 1.1 2015/02/13 19:28:40 pho Exp $ | |Suppress linker warnings about compatibility syscall wrappers by using |"capi" instead of "ccall". In Haskell FFI, "ccall" is actually an |interface to C ABI rather than C API. That is, GHC generates direct |references to the symbol even if it's actually defined as a cpp macro |or something like that, because GHC knows nothing about those macros |in foreign headers. Hence the following warnings: | | .../libHSunix-2.6.0.1.a(Signals.o): In function `s4SG_info': | (.text+0x3a3e): warning: warning: reference to compatibility | sigsuspend(); include for correct reference | .../libHSunix-2.6.0.1.a(Signals.o): In function `s5uV_info': | (.text+0x6adb): warning: warning: reference to compatibility | sigpending(); include for correct reference | |In other words, you can safely use "ccall" only when you are sure the |symbol you want to import is actually a symbol in the ABI sense, which |is not always the case for the POSIX API. | |--- libraries/unix/System/Posix/Signals.hsc.orig 2015-02-13 15:40:27.000000000 +0000 |+++ libraries/unix/System/Posix/Signals.hsc -------------------------- Patching file libraries/unix/System/Posix/Signals.hsc using Plan A... Hunk #1 succeeded at 598. Hunk #2 succeeded at 622. done => Verifying /data/pkgsrc/lang/ghc7/patches/patch-libraries_unix_cbits_execvpe.c => Applying pkgsrc patch /data/pkgsrc/lang/ghc7/patches/patch-libraries_unix_cbits_execvpe.c Hmm... Looks like a unified diff to me... The text leading up to this was: -------------------------- |$NetBSD: patch-libraries_unix_cbits_execvpe.c,v 1.1 2015/10/23 20:46:22 tnn Exp $ | |Don't conflict with builtin execvpe(3). | |--- libraries/unix/cbits/execvpe.c.orig 2013-04-18 21:36:40.000000000 +0000 |+++ libraries/unix/cbits/execvpe.c -------------------------- Patching file libraries/unix/cbits/execvpe.c using Plan A... Hunk #1 succeeded at 57. done => Verifying /data/pkgsrc/lang/ghc7/patches/patch-libraries_unix_include_execvpe.h => Applying pkgsrc patch /data/pkgsrc/lang/ghc7/patches/patch-libraries_unix_include_execvpe.h Hmm... Looks like a unified diff to me... The text leading up to this was: -------------------------- |$NetBSD: patch-libraries_unix_include_execvpe.h,v 1.1 2015/10/23 20:46:22 tnn Exp $ | |Don't conflict with builtin execvpe(3). | |--- libraries/unix/include/execvpe.h.orig 2013-04-18 21:36:40.000000000 +0000 |+++ libraries/unix/include/execvpe.h -------------------------- Patching file libraries/unix/include/execvpe.h using Plan A... Hunk #1 succeeded at 19. done => Verifying /data/pkgsrc/lang/ghc7/patches/patch-mk_config.mk.in => Applying pkgsrc patch /data/pkgsrc/lang/ghc7/patches/patch-mk_config.mk.in Hmm... Looks like a unified diff to me... The text leading up to this was: -------------------------- |$NetBSD: patch-mk_config.mk.in,v 1.1 2014/05/21 16:07:59 jperkin Exp $ | |Support shared libraries on 64-bit SunOS. | |--- mk/config.mk.in.orig 2013-04-18 21:22:46.000000000 +0000 |+++ mk/config.mk.in -------------------------- Patching file mk/config.mk.in using Plan A... Hunk #1 succeeded at 124. done => Verifying /data/pkgsrc/lang/ghc7/patches/patch-rts_Linker.c => Applying pkgsrc patch /data/pkgsrc/lang/ghc7/patches/patch-rts_Linker.c Hmm... Looks like a unified diff to me... The text leading up to this was: -------------------------- |$NetBSD: patch-rts_Linker.c,v 1.2 2014/05/21 16:07:59 jperkin Exp $ | |Support 64-bit x86 SunOS. | |--- rts/Linker.c.orig 2013-04-18 21:22:46.000000000 +0000 |+++ rts/Linker.c -------------------------- Patching file rts/Linker.c using Plan A... Hunk #1 succeeded at 75. Hunk #2 succeeded at 4015. done => Verifying /data/pkgsrc/lang/ghc7/patches/patch-rts_PosixSource.h => Applying pkgsrc patch /data/pkgsrc/lang/ghc7/patches/patch-rts_PosixSource.h Hmm... Looks like a unified diff to me... The text leading up to this was: -------------------------- |$NetBSD: patch-rts_PosixSource.h,v 1.1 2018/05/25 15:24:02 jperkin Exp $ | |Set _XOPEN_SOURCE correctly on SunOS. | |--- rts/PosixSource.h.orig 2013-04-18 21:22:46.000000000 +0000 |+++ rts/PosixSource.h -------------------------- Patching file rts/PosixSource.h using Plan A... Hunk #1 succeeded at 11. done => Verifying /data/pkgsrc/lang/ghc7/patches/patch-rts_StgCRun.c => Applying pkgsrc patch /data/pkgsrc/lang/ghc7/patches/patch-rts_StgCRun.c Hmm... Looks like a unified diff to me... The text leading up to this was: -------------------------- |$NetBSD: patch-rts_StgCRun.c,v 1.2 2015/02/05 00:59:10 pho Exp $ | |We have to make the visiblity of StgRun and StgReturn to "hidden" not |to trigger a bug in old binutils which results in: | | ld: rts/dist/build/RtsStartup.dyn_o: relocation R_X86_64_PC32 | against `StgRun' can not be used when making a shared object; | recompile with -fPIC | |With this workaround we can no longer override those two symbols using |LD_PRELOAD, but I think there is no reason to do so anyway. | |This has already been fixed in the upstream: |https://ghc.haskell.org/trac/ghc/ticket/7813 | |--- rts/StgCRun.c.orig 2013-01-27 13:21:52.000000000 +0000 |+++ rts/StgCRun.c -------------------------- Patching file rts/StgCRun.c using Plan A... Hunk #1 succeeded at 246. Hunk #2 succeeded at 281. done => Verifying /data/pkgsrc/lang/ghc7/patches/patch-rts_ghc.mk => Applying pkgsrc patch /data/pkgsrc/lang/ghc7/patches/patch-rts_ghc.mk Hmm... Looks like a unified diff to me... The text leading up to this was: -------------------------- |$NetBSD: patch-rts_ghc.mk,v 1.6 2016/05/07 10:02:06 joerg Exp $ | |This is pkgsrc specific: | |Install libffi.{a,la} too, so that the plist framework can deal with |.so name being different between platforms. Note that |"libffi$(soext)*" matches to libffi.{so,so.6,so.6.0} but not to |libffi.6.dylib so the pattern must be corrected to "libffi*$(soext)*", |otherwise libtool-expand will emit the following error on Mach-O |platforms: | | libtool-expand: `lib/ghc-7.6.3/libffi.6.dylib' was not found | |For what it's worth, this ain't the right way to install libtool |libraries. They should always be installed by running |"libtool --mode=install install -c foo.la /path/to/destination", which |can't be achieved without a major refactoring of the GHC build system. |And for that matter, GHC shouldn't install a local copy of libffi in |the first place. In fact GHC 7.8 seems to have a configure option |"--with-system-libffi" so we can hopefully remove this rather |unpleasant patch in the future. | |--- rts/ghc.mk.orig 2013-04-18 21:22:47.000000000 +0000 |+++ rts/ghc.mk -------------------------- Patching file rts/ghc.mk using Plan A... Hunk #1 succeeded at 112. Hunk #2 succeeded at 186. Hunk #3 succeeded at 510. done ===> Regenerating configuration scripts for ghc-7.6.3nb12 ===> Creating toolchain wrappers for ghc-7.6.3nb12 ===> Configuring for ghc-7.6.3nb12 ===> Extracting bootstrapping compiler for ghc-7.6.3nb12 ===> Preparing bootstrapping compiler for ghc-7.6.3nb12 checking for path to top of build tree... Shared object "libgcc_s.so.1" not found configure: error: cannot determine current directory *** Error code 1 Stop. make[1]: stopped in /data/pkgsrc/lang/ghc7 *** Error code 1 Stop. make: stopped in /data/pkgsrc/lang/ghc7