[Fink-users] cernlib and El Capitain

[Giuliano Franchetti]

Hi to all, 

I have installed El Capitatin, and recompiling everything. 
I experience an error in the compilation of the cernlib. 

Below is what appears on terminal. 

Thanks for the help. 


Giuliano 


===

rm kuipmcdf.c
rm -f archive/iconbox.o
gcc-5 -ftree-vectorize -c -O0 -g -pipe
-I/sw2/src/fink.build/cernlib2006-2006b-25/2006/src/packlib/kuip 
-I/sw2/src/fink.build/cernlib2006-2006b-25/2006/src/packlib/kuip 
-I/sw2/src/fink.build/cernlib2006-2006b-25/2006/src/packlib/kuip/code_motif 
-I/sw2/include/freetype2 -I/sw2/include -I/usr/X11R6/include 
-I/sw2/src/fink.build/cernlib2006-2006b-25/2006/src/include  -DCERNLIB_LINUX 
-DCERNLIB_MACOSX  -Dunix=unix -D__DARWIN__  -DFUNCPROTO=7 
-DNARROWPROTO-DCERNLIB_LINUX -DCERNLIB_UNIX -DCERNLIB_LNX   
-DCERNLIB_QMLXIA64 -DCERNLIB_QMGLIBC
-DCERNLIB_MACOSX -DCERNLIB_GFORTRAN iconbox.c -o archive/iconbox.o
In file included from iconbox.c:66:0:
icboxp.h:64:28: fatal error: X11/StringDefs.h: No such file or directory
compilation terminated.
Makefile:207: recipe for target 'archive/iconbox.o' failed
make[3]: *** [archive/iconbox.o] Error 1
make[3]: Leaving directory 
'/sw2/src/fink.build/cernlib2006-2006b-25/2006/src/packlib/kuip/code_motif'
Makefile:422: recipe for target 'code_motif/archive/objects.list' failed
make[2]: *** [code_motif/archive/objects.list] Error 2
make[2]: Leaving directory 
'/sw2/src/fink.build/cernlib2006-2006b-25/2006/src/packlib/kuip'
Makefile:421: recipe for target 'kuip/archive/objects.list' failed
make[1]: *** [kuip/archive/objects.list] Error 2
make[1]: Leaving directory 
'/sw2/src/fink.build/cernlib2006-2006b-25/2006/src/packlib'
Makefile:327: recipe for target 'install.lib' failed
make: *** [install.lib] Error 2
### execution of /tmp/fink.Z19Iw failed, exit code 2
### execution of /tmp/fink.UUXK1 failed, exit code 2
Removing runtime build-lock...
Removing build-lock package...
/sw2/bin/dpkg-lockwait -r fink-buildlock-cernlib2006-2006b-25
(Reading database ... 190300 files and directories currently installed.)
Removing fink-buildlock-cernlib2006-2006b-25 ...
Failed: phase compiling: cernlib2006-2006b-25 failed


Before reporting any errors, please run "fink selfupdate" and try again.

If you continue to have issues, please check to see if the FAQ on Fink's 
website solves the problem.  If not, ask on one (not both, please) of
these mailing lists:

The Fink Users List 
The Fink Beginners List ,

with a carbon copy to the maintainer:

Remi Mommsen 

Note that this is preferable to emailing just the maintainer directly,
since most fink package maintainers do not have access to all possible
hardware and software configurations.

Please try to include the complete error message in your report.  This
generally consists of a compiler line starting with e.g. "gcc" or "g++"
followed by the actual error output from the compiler.

Also include the following system information:
Package manager version: 0.39.1
Distribution version: selfupdate-rsync Tue Oct  6 06:53:27 2015, 10.11, x86_64
Trees: local/main stable/main
Xcode.app: 7.0.1
Xcode command-line tools: 7.0.0.0.1.1441394355
cernlib2006-2006b-25 is set to build with only one job.


---
Giuliano Franchetti 

Primary Beams / Beam Physik

Office: C27 3.018
phone:   +49 6159 71 1535
fax: +49 6159 71 3099
e-mail:g.franche...@gsi.de
http://web-docs.gsi.de/~giuliano

--   
GSI Helmholtzzentrum für Schwerionenforschung GmbH 
Planckstraße 1 
64291 Darmstadt 
www.gsi.de 

Gesellschaft mit beschränkter Haftung 
Sitz der Gesellschaft: Darmstadt 
Handelsregister: Amtsgericht Darmstadt, HRB 1528 

Geschäftsführung: 
Ursula Weyrich 
Professor Dr. Karlheinz Langanke 

Vorsitzender des Aufsichtsrates: St Dr. Georg Schütte 
Stellvertreter: Ministerialdirigent Dr. Rolf Bernhardt 

--
___
Fink-users mailing list
Fink-users@lists.sourceforge.net
List archive:
http://news.gmane.org/gmane.os.macosx.fink.user
Subscription management:
https://lists.sourceforge.net/lists/listinfo/fink-users

Re: [Fink-users] update-all error: ghc depends on clang36

[Alexander Hansen]

> On Oct 5, 2015, at 12:55, David Phillips  wrote:
> 
> 
> 
> Alas, the simple change to ghc to build with clang3.7 and llvm3.7 didn’t 
> work. Here’s the end of the build with the error message and the immediately 
> previous make. 
>  I need to run off to a meeting and so I will not get to the next step of 
> building with a single core instead of all of them to get a sensible error 
> message until this evening. Here’s the end of the build, though.
>  David
> 
> 
> "inplace/bin/ghc-stage1" -optc-m64 -optc-fno-stack-protector -optc-Wall 
> -optc-Wextra -optc-Wstrict-prototypes -optc-Wmissing-prototypes 
> -optc-Wmissing-declarations -optc-Winline -optc-Waggregate-return 
> -optc-Wpointer-arith -optc-Wmissing-noreturn -optc-Wnested-externs 
> -optc-Wredundant-decls -optc-Iincludes -optc-Iincludes/dist 
> -optc-Iincludes/dist-derivedconstants/header 
> -optc-Iincludes/dist-ghcconstants/header -optc-Irts -optc-Irts/dist/build 
> -optc-DCOMPILING_RTS -optc-fno-strict-aliasing -optc-fno-common -optc-DDTRACE 
> -optc-O2 -optc-fomit-frame-pointer -optc-DRtsWay=\"rts_p\" 
> -optc-funroll-loops -static -prof  -H32m -O -Iincludes -Iincludes/dist 
> -Iincludes/dist-derivedconstants/header -Iincludes/dist-ghcconstants/header 
> -Irts -Irts/dist/build -DCOMPILING_RTS -package-name rts -dcmm-lint  -DDTRACE 
> -i -irts -irts/dist/build -irts/dist/build/autogen -Irts/dist/build 
> -Irts/dist/build/autogen   -O2-c rts/sm/Evac.c -o 
> rts/dist/build/sm/Evac.p_o
> make[1]: *** read jobs pipe: No such file or directory.  Stop.
> make[1]: *** Waiting for unfinished jobs
> Makefile:64: recipe for target 'all' failed
> make: *** [all] Broken pipe: 13
> make: INTERNAL: Exiting with 1 jobserver tokens available; should be 12!
> ### execution of /tmp/fink.Q3xzN failed, exit code 2
> ### execution of /tmp/fink.hmAuW failed, exit code 2
> Removing runtime build-lock...
> Removing build-lock package...
> /sw/bin/dpkg-lockwait -r fink-buildlock-ghc-7.8.3-102.1
> (Reading database ... 328501 files and directories currently installed.)
> Removing fink-buildlock-ghc-7.8.3-102.1 ...
> Failed: phase compiling: ghc-7.8.3-102.1 failed
> 
> Before reporting any errors, please run "fink selfupdate" and try again.
> Also try using "fink configure" to set your maximum build jobs to 1 and
> attempt to build the package again.
> If you continue to have issues, please check to see if the FAQ on Fink's
> website solves the problem.  If not, ask on one (not both, please) of
> these mailing lists:
> 
>   The Fink Users List 
>   The Fink Beginners List .
> 
> Please try to include the complete error message in your report.  This
> generally consists of a compiler line starting with e.g. "gcc" or "g++"
> followed by the actual error output from the compiler.
> 
> Also include the following system information:
> Package manager version: 0.39.1
> Distribution version: selfupdate-rsync Mon Oct  5 15:43:09 2015, 10.11, x86_64
> Trees: local/main stable/main stable/crypto local/injected
> Xcode.app: 7.0.1
> Xcode command-line tools: 7.0.0.0.1.1441394355
> Max. Fink build jobs:  12
> 

I’m still in the middle of building.  Since I’ve got fewer cores that often 
results in success.  

The error message does indeed suggest that there’s a race condition in the 
build somewhere.




-- 
Alexander Hansen, Ph.D.
Fink User Liaison


--
___
Fink-users mailing list
Fink-users@lists.sourceforge.net
List archive:
http://news.gmane.org/gmane.os.macosx.fink.user
Subscription management:
https://lists.sourceforge.net/lists/listinfo/fink-users

Re: [Fink-users] Missing debs after update to 10.9 tree

[Viv Kendon]

On 5 Oct 2015, at 20:57, Alexander Hansen  wrote:
> 
>> On Oct 5, 2015, at 12:39, Viv Kendon  wrote:
>> 
>> On 5 Oct 2015, at 15:36, Alexander Hansen  
>> wrote:
>> 
 On Oct 4, 2015, at 16:46, Hanspeter Niederstrasser 
  wrote:
 
 On 10/4/2015 5:02 PM, Alexander Hansen wrote:
> 
>> On Oct 4, 2015, at 14:22, Viv Kendon  wrote:
>> 
>> continuing with “update-all” I got:
>> 
>> The following packages must be temporarily removed, but there are no 
>> .debs to
>> restore them from:
>> libev4 libev4
>> 
>> /sw/fink-64# ls 10.9/stable/main/binary-darwin-x86_64/libs/*ev4*
>> ls: 10.9/stable/main/binary-darwin-x86_64/libs/*ev4*: No such file or 
>> directory
>> 
>> And indeed, the debs are still in my 10.7 tree.
>> 
>> /sw/fink-63# ls 10.7/stable/main/binary-darwin-x86_64/libs/*ev4*
>> 72 
>> 10.7/stable/main/binary-darwin-x86_64/libs/libev4-shlibs_4.15-1_darwin-x86_64.deb
>> 72 
>> 10.7/stable/main/binary-darwin-x86_64/libs/libev4-shlibs_4.19-1_darwin-x86_64.deb
>> 208 
>> 10.7/stable/main/binary-darwin-x86_64/libs/libev4_4.15-1_darwin-x86_64.deb
>> 216 
>> 10.7/stable/main/binary-darwin-x86_64/libs/libev4_4.19-1_darwin-x86_64.deb
>> 
>> Should I manually move or copy them over to match the dangling symlinks? 
>> Is there a way to get any other missing debs fixed up, other than 
>> finding out like this, when they are actually needed?
>> 
>> 10.9.5, case-sensitive, fink -V
>> Package manager version: 0.39.1
>> 
>> gcc -v
>> Configured with: --prefix=/Applications/Xcode.app/Contents/Developer/usr 
>> --with-gxx-include-dir=/usr/include/c++/4.2.1
>> Apple LLVM version 6.0 (clang-600.0.57) (based on LLVM 3.5svn)
>> Target: x86_64-apple-darwin13.4.0
>> 
>> Got to this version of fink earlier today by doing fink selfupdate twice 
>> (first one gave me the warning about old symlink).  Unfortunately I 
>> don’t have the screen output from doing that to see what might have gone 
>> wrong.
>> 
>> many thanks,
>> — Viv
> 
> Hmm, the fink-0.39.x update was supposed to have taken care of that.  
> fink doesn’t try again if it detects a non-dangling symlink.
> 
> fink searches for .deb files in the 10.7 directory and just copies them 
> over to the 10.9-libcxx directory (including the full paths, of course).  
> I don’t happen to have a shell script handy to do that, though.
 
 I also had the failure of my debs being copied over (on 10.9).
 
 Of note, I had 10.9-libcxx in my Trees: line in fink.conf before the 
 update in order to easier update files on both trees.  Don't know if 
 that's relevant and the transmutation of multiple of symlinks (I had 
 10.9.old generated) confused the moving script.
 
 Hanspeter
 
>>> 
>>> A potential workaround would be to edit /sw/lib/fink/postinstall.pl as root 
>>> and change line 273 to 
>>> 
>>> if (1) {
>>> 
>>> Then run postinstall.pl .  After that run “fink reinstall fink” to remove 
>>> the manual modification so that fink doesn’t keep trying to copy .debs 
>>> every time you update it.
>>> 
>>> -- 
>>> Alexander Hansen, Ph.D.
>>> Fink User Liaison
>> 
>> Cool, I gave that a try and after about 8GB of copying I seem to have a full 
>> complement of .debs in 10.9-libcxx, to judge by the sizes of the dirs.
>> 
>> ls 10.9/stable/main/binary-darwin-x86_64/libs/*ev
>> 72 
>> /sw/fink/10.9/stable/main/binary-darwin-x86_64/libs/libev4-shlibs_4.15-1_darwin-x86_64.deb
>> 72 
>> /sw/fink/10.9/stable/main/binary-darwin-x86_64/libs/libev4-shlibs_4.19-1_darwin-x86_64.deb
>> 208 
>> /sw/fink/10.9/stable/main/binary-darwin-x86_64/libs/libev4_4.15-1_darwin-x86_64.deb
>> 216 
>> /sw/fink/10.9/stable/main/binary-darwin-x86_64/libs/libev4_4.19-1_darwin-x86_64.deb
>> 
>> looks much more promising.  And update-all is now proceeding.  
>> 
>> I notice it has copied over my “local” .debs, but not the corresponding 
>> finkinfo files.  That’s easy to fix of course.
>> 
>> Many thanks for the fix, much appreciated.
>> — Viv
> 
> Hmm.  Maybe we’ll want to exclude the local .debs.

If they are installed, I’d have said copy them over, else you get orphaned 
installs:

 fink list unison
Information about 9258 packages read in 2 seconds.
*i*  unison   2.48.3-1 File synchronization tool
*i*  unison-nox   2.48.3-1 File synchronization tool
*i*  unison2402.40.102-1   File synchronization tool
*i*  unison2452.45.28-1File synchronization tool
*i*  unison2483   2.48.3-1 File synchronization tool

(that was messing around when ocaml updated and needing a linux compatible 
version).  OTOH, anyone with stuff in local probably knows how to manually copy 
what they need.

Also, I think I’d have 

Re: [Fink-users] Octave-qtmac fails to build on 10.11 due to broken compiler

[Alexander Hansen]

> On Oct 5, 2015, at 13:20, Martin Costabel  wrote:
> 
> On 5/10/15 21:53, Gary K. Olson wrote:
>> Alex Hansen’s fix over weekend fixed the system-perl problem, so I
>> decided to build a few active packages that I hadn’t built under 10.10.
>> Several packages couldn’t build because of broken c compiler.  Here is
>> message on attempting to build octave-qtmac that was typical:
>> 
>>> 
>> checking for style of include used by make... GNU
>> checking for x86_64-apple-darwin-gcc... oct-cc
>> checking whether the C compiler works... no
>> configure: error: in `/sw/src/fink.build/octave-qtmac-3.8.2-6/octave-3.8.2':
>> configure: error: C compiler cannot create executables
>> See `config.log' for more details
> []
>> Failed: phase compiling: octave-qtmac-3.8.2-6 failed
>> 
>> I assumed that this is due to still calling for gcc4.9?
> 
> No, it is due to fink-octave-scripts not yet aware of the existence of El 
> Capitan. You can paper it over as follows:
> 
> Edit the two files /sw/bin/oct-cc and /sw/bin/oct-cxx and add 15 to the list 
> of recognized OS versions. Here are the diffs:
> 
> --- /sw/bin/oct-cc~   2015-10-02 21:03:28.0 +0200
> +++ /sw/bin/oct-cc2015-10-04 11:46:27.0 +0200
> @@ -3,3 +3,3 @@
> case $osversion in
> - 1[1234])
> + 1[12345])
>   COMPILER=/sw/var/lib/fink/path-prefix-clang/cc ;;
> --- /sw/bin/oct-cxx~  2015-10-02 21:03:28.0 +0200
> +++ /sw/bin/oct-cxx   2015-10-04 11:48:08.0 +0200
> @@ -3,3 +3,3 @@
> case $osversion in
> - 1[34])
> + 1[345])
>   COMPILER=/sw/var/lib/fink/path-prefix-libcxx/c++ ;;
> 
> With this fix, octave382 should build (it did for me, anyway).
> 
> -- 
> Martin
> 
> 
> 

I’ll get that updated—I figured it was safest to have builds error out promptly 
for unrelated OSes rather than to get partway through the long Octave build and 
then fail.

-- 
Alexander Hansen, Ph.D.
Fink User Liaison


--
___
Fink-users mailing list
Fink-users@lists.sourceforge.net
List archive:
http://news.gmane.org/gmane.os.macosx.fink.user
Subscription management:
https://lists.sourceforge.net/lists/listinfo/fink-users

Re: [Fink-users] Missing debs after update to 10.9 tree

[Viv Kendon]

On 5 Oct 2015, at 15:36, Alexander Hansen  wrote:

>> On Oct 4, 2015, at 16:46, Hanspeter Niederstrasser  
>> wrote:
>> 
>> On 10/4/2015 5:02 PM, Alexander Hansen wrote:
>>> 
 On Oct 4, 2015, at 14:22, Viv Kendon  wrote:
 
 continuing with “update-all” I got:
 
 The following packages must be temporarily removed, but there are no .debs 
 to
 restore them from:
  libev4 libev4
 
 /sw/fink-64# ls 10.9/stable/main/binary-darwin-x86_64/libs/*ev4*
 ls: 10.9/stable/main/binary-darwin-x86_64/libs/*ev4*: No such file or 
 directory
 
 And indeed, the debs are still in my 10.7 tree.
 
 /sw/fink-63# ls 10.7/stable/main/binary-darwin-x86_64/libs/*ev4*
 72 
 10.7/stable/main/binary-darwin-x86_64/libs/libev4-shlibs_4.15-1_darwin-x86_64.deb
 72 
 10.7/stable/main/binary-darwin-x86_64/libs/libev4-shlibs_4.19-1_darwin-x86_64.deb
 208 
 10.7/stable/main/binary-darwin-x86_64/libs/libev4_4.15-1_darwin-x86_64.deb
 216 
 10.7/stable/main/binary-darwin-x86_64/libs/libev4_4.19-1_darwin-x86_64.deb
 
 Should I manually move or copy them over to match the dangling symlinks? 
 Is there a way to get any other missing debs fixed up, other than finding 
 out like this, when they are actually needed?
 
 10.9.5, case-sensitive, fink -V
 Package manager version: 0.39.1
 
 gcc -v
 Configured with: --prefix=/Applications/Xcode.app/Contents/Developer/usr 
 --with-gxx-include-dir=/usr/include/c++/4.2.1
 Apple LLVM version 6.0 (clang-600.0.57) (based on LLVM 3.5svn)
 Target: x86_64-apple-darwin13.4.0
 
 Got to this version of fink earlier today by doing fink selfupdate twice 
 (first one gave me the warning about old symlink).  Unfortunately I don’t 
 have the screen output from doing that to see what might have gone wrong.
 
 many thanks,
 — Viv
>>> 
>>> Hmm, the fink-0.39.x update was supposed to have taken care of that.  fink 
>>> doesn’t try again if it detects a non-dangling symlink.
>>> 
>>>  fink searches for .deb files in the 10.7 directory and just copies them 
>>> over to the 10.9-libcxx directory (including the full paths, of course).  I 
>>> don’t happen to have a shell script handy to do that, though.
>> 
>> I also had the failure of my debs being copied over (on 10.9).
>> 
>> Of note, I had 10.9-libcxx in my Trees: line in fink.conf before the 
>> update in order to easier update files on both trees.  Don't know if 
>> that's relevant and the transmutation of multiple of symlinks (I had 
>> 10.9.old generated) confused the moving script.
>> 
>> Hanspeter
>> 
> 
> A potential workaround would be to edit /sw/lib/fink/postinstall.pl as root 
> and change line 273 to 
> 
> if (1) {
> 
> Then run postinstall.pl .  After that run “fink reinstall fink” to remove the 
> manual modification so that fink doesn’t keep trying to copy .debs every time 
> you update it.
> 
> -- 
> Alexander Hansen, Ph.D.
> Fink User Liaison

Cool, I gave that a try and after about 8GB of copying I seem to have a full 
complement of .debs in 10.9-libcxx, to judge by the sizes of the dirs.

 ls 10.9/stable/main/binary-darwin-x86_64/libs/*ev
 72 
/sw/fink/10.9/stable/main/binary-darwin-x86_64/libs/libev4-shlibs_4.15-1_darwin-x86_64.deb
 72 
/sw/fink/10.9/stable/main/binary-darwin-x86_64/libs/libev4-shlibs_4.19-1_darwin-x86_64.deb
208 
/sw/fink/10.9/stable/main/binary-darwin-x86_64/libs/libev4_4.15-1_darwin-x86_64.deb
216 
/sw/fink/10.9/stable/main/binary-darwin-x86_64/libs/libev4_4.19-1_darwin-x86_64.deb

looks much more promising.  And update-all is now proceeding.  

I notice it has copied over my “local” .debs, but not the corresponding 
finkinfo files.  That’s easy to fix of course.

Many thanks for the fix, much appreciated.
— Viv
--
___
Fink-users mailing list
Fink-users@lists.sourceforge.net
List archive:
http://news.gmane.org/gmane.os.macosx.fink.user
Subscription management:
https://lists.sourceforge.net/lists/listinfo/fink-users

Re: [Fink-users] update-all error: ghc depends on clang36

[Hanspeter Niederstrasser]

On Mon, October 5, 2015 2:55 pm, David Phillips wrote:
>
>
> Alas, the simple change to ghc to build with clang3.7 and llvm3.7 didn’t
> work. Here’s the end of the build with the error message and the
> immediately previous make.
>   I need to run off to a meeting and so I will not get to the next step of
> building with a single core instead of all of them to get a sensible
> error message until this evening. Here’s the end of the build, though.
>   David
>
>
> "inplace/bin/ghc-stage1" -optc-m64 -optc-fno-stack-protector -optc-Wall
> -optc-Wextra -optc-Wstrict-prototypes -optc-Wmissing-prototypes
> -optc-Wmissing-declarations -optc-Winline -optc-Waggregate-return
> -optc-Wpointer-arith -optc-Wmissing-noreturn -optc-Wnested-externs
> -optc-Wredundant-decls -optc-Iincludes -optc-Iincludes/dist
> -optc-Iincludes/dist-derivedconstants/header
> -optc-Iincludes/dist-ghcconstants/header -optc-Irts -optc-Irts/dist/build
> -optc-DCOMPILING_RTS -optc-fno-strict-aliasing -optc-fno-common
> -optc-DDTRACE -optc-O2 -optc-fomit-frame-pointer -optc-DRtsWay=\"rts_p\"
> -optc-funroll-loops -static -prof  -H32m -O -Iincludes -Iincludes/dist
> -Iincludes/dist-derivedconstants/header
> -Iincludes/dist-ghcconstants/header -Irts -Irts/dist/build -DCOMPILING_RTS
> -package-name rts -dcmm-lint  -DDTRACE -i -irts -irts/dist/build
> -irts/dist/build/autogen -Irts/dist/build -Irts/dist/build/autogen
>   -O2-c rts/sm/Evac.c -o rts/dist/build/sm/Evac.p_o
> make[1]: *** read jobs pipe: No such file or directory.  Stop.
> make[1]: *** Waiting for unfinished jobs
> Makefile:64: recipe for target 'all' failed
> make: *** [all] Broken pipe: 13
> make: INTERNAL: Exiting with 1 jobserver tokens available; should be 12!

This looks like the usual failure when Fink's GNU make doesn't like
something.  Try uninstalling the package 'make' and rebuilding ghc.

Hanspeter

-- 
More agile than a turtle, stronger than a mouse, nobler than a lettuce


--
___
Fink-users mailing list
Fink-users@lists.sourceforge.net
List archive:
http://news.gmane.org/gmane.os.macosx.fink.user
Subscription management:
https://lists.sourceforge.net/lists/listinfo/fink-users

Re: [Fink-users] Missing debs after update to 10.9 tree

[Alexander Hansen]

> On Oct 5, 2015, at 12:39, Viv Kendon  wrote:
> 
> On 5 Oct 2015, at 15:36, Alexander Hansen  wrote:
> 
>>> On Oct 4, 2015, at 16:46, Hanspeter Niederstrasser  
>>> wrote:
>>> 
>>> On 10/4/2015 5:02 PM, Alexander Hansen wrote:
 
> On Oct 4, 2015, at 14:22, Viv Kendon  wrote:
> 
> continuing with “update-all” I got:
> 
> The following packages must be temporarily removed, but there are no 
> .debs to
> restore them from:
> libev4 libev4
> 
> /sw/fink-64# ls 10.9/stable/main/binary-darwin-x86_64/libs/*ev4*
> ls: 10.9/stable/main/binary-darwin-x86_64/libs/*ev4*: No such file or 
> directory
> 
> And indeed, the debs are still in my 10.7 tree.
> 
> /sw/fink-63# ls 10.7/stable/main/binary-darwin-x86_64/libs/*ev4*
> 72 
> 10.7/stable/main/binary-darwin-x86_64/libs/libev4-shlibs_4.15-1_darwin-x86_64.deb
> 72 
> 10.7/stable/main/binary-darwin-x86_64/libs/libev4-shlibs_4.19-1_darwin-x86_64.deb
> 208 
> 10.7/stable/main/binary-darwin-x86_64/libs/libev4_4.15-1_darwin-x86_64.deb
> 216 
> 10.7/stable/main/binary-darwin-x86_64/libs/libev4_4.19-1_darwin-x86_64.deb
> 
> Should I manually move or copy them over to match the dangling symlinks? 
> Is there a way to get any other missing debs fixed up, other than finding 
> out like this, when they are actually needed?
> 
> 10.9.5, case-sensitive, fink -V
> Package manager version: 0.39.1
> 
> gcc -v
> Configured with: --prefix=/Applications/Xcode.app/Contents/Developer/usr 
> --with-gxx-include-dir=/usr/include/c++/4.2.1
> Apple LLVM version 6.0 (clang-600.0.57) (based on LLVM 3.5svn)
> Target: x86_64-apple-darwin13.4.0
> 
> Got to this version of fink earlier today by doing fink selfupdate twice 
> (first one gave me the warning about old symlink).  Unfortunately I don’t 
> have the screen output from doing that to see what might have gone wrong.
> 
> many thanks,
> — Viv
 
 Hmm, the fink-0.39.x update was supposed to have taken care of that.  fink 
 doesn’t try again if it detects a non-dangling symlink.
 
 fink searches for .deb files in the 10.7 directory and just copies them 
 over to the 10.9-libcxx directory (including the full paths, of course).  
 I don’t happen to have a shell script handy to do that, though.
>>> 
>>> I also had the failure of my debs being copied over (on 10.9).
>>> 
>>> Of note, I had 10.9-libcxx in my Trees: line in fink.conf before the 
>>> update in order to easier update files on both trees.  Don't know if 
>>> that's relevant and the transmutation of multiple of symlinks (I had 
>>> 10.9.old generated) confused the moving script.
>>> 
>>> Hanspeter
>>> 
>> 
>> A potential workaround would be to edit /sw/lib/fink/postinstall.pl as root 
>> and change line 273 to 
>> 
>> if (1) {
>> 
>> Then run postinstall.pl .  After that run “fink reinstall fink” to remove 
>> the manual modification so that fink doesn’t keep trying to copy .debs every 
>> time you update it.
>> 
>> -- 
>> Alexander Hansen, Ph.D.
>> Fink User Liaison
> 
> Cool, I gave that a try and after about 8GB of copying I seem to have a full 
> complement of .debs in 10.9-libcxx, to judge by the sizes of the dirs.
> 
> ls 10.9/stable/main/binary-darwin-x86_64/libs/*ev
> 72 
> /sw/fink/10.9/stable/main/binary-darwin-x86_64/libs/libev4-shlibs_4.15-1_darwin-x86_64.deb
> 72 
> /sw/fink/10.9/stable/main/binary-darwin-x86_64/libs/libev4-shlibs_4.19-1_darwin-x86_64.deb
> 208 
> /sw/fink/10.9/stable/main/binary-darwin-x86_64/libs/libev4_4.15-1_darwin-x86_64.deb
> 216 
> /sw/fink/10.9/stable/main/binary-darwin-x86_64/libs/libev4_4.19-1_darwin-x86_64.deb
> 
> looks much more promising.  And update-all is now proceeding.  
> 
> I notice it has copied over my “local” .debs, but not the corresponding 
> finkinfo files.  That’s easy to fix of course.
> 
> Many thanks for the fix, much appreciated.
> — Viv

Hmm.  Maybe we’ll want to exclude the local .debs.


--
___
Fink-users mailing list
Fink-users@lists.sourceforge.net
List archive:
http://news.gmane.org/gmane.os.macosx.fink.user
Subscription management:
https://lists.sourceforge.net/lists/listinfo/fink-users

Re: [Fink-users] Missing debs after update to 10.9 tree

[Viv Kendon]

On 5 Oct 2015, at 21:22, Viv Kendon  wrote:

> On 5 Oct 2015, at 20:57, Alexander Hansen  wrote:
>> 
>>> On Oct 5, 2015, at 12:39, Viv Kendon  wrote:
>>> 
>>> On 5 Oct 2015, at 15:36, Alexander Hansen  
>>> wrote:
>>> 
> On Oct 4, 2015, at 16:46, Hanspeter Niederstrasser 
>  wrote:
> 
> On 10/4/2015 5:02 PM, Alexander Hansen wrote:
>> 
>>> On Oct 4, 2015, at 14:22, Viv Kendon  wrote:
>>> 
>>> continuing with “update-all” I got:
>>> 
>>> The following packages must be temporarily removed, but there are no 
>>> .debs to
>>> restore them from:
>>> libev4 libev4
>>> 
>>> /sw/fink-64# ls 10.9/stable/main/binary-darwin-x86_64/libs/*ev4*
>>> ls: 10.9/stable/main/binary-darwin-x86_64/libs/*ev4*: No such file or 
>>> directory
>>> 
>>> And indeed, the debs are still in my 10.7 tree.
>>> 
>>> /sw/fink-63# ls 10.7/stable/main/binary-darwin-x86_64/libs/*ev4*
>>> 72 
>>> 10.7/stable/main/binary-darwin-x86_64/libs/libev4-shlibs_4.15-1_darwin-x86_64.deb
>>> 72 
>>> 10.7/stable/main/binary-darwin-x86_64/libs/libev4-shlibs_4.19-1_darwin-x86_64.deb
>>> 208 
>>> 10.7/stable/main/binary-darwin-x86_64/libs/libev4_4.15-1_darwin-x86_64.deb
>>> 216 
>>> 10.7/stable/main/binary-darwin-x86_64/libs/libev4_4.19-1_darwin-x86_64.deb
>>> 
>>> Should I manually move or copy them over to match the dangling 
>>> symlinks? Is there a way to get any other missing debs fixed up, other 
>>> than finding out like this, when they are actually needed?
>>> 
>>> 10.9.5, case-sensitive, fink -V
>>> Package manager version: 0.39.1
>>> 
>>> gcc -v
>>> Configured with: 
>>> --prefix=/Applications/Xcode.app/Contents/Developer/usr 
>>> --with-gxx-include-dir=/usr/include/c++/4.2.1
>>> Apple LLVM version 6.0 (clang-600.0.57) (based on LLVM 3.5svn)
>>> Target: x86_64-apple-darwin13.4.0
>>> 
>>> Got to this version of fink earlier today by doing fink selfupdate 
>>> twice (first one gave me the warning about old symlink).  Unfortunately 
>>> I don’t have the screen output from doing that to see what might have 
>>> gone wrong.
>>> 
>>> many thanks,
>>> — Viv
>> 
>> Hmm, the fink-0.39.x update was supposed to have taken care of that.  
>> fink doesn’t try again if it detects a non-dangling symlink.
>> 
>> fink searches for .deb files in the 10.7 directory and just copies them 
>> over to the 10.9-libcxx directory (including the full paths, of course). 
>>  I don’t happen to have a shell script handy to do that, though.
> 
> I also had the failure of my debs being copied over (on 10.9).
> 
> Of note, I had 10.9-libcxx in my Trees: line in fink.conf before the 
> update in order to easier update files on both trees.  Don't know if 
> that's relevant and the transmutation of multiple of symlinks (I had 
> 10.9.old generated) confused the moving script.
> 
> Hanspeter
> 
 
 A potential workaround would be to edit /sw/lib/fink/postinstall.pl as 
 root and change line 273 to 
 
 if (1) {
 
 Then run postinstall.pl .  After that run “fink reinstall fink” to remove 
 the manual modification so that fink doesn’t keep trying to copy .debs 
 every time you update it.
 
 -- 
 Alexander Hansen, Ph.D.
 Fink User Liaison
>>> 
>>> Cool, I gave that a try and after about 8GB of copying I seem to have a 
>>> full complement of .debs in 10.9-libcxx, to judge by the sizes of the dirs.
>>> 
>>> ls 10.9/stable/main/binary-darwin-x86_64/libs/*ev
>>> 72 
>>> /sw/fink/10.9/stable/main/binary-darwin-x86_64/libs/libev4-shlibs_4.15-1_darwin-x86_64.deb
>>> 72 
>>> /sw/fink/10.9/stable/main/binary-darwin-x86_64/libs/libev4-shlibs_4.19-1_darwin-x86_64.deb
>>> 208 
>>> /sw/fink/10.9/stable/main/binary-darwin-x86_64/libs/libev4_4.15-1_darwin-x86_64.deb
>>> 216 
>>> /sw/fink/10.9/stable/main/binary-darwin-x86_64/libs/libev4_4.19-1_darwin-x86_64.deb
>>> 
>>> looks much more promising.  And update-all is now proceeding.  
>>> 
>>> I notice it has copied over my “local” .debs, but not the corresponding 
>>> finkinfo files.  That’s easy to fix of course.
>>> 
>>> Many thanks for the fix, much appreciated.
>>> — Viv
>> 
>> Hmm.  Maybe we’ll want to exclude the local .debs.
> 
> If they are installed, I’d have said copy them over, else you get orphaned 
> installs:
> 
> fink list unison
> Information about 9258 packages read in 2 seconds.
> *i*  unison   2.48.3-1 File synchronization tool
> *i*  unison-nox   2.48.3-1 File synchronization tool
> *i*  unison2402.40.102-1   File synchronization tool
> *i*  unison2452.45.28-1File synchronization tool
> *i*  unison2483   2.48.3-1 File synchronization 

Re: [Fink-users] Failed: phase compiling: flac-1.3.1-1 failed ...

[Daniel Macks]

On Fri, 25 Sep 2015 23:16:04 -0400, Daniel Macks  wrote:
On Fri, 25 Sep 2015 16:47:33 -0700, Saleh Elmohamed
>  wrote:
> > /bin/sh ../../libtool --tag=CC --mode=link gcc -I/sw/include -O3 > 
> -funroll-loops -Wextra -no-undefined -version-info 11:0:3 -L/sw/lib > 
> -o libFLAC.la -rpath /sw/lib bitmath.lo bitreader.lo bitwriter.lo
[...]
> > Undefined symbols for architecture x86_64:
> > "___cpuid", referenced from:
> > _FLAC__cpu_info in cpu.o
> > _FLAC__cpu_info_x86 in cpu.o
> > "___cpuid_count", referenced from:
> > _FLAC__cpu_info in cpu.o
> > _FLAC__cpu_info_x86 in cpu.o
> > ld: symbol(s) not found for architecture x86_64
>
> I noticed this also earlier this week, only on 10.7 like you (not on 
> 10.10). Best I can tell it's due to a very different codepath taken 
> when bswap16 is not detected. Perhaps compounded by using a header 
> test (cpuid.h) that has a generic name to get functions that are 
> specific to non-darwin and/or only available in later xcode versions 
> than 10.7 has. If I have time I'll try to figure out how to shim the 
> missing function or give the "correct" result for the buggy test. 

I just committed a fix that gets over this hump on 10.7. No idea 
whether it actually makes a library that works well for...whatever you 
plan to do with it. 

dan

--
Daniel Macks
dma...@netspace.org


--
___
Fink-users mailing list
Fink-users@lists.sourceforge.net
List archive:
http://news.gmane.org/gmane.os.macosx.fink.user
Subscription management:
https://lists.sourceforge.net/lists/listinfo/fink-users

Re: [Fink-users] update-all error: ghc depends on clang36

[Alexander Hansen]

> On Oct 5, 2015, at 12:10, David Phillips  wrote:
> 
> 
> Folks, 
>  I just upgraded to OSX 10.11 following Alexander Hansen’s instructions with 
> almost no troubles. However, now, when I perform a fink update-all, I receive 
> the following error message:
> 'Can't resolve dependency "clang36" for package "ghc-7.8.3-102" (no matching 
> packages/versions found)
> Exiting with failure.’
> I think I only have ghc installed because pandoc depends on it, but I’d 
> rather not uninstall the whole chain if I can avoid it as I do use pandoc 
> from the fink build.
>   Also I’m not sure who the maintainer for ghc is at this point to cc on this 
> message.
> 
>  David

Fionna:LWP-Pod-20150916 hansen$ fink info ghc | grep Maintainer
 Maintainer: None 

We’ll have to see whether ghc can be updated for clang37.  If you’d like to 
explore this locally on your own system, you can try the following:

1) Copy /sw/fink/dists/stable/main/finkinfo/languages/ghc-10.11.info and 
/sw/fink/dists/stable/main/finkinfo/languages/ghc.patch over to 
/sw/fink/dists/local/main/finkinfo
2) Edit /sw/fink/dists/local/main/finkinfo/ghc.info 
a) Update the revision to 102.1
b) Use clang37 and llvm37 instead of clang36 and llvm36 in the 
BuildDepends
c) Edit the SetCC line to use “llvm-3.7” instead of “llvm-3.6” 
c) Edit the ConfigureParams block to use “llvm-3.7” instead of 
“llvm-3.6” 

Since you actually use it, you’re more likely to see problems than I am.


-- 
Alexander Hansen, Ph.D.
Fink User Liaison


--
___
Fink-users mailing list
Fink-users@lists.sourceforge.net
List archive:
http://news.gmane.org/gmane.os.macosx.fink.user
Subscription management:
https://lists.sourceforge.net/lists/listinfo/fink-users

[Fink-users] Octave-qtmac fails to build on 10.11 due to broken compiler

[Gary K. Olson]

Alex Hansen’s fix over weekend fixed the system-perl problem, so I decided to 
build a few active packages that I hadn’t built under 10.10. Several packages 
couldn’t build because of broken c compiler.  Here is message on attempting to 
build octave-qtmac that was typical:

> 
checking for style of include used by make... GNU
checking for x86_64-apple-darwin-gcc... oct-cc
checking whether the C compiler works... no
configure: error: in `/sw/src/fink.build/octave-qtmac-3.8.2-6/octave-3.8.2':
configure: error: C compiler cannot create executables
See `config.log' for more details
### execution of /tmp/fink.TaewQ failed, exit code 77
### execution of /tmp/fink._N1rF failed, exit code 77
Removing runtime build-lock...
Removing build-lock package...
/sw/bin/dpkg-lockwait -r fink-buildlock-octave-qtmac-3.8.2-6
(Reading database ... 479268 files and directories currently installed.)
Removing fink-buildlock-octave-qtmac-3.8.2-6 ...
Reading buildlock packages...
/sw/bin/dpkg-lockwait -i /sw/fink/debs/fort77_1.18-19_darwin-x86_64.deb
Selecting previously deselected package fort77.
(Reading database ... 479267 files and directories currently installed.)
Unpacking fort77 (from .../fort77_1.18-19_darwin-x86_64.deb) ...
Setting up fort77 (1.18-19) ...

Failed: phase compiling: octave-qtmac-3.8.2-6 failed

I assumed that this is due to still calling for gcc4.9? Any help would be 
appreciated. Thank you.

Gary K Olson

my kit:
Package manager version: 0.39.1
Distribution version: selfupdate-rsync Sun Oct  4 19:03:20 2015, 10.11, x86_64
Trees: local/main stable/main local/injected
Xcode.app: 7.0.1
Xcode command-line tools: 7.0.0.0.1.1441394355
Max. Fink build jobs:  4--
___
Fink-users mailing list
Fink-users@lists.sourceforge.net
List archive:
http://news.gmane.org/gmane.os.macosx.fink.user
Subscription management:
https://lists.sourceforge.net/lists/listinfo/fink-users

Re: [Fink-users] update-all error: ghc depends on clang36

[David Phillips]

Alas, the simple change to ghc to build with clang3.7 and llvm3.7 didn’t work. 
Here’s the end of the build with the error message and the immediately previous 
make. 
  I need to run off to a meeting and so I will not get to the next step of 
building with a single core instead of all of them to get a sensible error 
message until this evening. Here’s the end of the build, though.
  David


"inplace/bin/ghc-stage1" -optc-m64 -optc-fno-stack-protector -optc-Wall 
-optc-Wextra -optc-Wstrict-prototypes -optc-Wmissing-prototypes 
-optc-Wmissing-declarations -optc-Winline -optc-Waggregate-return 
-optc-Wpointer-arith -optc-Wmissing-noreturn -optc-Wnested-externs 
-optc-Wredundant-decls -optc-Iincludes -optc-Iincludes/dist 
-optc-Iincludes/dist-derivedconstants/header 
-optc-Iincludes/dist-ghcconstants/header -optc-Irts -optc-Irts/dist/build 
-optc-DCOMPILING_RTS -optc-fno-strict-aliasing -optc-fno-common -optc-DDTRACE 
-optc-O2 -optc-fomit-frame-pointer -optc-DRtsWay=\"rts_p\" -optc-funroll-loops 
-static -prof  -H32m -O -Iincludes -Iincludes/dist 
-Iincludes/dist-derivedconstants/header -Iincludes/dist-ghcconstants/header 
-Irts -Irts/dist/build -DCOMPILING_RTS -package-name rts -dcmm-lint  -DDTRACE   
  -i -irts -irts/dist/build -irts/dist/build/autogen -Irts/dist/build 
-Irts/dist/build/autogen   -O2-c rts/sm/Evac.c -o 
rts/dist/build/sm/Evac.p_o
make[1]: *** read jobs pipe: No such file or directory.  Stop.
make[1]: *** Waiting for unfinished jobs
Makefile:64: recipe for target 'all' failed
make: *** [all] Broken pipe: 13
make: INTERNAL: Exiting with 1 jobserver tokens available; should be 12!
### execution of /tmp/fink.Q3xzN failed, exit code 2
### execution of /tmp/fink.hmAuW failed, exit code 2
Removing runtime build-lock...
Removing build-lock package...
/sw/bin/dpkg-lockwait -r fink-buildlock-ghc-7.8.3-102.1
(Reading database ... 328501 files and directories currently installed.)
Removing fink-buildlock-ghc-7.8.3-102.1 ...
Failed: phase compiling: ghc-7.8.3-102.1 failed

Before reporting any errors, please run "fink selfupdate" and try again.
Also try using "fink configure" to set your maximum build jobs to 1 and
attempt to build the package again.
If you continue to have issues, please check to see if the FAQ on Fink's
website solves the problem.  If not, ask on one (not both, please) of
these mailing lists:

The Fink Users List 
The Fink Beginners List .

Please try to include the complete error message in your report.  This
generally consists of a compiler line starting with e.g. "gcc" or "g++"
followed by the actual error output from the compiler.

Also include the following system information:
Package manager version: 0.39.1
Distribution version: selfupdate-rsync Mon Oct  5 15:43:09 2015, 10.11, x86_64
Trees: local/main stable/main stable/crypto local/injected
Xcode.app: 7.0.1
Xcode command-line tools: 7.0.0.0.1.1441394355
Max. Fink build jobs:  12



> On Oct 5, 2015, at 3:30 PM, Alexander Hansen  
> wrote:
> 
> 
>> On Oct 5, 2015, at 12:10, David Phillips  wrote:
>> 
>> 
>> Folks, 
>> I just upgraded to OSX 10.11 following Alexander Hansen’s instructions with 
>> almost no troubles. However, now, when I perform a fink update-all, I 
>> receive the following error message:
>> 'Can't resolve dependency "clang36" for package "ghc-7.8.3-102" (no matching 
>> packages/versions found)
>> Exiting with failure.’
>> I think I only have ghc installed because pandoc depends on it, but I’d 
>> rather not uninstall the whole chain if I can avoid it as I do use pandoc 
>> from the fink build.
>>  Also I’m not sure who the maintainer for ghc is at this point to cc on this 
>> message.
>> 
>> David
> 
> Fionna:LWP-Pod-20150916 hansen$ fink info ghc | grep Maintainer
> Maintainer: None 
> 
> We’ll have to see whether ghc can be updated for clang37.  If you’d like to 
> explore this locally on your own system, you can try the following:
> 
> 1) Copy /sw/fink/dists/stable/main/finkinfo/languages/ghc-10.11.info and 
> /sw/fink/dists/stable/main/finkinfo/languages/ghc.patch over to 
> /sw/fink/dists/local/main/finkinfo
> 2) Edit /sw/fink/dists/local/main/finkinfo/ghc.info 
>   a) Update the revision to 102.1
>   b) Use clang37 and llvm37 instead of clang36 and llvm36 in the 
> BuildDepends
>   c) Edit the SetCC line to use “llvm-3.7” instead of “llvm-3.6” 
>   c) Edit the ConfigureParams block to use “llvm-3.7” instead of 
> “llvm-3.6” 
> 
> Since you actually use it, you’re more likely to see problems than I am.
> 
> 
> -- 
> Alexander Hansen, Ph.D.
> Fink User Liaison
> 


--
___
Fink-users mailing list
Fink-users@lists.sourceforge.net
List archive:

Re: [Fink-users] Octave-qtmac fails to build on 10.11 due to broken compiler

[Martin Costabel]

On 5/10/15 21:53, Gary K. Olson wrote:
> Alex Hansen’s fix over weekend fixed the system-perl problem, so I
> decided to build a few active packages that I hadn’t built under 10.10.
> Several packages couldn’t build because of broken c compiler.  Here is
> message on attempting to build octave-qtmac that was typical:
>
>>
> checking for style of include used by make... GNU
> checking for x86_64-apple-darwin-gcc... oct-cc
> checking whether the C compiler works... no
> configure: error: in `/sw/src/fink.build/octave-qtmac-3.8.2-6/octave-3.8.2':
> configure: error: C compiler cannot create executables
> See `config.log' for more details
[]
> Failed: phase compiling: octave-qtmac-3.8.2-6 failed
>
> I assumed that this is due to still calling for gcc4.9?

No, it is due to fink-octave-scripts not yet aware of the existence of 
El Capitan. You can paper it over as follows:

Edit the two files /sw/bin/oct-cc and /sw/bin/oct-cxx and add 15 to the 
list of recognized OS versions. Here are the diffs:

--- /sw/bin/oct-cc~ 2015-10-02 21:03:28.0 +0200
+++ /sw/bin/oct-cc  2015-10-04 11:46:27.0 +0200
@@ -3,3 +3,3 @@
  case $osversion in
-   1[1234])
+   1[12345])
COMPILER=/sw/var/lib/fink/path-prefix-clang/cc ;;
--- /sw/bin/oct-cxx~2015-10-02 21:03:28.0 +0200
+++ /sw/bin/oct-cxx 2015-10-04 11:48:08.0 +0200
@@ -3,3 +3,3 @@
  case $osversion in
-   1[34])
+   1[345])
COMPILER=/sw/var/lib/fink/path-prefix-libcxx/c++ ;;

With this fix, octave382 should build (it did for me, anyway).

-- 
Martin




--
___
Fink-users mailing list
Fink-users@lists.sourceforge.net
List archive:
http://news.gmane.org/gmane.os.macosx.fink.user
Subscription management:
https://lists.sourceforge.net/lists/listinfo/fink-users

Re: [Fink-users] update-all error: ghc depends on clang36

[David Phillips]

Building GHC with only one core following Alexander's instructions for 
modifying the info file for GHC worked without trouble - if a bit slowly with 
only one core!
  Thanks for the help. 
 David


> On Oct 5, 2015, at 4:09 PM, Alexander Hansen  
> wrote:
> 
> 
>> On Oct 5, 2015, at 12:55, David Phillips  wrote:
>> 
>> 
>> 
>> Alas, the simple change to ghc to build with clang3.7 and llvm3.7 didn’t 
>> work. Here’s the end of the build with the error message and the immediately 
>> previous make. 
>> I need to run off to a meeting and so I will not get to the next step of 
>> building with a single core instead of all of them to get a sensible error 
>> message until this evening. Here’s the end of the build, though.
>> David
>> 
>> 
>> "inplace/bin/ghc-stage1" -optc-m64 -optc-fno-stack-protector -optc-Wall 
>> -optc-Wextra -optc-Wstrict-prototypes -optc-Wmissing-prototypes 
>> -optc-Wmissing-declarations -optc-Winline -optc-Waggregate-return 
>> -optc-Wpointer-arith -optc-Wmissing-noreturn -optc-Wnested-externs 
>> -optc-Wredundant-decls -optc-Iincludes -optc-Iincludes/dist 
>> -optc-Iincludes/dist-derivedconstants/header 
>> -optc-Iincludes/dist-ghcconstants/header -optc-Irts -optc-Irts/dist/build 
>> -optc-DCOMPILING_RTS -optc-fno-strict-aliasing -optc-fno-common 
>> -optc-DDTRACE -optc-O2 -optc-fomit-frame-pointer -optc-DRtsWay=\"rts_p\" 
>> -optc-funroll-loops -static -prof  -H32m -O -Iincludes -Iincludes/dist 
>> -Iincludes/dist-derivedconstants/header -Iincludes/dist-ghcconstants/header 
>> -Irts -Irts/dist/build -DCOMPILING_RTS -package-name rts -dcmm-lint  
>> -DDTRACE -i -irts -irts/dist/build -irts/dist/build/autogen 
>> -Irts/dist/build -Irts/dist/build/autogen   -O2-c rts/sm/Evac.c 
>> -o rts/dist/build/sm/Evac.p_o
>> make[1]: *** read jobs pipe: No such file or directory.  Stop.
>> make[1]: *** Waiting for unfinished jobs
>> Makefile:64: recipe for target 'all' failed
>> make: *** [all] Broken pipe: 13
>> make: INTERNAL: Exiting with 1 jobserver tokens available; should be 12!
>> ### execution of /tmp/fink.Q3xzN failed, exit code 2
>> ### execution of /tmp/fink.hmAuW failed, exit code 2
>> Removing runtime build-lock...
>> Removing build-lock package...
>> /sw/bin/dpkg-lockwait -r fink-buildlock-ghc-7.8.3-102.1
>> (Reading database ... 328501 files and directories currently installed.)
>> Removing fink-buildlock-ghc-7.8.3-102.1 ...
>> Failed: phase compiling: ghc-7.8.3-102.1 failed
>> 
>> Before reporting any errors, please run "fink selfupdate" and try again.
>> Also try using "fink configure" to set your maximum build jobs to 1 and
>> attempt to build the package again.
>> If you continue to have issues, please check to see if the FAQ on Fink's
>> website solves the problem.  If not, ask on one (not both, please) of
>> these mailing lists:
>> 
>>The Fink Users List 
>>The Fink Beginners List .
>> 
>> Please try to include the complete error message in your report.  This
>> generally consists of a compiler line starting with e.g. "gcc" or "g++"
>> followed by the actual error output from the compiler.
>> 
>> Also include the following system information:
>> Package manager version: 0.39.1
>> Distribution version: selfupdate-rsync Mon Oct  5 15:43:09 2015, 10.11, 
>> x86_64
>> Trees: local/main stable/main stable/crypto local/injected
>> Xcode.app: 7.0.1
>> Xcode command-line tools: 7.0.0.0.1.1441394355
>> Max. Fink build jobs:  12
> 
> I’m still in the middle of building.  Since I’ve got fewer cores that often 
> results in success.  
> 
> The error message does indeed suggest that there’s a race condition in the 
> build somewhere.
> 
> 
> 
> 
> -- 
> Alexander Hansen, Ph.D.
> Fink User Liaison
> 

--
___
Fink-users mailing list
Fink-users@lists.sourceforge.net
List archive:
http://news.gmane.org/gmane.os.macosx.fink.user
Subscription management:
https://lists.sourceforge.net/lists/listinfo/fink-users

Re: [Fink-users] update-all error: ghc depends on clang36

[Alexander Hansen]

> On Oct 5, 2015, at 19:25, David Phillips  wrote:
> 
> 
> Building GHC with only one core following Alexander's instructions for 
> modifying the info file for GHC worked without trouble - if a bit slowly with 
> only one core!
>  Thanks for the help. 
> David
> 

For a further data point, do you have fink’s make installed? 

 On 10.11 it appears to trigger race conditions more often than does Apple’s 
make, so if you have a similar error in the future you might try removing 
Fink’s make before reducing your build jobs.
-- 
Alexander Hansen, Ph.D.
Fink User Liaison


--
___
Fink-users mailing list
Fink-users@lists.sourceforge.net
List archive:
http://news.gmane.org/gmane.os.macosx.fink.user
Subscription management:
https://lists.sourceforge.net/lists/listinfo/fink-users