FW: ports/115449: [MAINTAINER] astro/boinc-setiathome-enhanced:update to 5.27

Tue Aug 21 12:48:22 PDT 2007

Ladan, R.C. schreef:
> [ Better discuss it via non-$REALJOB email address... ]
> 
> -----Oorspronkelijk bericht-----
> Van: Martin Wilke [mailto:miwi at FreeBSD.org]
> Verzonden: di 8/21/2007 12:29
> Aan: Ladan, R.C.
> Onderwerp: Re: ports/115449: [MAINTAINER] astro/boinc-setiathome-enhanced:update to 5.27
>  
> On Mon, Aug 20, 2007 at 09:33:36PM +0200, Ladan, R.C. wrote:
>>> -----Oorspronkelijk bericht-----
>>> Van: Martin Wilke [mailto:miwi at FreeBSD.org]
>>> Verzonden: ma 8/20/2007 5:05
>>> Aan: Ladan, R.C.
>>> Onderwerp: Re: ports/115449: [MAINTAINER] astro/boinc-setiathome-enhanced:  update to 5.27
> 
>>> -----BEGIN PGP SIGNED MESSAGE-----
>>> Hash: SHA1
>>> On Mon, 20 Aug 2007 11:13:38 +0200
>>> "Ladan, R.C." <r.c.ladan at REAL.JOB> wrote:
>>> |Hi Martin,
>>> |
>>> |Looking at the log, it seems you are using the 5.13 version of Makefile
>>> |and distinfo instead of the 5.27 version.
>>> |The new version doesn't need the patch to sqlblob.h anymore as it is
>>> |integrated into the source code.
>>> |
>>> |Regards,
>>> |Rene
>>> Hi Rene,
>>> That's is really a build error :(
>>> http://head.miwibox.org/tb/index.php?action=describe_port&id=1806
>>> http://amd64.miwibox.org/tb/index.php?action=describe_port&id=1245
>> Nope!  I can still build the port using the base-gcc 4.2.1 on my recent i386-CURRENT box.
>> After comparing the build logs, it seems that something is wrong with your build environment:
> 
>> --- test-i386-gcc421-fbsd7	2007-08-20 20:48:28.000000000 +0200
>> +++ /home/rene/boinc-setiathome-enhanced-5.27.log	2007-08-20 21:02:01.000000000 +0200
>> @@ -1,9 +1,418 @@
>> -Script started on Mon Aug 20 20:45:55 2007
>> -root at 195-241-234-69:/usr/ports/astro/boinc-setiathome-enhanced#make
> 
> 
>> +building boinc-setiathome-enhanced-5.27 in directory /usr/local/tinderbox/7-Head
>> +maintained by: r.c.ladan at gmail.com
> 
>> [... deps ...]
> 
>> @@ -15,7 +424,7 @@
>>  checking target system type... i386-portbld-freebsd7.0
> 
>>  checking for a BSD-compatible install... /usr/bin/install -c -o root -g wheel
> 
>>  checking whether build environment is sane... yes
> 
>> -checking for a thread-safe mkdir -p... /usr/local/bin/gmkdir -p
> 
>> +checking for a thread-safe mkdir -p... ./install-sh -c -d
>>                                         ^^^^^
>>  (this is because of ports/sysutils/coreutils)
> 
>>  checking for gawk... no
> 
>>  checking for mawk... no
> 
>>  checking for nawk... nawk
> 
>> @@ -43,8 +452,8 @@
>>  checking for tr... /usr/bin/tr
> 
>>  checking for ar... /usr/bin/ar
> 
>>  checking for grep... /usr/bin/grep
> 
>> -checking for autoconf... /usr/local/bin/autoconf
> 
>> -checking for autoheader... /usr/local/bin/autoheader
> 
>> +checking for autoconf... no
>> +checking for autoheader... no
>>                                   ^^^^^
>> (because of devel/autotools)
> 
>>  checking for astyle... no
> 
>>  checking for indent... /usr/bin/indent
> 
>>  checking for cp... /bin/cp
> 
>> @@ -329,11 +738,9 @@
>>  checking mach/mach_time.h usability... no
> 
>>  checking mach/mach_time.h presence... no
> 
>>  checking for mach/mach_time.h... no
> 
>> -checking xmmintrin.h usability... yes
> 
>> +checking xmmintrin.h usability... no
>>  checking xmmintrin.h presence... no
> 
>> -configure: WARNING: xmmintrin.h: accepted by the compiler, rejected by the preprocessor!
> 
>> -configure: WARNING: xmmintrin.h: proceeding with the compiler's result
> 
>> -checking for xmmintrin.h... yes
> 
>> +checking for xmmintrin.h... no
>>  checking emmintrin.h usability... yes
> 
>>  checking emmintrin.h presence... yes
> 
>>  checking for emmintrin.h... yes
> 
> 
>> [... remaining config/build ...]
> 
>> The file xmmintrin.h is located in /usr/include and actually part of GCC since at least 2002.  It defines all the mm_* functions, this is where all (!) your boxes fail.  T
> 
>> Maybe you've set something strange in a config file?  I cannot really imagine that make *world does CPU-detection other than what you tell it in/at (... first install? )
> 
>> Because the tests for xmmintrin.h fail but those for emmintrin.h succeed at your side, configure #defines USE_INTRINSICS which tell the code to use the mm_* functions.  Consequently, the build fails.
> 
>> Regards,
>> Rene
> 
> Rene both tb's is from 19.08.07 with gcc 4.2.1, this build failed localy here build from 17.08 reproduct can you all with
>  ports-mgmt/porttools and "port test".
> 
I just did that (again, also when testing the update locally), it just builds fine.
The log is at http://home.tiscali.nl/rladan/freebsd/port-test-sah527.bz2
SIZE = 8322
SHA256 = cf4e227af94b25187f0ba5d7f2682902c75e8f2a69a1400ca9ed8ee2e864edfb

If you look at http://setiathome.berkeley.edu/results.php?userid=211311
(could be down right now), you can see that the 5.27 client produces valid results.

I also looked for xmmintrin.h in /usr/src/ObsoleteFiles.inc but didn't find anything
in there.
-- 
GPG fingerprint = E738 5471 D185 7013 0EE0  4FC8 3C1D 6F83 12E1 84F6 (subkeys.pgp.net)

"It won't fit on the line."
		-- me, 2001