Ld: cycle in dylib re-exports with /usr/X11/lib/libGL.dylib liker error

Hello,
I found out that some open source applications when tied to be installed from source or trough fink or mac ports simply fail at linking with the error:
ld: cycle in dylib re-exports with /usr/X11/lib/libGL.dylib
However, when I force the linker to use /System/Library/Frameworks/OpenGL.framework/Versions/A/Libraries/libGL.dylib instead it works.
I am wondering what is the problem since /usr/X11/lib/libGL.dylib has to be exported from /System/Library/Frameworks/OpenGL.framework/Versions/A/Libraries/libGL.dylib.
It is very painful to force configure to use right version of libGL.dylib. Since the two libraries are of
different size the should be different. Simple replacement of /usr/X11/lib/libGL.dylib by /System/Library/Frameworks/OpenGL.framework/Versions/A/Libraries/libGL.dylib would work, but what are the consequences for the other applications?
I would be very happy is someone from OS X developement would reply to this and address this problem in the future.
Regards,
Slobodan

Post to the Unix forum under OS X Technologies and you should get a knowledgeable answer.

Similar Messages

Problems with /usr/ccs/lib/cpp

HI,
I have problems with /usr/ccs/lib/cpp
I want get the pre-compiled program
I used a simple program
$ cat x.p
#include <stdio.h>
main()
printf ( "Process\n" );
When I used the command
$ /usr/ccs/lib/cpp x.p > y.c
The file y.c has 0 bytes.
Pls your help. I am using Sun forte 6 Update 2

You should also consider whether you should be using cpp at all.
/usr/ccs/bin/cpp is the old K&R prepocessor, not the standard C preprocessor that is part of the C compiler. To use the standard C preprocessor, run "cc" using the -P or -E option.
The -P option creates a .i file and includes only the prepocessed code.
The -E option writes to standard output, and include file and line directives showing the source of all included text.

Can I copy fonts from /usr/X11/lib/X11/fonts to /Library/Fonts ?

Hi,
Short Version
The short version of my question is in the subject (specifically, I'm referring to files of type afm, pfb, pfa and maybe some other sort of pf? located in /usr/X11/lib/X11/fonts/Type1/)
Long Version
Here's the long version, along with an explanation of why I'm interested. Please keep in mind that the only thing I know about fonts is what is contained in the xfig documentation.
I'm using LaTeX Equation Editor. (It, in turn, is using the TeXlive 2007 distribution of (La)TeX. Yes that replaced some "system files" like /usr/bin/pstopdf, but I recovered them from another Leopard Mac.) However, the equations which LaTeX Equation Editor is rendering just look wrong; for example, the equation $a=b$ renders in a boxy upright font, whereas the symbols are supposed to be italicized (math mode). Some symbols (for example $\epsilon$ a.k.a. ε) just don't appear. However, I found that the directory /tmp/random string of hyphenated numbers/ contained equation.tex, equation.eps and equation.pdf (among other files). Examining equation.eps using gv renders the equation correctly. However, viewing equation.pdf in Preview reproduces the faulty rendering. The faulty pdf is reproduced by regenerating equation.pdf using epstopdf (as supplied by TeXLive). Using (Apple's) pstopdf instead results in a slightly better result, but symbols like ε are still missing.
Just using the eps file is not useful (in addition to being kludgy) because the problem is also reproduced if I import the eps file into Keynote. Using Grab to take a screen shot of gv's successful rendering eliminates the transparent background, which results in an ugly import into Keynote.
Incidentally, I did not have this problem when using Tiger.
I'm assuming, particularly since gv didn't even work properly until I had installed fonts following the xfig documentation, that this is a font issue, and that, in particular, my X11 fonts are not transferring to Aqua. Is there a way of transferring the X11 fonts I installed for gv etc. to Aqua and if so how? Do I have to worry that font files in Aqua appear to be named after the name of the fonts, whereas under X11, the map from font name to font file appears to be the purpose of the extra file /usr/X11/lib/X11/fonts/fonts.dir?
Thanks!

You must in fact add your custom fonts, as the default font set will be the same on all macs running Lion.
You do not offer details, but I may confess what I do in order to have the same font set on all macs at home, just in case someone needs to work on another computer. Usually, they are linguistic fonts.
- I do not install additional fonts, bar the first time when I install a new Mac OS
- save all fonts stored in System/Library/Fonts and /Library/Fonts unto an xternal drive, just in case something wrong happems
- I have a folder with my custom fonts, which are added to all macs around
I do not know what you want to do, but I offered a hint.

Export with consistent=y raise snapshot too old error.

Hi,
Oracle version:9204
It raises
EXP-00056: ORACLE error 1555 encountered
ORA-01555: snapshot too old: rollback segment number 2 with name
"_SYSSMU2$" too small
when I do an export with consistent=y option.
And I find below information in alert_orcl.log
Wed Apr 20 07:50:01 2005
SELECT /*NESTED_TABLE_GET_REFS*/ "XXX"."TABLENAME".* FROM
"XXX"."TABLENAME"
ORA-01555 caused by SQL statement below (Query Duration=1140060307
sec, SCN: 0x0000.00442609):
The undo parameters:
undo_retention=10800(default value)
undo_retention is larger than the seconds run export(only 1800
seconds),so I think the default value is enough.
undo_management=auto(default value)
Maybe the rollback tablespace is too small(about 300M)? But I think oracle should increase the size of datafile in this mode.Is that right?
undo_tablespace=undotbs1
undo_suppress_errors=false
I think I must miss something.
Any suggestions will be very appreciated.
Thanks.
wy

UNDO_RETENTION is a request, not a mandate. If your UNDO tablespace is too small, Oracle may have to discard UNDO segments before UNDO_RETENTION is reached.
How much UNDO is your database generating every second?
SELECT stat.undoblks * param.value / 1024 / 1024 / 10 / 60 undo_mb_per_sec
FROM v$undostat stat,
v$parameter param
WHERE param.name = 'db_block_size'Justin
Distributed Database Consulting, Inc.
http://www.ddbcinc.com/askDDBC

Problems with /usr/local/lib after installing

After I installed Leopard, I noticed that some of my files in usr/local/lib seemed to have disappeared. When I tried to open /usr/local/lib I got the following error: 'lib: Too many levels of symbolic links.' and it appeared that there was also a 'lib 1' directory in /usr/local which had all of my files in it. I managed to fix it using:
sudo rm -rf /usr/local/lib
sudo cp -R '/usr/local/lib 1' /usr/local/lib
Just wanted to give a heads up since this was rather annoying to figure out. Other than this I have very few issues with leopard, so far limited to:
-DivX playback doesn't appear to work
-after login it appears to take a while for finder to start
-a few problems mounting external devices, such as my USB drive and iPhone

I’ve found this on my system as well! I had some trouble installing MacFuse under Leopard. The installer just said "Das Installations konnte einige Dateien nicht nach '/' kopieren. Wenden Sie sich an den Hersteller", which basically means that it couldn't write to the root directory.
Then I found out about the /usr/local/lib symlink. This is very disturbing and clearly a major bug. I think it's a problem with the installation process of the developer tools. I’ve upgraded to Leopard two weeks ago. Everything worked. Then I’ve installed XCode 3.0 this and then the problem started. The circular symlink in /usr/local/lib has the timestamp of the XCode 3.0 installation…

Compiling error with /usr/bin/ld: cannot find -l error

With the help of ToddLittle, I have solved a lot of problem with tuxedo as I am a freshman to it.Thank you ToddLittle!
However I meet a new problem with compiling my new program. I got this error:
/usr/bin/ld: cannot find -l/home/cjf/oracle/tuxedo12.1.1.0/include
where /home/cjf/oracle/tuxedo12.1.1.0 is my tuxedo installation path.
I build it under ubuntu 12.04 64 bit and my IDE is eclipse.
How to solve it? thanks!

Usually it is best to execute the script that is installed in the Tuxedo installation directory that sets up all the necessary environment variables. That script is $TUXDIR/tux.env. Source that script before trying to use buildclient or buildserver or any other Tuxedo command to solve most problems like this.
PS You might want to check out Solaris Studio if you want to use an IDE as there is a Tuxedo plug-in for it that really helps in developing Tuxedo applications.
Regards,
Todd Little
Oracle Tuxedo Chief Architect

X11: libfreetype.6.dylib

When I try to use the latest version of mplayer I get the following error: Incompatible library version: mplayer requires version 10.0.0 or later, but libfreetype.6.dylib provides version 6.3.0
Does someone know where I can find the 10.0.0 version of this library? All help is welcome, it's rather important.

actually I should provide more info.
I am trying to play EAC3 (only) audio tracks on my macbook. The only way I can think of to do this is to install mplayer. I have tried this (both from the official release site, and the Overlook build SVN-071205-01 [the most recent] from www.haque.net). Both crash immediately with the same error message:
dyld: Library not loaded: /usr/X11/lib/libfreetype.6.dylib
X11 is definitely installed (and reinstalled) on my macbook. When I initially got this message I updated to v 1.1.3. This had no impact. So I ran
sudo find / -name "libfreetype*"
I got this result:
/Applications/VLC.app/Contents/MacOS/modules/libfreetype_plugin.dylib
/usr/X11R6/lib/libfreetype.6.3.dylib
/usr/X11R6/lib/libfreetype.6.dylib
/usr/X11R6/lib/libfreetype.a
/usr/X11R6/lib/libfreetype.dylibsudo find / -name "libfreetype*"
based on that I copied the x11r6 directory and created a new one X11. When I ran mplayer I got the new error message ... ie, I need version 10 of libfreetype.

System Copy:Failure when starting export with SAPINST

We are busy with a Unicode conversion and needs to start the export with sapinst. All pre unicode conversion steps are done, we upgraded from 4.6C to ECC6.0. It is a distrubuted system running on MSSQL. Started the export on the DB server and got the follwing error below.
Regards
Roger
INFO 2007-02-08 20:09:27
Execute step stepPreconditions of component |NW_Export|ind|ind|ind|ind|0|0|NW_ABAP_Export_Dialog|ind|ind|ind|ind|4|0|NW_ABAP_Export|ind|ind|ind|ind|0|0.
INFO[E] 2007-02-08 20:09:27
FSL-02077 File system export (share) saploc does not exist.
INFO[E] 2007-02-08 20:09:27
FSL-02077 File system export (share) sapmnt does not exist.
ERROR 2007-02-08 20:09:27
CJS-00030 Assertion failed: in NW_ABAP_Export.stepPreconditions: Executable directory 'G:\usr\sap\HD3\SYS\exe\nuc\NTI386' must exist
ERROR 2007-02-08 20:09:27
MUT-03025 Caught ESAPinstException in Modulecall: ESAPinstException: error text undefined.
ERROR 2007-02-08 20:09:27
FCO-00011 The step stepPreconditions with step key |NW_Export|ind|ind|ind|ind|0|0|NW_ABAP_Export_Dialog|ind|ind|ind|ind|4|0|NW_ABAP_Export|ind|ind|ind|ind|0|0|stepPreconditions was executed with status ERROR .
INFO 2007-02-08 20:09:36
An error occured and the user decide to stop.\n Current step "|NW_Export|ind|ind|ind|ind|0|0|NW_ABAP_Export_Dialog|ind|ind|ind|ind|4|0|NW_ABAP_Export|ind|ind|ind|ind|0|0|stepPreconditions".

Hi All
These are the steps I had to perform when I faced the issue -
Background:
Faced the issue when I was exporting the DB instance (SQL 2000) using SAPINST.
Steps:
1. We are running NW04S Portal System.Copied the E:\usr\sap\DEV.... files from CI to DB server.
2. Defined sapmnt/saploc shares for E:\usr\sap folder on DB server.
3. Created the folder E:\usr\sap\DEV\SYS\exe\uc\NTI386\mssjdbc and copied the files base.jar,util.jar,sqlserver.jar,spy.jar files from E:\usr\sap\DEV\JC00\j2ee\jdbc folder.
4. Retry the SAPINST. Final confirmation screen appeared.
5. Checked all the parameters for correctness and clicked on start...export started and finished completely.
All these steps I have done just looking at the log file.One thing I am not still clear is why SAPINST is looking for \usr\sap/...files/folder in DB instance instead of App server.
Would appriciate if any one can clarify this.
Thanks
Naresh P.

Any restriction on Mac that binaries running from root dirs like /usr/libexec/ cannot load dylib's at run-time?

Hi all,
Is there any restriction on Mac OS X 10.6/10.7 that binaries running from root directories like /usr/libexec/ cannot load dylib's at run-time using dlopen() or NSAddImage()?
I am trying to run a binary like "sample" from a directory inside /usr/libexec/ which tries to load a dylib like "sample.dylib" which is present in the same directory as "sample".
It resulted in a segmentation fault with the following error and stack-trace.
Program received signal EXC_BAD_ACCESS, Could not access memory.
Reason: KERN_INVALID_ADDRESS at address: 0x0000000000003430
0x0000000000003430 in ?? ()
Using gdb, I could find out that the program crashed inside dlopen() (the path passed to dlopen() being a simple filename like "sample.dylib").
The dlopen() at the top of the stack-trace pasted below gets called around 10 times; it does not crash during this 2nd level dlopen() any-time.
#0 0x00007fff81be9e93 in dlopen ()
#1 0x00007fff82838f7f in map_images_nolock ()
#2 0x00007fff82821bc0 in map_images ()
#3 0x00007fff5fc039b9 in __dyld__ZN4dyldL18notifyBatchPartialE17dyld_image_statesbPFPKcS0_jPK15dyld_imag e_infoE ()
#4 0x00007fff5fc0c839 in __dyld__ZN11ImageLoader4linkERKNS_11LinkContextEbbRKNS_10RPathChainE ()
#5 0x00007fff5fc04d48 in __dyld__ZN4dyld4linkEP11ImageLoaderbRKNS0_10RPathChainE ()
#6 0x00007fff5fc08f68 in __dyld_dlopen ()
#7 0x00007fff81be9ec0 in dlopen ()
#8 0x000000010002694f in AAA ()
#9 0x0000000100026730 in XXX ()
#10 0x0000000100026494 in YYY ()
#15 0x00000001000566a9 in main (argc=7, argv=0x7fff5fbff810) at ../../main.c:50
After this 2nd level dlopen() gets called 10-15 times, the stack will unwind from map_images_nolock () till __dyld_dlopen (). But when it tries to finish all the instructions inside __dyld_dlopen () and step out of it, the program crashes with the EXC_BAD_ACCESS signal.
Please find below the final stack-traces obtained during debugging when I put breakpoints for AAA() and dlopen() and went on stepping inside till the program crashed.
(gdb) bt
#0 0x00007fff5fc04d48 in __dyld__ZN4dyld4linkEP11ImageLoaderbRKNS0_10RPathChainE ()
#1 0x00007fff5fc08f68 in __dyld_dlopen ()
#2 0x00007fff81be9ec0 in dlopen ()
#3 0x000000010002694f in AAA ()
#4 0x0000000100026730 in XXX ()
#5 0x000000010002663c in YYY ()
#11 0x00000001000566a9 in main (argc=7, argv=0x7fff5fbff7f0) at main.c:50
(gdb) step
Single stepping until exit from function __dyld__ZN4dyld4linkEP11ImageLoaderbRKNS0_10RPathChainE,
which has no line number information.
0x00007fff5fc08f68 in __dyld_dlopen ()
(gdb) bt
#0 0x00007fff5fc08f68 in __dyld_dlopen ()
#1 0x00007fff81be9ec0 in dlopen ()
#2 0x000000010002694f in AAA ()
#3 0x0000000100026730 in XXX ()
#4 0x000000010002663c in YYY ()
#10 0x00000001000566a9 in main (argc=7, argv=0x7fff5fbff7f0) at ../../main.c:50
(gdb) step
Single stepping until exit from function __dyld_dlopen,
which has no line number information.
Program received signal EXC_BAD_ACCESS, Could not access memory.
Reason: KERN_INVALID_ADDRESS at address: 0x0000000000003430
0x0000000000003430 in ?? ()
To summarise:
For some reason, a user wants to install a product and its binaries under /usr/libexec/.
This binary tries to dynamically load a dylib using dlopen("sample.dylib", RTLD_LAZY | RTLD_LOCAL).
It looks like, the dynamic linker then opens dependent dylibs and runs module initialisers. Then crashes when returning (which sounds like a trashed stack).
Can anyone please tell me whether there is some OS-imposed restriction or OS bug which can explain this type of behaviour?
(Note: The crash does not occur if the binary and dylib are installed under a normal user's home directory and run from there.)
Many thanks,
Vinay

The otool output for the binary and dylib are pasted below.
localhost:Latest dev$ otool -L /usr/libexec/Dipu/sample/sample
/usr/libexec/Dipu/sample/sample:
/System/Library/Frameworks/IOKit.framework/Versions/A/IOKit (compatibility version 1.0.0, current version 275.0.0)
/System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation (compatibility version 150.0.0, current version 476.18.0)
/usr/lib/libstdc++.6.dylib (compatibility version 7.0.0, current version 7.4.0)
/usr/lib/libgcc_s.1.dylib (compatibility version 1.0.0, current version 1.0.0)
/usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 111.1.4)
localhost:Latest dev$ otool -L /usr/libexec/Dipu/sample/sample.dylib
/usr/libexec/Dipu/sample/sample.dylib:
libProduct.dylib (compatibility version 1.0.0, current version 1.0.0)
/System/Library/Frameworks/Carbon.framework/Versions/A/Carbon (compatibility version 2.0.0, current version 136.0.0)
/System/Library/Frameworks/IOKit.framework/Versions/A/IOKit (compatibility version 1.0.0, current version 275.0.0)
/usr/lib/libstdc++.6.dylib (compatibility version 7.0.0, current version 7.4.0)
/usr/lib/libgcc_s.1.dylib (compatibility version 1.0.0, current version 1.0.0)
/usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 111.1.4)
/System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation (compatibility version 150.0.0, current version 476.18.0)

/ /System/Library/CoreServices/ReportCrash /usr/lib/libgenkit.dylib /usr/share/icu/icudt51l.dat /private/var/folders/_p/d1ckn47x4wsghc2vb1rk82vm0000gn/C/com.apple.LaunchServic es-044501.csstore /usr/lib/dyld /private/var/db/dyld/dyld_shared_cache_x86_64 /

Ever since I upgrade to Maverick I keep getting this error. Any have an idea. Its really messing up my computer.

Hi,
Your doing what and when this appears would be useful.
It also looks like it has extra spaces dotted about in this copy of the info.
There appear to be several spaces before some of the forward slashes that are not expected.
On the other hand it could be several different lines of a report as when I look ReportCrash is an Executable item and cannot be opened further.
/ /System/Library/CoreServices/ReportCrash
/usr/lib/libgenkit.dylib /usr/share/icu/icudt51l.dat
/private/var/folders/_p/d1ckn47x4wsghc2vb1rk82vm0000/C/com.apple.LaunchServic es-044501.csstore
/usr/lib/dyld /private/var/db/dyld/dyld_shared_cache_x86_64 /d
Did it read like the above originally ?
It would appear that Report Crash is triggered and it might have something do do with something in Launch Services but then again they might not be related other than an app reporting it is using those two things .
8:40 pm Saturday; May 3, 2014
 iMac 2.5Ghz i5 2011 (Mavericks 10.9)
 G4/1GhzDual MDD (Leopard 10.5.8)
 MacBookPro 2Gb (Snow Leopard 10.6.8)
 Mac OS X (10.6.8),
 Couple of iPhones and an iPad

/usr/lib/System.dylib contains just ppc code on ppc Mac OSX

Hello!
Trying to compile universal binaries on ppc Mac OSX fails sometimes because when a project links against /usr/lib/System.dylib for both architectures, it fails for the i386 part. On i386 Mac OSX is works fine because there System.dylib contains ppc AND i386 code.
I installed the latest XCode 2.2.1 from the DVD (adding all Cross Development Support) and ran the ppc Combo Update 10.4.6 but still ended up with a flat System.dylib.
Is there a way to install ppc Mac OSX so that all its libraries and executables contain i386 code as well?
Regards
Philipp

Jeff,
Unless someone comes up with a better suggestion (it is almost midnight with our forum staff friends), you may consider a reinstallation (uninstall, run the Cleaner Tool, and reinstall).
http://www.adobe.com/support/contact/cscleanertool.html

Error message ''Disk Full' when trying to export with CS6, works fine in 5.5

I get an error message when I try to export media in Premiere CS6 that I don't have in CS5.5
I use export media for a AVCHD 60fps sequence with a few photoshop images on top of the main track. In the export settings I choose ''match sequence settings'' which gives me a .mpeg I- frame, 1920 x 1080 P, 60fps, quality 50, and a I tick the ''use maximum render quality'' check box at the bottom.
It used to work fine but now I get a message: Disk Full where there is 1.5 TB free on the disk?
I have the exact same sequence made in Premiere Pro CS 5.5 and it exports fine.
It started to act weird after I used the queue for the first time and it went to the CS6 encoder.
Since I used that once and it gave me that message for the first time, even when I try to do the same thing directly in Premiere CS6 , I always get the same error message.
Question 2, is there a place to change the settings from 50 to 100 for the quality of this export profile?
By the way I can export with other profiles without problems but this setting is the only one that I've seen that lets me keep my 60 frames per second in the export
process. (Example: I can export this sequence in the H.264, 1920 x 1080, 30fps, high quality preset, using the maximum render quality in CS6)

Hi Guys,
Thanks for the input but if you read my message correctly and until the end I already can export to other settings and I want an answer to this problem. There's no logical explanation to why this suddenly wouldn't work with the same sequence that it used to do fine with and that CS 5.5 handles fine.
As I said earlier this is the only setting that lets me export my sequence with the 60 fps, that I know of, for further and easier manipulation. As my main camera provides me with these 60 fps I'd really like to be able to have them left in that transition phase and, no, I'm not gonna give up until I know, really, what's going on and solve that problem. If this is a bug it needs to be fixed, if not I want to know what's happening.
Also there's no answer to my question 2: Question 2, is there a place to change the settings from 50 to 100 for the quality of this export profile?
So for you experts out there the question is still hanging.

How do I stop Illustrator automatically adding numbers to file name when exporting with artboard?

I need to export with artboard. Illustrator automatically adds a number to my original file name. It won't even let me replace the old file. It is extremely annoying and makes flow with other programs really hard because they can't auto update, since the file name changes. When you have multiple files, you can't be manually renaming all the time. Anyone knows how to fix this? Thanks.

This has been discussed a few times, and here is a summary of the usual solutions.
http://forums.adobe.com/message/2989508#2989508
What is it EXACTLY you are trying to do. Are you trying to export a single page from filename.ai and make it filename.psd, (without any number afterwards). In that case you could delete the extra artboards prior to export, and then you won't have to checked use artbaords as there is only one page. Otherwise I would plan ahead so your end result will be filename-01.jpg, filename-02.jpg, and then everytime you use export artbaords those names will replace.
I am happy with the function as is, and would prefer to have a serial number automatically after the name when checking use artboards. But sounds liek you and others desire more options and thte ability to turn off adding to the filename.

I try to export my Imovie and I always get the error code"Unable to prepare project for publishing...File already open with write permission" it even does it when I am just trying to export to my external hard drive to save the file

I try to export my Imovie and I always get the error code"Unable to prepare project for publishing...File already open with write permission" it even does it when I am just trying to export to my external hard drive to save the file

Thanks!!! I had same error (with -49) and the mac had an external hard drive attached.. and it was doing a backup. Stopped, disconnected and the project exported!

HELP! What are the settings to export with the highest quality (I will show my project in a cinema.)

Hello,
I have a project I will be showing at a local cinema, and I need to have the best possible quality for that......
My source footage is SD, so I am exporting it as a MPEG2-DVD project, the first time I exported (with the default MPEG2-DVD settings) and burned it, it looked good (at least on my 34" TV) but I am not sure if that quality will be as good when projected on a cinema screen...
So, what I need to know is, which of the MPEG2-DVD export settings I must change in order to get the quality I am looking for.....
BTW I am using Adobe Premiere CS4 and the whole Creative Suite 4....
Thanks in advance
Al

If it looks good on your tv it will look good at the cinema.
The resolution will still be 720x480/576 if a dvd player is hooked up to the projector.

Ld: cycle in dylib re-exports with /usr/X11/lib/libGL.dylib liker error

Similar Messages

Maybe you are looking for